Tikal

The primary theoretical question addressed in this book focuses on the lingering concern
of how the ancient Maya in the northern Petn Basin were able to sustain large
populations in the midst of a tropical forest environment during the Late Classic period.
This book asks how agricultural intensification was achieved and how essential resources,
such as water and forest products, were managed in both upland areas and seasonal
wetlands, or bajos. All of these activities were essential components of an initially
sustainable land use strategy that eventually failed to meet the demands of an escalating
population. This spiraling disconnection with sound ecological principles undoubtedly
contributed to the Maya collapse. The books findings provide insights that broaden the
understanding of the rise of social complexity the expansion of the political economy,
specifically and, in general terms, the trajectory of cultural evolution of the ancient
Maya civilization.
DAVID L. LENTZ is Professor of Biological Sciences and Executive Director of the
Center for Field Studies at the University of Cincinnati. He has authored more than ninety
publications that have appeared as journal articles, book chapters, and three books,
including this volume. His research focuses on anthropogenic landscape changes of the
past and paleoethnobotanical studies in Mesoamerica, Central Asia, and North America.
NICHOLAS P. DUNNING is Professor of Geography at the University of Cincinnati. He
is a geoarchaeologist and cultural ecologist specializing in the Neotropics. He has
published several books and more than ninety articles and book chapters, including those
in this volume.
VERNON L. SCARBOROUGH is Distinguished University Research Professor and
Charles Phelps Taft Professor in the Department of Anthropology at the University of
Cincinnati. His work emphasizes sustainability and global water systems through an
examination of past engineered landscapes, using comparative ecological and
transdisciplinary perspectives. He has published ten books eight of them edited,
including this volume and authored more than ninety book chapters or journal articles.
 Tikal

Paleoecology of an Ancient Maya City

Edited by
David L. Lentz
University of Cincinnati
Nicholas P. Dunning
University of Cincinnati
and
Vernon L. Scarborough
University of Cincinnati
32 Avenue of the Americas, New York, NY 10013-2473, USA
Cambridge University Press is part of the University of Cambridge.
It furthers the Universitys mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107027930
Cambridge University Press 2015
This publication is in copyright. Subject to statutory exception and to the provisions of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published 2015
Printed in the United States of America
A catalog record for this publication is available from the British Library.
Library of Congress Cataloging in Publication data
Tikal : paleoecology of an ancient Maya city / edited by David L. Lentz, University
of Cincinnati, Nicholas P. Dunning, University of Cincinnati, Vernon L. Scarborough,
University of Cincinnati.
pages cm
Includes bibliographical references and index.
ISBN 978-1-107-02793-0 (hardback)
1. Paleoecology Guatemala Tikal Site. 2. Wetland agriculture Guatemala
Tikal Site. 3. Tikal Site (Guatemala) I. Lentz, David L. (David Lewis),
1951 II. Dunning, Nicholas P. (Nicholas Pierce), 1957 III. Scarborough, Vernon
L. (Vernon Lee), 1950
QE720.2.G9T55 2015
560.450972812dc23 2014015323
ISBN 978-1-107-02793-0 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of URLs
for external or third-party Internet Web sites referred to in this publication and does not
guarantee that any content on such Web sites is, or will remain, accurate or appropriate.
 Contents
List of Figures
List of Tables
Contributors
Editors
Foreword by Payson Sheets
Acknowledgments
1 Tikal Land, Water, and Forest: An Introduction
Nicholas P. Dunning, David L. Lentz, and Vernon L. Scarborough
2 The Evolution of an Ancient Waterworks System at Tikal
Vernon L. Scarborough and Liwy Grazioso Sierra
3 At the Core of Tikal: Terrestrial Sediment Sampling and Water Management
Brian Lane, Vernon L. Scarborough, and Nicholas P. Dunning
4 Bringing the University of Pennsylvania Maps of Tikal into the Era of
Electronic GIS
Christopher Carr, Eric Weaver, Nicholas P. Dunning, and Vernon L.
Scarborough
5 Examining Landscape Modifications for Water Management at Tikal Using
Three-Dimensional Modeling with ArcGIS
Eric Weaver, Christopher Carr, Nicholas P. Dunning, Lee Florea, and Vernon
L. Scarborough
6 Life on the Edge: Tikal in a Bajo Landscape
Nicholas P. Dunning, Robert E. Griffin, John G. Jones, Richard E. Terry,
Zachary Larsen, and Christopher Carr
7 Connecting Contemporary Ecology and Ethnobotany to Ancient Plant Use
Practices of the Maya at Tikal
Kim M. Thompson, Angela Hood, Dana Cavallaro, and David L. Lentz
8 Agroforestry and Agricultural Practices of the Ancient Maya at Tikal
David L. Lentz, Kevin Magee, Eric Weaver, John G. Jones, Kenneth B.
Tankersley, Angela Hood, Gerald Islebe, Carmen E. Ramos Hernandez,
and Nicholas P. Dunning
9 Fire and Water: The Archaeological Significance of Tikals Quaternary
Sediments
Kenneth B. Tankersley, Nicholas P. Dunning, Vernon L. Scarborough, John G.
Jones, Christopher Carr, and David L. Lentz
10 Fractious Farmers at Tikal
David Webster and Timothy Murtha
11 Material Culture of Tikal
Palma J. Buttles, Carmen E. Ramos Hernandez, and Fred Valdez Jr.
12 A Neighborly View: Water and Environmental History of the El Zotz Region
Timothy Beach, Sheryl Luzzadder-Beach, Jonathan Flood, Stephen Houston,
Thomas G. Garrison, Edwin Romn, Steve Bozarth, and James Doyle
13 Defining the Constructed Niche of Tikal: A Summary View
David L. Lentz, Nicholas P. Dunning, and Vernon L. Scarborough
References
Index
Figures
1.1. Tikal chronological chart
1.2. Map of the Maya Lowlands region
1.3. Map showing the situation of Tikal in a landscape of uplands and bajos
1.4. Map of central Tikal showing the location of Perdido Pocket Bajo and
Arroyo Corriental Bajo
2.1. Map showing the main catchments and associated reservoirs at Tikal
2.2. Scaled cross-sectional schematic of the southern half of the Tikal ridge-top
2.3. Map of the Temple, Palace, and Hidden Reservoir chain with the ancient
arroyo course and associated excavations
2.4. Cross-sectional composite of Palace Dam profiles
2.5. Artists reconstruction of Palace Dam with posited vertically stacked
sluices
2.6. Map of Late Preclassic construction around the TemplePalaceHidden
Reservoir chain
2.7. Analogous ravine spring setting at Naranjo today
2.8. Map of Corriental Reservoir coring and excavation locations
2.9. Map of Perdido Reservoir coring and excavation locations
3.1. Image of the ESP device and workers operating the foot lever for
extraction
4.1. The lower half of the Great Plaza Quadrangle map
4.2. The arrangement of the nine maps in the TR11 Detail Map series
4.3. Accuracy test points in a plaza group on the Temple IV Quadrangle map, a
chultun opening; and the center of the ruins of a range structure
4.4. Morris Jones demonstrating the use of the plane table and telescopic
alidade, and a local worker holding a stadia rod in a ruin
4.5. Map shows the Penn Project point of beginning; the Datum Benchmark,
and UCAPT point of beginning, the PETTY CO.B.M.
4.6. The horizontal position accuracy of the TR11 Detail Maps
4.7. Maps of the areas used for the vertical accuracy assessment
4.8. The elevation profiles along the Perdido transect
4.9. Vertical accuracy assessment comparing 103 benchmark and spot
elevations from the Tikal Report No. 11 detail maps to elevations from the NASA
AIRSAR mission
5.1. Structures that appear to be placed to block the flow of water and form
aguadas
5.2. Watershed for Temple, Palace, and Hidden Reservoirs
6.1. Map of a portion of the East Brecha showing aguadas and excavations
6.2. Soil pit in the Bajo de Santa Fe west of the Aguada de Terminos
6.3. Map of the Aguada de Terminos area
6.4. North profile of Op. 5C and Op. 2B
6.5. Trench profile (Op. 16) in Parcela 2, El Pinal, Bajo de Santa Fe
6.6. West profile of Op. 5F in the Aguada de Terminos
6.7. Pollen frequency diagram for the Aguada de Terminos (Op. 5F)
6.8. Pedons and carbon isotope profiles in the Bajo de Santa Fe east of Aguada
de Terminos
6.9. Perdido Pocket Bajo pedons and carbon isotope profiles
6.10. South profile of Op. 8D in Perdido Pocket Bajo
7.1. Species accumulation curves for trees and vines
7.2. Accumulated frequency of upland forest species for trees
7.3. Scanning electron micrographs of Brosimum alicastrum, Manilkara
zapota, and Protium copal
7.4. Scanning electron micrographs of wood samples recovered from a midden
in El Jaguar
8.1. Diagram of Grupo de Jaguar at Tikal including structures 7D20-44
8.2. Flotation apparatus used to extract archaeological plant remains from soil
8.3. Kim Thompson, Demetrio Crdova, and Marielos Corado conduct
botanical surveys
8.4. Upland biomass calculations
8.5. Voronoi diagram of hypothetical limits of extractive areas during the Late
Classic and Late Preclassic periods
8.6. Pollen profile from Aguada Vaca de Monte
8.7. Household 1 at Cern
8.8. Plant macroremains from Tikal
8.9. Excavation profile (north wall) from Corriental Reservoir
9.1. Location of Bajo la Justa, Tikal, and other sites mentioned in the text
 9.2. Scanning electron micrograph of a bipyramidal euhedral quartz crystal
9.3. Scanning electron micrograph of a bipyramidal euhedral zircon crystal
9.4. The trace element content of Maya Lowland sediment samples compared
to volcanoes in Guatemala, Mexico, and El Salvador
9.5. The trace element content of reservoir sediment from Tikal
11.1. Lithic artifacts of the UCAPT: obsidian edge altered blade, corner
modified obsidian blade, carved greenstone fragment, ovoid chert scraper, chert
agricultural tool
11.2. Miscellaneous UCAPT artifacts: incised bone awl fragment, carved chert
eccentric, shell button/ear flare segment, shell button/ear flare segment,
perforated shell pendant
11.3. Miscellaneous UCAPT artifacts: ceramic ear flare fragment, ceramic
stamp, and polychrome sherd with figure
11.4. Complete vessel from Operation 13: unnamed red bowl with rounded
base
11.5. Complete vessel from Operation 13: unnamed orange-polychrome
cylinder
11.6. Complete vessel from Operation 13: unnamed orange-polychrome tall
bowl
11.7. Nearly complete vessel from Operation 13: unnamed orange-polychrome
tripod dish
12.1. Location map of El Zotz and surroundings
12.2. Zotz Aguada excavation map
12.3. Zotz Aguada stratigraphy
12.4. Zotz Aguada 13A-2 Photo
12.5. Diagrams of pollen morphs (above) and biosilicates (below) from Zotz
12.6. Cross sections of Zotz Aguada
Tables
3.1. Volumetric estimates of Tikal reservoirs
4.1. The summary of horizontal and vertical accuracy assessments
6.1. Aguada de Terminos Op. 5C profile description
6.2. Selected physical and chemical characteristics of the A horizons of 16 soil
pedons
7.1. Top ten plant families in 5.95 ha forest survey in Tikal National Park
7.2. Diversity indices for all trees greater than 6 cm diameter at breast height
(DBH)
7.3. Most abundant species by stem density and dominant species by basal area
7.4. Sample areas at Tikal National Park and associated richness of trees and
vines
7.5. Ten most abundant tree species and eleven families represented in the
paleoethnobotanical remains
7.6. Jaccard Dissimilarity Matrix comparing paleoethnobotanical remains
recovered at Tikal with the modern forest and each habitat at the generic level
7.7. Socioeconomic value of woods present at Tikal as trees in the modern
forest and as paleoethnobotanical remains
8.1. Aboveground biomass of modern Tikal forests and estimates for forests
during the Late Classic period
8.2. Fuel and timber needs of Tikal during the Late Classic period
8.3. Comparison of climatic conditions of Tikal and Cern
8.4. Ancient plant remains identified at Tikal
9.1. Radiocarbon dates obtained from the Quaternary sediments of Tikal and
vicinity
9.2. X-ray diffraction analysis of reservoir, aguada, and bajo sediments from
Tikal and vicinity
9.3. Trace element content of Tikal reservoir sediments and comparative
volcanic sources
9.4. Stable carbon isotope values for insoluble soil organic matter from
reservoirs
10.1. Tikal landscape (slope + topographic position)
11.1. Obsidian artifacts from various excavations
11.2. Chert artifacts from various operations
11.3. Major temporal periods and associated types in the UCAPT collection
11.4. Quantity of sherds and chronological assessment per operation and subop
 Contributors
Timothy Beach is Centennial Professor in the Department of Geography and
Environment at the University of Texas at Austin. He is an elected AAAS Fellow
and has held fellowships from the Guggenheim Foundation and Dumbarton Oaks.
His research focuses on soil and agricultural systems, geomorphology, water,
environmental change, paleoclimate, and geoarchaeology.
Steven Bozarth is a researcher in the Department of Geography at the University of
Kansas. His research interests include palynology, opal phytolith analysis,
paleoenvironmental reconstruction, and ancient agriculture.
Palma J. Buttles is Manager and Senior Member of the Technical Staff at the
Software Engineering Institute at Carnegie Mellon University and Research Fellow
at the Texas Archaeological Research Laboratory, University of Texas at Austin. A
recent paper is People Capability Maturity: Contributing to Organizational
Success, Joburg Centre for Software Engineering Annual Report (2009), University
of the Witwatersrand, Johannesburg, South Africa.
Christopher Carr is a doctoral student at the University of Cincinnati in the
Department of Geography. A recent publication is entitled Map of the Ruins of
Tikal, El Petn, Guatemala and Georeferenced Versions of the Maps Therein, which
appeared in Tikal Report 11 (tDAR ID: 390922) http://core.tdar.org/project/390922.
Dana Cavallaro recently received her M.S. degree from the Department of
Biological Sciences at the University of Cincinnati. Her thesis is entitled
Reconstructing the Past: Paleoethnobotanical Evidence for Ancient Maya Plant Use
Practices at the Dos Pilas Site, Guatemala.
James Doyle is Post-Doctoral Associate in Pre-Columbian Studies at the Dumbarton
Oaks Research Library and Collection. He is the author of the article Re-Group on
E-Groups: Monumentality and Early Centers in the Middle Preclassic Maya
Lowlands, which appeared in Latin American Antiquity in 2012.
Jonathan Flood is a doctoral candidate at the University of Texas at Austin in the
Department of Geography and the Environment. His research interests include
archaeology, geoarchaeology, water quality, human ecology, fluvial geomorphology,
groundwater hydrology, and water chemistry.
Lee Florea is Professor at Ball State University in the Department of Geological
Sciences. His research interests include hydrogeology, geomorphology,
geochemistry, geophysics, carbonate geology, and biology with a focus on sinkholes,
caves, and springs.
Thomas G. Garrison is Lecturer in the Department of Anthropology and
Archaeology at the University of Southern California. He has been the director of the
El Zotz Archaeological Project since 2012 and is currently editing An Inconstant
Landscape: The Archaeology of El Zotz, Guatemala (Colorado) with Stephen
Houston.
Liwy Grazioso Sierra is Professor of Maya Archaeology at the Universidad de San
Carlos de Guatemala. A recent work by her is Ro Azul, belleza enclavada en el
trifinio Guatemala-Mxico-Belice, which appeared in Publicaciones
Mesoamericanas.
Robert E. Griffin is an Assistant Professor in Atmospheric Science at The
University of Alabama in Huntsville. His research interests include environmental
archaeology, GIS, and remote sensing across Central America.
Angela Hood is Senior Archaeologist at GAI Consultants, Inc., in Pittsburgh,
Pennsylvania. She is a recent graduate of the University of Cincinnati; her M.A.
thesis is entitled Testing the Veracity of Paleoethnobotanical Macroremain Data: A
Case Study from the Ceren Site, El Salvador.
Stephen Houston serves as Dupee Family Professor of Social Sciences at Brown
University. Among his recent publications are The Life Within: Classic Maya and the
Matter of Permanence (Yale) and an edited volume, The Shape of Script: How and
Why Writing Systems Change (SAR Press). A MacArthur Fellow, he has also held
fellowships and grants from the Guggenheim Foundation, the National Endowment
for the Humanities, and the National Science Foundation.
Gerald Islebe is Mendeley Member in Biological Sciences from El Colegio de la
Frontera Sur in Chetumal, Mexico, with interests in paleoecology, biogeography, and
global change. Among his recent publications is Holocene Vegetation and Climate
History of Central Quintana Roo, Yucatn Pennsula, Mexico in Review of
Palaeobotany and Palynology with Alicia Carrillo-Bastos.
John G. Jones is Senior Paleoethnobotanist with Archaeological Consulting
Services Ltd. in Tempe, Arizona. One of his recent papers is entitled Pollen
Evidence for Early Settlement and Agriculture in Northern Belize, which appeared
in Palynology.
Brian Lane is a doctoral student at the University of Oregon in the Department of
Anthropology. His research interests include the environment, Pacific Island studies,
Oceania archaeology, evolutionary archaeology, and Maya archaeology.
Zachary Larsen recently received his M.S. degree from Brigham Young University,
Provo, Utah, in the Department of Environmental Science with a focus on soil
archaeology. His research focus is on the geochemical analysis of the agricultural
and household activities at Uci, Yucatan, Mexico.
Sheryl Luzzadder-Beach is a Professor and Chair of the Department of Geography
and Environment at the University of Texas at Austin. A recent work by her is
Arising from the Wetlands: Mechanisms and Chronology of Landscape
Aggradation in the Northern Coastal Plain of Belize, with Timothy Beach, which
appeared in Annals of the Association of American Geographers.
Kevin Magee received his Ph.D. in Geography from the University of Cincinnati in
2011. His research expertise is in object-based imagery analysis methods, abstract
and vague retrieval models for image classification, and remote sensing for natural
and cultural resource management.
Timothy Murtha is Professor of Landscape Architecture at Pennsylvania State
University. His research interests include settlement patterns, settlement ecology,
demography and landscape archaeology, GIS, integrated modeling, and spatial
analysis.
Carmen E. Ramos Hernandez is an archaeologist who works for the Guatemalan
Institute for Anthropology and History. Her research focuses on ancient Maya
material culture.
Edwin Romn is Co-director of El Zotz Archaeological Project and a doctoral
student in the Teresa Lozano Long Institute of Latin American Studies at The
University of Texas at Austin. His research interests include Maya architecture,
political organization, and indigenous epistemology.
Payson Sheets is Professor in the Department of Anthropology at the University of
Colorado, Boulder. His area of specialization is Mesoamerican archaeology, hazards
research, lithic technology, ancient adaptations, remote sensing, and geophysical
applications.
Kenneth B. Tankersley is Associate Professor of Anthropology and Geology at the
University of Cincinnati. One of his recent works, Evidence for Volcanic Ash Fall
in the Maya Lowlands from a Reservoir at Tikal, Guatemala, appeared in the
Journal of Archaeological Science in 2011.
Richard E. Terry is Professor of Plant and Wildlife Sciences at Brigham Young
University. One of his recent works, Chemical Signatures of Middens at a Late
Classic Maya Residential Complex, Guatemala, coauthored with Markus Eberl and
Marco Alvarez, appeared in Geoarchaeology: An International Journal in 2012.
Kim M. Thompson recently completed her Ph.D. in the Department of Biological
Sciences at the University of Cincinnati and currently serves as a Lecturer in the
Department of Biological Sciences at Ohio University. Her research focuses on the
characterization of the Neotropical forests of the ancient Maya and genetic variation
in the valued fruit, timber, and latex tree, Manilkara zapota.
Fred Valdez Jr. is a Professor of Anthropology at The University of Texas at Austin.
A recent work by him is An Alternative Order: The Dualistic Economies of the
Ancient Maya, with V. Scarborough, which appeared in Latin American Antiquity.
Eric Weaver is a doctoral student in the Department of Geography at the University
of Cincinnati. His research interests include paleoclimatology, karst studies, and
Maya water management.
David Webster is a Professor in the Department of Anthropology at Pennsylvania
State University. Among his recent publications is Archaeology of Ancient Mexico
and Central America: An Encyclopedia (Routledge 2010) with Susan Toby Evans.
 Editors
Nicholas P. Dunning is Professor of Geography at the University of Cincinnati. He
is a geoarchaeologist and cultural ecologist specializing in the Neotropics. He has
published several books and more than ninety articles and book chapters including
those in this volume.
David L. Lentz is Professor of Biological Sciences at the University of Cincinnati
and Executive Director of the UC Center for Field Studies. He has published more
than ninety articles and three books, including this volume. Previous works include:
Imperfect Balance: Landscape Transformations in the Precolumbian Americas
(Columbia) and Seeds of Central America and Southern Mexico (New York
Botanical Garden Press), written with Ruth Dickau. A Fellow of the Linnean Society
of London and former Fulbright Scholar, he has received grants for his research on
the evolution of early plant domesticates and ancient landscape studies from the
National Science Foundation, the National Endowment for the Humanities, the
Wenner-Gren Foundation, the National Geographic Society, the Foundation for the
Advancement of Mesoamerican Studies, and other sources.
Vernon L. Scarborough is Distinguished University Research Professor and
Charles Phelps Taft Professor in the Department of Anthropology at the University
of Cincinnati. His work emphasizes sustainability and global water systems through
an examination of past engineered landscapes using comparative ecological and
transdisciplinary perspectives. In addition to editing Water and Humanity: A
Historical Overview for UNESCO, he is a steering committee member of IHOPE-
Global (Integrated History for the Future of the People of Earth ihopenet.org)
located at Uppsala University and an active organizer of the subgroup IHOPE-Maya.
He is a Senior Editor for WIREs Water Journal (Wiley-Blackwell) and a Series Editor
for the series New Directions in Sustainability and Society (Cambridge University
Press), having published ten books eight of them edited (inclusive of this volume)
and more than ninety book chapters or journal articles of his own.
 Foreword

I am honored to be asked to write the Foreword to such an important volume on the

ancient Maya site of Tikal, Guatemala, and how its inhabitants used and misused their
environment. The authors in this book address one of the towering issues facing our
society today: sustainability. Ancient and recent people living on our planet have had
impressive successes in achieving sustainable adaptations for hundreds and often for
thousands of years, yet others have misunderstood the consequences of maladaptations
and have collapsed. The ancient Maya have provided us with numerous examples of these
extremes. For instance, the Maya commoners in the village of Chan, in what is now
Belize, achieved a sustainable adaptation for more than two thousand years (Robin 2012),
and the lead editor of this volume, David L. Lentz, contributed research results at Chan
that help explain how that success was achieved. The city that rose near Chan,
Xunantunich, exemplifies the other extreme, as it only lasted for two centuries and
collapsed completely. The factors leading to success or failure in achieving a sustainable
adaptation are explored in this current volume.
The experience of Tikal is a mixed one, between that of Chan and Xunantunich, with
sustainability achieved early in its time as a complex society and maintained for a few
centuries. However, they crossed the threshold of sustainability, the tipping point, and
during their last centuries created such fundamental problems that they ceased to exist by
A.D. 900. Understanding the factors in sustainability successes and failures has been
beyond archaeologists abilities until recently. Fortunately, the sophistication of todays
researchers conducting multidisciplinary research including the natural sciences, social
sciences, and humanities allows serious exploration of this topic in past societies, with
direct implications for our society today and into the future.
The authors herein document how the ancient Maya in the Early Classic and into the
Middle Classic (see Figure 1.1) periods maintained an adaptation that supplied sufficient
food, firewood, wood for construction, and other resources, without degradation of soils or
rain forest. They developed a sophisticated system of agroforestry by favoring the
productive trees and avoided deforestation in their early centuries. They developed
sophisticated means of managing water in a challenging environment without surface
rivers or lakes. Sediment cores and excavations within reservoirs, bajos, and aguadas
provided information on natural and cultivated flora. In those sediments were smectite
clay concentrations demonstrating the frequent airfall depositions of volcanic ash that
increased soil fertility of the tropical soils. The population was growing but did not exceed
the carrying capacity of the environment until around the sixth century A.D.
Because farmers interacting with their crops and soils on a daily basis are (and were in
the past) acutely aware of edaphic conditions, I can only suspect the Tikal elite were
unaware of the longer-term consequences of passing that tipping point into unsustainable
conditions. Scholars have much to learn about elite-commoner relationships and their
changes through time. A greater disconnection between the world of the elites and that of
the farmers is probable. Also, Tikal was living in, and contributing to, a more bellicose
environment in the sixth century and later, and the elite may have decided that population
increase beyond sustainability was advantageous in warfare in the short run. At any rate,
this volume documents the factors in the Late Classic period that contributed to the
decline and ultimately to the total collapse of civilization at Tikal. Drought added further
stress to a system already on a downward spiral. At most, the drought accelerated the
demise, but it was not a sole cause. I suspect had Tikaleos done the terracing that was so
prevalent at Chan and Caracol, the Late Classic decline might have been slowed, but not
stopped.
What is ironic is that humanistic achievement at Tikal reached its apex during this time
of stress and failure of the adaptive infrastructure, as judged by the architecture, sculpture,
and other fine arts. The same disjuncture occurred at Tikals sister city, Copn, at about the
same time, with the same consequences.
This book is an essential resource for Mayanists interested in the great trajectory of one
of the worlds civilizations. Maya achievements were stellar in art, architecture,
astronomy/astrology, epigraphy, numerology, cosmology, and many other domains. As
exemplified by Chan for two millennia and by Tikal for a few centuries, sustainable
adaptations were achieved in spite of challenges in their environments. Decisions were
made, particularly in the sixth century, that undermined that success, and the consequences
of overpopulation relative to adaptation, leading to environmental degradation, are
presented in this volume. Therefore, this volume is an essential resource for scholars
worldwide in a wide range of disciplines who are studying sustainability. It should be
required reading of politicians as well, but I am not optimistic that they would want to
consider the difficult decisions that are necessary to maintain sustainability in our present
world. The mantra of growth being essential to the futures of societies is deeply
ingrained in our societies, yet it is antithetical to sustainability. It is a rare politician who
has the courage to realize and express this. Can we learn from this Tikal case, or do we
continue overpopulating and degrading our environment to our own long-term detriment?
Payson Sheets
 Acknowledgments
Funds for this research project were provided by the National Science Foundation (grant
#BCS-0810118), the Wenner-Gren Foundation (grant #7799), the Alphawood Foundation
of Chicago, the Foundation for the Advancement of Mesoamerican Studies, and the
University of Cincinnati. Without these funds this project could never have proceeded and
we are deeply grateful for their support.
Eric Ponciano of the Guatemalan Ministry of Culture and Sports; Juan Carlos Prez
Caldern, Director General of the Institute of Anthropology and History of Guatemala
(IDAEH); and Mnica Urquiz, Head of the Department of Prehispanic and Colonial
Monuments at IDAEH, provided invaluable insight, support, and guidance during the
permitting, field research, and reporting phases of this project. Ana Luca Arroyave served
as the observant and helpful IDAEH inspector for the 2010 field season. The
administrators of Tikal National Park, Armando Guilln, Elmer Tun, and Fredy Sosa, were
amazingly efficient and unquestionably helpful during our two field seasons (2009 and
2010) at Tikal. Likewise, staff members at Tikal National Park, namely, Benito Burgos,
Marco Tulio Castellanos, Mirta Cano, Luis Garca, and Beto Tesucn, were extremely
kind in sharing their expertise and knowledge of the park.
Two of Guatemalas preeminent archaeologists, Drs. Barbara Arroyo and Vilma Fialko,
freely shared ideas and information; without their advice and support, this project might
not have come to fruition. Also, Oswaldo Gmez, another archaeologist working at Tikal,
was extremely kind and shared data and advice in a spirit of true academic camaraderie.
Our team of professional field archaeologists, including Ana Luisa Arriola, Sheryl
Carcuz, Mara de los ngeles Corado, Claudine Escobar Durand, Marielos Corado,
Raquel Macario, Silvia Alvarado, Walter Burgos, Mauricio Daz, and Blanca Mijangos,
supervised the excavations and prepared drawings of artifacts and excavation profiles. We
are deeply grateful for their efforts and talents. Eric G. Coronel kindly conducted
phosphate testing on soil samples. We thank our colleague Benjamin Thomas for
collecting key GPS points during the 2009 field season. Dr. Thomas Sever, currently at the
University of Alabama at Huntsville, provided expert remote sensing advice and shared
IKONOS images that were invaluable to the project. The late Dr. T. Patrick Culbert
analyzed our ceramic collections during the 2009 season and provided a historic and
philosophical connection with the earlier Penn Project at Tikal. We were honored to have
shared his final field season at Tikal (www.cambridge.org/Tikal; See S. Fig. 1).
We were fortunate to have had the field assistance of dozens of workers from the
communities of El Caoba, El Ramate, Capulinar, and Zocotzal, and we thank them for
their hard work, knowledge of the forests, and joyful spirits. Demetrio Crdova, Marcos
Caal, Santo Chico, Elmer Yat, and Jos Choc were particularly helpful
(www.cambridge.org/Tikal; See S. Fig. 2).
Identifications of herbarium specimens were assisted by Ron Liesner, Richard Abbot,
Charlotte Taylor, Henk van der Werf, and John Pruski of the Missouri Botanical Garden.
The suggestion to sequence DNA from our unknown herbarium specimens was by
Richard Abbot. Cassandra Gallagher, Katlyn Hahn, Amanda McGuire, Bradley Meyer,
Grace Morris, Blair Mynear, Nicole OConnell, Anne Schmidt, Alex Zumberger, Sarah
Shalloe, and Venicia Slotten assisted with processing, identifying, and imaging
paleoethnobotanical remains. Linnea Lentz and Victoria Lentz provided highly useful
assistance in the lab. Dr. Necati Kaval was of great assistance in the use of the
environmental scanning electron microscope. Trent Leslie provided some key assistance
in using R for data analysis.
We thank Kathleen Carr, Kenneth Hinkel, Kevin Raleigh, Jeffrey Marion, and Mary
Pohl for insightful editorial assistance on earlier drafts of this volume. Finally, we are
especially grateful for the editorial efforts of Lorelle Lentz, who has the uncanny ability to
turn verbal chaos into well-ordered prose.
 1 Tikal Land, Water, and Forest

An Introduction
Nicholas P. Dunning, David L. Lentz and Vernon L. Scarborough

The towering temples of Tikal have become icons of ancient Maya civilization, luring tens
of thousands of visitors annually to experience their jungle-shrouded mysteries. Thanks to
decades of careful archaeological investigations and epigraphic translation of ancient
texts, we now have a good grasp of the human history of this once-glorious Maya city,
especially as seen through the eyes of its rulers who commissioned its monumental art and
architecture. If one stands atop one of the citys pyramids today, only a few of the largest
temples and palaces stand out clear of the forest. The hundreds of more humble residential
complexes that sprawled across the hills and valleys are invisible. To the south the forest
spreads to a horizon marked by a range of conical hills, the Serrana Macanche,
overlooking the central Petn lakes beyond. To the north the forest blankets a rolling
landscape and a prominent escarpment marking the outer edge of the Mirador Plateau. To
the west and east the land drops away into deep, convoluted basins, or bajos, that hemmed
in the flanks of the ancient city. In September, this landscape is a lush green, often
steaming with recent rains. In March, the green has waned, the taller trees are barren of
leaves, and browns and grays mix with a dulling green. Such is the pulse of life in the
seasonal wet/dry forest of the Maya Lowlands.
For the ancient inhabitants of the interior of the Maya Lowlands, the annual oscillation
from wet to dry season was of critical importance. A delayed onset of rains, or a prolonged
or severe canicular drought, could play havoc with crop production as well as the
necessary annual harvest of water. For not only was the agricultural system that
supported Tikal dependent on rainfall, a lack of substantial and dependable surface water
sources made the collection of rainwater a requisite of urbanization and survival. Longer-
term droughts threatened the survival of the city and its inhabitants. Crops failed. The
productivity of forest resources waned. And the life-sustaining reservoirs would dry.
What did the rulers of ancient Tikal see when they looked across this landscape in AD
700? Was it a landscape denuded of forest, with fields stretching in countless succession?
Was it a mosaic of fields, orchards, and forest? Were soils degraded or well managed? Did
the citys reservoirs fill reliably and amply? Until recently, our ability to answer such
questions has been severely limited by a lack of data, and these were the queries driving
the impetus for the University of Cincinnati Archaeological Project at Tikal (UCAPT).
The role of environmental science in addressing aspects of complex society has
markedly changed over the last two decades. Archaeology has always been in partnership
with a suite of biophysical sciences in attempting to explain the behaviors of prehistoric
peoples, but with the study of cultures identified by developed writing systems and
complex iconographic imagery the scholarly emphasis was often redirected to a careful
assessment of what ancient city states deliberately and most tangibly said about
themselves. Too, the frequently spectacular investments made by the archaic state in
architectural monumentality immediately attracted the high-profile attention of both the
archaeologist and the museum director. Understood as a record of a narrow but influential
segment of society, these ancient elite and material drivers of social complexity dominated
our view of those initial and subsequent experiments in statecraft globally.
The Maya Lowlands of Central America have been especially affected by this research
trajectory with past generations of exploration coveting the often-serendipitous discovery
of a pyramid complex, palace structure, or set of Classic period stelae. The forces drawing
the archaeologist to those tombs/rooms with yes, wonderful things (Howard Carters
oft-quoted, perhaps apocryphal words upon opening Tutankhamens tomb) continue to
influence interpretations likely disproportionately affecting an assessment of the overall
social and economic development of ancient Maya culture. This is not to say that what is
now understood about the ancient Maya as derived from the hard-won contexts (indeed,
the decipherment of Maya writing is a remarkable intellectual accomplishment) is
anything but highly significant and relevant; that understanding is now, however,
markedly augmented by paleogeographers, biological ecologists, computational modelers,
and a suite of collaborative scientists.
One such project directed by a biologist, a geographer, and an anthropological
archaeologist (www.cambridge.org/Tikal; See S. Fig. 3) is the subject of this volume. The
UCAPT has been a multiyear focused program to capture aspects of the coupled
nature/human dynamic at one large site prominent in the Maya area from its early well-
established blossoming during the Late Preclassic period (400 BCAD 200) to its apogee
of florescence and regional control in the Late Classic period (A.D. 550800) to its great
fragmentation or collapse by A.D. 900 (Figure 1.1). The chapters presented cover the
methods and techniques employed in the interpretation of the quantity and quality of data
obtained. Each author was encouraged to analyze his or her data sets without regard to
preconceived outcomes and directed only by initial hypotheses, though once field and
laboratory assessments were established, significant attention was invested in correlating
patterns of information among researchers and data in conveying meaningful
interpretations.
Figure 1.1. Tikal chronological chart including period, Maya Long Count, modern
Gregorian calendar dates, and ceramic phases. Redrawn from Houston and Inomata
(2009).
Our work followed in the footsteps of a long procession of scholarly investigations of
Tikal. Of special merit for grounding our studies were the decades of work initiated and
developed by the University of Pennsylvania Museum of Archaeology and Anthropology
and their research investments at Tikal. Several scholarly tomes have been written on this
long-term project, and we were extremely fortunate to draw on those publications and
archived notes. We also benefited from subsequent work sponsored by the Instituto de
Antropologa e Historia de Guatemala, including a professional visit from Vilma Fialko
one of the principals responsible for excavation and mapping in the Lost World area of the
site in proximity to several water management features. Of the many sources made
available to our teams, perhaps the most significant for a project interested in the ancient
engineered landscape derived from a water, soil, and plant perspective has been the
published Map of the Ruins of Tikal, El Petn, Guatemala by Carr and Hazard (1961).
Representing the principal platform controlling nearly all of our synthetic sitewide
assessments, these maps were carefully scrutinized for accuracy, digitized, and
georeferenced, allowing us to incorporate our work into a GIS framework (see Carr et al.,
this volume). The monograph by Dennis Puleston (1983) further complemented our
locational efforts, particularly along his East Transect extending beyond the 16 km2 block
associated with the Penn maps and Central Tikal.
In addition to the impetus pushing the wave of new work in the Maya Lowlands that
emphasizes the role of ecological assessment from subsistence exploitation strategies to
climate change impacts, our project was hatched in the context of a series of empty coffee
cups lingering still on the University of Cincinnati campus and within our three respective
departments. For friends sheltered by an interdisciplinary umbrella and within immediate
physical proximity to one another, numerous interactions both informal and formal
allowed an uninterrupted flow of synergetic exchanges and activities among faculty and
students from the outset.
As developed in the NSF proposal (Grant BCS-0810118) supporting much of this
research, our goals were fourfold: to (1) assess the impact of Maya agroforestry practices
on the Tikal environs across its nearly 2000 year occupation history, (2) examine how
changes in water management adaptations effected and were affected by broader political-
economic changes in Maya society, (3) assess the importance of bajos (seasonal
wetlands) and their role in forest resource extraction and agricultural activity, and (4)
determine how the Tikal reservoirs, which potentially represented a carefully designed
water storage and hydraulic distribution system, may also have functioned as locations for
an array of civic-ceremonial activities (Lentz et al. 2009).
Previous research at Tikal and elsewhere in the Maya Lowlands helped frame these
questions. Archaeological research at Tikal spans more than a century. Systematic
explorations began in 1881 when Alfred Maudsley produced the first useful maps and
photographs of the site (Coe 1963). Other great Mayanists to work at the site were Teobert
Maler (1911), Sylvanus Morley (1938), Alfred Tozzer (1911), Edwin Shook (1951), and
Tatiana Proskouriakoff (1946). In 1956, the University of Pennsylvania launched a
multidecade Tikal Project, designed not only to restore and preserve one of the worlds
outstanding architectural monuments, but to ascertain more about the history of the site
and learn the reasons for its meteoric rise and catastrophic collapse. Most of the basic data
from those excavations appeared in a series of reports published by the University
Museum (Tikal Reports 1986; Culbert 1993; Puleston 1983; Haviland 1985; Coe 1990;
Jones 1996). Efforts to synthesize the Tikal data have been published in subsequent works
(e.g., Harrison 1999; Sabloff 2003). Research continued under the Proyecto Nacional
Tikal de Guatemala (Laporte 1987, 1993, 1995, 1997, 2003; Laporte and Fialko 1985),
whose efforts contributed significantly to our understanding of the prehistory of the site.
Recently, Webster and associates completed a study of the earthworks surrounding Tikal
(Webster et al. 2007). As the result of these earlier efforts Tikal is well known
archaeologically and extensively mapped in fine detail.
In stark contrast to the vast knowledge of the architectural features, settlement patterns,
ceramics, lithics, and other artifact assemblages from the site, much less is known about
the water and land management and ancient agroforestry practices at Tikal. Part of the
reason for this situation is historical: Archaeologists of the 1950s were less interested in
the environment and more interested in the durable aspects of material culture.
Accordingly, excavations were not organized to record evidence of ancient subsistence
and water use activities. While the detailed maps made by the Penn Project allowed for
initial modeling of water flow and collection at Tikal (Scarborough and Gallopin 1991), a
lack of published excavation data from the reservoirs limited understanding of how the
ancient water management system worked. A notable exception to this dearth of
information on ancient Maya water use activities in the area, however, was Fialkos work
(2000) in her survey of the Holmul River that discussed modifications of the rivers
channel for water control as it passes near Tikal.
Our project was designed partially to remedy this dearth of knowledge, aiming to
collect basic data on the present and past forests of Tikal and new field data on Tikals
elaborate water management system and on the history of land use and environmental
change at Tikal and surrounding areas. Before summarizing the work we set out to do, a
brief review of Tikals natural and historical setting is in order.

Environmental Context
Tikal is situated near the southern end of the Elevated Interior Region (EIR) of the Maya
Lowlands, a large, complex physiographic region that lies at the heart of the Yucatan
Peninsula (Figure 1.2) (Dunning et al. 2012). The EIR is composed of variably exposed
beds of Late Cretaceous and Lower Tertiary carbonate rock, uplifted and deformed over
many millions of years. Drainage within this karst region is almost exclusively internal,
and the permanent groundwater table typically lies many tens of meters below the surface,
essentially inaccessible to the ancient Maya except through infrequent deep cave systems.
The southern end of the EIR is composed of the Central Petn Karst Plateau and the Three
Rivers Region. Embedded within the karst plateau is a structural subsidence basin (the
Mirador Basin) the southern and eastern edges of which are marked by the Buenavista
Escarpment. This escarpment dominates the skyline northwest of Tikal.
Figure 1.2. Map of the Maya Lowlands Region showing the situation of Tikal in the
Elevated Interior Region and sites mentioned in the text.
The eastern side of the Central Petn Karst Plateau is fractured into a complex horst and
graben landscape dubbed the Three River Region (Dunning et al. 2003). The horst uplands
have weathered into rugged karst forms. The grabens are home to large, clay-floored
complex depressions with seasonal wetlands (bajos), including the sprawling Bajo de
Santa Fe, Bajo de la Joventud, Bajo de la Justa, Bajo de Azcar, and Bajo de los
Alacranes. These megabajos are interlinked by two drainage systems, the Rio Azul and
Rio Holmul, which also connect with a number of smaller, higher elevation depressions.
Two smaller, but still sizable bajos hem in the Tikal uplands on the southwest and west:
Bajo Socotzal (or Tzocotzal) and Bajo Ixtinto (Figure 1.3). Another sizable depression,
Bajo Bejucal, lies northwest of Tikal at the foot of a section of the Buenavista Escarpment.
Numerous smaller depressions, ranging in size from a few square kilometers to a few
hundred square meters (which we have dubbed pocket bajos), pockmark the karst
uplands, including within the central residential area of Tikal (e.g., Perdido Pocket Bajo
and Arroyo Corriental Pocket Bajo; Figure 1.4).
Figure 1.3. Map showing the situation of Tikal in a landscape of uplands and bajos.
Large square represents the boundaries of Tikal National Park. The small square is the
limit of the University of Pennsylvania site map.

Figure 1.4. Map of central Tikal showing the location of Perdido Pocket Bajo and
Arroyo Corriental Bajo.
Today, most of the bajos in the region are seasonal swamps, characterized by periods of
standing water during parts of the rainy season and edaphic desiccation during the dry
season. Hence, the swamp forest that occupies much of the land surface within the
regional bajo system is composed of woody and herbaceous species that are adapted to
severe soil moisture differences. Given the prevalence of bajos around Tikal, it is likely
that these features played a significant role in the history of regional settlement. Over the
past decade, evidence has accumulated indicating that bajos have experienced
considerable environmental change over the past several millennia likely the result of
both climatic and anthropogenic influences (Beach et al. 2003, 2008, 2009; Dunning et al.
2002, 2006; Hansen et al. 2002).
Soils within the region fall largely into four orders: Mollisols, Vertisols, Entisols, and
Histosols. The Mollisols are Rendolls (Rendzinas), including Typic, Lithic, Cumulic, and
Vertic subgroups. These are generally shallow, base-rich, clayey soils mantling hard
limestones and lying on generally well-drained upland terrain. The skeletal Lithic
Rendolls and even thinner Entisols occur on steeper slopes. Cumulic Rendolls occur along
the base of slopes and merge into Vertic Rendolls along the margins of bajos. Vertisols are
the dominant soil type observed in the bajos. These are deeper clay soils exhibiting
profound dry season cracking, and prevalent shrink-swell activity that results in a rugged,
hummocky soil surface. A few areas of Histosols (organic mucks) occur in places of
perennial soil moisture within some larger bajos. Histosols may have been more
widespread within the regions bajos at some time in the past (Dunning et al. 2006).
Within the Holocene, we now realize that climatic conditions across the region were far
from stable. Global and pan-Caribbean mesocycles of climatic fluctuation were felt across
the Maya Lowlands marked by cooler and drier periods followed by warmer and moister
conditions lasting hundreds of years (Brenner et al. 2002; Haug et al. 2003; Mueller et al.
2009). Shorter-term climatic cycles may also have significantly affected rainfall patterns
in the region, most notably a 208-year cycle of solar energy pulses (Hodell et al. 2001;
Wahl et al. 2006). Importantly for the Maya, such cycles may have increased the
frequency of drought in areas of the lowlands, with potentially devastating results on
Maya populations (Hodell et al. 2005; Gill 2000; Haug et al. 2003).
As elsewhere in the Maya Lowlands, the regions hybrid tropical/sub-tropical climate
generates severe disparities in rainfall (typically about 90 percent of the regions 1,500
2,000 mm of rainfall arrives during the MayDecember wet season) and in the availability
of water. Even during climatically stable (normal) times, however, the Maya Lowlands
are characterized by significant interannual variation in rainfall on the order or 3040
percent a fact that has played an important role in agricultural strategies over time
(Dunning and Beach 2004). Although drought risk and seasonal water shortages have
received the greatest attention in scholarly investigation of Maya cultural ecology, the
region is also frequently plagued by excessive and damaging precipitation, most notably in
the form of tropical storms and hurricanes, which unfortunately often arrive during the
harvest season with devastating consequences (Dunning and Houston 2011).
Today, lack of accessible water poses significant challenges for occupation during the
dry season. Although flow within the regions streams and rivers effectively ceases during
the dry season, deep pools of water remain in sections of the Rio Holmul. Bajos are also
the most common location for aguadas (ponds), particularly along their margins, where
defined by underlying bedrock fractures. These aguadas are likely to have originated as
small karst solution or collapse features that have partially infilled with clay sediment,
allowing them to retain water (Siemens 1978). Near Tikal many aguadas were clearly
subject to human modification to enhance their water-holding capacities. Aguadas are a
less frequent feature within the undulating uplands and smaller bajos surrounding Tikal.
While many of these aguadas may also have a natural, karst solution origin, others exhibit
evidence of strong human modification such as the construction of clay berms around their
peripheries. Some aguadas may be entirely artificial, probably originating as quarries that
were subsequently modified to retain more water.
Today, well-drained upland terrain is mantled by several subtypes of upland forest
(following Brokaw and Mallory 1993) that has been called variously climax forest
(Lundell 1937), deciduous seasonal forest (Wright et al. 1959), and montaa (a term
used widely in the Petn; Ford 1986). The canopy is typically 1530 m high with some
taller trees. Manilkara zapota (L.) P. Royen, Brosimum alicastrum Sw., Pouteria reticulata
(Engl.) Eyma, Drypetes brownie Standl., Sabal mauritiiformis (H. Karst.) Griseb. & H.
Wendl., Blomia prisca (Standl.) Lundell, Trichilia minutiflora Standl. and Pseudomedia
oxyphyllaria Donn. Sm. are among the dominant tree species (Schultze and Whitacre
1999). Considerable spatial variability exists in the species composition depending in part
on microtopographic situation. Areas with bajos that are subject to the greatest extremes in
soil moisture levels are characterized by scrub swamp forest (following Brokaw and
Mallory 1993). Other terms used for this forest assemblage include bajo forest (Lundell
1937), low marsh forest (Wright et al. 1959), and tintal bajo (Ford 1986). Trees in this
assemblage are typically 310 m in height, with Haematoxylum campechianum L. and
Croton billbergianus Mll. Arg. most prevalent; a thick ground cover of sedges, especially
sawgrass (Scleria bracteata Cav.), is often present. A large variety of transitional forest
types exist between the upland forest and scrub swamp forest extremes. Such transitional
forests typically resemble upland forest in their structure, but with a lower canopy and
species composition that also includes examples of scrub forest types. Often the midstory
is dominated by escoba palm (Cryosophila stauracantha [Heynh.] R. Evans), so that this
assemblage has been referred to as escobal transition zone (Lundell 1937) and escoba
bajo (Ford 1986; Kunen et al. 2000). Transitional forest types are typically characteristic
of pocket bajos and frequently located on the aprons of gently sloping lands with deep
colluvial soils found along the margins of many bajos both big and small.
Earlier research on wood and charcoal recovered in previous excavations at Tikal
indicates that its ancient inhabitants practiced some type of conservative forest
management, but that preferred woods may have become scarce during the citys Late
Classic population peak (Lentz and Hockaday 2009). More broadly, forest and land
resources clearly were under great pressure in the Late Classic throughout much of the
Elevated Interior Lowlands (Dunning et al. 2012; Turner and Sabloff 2012).

Historical Context
The pre-Hispanic human occupation of the Maya Lowlands can be divided into the
following chronological periods (noting that various scholars have put forward somewhat
different chronologies, sometimes preferring differing terminology such as Formative
rather than Preclassic, and with chronologies that vary among regions and sites):
PaleoIndian (Pre-7000 BC), Archaic (70002200 BC), Preclassic (2200 BCAD 250),
Early Preclassic (22001000 BC), Middle Preclassic (1000400 BC), Late Preclassic (400
BCAD 100), Terminal Preclassic (AD 100250), Classic (AD 250900), Early Classic
(AD 250600), Late Classic (AD 600770), Terminal Classic (AD 770900), Postclassic
(AD 9001500), Early Postclassic (AD 9001250), and Late Postclassic (AD 12501500).
While PaleoIndian, Archaic, and Early Preclassic materials and occupations have been
found widely distributed within the Maya Lowlands, no materials from these periods have
thus far been recovered at Tikal. Sometime prior to about 700 BC, small populations
began to reside at Tikal, and by 600 BC the first monumental architectural constructions
appeared, most notably in the area of the North Acropolis and in the Mundo Perdido
complex (Laporte 2003). By the Late Preclassic period (ca. 350 BC), Tikal had developed
into a significant entity in the emerging political landscape of the Maya Lowlands, though
comparatively small relative to some of the Preclassic giants of the Mirador Basin.
However, unlike larger centers in the nearby Mirador Basin, Tikal survived the turmoil of
the second century AD and emerged as a major center of the Early Classic period
(Dunning et al. 2014).
Tikals rise in the Early Classic apparently attracted the interest of the dominant
political economic power in Mesoamerica, Teotihuacan, and in AD 378 a Mexican lord
was placed on the throne of Tikal. The new Maya-Mexican regime prospered and Tikals
regional dominance increased (Martin and Grube 2008). Tikal experienced variable
fortunes during the Classic era, including a notable downturn in its historical trajectory or
hiatus in the sixth century AD, a phenomenon clearly associated with major military
defeats (Martin and Grube 2008), which may have been brought on in part by a
devastating volcanic eruption in AD 536 (Dahlin and Chase 2014). After the stagnation
and decline of the Hiatus, Tikals fortunes improved dramatically after AD 695. Between
695 and ca. AD 800, Tikal rulers engaged in an unprecedented building campaign,
including numerous expansions of the royal palace (Central Acropolis) and the erection of
six towering temple pyramids (Temples I thru VI) that have become the icons of this
ancient city (Martin and Grube 2008). Tikal was repeatedly victorious in military ventures,
most notably against archrival Calakmul, and was likely the dominant political entity in
the Petn in the eighth century AD. Tikals fortunes declined dramatically in the early
ninth century and by AD 900 the city was largely abandoned except a small residual
population that persisted for another two hundred or so years.

The University of Cincinnati Archaeological Project at

Tikal: 20092010
Over the course of the 2009 and 2010 field seasons, excavations, coring, mapping, and
collections were undertaken at numerous loci. Operation numbers were largely tied to
geographic locations as well as type of investigation.

Operation 1 and sub ops: Excavation, vegetation survey, and coring in and around
Corriental Reservoir.
Operation 2 and sub ops: Excavation in the Arroyo Corriental and Arroyo Corriental
 Pocket Bajo.
Operation 3: Coring in the Tikal Reservoir.
Operation 4: Coring in Aguada Pucte.
Operation 5: Excavation and coring in and around the Aguada de Terminos.
Operation 6: Excavation and coring in the Palace Reservoir.
Operation 7: Excavation and coring in the Temple Reservoir.
Operation 8: Excavation and coring in and around Perdido Reservoir and the Perdido
Pocket Bajo.
Operation 9: Mapping and excavation of residential units near Corriental Reservoir.
Operation 10: Excavation around Pulestons Peninsula on the East Brecha in the
Bajo de Santa Fe.
Operation 11: Coring in Aguada Vaca del Monte.
Operation 12: Excavation in Aguada Pital.
Operation 13: Mapping, vegetation survey, and excavation of a residential unit near
Perdido Reservoir.
Operation 14: Mapping, vegetation survey, and excavation in residential units
northeast of Corriental Reservoir.
Operation 15: Mapping, vegetation survey, and excavation in residential groups near
Aguada de Terminos.
Operation 16: Excavation in the Hidden Reservoir.

Field investigation of the entirety of Tikals water management system is a Herculean

task beyond the means of our project. We, therefore, targeted our investigations to sample
reservoirs in a variety of topographic and cultural settings. The linked system of three
elevated central precinct reservoirs (Temple, Palace, and Hidden) received extensive
attention, as did two midlevel tanks (Corriental and Perdido) in Tikals central residential
area (see chapter by Scarborough and Grazioso Sierra). Tikal Reservoir and Aguada Pital
received more cursory investigation. These reservoirs and others in central Tikal were
incorporated into a GIS database for the purpose of hydrologic modeling of the water
system (see chapter by Weaver et al.).
Several reservoirs in outlying groups associated with the Bajo de Santa Fe were
targeted, with the greatest attention given to the Aguada de Terminos, and lesser amounts
of time given to Aguada Pucte, Aguada de Elmer, and Aguada Vaca del Monte (see
Chapter 6 by Dunning et al.). Several reservoirs were also investigated from a
geoarchaeological perspective to examine the origins of their sediments and to gain a
better understanding of ancient land use in their respective watersheds (see chapter by
Tankersley et al.). Geoarchaeological investigations were also carried out in parts of the
Bajo de Santa Fe within Tikals sprawling urban residential zone, as well as in two much
smaller, more elevated pocket bajos (Perdido and Arroyo Corriental), in order to garner
paleoenvironmental data bearing on the natural history and human use of these
depressions (see chapter 6 by Dunning et al.).
To understand Maya forestry practices within and around Tikal three related projects
were undertaken. Several residential groups near Corriental Reservoir, Perdido Reservoir,
and Aguada de Terminos were systematically sampled via soil phosphate testing to detect
middens for subsequent excavations designed to recover paleoethnobotanical samples.
Recovered samples were analyzed and compared with Central American reference
collections housed at the Department of Biological Sciences at the University of
Cincinnati to develop a detailed picture of the forests in use by the ancient residents of
Tikal (see chapters by Thompson et al. and Lentz et al.). These data were compared with
extensive surveys of the present-day Tikal forests, pollen data from two aguadas in the
Bajo de Santa Fe and nearby Lake Petn Itza, and satellite imagery of the greater Tikal
area to assess the changing composition of the ancient forests and their biomass
productivity (see chapter by Lentz et al.). Additional ongoing work will examine the
genetic variability of the forest.
The results of our investigations are reported in detail in the following chapters.
Additionally, we sought the perspectives of two colleagues, David Webster and Timothy
Murtha, who have worked extensively at Tikal, to add their analyses and views on how the
ancient city and its inhabitants functioned and survived. Also, we were interested in the
prehistory of one of Tikals nearest neighbors and onetime rival, El Zotz. This site has
been the subject of recent multidisciplinary investigations (see chapter by Beach et al.)
and we found their insights to be illuminating vis--vis developments at ancient Tikal.
2 The Evolution of an Ancient Waterworks System
at Tikal
Vernon L. Scarborough and Liwy Grazioso Sierra

Maya archaeology has undergone another revolution in theory and methods over the last
two decades. This revolution stems from a renewed and accelerated emphasis on human
ecology, one with historically significant antecedents. These sophisticated scientific
models are now guiding much of the new research on the ancient Maya as a greater
cultural group (Scarborough 2010). During the 1970s, 1980s, and into the 1990s, the
greatest breakthroughs in Mesoamerican archaeology were in epigraphy, with the
decipherment of perhaps 80 percent of the known hieroglyphic texts describing the ancient
Maya, a worldview reclaimed by a highly skilled and innovative group of reflective
scholars (Coe 1992). The emphasis of the ancient scribes was to record elites assessments
of the world they occupied and the etiquette they used in coping with it. This perspective
has changed our interpretations of how the 1 percent of Maya society was organized and
viewed itself, but it has revealed comparatively little about the lives of the nonelite and the
economic scaffolding supporting statecraft. Although the corpus of information and
knowledge about the highly tiered societal elite and their ideological orientations
continues to grow (Jackson 2013), a next generation of scholarly activity is emerging from
a formidable community of ecologically oriented researchers working in a milieu of
interdisciplinarity. Although this volume represents only one urban center, it captures in
microcosm the many regional projects now working in the Maya Lowlands, each drawing
on shared findings in attempting to assess best the dynamic human-nature couplings of the
past (cf. Chase and Scarborough 2014). This chapter details the importance of water
management at the grand site of Tikal.

The Water System at Tikal

Water management has a long and influential history in identifying the kinds and degrees
of social complexity in ancient contexts (Scarborough 2003). Our work introduces the
functional components that we have documented and how these joined to produce the
hydraulic system at the hilltop center of Tikal Central Tikal or its urban core
(Scarborough et al. 2012) (Figure 2.1). Here we emphasize how the water system
developed over a 1,000-year period and the self-organizing processes that evolved in a
tropical rain forest.
Figure 2.1. Map showing the main catchments and associated reservoirs at Tikal. Our
work examined five reservoirs in the southern half of Central Tikal zone Temple, Palace,
Hidden, Corriental, and Perdido Reservoirs. The blue lines identify prominent arroyos.
Reprinted with permission from AAAS.
As long ago as 1991 Scarborough and Gallopin revealed aspects of this complex system
a study that continues to aid our current assessments even though that work was based
solely on archival sources and the principal mapping efforts of Carr and Hazard (1961).
Our recent mapping, soil coring, and formal excavation program has significantly
expanded that study and identified several previously unknown elements necessary for a
comprehensive understanding of the system (Scarborough et al. 2012). Important findings
include (1) the largest ancient dam now identified in the Maya area of Central America
and the posited manner by which reservoir waters were released from the dam; (2) the
construction and use of a cofferdam for dredging/cleaning the largest central precinct
reservoir at Tikal; (3) ancient seeps/springs that spurred the initial colonization of the
ridge on which Tikal was built; (4) the construction of pollution-control features allowing
surface runoff from paved surfaces to pass through sand filtration boxes before entering
and permitting a potable reservoir supply; (5) a switching station that accommodated
seasonal reservoir-water infilling and subsequent release during the driest periods of the
year; and (6) the deepest well-defined canal segment cut into limestone bedrock in the
Maya Lowlands. Our research was designed to integrate these various engineering
achievements and identify the system that sustained the urban complex of Tikal through
deep time. By assessing the adaptive behaviors of our distant human ancestors and their
engineered landscapes, we can better inform our own suite of environmental options and
degrees of resilience (Chase and Scarborough 2014).
In this chapter we identify the component parts of the water system and their
articulation, the interplay between the coevolving built environment and human society,
and assess links to external factors such as climate change or population growth (both
inclusive of and separate from deliberate or conscious societal decision making). The self-
organizing principles shaping the growing literature around resilience theory and the
rate and process by which societies cycle through deranging periods of transformation
implicitly guide the chapter (Scarborough and Burnside 2010). The role of niche
construction a term borrowed from ecology (not unlike resilience) also proves
useful for understanding the development of the engineered landscapes (Kendal et al.
2011; see Wilkinson et al. 2012; Wilkinson 2014). For example, evidence suggests that
upon colonization of the uptown ridgetop center during the Middle to Late Preclassic
period (600 BCAD 150), in part made possible because of the aforementioned
seep/spring activity, productive agricultural activity accommodated significant population
growth trajectories and monumental construction programs in the urban core (see Webster
and Murtha, this volume). Nevertheless, such accomplishment meant that the slow
recharge of the natural seeps now provided an inadequate water supply for the growing
demand. The clever niche construction adjustment made was by way of additional
pavements in the site center, which when canted and directed to open quarries within
drainage ways the latter the building material for the civic architectural efforts
permitted the containment of seasonal surface runoff. Although the recharge surfaces for
the seeps were sealed by these pavements and the natural filtering of percolating
subterranean water sources curtailed, vast amounts of precipitation could now be directly
accessed (Scarborough and Gallopin 1991). By locating banks of sand (a resource not
naturally found in the Tikal uplands) into filtration boxes at the ingresses of reservoirs,
the urbanites introduced a human-made cleansing system to provide adequate potable
water supplies to this next phase of the engineered landscape and increased population. To
explain this phase transition further, data from the paleoclimatological community
suggest a droughtlike period affecting the Maya Lowlands toward the end of the Late
Preclassic period (e.g., Dunning et al. 2013; Mueller et al. 2009). Here we see the
adaptability and resilience of the lowland Maya. Nevertheless, a path dependency and a
vulnerability to a similar yet more severe set of drought variables some seven hundred
years later are shown to bring their pivotal experiment in tropical low-density urbanism to
a near end (Gunn et al. 2007; Haug et al. 2003; Hodell et al. 1995; Hodell and Guiderson
2001; Kennett et al. 2012; Medina-Elizalde and Rohling 2012; Yaeger and Hodell 2008).1

The Parts
The Central Tikal water system was defined by several interdependent landscape
components highly associated with the seasonal rainfall regime. The Late Classic (AD
650850) system consisted of three reservoir types: (1) large summit-ridge reservoirs, (2)
bajo-margin tanks, and (3) small residential reservoirs (Figure 2.2). All of these features
worked together to provide Tikal with the concentrations of water necessary to sustain the
urban center year round (Scarborough et al. 2012, online support information figure S1)
(Figure 2.2).
Figure 2.2. Scaled cross-sectional schematic of the southern half of the Tikal ridgetop,
with reservoirs positioned by elevation and dated sedimentation histories (abstracted by C.
Carr). Reprinted with permission from PNAS (see Scarborough et al. 2012, online support
information tables S1 and S2).

Summit-Ridge Reservoirs
Within the central precinct of the site area, six sizable reservoirs were built by damming
incised natural arroyos generally descending west to east and perpendicular to the ridgetop
on which most of the monumental architecture at the center was situated (Scarborough and
Gallopin 1991). These reservoirs were formed by positioning a series of causeways or
dams across their longitudinal axes as well as excavating downward and expanding their
widths to accommodate a more generous water supply; the quarried stone was used to
construct some of the most towering pyramids and elaborate acropolises in the Maya
world. Although we now have more accurate and enlarged capacity figures for several of
the reservoirs associated with the southern half of the center (Scarborough et al. 2012),
preliminary estimates derived from Scarborough and Gallopin (1991) for all of the
summit-ridge reservoirs suggest a volume of 105,108243,711 m3. Given the abundance
of seasonal rainfall, individual households are assumed to have accessed adequate and
immediate water supplies during eight months of the year. However, during the annual
four months of droughtlike conditions, or approximately 120 days, reservoir access was
crucial. If the overall consumption requirement of the greater population at Tikal was in
the neighborhood of 15 liters/day, or the average consumption rate of a Cambodian citizen
occupying a similar environment today (United Nations Development Programme 2006),
or a figure of 17 liters/day from an ethnoarchaeological example derived from the
somewhat more arid present-day community of Xculoc, Campeche (Becquelin et al.
1994), and the lower estimate for the capacity of the summit-ridge tanks is used, a
population estimate of those drawing from just these reservoirs is about 50,000 people. If
we further cut that figure in half to accommodate any number of deranging factors
(leakage, evaporation, plaster production, elaborated bathing needs, and agricultural
consumption in the urban core), a figure of 25,000 remains. Nevertheless, if we were
accept Havilands (1970) estimate of 11,000 people within the 16 km2 Central Tikal zone,
then the waterworks were likely overbuilt to withstand infrequent but heavy storm runoff
as well as to accommodate enough additional water for nonlocal, seasonal work gangs
during peaks in construction activity (to say nothing of pilgrims and related visitors).
Some might speculate that the Maya wildly overbuilt; they simply did not need to equip
the Central Tikal site area with such a sophisticated system of dams, reservoirs, sluices,
canal segments, and even a filtration system (cf. Weaver et al., this volume). Several
papers during the late 1970s suggested that the great building boom of the Late Classic
was stimulated by an elite need to redirect resources toward monumental construction
projects designed to show political and ideological control and competence during a
period of ecological failings and population overshoot (Culbert 1977; Hosler et al. 1977;
Sharer 1977). The elaborate summit-ridge reservoirs may have been a partial reaction to
these building efforts, given the quarried fill necessary to erect the noteworthy edifices.
Nevertheless, to underestimate the labor and forethought necessary in building and
expanding the waterworks put to functional ends is probably not advisable either. Given
their engineering acumen, effort, and time, the system was constructed to accommodate a
projected population density in a deliberate and accomplished manner. The filtration
system alone speaks legions about planning and maintenance, and the need for potable
supplies year round.
At another interpretive level, reservoirs that are recharged by immediate, relatively
small, and closed microwatersheds like those at Tikal (Scarborough 1993, 1994) can be
used as indices of past rainfall regimes. If Weaver et al. (this volume) can demonstrate that
(1) the tanks at Tikal were not filled to capacity for a portion of the year on the basis of
new runoff modeling for present-day Tikal, and (2) the Maya did build their reservoirs to
hold the maximum capacity of water then available, it seems likely that rainfall rates at the
time of construction and principal use were higher than those rates today. Because we
have only proxies for the actual amount of precipitation in the Maya area in ancient times
(Kennett et al. 2012), with our indices based on a normal identified by current rainfall
rates, even approximated precipitation amounts can only be estimated in generalized terms
as acute drought, droughtlike, or something less so. Given the potential for reservoir
studies in this regard, tanks may well represent our best measure of actual rainfall over a
closed, specific catchment and therefore permitting an assessment of precipitation in a
manner not unlike a rain gauge functions. Reservoirs like those at Tikal and throughout
the Maya area away from riverine recharge may well provide quantitative controls over
the actual severity of drought conditions less fixed to relative measures tied to present-day
figures.

Dams
Our recent work (Scarborough et al. 2012) indicates that sizable earthen and stone-fill
dams were positioned to span the major southern arroyo cutting down from the ridgetop at
Tikal. These dams expectedly functioned as grand walkways straddling the ravines and
adding to an elaborate causeway system triangulating at the summit of uptown Tikal
(Figure 2.1). Our excavations focused on the largest of the dams Palace Dam resulting
in a tentative interpretation of the features control and release of waters (Figures. 2.3, 2.4,
and 2.5). Several scenarios were investigated in attempting to understand the manner by
which water was stored and subsequently released. The Late Classic feature was a
substantial gravity dam rising 10 m, spanning 80 m, and as much as 60 m at its basal
width (Figure 2.1; Scarborough et al. 2012), though the end points of the dams length
were likely the original, steeply incised arroyo banks (that is, pediments of bedrock)
before the sidewall quarrying of the ravine resulted in the final width and volume of the
Palace Reservoir (www.cambridge.org/Tikal; See S. Fig. 4) (Scarborough et al. 2012,
online support information figure S4).

Figure 2.3. Map of the Temple, Palace, and Hidden Reservoir chain with the ancient
arroyo course and associated excavations. Temple Reservoir seep/spring and subsequent
cofferdam are noted as well as the prominence of the Palace Dam (abstracted by C. Carr).
Reprinted with permission from PNAS.

Figure 2.4. Cross-sectional composite of Palace Dam profiles for operations 6L, 6Q,
6W, 6U, and 6V (prepared by E. Weaver). Reprinted with permission from PNAS.
Figure 2.5. Artists reconstruction of Palace Dam with posited vertically stacked sluices
(prepared by R. Weaver). Reprinted with permission from PNAS and R. Weaver.
Scenario I: Given the pressure on the interior wall of the dam, the waters held back may
not have been sluiced through the feature at all, rather simply contained (Harrison 2012).
In that event, the collected waters for the numerous citizens of Tikal would have been
seasonally fouled, compounded by the additive effects of repeated years of polluting
sediments washed into the tank. Frequent dredging would have been necessary whether or
not the feature was properly sluiced through the dam; but assuming it was not routed
through the dam, a further concern would have been causeway-height overflow during
periods of extreme seasonal rainfall events. Nearly four times the precipitation necessary
to fill each of the five or six reservoirs at the summit ridge of uptown Tikal was estimated
to flow into the reservoir system for about eight months by Scarborough and Gallopin
(1991; Gallopin 1990). However, Weaver and colleagues (this volume) now suggest that
the catchments were not as paved and impervious as originally thought, significantly
reducing the seasonally infilling waters. Nevertheless, waters would have occasionally
crested and breached the Palace Reservoir Dam or Causeway, closing off a principal
traffic access between the central and southern portions of the city and necessitating
considerable erosional repair. No spillway is discernible. Although the Palace Reservoir
had been clearly dredged before the near-total abandonment at Tikal the only sampled
tank to show such complete sediment removal2 it is difficult to imagine how the
reservoir could have been cleaned without drawing the water levels down incrementally
and in a controlled release. Given the sophistication of Maya waterworks at
contemporaneous Palenque (French and Duffy 2010; French et al. 2012) and the
maximum volume capacity difficulties associated with an estimated 38,000 m3 of water
into the Palace Reservoir (Gallopin 1990) immediately after a major storm, simple bucket
extraction seems an impractical solution.
Scenario II: Another possible means for extracting water from the Palace Reservoir
given the formidable scale of the dam was the implementation of the cistern sluice made
famous by the ancient Sinhalese as early as the third century AD (Gunawardana 1978;
Scarborough 2003). To contain the water pressure against an open sluice in the sidewall of
the dam, a retractable, watertight vertical shaft (possibly nested pots with cutout bottoms)
always positioned at the waterline was connected to an underground horizontal tunnel
that directed a controlled release below the dam for issuance outside and below the
feature. Our trenching and coring efforts, however, were unsuccessful in finding evidence
for such a feature, though we excavated at locations understood to be the lowest points in
the reservoir floor at or near the foot of the dam.
 Scenario III: A third possibility is that the Maya had developed materials strong enough
to accommodate several sizable sluice gates near the base of the dam. Although we
searched for such gates, especially at that same basal interface between the reservoirs
lowest channelized segments noted in Scenario II near the original V-shaped base of the
ancient arroyo and along the skirt of the dam, we were unable to identify a sluice gate.
Although hardwoods or perhaps marblelike limestone could be shaped to cover sizable
sluice openings, the difficulties in sliding covers back and forth at a basal depth with a
near-full volume of water pressure above would have proved impractical, if not
impossible.3

Explanation
Our dam trenching efforts did reveal another possible technology employed by the ancient
Maya, a technology first reported in the Old World by Mongol Islamists in AD 1280 at the
Saveh gravity dam site, 160 km south of Tehran (Vogel 2004). Dating to the eighth or
ninth century, the posited Palace Dam sluice system would have consisted of several
vertically stacked orifices (Figures 2.4 and 2.5), each perhaps 30 cm in diameter,
positioned in rows across the interior face of the dam. Our excavations exposed a poorly
preserved sluicelike opening associated with fragments of tabular limestone and packed in
bajo clay about midway up the dam wall (Scarborough et al. 2012, online support
information figure S6). Although no other similar features were identified in our 2-m-wide
trench, discrete pockets of bajo clay a highly effective sealant were noted in at least
one other case associated with the interior wall of the dam. Given the pressures against the
Palace Dam, especially at the height of the rainy season, we suggest that water was
released through a set of diminutive sluices located at the dropping waterline. As water
levels descended, successively lower-level sluices were opened, eliminating the need for
the highly pressurized release from one or two large sluice gates located near the base of
the dam and only allowed in modern dams with reinforced concrete and tempered steel.
Just the simple removal and replacement of a sluice gate cover at the bottom of the dam
face, as noted, would have required sophisticated pulley systems, to say nothing about the
size and plate strength of the sluice gate covers.
Although several dams were probed by our efforts at Tikal, no others were as large or
sophisticated as the Palace Dam. We suspect that the technology we suggest was not
unique to Tikal and that other large dams elsewhere in the Maya Lowlands, as well as the
comparable sized dam at Xoxocotln, Oaxaca (OBrien et al. 1982), and the huge Purrn
Dam of the Tehuacn Valley, Puebla (Aiuvalasit et al. 2010), were of similar technological
complexity.

Cofferdam System
Immediately to the west of the Palace Reservoir at the original head end of the natural
arroyo is the Late Classic Temple Reservoir (Figure 2.3). Relatively small compared to the
volume of its lower complement, it was also dammed by the Temple Causeway before
water was released into the Palace Reservoir. Given the chain of reservoirs descending
along the path of the ancient arroyo (Temple-Palace-Hidden), it was practical to use the
uppermost Temple Reservoir as a holding tank for cleaning the larger Palace Reservoir.
We do not know the frequency of its use when put to this end, but it is clear that the Palace
Reservoir was dredged, as evidenced by the relatively slight amount of sediment contained
in the tank during our excavations about a meter at the most basal reaches, even near the
foot of the Palace Dam.4
The main Temple Reservoir was likely constructed in the Late Classic period, though
the southern silting tank was constructed earlier at the original head end of the ancient
arroyo (Scarborough et al. 2012, online support information figure S3). Given that
Harrison (1999) indicates that Temple III dates to AD 810 and is in immediate proximity
to and oriented toward Temple Reservoir, it is likely that the fill for its construction was
taken from the tank depression, as suggested by three of the four Accelerator Mass
Spectometry (AMS) dates retrieved from sediments in the main Temple Reservoir tank
[the one intrusive Late Preclassic date likely washed from buried plaza surfaces associated
with the initiation of the nearby North Acropolis (Coe 1990), Lost World complex
(Laporte and Fialko 1985) and an external ramp flanking the west side of the reservoir
(Fialko 1988)] (Figures 2.1 and 2.2).
The literature contains limited information concerning the identification of cofferdam
construction elsewhere in Mesoamerica, but a sizable feature of this purported function
was found immediately above the huge Purrn Dam, near Tehuacn, Mexico (Woodbury
and Neely 1972). An Early Classic date was originally assigned to the construction of the
major dam, though recent reassessments now suggest a Late Preclassic building date
(Aiuvalasit et al. 2010).
Although the major construction and more intensive modifications to the Tikal
reservoirs occurred during the Late Classic, because of the significant presence of
Preclassic architecture close to the reservoirs and the existence of spring activity the
likely rationale for the location of the summit-ridge reservoirs we suggest that several of
the main reservoirs were at least initiated by the Late Preclassic (Figure 2.6). The
canalized course at the bottom of Palace Reservoir and the way in which its bedrock sides
were worked suggest landscape alterations very early. The excavation and shaping of these
reservoirs must have taken place at the same time as the first standing architectural stages
in the area. A Late Preclassic date as the starting point for these main reservoirs is
consistent with several of the hydraulic features detected in the other reservoir settings
tested farther downslope.
Figure 2.6. Map of swathe of Late Preclassic construction wrapping around the head
end of the subsequent TemplePalaceHidden Reservoir chain in uptown Tikal.

Colonizing Springs
Of special importance was the discovery of an out-flowing spring at 1.6 m below the
surface, with dated Mid- to Late Preclassicsediments at 521216 BC (Scarborough et al.
2012, online support information S3; see Johnstone 2004 for other examples elsewhere in
the Maya Lowlands) (Figures 2.2 and 2.3). Given the difficulties in accessing potable
water for the first colonists to ridgetop Tikal, headwater spring activity potentially
permitted early Late Preclassic populations year-round sustenance (see Figure 2.2). The
spring(s) was located in what was later to become a silting tank for the main Temple
Reservoir, a deliberately excavated basin first constructed to contain spring waters. By
way of illustration, at the archaeological site of Naranjo today (40 km east), the entire
water supply for well more than one hundred workmen during the four-month-long dry
season is provided by naturally filtered, head-end arroyo spring activity (personal
observation and V. Fialko, personal com 2010; Figure 2.7).
Figure 2.7. Analogous ravine spring setting at Naranjo today: (a) photograph of
containment dam and (b) photograph of surface welling waters near dam (note plastic
tubing in and around pool drawing potable water by simple gravity flow).
The role of wetlands associated with origin myths in Mesoamerica Place of Reeds
has received considerable attention (Sharer and Traxler 2006). Stuart (2000) suggests that
Teotihuacan was home to this myth with respect to Copns association with that colossus,
but given Tikals scale and considerable linkages to Teotihuacan the latter a site with a
hundred springs located more than 10 km from the nearest lake margin (Sanders, Parsons,
and Santley 1979; Scarborough 2003) such a myth may have been seated at Tikal as
well, and perhaps other early centers in the Petn (Garrison and Dunning 2009).
Colonization of a hillock some 25 km from the large lake system of Petn Itz was only
made feasible by such elevated spring waters. The subsequent enlargement and damming
of seasonally flowing ravines came later, but again emphasizing the attraction to the
hillock on which Tikal was positioned. A recurrent query is why hills and ridges away
from large volumes of naturally available water were the seats for great urban experiments
in the southern Maya Lowlands (Isendahl et al. in press). It is not until after the ninth
century collapse of the great southern lowland cities that sites like Tayasal and Topoxte
were located on the Petn lakes (see Sharer and Traxler 2006).5

Filtering System
Identified within several of the reservoirs excavated and cored inclusive of Corriental,
Temple, and Palace Reservoirs were fine lenses of sand alternating with clay deposits.
Subsequent laboratory assessment demonstrated that the sands analyzed from Corriental
Reservoir were mined as far away as 35 km northeast of the site area and associated with
unique quartz sand grains (Scarborough et al. 2012, online support information figure S7).
Our reconstructions strongly indicate that such sands are a consequence of deliberate
placement at the ingress gates to these tanks and that severe precipitation episodes such as
tropical storms washed away filtering devices containing these otherwise water-purifying
sands. Perdido Reservoir, a small formal tank put to agricultural use and located at the
southwest foot-slope margins of Central Tikal, has more limited evidence for sand lensing,
which is an expected result of a reservoir that does not provide potable waters.6 We have
struggled to explain the presence of sand at Tikal as it does not occur naturally on or
immediately near the ridge on which the uptown site is located. Sand is not a component
in Maya plaster preparation (Villaseor 2010).
 Although the Romans are known to have occasionally used sand filters in concert with
charcoal (Hodges 2002: 275), Western Europe did not use them until the late 1800s.
Cholera and typhoid are known to have spread throughout London by way of sump pits
contaminating well waters (Johnson 2006; McNeill 1976), with the late development of
filtration systems following only then. Other archaic states invested in water purity earlier
than the ancient Maya, but the Maya system remains early and unique in the tropical
world. Surely boiling or the addition of fermenting fruit further accommodated their
populations (Dahlin and Litzinger 1986). Nevertheless, the waterborne disease load was
elevated for most individuals, though these limited precautions/improvements contributed
to well-being at least to a degree.7

Switching Station and Canal Construction (Corriental

Reservoir)
Carr and Hazard (1961) introduce the notion of a seasonally controlled sluice gate
conjectured on the south side of Corriental Reservoir (Figures 2.1 and 2.8), one of four
bajo-margin tanks at the foot of the Tikal ridge and positioned loosely in the cardinal
directions. Our recent surveying, excavation, and coring program invested heavily in both
Perdido and Corriental Reservoirs (Figure 2.1). Elevational controls linking the recovery
of stratigraphic profiles (fence diagrams) reveal the infilling processes (Corrado 2014;
Nagy 2012; Scarborough et al. 2012; Tankersley et al. 2011: Figures 2.3 and 2.4) with
various periods of occupation and use at higher flanking catchment locations draining into
these tanks. Generally, bajo-margin reservoirs were subject to relatively little dredging at
Tikal, likely a testament to managed land use by way of erosion controls and posited
silting tanks, the latter best identified immediately south of Temple Reservoir. Given the
near-absence of well-defined terracing to curb slope-wash loss, it is likely that numerous
economic and shade trees stabilized much of the terrain (Lentz et al., this volume).
Figure 2.8. Map of Corriental Reservoir coring and excavation locations. The ingress
gates are identified as well as the switching station (abstracted by C. Carr). Reprinted
with permission from PNAS.
The East Gate at Corriental Reservoir was likely constructed later in the Late Preclassic
period given superposition and dated sediments in association from elsewhere in the
reservoir (Tankersley et al. 2011) (Figures 2.2 and 2.8). A steep-sided, 2-m deep V-shaped
canal segment was dug into the elevated limestone bedrock leading into this break in the
circumscribing Corriental Reservoir berm (Scarborough et al. 2012, figure 5C). Although
the berm was not in place until AD 400570, flow into the tank at the East Gate location is
suggested by an initial Late Preclassic date. Excavation within the original control gate
area indicated that a meter drop in elevation sloped into the bottom center of the tank
about 45 m horizontal to the west. Because a second primary catchment carried waters
into the basin from a Northwest Gate, it is posited that an expedient dam was erected to
contain additional waters when the margins of the East Gate approached cresting levels.
Once the tank filled to some unknown height, it could be drawn down by a slow but
deliberate breaching of the seasonally constructed dam; the tanks relative volume at this
time was slight.
Given that several climatic models indicate that the end of the Late Preclassic was a
droughtlike period throughout much of the Maya Lowlands (e.g., Dunning et al. 2014), the
construction of a water-augmenting, switching station and the efforts made to contain as
much runoff as possible in a seasonally released holding tank system follow. Nevertheless,
by the Late Classic period rainfall appears to have increased regionally and the tank
system dynamics to have been modified accordingly. Charcoal flecks recovered from a
berm core taken from the interface between the earliest introduced berm matrices and the
original paleosol indicate the late Early Classic construction date noted previously (Figure
2.2). Although there is little evidence for major dredging of the tank, slight (though
continuous) sedimentation into the tank occurred. Slope erosion was less significant, as
noted previously, but over several centuries even slight deposition raised the reservoir
floor with impervious fine clays. Rather than dredge, the Maya elected to scrape soil and
stone outside the reservoir in producing a berm that probably augmented capacity, but
appears to have acted as much to shunt away more abundant Classic period runoff waters.
This observation is further supported by the careful infilling of the V-shaped canal
segment created earlier in the Late Preclassic. In this case, the canal segment was filled
and sealed with a masonry concretelike matrix equivalent to that in thick load-bearing
walls and plaza floors of the central precinct. The catchment channel that had been
carefully cut into bedrock and directed into the reservoir was deliberately decommissioned
(Scarborough et al. 2012). When combined with the shunting effects of the upslope berm
construction wrapping around the reservoir and dating to the onset of the Late Classic, it is
apparent that flooding rains became more of a concern than their scarcity.
During the Late Classic period, a more permanent dam was located at the present East
Gate opening to contain seasonal runoff primarily from the Northwest Gate ingress and its
sizable catchment area (Scarborough et al. 2012). Inflow from the South Gate is also
apparent from an evaluation of sediment fractions carried into the center of the reservoir
from that ingress and identified by an alluvial apron revealed by our subsurface cores at
this time (Nagy 2012). Carr and Hazard (1961) do suggest that the South Gate was a
switching station during the Late Classic period and propose that sizable construction
blocks lining the margins of the egress and the downstream reaches of the East Gate
outflow are the ruins of a permanent Late Classic plug closing that side of the reservoir.
The switching station to the East may have been switched to the South.

Perdido and Other Tanks

Perdido is the second bajo-margin tank identified on the southern half of Tikal that was
examined extensively (Figures 2.1, 2.2, and 2.9). One of the smallest of the formal tanks
identified, it may have functioned rather exclusively as a source of agricultural irrigation
waters during the dry season and an aid to control field flooding in the wet. As inferred by
the artifact debris recovered from our excavations, several times more artifact waste was
found to have washed into Perdido when compared to others. Only one ingress gate and
one egress gate were identified, and much of the circumscribing berm was ill defined. The
shallow depth of the plaster flooring (1.5 m below the surface Op. 8A) was overlain with
several layers of continuously bedded sediment suggestive of limited dredging, though
perhaps a natural flushing of debris occurred at the outset of the rainy season. A soil pit
located 100 m south and 30 cm below the approximate bottom of the southern egress (Op.
8D), as well as other investigations in the Perdido Pocket Bajo, suggest that bajo-margin
fields were likely irrigated from the reservoir (Dunning et al., this volume; Scarborough et
al. 2012, online support information S2B).
Figure 2.9. Map of Perdido Reservoir coring and excavation locations (abstracted by C.
Carr).
Although the Tikal Reservoir (Figure 2.1) was probed using a Livingston piston coring
device, the tank had been so compromised by historical camp preparations (Shook 1962)
that the only date we were able to obtain was that associated with pre-Maya occupation
14301260 BC (Figure 2.2) (Scarborough et al. 2012, online support information S2B).
The Tikal Reservoir was the largest of all bajo-margin reservoirs and the terminus of
waters released from the TemplePalaceHidden Reservoir chain and the ancient natural
arroyo. The early date likely reflects a paleosol associated with an initial base landscape,
with subsequent modifications by the ancient Maya grander and at least an order of
magnitude more expansive than that identified at the agricultural tank to the southwest of
the center Perdido. Agricultural plots are inferred to have been a significant focus for
this eastern flank of the site area.
Hidden Reservoir is a poorly understood basin on the downslope side of the Palace
Dam (Figure 2.1). Our probing was limited to three excavation units revealing a
preponderance of Late Classic sherds (Figure 2.2). The Mendez Causeway defines the
containment dam for this shallow basin, but the basin may not have been a significant
holding tank given its limited height; it appears to have acted more as a check dam used
in regulating the velocity and amount of release from the major Palace Dam construction
before cascading precipitously toward the Tikal Reservoir. Waters flowing downhill from
Hidden are likely to have acquired sediment and organic waste. As noted, the Tikal
Reservoir is viewed primarily as an agricultural tank.
Like Temple Reservoir, the Inscriptions Reservoir appears immediately associated with
one of the six major pyramid temples Temple VI or inscriptions at Tikal (Figure 2.1).
Our investigations were limited to two cores placed near the center of the basin, which
revealed four pre-Maya AMS dates suggesting the feature may have been a natural sink
that was subsequently used by the Maya as a tank (Figure 2.2) (Scarborough et al. 2012,
online support information S2B).8 The possibility does exist that an Early Preclassic
village may have colonized the setting (Harrison 1999: 161).
Madeira Reservoir was the subject of a portion of a dissertation (Iglesias 1987), with
ceramic dating suggesting Late Preclassic through Late Classic usage. Our teams did not
reexamine the basin as it is likely a residential reservoir recharged, in part, from the Lost
World Pyramid courtyard spaces above it. Given the jade and related exotics retrieved
from those excavations, it is likely that this particular tank had ritual significance.

Niche Construction
Recent attention to the biological concept of niche construction has captured the attention
of anthropological archaeologists (OBrien and Laland 2012; Wilkinson et al. 2012). Not
unlike the notion of landesque capital (Brookfield 1984), the concept emphasizes
deliberate modifications to the landscape, but also the many unintended consequences that
can both enhance and degrade a built environment. An urbanscape is an especially
challenging engineered landscape because of the frequency and monumentality of its
alterations. Not only does a heavily affected environment demand a constant reset
adaptation frequently associated with seasonal fluctuations in precipitation, but it is also
influenced by the vagaries of human agency both at the kingly level of decision making
and by quotidian activities. Our charge is to improve understanding of these significant
human-nature interactions through time in Central Tikal.

Pre-Maya
The summit area on which Tikal was positioned provides very little evidence of pre-Maya
antiquity. Two dates from the ancient arroyo stream sediments within the Palace Reservoir
area indicate that enough particulate charcoal was in the air to suggest very early burning.
One of these dates was associated with a stratigraphic sediment inversion in the ancient
arroyo channel, but the other was from a sealed context under the Palace Dam wall both
dates span a period from about 1800 to 1600 BC (Figure 2.2). It is possible that this
charcoal deposition was human induced, but given the absence of artifacts, the charcoal
may as likely have been associated with natural forest ignitions. Earlier dates deeply
buried farther downslope in the bajo-margin tanks indicate the amount of earthmoving and
erosion that have occurred at the summit heights at Tikal since colonization.
Several late Pleistocene and early Holocene dates were recovered at basal levels in both
our cores and formal excavations well below the 45-m climb to the summit of uptown
Tikal. Our oldest dates are from pretank deposits underlying much of Perdido Reservoir
(Figure 2.2). Three dates in excess of 15,000 years BP are recorded from beneath the tank
and associated with the underlying bajo clays. Another early pretank date of nearly seven
thousand years ago was collected at the ingress to Perdido. We are unable to assess the
amounts of water possibly retained by these ancient bajo sediments that today are
seasonally inundated at lower elevations especially in the huge adjacent Bajo de Santa Fe.
Underlying Corriental Reservoir was one very early Holocene date of approximately nine
thousand year ago associated with shallow soil overlying limestone bedrock. A thicker soil
(2,300-year-old surface) later developed in a small wetland or pocket bajo embedded
within the ancient natural drainage way (Figure 2.2). A suite of early dates are also
associated with the Inscriptions Reservoir, the earliest identified by a very early Holocene
date of 11,600 years ago. Two other dates were affected by severe argilliturbation, but
indicate sediments likely associated with natural wild fires dating to about 4,000 years BP.
Because formal trenching was not conducted in Inscriptions Reservoir, we cannot know
for sure how it functioned or when (or whether) it was constructed/modified.
Nevertheless, a single date from near the top of our core may suggest an initial date for
water capture between 1400 and 1000 BC. Harrison (1999: 1601) indicates that the
earliest trash ceramics retrieved from Tikal is from this immediate area, and the
remarkable historical text on the roof comb of the Temple of the Inscriptions provides the
earliest known date from Tikal 1139 BC (casting back in time, i.e., a retrospective date
from a Late Classic text). Clearly, more time and energy invested at this location are
warranted.

Late Preclassic Period

Although Middle Preclassic occupation at Tikal is archaeologically identifiable, these
early colonists made limited modification to their land and water resources when
compared to the subsequent spikes in landscape change. Nevertheless, human niche
construction involved the deliberate decision to occupy the ridgetop on which Tikal
ultimately ascends. The springs associated with the head end of the original arroyo now
defined by the descending TemplePalaceHidden Reservoir chain were a colonizing
attraction (Figures 2.1 and 2.2). Although we do not know whether other springs were
present at Tikal at the time, it is clear that those discovered in the Temple Reservoir silting
tank significantly influence the construction history of early Tikal. Figure 2.6 shows the
location of some of the most important Late Preclassic occupation at the site a swathe of
prepared hilltop accommodating the North Acropolis (Coe 1990) to the Lost World
complex (Laporte and Fialko 1985) as well as a ramp or causeway built immediately west
and above the silting tank (Fialko 1988). One stratigraphic sediment reversal that washed
into the main tank of Temple Reservoir again provides evidence of Late Preclassic
occupation in the vicinity prior to its Late Classic major construction.
Although our identification of head-end springs associated with a steep-sided arroyo at
Naranjo today indicates the availability of such colonizing waters, not all ridges or
hillocks provided this advantage. Niche construction does imply the identification and
employment of local resources and frequently sets one microenvironment off from another
for human harvesting or exploitation. As suggested earlier (endnote 5), Tikal appears to
have accommodated trade and exchange routes through the peninsula via natural drainages
and perhaps their subsequent modification for example, on the Ro Holmul (Fialko
2000). Other hillocks in the area may have been as strategically attractive, but elevated
spring access set the Tikal ridgetop apart.
Considerable iconographic and related information argues for origin myths that identify
a Place of Reeds as a colonizing location. Clearly the bajo margins accommodate
aspects of this metaphorical or symbolic place, but so, too, does the elevated spring
location at the summit of Tikal. When coupled with the water mountain imagery
manifested most clearly by the Aztec (Broda et al. 1982; Scarborough 1998) and present-
day Nahua speakers (Sandstrom 2014), place making at Tikal was preadapted to a widely
accepted Mesoamerican worldview.
Both the ideological reinforcement and the practicalities of potable water for drinking
supplies to laborers, visitors, and residents made Tikal an attractive center. However, with
successful Late Preclassic populations spiking, alternative water sources were needed. The
growth associated with uptown Tikal, especially the labor investments, would have
resulted in an overexploitation of the springs. Perhaps as a consequence of paving sizable
areas associated with Late Preclassic summit construction and the witnessing of the
quantities of water shed, or by drawing upon emerging Mesoamerican management
technologies (Scarborough 2006), the deliberate channeling of runoff into elevated basins
for later dry-season consumption developed. Several ends converged in the clever set of
niche construction adaptations employed by the ancient Maya at this temporal juncture.
First, the continued monumental growth of the urban center resulted in a significant
demand for quarried stone and available earthen fill. Much of the building matrix
necessary for the earlier architecture at Tikal was obtained from cut-and-fill operations
designed to level the naturally uneven and craggy hillock summit. Deforestation at the top
of the ridge was the likely result, with some runoff collection used to accommodate
seasonal building projects. The additional need for construction fill was satisfied by
deliberately quarrying into arroyo margins and at select locations where quarry stone both
aided in nearby pyramid or palace structure erection and provided cavities for collecting
water. With more pavements, more water was collected via expanded water catchments
directed into artificial basins.
Our excavations into the basal depths of the Palace Reservoir suggest that a low-lying
bench carved into limestone bedrock helped to allow access to the bottom of the original
arroyo, a steep 10-m descent from the top of the northern ravine face (Scarborough et al.
2012, online support information figure S4). This channel modification has a Late
Preclassic date of 35855 BC (Figure 2.2), perhaps suggesting that the arroyo was altered
and controlled at this early date. Fine laminated sediments overlie the meter-elevated
bench surface indicating the prospect of a low-lying dam after the abandonment of basal
access to the original arroyo channel.
Second, the Maya were incentivized by a growing deficit of water to accomplish their
many construction activities and maintain their daily living requirements. Not only were
there more demands on the finite spring flow, but the pavements throughout the site area
were sealing off any recharge potential that otherwise permitted the natural percolation of
groundwater through the formerly unaltered natural surfaces at Tikal. Although an
unintended consequence of the construction efforts at the site, the complementary
enlargement of the reservoir basins set in motion a series of infrastructure modifications.
A third development acting as a challenging stressor was the increasing incidence of
drought-prone years during the Late Preclassic period (e.g., Dunning et al. 2014).
Elsewhere, we note that Late Preclassic drying correlates with increased attention to water
needs at Tikal (Scarborough et al. 2012). Although Webster and Murtha (this volume)
reject the centralizing effects of reservoir construction at Tikal, we view the early
concentration of the water resource as critical in the initial and subsequent development of
Tikal.
Both Corriental and Perdido Reservoirs reveal suggestive Late Preclassic
manifestations with stratified, well-dated sediments dating to a potential Late Middle
Preclassic usage in the case of Corriental, followed by three of the four dates originating
from levels dating squarely to the Late Preclassic (Scarborough et al. 2012; Tankersley et
al. 2011) (Figure 2.2). An additional date from alluvial deposits in the drainage below
Corriental may indicate Late Preclassic release from the reservoir (Figure 2.2). The
elaborate switching station described earlier was likely built at this time, a period
associated with Late Preclassic drought. Perdido provides a Late Preclassic stratigraphic
sediment reversal date above a well-defined plaster floor temporally placed to the Early
Classic, so a Late Preclassic presence may be demonstrable on this southwestern flank of
Central Tikal as well (Figures 2.2 and 2.9).9

Early Classic Period

During the Classic period, several landscape adaptations were developed. A possible Early
Classic dam perhaps the obstruction accounting for the sediment banding over the Late
Preclassic bench noted previously was placed across the arroyo (and was buried under
the subsequent Late Classic Palace dam) (Figure 2.4). Although the well-designed uptown
pavements canting runoff into sizable quarries, dammed arroyos, or deliberate tanks had
been initiated, it was not until the subsequent Late Classic period that the system was
significantly enhanced and completed. The Early Classic adaptation was one to the
immediate loss of springs due to the construction and pavement of the formerly exposed
and rainwater rechargeable, porous surface bedrock in the uptown central precinct.
Generally speaking, this period of hydrologic and landscape modification appears poorly
developed, though there is little evidence that significant setbacks in civic architectural
construction or other material indicators occurred (Harrison 1999; Sabloff 2003). The
water system at the summit of the site was probably not greatly different from that
identified with the Late Preclassic period.
Farther downslope, however, several enhancements to Late Preclassic construction
occurred, suggesting a less centralized focus on water and land-use management and
perhaps an expansion of the population away from the uptown spring focal area. Both
Perdido and Corriental Reservoirs register significant modifications during this period.
Perdido manifests a plaster floor (Figures 2.2 and 2.9), and although the tank may
manifest Late Preclassic remains, it was most functional at this time (see Buttles et al. this
volume). Investigations in the Perdido Pocket Bajo lying topographically below the
reservoir indicate agricultural land use, including probable intensive cultivation of maize
likely facilitated by irrigation (Dunning et al., this volume).
Corriental Reservoir has a similar developmental history in that the northern berm was
built more for shunting water away from a rapidly infilling reservoir during an uptick in
rainfall regionally (AD 350540) (Figure 2.2). Although we do not have precise dates, the
dense masonry fill blocking the sectioned canal segment formerly feeding the Late
Preclassic basin was put in place suggesting increased amounts of water beyond the
holding capacity of the basin and the needs of the immediate locality. Although we do not
have clear evidence for agricultural plots below and beyond the reservoir, Classic period
occupations do appear with some regularity in this zone.
Both of these bajo-margin tanks suggest their inception during the Late Preclassic
period induced by population growth and perhaps by the effects of a panMaya Lowland
drying trend. With the expansion of the population away from the drying spring systems
of the summit hillock, a preoccupation with greater intensification of downslope
agricultural potential perhaps induced by upland soil erosion related to uptown Tikal
construction (see Murtha and Webster, this volume) and a projected increase in rainfall
following the Late Preclassic, we propose that dependable seasonal water abundance
triggered new landscape adaptations.

Late Classic Period

The Late Classic period ushers in a major spike in water management scale and
technological complexity. The greatest investment in the TemplePalaceHidden
Reservoir chain occurs at this time with the major construction of the Palace Dam and its
impoundment of the most water in one reservoir at Tikal, as well as the quarrying
associated with Temple Reservoir. Our dates suggest that all of this occurs after AD 660
and no later than AD 880, with a major collapse of a portion of the Palace Dam occurring
no later than AD 880 (Figure 2.2). Of particular interest is the ascension of the great king
Jasaw Chan Kawiil, in AD 682, after the century-long or longer hiatus at Tikal
(Harrison 1999; Martin and Grube 2008). Harrison (1999; 2012) indicates that his
house, Structure 5D-57, was placed immediately above the Palace Reservoir and
oriented southward overlooking the water rather than his spectacular Temple I interment to
the immediate north (we surmise that the fill for both Temple I and II was taken from that
original ancient arroyo). But it was his son, Yikin Chan Kawiil, reigning from AD 734 to
746, who may be responsible for most of the reservoir construction in the central precinct
area. Harrison (1999, 2012) further demonstrates that Yikin Chan Kawiils residence,
5D-52, was placed yet closer to the edge of Palace Reservoir, again oriented toward the
water and only accessible from that south side, and positioned to block the view of his
dead fathers house. When combined with evidence that this king was responsible for the
construction of at least two of the really huge temples at Tikal inclusive of Temple IV and
Inscriptions, the argument is strengthened. To obtain the fill for both great buildings, and
others, it is conjectured that the Palace Reservoir was enlarged and the Temple Reservoir
likely initiated. Temples III and V are believed to have been built later, with a date of AD
810 for Temple III (Harrison 1999: 175). Given the proximity of Temple Reservoir to this
latter pyramid and the fill it necessitated, it is as likely that the cofferdam adaptation to the
Temple-Palace-Hidden chain was a very late attempt to improve a growing demand for
clean water.
Of further note is the suggestion by Harrison (1999: 126) also mentioned by Martin
and Grube (2008: 21, 47), though guardedly that Palenque was allied with Tikal after the
hiatus, if not before. If so, such a political union against Calakmul and a Snake
Confederation surely entailed shared knowledge inclusive of hydraulic technologies.
French, Duffy, and their colleagues (2010, 2012) have demonstrated some of the most
sophisticated use of water technology known for the New World at Palenque (also see
Liendo 1999). Although water pressure there was driven by the numerous artesian springs,
the development of high-quality and enduring water conveyance techniques may have
influenced the kind of sluice structures posited for our Palace Dam reconstruction (Figures
2.3, 2.4, and 2.5).
Last, Late Classic waterworks farther downslope appear to remain the same as the Early
Classic period. Ceramics dating to this time are relatively abundant in these bajo-margin
tanks. The catchment area leading into the Perdido Reservoir was significantly
channelized at its upper end, with two well-defined feeder canals or chutes apparent
(Figure 2.1). Our resurvey and excavations confirmed the construction of one of these
steeply descending canal segments dubbed the water slide issuing from the planed
summit plateau associated with impervious pavements within the Mundo Perdido zone.
The segment was 10 m wide by 3 m deep cut into bedrock and terminating in a shallow
silting tank and a 90-degree turn. Immediately downstream from the bend was a feature
composed of evenly spaced boulder pylons a weir. This combination of channel
redirection, silting tank, and weir is suggestive of the need to control seasonally violent
runoff in the steeply sloping terrain flanking Central Tikal likely dating to the Late Classic
period.

Conclusion
Tikal was a low-density urban community typical of several archaic states noted in the
semitropical world (Scarborough and Lucero 2011; see Fletcher 2009; Marcus and Sabloff
2008). According to Murtha and Webster (this volume) the population or labor force for
Tikal was slight, with densities in the neighborhood of 4060 person/km2 at AD 600 that
rose from 1012 person/km2 during the Late Preclassic period. The projected population
high of 10,00015,000 at the onset of the Late Classic period is vastly lower than
projections from carefully surveyed urban settings like La Milpa approaching 50,000
(Tourtellot et al. 2003: 43) or Caracol at twice that estimate (Chase et al. 2012); such
urbanscapes did not enclose as much territory as Tikals.10 Our water consumption figures
based on the reservoir capacities at Tikal are more in line with the latter population
estimates; however, we do not wish to become embroiled in this debate. The point we
wish to make is that planning and labor costs to construct and maintain the water system at
Tikal were sophisticated and expensive, respectively.
The parts of the water system were articulated in different ways at different times at
Tikal. Many of the most sophisticated developments were likely put in place late in Tikals
history from the Temple cofferdam to the posited sophistication of the Palace Dam.
Nevertheless, inventiveness likely began early with the advent of the Corriental switching
station and the projected use of sandbox filtering in place by the Early Classic period. In
attempting to group the changes to the engineered landscape by temporal period, we have
introduced the organizing principle of human niche construction. The colonizing efforts of
the first Maya were toward springs near the summit of the Tikal ridgetop, though low-
lying residents were likely in proximity to bajo-margin aguadas or natural sinks. The
southernmost arroyo cutting west to east was likely not significantly modified until after
the hiatus, though Early Classic damming is indicated. With the truly major investments
made in Late Classic temple construction all six temple-pyramids that remain the
grand summit water system was established. Building these huge freestanding temples
required excavated fill. The planning marvel from our vantage is not simply the towering
monuments to stately tombs or cenotaphs, but the intricate interplay between the
populaces need for a potable water source and the cultural practicalities and aesthetics the
Maya demanded from their engineered landscape.
Notes
1 Compare Fletcher 2009 and Buckley et al. 2009 by way of a not-dissimilar tropical
drought spelling the demise of the fourteenth-century Cambodia center of Angkor, another
low-density urban setting.
2 Figure 2.2 indicates that the most recent date retrieved from the basal sediments of the
basin is Late Preclassic.
3 Don Fernando Tesucn, a resident of a nearby village in proximity to the Tikal Park,
visited the University of Cincinnati in the early 1990s as an assistant to the linguist
Andrew Hofling, then in the Department of Anthropology. Don Fernando was one of the
very few Itz language speakers alive and provided fundamental aid for the subsequent
Itz dictionary. He had worked at Tikal during early restoration efforts and reported sand
deposits again stratified from under the Maler Causeway (a possible dam) and within
an underlying corbel vaulted conduit. Our excavation probes did not test for this posited
sluice as we were unsuccessful in locating Don Fernando during our two field seasons.
4 Of note were the mapping and excavations at the ends of the two projecting berms
running east and west into the upper southern end of the main Temple Reservoir and
separating it from a silting tank. The channel between the berms revealed a lateral though
near-vertical drop of 3 m from the silting tank basal margins into the low-lying basal floor
of the main Temple Reservoir, with an additional 3-m drop to its center. Under the
precipitous drop immediately below the constricting berm channel were human remains
(identified but not removed) associated with at least one poorly preserved Late Classic
cylinder vessel (see Valdez and Buttles, this volume). Because the westernmost berm was
sculptured from the original bedrock and the easternmost projection built up by infilling a
portion of the original arroyo (Scarborough et al. 2012, online support information figure
S3), the burial was likely dedicatory, associated with the final construction events of
Temple Reservoir. The walkway access of the berms into the interior of Temple Reservoir,
together with the movement of water from the silting tank into the main basin over the
dedicatory burial, would have had significant symbolic import.
5 We conjecture that Tikal, Calakmul, Naranjo, Caracol, and several other interior cities
must have coped with inadequate dry-season water availability by stabilizing bajo-margin
soils and controlling sediment infilling (Beach et al. 2008; Beach et al. 2009; Dunning et
al. 1999; Dunning et al. 2002; Scarborough and Burnside 2010); bajo settings used, in
part, as shallow stream waterways allowing the movement of goods and services via canoe
traffic (see Garrison and Dunning 2009). Surveys by Fialko (2000) along the intermittent
Holmul River skirting the margins and bajos of Tikal suggest a sizable Late Preclassic
occupation; and with the subsequent centralization of Tikals population and monumental
architectural investments during the Late Classic period, such streams and connecting
canal segments likely carried exchange goods to and through greater Tikal (Gunn et al.,
2014). Too, Tikals ridgetop location would have been highly defensible on the basis of its
elevation and internal control of a significant potable water source; and with its Late
Classic population spike, the center would have had the occupational densities to
discourage military advances.
The suggestion that the soilscapes evolved with human niche construction consciously
or otherwise may also have contributed to Tikals long-term success (Murtha and
Webster, this volume), especially during its Late Classic growth surge. But to attract a
population at the outset away from the abundance of lakeshore access forces other
explanations for Tikals place making, as the soils were not markedly better suited at the
outset of settlement than in several other areas of the Maya Lowlands (see Dunning and
Beach 2010). From our vantage, it was trade facilitated over poorly understood water and
land traffic routes and the original abundance of spring waters at a defensible elevation
that made Tikal as well as other early centers in the Maya area the subsequent center
that they become (Scarborough and Valdez 2014).
6 Perdido Reservoir excavations do reveal sand lensing frequently mixed with other grain
sizes. The function of the small bajo-margin reservoir may have changed with time or
neglect from a more potable source of water to one that functioned primarily for
agricultural ends.
7 Harrison (1999: 193) makes the point that the incidence of midden deposition at
uptown Tikal was slight. Although the Maya may have had other motives for recycling
artifact waste, its absence from plaza surfaces would have reduced waterborne disease
loads given their dependency on an otherwise complex water harvesting system.
8 Dunning (personal communication 2013) asserts that Inscriptions was a natural
sinkhole in which deep, cumulic soils developed over many thousands of years and was
never converted into a reservoir. Instead, he believes the Maya used this unique
microenvironment to grow some kind of habitat sensitive crop such as cacao (see also
Thompson et al., this volume and Lentz et al., this volume).
9 Dunning (personal communication 2012) has cautioned us about the Late Preclassic
sedimentation history into the Corriental basin, indicating that those early dates may well
predate actual reservoir construction. We do not dismiss this assessment, but suggest that a
natural basin was likely of use to these downslope populations at this time, whether or not
it was modified significantly. The Perdido basin is also argued to have had a Late
Preclassic usage, though tightly controlled dating continues to elude us (see Buttles et al.,
this volume).
10 Population estimates derived from Haviland (1972) and reported by Rice and Culbert
(1990) suggest a density of 600700 persons/km2 in his 63-km2 central zone or a
population of 40,000 during the Late Classic. Tikal surely influenced a yet greater
immediate area.
 3 At the Core of Tikal: Terrestrial Sediment
Sampling and Water Management
Brian Lane, Vernon L. Scarborough and Nicholas P. Dunning

Introduction
The study of hydraulic systems at Tikal incorporated a variety of data sources, including
published reports and unpublished field notes from previous investigations as well as
specifically targeted fieldwork (Scarborough et al. 2012; Scarborough and Grazioso
Sierra, this volume). Of the many methods of field data recovery available to the project,
terrestrial percussion coring was instrumental. The Environmentalists Subsurface Probe
that was utilized allowed for quick and systematic recovery of stratigraphic and
sedimentary data over a large area.
Sediment sampling through the application of various coring techniques has seen
extensive use around the world in various contexts. The strength of coring is the ability to
extract intact columns of soil and sediment while preserving stratigraphic context. In
projects that require widespread horizontal and vertical control, terrestrial coring methods
are invaluable in balancing data extraction against time and labor. At Tikal, terrestrial
coring, in conjunction with limited wet coring, supplied a significant evaluative tool for
decoding the history and change within the water systems and their relationship to the
surrounding landscape.
A great variety of devices and techniques exist for extracting soil cores. Most common
in archaeological and paleoenvironmental studies are various forms of wet or subphreatic
sampling. These techniques require a device that can successfully extract intact samples
from saturated or submerged sediments. In this study of the central reservoirs at Tikal,
however, our coring program necessitated a means of also extracting dry and partially
moist terrestrial soils.
The intended goals for this coring program were to fill in stratigraphic gaps between
widespread test excavations as well as to provide a means of collecting intact samples for
analysis in a laboratory setting. Our approach allowed for both field descriptions of each
sample and laboratory analysis of sediments. Horizontal control was maintained to
accommodate detailed measurements of reservoir depths for projected volumetrics as well
as gradient controls for assessing waters movement. This chapter describes each of these
benefits as realized at Tikal through the application of specific terrestrial percussion coring
methods or dry coring as well as aspects of the more conventional Livingston coring
technique or wet coring. The chapter is broadly divided into two sections. First, a
general review of coring examines the rationale behind this data collection method. In the
latter portion of the chapter, we outline the rationale and the process of data collection
during the study of water systems. We conclude with a brief examination of the results of
the coring program and how these data broaden our understanding of water management
at Tikal. Our conclusion from this chapter is that this particular percussion coring
technique offers an invaluable source of data collection that can both supplement
traditional archaeology as well as offer avenues to unexpected assessments of the past.

Coring Background
Coring is the removal of intact columns of soil or sediment. The wide use of coring in a
variety of disciplines is a testament to the strength of the method. As Stein (1986) pointed
out, there are several options available to researchers desiring to test subsurface sediments.
Coring differs from augering in that it is not designed to mix the soil matrix, but instead is
capable of retrieving intact columns that better approximate the original stratigraphic
context (Stein 1986). The ability to preserve stratigraphy during extraction allows coring
to function as a more useful option for sampling a large area.
Stein (1986) cites some of the earliest forms of coring/augering in archaeology dating
as far back as the late 1930s. Principally, this subsurface testing took the form of auger
devices used to test subsurface matrices as well as aid in building chronologies. These
methods of subsurface sampling saw limited use until the 1970s, when augering and
coring became a more established method in archaeology, as ecological and site formation
considerations emerged. Additionally, the ability to encounter carbonized material
increased the utility of augering and coring to enhance chronology building. AMS dating
allowed for smaller-diameter devices to be a viable option as a means of sampling for
carbonized material.
Coring and other subsurface sampling techniques can fill a variety of roles in
archaeology. Among the primary reasons for their employment is the efficient use of time
in subsurface sampling across a large area, often possible with a minimal number of
operators. Thus, a large area can be efficiently sampled for stratigraphic information as
well as identifying potential deposits of interest by a small crew in a limited amount of
time.
The ability to visualize the subsurface composition of an area helps the researcher gain
a preliminary glimpse of what lies beneath the surface as well as fill in gaps that intensive
test excavations cannot fill in a timely or labor efficient manner. Information concerning
geomorphology, taphonomy, pedology, and subsurface variation can be quickly gleaned
through systematic use of coring techniques when combined with exacting surface survey
and elevation controls. Precise elevation data of the surface allow for a series of fence-
post diagrams or cross-sectional profiles tied to fixed elevation data that permit sitewide
stratigraphic comparison.
Through an understanding of the utility and limits of the spectrum of techniques
available, researchers can choose the most appropriate means of sampling for their needs.
Issues concerning the diameter of the sample, physical properties of the deposits, depth
capabilities, and ease of use are all factors that are considered when choosing a coring
device. Sampling of sediments from lacustrine or wet deposits is the most commonly
reported form of coring today in the archaeological and paleoecological literature. The
retrieval of wet cores often has the goal of recovering multiple proxy data, especially
ancient pollen because of the favorable preservation conditions associated with
submergence. To achieve these goals, larger core diameters are preferred. Retrieval is
aided by the typically soft nature of the matrices sampled. Wet cores can often be
extracted by a small team with little to no mechanical aid even in remote areas. The
modified Livingston corer also utilized in this project exemplifies this type of sampling.
Terrestrial cores, however, typically require a smaller diameter to cut into the drier,
often harder matrices, and there remain a variety of means to extract these samples. Hand-
driven probes, such as split-spoon augers, extract intact columns and allow an operator to
cut open the sample to view the undisturbed stratigraphy in the field. However, these
probes often lack the ability to preserve the entire column of sediment for longer periods.
Such techniques are thus subject to only cursory field analyses by way of examining the
interior of the undisturbed sample as it is found inside the device. Nevertheless, general
descriptions of stratigraphy and soils can be gained via these devices.
Other forms of coring device allow for the sampling of relatively intact columns of
sediment inside metal or plastic sheaths and permit long-term storage or transport of the
cores for laboratory analysis. Composition, chemical testing, detailed stratigraphic and
textural analyses, isotopic studies, as well as a variety of additional laboratory-based
pedological or geological studies can be undertaken with the cores recovered with these
devices. Transparent sample tubes offer the best of both worlds in allowing for initial field
description as well as the option for storage and transportation. This latter type of
terrestrial coring was chosen to serve the goals of the study at Tikal best.

Coring Operations
To understand the larger system of water management at Tikal, multiple operations with
parallel goals were planned and executed that targeted specific reservoirs and the
surrounding terrain. Several reservoirs at Tikal have been the subject of archaeological
investigations in the past (Carr and Hazard 1961; Iglesias Ponce de Leon 1987; Gallopin
1990; Scarborough and Gallopin 1991). Many of these previous data enter into the broader
synthesis, but several reservoirs remained relatively unexamined until this project.
Corriental and Perdido Reservoirs were two such reservoirs, subject to more peripheral
evaluation by several prior investigations. These gaps in the data left a need for accurate
stratigraphic information, life-use histories, reevaluation of volumetric capacities, and
study of catchment areas in order to understand the hydrology of ancient Tikal.
The main role of terrestrial coring was to gain stratigraphic control over extended areas
in between formal excavation units. Vertical differences between coring locations were
carefully monitored by our Electronic Distance Measurement (EDM) total station in order
to preserve an accurate view of the depths of stratigraphic breaks across the reservoirs.
These spatial data served as the basis for delineating the operational depths of the
reservoirs as well as for volumetric calculations. Additional goals of this subproject were
contingent on detailed compositional and geochemical tests of more reservoirs as well as
drainage courses. Terrestrial coring proved successful as a method to reduce the amount of
time and labor required to recover intact soil samples, while collecting stratigraphic data at
a meaningful scale.
 A parallel coring program focusing on broad wetland landscape changes was also
undertaken as part of the project at Tikal. Recovering detailed stratigraphic and
palynological data was of primary importance for the coring of wet and seasonally
inundated sediments. These cores targeted several inundated aguadas (ponds/hinterland
reservoirs) within the Bajo de Santa Fe depression to the east of the site center (see
Dunning et al., this volume).

The Field Recovery

The device chosen for our work at Tikal was the JMC Environmentalists Subsoil Probe
(ESP), which Scarborough has used in multiple contexts (Scarborough et al. 1994, 1999,
2000) (Figure 3.1). The same model has been utilized around the world in other
archaeological investigations (Ballantyne 2009; Cannon 2000; Tankersley and Ballantyne
2010; Junker 1999; Scarborough et al. 1994). This particular device is capable of sampling
to a depth of 4.57 m from the surface depending on the composition of the soil (further
depth is likely achievable with an additional extension segment). Sample diameter varies
by specific model; throughout this project, we utilized a sampling tube with a 2 cm
diameter. This width was optimal for the viscous, expanding clay soils encountered,
because of their resistance to vertical extraction; a larger core diameter would have likely
prevented extraction from deeper buried deposits. The initial metal core sampling segment
associated with the device has a cutting bit attached and encased a 91.5 cm clear plastic
liner that allowed for both removal of an intact sediment sample and its later storage.
Subsequent core sample segments descended in a stacked manner retrieved from the same
precise location or hole. Any invasive matrix (hole slop) introduced into the sample
after each segment was easily identified and removed in the field. Each location was
recorded as a separate lot of a coring suboperation, and each sample was labeled as it was
extracted from the earth.

Figure 3.1. Image of the ESP device and workers operating the foot lever for extraction.
A single operator can use the ESP device, but for ease of use and efficiency two to three
individuals are ideally employed to extract, record, and transport the core samples. Once a
sampling site is located, the leaf litter is cleared and the main frame of the device set into
place. A plastic liner is then inserted into the initial or primary metal sampling tube. A
slide hammer fits to the top of the apparatus and a single person can drive the sampling
tube into the ground. The labor saving aspect of this specific device is the lever system at
the base, used to extract the sampling tube from the earth. A lone operator can work the
foot-operated pedal, or in difficult conditions when extraction is complicated, a second
person can step in and lever the pedal while the other person steadies the frame. The clear
plastic liner is easily removed from the outer metal sampling tube, with further depth
achieved by inserting another plastic liner and adding metal extension tubes in a stacked
sequence. A maximum depth was regularly achievable of 4.57 m bsd (below surface
datum). This was done by utilizing the primary sampling tube and four extension rods,
each attached using metal pins designed to hold the segments together. The sharpened
tungsten bit on the business end of the primary sampling tube proved capable of cutting
through sizable roots and large cobbles of limestone during sampling.
The terrestrial coring program was primarily limited to Corriental and Perdido
Reservoirs. Because of the relatively large areas involved and the generally uniform
shapes of each reservoir, transects were laid through each to create cross-sectional
profiles. Sampling in Corriental Reservoir was initially designed to cross east to west,
connecting the proposed watercourse from the northwest ingress to the east outlet, and
again north to south, to an additional overflow outlet to create a second cross section. The
intervals used between coring locations were uniform within each reservoir, based on
reservoir size and quantity of available plastic sample liners.
Corriental Reservoir was initially sampled along three transects at 20 m intervals
because there was a limited supply of plastic liners during the first field season. Each
transect radiated from the centrally located, formally excavated control pit (Op. 1C)
located in the heart of the reservoir and connected to an ingress or outlet opening in the
berm wall (Scarborough and Grazioso Sierra, this volume; Figure 2.8). Sediments from
the berm of the reservoir wall were also regularly sampled to gain a better understanding
of the construction/depositional history. The second field season saw the addition of two
shorter transects in order to fill gaps in the data left by the original sampling. These
additional data points made for a more accurate description of the operational basin and
volume.
The second principal location for systematic coring was in Perdido Reservoir
(Scarborough and Grazioso Sierra, this volume; Figure 2.9). In this reservoir, four
transects were laid out in the form of a # sign, with two parallel transects north-south
and two more east-west. Each transect began on the berm from one side and extended to
the berm opposite. Samples were extracted at 10 m intervals because of the relatively
small size of the reservoir. On the basis of the regularity of the stratigraphy throughout, as
revealed in initial excavation (Op. 8A), cores in this reservoir were limited to two or three
extensions until the underlying dark thick bajo clay was encountered. This layer was
understood to represent the premodification soil surface. An additional core was taken
from the bottom of the Op. 8A pit in an attempt to determine the depth to bedrock;
however, within the extra 91 cm probed no stratigraphic change was encountered.
As a result of the limited profile exposure provided by a single core, whenever possible
coring transects were positioned to align with test excavations. The more precise
stratigraphic exposure of formal excavation was then compared to the cores. Joining the
coring transects with excavation unit profiles allowed the coring technicians to identify
known strata and variations across the sample area.
The portability of the coring device allowed for the collection of opportunistic samples
in other reservoirs. The center of the Inscriptions Reservoir was sampled in order to allow
preliminary comparisons with the nearby Corriental Reservoir. Similarly, two cores that
were taken within the Temple Reservoir basin and one in the upper silting tank later led to
an increased interest in the Temple Reservoirs upper basin. Further samples were
recovered from the drainages leading into Corriental Reservoir with the now-realized hope
of creating a profile of the sediments issuing into the reservoir (Nagy 2012). The ultimate
aim with these last two samples was to test the hypothesis that the eastern opening acted
as a switching station and seasonally allowed water and sediment to pass into the reservoir
from the northeast. Nagys thesis also identifies the ingress function of a south gate at
Corriental, a sluiceway suggested as a switching station by the original University of
Pennsylvania transit mapping teams (Carr and Hazard 1961).

Strengths, Limitations, and Other Considerations

A common concern with coring techniques is that a single core offers limited insight into
the subsurface morphology over a large area. Used systematically, this technique permits
multiple cores and the kind of spatial representation many projects require, and sampled
sediments can often be transported to a laboratory for later analysis. Small diameter sizes
limit resolution of the subsurface composition to some degree, especially if mottling or
specific characteristics of the soil or sediment are absent in the recovered sample. If
formal test excavations are also employed, these considerations are largely overcome.
A typical impediment to coring or augering is that of working blind. An operator does
not necessarily know what is encountered below the surface in terms of objects that may
impede or slow the extraction of a sample. With this specific percussion device, roots and
small rocks were easily cut with the force of the slide hammer and the tungsten bit. An
operator drawing on the vibrations felt through the slide hammer could distinguish
between small and large objects. The larger the stone, the more resistance felt and heard
through the device, as large buried boulders and bedrock have a distinct feeling and sound
(sound occurs when the slide hammer makes contact with the striking platform). If
bedrock or large boulders were suspected, additional cores were taken nearby to
corroborate the original finding.
Of minor note is the occurrence of loose matrices traveling vertically within the plastic
sample tube during retrieval and transportation. This situation occurs mostly in soils with
higher proportions of sand or granules. In this instance, fine and coarse material can move
in the space between the column of soil/sediment and the interior of the plastic sample
tube. As the core dries and shrinks, more space opens between the plastic and the
sediment. In the laboratory, the columns of soil can be cut in half longitudinally, and thus
any movement along the outside of the column can be ignored. The only potential problem
faced is that this movement may present difficulties in the field while describing matrices
as seen through the plastic sample liners.
A central concern during our reservoir investigations was that of incomplete recovery or
the possible compression of the sample. Wright (1993) points out that with wet cores in
particular, what is commonly termed compression may in fact be the result of incomplete
recovery of the sediment sample due to internal friction within the sampling tube or
internal pressure that makes the collected soil act as a plug and prevents further soils from
entering the sampling tube. This is an unlikely scenario because the ESP device does not
create an airtight seal with the sampling tube. Nevertheless, Cannon (2000) suggests that
with the ESP, the percussive act of driving the sampling bit may actually be capable of
compressing porous terrestrial soils. In order to compensate for these conditions, the depth
that was sampled in our cores was recorded with a tape measure separately from the length
of the samples recovered, with initial profiles of each column of soil recorded in the field.
It is likely that a combination of incomplete recovery due to internal friction as well as
some slight compression that reduced the overall length of some samples occurred.
Nevertheless, when our core samples were compared to our open control excavations as
well as adjacent cores we were able to ensure interpretative accuracy.
Further, shrinking was noted as the samples dried within the sample tubes. This
highlights the need to make detailed notes in the field shortly after the samples are
extracted. These observations record the original depths sampled on the basis of the length
of the sampling tube and lengths of core segments; the dual measurements allow for later
corrections. To monitor the drying processes further, each sample tube was cut down to
the exact length of the soil sample after extraction to minimize sample movement with
shrinkage in the tube, but then related back to the length of the original plastic sheath.
An occurrence of note was the expansion of some thick, shrink-swell clay soils
(Vertisols) from within Corriental Reservoir. In this instance, the sample expanded within
the plastic sheath and was effectively wedged within the primary metal sampling tube.
Once the plastic sample case was forcibly removed, the operators observed that the plastic
had shattered inside the metal tube from the internal expansion pressure of the clay.
As with any method, there are some limitations to dry coring, but the net effect of
gaining a reasonably complete view of the subsurface topography in a labor- and time-
efficient manner makes it worth consideration. When complemented by a program of
soil/sediment control pits and formal feature excavation, the coring effort allows an
accurate assessment of past depositional events. In the study of the reservoirs at Tikal,
coring proved itself as an integral part of the project.

Goals and Analyses

Initial efforts focused on core extraction and field analysis. This stage accounted for the
baseline descriptions, measurements, and profiles made for each core. Largely, this
documentation and initial examination aimed at providing immediate stratigraphic insight
into each reservoir. These analyses were done concurrently with the excavation of the test
pits so that the information gained from either operation could aid in guiding further field
testing.
An example of how coring shaped the excavation program is provided by the few cores
retrieved at the Temple Reservoir, which gave insight on where to place formal
excavations. One core was located in the upper tank and the second in the primary basin of
the reservoir. In part because of the high moisture content in the upper tank core, the first
test excavation was opened in this particular section leading to the eventual identification
of possible ancient spring activity. This finding highlights the utility of informing
researchers of conditions below the surface quickly and with minimal investment.
The goals for these data required a controlled setting for analysis. The first was the
reassessing of operational volumes of Perdido and Corriental Reservoirs. Second was a
means of providing intact soil/sediment columns for systematic AMS dating as well as
geochemical and compositional analyses. These subsequent tests were designed to create
more precise sediment profiles in addition to generating a better understanding of the life
histories of each reservoir (sedimentation, possible dredging, etc.). The additional cores
extracted from locations outside these two primary reservoirs were also subjected to these
analyses.
The reevaluation of volume was based on the shape and depth of the reservoirs as
estimated from operational depths provided by the cores and test excavations. Instead of
using current topography or employing a central depth or depths from the edges of
reservoirs clearly an unrealistic assessment of subsurface shape the depths from every
core were analyzed and entered into a geographic information system (GIS). By creating a
vertical series of false bottoms through time based on sedimentation and dredging
histories for each reservoir, the program was able to calculate accurately the amount of fill
in terms of these data (see Weaver et al., this volume).
To determine the depth used for estimations of fill capacity, the stratigraphy of the cores
was correlated to the exposed layers found in the test excavations. The base of each
reservoir was identified in most sample locations. In Perdido Reservoir, the operational
bottom manifested as a thin deteriorated plaster lining on top of the dark viscous bajo clay
(with associated dating). This slight addition of depth was used in the estimates for
Perdido Reservoir when compared to earlier projections (Carr and Hazard 1961;
Scarborough and Gallopin 1991). In Corriental Reservoir, operational depth was
determined on the basis of a combination of AMS dating, pedological analysis, and
stratigraphic breaks. There was little evidence of dredging or high amounts of siltation
during the use of Corriental Reservoir (Scarborough et al. 2012).
Reassessments of the capacities for each of the two reservoirs reveal part of the
hydrological capacity of Tikal. Previous estimates were made by Carr and Hazard (1961),
as well as a later systematic reassessment by Gallopin (Table 3.1). The latter calculation
used the assumption that the reservoirs would fill to their physical capacity. The coring
operation effectively demonstrated that there is an approximate 6,000 m3 of sediment in
the Corriental Reservoir over the first operational bottom, and a similar 2,000 m3 in
Perdido Reservoir. It is unlikely that these reservoirs were filled to absolute brimming
capacity, and although the depths of the reservoirs have been increased, the amount of
water stored is likely similar to Gallopins (1990) estimates (Scarborough and Weaver,
personal communications).
Table 3.1. Volumetric estimates of Tikal reservoirs

Reservoir Carr and Hazard Gallopin

 Corriental 17,380 m3 57,559 m3

Perdido 3,070 m3 4,605 m3

Additional data extracted from the core samples were obtained by targeted laboratory
tests on subsamples of various cores taken from each reservoir or from specific
stratigraphic contexts within individual cores. Basic descriptions including, color, texture,
composition, and particle size analysis, were among the first set of procedures to be
performed on the core samples. We preserved the ability to perform contextualized
description of stratigraphy because of the intact nature of the cores preserved in each
sample tube. Additional description of structure and composition was more accurate in the
laboratory than that made possible with conventional bagged soil samples.
Further testing of the samples included AMS dating, loss-on-ignition to quantify
organic matter, chemical and element testing using Inductively Coupled Plasma (ICP)
spectrometry, X-ray diffraction, and X-ray fluorescence. Magnetic susceptibility, scanning
electron microscopy, and isotopic tests (both stable carbon and nitrogen) were also
performed (Scarborough et al 2012; Tankersley et al. 2011; Nagy 2012). Arbitrary 10 cm
increments were used in these tests in order to offer a complete description of the entire
column. For specific samples where pollen was recoverable, limited palynological work
was also done; however, the dry terrestrial sediments proved poorly suited for pollen
preservation.
Through the course of these detailed sediment analyses, surprising revelations were
made. One of the unexpected discoveries was the presence of volcanic tephra in the
microstratigraphy of the cores retrieved from near the centers of the two reservoirs
Corriental and Perdido (Tankersley et al. 2011; see Tankersley et al., this volume).
These tests created detailed profiles that were characteristic of each reservoir that we
systematically sampled. Testing also accommodated detailed glimpses into less
systematically sampled locations such as the Inscriptions and Temple Reservoirs and from
specific drainages. High-resolution data from the transported samples combined with the
benefits provided from field analysis of the cores even in the context of limiting
difficulties exemplify the strengths of this method in conjunction with traditional test
excavations and survey. The end product is a more complete picture of the landscape and
hydrology of ancient Tikal.

Conclusion
Percussion coring techniques of terrestrial deposits were an invaluable sampling method
for this project. Recovery of numerous samples has allowed for a more accurate
estimation of ancient volumetric capacities and an understanding of the cultural,
geomorphic, and pedological history and composition of both Corriental and Perdido
Reservoirs. The large number of cores extracted allowed for a variety of detailed tests in
the laboratory and provided the basis for new discoveries concerning the amount and
significance of volcanic tephra falls across the Maya Lowlands. Of additional significance
was the revelation that nearly all the core matrices retrieved from the reservoirs could be
dated. The light but persistent charcoal rain from continuous domestic fires, plaster
production, ceramic kilns, and at least some agricultural burning has provided important
dating access. Because of the relative ease of use, timeliness, and long-term preservation
and transport of cores, this specific technique proved an integral facet in the process of
understanding of the reservoirs at Tikal.
4 Bringing the University of Pennsylvania Maps of
Tikal into the Era of Electronic GIS
Christopher Carr, Eric Weaver, Nicholas P. Dunning and Vernon L. Scarborough

Introduction and Overview

In 1956 the University Museum, University of Pennsylvania, began what would be a
fifteen-year project at the ancient Maya site of Tikal, Guatemala (hereafter, the Penn
Project). As one of its first efforts, the Penn Project produced a series of paper maps of
the site. This chapter reports on our work to convert these paper maps, and the wealth of
information they contain, into an electronic format for use in Geographic Information
Systems (GIS). After this conversion, we assessed the accuracy of the maps using Global
Positioning System (GPS) receivers, land survey equipment, and aerial radar altimetry.
Our assessment is that the maps are very accurate. This chapter explores the factors that
contribute to the maps accuracy.

The Penn Project at Tikal and Prior Mapping Efforts

The Penn Project, under contract from the Guatemalan government, had as its objectives
archaeological and biological research, site restoration, and touristic development (Shook
1958; Coe and Haviland 1982). Edwin M. Shook, field director, reports that much of the
first two seasons (1956, 1957) was devoted to building camp, digging wells
(unsuccessful), bushing, and road building (Shook 1958). Once such vital needs as
housing and provisioning had been secured in league with a rustic system of internal
roadways, it became mapping that received absolute priority (Coe and Haviland 1982:
23). Using nine survey crews over four field seasons (195760), the maps of the central 16
square kilometers of Tikal were completed and published in 1961 as Tikal Report No. 11
(Carr and Hazard 1961), the subject of this chapter (hereafter, the TR11 Maps). The
center map of the TR11 Map series, the Great Plaza Quadrangle map, is shown in Figure
4.1. The names and the arrangement of the TR11 Maps are given in Figure 4.2. As the
Penn Project anticipated, once the map had been issued, it became the principal vehicle
for the selection of subjects to be excavationally investigated. Its scrutiny served
particularly to reveal provocative patterns and anomalies thereafter dug (Coe and
Haviland 1982: 25).
Figure 4.1. The lower half of the Great Plaza Quadrangle map. This is the centermost
map of the TR11 Detail Maps; see inset map in lower right corner. Ruins generally are
shown by the schematic method of representation, after Parris. Larger ruins are shown as
architectural plan views. The inset map, lower left, gives the work areas for the survey
teams. Topography is shown by 1-meter interval contour lines. Image courtesy of the Penn
Museum.

Prior Mapping and the Schematic Representation of Ruins

As this chapter will show, the TR11 Detail Maps are very accurate and complete. The
questions become, What factors joined together to make the maps so good? And, What
can we learn for future projects?
Part of the reason the TR11 Detail Maps are so complete and accurate is that the Penn
Project team learned valuable lessons from mapping efforts at other ancient Maya sites
in particular Mayapn (19501), Piedras Negras (1932, 1939), and Chichn Itz (1924,
1929, 1932). The mapping work done at these three sites pointed the way to producing a
complete and accurate map and demonstrated the value of a good map to an
archaeological project. The benefits derived from these previous surveys also helped to
justify the cost and effort the Penn Project mapping was to require in Tikal.
The Mayapn mapping provided several valuable lessons for the Penn Project. Prior to
Edwin Shooks role as the first field director of the Penn Project at Tikal, he was an
archaeologist on the Mayapn project. Shook introduced the Mayapn project surveyor,
Morris R. Jones, United States Geological Survey (USGS), to the Mayapn site (Shook
1950). Mapping Mayapn gave Jones, a surveyor previously working in the U.S. Midwest
(Lee 2004), experience in recognizing and recording ancient Maya ruins and working in
rough vegetation. The Mayapn map (Pollock et al. 1962) used a simple rectangle method
of representation for the ruins, showing general size and orientation of the mounds, but no
details and no height. The Mayapn map served as a key to, and of invaluable aid in, the
past seasons work (Ruppert and Smith 1951: 660): money well spent.
The 1932 Piedras Negras map (and its 1939 revision) was the acknowledged model for
the TR11 Maps (Carr and Hazard 1961: 4). The field director at Piedras Negras, Linton
Satterthwaite, later to work at Tikal, oversaw the generation of a new map for the site
starting in 1932 (Satterthwaite 1943). The new map replaced the 1901 map of Toebert
Maler (Maler 1901). Three factors were critical to the high quality of the 1932 Piedras
Negras map. First, as at Mayapn (Jones 1952), the goal was to make a complete map
to show the shapes, heights, orientation and assemblage of all terraces, platforms,
mounds, and standing structures, small as well as large, and was to include peripheral
areas not thus far recorded (Satterthwaite 1933: 11). In other words, the goal was not just
to map the grandest structures and not just in the central precinct, but to include all
structures as well as outlying commoner areas. Second, as at Mayapn, the mapping was
done by someone specifically trained for this kind of work. In this case the survey and
mapping effort was lead by the architect Fred P. Parris (Satterthwaite 1933: 11). Third,
Parris used a schematic representation of the ruins. Satterthwaite (1933: 15) notes that it
was just not possible for the Piedras Negras project to make small-interval contour maps
of each mound. Instead, Parris, no doubt calling on his training in architectural drawing,
developed a schematic method of representing the mounds. Satterthwaite describes at
length Parriss method of schematic representation, locating many points both
horizontally and vertically (but not as many points as required for contour lines) to show
the shape of the mounds, terraces, and so on, including side slopes and various
indentations (e.g., from the collapse of corbel vault ceilings). This schematic method of
representation is essentially a plan view of the ruins. Plan views are a basic method of
representation used by architects and engineers.
The schematic representation of ruins is a compromise between speed of execution and
information content. With the schematic representation the shapes of the ruins are
stylized, but at the same time pictorially presented as they actually appear, and not merely
as symbols (Carr and Hazard 1961: 4). The simplest representation of ruins, a standard
symbol, such as a dot or square, would be fast to map but have low information content
(no size, no orientation, no height). A simple rectangle drawn to scale, as at Mayapn or
the Maler (1901) map of Piedras Negras, adds information on size and orientation, but the
simple rectangle may overgeneralize by forcing nonrectangular ruin shapes to fit a
rectangular symbol. The otherextreme, contour lines over the mound, takes a great deal
more time to measure and draw and requires clearing much more vegetation to obtain the
required sight lines. The surveyor would be hard pressed to draw enough microcontour
lines to match the detail in a schematic as drawn by an architect or engineer. For example,
a later project at Nohmul, Belize, started with contour mapping of mounds for the site map
but switched to schematic mapping to save time (Hammond 1988). In justifying the switch
from contour to schematic, Hammond notes the schematic method of representation has
disadvantages, principally in the lack of any independent control on the interpretative
detail, it does successfully present relative dimensions and overall configuration and, if
accepted as the shorthand method that it is, allows much larger areas to be mapped than
with a purely topographic technique (Hammond 1988: 3). Further, the mapping field
notes can include top-of-mound elevations, as was done at Piedras Negras and Tikal, thus
making the schematic less relative.
 Malers 1901 map of Piedras Negras uses the simple rectangularized method of
representation of ruins rather than the fully evolved schematic representation of Parris. See
Carr and Hazard (1961) for a comparison of the Maler and Parris methods of
representation of ruins. The rectangular method has come to be known colloquially as the
Maler Convention among Mayanists. The 1932 map of Piedras Negras introduced the
schematic method of representation of ruins. The 1939 version of the Piedras Negras map
contains the acknowledgment schematic mound forms after Parris (Satterthwaite 1948:
13 and appendix 6). Thus, the schematic method of representation of ruins might rightly
be named the Parris Convention for its first user, Fred Parris.
Another map providing inspiration for the Penn Project is from the site of Chichn Itz
(Ruppert 1935, 1952). The Chichn Itz project, as at Mayapn, enlisted the leadership of
an experienced surveyor from the USGS to make its map, Jerome O. Kilmartin (Kilmartin
1924). Kilmartin worked in the challenging brush of northern Yucatan for two field
seasons, 1924 and 1929. Later, Kilmartin was a liaison to both Morris R. Jones at
Mayapn and the Penn Project at Tikal. Kilmartin did not use the schematic method of
representation for the Chichn Itz map. A small extension of the map in 1932 by John P.
ONeill did use the schematic method of representation of ruins (Ruppert 1935, illustration
350; Ruppert 1952, illustration 151). It is possible that ONeill, a volunteer on the Chichn
Itz project, learned the schematic method of representation from Fred Parris. Sylvanus
Morley reports Parris, a Penn Museum employee, and ONeill received training on
astronomical surveying during this period, possibly together (Morley 1937, v. 1, p. x).
The ancient Maya site of Uaxactn, just north of Tikal, also was extensively mapped in
the 1920s and 1930s. The Map of Uaxactn (Ricketson and Ricketson 1937, figure 198)
uses the Maler-like rectangularized representation of ruins, not the schematic method.
Another map of Uaxactn, the transect map of the House-Mound Survey (Ricketson and
Ricketson 1937, figure 2), also did not use the schematic method of representation. The
House-Mound Survey, however, may have been an inspiration for another Penn Project
map series, the TR13 settlement survey maps that radiated out from central Tikal,
including a link to Uaxactn (Puleston 1983).

The Penn Project Survey Teams

When it came time to start the Penn Project, field director Shook already had firsthand
experience from Mayapn on the value of a complete and accurate map and the required
fieldwork and production steps needed to make it. Further, the second person on the Penn
Project staff list, Satterthwaite (Coe and Haviland 1982), had complementary experience
at Piedras Negras and with schematic representation. Building on that experience Shook
selected principal survey team leaders with the relevant training and temperament to work
at the accuracy required for a mapping project of this scale: a civil engineer, James
Hazard, University of Toronto (1957 and 1958 seasons), and an architect, Robert Carr,
Penn State (1959 and 1960 seasons). Other survey teams were led by University of
Pennsylvania architecture graduate students Newton N. Levine and Richard S. Wurman,
along with forestry graduate student Hans M. Gregersen, Penn State, and engineer
Eduardo Martinez, Mexico all of whom would have had surveying and drafting courses
as part of their studies (R. Carr, personal communication, August 1, 2012). Archaeologist
William Coe was also involved. Coe helped Shook set up the mapping strategy, did
mapping the first year (Wilson 2012; Carr and Hazard 1961), and provided ongoing
support (R. Carr, personal communication, September 6, 2012). Just as important, the
project hired a crew of local workers, mostly chicleros, who knew the terrain and who
provided continuity on the survey teams from year to year (Wilson 2012). The
multinational nature of this effort was exemplified by the nationalities of the two authors
of Tikal Report No.11 James Hazard (Guatemala) and Robert Carr (United States).
Shook recruited his colleague from Mayapn, Morris Jones, to train these survey team
leaders. Jones was in Tikal for the first month to set up the mapping procedures with
Shook and to train Hazard, Coe, and others (Figure 4.4). The Penn Project also was in
continuous contact (195669) with Kilmartin, the Chichn Itz surveyor, for support and
advice (Jerome O. Kilmartin Collection in the George E. Stuart Collection of
Archaeological Materials, UNC libraries). When Carr took over the survey effort for the
second two years, Jones again trained him. Jones trained Carr in the use of the plane table
over several weekends in Washington, D.C., where they both lived (R. Carr, personal
communication, August 1, 2012). Carr does not recall that anyone trained him in the
schematic method of representation of ruins. When Carr arrived in Tikal he reviewed the
maps Hazard and the others had already made. As an architect, he immediately grasped
the concept and what he needed to do to complete the mapping the schematics were plan
views of the ruins (R. Carr, personal communication, September 6, 2012). But at the end
of the day, it was Edwin Shook who provided the overall inspiration and direction for the
mapping. Hazard, speaking of his introduction to the rigors of mapping and the high
quality standards of the project, relates, Ed loved mapping and he had a lot of experience
so the rest was easy! (J. Hazard, personal communication, March 13, 2013).
The Penn Project continued to utilize the expertise of the USGS when it came time to
prepare the plane table maps for publication, enlisting C. B. Brady, USGS (Carr and
Hazard 1961: 1). Carr met with Brady several times a week in Washington, D.C., during
the production process to assure the accuracy of the maps (Wilson 2012). Nevertheless,
some items were missed. For example, there are multiple typos in the latitude-longitude
noted on the map borders (cf. Love 1927).

The Challenge of the Tropical Forest

The Penn Project acknowledged the need for people able to recognize well-covered ruins
and to work in the dense topical forests of Central America. Tatiana Proskouriakoff
relates: Today, large tracts of forests are unpeopled, and a traveler in the Petn district of
northern Guatemala must make his way on foot or on muleback, hacking with his machete
at the fast-growing vines that constantly strive to obliterate the man-made trail. He may
thus pass through the very heart of a once populous city unaware that the sharp little rise
he crosses, which seems nothing more than a curious whim of nature, is in fact the ruin of
some ancient edifice (Proskouriakoff 1963: xiv). Or Scarborough: We all know the
difficulties in rediscovering or mapping ancient ruins in the tropics today, in dense
secondary vegetation, an entire pyramid can be overlooked if a fortuitous survey line does
not cross over a portion of the structure (Scarborough 2003a: 82). Shooks preface to
TR11 confirms that conditions in Tikal were no different, saying the report describes the
strenuous task of field mapping and the problems encountered in accurately depicting the
archaeological remains where the natural elements have exerted their destructive force for
a thousand years (Carr and Hazard 1961: iv). These conditions persist today in many
parts of the Maya realm (and for us in Tikal), though deforestation has overtaken sizable
areas including those abutting Tikal National Park on the south.

The Penn Project Survey Procedures

The survey at Tikal generally followed the proven procedures from Mayapn and Piedras
Negras. The extent of the TR11 Maps was somewhat arbitrarily determined. Shook made
a general reconnaissance looking for the limits of the ancient city of Tikal in 1956. Ruins,
house mounds, temples, and palaces continued as far as he searched. In Shooks search
no recognizable limits were observed (Carr and Hazard 1961: iii). Shook arbitrarily
selected a 4 km-by-4 km square, roughly centered on the Great Plaza, as the map limits
(Figure 4.2). The inner 3 km square was mapped with a plane table, at a scale of 1:2,000,
with a 1 m contour interval. This square was presented in the nine detail maps, each detail
map 1 km on a side (Figure 4.2). The Penn Project also repeated the practice of mapping
ancient cultural features of even the most modest size including the outline of house
mounds, platforms, altars, standing walls, and causeways. The TR11 Detail Maps
represent ruins in schematic form, with relative indication of height (the Parris
Convention), and use standardized symbols (i.e., not schematics) to show the locations of
ancient Maya altars, stelae, chultuns, and quarries. Notably, Edwin Shook was an
engineering student before he was hired to draft maps of Maya sites for the Carnegie
Institute of Washington (Veronda 1998). Shook would have been comfortable with, maybe
even encouraging of, this schematic representation method. The Penn Project survey
benchmarks, a steel stake generally set in concrete, are also shown, as are spot elevations
in many plazas.
Figure 4.2. The arrangement of the nine maps in the TR11 Detail Map series Bejucal,
North Zone, Encanto, Temple IV, Great Plaza, Camp, Perdido, Corrential, and Temple of
the Inscriptions. Not labeled, the 4 km by 4 km area is the Ruins of Tikal map. The
figure is oriented to UTM grid north (very close to true north). The TR11 Maps are
oriented to the magnetic north of ca. 1957. For reference, the road into and out of the Park,
along with the runway, are shown. The triangles mark the calculated locations of the
virtual control points used to georeference the ten maps. The triangle in the Camp Quad
marks the Petty Benchmark, our point of beginning. The dots mark the locations of the
96 map features used to assess horizontal map accuracy.
A 0.5 km-wide outer band surrounding the detailed area was mapped with less precise
rapid reconnaissance methods (Brunton compass and pacing). A simplified version of the
TR11 Detail Maps, along with the 0.5 km band, constitutes the tenth map in the TR11
Map series. This simplified map, the Ruins of Tikal, is at a 1:6,250 scale, with a 5 m
contour interval. Ruins on the simplified map are represented as simple black rectangular
symbols (the Maler Convention, Figure 4.3, right).
Figure 4.3. Left image shows two map accuracy test points in a plaza group on the
Temple IV Quadrangle map; a chultun opening, item 4; and the center of the ruins of a
range structure, item 13. The ruins are drawn using the schematic method of
representation the Parris Convention. The right image is a Maler Convention
representation of the same ruins (from the TR11 Ruins Map). The inset photograph is
the opening of a typical chultun. Error vectors for the two map accuracy test points on the
left image show how far the mapped position has to move to match the true position as
determined by GPS. In the case of the chultun, its mapped position has to move in the
northwest direction about 7 meters. The X in the center of the plaza marks a spot
elevation of 251.4 meters above the Penn datum. Basemaps courtesy of the Penn Museum.
The key instruments for the Penn Project survey were a plane table, telescopic alidade
with magnetic compass, and stadia rod (Figure 4.4). With these tools the survey team
leader drew the map on the plane table as the sights were made. Data collection and
drafting were combined in a single operation. Since the map was created by the surveyor
in the field, error checks could easily be made by visual comparison between the map and
the surrounding area (Low 1952). While not shown on the published maps, the field books
of the survey team leaders contain the elevations of the ruins (Carr and Hazard 1961: 6).
Figure 4.4. The TR11 Detail Maps were made with the thenstate of the art technology
plane table and telescopic alidade. The photo on the left shows Morris Jones, USGS,
demonstrating to William Coe and Vivian Broman their use (Courtesy of the Penn
Museum, image #C57-008-0008). The right photo is an unidentified local worker holding
a stadia rod in a ruin (Courtesy of the Penn Museum, image #64-037-0076). The stadia rod
is placed on the point being measured and is sighted through the alidade. The graduations
on the stadia rod are used to calculate the distance and elevation of the point.
As in Mayapn, the Penn Project group attempted to tie the maps into the global system
of latitude, longitude, and elevation. Unfortunately, at Tikal there was no nearby
government network of benchmarks with geographic coordinates. Instead, TR11 cites an
approximate location of the site based on an astronomic observation on Temple I from a
1923 expedition (Love 1927) and an oil company radio navigation system (SHORAN) in
ca. 1959 (Carr and Hazard 1961: 1). Both location measurements were very good for their
time, but do not meet current accuracy standards (thus, we updated these location
measurements).
The use of benchmark monuments and survey loops was a key to the accuracy of the
TR11 Detail Maps. To prevent the accumulation of error as a map is extended out from the
starting point, surveyors use a principle of working from the whole to the part
(Schofield 2007). The whole refers to the networks of benchmarks covering the whole
area of the map. The benchmarks are connected together in survey loops. The loops are
mathematically checked to assure they return to the start point. If any loop does not return
to the start point, the surveyor knows there is an error and can resurvey the loop. When the
loop closes with the required accuracy, the surveyor knows the benchmarks are properly
located the whole is established (Wolf and Ghilani 2006). Now the part can be
tackled. The part is the local surveying carried out in the area immediately around each
of the individual benchmarks (for their local surveying, the Penn Project used the
magnetic compass in the alidade to align the plane table to magnetic north, rather than the
alternative of backsighting on another benchmark).

Georeferencing the Maps

Integrating historical maps in GIS to analyze the spatial information they contain, or to
layer them with other spatial data, requires that the maps be georeferenced. That is,
selected control points on a scan of the original map must be aligned with their actual
geographic coordinates (Rumsey and Williams 2002: 4).

Georeferencing Methods
Podobnikar (2009) groups georeferencing methods into two general types
transformation between two coordinate systems where the transformation parameters are
known, and transformation using identical ground points where the transformation
parameters are not known. We were able to use the first method to achieve a great savings
in time and effort.
The second method, though, is generally more common in the field of geography and
GIS. The second method takes control points (points whose locations are known in a
modern coordinate system, such as UTM, Universal Transverse Mercator), and matches
them with those same locations on the historic map. Church spires, bridges, road
crossings, and towers are typical points used in this matching scheme. The locations of
these points are typically known from modern georeferenced maps and aerial photographs.
Unfortunately, the Tikal area does not have modern churches, bridges, and so on, to show
on modern maps, satellite imagery, or aerial photographs, and almost all the ancient Maya
constructions are covered by a thick forest canopy. With the exception of a few cleared
structures, satellite images of the area show a sea of green.
It is also possible to establish control points for georeferencing by using a GPS to
measure the location of points within the boundary of each map. The tangled understory
and limited visibility in the Petn forest make determining where you are, so that you can
GPS a control point, extremely difficult, time-consuming, and expensive. To navigate in
the forest it is best to have a georeferenced map stored in the GPS (the task may be close
to impossible without such a map, as we discovered when working in an area where the
TR11 Detail map in our GPS was not functioning).
Fortunately, because of the accurate surveying of the Penn Project, the time and cost
required to collect multiple control points on each map could be eliminated. Instead, we
could georeference by transformation between coordinate systems (in our case between
the Penn Project coordinate system and the UTM coordinate system). This method
requires the location of only one point, along with a reference direction and the map scale.
We used trigonometry to create transformation equations to calculate the positions of the
corners of the TR11 Maps (the virtual control points) in the UTM coordinate system.
This involved converting distances and angles to the map corners in the on-the-ground
Penn coordinate system to the projected UTM coordinate system. These calculations
follow the method of reduction of angles and distances to the UTM grid and the
computation of UTM coordinates of traverse stations found in the Elementary Surveying
text by Wolf and Ghilani (2006, chapter 20). Finally, point-matching-type georeferencing
was used to snap the corners of the maps on to the virtual control points.

Penn Project Coordinate System

The Penn Project used a standard land survey practice to start their mapping; they
established a local coordinate system. Their start point, point of beginning in survey
terms, was the Datum Benchmark behind the camp kitchen (Figure 4.5). This point of
beginning was the reference location for their horizontal and vertical surveying. Their
reference direction was magnetic north. This point of beginning, the orientation we call
Penn Project magnetic north (Figure 4.5), and their distance scale define a local
coordinate system for their survey measurements. Our task was to transform this local
coordinate system to the global coordinate system of latitude and longitude that is, to
georeference the maps. Actually, our transformation was to the closely related UTM
coordinate system. The UTM coordinate system projects physical locations from the
roughly spherical surface of the earth (the spherical system used in latitude-longitude
coordinates) onto a flat surface, with Cartesian coordinates. On the basis of the availability
of satellite imagery with nominal 1 m pixel spacing, our goal was to georeference the
maps to nominal 1 m accuracy.

Figure 4.5. Georeferencing by land survey methods requires a point of beginning, a

reference direction, and the map scale. The left image from the Camp Quadrangle map
shows the Penn Project point of beginning; the Datum Benchmark, our point of
beginning; the PETTY CO.B.M (triangles); and the map scale. Basemap courtesy of the
Penn Museum. The insert photo is our GPS on the Petty Benchmark. The right diagram
shows the relationship between the reference direction used by the Penn Project and the
reference direction used in our georeferencing, UTM grid north.

Point of Beginning Location and GPS Procedure

We measured the location of the point of beginning for the TR11 Maps using GPS. We
were unable to locate the Penn Project point of beginning, the Datum Benchmark
indicated on the map (Figure 4.5). Fortunately, an adjacent benchmark, marked on the map
as PETTY CO. B.M, was still present on the ground (the Petty Benchmark). This
alternate benchmark became the point of beginning for our calculations. The Petty
Benchmark was set by an oil prospecting company, the Petty Geophysical Company, in ca.
1957, and was located using a predecessor to GPS, the land-based radio navigation system
called SHORAN (short-range navigation). As a check we compared our 2010 GPS
location to the SHORAN derived location. When converted from the Petn version of
datum NAD27 (Tikal, ca. 1990) to datum WGS84, the Petty Benchmark location (Carr
and Hazard 1961: 1) differs from our GPS reading by less than 20 m. On a 2008 visit to
Tikal, Robert Carr used his firsthand knowledge to relocate the Datum Benchmark. It has
a brass cap stamped Camp Kitchen 200 with the 200 referring to the assumed
elevation of the benchmark above sea level, in meters (Robert Carr, personal
communication, August 1, 2012).
The inset photo in Figure 4.5 shows the Petty Benchmark along with our GPS unit. To
maximize position accuracy, we conducted multiple GPS sessions over a six-week period.
The multiple sessions were averaged together to determine the location of the Petty
Benchmark. On the basis of the range of the positions, we estimate our position accuracy
to be better than 3 m in the horizontal and vertical directions. For this location
measurement, and for the later map accuracy checks, we used a consumer-grade GPS, a
Garmin 60CSx (Garmin 2007). When collecting point locations with the GPS, we
activated WAAS (wide area augmentation system) real-time differential correction and
averaged one hundred or more readings at one-second intervals. While WAAS differential
corrections are not specifically calibrated for Guatemala, WAAS is calibrated for the
adjacent area of Mexico. For the first quarter of 2010 (the time we were in the field) the
U.S. Federal Aviation Administration reports that GPS under ideal conditions, with WAAS
activated, had a 95 percent probable horizontal accuracy of 0.701 m in Merida, Mexico,
and 0.883 m in Tapachula, Mexico (WAAS, 2010: 10). Tikal is roughly on a line between
Merida and Tapachula. With this level of accuracy available from GPS with WAAS real-
time differential correction, we did not need postprocess differential correction from the
then-nonfunctional CORS (continuously operating reference station) station at the nearby
Santa Elena, Petn, airport.
On the basis of our GPS readings we estimate the location of the Petty Benchmark in its
2010 position to be 02 22 102.7 east by 19 06 336.3 north (UTM zone 16 north, WGS84
datum), with an elevation of 244.3 m above sea level (base of monument). Our map
accuracy checks suggest the Petty Benchmark has been moved in the past fifty or so years
since the Penn Project mapped it. For the final georeferencing (what we call SHORAN
version 11) we used what we calculate to be the original position of the Petty Benchmark
as mapped ca. 1957. We derived this original position by minimizing the offset in the
centroid of the error vectors for the map accuracy test points. This calculation shows the
Petty Benchmark may have been moved 4.7 m to the west and 0.4 m to the north. This
calculated move is consistent with our tape measure readings to the centerlines of the
roads east and south of the monument. The severe out-of-level position of the benchmark,
Figure 4.5, in this area of modern construction, is also consistent with a move.

Reference Direction Penn Project North to UTM Grid

North
We needed a reference direction in addition to the point of beginning for the map. The
Penn Project used magnetic north as its reference direction a direction read directly from
the magnetic compass that was part of their telescopic alidade. It turns out the magnetic
compass of their alidade was slightly out of adjustment. The survey crew caught this
problem by determining the direction of true north by the alignment of the star Polaris (the
North Star). On August 1, 1960, Carr determined true north from Polaris observation.
The grid north of the map was found to be 559 E. of true north, slightly less than the
calculated magnetic declination (Carr and Hazard 1961: 3). On the basis of the star sight,
the Penn Project set up a pair of monuments aligned to what we call the Penn Project
magnetic north. Presumably, they regularly checked and aligned the compass in their
alidades against these monuments (by always setting up their compass against these
benchmarks, they did not have to adjust for the movement of magnetic north over the
duration of their measurements four years). We used the reading from their Polaris sight
to determine the correction between their magnetic north and true north (559), not the
true value printed on the TR11 Maps (Figure 4.1). The true correction between magnetic
north and true north is 648 for 1960 in the Tikal location (NGDC, 2010). This difference
is less than 1 degree from the Penn Project magnetic north but is significant when we were
interested in nominal 1 m accuracy for our georeferencing.
Yet another correction was needed: true north to UTM grid north. In the UTM
projection system, the north-south lines on the map are parallel to each other and
perpendicular to the equator. Of course, the north-south lines should not be parallel since
these converge at the North Pole. The difference between the direction to true north and
the direction perpendicular to the equator is the convergence angle. We used the
convergence angle for UTM zone 16n, WGS84, for the Tikal area 047 (NGS 2010).
The diagram, Figure 4.5, shows how the various directions relate. The Penn Project
magnetic north is 6.76 east of UTM grid north.

Map Scale Ground Distance to UTM Grid Length

This georeferencing method requires yet another piece of data the map scale. Distances
on the ground, ground length, need to be converted to ellipsoid length and then to UTM
grid length. As with the difference between Penn Project north and magnetic north, this
length correction is small, but we were working toward 1 m accuracy. This correction
converts the 4,000 m ground length of the sides of the area covered by the TR11 Maps to
4,002.1 m. That is, the virtual control points marking the four corners of the Ruins Map
(Figure 4.2) are not 4,000 m apart as measured on the ground, but rather, 4,002.1 m apart
when mapped in UTM zone 16 north. Surveyors call this distance correction the
combination factor (Wolf and Ghilani 2006). The combination factor is 1.00052060 for
our average elevation above the ellipsoid and location within UTM zone 16n.
Using trigonometry and these three pieces of data (the point of beginning coordinates,
the reference direction, and the map scale), we created transformation equations to
calculate the locations of the corners of the maps, the virtual control points, in the UTM
coordinate system. There are twenty virtual control points, plus the point of beginning
(Figure 4.2).

Preparation of the Paper Maps for Georeferencing

The paper maps must be scanned into an electronic format to begin the process of
georeferencing. The University of Pennsylvania Museum (the copyright holder) scanned
the maps (at 300 dpi) as part of the process of making an electronic copy of TR11. We
separated the ten individual maps from the PDF format document and saved each map
individually in tiff format. The original paper maps use two colors (brown for contour
lines, black for ruins and all other features). The Penn scan changed these to gray scale to
reduce the file size. We removed the borders of the scanned maps by cropping. With the
borders removed, the TR11 Detail Maps can be georeferenced in GIS as a seamless
whole.

Georeferencing with Virtual Control Points

Using the georeferencing tool in ESRI ArcMap, the four corners of each of the TR11
Maps were snapped together, or transformed to, the appropriate virtual control points
(Figure 4.2). The Penn Project supplied fiducial marks in the corners of the TR11 Maps
(the marks are in a slightly different position from the corners of the grid printed on the
maps see, for example, the lower right corner of the map in Figure 4.1). Using the four
fiducial marks on each map, the georeferencing with affine transformation yields a
horizontal RMS (root mean square) error in the positions of the fiducial marks of 0.12 to
0.46 m for the TR11 Detail Maps and 0.92 m for the overview map. This is within our
goal of 1 m accuracy.

Accuracy Assessment of the TR11 Maps

How well did the Penn Project do at making the TR11 Maps? How do the TR11 Map
positions compare to those derived with a modern GPS? How much confidence should we
have when using the maps? To answer those questions we field-checked the accuracy of
the TR11 Detail Maps (Carr et al. 2011). During the 2010 field season of the University of
Cincinnati Project (the UC Project), we used a GPS receiver, Total Station survey
system, and the results of a radar altimetry mission to measure the horizontal and vertical
location of points throughout the 9 km2 area covered by the TR11 Detail Maps.

Horizontal Position Accuracy Method

To evaluate the horizontal position accuracy of the maps we compared the GPS and
mapped locations of ruins throughout the area of the TR11 Detail Maps. These map
accuracy test points must be clearly identifiable on both the ground and the map (Figure
4.3). A chultun (narrow-mouthed storage pit or cistern), assuming you can find it, is a
perfect test point as there is no doubt as to the location of its center. To determine the on-
the-ground coordinates of a map accuracy test point, the GPS was held stationary over the
test point and an average of 100 or more readings was made. While the GPS was
averaging, a sketch of the relevant feature and context was recorded. The sketch was
compared to the TR11 Detail Maps in the field to ensure that the correct point was
identified. On the basis of evaluation of more than 96 mapped features (Figure 4.2) we
find that the schematic method of representation of ruins (the Parris Convention) very
clearly shows the feature shape down to small indentations in the tops of many ruins or
to the smallest quarry scarp.
Figure 4.3 (left) shows two map accuracy test points plotted on the georeferenced map.
Chultun 5B-4 is noted on the upper center by the open circle symbol (in the Penn Project
numbering scheme, the 5B represents the map square and the 4 represents the item
number within that square). The dot to the left of the chultun is where GPS indicates the
chultun is located. Assuming the GPS is correct, the map is in error by a few meters.
Specifically, the chultun symbol on the map needs to be moved in a northwest direction
about 7 m. This is the error vector for this point. The center of range structure 5B-13 is
another map accuracy test point. The error vector for this point is roughly northeast and 6
m in length.
In all, 96 map accuracy test points (Figure 4.2) were collected to evaluate the horizontal
position accuracy of the TR11 Detail Maps. We used an opportunistic sampling plan to
collect points throughout the area of the TR11 Detail Maps. As noted by Dunning (1992),
it is very inefficient to use a random sampling plan in the Maya area because of the
difficulties in moving through the vine-filled tropical forest with its scarcity of roads and
trails. The opportunistic sampling plan does, unfortunately, limit the statistical analysis we
can do. Our points were collected first in the area where the UC Project was already
working. To cover other areas, existing park roads and trails were used for access. The last
100 or 200 m required bushwhacking through the forest to reach the test point location.

Horizontal Position Accuracy Results

We derived 96 error vectors from the 96 test point-map position pairs. The 96 mapped
positions are based on the SHORAN version 11 georeferencing. The errors in the 96 test
points are shown as error vectors and as a distance histogram in Figure 4.6.

Figure 4.6. The horizontal position accuracy of the TR11 Detail Maps. The left polar
plot shows the error vectors for the 96 map accuracy test points. The error vector
represents the direction and distance the map position has to move to match the GPS-
derived position (angle is from grid north; distance is in meters). The error vectors are
based on the SHORAN version 11 georeferenced positions of the maps. The right graph is
a histogram of the error vector lengths.
On the left plot in Figure 4.6 the vector line is not plotted. For clarity, only the vector
end point is shown. The error vectors can be imagined as a line from the center of the
circle to each of the 96 points. Looking at the distribution of the error vectors, there does
not seem to be a direction bias. In other words, all compass directions seem to be
represented about equally in error vectors. The spread of points around the center
represents a random error component.
The right plot, the histogram of the horizontal error, looks at just the length of the error
vector. Half the errors are less than 5.6 m. USGS rates map accuracy at the 90 percent
level. These data show 90 percent of points are within 11.2 mof their GPS reading. The
error measures are summarized in Table 4.1. At the map scale (1:2,000) the median error
of 5.6 m represents just a 3 mm error in plotted position.
Table 4.1. The summary of horizontal and vertical accuracy assessments. For the
96 horizontal accuracy test points the median error is 5.6 meters. For the 103 vertical
accuracy test points, the median error is 2.1 meters. The small area check with the Total
Station shows 0.4 meter median deviation.

Reference data Accuracy Assessment Results

Horizontal Vertical Vertical

GPS w/ WAAS differential Total NASA

correction Station AIRSAR
survey radar
system elevation

Number of test 96 24 103

points

Offset in 49.4 m 128.3 m 57.5 m

elevation datum

Minimum 0.3 m 0.0 m 0.0 m

deviation

Median deviation 5.6 m 0.4 m 2.1 m

Mean deviation 6.3 m 0.5 m 3.2 m

90 Percent 11.2 m 1.1 m 6.5 m

deviation

Maximum 15.7 m 1.4 m 15.4 m

deviation

Vertical Accuracy by Total Station and AIRSAR Method

To assess the vertical accuracy of the TR11 Detail Maps, we used land survey equipment,
a Total Station, and airborne radar altimetry. We extended the Total Station survey work
used on another part of the UC Project into a suitably long transect. The Total Station gave
very accurate vertical measures along the traverse, but in a very limited area. The main
components of our Total Station system were a Sokkia Set5 Total Station (theodolite with
laser range finder), a TDS Racon electronic data collector, and a CST target prism.
To get a sense of the overall vertical accuracy of the TR11 Detail Maps we used the
results of an AIRSAR, airborne synthetic aperture radar, mission flown by NASA
(National Aeronautics and Space Administration). The AIRSAR mission gave us vertical
readings over the entire area covered by the TR11 Detail Maps (AIRSAR 2008).
The left map in Figure 4.7 shows the path of the transect used to measure vertical
accuracy by Total Station. The transect starts at one of the UC Project excavations in the
field zone downstream of the Perdido Reservoir, passes through the reservoir and up a
Park road (formerly an ancient Maya quarry) to the ridge top just south of Temple IV.
From there, the transect turns southeast to follow a Park trail into the Mundo Perdido
plaza, and ends at a Penn Project benchmark (Figure 4.8). We made twenty-four elevation
readings with the Total Station along the transect. The distance along the transect is shown
and is used as the x-axis in the profile graph (Figure 4.8).

Figure 4.7. Maps of the areas used for the vertical accuracy assessment. The left map
shows the Total Station transect from the Perdido Reservoir area (lower left of map) up
and over to the Mundo Perdido plaza area, ending at a benchmark (the Perdido transect).
Distance along the transect is noted. The elevation from the Total Station will be
compared to the elevation from the contour lines on the map. The basemap is a seamless
combination of the detail maps Temple IV, Great Plaza, Corriental, and Perdido. Basemap
courtesy of the Penn Museum. The right map is the DEM from the NASA AIRSAR radar
altimetry mission with the locations of the TR11 maps superimposed. The triangles show
the locations of the 103 benchmark and spot elevations on the TR11 Detail Maps, which
were compared to the elevations from AIRSAR.
Figure 4.8. The elevation profiles along the Perdido transect. The profiles are from our
Total Station (solid line) and from the TR11 Detail Maps (dashed line). The vertical axes
of the profiles are offset to account for the different elevation references. The profiles are
in very close agreement in the first part in the virgin forest and along the Park service
road. The profiles diverge, but follow the same trend, in the tourist and plaza areas. The
insert photo is the Penn Project benchmark at the end of the transect, in Mundo Perdido
plaza. It appears the ground around the benchmark has been lowered 0.4 meter in the time
between when the benchmark was set ca. 1957 and our work, 2010.
The AIRSAR elevation data are from the mission flown March 7, 2004, and processed
by NASA as ts1876 (AIRSAR 2008). The mission recorded C-band radar images over
Tikal from two different directions. C-band radar reflects back from the first solid surface
it encounters in our case, the top of tree canopy or, in cleared areas, the ground or a
building. By overlaying the two images, and using interferometric methods, the elevation
of the first surface is determined. The elevations are presented as a DEM (digital
elevation model) of heights above the WGS84 ellipsoid. The AIRSAR processing
vertically aligned the DEM to the Shuttle Radar Topography Mission (SRTM) data at 9
points. The elevation errors in the alignment were x-bar 0.04 m, standard deviation 1.7 m.
In Figure 4.7 (right) the DEM light colors are the highest elevations and the dark colors
are the lowest elevations (the ridge top marking the location of the Great Plaza is clearly
visible). The inner nine squares mark the area covered by the TR11 Detail Maps. The
airport runway (see also Figure 4.2) is visible as a darker line in the northeast part of the
map (darker because the radar first return is from the runway surface, which is cleared of
vegetation, so is lower than the adjacent area). The dotted line in the southwest is the path
of the Total Station survey transect.

Vertical Accuracy by Total Station Results

Figure 4.8 shows the elevation profile along the Perdido Transect. The solid line is the
elevation profile plotted from the elevations of the 24 points along the Total Station
traverse. The dashed line is the profile taken from the contour lines on the TR11 Detail
Maps. The elevations use different vertical data so the Penn Project elevations were
shifted vertically relative to the Total Station elevations. The amount of shift was
calculated to minimize the overall difference between elevations in the first 800 m of the
transect, the area expected to be little disturbed by archaeological or Park operations in the
intervening 50+ years. The point-by-point difference between the profiles is small in the
areas expected to be little disturbed, the virgin forest and Park service road. The
difference between the profiles is greater along the tourist trail and in the Mundo Perdido
plaza area (where large-scale excavations and restoration have taken place). See Table 4.1
for a summary of the comparison between elevations from the TR11 Detail maps and from
the Total Station.
The inset photograph in Figure 4.8 shows the Penn Project benchmark in the Mundo
Perdido plaza (the benchmark is just southeast of Mundo Perdido, structure 5C-54; the
benchmark cap is stamped PYRAMID). The benchmark is marked on the Great Plaza
quad map with an elevation of 262.4 m. On the basis of the appearance of the benchmark
and an understanding of how benchmarks usually are made, it appears that about 0.4 m of
soil has been removed from the Plaza surface between the time James Hazard mapped the
benchmark in 1957 and when we remeasured it in 2010. The vertical accuracy graph
shows our Total Station elevation of the ground surface is 0.6 m below the Penn Project
elevation very close to the ground degradation noted at the benchmark. The Penn Project
surveying has just 0.2 m (0.6 m 0.4 m) cumulative elevation error over this 1,300 m long
transect.

Vertical Accuracy by AIRSAR Results

We assessed the elevation accuracy of the entire TR11 Detail Map set by comparing the
TR11 Map elevations to their corresponding AIRSAR-derived elevations at 103 test points
spread throughout the maps (Figure 4.7, right). The elevation comparison was made at
benchmarks and spot elevations on the TR11 Detail Maps. Elevations at all 25
benchmarks marked on the TR11 Detail Maps were used (e.g., triangles in Figure 4.5,
left). Seventy-eight spot elevations were used (e.g., X mark in Figure 4.3, left). All spot
elevations in outlying areas were used. We did not use all spots in the site core since there
were many spot elevations close together. The spot and benchmark elevations are given to
the nearest 0.1 m, so we used these rather than the less precise 1 m interval contour lines.
The linear fit (Figure 4.9) by ordinary least squares regression (OLS), between TR11
elevation and AIRSAR elevation, shows essentially a perfect fit between changes in TR11
Detail Map elevations and AIRSAR elevations. The slope of the regression line is 0.994.
A theoretically perfect fit would have a slope of 1 (1 m change in the Penn system equals
1 m change in the radar elevation). Further, the points are tightly grouped around the
regression line R2 = 0.96. The fit line is:
[AIRSAR elevation in ellipsoid datum] = 0.994 [Penn elevation in Penn datum] +
57.5 m.
Figure 4.9. Vertical accuracy assessment comparing 103 benchmark and spot elevations
from the TR11 Detail Maps to elevations from the NASA AIRSAR mission. Ordinary
least-squares regression shows very good agreement between the two data sets (slope
essentially 1, R-squared = 0.96). There is an offset between the two elevation sets
reflecting different datums and different surfaces measured (TR11 measured ground
surface; AIRSAR measures first return surface). The inset, from North Zone map, shows
the locations of a benchmark and a spot elevation used in the comparison. Basemap
courtesy of the Penn Museum.
The OLS-based equation tells us the offset between the two elevation measurement
sets: the 57.5 m intercept value. This offset has two potential explanations. First is a
difference in vertical datum. The second is a difference in the surface measured. To
understand the offset, we converted the elevations to a common vertical datum, the geoid
height datum commonalty called mean sea level (MSL).
The Penn Project used an arbitrary vertical reference which assigns a 200.0 m elevation
to the Datum Benchmark by the camp kitchen (Carr and Hazard 1961: 2; Figure 4.5, left).
On the other hand, the AIRSAR elevations are relative to the WGS84 ellipsoid (AIRSAR
2008). On the basis of GPS elevation readings at the Petty Benchmark we determined the
conversion between Penn Project elevations and mean sea level elevations. The GPS-
averaged elevation is 244.3 m (base of monument), relative to MSL. This compares
favorably to the elevation scaled from the Guatemalan government topographic map
(Tikal ca. 1990) of 245 meters above mean sea level. The elevation of the ground at the
base of the Petty Benchmark, in the TR11 datum, scaled from the Camp Quadrangle map,
is 194.9 m (Figure 4.5, left). Therefore, to convert from the TR11 elevation datum to
elevation above sea level, add 49.4 m (244.3 m 194.9 m).
To convert the AIRSAR elevations to MSL we used the geoid separation between the
WGS84 ellipsoid and MSL. The Geoid Height Calculator (UNAVCO 2012) gives the
geoid separation as -4.9 m for the Tikal area. Converting both the AIRSAR and the Penn
elevations to the MSL datum gives the new OLS equation:
[AIRSAR elevation in MSL datum] = 0.994 [Penn elevation in MSL datum] + 13.3 m.
This equation shows the AIRSAR elevations are 13.3 m above the Penn elevations when
both are expressed relative to mean sea level. This difference will, largely, represent the
difference in the surface measured ground for the Penn elevations and first surface for
AIRSAR. In other words, the average height of objects (trees, buildings) above the surface
in the area of the TR11 Detail Maps is 13.3 m. For the mostly tree covered area, this
seems reasonable.
Further, the OLS-based equation was used to convert the 103 TR11 elevations to the
AIRSAR vertical reference. The differences between the elevations were calculated. The
median error is 2.1 m. Some of this error is due to a difference in surface measured:
ground surface for TR11 versus first return for AIRSAR. See Table 4.1 for additional
measures of accuracy.
Overall, the excellent agreement between the 1960s era TR11 Detail Map elevations
and the 2004 radar elevation indicates that the TR11 Detail Map elevations are accurate.
The TR11 Detail Maps have the advantage over the AIRSAR data that they are ground
measurements and contain the locations of ancient Maya ruins. The AIRSAR data have
the advantage that they cover a much larger area. To date, airborne radar imaging has not
been able to detect ruins under tree canopy, but airborne Lidar, a laser-based imaging
technique, is showing promise at finding ruins under tree canopy (Chase et al. 2012).

Summary
With the TR11 Maps in electronic format, and their accuracy defined, the wealth of
information they contain on topography, architecture, and site planning can be used to
address current questions. The Penn Project data contained in the maps and associated
excavation reports represent a valuable resource that would be very difficult to duplicate
today for cost reasons and because of current limits on access and vegetation clearing in
this Guatemalan National Park. Electronic copies of the georeferenced maps are available
for download at the Digital Archaeological Record, tDAR (Carr 2013).

Potential Uses of the Data

On the most basic level, in their original paper form, these maps are large, delicate, and
unwieldy (and out of print). Digitizing and georeferencing the TR11 Maps into electronic
form increase their accessibility. The Penn Project mapped the location of 2120 structures
(Carr and Hazard 1961: 10). These locations can be converted into a spatial database and
automated analysis conducted (for example, to analyze structure orientation or changes in
structure density with distance from the site center). The digitized contour lines can be
electronically analyzed to identify drainage lines and watershed boundaries (see chapter
by Weaver et al., this volume). The contour lines can be converted into a digital elevation
model (DEM). With the DEM and the structures, it is possible to create a virtual fly-
through of Tikal. The fly-through offers an overall impression of how the ancient Maya
positioned their structures and open spaces within the landscape.
The georeferenced TR11 Maps can be used with GPS locations of current research
(e.g., Tankersley et al. 2011; Scarborough et al. 2012; the research reported in this volume)
to place the current efforts within an overall spatial context. The TR11 Maps can be
loaded into a GPS and used to guide a researcher to the location of past excavations, such
as the Penn Project work on small structures in the periphery of Tikal, such as TR19
(Haviland 1985), Excavations in Small Residential Groups of Tikal. The TR11 maps can
be combined with satellite imagery and correlations examined between ancient Maya
structures and tree canopy spectral signatures. Or, GPS tracks of the current network of
tourist trails in the Tikal site center can be overlain on the TR11 Maps as a foundation for
updated tourist maps of Tikal National Park. The technology available for mapping today
has advanced considerably, but the level of care and high standards used to produce the
TR11 maps provide valuable guidance for future mapping efforts in the Maya Lowlands.

Conclusions
The Penn Project TR11 Maps are very accurate both horizontally and vertically. The maps
are accurate because the project focused on accuracy the directors of the Penn Project
funded high-quality crews with the proper skills using the best practices and had a clear
vision for the desired result. The maps are close to complete the Penn Project relocated
and mapped nearly all surface ruins within the surveyed area using the schematic method
of representation. These accurate TR11 Maps are now georeferenced to a global
coordinate system and their wealth of knowledge is available in electronic format.
 5 Examining Landscape Modifications for Water
Management at Tikal Using Three-Dimensional
Modeling with ArcGIS
Eric Weaver, Christopher Carr, Nicholas P. Dunning, Lee Florea and Vernon L.
Scarborough

The karst terrain of Tikal, Petn, Guatemala, limits the amount of water that remains on
the surface from the highly seasonal precipitation that falls on the surrounding tropical
forest. This scarcity of surface water was a constant struggle for the Maya of Tikal, as it
was for most of the Maya Lowlands. Despite this obstacle the city-state managed to
flourish for more than a millennium, while using Stone Age equipment and without the aid
of beasts of burden. The influence that this scarcity held on the Maya is evident in every
aspect of their life religious, political, and household driven by the constant awareness
of the limited nature of this precious resource and the need to obtain and conserve it (Faust
2001). One of the creation stories of the ancient Maya describes a great ocean that existed
before the world formed on the back of a crocodile or turtle. In this story, the first humans
First Mother and First Father emerged from a resurgent spring at the mouth of a cave,
cheen, at the base of a water mountain, or witz (Tedlock 1985). It has been suggested that
many elements of Maya architecture are representative of the creation story and acted as a
constant reminder to the commoners that the king was a direct descendant of the First
Mother and Father (Dunning et al. 1999). This connection is strongly evidenced by the
duality of architectural terms among the Maya such as nab, meaning a plaza or a pool of
water; witz, meaning hill or pyramid; and cheen, meaning a city center or a water cave.
This use of sacred architecture can be described as a form of geomancy by which the built
environment is carefully tuned to a cosmic model (Moyes 2006).
While the architecture of the Maya served political and religious purposes, it would be
an error to ignore a primary and utilitarian purpose of these structures manipulating and
containing the flow of water. Like the terraformers often described in current works of
science fiction, using their understanding of the hydrological cycle the ancient Maya
accretionarily modified the surrounding inhospitable landscape to create a more life-
supportive environment (Scarborough 1998). Those whose architecture was most
connected with the key water sources were also most likely to be the earliest arrivals at the
site; in this sense, perhaps the first settlers are very similar in characteristics to the First
Mother and Father of the creation story (Dunning et al. 1999). The hydrological cycle of
the Maya begins with the belief that rain clouds formed in hollowed mountains and were
released to the sky out of caves (Groark 1997); both of these elements are commonly
found at Maya sites manifested as pyramids and caves, some of which were natural and
others artificially constructed. Causeways that cut across the landscape are used to channel
and disperse water in a similar fashion as stream channels. Reservoirs, aguadas, and
chultuns acted as lakes, ponds, and sinkholes. The addition of plaster on these structures
created a surface far more impervious than the surrounding land.
Through intentional and unintentional efforts the Maya modified the landscape in ways
that manipulated the movement and containment of water. The most obvious types of
intentional modifications were through reservoirs, aguadas, and chultuns (Dunning et al.
1999). Some of these features may have originated as quarries during the construction of
the city. It is entirely plausible that the Palace, Temple, and Hidden Reservoirs are largely
a result of the construction activity in the Great Plaza, which includes two pyramid
structures (Scarborough and Grazioso, this volume). The impervious surfaces of central
Tikal had the advantage of repelling water and forcing it to be channeled to desired
locations throughout the site; however, they may have also prevented the recharge of
springs that could have helped to replenish the reservoirs with fresh water (Scarborough et
al. 2012).
Despite efforts to restore portions of Tikal, most of the city remains in ruin and covered
by the tropical forest. The reservoirs, no longer maintained, cease to hold the water that
once supplied the great city. Standing on the site today it is difficult to visualize in total the
dynamic hydraulic processes that once occurred. The lack of perennial streams makes this
effort even more difficult in the dry season, when the landscape and its arroyos lie
parched. Aerial and satellite images currently available for the area are not yet at the
quality to provide anything more than faint clues about the surface hydrology of the area
(Thomas 2010). Fortunately, hydrologic models of the site can be developed using the
available data through a Geographic Information System (GIS).
The initial step in creating a hydraulic model for Tikal was to digitize and georeference
the paper University of Pennsylvania Tikal Project maps (Carr and Hazard 1961) using
ArcGIS. The Penn Project began the mapping of greater Tikal in 1957 and completed the
survey in 1960. Fortunately, lead surveyors Robert Carr and James Hazard had the goal of
producing a high-quality map to help facilitate the long-term planning of archaeological
research at Tikal (Carr et al., this volume). The final map included 9 km2 of surveyed
territory, drawn with 1-m contour intervals and a 1:2,000 scale. An additional 7 km2 of
greater Tikal used 5-m contour intervals (Carr and Hazard 1961). Approximately 2,120
structures were mapped in the survey and provide a representation of Central Tikal in the
Late Classic period (Coe 1962).
A portion of our fieldwork in 2009 and 2010 included Total Station surveys of areas
previously surveyed by the Penn Project (Corriental Reservoir and Perdido Reservoir).
Additionally, in 2010 we located some of the structures that Carr and Hazard had
identified using coordinates derived from the georeferenced maps. On the basis of these
observations it was assessed that the location of these structures maintained a horizontal
positional accuracy with a median of +/5.6 m (Carr et al., this volume). In terms of this
evidence, we concluded that the survey quality of the Tikal map is essentially comparable
with what is possible using todays survey equipment. While an all-inclusive survey using
a Total Station would increase the accuracy of the map, it would incur a substantial
expense and be extremely destructive to the natural environment with minimal yield of
new data. Until LiDAR is introduced into the Tikal data set, it is almost certain that the
Carr and Hazard map will remain the best source for topographic information at the site.
On the basis of the determination of the essential accuracy of the Penn Project survey
and map it was possible to create a hydraulic model based on the digitized and
georeferenced maps. The process of creating a hydraulic model of Tikal began with the
creation of a Digital Elevation Model (DEM). To help understand some of the
modifications that were made at Tikal, two models were created. The first model,
representing the Late Classic Maya or the cultural landscape, was made by utilizing the
digitized 1-m contours of the Penn Project map and overlaying structures. The second
model, the pre-Maya model or natural landscape, took the digitized map of the first model
and extrapolated from this what the site may have been like before the erection of
structures. Essentially, this meant reconnecting elevation lines that had been broken by the
addition of buildings and continuing the elevation lines in a typical incline for a hill. An
interpolation method, based on the digitized contours, was used to create a hydrologically
correct integer DEM. Sinks are defined in this chapter as errors in data during the
production of a DEM that result in artificial depressions. Although in DEM preprocessing
it is common practice to fill these in, it was not necessary to do so in this case because this
issue is particular to satellite imagery and does not with the exception of typos occur
with land-based surveys. A Triangulated Irregular Network (TIN) was developed to
incorporate the additional features of the site into the DEM. Aguadas were added to the
terrain of this model on the basis of Gallopins thesis, as significantly more aguadas (98)
were documented in this effort than in the original Penn Project survey (Gallopin 1990;
Scarborough and Gallopin 1991). All of the structures documented by the Penn Project
were added to the Late Classic Maya model. Since it was not possible to determine
elevation heights for each structure on the basis of the map, we decided to incorporate a
height for all structures sufficient to act as a barrier to water. These buildings were then
burned into the DEM in order to create a barrier to inhibit water flow. The addition of
structures in the hydraulic model of Tikal, or any Maya city, is critical because almost
every element of the architecture served to manipulate the flow of water (Coe 1962).
Plazas were added to sizable household groups as well as in Central Tikal in order to
flatten out these locations. Causeways, significant sources of water movement in Central
Tikal, were added to the map and were designed to allow water to flow through. Quarries
and the 197 documented chultuns, while not included, would be a beneficial future
addition to this model.
The hydraulic model used in this analysis of the Tikal terrain is based on the ArcHydro
model, which is implemented in the third-party extension to ArcGISArcHydro
(Maidment 2002). ArcHydro is able to determine flow direction, flow accumulation,
drainage lines, and catchment areas and to delineate watersheds on the bass of the
elevation derived from the DEM. Because of the karstic landscape, Tikal must be treated
in ArcHydro as having a deranged drainage system. A deranged drainage system differs
from the more common dendritic system in that it has natural locations where stream flow
is prohibited from advancing farther until filled. In a karst setting this would occur at
sinkholes. Additionally, for Tikal this would also include reservoirs and aguadas (Figure
5.1). With a few exceptions, default parameters were used in the creation of these models.
As with the DEM, the common process of filling in sinks was avoided. A general guide
determining stream definition is to base it on 1 percent of the maximum flow
accumulation (Maidment 2002). This model utilized the more-developed USGS Elevation
Derivatives for National Applications (EDNA), which indicates the threshold to be
5,000*30*30 m cells for catchment definition. By constructing the hydrologic model of
Tikal, a further understanding of the distribution of water can be developed. Watersheds
for the area were delimited and the water flow routes were determined. These routes
provide clues as to why the Maya would have chosen to develop and settle where they did.
These routes can further explain the choice of location for chultuns and aguadas and may
even be able to suggest locations of previously missed aguadas (Figure 5.2). The model
suggests that certain structures may have been placed strategically to manipulate the flow
of water, by either redirecting or inhibiting its movement.

Figure 5.1. Many structures appear to be placed to block the flow of water and form
aguadas (an example of which can be seen in the upper right-hand corner). Two potential
locations for undocumented aguadas are marked by an X.
Figure 5.2. Watershed for Temple, Palace, and Hidden Reservoirs demonstrating how
the Maya used structures to assist in managing the flow of water. Watershed was
delineated with ArcHydro software that utilized a digitized topographic map where the
structures were added to act as barricades and plazas flattened the surface area.
The same data used for creating a hydraulic model can be used in ArcScene, an ArcGIS
extension, to create a three-dimensional model of the landscape.
(www.cambridge.org/Tikal; See Video 1) While three-dimensional data can be used in a
two-dimensional interface, a thorough comprehension is more easily derived by
visualizing these data in a three-dimensional environment. This is particularly true with
exploratory research. In the three-dimensional model, more aptly described as a 2.5-
dimensional model, each structure was digitized individually. Residential structures were
also digitized as points, which were placed at each structures center. A three-dimensional
symbol was designed to represent the structures using Google SketchUp. The symbol size
was increased on the basis of the size of the area of the structure. The temples were built
as polygons on the basis of their height. A 30-m resolution IKONOS satellite image was
then used as a base image to drape over the landscape. This image was chosen for
demonstration purposes, primarily for aesthetic and locational reasons. In the future, other
images could be used and provide much more information, for example, a Normalized
Difference Vegetation Index (NDVI) image of the area.
A digital model of the landscape of Tikal, while an impressive visual tool for
exploratory research, is capable of being much more. Within the constructed model is the
built-in capability of a continuously growing database. A structure that may begin as a
point in the model may later be further delineated as structure types, such as house
mounds or palaces. A long-term goal would to be to add temporal elements to the database
capable of displaying the chronology of the construction of Tikal and how this
modification affected the hydrology over time. Structures can also be built into polygons
where they would provide information specific to their footprint. These topologies, point
and polygon, provide the capability to capitalize on the powerful spatial analysis tools of
ArcGIS. Catchment and watershed areas provide an interesting addition to options for
defining borders for analytic purposes.
 A primary interest behind developing the database was that it could be used as one step
toward the development of a hydrologic budget for Tikal. Specifically, such a model
would examine the hydrology from a mass balance approach all water gained and lost
through natural and anthropogenic pathways over a period of time (Arnold and Allen
1996). Early modeling attempts (Gallopin 1990) suggest that a series of three reservoirs in
Central Tikal (Temple, Palace, and Hidden) were filled to capacity and occasionally
overflowed. Ground observations support this hypothesis (Scarborough et al. 2012) and
include the use of shunts to bypass earthen dams. However, the Gallopin and similar
models make several key assumptions for values of precipitation and evaporation. Recent
daily climate data at Tikal, accumulated during the past decade, paves the way for a more
robust modeling endeavor using a linked-reservoir approach in which both the volume and
the throughput in each reservoir are variable through time and are computed using a series
of mass-balance equations. Changes in mass are manifested as changes in reservoir
volume. The interactions between reservoirs are connected using linking equations. Once
developed and tested, this linked-reservoir model can be combined with Monte Carlo
simulations of realistically distributed, but randomized precipitation and evaporation data.
Such an approach can rigorously evaluate the function of Maya reservoirs through
changing climate regimens.
Current GIS technology makes hydraulic modeling of ancient sites possible. The
quality and capability of this analysis are limited only by the available data, a situation
that will vary from site to site. Tikal is an excellent site to explore the capabilities of what
can be produced with hydrologic modeling, primarily because of the high quality of the
site survey conducted by the Penn Project. The development of comparative models that
reflect the natural landscape and the evolved cultural landscape provides the opportunity
to examine the contrast between the two in order to understand how and why the
landscape has been altered. It is also paramount for our own future to understand why a
civilization like the Maya, so accustomed to struggling to maintain their rare and precious
resources, eventually were no longer able to maintain this balance. Further advancement
of the hydrologic budget and an understanding of the climate differences between Modern
and Late Classic Tikal can help shed light on the details of ancient Maya water
management and implications for the sustainable reoccupation of the Maya Lowlands.
 6 Life on the Edge: Tikal in a Bajo Landscape
Nicholas P. Dunning, Robert E. Griffin, John G. Jones, Richard E. Terry, Zachary Larsen
and Christopher Carr

Introduction
Anyone who tries to cross the Petn on foot, including early Spanish entradas (Means
1917: 108), quickly becomes aware of the widespread presence of bajos and their dense,
inhospitable vegetation.1 Geographer Karl Sapper (1896) is the first scholar to report the
preponderance of bajos, karst depressions that are typically inundated during the rainy
season and desiccate during the dry season, in Guatemalas Petn District and adjacent
areas of Mexico and Belize. Many of the largest ancient Maya cities in this region are
situated near the edges of large bajos, an aspect of the regional settlement patterns that has
long intrigued scholars because of the perception of bajos as problematic, resource-poor
environments (Lundell 1937; Morley 1937; Bullard 1960). The term bajo is used widely
in the Maya Lowlands to refer to any low-lying, wet terrain. Here, we are specifically
dealing with upland or interior bajos, those lying within the elevated interior region of
the Yucatan Peninsula at elevation of 80 m or higher (Dunning et al. 2012). Over the
years, numerous scholars have proposed solutions to the bajo conundrum, suggesting
variously that bajos had once been lakes or perennial (not seasonal) wetlands (Cooke
1931; Ricketson 1937; Harrison 1977; Jacobs 1995; Dunning et al. 2002), are more
environmentally diverse and potentially productive than commonly realized (Culbert et al.
1990; Kunen et al. 2000), or were heavily modified by human engineering to make them
agriculturally productive (Adams 1980; Harrison 1990). A common thread to these diverse
studies is that there was something about the bajos that made them considerably more
attractive as resources and settlement locations in the past than is readily apparent today. A
common problem with many of these studies has been a tendency to generalize findings
from one bajo or set of bajos to the great majority of bajos and their potential role as
important locations in the development (or decline) of Maya civilization. However, as
more and more investigations have been made into both the current and past environments
found within bajos, it has become clear that while some shared trajectories can be found,
each depression has to a significant extent its own unique environmental history (Dunning
et al. 2006). We, thus, expected the Bajo de Santa Fe and smaller depressions around Tikal
to hold their own stories.
The largest bajos in the northeastern Petn (e.g., Azcar, Santa Fe, and La Justa)
occupy structural troughs (grabens) along SW-NE trending faults (Dunning et al. 1998).
Tikal lies near the southwestern margin of the Bajo de Santa Fe. In addition to these
sprawling depressions, each of which covers hundreds of square kilometers, numerous
much smaller bajos pockmark more elevated karst surfaces throughout the region. These
smaller features range in size from a few hundred square meters to a few tens of square
kilometers. We use the term pocket bajo to describe localized depressions of less than 2
km2, of which there are numerous examples in the greater Tikal urban area. These smaller
bajos likely originate with limestone dissolution and collapse, followed by infilling with
fluvial, colluvial, and eolian sediments. We conducted field studies in two such features on
the southern flank of Tikal: Perdido Pocket Bajo and Arroyo Corriental Pocket Bajo (see
Figure 1.3). Additionally, we conducted field investigations in the Bajo de Santa Fe.
The goals of our investigations included (1) establishing a better understanding of the
environmental history of the Bajo de Santa Fe and upland pocket bajos within the greater
Tikal urban zone, (2) documenting the nature of ancient Maya land use within bajos, and
(3) examining the relationship between ancient water management features and bajo lands.
Before detailing some of our recent work, a brief review of previous work in the Bajo de
Santa Fe and other Tikal bajos is warranted.

The Bajo de Santa Fe and Previous Investigations

The Bajo de Santa Fe is a sprawling, convoluted depression covering approximately 480
km2. This bajo lies east and northeast of Tikal and is part of the Rio Holmul drainage
(Figure 1.1). The bajo includes a complex array of ecosystems including islands of upland
forest, a variety of palm-dominated forests (typically transitional between the uplands and
lower lying zones), large areas of swamp scrub forest (Tintal or Arbusto forest), and areas
of pine and palmetto savanna (see Chapter 1, this volume, and Lundell 1937, 1961;
Brokaw and Mallory 1993; and Kunen et al. 2000 for descriptions of forest types).
Remains of ancient Maya settlement extend around much of the western margins of the
Bajo de Santa Fe, and the urban zone of Tikal extends to the edges of a southwest arm of
the bajo about 2 km east of Temple IV. The floor of the bajo there lies some 70 m lower in
elevation than the Great Plaza. Dennis Pulestons (1967, 1973, 1983) settlement survey
followed the East Brecha of Tikal National Park, extending partway into the southern
portion of the Bajo de Santa Fe. In both 2009 and 2010, our investigations centered on
portions of the East Brecha and areas to its north (Figure 6.1). Other investigations have
preceded ours in the bajo.

Figure 6.1. Map of a portion of the East Brecha showing aguadas and excavations.
 The soils of the Bajo de Santa Fe were mapped as belonging to the Yaloch Series by
Simmons, et al. (1959): slightly acidic, heavy clay soils with prominent A horizons,
negligible B horizons, and deep C horizons. However, it is unclear whether any sampling
took place in the Bajo de Santa Fe or whether it was mapped by extrapolation from other
bajos. Laws (1961) examined soil samples associated with Tintal and pine savannas in the
bajo and concluded that both were dominated by montmorillonite clay, which contributed
to notable shrinking and swelling. Cowgill (Cowgill and Hutchinson 1963) dug a 511-cm-
deep soil pit in the Bajo de Santa Fe near the Aguada de Terminos east of Tikal and did
extensive laboratory analyses of soil samples (to be discussed further later).
Olson (1977) examined soils in portions of the Bajo de Santa Fe and classified those
lying in interior areas underlying Tintal or Scrub forest as Vertisols, noting their high
montmorillonite clay content, shrink-swell behavior, and tendency to develop gilgai.
Following these observations, Puleston (1973, 1977) cautioned against interpreting
patterned ground within the Bajo de Santa Fe and other wetlands as clear evidence of
wetland cultivation by the Maya. Some ephemeral linear patterns were observed from the
air in the Bajo de Santa Fe in the 1970s, but were not captured on film (Harrison 1978;
Dahlin et al. 1980). Olson also noted that soils along the periphery of bajos had deep,
cumulic fertile A horizons, classifying these as Mollisols (Argiudolls), usually mantled in
palm-dominated forest. Shallow Mollisols (Rendolls), the erosion of which was the source
of the accumulating A horizons downslope, are found on sloping terrain surrounding bajos
as well as on islands within larger bajos.
Adams (1980) reported patterning within the Bajo de Santa Fe (among many other
bajos) on the basis of interpretation of Synthetic Aperture Radar (SAR) data, though no
supporting imagery was provided. Later analyses suggested that the SAR study was
seriously flawed by the poor resolution of the imagery, which could not have detected
many of the features claimed to have been observed (Pope and Dahlin 1989, 1993).
Sever and Irwin (2003) documented several linear features in the Bajo de Santa Fe
using a combination of higher-resolution remotely sensed imagery. However, two of these
features were subsequently found to be lines created in an IKONOS image when three
partial scenes were improperly stitched together by technicians at IKONOS (a procedure
done without the knowledge of the personnel at NASA who ordered the scene). One
linear feature was visited on the ground using GPS guidance, but its nature remains
unclear perhaps attributable to either a buried causeway or a completely aggraded
channel now manifest only as a change in vegetation along its course (Tom Sever,
personal communication, 2009).
Weller (2006) reported additional linear features in the Bajo de Santa Fe on the basis of
analysis of a combination of imagery types, as well as documenting numerous bajo islands
with evidence of ancient settlement clusters. Notably, some of the linear features appear to
connect bajo islands with stream channels. We also noted several of these features on
satellite imagery in remote areas of the bajo, but were not able to access these on the
ground. This association of linear features, river channels, and bajo islands is analogous to
linear features excavated in 2007 and 2008 in the Bajo de Azcar north of San Bartolo
which are likely a combination of natural stream channels and shallow canals excavated to
extend the length of these channels (Dunning and Griffin 2008).
 Fialko (2000, 2005) conducted multiyear ground survey and excavations in parts of the
Rio Holmul drainage basin, which encompasses much of the Tikal urban area and portions
of the Bajo de Santa Fe. The uppermost reaches of the Rio Holmul and its tributaries south
of Tikal were found to be composed of seasonal channels of various sizes, some of which
were discontinuous, punctuated by karst sinkholes. Within the Bajo de Santa Fe, the Rio
Holmul becomes more substantial, though still highly seasonal. Deeper pools at various
points along the river retain water throughout most dry seasons. For the most part, surface
drainage around and within the Bajo de Santa Fe is weakly developed and poorly
organized, reflecting the preponderance of internal, karst-controlled drainage, generally
low slope gradients, and deep sediment accumulation within the bajo.

Investigations in the Bajo de Santa Fe: 20092010

Our investigations in the Bajo de Santa Fe took place largely along Tikal National Parks
East Brecha, taking advantage of the topographic and settlement mapping done decades
ago by Dennis Puleston (1973). This brecha enters the Bajo de Santa Fe some 2 km east of
Temple IV and crosses several swaths or embayments of the bajo floor as well as three
peninsulas of higher ground that protrude into the bajo (Figure 6.1). In order to utilize
Pulestons work better, his East Transect map was digitized before fieldwork began. As
part of our work, numerous hand-held GPS points were taken at visible features along the
transect to georectify the maps and integrate these data into a project GIS along with the
Penn Project maps of central Tikal and several types of satellite imagery (see Carr et al.,
this volume). This process included corrections of measurement errors on the original
maps, though these proved remarkably accurate given the informal manner in which they
were made.
Time and resources did not allow us to excavate a deep soil pit in the floor of the Bajo
de Santa Fe; however, since some of our own investigations took place in the general
vicinity of the Cowgill and Hutchinson (1963) pit, we can reasonably incorporate their
data into our assessment of the environmental history of that portion of the bajo nearest
Tikal. This 511-cm-deep pit was first excavated under the supervision of Ursula Cowgill
in 1959 several hundred meters NW of the Aguada de Terminos. Unfortunately, we were
unable to relocate the exact location (Figure 6.1). The pit was partially reopened in 1962
in order to obtain additional soil samples. The soil revealed in the pit is a deep Vertisol
(Figure 6.2), with pronounced evidence of prolonged shrink-swell activity
(argilloturbation). Examination of the pits stratigraphy, in particular the lack of any
discernible alluvial or lacustrine deposits, led these investigators to conclude that the deep
clays that dominated the bajo soil and underlying sediment in this location were produced
by mass wasting of the limestone bedrock composing the uplands surrounding the bajo.
However, it must be kept in mind that this study amounted to a point sample; thus the lack
of alluvial or lacustrine deposits at this one place is significant only at this location, a
limitation noted by Cowgill and Hutchinson. As we are reporting elsewhere, there is
growing evidence that a significant portion of the smectite clays composing much of the
soil in the Petn may be ultimately derived from eolian volcanic ash (Tankersley et al.
2011; Tankersley et al., this volume).
Figure 6.2. Soil pit in the Bajo de Santa Fe west of the Aguada de Terminos (based on
Cowgill and Hutchinson 1963 Figures 2 and 3).
Comparatively little attention was paid to a layer of carbonaceous material of
unknown nature and origin found at a depth of about 5 m (Cowgill and Hutchinson 1963:
12). This layer yielded a radiocarbon date of 11,560 360 BP and may be analogous to
related late Pleistocene soil surfaces that have been found in several other bajos (Dunning
et al. 2006).
Cowgill and Hutchinsons analysis of the statigraphy and clays led them to believe that
these were produced by long-term mass wasting on the surrounding uplands and showed
negligible impact from Maya activities. Our own excavations targeted footslope areas
along the bajo margins, such as Op. 5C, just east of the Aguada de Terminos (Figure 6.3).
Not surprisingly, such locations rather dramatically show the effects of ancient forest
clearance and agriculture. Op. 5C revealed a buried well-developed ancient soil surface
(Ab horizon) lying below 50 cm of colluvium eroded from the ridge on which Grupo de
Terminos is located (Figure 6.4a). Radiocarbon dating of organic matter within this buried
soil produced a date of 2850 40 BP, indicating that this surface formed gradually during
pre-Maya and Preclassic times, a finding consistent with buried soils found in comparable
contexts throughout the Petn and northern Belize (Dunning and Beach 2000; Beach et al.
2008). The colluvial sediment burying the soil contained numerous sherds, of which those
few that were chronologically diagnostic were all Late Classic types. Excavations were
also made in a similar footslope position (Op. 10A) between kilometers 7 and 8 of the East
Brecha where it crosses a steeply elevated peninsular ridge that protrudes northward into
the Bajo de Santa Fe. Much of this ridge is densely covered in ancient settlement remains.
We refer to this ridge as Pulestons Peninsula. Not surprisingly, these excavations also
revealed an early soil surface buried under anthropogenic colluvium.

Figure 6.3. Map of the Aguada de Terminos area.

Figure 6.4. a) North profile of Op. 5C on a footslope at the edge of the Bajo de Santa Fe
near Aguada de Terminos. A: Black (10YR2/1) clay; large hard crumbs; scattered sherds;
AC: Very dark gray (10YR3/1) clay; subangular blocks; C: Dark gray (2.5Y4/1) sandy
clay [20 percent coarse sand and limestone gravel]; slickensides; scattered sherds; 2Ab:
Black (Gley 2.5/N) clay; massive; slickensides; 2ACb: Light brownish gray (2.5Y6/1)
sandy clay; massive; slickensides; 2Cb: Pale yellow (2.5Y8/2) clay in matrix of weathered
limestone.
b) North profile of Op. 2B on a footslope at the edge of Arroyo Corriental Pocket Bajo. A:
Black (10YR2/1) clay; large hard crumbs; scattered sherds; AC: Very dark gray
(10YR3/1) clay; massive; C: Very dark grayish brown (10YR3/2) sandy clay; sherds; 2Ab:
Very dark brown (10YR2/2) clay; massive; 2ACb Dark gray (10YR4/1) clay; massive;
15 percent rock fragments.
In 2001, geoarchaeological excavations were undertaken in another part of the Bajo de
Santa Fe under the aegis of another project (Dunning et al. 2002). In 2000, settlement
pattern reconnaissance led by Vilma Fialko (2005) along the peripheries of Tikal National
Park encountered areas of intriguing surface patterning in a pine/palmetto savanna area in
the Bajo de Santa Fe near the NE corner of the park, an area removed from upland areas
except for a nearby large bajo island dubbed simply La Isla. In 2001, trenches and pits
were dug in two sections of El Pinal and test excavations were made in settlement
remains found on La Isla. The trenches clearly revealed that the ephemeral surface
patterning was the product of argilloturbation (clay heaving) and gypsum accumulation
(gyp heave) and not the result of human engineering (a process known from other
wetlands in the Maya Lowlands and elsewhere in the world; e.g., Paton 1973; Jacob 1995;
Pope et al. 1996). Intriguingly, the trenches and pits (including one pit excavated to a
depth of more than 4 m) in one area, El Pinal Parcela 1, failed to produce any Ab horizons,
whereas trenches in the other section of savanna about 1 km distant (El Pinal Parcel 2 or
Pinal Central) revealed a clear Ab horizon somewhat distorted by clay heaving and
gypsum swelling (e.g., Op. 16: Figure 6.5). Parcela 2 is lower in elevation than the first, a
characteristic that may help account for the presence of the Ab horizon in only this area
and not the other, though slope gradients are very low across this entire section of the bajo.
Organic matter within the Ab produced a radiocarbon date of 2550 40 BP. Soils
throughout this area of the bajo are leached, acidic, and infertile (low CEC), making
agriculture here unlikely. We tentatively classified these soils as Dystraquepts (Soil Survey
Staff 2006). It is distinctly possible that these pine savannas were a significant source of
fuel wood for ancient Tikal (see Lentz et al., this volume).

Figure 6.5. Trench profile (Op. 16) in Parcela 2, El Pinal, Bajo de Santa Fe (after
Dunning et al. 2002).
In sum, evidence accumulated to date has produced no evidence for the former
existence of lakes or perennial wetlands in the Bajo de Santa Fe. However, it should be
noted that only a few isolated places in this sprawling, complex wetland system have been
subject to geoarchaeological investigation, and it remains possible that wetter conditions
once existed in parts of the Bajo de Santa Fe as they did in some other bajos (Dunning et
al. 2006). We did uncover evidence of anthropogenic soil erosion and associated
colluviation on the margins of the bajo nearest to Tikal. Comparable processes have been
documented in many other bajos and linked to the later distribution of palm forests along
the edges of these bajos an association that likely holds true for the Bajo de Santa Fe as
well.
Aguada Investigations in the Bajo de Santa Fe
Aguadas (ponds) are a widespread feature across many parts of the Maya Lowlands,
including the NE Petn. The majority of these features are believed to have originated as
natural karst sinkholes, plugged with accumulated clay sediments, though often modified
by the Maya to enhance water storage (Siemens 1978; Akpinar et al. 2012). During the
2009 and 2010 seasons, a total of eight aguadas were investigated to varying extents along
the East Brecha and adjacent areas of the Bajo de Santa Fe (Figure 6.1). Only surface
observations were made at four aguadas: Aguada Benito, Aguada de Carlos, Aguada La
Sarteneja, and Aguada La Presa. The most notable cultural feature observed was an
interior stone wall and attached pier in Aguada de Carlos. Sediment cores were extracted
from three: Aguada Pucte, Aguada de Terminos, and Aguada Vaca del Monte. Excavations
were conducted at two: Aguada de Terminos and Aguada de Elmer.
Sediment cores were obtained with a modified Livingston Corer from near the margins
of Aguada Pucte and Aguada de Terminos in April 2009, when both of these aguadas
contained more than a meter depth of water. In March 2010, cores were obtained from
Aguada Vaca del Monte, which had no standing water, but surface sediments remained
saturated. Coring was attempted in Aguada La Sartenaja, which contained standing water,
but rock was encountered at shallow depths within the aguada floor in all places that were
probed.
The Aguada de Terminos is situated along the southern flanking arm of the Bajo de
Santa Fe that extends farthest westward toward the Tikal site center (Figure 6.1). The floor
of the bajo lies north and west of the aguada; terrain rises gradually to the south and more
steeply to the east onto a peninsular ridge that protrudes northward into the bajo. In April
2009, the aguada contained about 1.5 m of water. We extracted a sediment core from near
the northwest corner of the aguada. In March 2010, the aguada was completely dry. Op.
5F was excavated near the center of the aguada to a depth of 110 cm (Figure 6.6; Table
6.1). The 10010 cm level was sterile marly clay above hard limestone. The 90100 cm
level was green-gray clay with many sherds. A measured radiocarbon date of 1940 40
BP or a calibrated age range of 50 BCAD 120 (Late Preclassic) was obtained from
charcoal collected at a depth of 100 cm. Sediments between 8 and 90 cm were composed
of organic clay that accumulated within the aguada over a long period. Additional
radiocarbon dates from the 2009 core and close examination of the Op. 5F profile indicate
that sediments that began to accumulate late in the Late Classic period (eighth century AD
and later) lie immediately atop sediments dating to the Late Preclassic, clearly indicating
that the aguada was dredged during much of the Classic period, creating a notable
unconformity in the sediments. Presumably the dredging occurred to maintain water
volume, though the excavated sediment could well have been used to enhance soils behind
nearby agricultural terraces (see later discussion).
Figure 6.6. West profile of Op. 5F in the Aguada de Terminos.
Table 6.1. Aguada de Terminos Op. 5C profile description

Horizon Depth Color (Munsell) Description

(cm)

Oi 08 Dark brown Fibric organic matter;

sherds

A 830 Black Organic clay; large hard

(10YR2/1) crumbs; sherds

C1 3070 Dark greenish Clay; massive;

gray (10Y3/1) sherds

C2g 7090 Dark gray Clay; massive;

(Gley 4/N) sherds

C3 90 Greenish gray Massive clay with abundant cherty

100 (10Y5/1) pebbles, sherds, and lithics

C4 100 Light greenish Clayey marl with

10 gray (10Y8/1) cherty pebbles
 The large quantity of chert cobbles, flakes, and other rocky debris found close to some
sides of the Aguada de Terminos indicate that it likely originated as a quarry probably
sometime during the Late Preclassic. Quarrying is a common origin for many ancient
Maya aguadas and urban reservoirs (Akpinar et al. 2012; see also Scarborough and
Grazioso Sierra, this volume). In 2009, we mapped and excavated within an area of chert
processing immediately southwest of the aguada. Some of the chert cobbles and large
primary flakes in this area were used to construct low agricultural terraces on adjacent
sloping land. One of these terraces lying southeast of the aguada was excavated in 2009
(Op. 5B), but did not produce datable material. The excavated terrace is of simple design,
consisting of a core of larger cobbles covered with sloping heaps of small stones. Given
that the terraces are on a low gradient slope (23 percent), the soil behind the terraces
may be surmised to have been added by human labor after the terrace was constructed,
perhaps including muck dredged from the aguada rather than resulting from natural
accumulation.
Pulestons map of the East Brecha shows an area labeled ridge and furrow terrain
lying at the north end of the ridge to the east of the Aguada de Terminos. This area
actually consists of numerous low ridges and mounds constructed from chert-rich cobbles,
chert flakes, and small stones and covers considerably more area than is shown on the
map. Pulestons (1967) notes indicate that he thought that many of these walls functioned
as agricultural terraces and that he began excavations, but was forced to stop by heavy
rain. Our own investigations found that many, though not all, of the walls are positioned
on the slope of the natural ridge as terraces. Op. 5E was a 1 x 3 trench excavated across
one terrace, revealing this feature to have the same simple design as that seen in the Op.
5B terrace (Figure 6.3).
A collection was made of ceramic sherds left on the surface near looters trenches into a
small pyramidal structure in the Grupo de Terminos (a Plaza Plan 2 complex situated on
the peninsular ridge east of the aguada). All of the chronologically diagnostic sherds were
of Late Classic date (Culbert 2009).
Pollen was originally recovered from the 2009 Aguada de Terminos core and a
preliminary analysis completed. However, the ability to dig a pit in the aguada floor in
2010 allowed for a finer stratagraphic sampling and reanalysis. Pollen was also collected
and analyzed from one of two side-by-side cores extracted from the Aguada Vaca del
Monte in 2010.

Pollen Analysis: Aguada de Terminos

The pollen sequence from Aguada de Terminos represents sediment samples collected
from the west profile of Op. 5F in the middle of the aguada. The pollen-bearing portion of
the sequence is represented by the top 100 cm, with basal dates as early as 170 BC (Figure
6.7). A total of 21 pollen samples (5 cm interval) were analyzed from this profile. The
sequence is well dated and fairly detailed, offering a picture of past vegetation in the
aguada area. The basal zone (Zone 1) extends from 100 to 90 cm below surface. As noted,
the profile exhibits an abrupt discontinuity at about 90 cm BS. The sediments below this
break produced a charcoal radiocarbon date of BC 50AD 120 and clearly date to the Late
Preclassic period, whereas overlying sediments (Zone 2 and 3) likely date to the later Late
Classic, Terminal Classic, and beyond.

Figure 6.7. Pollen frequency diagram for the Aguada de Terminos (Op. 5F).
Zone 1 reflects significant Preclassic Maya cultivation efforts in close proximity to the
aguada with pollen from Zea mays (maize) occurring in each sample. Interestingly, pollen
from Canna (achira) was also noted in two samples from this zone. Canna is native to
South America (Stevenson and Wassmer 2004), but became an important cultivar
throughout the Caribbean in pre-Columbian times (see Lentz et al., this volume). Canna
cultivation has not been documented elsewhere in Central America, and further evidence
is needed before we can be certain of its status in the region. Solanaceae (nightshade
family) pollen also occurs almost exclusively in the basal zone from Terminos. This is a
notoriously difficult family to identify beyond specific subfamily groups, but the grains
here are consistent with Physalis/Capsicum (ground cherry or tomatillo/chili pepper
types). It is important to understand that a number of noneconomic weeds could also be
represented by these ordinarily rare grains. Elevated percentages of weedy taxons,
including Cheno-Ams, Poaceae, and Polygonaceae, support the notion that widespread
agriculture was taking place at Terminos at this time. Arboreal elements are somewhat
depressed consistent with the selective removal of unwanted trees.
The aguada was likely to have been cleaned periodically while it was in regular and
continued use, reflected by the absence of Salvinia (a water fern) spores. This floating
plant would have been undesirable to the users of the pond as it would have removed
nutrients from the system and would have choked the waterway. Salvinia is absent in Zone
1 (Late Preclassic), but begins to appear in the aguada sediments midway through Zone 2,
probably indicating the onset of neglect of Terminos as a water source. We interpret this
neglect as likely occurring during the onset of the Terminal Classic when Tikal was
beginning to experience turmoil. A date of AD 1298 to 1413 corresponding to the
sediment zone from 65 to 70 cm BS is likely in error and is inconsistent with the sediment
history and other dates from the sequence.
The middle zone (Zone 2) at Terminos, occurring from 90 to 40 cm BS, reflects
continued, but diminishing human presence in the aguada area. The upper section of this
zone is dated at AD 10241156, that is, within the Early Postclassic. Sediments in the
lower and middle parts of this zone likely represent the later Late to Terminal Classic
periods. Agricultural cultivation continued at Terminos as reflected by the occurrence of
maize pollen in nearly every sample. Disturbance vegetation continues to occur in high
percentages, although Asteraceae pollen decreased and Cheno-Ams grains increase. There
is an increase in both Moraceae (largely Brosimum [breadnut]) and Myrtaceae (myrtle
family, including guava and allspice, but also noneconomic taxons). These increases could
reflect either a resurgence of forest on nearby areas of upland terrain, including succession
on abandoned architectural groups, and perhaps the deliberate management of these
potential economics (which is also suggested by the presence of Sapotaceae pollen) by
surviving populations clustered around the aguada and other water sources. Increases in
forest taxa in the upper part of the zone likely reflect the onset of complete forest
regrowth. Charcoal concentrations are fairly high throughout the entire zone, though
peaking in what is likely the Terminal Classic. These charcoal levels may reflect regional
as well as local patterns.
The uppermost zone (Zone 3), occurring from 40 to 0 cm below surface, dates to the
Postclassic period after AD 10241156. Here we see a total absence of Zea mays pollen
indicating that maize agriculture either had ceased or was reduced in the Terminos area.
There is a gradual reduction in charcoal concentrations throughout this section, again
reflecting a reduction in burning in the region.
Aguada Vaca del Monte lies about 100 m east of an elongated island lying in the bajo
north of the East Brecha (Figure 6.1). One core extracted from the floor of the aguada
yielded a basal radiocarbon date of AD 62090. Pollen analysis on this core revealed
evidence for forest clearance and cultivation including Zea mays in the vicinity of the
aguada during the Late and Terminal Classic periods followed by Postclassic reforestation
(see Figure 7.3 and Lentz et al., this volume). Late/Terminal Classic cultivation was also
indicated in the vicinity of Aguada Pulgada, located deeper in the Bajo de Santa Fe and
adjacent to a large island near the northeast corner of Tikal National Park (Dunning et al.
2002; see also Lentz et al., this volume). The pollen recovered from all three aguadas is
indicative of significant forest clearance and agriculture in the vicinity of these ponds in
the Late Classic period. The degree to which the agricultural pollen reflects adjacent
uplands or the bajo itself, or both, remains more problematic.

Carbon Isotope Sampling and Analyses: Bajo de Santa Fe

and Perdido Pocket Bajo
Changes in the ratios of stable carbon (C) isotopes in the soil organic matter (SOM) have
been used successfully to determine the vegetative histories of soils that have changed
from predominantly C3 forest vegetation to C4 plants associated with ancient maize
agriculture (Burnett et al. 2012; Fernandez et al. 2005; Lane et al. 2004, 2008; Webb et al.
2004; Wright et al. 2009). Most tree and vine species common to the Petn forest of
Central America utilize a C3 photosynthetic pathway while maize and some other tropical
grasses, and some sedges and other tropical species, utilize a more efficient C4 pathway.
As CO2 diffuses into stomata and through cell membranes, there is isotopic discrimination
against the heavier 13CO2 during the photosynthetic process; thus, all plant tissues are
depleted of 13C in comparison to its natural abundance in atmospheric CO2. The 13C of
C3 plants is near 27 per mil (27). C4 plants are much less discriminatory toward the
heavier 13CO2 and exhibit a 13C closer to 12. Soil humus of both surface and root-
zone soil contains the decayed remains of all types of vegetation that grew for thousands
of years; therefore, a vegetative history of the soil is recorded in the C isotope ratios of
humic substances (Boutton 1996; Boutton et al. 1998; Webb et al. 2004). The cultivation
of ancient C4 maize can be observed by a greater portion of 13C in the humus against a
background of 13C depleted humus derived from native C3 forest vegetation (Ehleringer
1991; Liu et al. 1997). An important factor in the interpretation of soil 13C values deals
with the naturally occurring soil microbial diagenesis of SOM, which results in moderate
isotopic fractionation (Blair et al. 1985). The metabolic pathways of soil microbial
decomposition can cause increases in 13C of 12.5 in deeper soil horizons, and certain
soil conditions in tropical regions produce increased microbial activity, which can induce
increases as great as 3 or 4 (gren et al. 1996; Boutton 1996; Martinelli et al. 1996).
Hence, an isotope enrichment of 4 or greater within a pedon has been set as a
benchmark for definitive C4 plant growth against natural C3 isotopic ratios (Burnett et al.
2012; Fernandez et al. 2005; Johnson et al. 2007; Lane et al. 2004; Webb et al. 2004;
Wright et al. 2009).
Soil cores were collected at Tikal National Park, using a clay specific bucket auger
(AMS, American Falls, Idaho) along two sampling transects: One followed a portion of
the East Park Brecha near Aguada de Terminos; the second traversed a portion of the
Perdido Pocket Bajo about 0.5 km south of the Perdido Reservoir (Figure 1.4).2
The 16 pedons (soil cores) belonged to the soil order Mollisols (Soil Survey Staff
2006). The shallow (<45 cm) summit and backslope pedons were Rendolls, while the
deeper soils exhibited vertic and aquic properties (transitional to Aquic Vertisols). The
color range is from 7.5 YR 5/1 to 10 YR 7/1, with dark A horizons indicating high organic
matter, and the subsurface B and C horizons exhibiting argilloturbation with intermixed
dark and light and visible slickensides. Several of the subsurface intermixed C and B
horizons displayed gleying and iron and manganese concretions as a result of poor
drainage within the bajos. The pedon depths vary from 20 to 180 cm, with limestone
bedrock encountered in pedons less than 50 cm deep.
The typical 13C values of surface horizons ranged from 26 to 30, reflecting the
current C3 forest vegetation. A shift in the 13C of near 4 or greater between the surface
horizon and soil of the ancient root zone was deemed sufficient evidence of a one-time C4
plant dominated landscape associated with maize production.
The Terminos pedons were collected along a portion of the East Brecha extending
from just west of the Grupo de Terminos eastward through an arm of the Bajo de Santa Fe
to the base of Pulestons Peninsula (Figure 6.8). The 13C results from the eight Terminos
pedons sampled along a portion of the East Transect mimic those from the Perdido bajo.
Samples were collected from several types of topographic slope contexts: Pedons 13
from summit contexts, Pedons 4 and 8 from backslope locations, and Pedons 5 through 7
were sampled at foot- and toeslope locations.

Figure 6.8. Pedons and carbon isotope profiles in the Bajo de Santa Fe east of Aguada
de Terminos.
Terminos Pedons 1 through 4 were collected on the summit and backslopes in close
proximity to the Grupo de Terminos settlement cluster located on a limestone peninsula.
These pedons were less than 75 cm deep and the change in 13C with depth was less than
1.8 (Table 6.2). The vegetative history contained in these four pedons is an indication
that maize was not grown long-term on the hill or among the household structures.
Terminos Pedons 5 and 6 exhibited 13C shifts of 3.8 and 5.7, respectively. These two
profiles were more than 90 cm deep and contained isotopic evidence of an ancient C4
vegetative history suggesting ancient forest clearance for maize production. Pedons 7 and
8 were located at the edge of Pulestons Peninsula downslope from other groups of house
mounds. Pedon 7 was deeper than 90 cm and exhibited a 13C shift of 2.2. Pedon 8 was
a shallow backslope soil with a 13C shift of 0.8. These two pedons at the edge of the
rural settlement lacked evidence of ancient maize agriculture.
Table 6.2. Selected physical and chemical characteristics of the A horizons of 16
soil pedons

Pedon

Pedon Easting Northing Depth Slope Clay Texture* pH P

m M cm position g/kg class mg/kg

Perdido 220185 1905099 30 Summit 308 SiCL 6.8 5.11
1

Perdido 220233 1905096 30 back 383 CL 6.9 5.88

2

Perdido 220276 1905090 15 back 479 C 7.0 5.7

3

Perdido 220139 1905101 30 back 526 C 7.0 4.11

4

Perdido 220058 1905099 105 foot 526 C 6.0 2.4

5

Perdido 219972 1905112 90 foot 827 C 7.1 4.05

6

Perdido 219877 1905116 45 toe 858 C 7.3 2.79

7

Perdido 219656 1905104 195 toe 878 C 5.6 2.85

8

Terminos 226503 1905532 60 summit 552 C 7.1 5.88

1

Terminos 226448 1905563 45 summit 578 C 6.7 2.91

2

Terminos 226679 1905784 15 back 428 C 7.6 2.23

3

Terminos 226711 1905521 75 back 801 C 5.8 0.57

4

Terminos 226778 1905518 90 foot 853 C 5.3 0.63

5

Terminos 227321 1905476 90 toe 927 C 5.8 0.19

6

Terminos 227717 1905444 90 toe 928 C 5.2 0.25

7

Terminos 228000 1905430 60 back 931 C 5.7 0.3

 8

* SiCL = silty clay loam

CL = clay loam
C = clay
The Perdido Pocket Bajo lies about 400 m SSW of the Mundo Perdido Plaza at an
elevation of about 220 m above the Penn Project benchmark (see Carr et al., this volume),
and covers an area of approximately 0.4 km2 (Figure 1.3). Samples were collected along
an east-west transect extending from sloping terrain on the east side of the bajo westward
onto the bajo floor (Figure 6.9). Perdido pedons 1 through 4 were approximately 15 cm
deep over limestone. These shallow Rendolls were sampled within patios and near
household structures of Group C7. The change in 13C between surface and root zone
horizons of the pedons ranged from 0.8 to 1.6, reflective of both ancient and current C3
forest vegetation (Table 6.2). There was no C isotope evidence of ancient maize
production within the household group. Pedons 5 and 6 were located at the edge of
thepocket bajo. The cores were collected to depths of 105 and 90 cm, respectively. The
change in 13C from the surface horizon to horizons below 45 cm shows 3.3 and 6
shifts, respectively. Pedon 5 was closest to household structures and the shift in 13C
provided some evidence of ancient C4 vegetation, but pedon 6 located farther into the
pocket bajo exhibited strong isotopic evidence of ancient C4 vegetation associated with
forest clearance and maize agriculture. Perdido Pedon 7 was only 45 cm deep and had a
13C shift of 2.7, which is inconclusive for the presence of C4 plants. Pedon 8 was
sampled to a depth of 195 cm and exhibited two peak shifts in 13C of 4.3 at the 30 cm
depth and 4.7 at the 75 cm depth. This pattern in the C isotope data suggested two
separate periods of C4 dominated plant cover associated with ancient maize agriculture. Of
the eight Perdido pedons, number 6 contained a 13C enrichment of 6.05, and pedon 8
exhibited an enrichment of 4.7. These data indicated that C4 plants (likely maize) were
grown for long periods within the Perdido Pocket Bajo.
Figure 6.9. Perdido Pocket Bajo pedons and Carbon isotope profiles.
In sum, three Perdido pedons and two Terminos pedons from bajo toeslopes exhibited
weak to strong evidence of ancient C4 maize agriculture. The soil horizons with the
greatest enrichment in 13C were located at depths of 30 to 75 cm. These soil depths
represent the ancient rooting zone. Soil scientists have reported that a large portion of the
C photosynthesized by maize is converted to SOM through rhizodeposition of root tissues
and root exudates (Balesdent and Balabane 1996; Molina et al. 2001). Detritus from the
C3 forest vegetation that replaced ancient C4 crops and weeds has deposited 13C depleted
C at the surface. From the results of the isotope ratio analysis, 4 of the 16 pedons show
strong evidence of ancient maize agriculture. One pedon showed weak evidence with a
shift of 3.3. The weak isotopic signature could be attributed to a briefer period of maize
agriculture or possibly even a transition area between ancient native forest and ancient
agricultural fields. Even though some of the soils of the Tikal bajos fall under the soil
order of Mollisols, one of the most fertile of the 12 soil orders, significant limitations exist
with these soils. The high clay content and poor drainage cause these soils to be extremely
sticky and muddy in the wet season and extremely hard in the dry season. It is probable
that large amounts of labor and management would have been needed to benefit from the
potential fertility of these soils. It is important to note the location of the pedons along
each transect in conjunction with the map and in situ information. These two separate
locations of similar soil, landscape, and slope provided consistent results, indicating that
they are accurate and can be predictive of similar areas within the region (Wright et al.
2009). All of the pedons with vegetative histories of ancient maize agriculture are within
the karst depressions and their margins. However, the slopes near the bajos and bajo
margins include greater slopes including back and shoulder slopes, which under ancient
agricultural conditions could have posed a highly erodible landscape, which caused
isotopic evidence within the soil to be lost. These findings agree with those derived from
the isotopic study of sediments in some of Tikals reservoirs, which indicate little maize
was grown on the sloping land in the watersheds feeding the Perdido and Corriental
Reservoirs, but that maize was being grown (among other crops) near Aguada de
Terminos and Aguada Vaca del Monte along the margins of the Bajo de Santa Fe
(Tankersley et al., this volume).
The preceding findings corroborate those of Richard Burnett, who conducted a similar
study at the site of Ramonal (also known as Chalpate) in 2005 (Burnett et al. 2012).
Ramonal is a small site lying some 9 km ESE of the Tikal site center and 2 km south of
the East Brecha; it was likely politically independent through the Late Preclassic period
then incorporated into greater Tikal in the Classic (Lou 1997; Fialko 2005; see also Lentz
et al., this volume). The site lies on rolling terrain of low ridges and small pocket bajos,
and soils range from shallow Rendolls to deeper Argiudolls. C isotopic analysis revealed
strong evidence of prolonged maize agriculture in footslope and toeslope locations around
and within the pocket bajos, but no evidence on adjacent uplands (though again with the
caveat that soils in these locations were likely truncated by ancient erosion).

Perdido Pocket Bajo: Excavations

The Perdido Pocket Bajo lies near the Tikal site center downslope of the Mundo Perdido
complex. The Perdido Reservoir is perched near the northeast edge of the bajo. Several
arroyos draining the central Tikal ridge disgorge into the bajo, including those from the
Mundo Perdido Plaza that were heavily channelized by the ancient Maya (see
Scarborough and Grazioso Sierra, this volume). That engineering was apparently in part
designed to allow water to be alternately shunted into the reservoir or directly into the
bajo. Water stored in the reservoir could have subsequently been selectively released onto
the bajo floor at a later date.
Two pits were excavated into the bajo floor: Ops. 8D and 8I. The profile revealed in
Op. 8D is striking (Figure 6.10). The lower portion of the profile
(www.cambridge.org/Tikal; See S. Fig. 5) comprises a typical bajo Vertisol, but the major
part of the A horizon of this buried soil has been stripped away, most likely scoured by a
major flood event probably a hurricane (Dunning and Houston 2011). Subsequent floods
are recorded in a series of stratified alluvial deposits. The earliest stratum included a
charcoal fragment that produced a calibrated radiocarbon date of AD 400570. These
strata likely represent later high-energy flow events (strong enough to transport coarser
sediment, but not strong enough to scour the bajo surface) that were directed away from
Perdido Reservoir. Op. 8I, located on slightly higher ground, exhibited a similar profile
with truncated bajo soil and overlying alluvial deposits, though a lesser number than Op.
8D (Dunning et al. 2011). The reduced number of alluvial strata in Op. 8I likely indicates
that some high-energy flow events did not reach this slightly more elevated location.
Figure 6.10. South profile of Op. 8D in Perdido Pocket Bajo.
The C isotope analysis presented earlier makes a compelling case for sustained maize
agriculture in at least part of the Perdido Pocket Bajo. The position and nature of Perdido
Reservoir suggest that it was used to provide supplementary water for crops within the
bajo (see Scarborough and Grazioso Sierra, this volume). Such water could have made the
difference between a successful and a failed crop during years with particularly strong
midsummer canicular droughts, or perhaps even allowed for some dry season cropping.
However, evidence from within the bajo soils suggests that these crops were occasionally
at risk because of storm flooding.

Arroyo Corriental Pocket Bajo

Arroyo Corriental Pocket Bajo lies a couple of hundred meters downstream from
Corriental Reservoir along the course of the Arroyo Corriental (Figure 1.3). Between the
reservoir and pocket bajo, the arroyo is deeply entrenched where it snakes between a
series of steep hills. Op. 2A was a 1 x 4 m trench excavated into alluvial deposits lying
along the arroyo (Dunning et al. 2009). A soil surface (Ab) buried 90 cm below the
surface produced a calibrated radiocarbon date of 19080 BC. This Preclassic soil was
progressively buried by later high-energy floods.
Op. 2B was a 250-cm-long trench dug into the footslope at the base of a steep hill
overlooking the Arroyo Corriental Pocket Bajo (Figure 6.4b). A buried soil surface (Ab)
exposed in this trench produced a calibrated radiocarbon date of 90 BCAD 80. Later
upslope erosion and colluviation buried the soil similarly to that witnessed in comparable
locations at Tikal and elsewhere in the Maya Lowlands (Beach et al. 2006). These findings
corroborate the idea that the Maya at Tikal began to clear and destabilize sloping land to a
significant degree within the Preclassic.

Discussion: Ancient Land Use and Settlement in a Bajo

Landscape
The high density of ancient settlement along the margins of the Bajo de Santa Fe has been
noted in a number of previous studies (e.g., Bullard 1960; Carr and Hazard 1961; Cowgill
and Hutchinson 1963; Puleston 1973, 1983; Fry 1969, 2003; Fialko 2000, 2005; Dunning
et al. 2002; Weller 2006). The brief outline of the history and distribution of settlement
around the bajo draws upon those earlier studies as well as our own observations including
new paleoenvironmental data presented previously.
There is relatively little evidence for Middle Preclassic settlement around the Bajo de
Santa Fe. However, during the ensuing Late Preclassic, minor centers sprang up in many
parts of the Upper Holmul drainage, including the margins of the Bajo de Santa Fe, and
settlement clusters likely directly associated with Tikal appear along the nearest bajo
margins. Fialko (2005) notes that many of the minor centers were situated at the
confluences of tributary streams or locations where stream channels enter into bajos, both
locations where deep water pools are often to be found. As noted earlier, land clearance
clearly accelerated in the Late Preclassic, and soil erosion on the uplands increased
significantly as sloping land surfaces became destabilized (see also Murtha and Webster,
this volume). Our excavations indicate that the Aguada de Terminos, opened initially as a
chert quarry, was transformed into a reservoir during the Late Preclassic, perhaps in
response to the increasing water needs or a growing population on Tikals periphery or
increasing regional aridity (Dunning et al. 2014). The reservoirs in central Tikal were also
initiated about this same time (Scarborough and Grazioso Sierra, this volume). Terminos
was also a focus of agricultural activity, including both maize and root crop cultivation in
the Late Preclassic. Presumably, cultivation of the many pocket bajos within the Tikal
urban area was also under way in the Preclassic. Intriguingly, pedon 8 in the Perdido
Pocket Bajo contained isotopic evidence for ancient maize cultivation at both 30 cm and
75 cm depths suggestive of two distinct periods of sustained maize cultivation separated
by a time dominated by aggradation. The lower zone likely represents cultivation of the
Preclassic soil surface buried by sediments produced by upslope erosion, followed by
sustained maize cultivation again after landscape stabilization likely in the Classic
period.
Fialkos (2005) analysis of settlement in the Upper Holmul drainage indicates that by
the Early Classic, minor centers and other settlements in the region were incorporated into
the political and economic dominion of Tikal, a situation that held through the end of the
Late Classic. However, Fry (2003) notes that during the Middle Classic Hiatus many of
the peripheral and minor centers around Tikal show a greater degree of economic
autonomy than during preceding and later times. By the seventh century AD, settlement
intensified around the Bajo de Santa Fe with all tested settlement clusters indicating
significant occupation and building activity. Earlier work in the Bajo de Santa Fe on La
Isla and our own collections from looters trenches on Isla Vaca del Monte indicate that
these islands were both densely occupied in the Late Classic and it is likely that other bajo
islands share this pattern. Frys (1969, 2003) analysis indicates that considerable localized
ceramic and lithic production was carried out in settlement groups in Tikals periphery,
though there was clearly some degree of economic integration with central Tikal. The
primary economic activity of most households on the margins of the Bajo de Santa Fe was
likely agriculture. As noted previously, some of the most productive soils in the region are
found along the footslopes and toeslopes surrounding the bajo. Carbon isotope analyses of
soils sampled along portions of the East Transect indicate that significant quantities of
maize were being grown in this area of the bajo. Pollen from the Aguada de Terminos and
Vaca del Monte supports this assertion. What land use might have been like deeper within
the Bajo de Santa Fe, where both soils and drainage are more problematic, is less clear.
However, timber extraction, including Haematoxylum campechianum, was clearly taking
place and likely accelerated in the Late Classic (Lentz and Hockaday 2009; Lentz et al.,
this volume).
Little direct evidence has been recovered at Tikal for cultivation of sloping uplands
surrounding the reservoirs. However, considerable erosion of land surfaces is indicated by
substantial sediment deposition in investigated pocket bajos as well as the margins of the
Bajo de Santa Fe. Pocket bajos were clearly significant foci of agricultural production in
the Classic period. As noted by Webster and Murtha (this volume), by the eighth century
AD, the Tikal landscape was full likely saturated with people and with crop
production pushed to its limit. In this scenario, the prime lands of deep, colluvial soils
rimming bajos were undoubtedly intensively cultivated.
Evidence collected to date indicates that settlement around the Bajo de Santa Fe
followed a similar trajectory to that of central Tikal. In the Terminal Classic period, many
settlement clusters and minor centers were only lightly occupied or abandoned altogether.
Martin (2003) notes that many minor centers surrounding Tikal begin to show increasing
separation through the erection of their own stelae. Fialko (2005) notes that the Terminal
Classic minor centers that remained occupied longest were those located at stream
confluences or other natural water sources, likely reflecting increasing problems of
regional drought frequency and severity. The continuation of a significant, though
declining, level of population indicated by the pollen record at Aguada de Terminos
suggests that this water source also offered a refuge. No indications of Postclassic
settlement have been recovered archaeologically around the Bajo de Santa Fe. However,
pollen evidence from Aguada de Terminos indicates the probable persistence of some
residual population in the area for a century or so.

Conclusions
The ancient Maya city of Tikal lies near the southwestern margin of the sprawling Bajo de
Santa Fe. Ancient residential settlement is relatively dense along the flanks of the bajo and
on islands of higher ground within it. Archaeological, geoarchaeological, and
palynological evidence suggests that the residents in this area were engaged in agriculture
focused on deep, cumulic soils along the bajo margins including the cultivation of maize
and root crops. The nature of Maya activities within the central portions of the bajo
system, where hydrological extremes are more pronounced, is more problematic, but it
likely included timber extraction. Chert exposed around the edges of the bajo was clearly
also exploited. Rocky debris from chert quarrying was used to construct terraces and field
walls flanking at least one part of the bajo near the Aguada de Terminos. Portions of the
Bajo de Santa Fe and its complex watershed were evidently hydrologically manipulated,
as seen in the presence of reservoirs (aguadas) and possible navigation channels. On
satellite imagery, canals appear to connect several islands within the bajo with the channel
of the Rio Holmul, a large seasonal stream that winds through the depression but were not
reached on the ground by our teams. Our understanding of the environmental history of
the Bajo de Santa Fe remains far from complete.
The upland terrain over which urban Tikal sprawls is riddled with karst depressions of
varying sizes that we have dubbed pocket bajos. Our investigations included two of
these depressions downstream from the Corriental and Perdido Reservoirs.
Geoarchaeological investigations within these pocket bajos suggest that Preclassic
agriculture on surrounding lands produced significant soil erosion and resulting
sedimentation. As the Classic period unfolded, reservoirs constructed at the hydraulic
heads of these depressions were used to control flooding and, perhaps, were used for
limited amounts of irrigation.
The karst landscape of the central Yucatan Peninsula, including the region surrounding
Tikal, includes thousands of solution and collapse depressions ranging in size from a few
tens of square meters to hundreds of square kilometers. The predilection of early and large
Maya communities to be established on the margins of large depressions (bajos) indicates
that these locations were perceived as valuable. Recent investigations have uncovered
compelling evidence that some bajos contained shallow lakes or perennial wetlands when
the Maya first settled on their margins. To date, no similar paleoenvironmental data have
been recovered from the Bajo de Santa Fe. However, only small areas of this sprawling
depression have been sampled, and it is possible that lacustrine environments once existed
in unsampled areas (e.g., perhaps in the lowest-elevation areas in the northeastern portion
of the bajo). Regardless, this bajo and the numerous smaller depressions that dot the Tikal
landscape clearly held adaptive advantages for Maya peoples. Manipulation of the
topography and surface drainages was one such advantage, expressed at Tikal in the
evolution of a system of diverting and capturing runoff that was needed to concentrate
large numbers of people in the seasonally arid region. The slope breaks found at bajo
edges are also areas of natural cumulic soils well suited to agricultural intensification.
Bajo margins are also natural ecotones, characterized by transitional vegetation types, but
also in close proximity to the forest and soil resources of both the well-drained uplands
and the slow-draining bajos. In combination, bajo-margin locations presented the Maya
with a suite of resources and opportunities that made these attractive and advantageous
places for settlement, as seen in the development of Tikal.
Notes
1 From the account of Fray Andrs de Avendao y Loyola on the failed wet season
Entrada to reach the Itza kingdoms of the Peten in 1695 (after Means 1917: 108): We
walked a distance of four leagues to another place called Nohbecan, which place, for a
league before and a league after, consists of great overflowed stretches which in this
language are called akalches. It can be well understood what pain we endured with our
sore legs and feet passing through this two leagues of water and mud which at least came
up to our knees, leaving us almost crippled.
2 Each core was partitioned in 15-cm depth intervals for stable isotope analysis. The
samples were transported to the BYU Soil and Plant Analysis laboratory. Samples were
dried in air, crushed, and sieved (2 mm/10 mesh). The soil textural class of each surface
horizon was determined by the hydrometer method (Gee and Bauder 1986). General
taxonomic designations were determined according to the USDA taxonomic classification
(Soil Survey Staff 2006). Each sample was thoroughly mixed and a 5-g subsample was
further ground by mortar and pestle to pass through a 250 m (60 mesh) sieve in
preparation for total C, total N, and isotope analyses. Carbonates were removed from 3-g
samples by acidification with 1.0 M HCl in a water bath at 70 C until effervescence
ceased. Humic and fulvic acid fractions of the SOM were extracted (Webb et al. 2004;
Wright et al. 2009), leaving the humin fraction for isotopic analysis. The humin fraction is
considered to contain the oldest C in the soil (Bender 1968). The stable C isotope ratios of
the humin fraction of the SOM of each soil horizon were then determined with the use of
an isotope ratio mass spectrometer (Thermo Finnigan, Waltham, Massachusetts) coupled
with an elemental analyzer (EAIRMS) (Costech, Valencia, California). The ensuing results
were plotted to facilitate visual analysis and reported as 13C in per mil notation ().
 7 Connecting Contemporary Ecology and
Ethnobotany to Ancient Plant Use Practices of the
Maya at Tikal
Kim M. Thompson, Angela Hood, Dana Cavallaro and David L. Lentz

Introduction
The ancient Maya left a grand architectural mark in the Petn with their temples, palaces,
stelae, and reservoirs at Tikal. Did they also leave an ecological mark? This question
directed our efforts at Tikal to understand how the ancient Maya developed a landscape to
fulfill the everyday nutritional, structural, medicinal, and spiritual needs of the populace
from plant resources. We studied the modern forests that cover much of the ancient polity
of Tikal, and the ground beneath it, to discern remnants of their silvicultural practices,
appreciating that any information we gleaned would offer insights into the sustainable
practices that supported large-scale occupations of the landscape and inform conservation
efforts in these endangered forests (Carr 2008; Fedick et al. 2008; Gomez-Pompa 1987;
Mercer et al. 2005; Nations 1992; Puleston 1973). Forests in areas once occupied by the
Maya have been studied for these very purposes (Brokaw and Mallory 1993; Lentz et al.
2002; Ross 2011; Ross and Rangel 2011), and interpretations of ancient Maya influence
on the modern landscape have been supported by ethnohistoric accounts, ethnographic
studies of modern Maya plant use practices, and the recovery of paleoethnobotanical
remains. We married three lines of discovery (paleoethnobotanical, ethnographic, and
ecological) to elucidate how the forests of Tikal might act as indicators of ancient Maya
silviculture. Silviculture describes a continuum of tree management intensity from home
gardens to managed fallows and forests (Peters 2000). Our study had three objectives: (1)
to reveal woods used by the ancient Maya at Tikal, (2) to characterize the modern tree
community of Tikal, and (3) to describe correlations among the modern forest,
ethnobotanical records of modern and historic Maya, and ancient plant use.
A small number of species typically dominate Neotropical forests and have been
referred to as oligarchies (Pitman et al. 2001). In Belize, Campbell et al. (2006) explored
whether the particular oligarchies present in forests once occupied by the ancient Maya are
relicts of their protection and management and identified a correlation with the forest
gardens of contemporary Maya. Other studies have examined ancient Maya settlements
that differ in the density of residential structures for disproportionate occurrence and
pairing of tree species, revealing significant associations (Ross 2011; Ross and Rangel
2011). Tikal, one of the preeminent lowland Maya polities, is well known for its cut-stone
architecture and extensive building across an expansive anthropogenic landscape that,
given its large populations, must have relied on careful resource management (Coe 1965;
Puleston 1973). If traces of prudent ancient Maya land-use management practices have
indeed been imprinted in the current ecosystem, then species economically valuable to the
ancient Maya, those that are heavily represented in paleoethnobotanical remains at Tikal,
should be among the most dominant trees in the modern forests of Tikal.
Orchards, kitchen gardens, and managed forests have been hypothesized as components
of ancient Maya subsistence and could have been implemented using the lithic technology
of the ancient Maya (Denevan 2006; Ford and Emery 2008; Gomez-Pompa et al. 1987;
Harrison and Turner 1978). In fact, indigenous people have been managing tropical forests
across the globe for millennia (Erickson 2006; Wiersum 1997). The gardens and forested
areas tended by present-day indigenous Maya, including the Chorti, Huastec, Itza, Qeqchi
, Lacandon, and Yukatek, have been described as species-rich with a complex structure
(Alcorn 1984; Atran et al. 2004; Campbell et al. 2008; CorzoMrquez and Schwartz 2008;
Diemont et al. 2006; Gomez-Pompa et al. 1987; Rico-Gray et al. 1990, 1991; Wisdom
1940). Numerous plants are valued within each home garden, and there is a high diversity
of species cultivated in gardens within each community. A single garden may have as
many as 180 species in multiple levels from herbs to fruit trees and often includes
representatives of the forest canopy (Corzo Marquez and Schwartz 2008; Gillespie et al.
1993; Lundell 1938). Although contemporary Maya gardens include many non-native tree
species that were introduced by Europeans, native trees are valued and many that are
found in the forest are also maintained in home gardens, in small orchards, and within
reserve areas set aside for agroforestry (Atran 1999; Corzo Mrquez and Schwartz 2008;
Ford 2008; Mercer et al. 2005; Rico-Gray et al. 1991).
Contemporary Maya cultures, which maintain many of their ancestors traditions,
provide insights into practices of the ancient Maya (Atran et al. 2004). The observation
that valued species are naturally abundant in forests that were once occupied by the
ancient Maya led to hypotheses that their dominance was a direct result of ancient
agroforestry activities (Folan et al. 1979; Gomez-Pompa 1987; Lundell 1937). It is
possible that the forest has had a long-term effect on Maya practices just as their needs for
plant resources may have had long-term impacts on the forest. A variety of plant resources
would have been essential to meet not only food requirements, but construction,
medicinal, ritual, social, and other needs of the Tikal inhabitants (see Lentz et al., this
volume). Economic values placed on planted and protected trees by indigenous Maya
groups have been studied, and the results of these investigations demonstrate that the
diversity of tree species parallels the diversity in their values for the people who manage
them (Alcorn 1984; Atran et al. 2004; La Torre-Cuadros and Islebe 2003; Rico-Gray et al.
1991).
The abandoned ruins of Tikal, once heavily populated, still contain remnants of former
residential structures, of their ancient residents trash, and concomitant evidence of lives
lived in the past. Plant materials burned for ceremonial or practical purposes offer clues to
the resources of the ancient Maya when they become preserved in archaeological contexts
as carbonized or charred remains (Johnston and Gonlin 1998; Stanton et al. 2008). The
paleoethnobotanical record reveals a diversity of species utilized by the ancient Maya
(Beltrn Frias 1987; Cliff and Crane 1989; Coe et al. 1986; Hurst et al. 1989; Lentz 1991a,
b, 1994, 1999; Lentz et al. 1996; McKillop 1994, 1996; Miksicek 1983, 1986; Miksicek
1991; Turner and Miksicek 1984), and these shed light on resource use, forest
management, and environmental changes in Maya prehistory (Lentz and Hockaday 2009).
 In this study, we compared species represented in the paleoethnobotanical record with
the modern composition of the forest and ethnographic records to test the hypothesis that
the tree community at Tikal holds a connection to plant use practices of the ancient Maya.
Our main hypothesis in this study was that the ancient Maya of Tikal managed their
forests in ways that conserved useful species. We expected to find that species
predominant in the paleoethnobotanical record at Tikal were also dominant among the tree
species of the modern forests that cover the ancient city of Tikal. Furthermore, we
hypothesized that the dominant trees in the modern forest would be of high socioeconomic
value to contemporary Maya groups.

Paleoethnobotanical Samples
Shovel test pits were excavated by the University of Cincinnati Archaeological Project at
Tikal (UCAPT) in the 2009 field season behind house groups for the recovery of
carbonized paleoethnobotanical remains. To enhance our recovery of archaeological plant
remains, we employed soil phosphate testing to direct our excavation efforts in the 2010
season. Elevated phosphate levels are reliable indicators of the presence of organic
materials, including carbonized plant remains (Eberl et al. 2012; Parnell et al. 2002; Terry
et al. 2000). We systematically collected soil samples for phosphate testing at 10 m
intervals in 0.01 ha plots across four operations totaling 3.12 ha. Excavations were
conducted in areas with high phosphate signatures (Figure 8.1). Carbonized plant
macroremains were collected as they were encountered during excavation. Soil samples
were also collected from excavations for flotation processing, and plant remains were
extracted using a SMAP flotation system (Pearsall 1989). All charred plant materials
designated for radiocarbon dating were analyzed for botanical identification prior to
processing. Carbonized plant remains collected by the University of Pennsylvania project
at Tikal from 1956 to 1969 were identified by Lentz and Cavallaro in an earlier study, and
many of these data were drawn upon to extend our understanding of plant use activities at
the ancient city. These included 64 wood morphospecies, 32 identified to species level, in
24 plant families. Dating of plant remains to major periods of the Maya era was
accomplished using associations with ceramic assemblages and AMS radiocarbon dating
of contexts.

Species Identification of Paleoethnobotanical Remains

Preliminary identifications to broad taxonomic groups were made with the aid of a
stereomicroscope with a magnification of 850X. For more detailed analyses of seeds,
wood, and other macroremains, images of 228 samples were produced with a Philips
XL30 Environmental Scanning Electron Microscope equipped with a field emission
electron source and visualized with the software AnalySIS 3.2, which was also used for
measuring vessel diameter. Identification of samples was accomplished using diagnostic
features following guidelines established by the International Association of Wood
Anatomists (IAWA Committee). Each specimen was compared to Lentzs reference
collection of tropical woods and images and descriptions published by others (Alves and
Angyalossy-Alfonso 2000; Carlquist 2001; Cassens and Miller 1981; Perna and Melandri
2006; Folorunso 2011; Gasson et al. 2009; Gutirrez et al. 2009; Heintzelman and Howard
1948; Inside Wood 2004; Kribs 1959; Kukachka 1978; Lens et al. 2005, 2008; Len 2003,
2005; Luchi 2011; Marchiori and Brum 1997; Rebollar and Quintanar 2000; Richter 1981,
1985; Richter and Dallwitz 2000; Stern 1954; Uribe 1988; Wheeler 2011).

Sampling of Trees and Vines

We conducted a survey of the modern Tikal forest during two field seasons of the UCAPT.
We characterized the composition of the tree community in three distinct habitats in a total
area of 5.95 ha at Tikal National Park. Transects were established as 50 m x 10 m plots
(0.05 ha) over a total length of 1.55 ha in 2009 to identify trees in the upland forest.
Circular plots of 0.1 ha were also established around three house groups in the same area.
During the 2010 field season, we identified trees in each of the plots we had established
for excavation in plazuela groups. To capture the broad diversity of forest species, we also
established transects in upland and transitional forests and the seasonal swamp forests
known as bajos. Geographic coordinates of each plot were taken with a Magellan
Platinum GPS unit. The diameter at breast height (DBH), height calculations, and
characteristic features of each tree and vine greater than 6 cm DBH were recorded in each
plot. Height of the shortest and tallest trees in each sample area was calculated using a
hand-held clinometer and measuring tape. Common names in Spanish and Mayan
languages, as well as known uses of the identified trees and vines, were provided in the
field by local workers knowledgeable about forest trees. Samples were taken from a
minimum of one tree corresponding to each common name. Plant samples were pressed
and dried over a light source and maintained in the University of Cincinnati Herbarium
(CINC). Duplicates were delivered to the Universidad de San Carlos de Guatemala
Herbarium.

Species Identification for Modern Forest Survey

Taxonomic identifications were determined in the field or lab according to leaf and bark
characteristics, odor, presence and color of sap, characteristics of fruits and flowers when
present or known, and color of the inner wood. Identifications were confirmed in the lab
using the Flora of Guatemala (Standley, Steyermark and collaborators 194677) and other
systematic treatments (Acevedo-Rodrguez 2003; House et al. 1995; Mitchell 1987;
Pennington 1990, 1991; van der Werff 2002).
Samples with unknown or uncertain species identifications were taken to Missouri
Botanical Garden Herbarium for further analysis or corroboration under the guidance of
Ron Leisner. Identities of 26 species were established with DNA sequencing using
universal primers to amplify conserved regions from nuclear and chloroplast genomes.
DNA was extracted from leaf samples and two primers were used to amplify conserved
genes.1 Sequencing was accomplished at the University of Washington High-Throughput
Genomics Unit and visualized using FinchTV1.4.0 (Geospiza). Sequence similarities were
sought using the Basic Local Alignment Search Tool (BLAST) in the National Center for
Biotechnology Information (NCBI) database (Altschul et al. 1990). Failure to identify
unknown species from the modern forest occurred when there was insufficient plant
material, specimens lacked reproductive structures, or we were unable to collect
specimens and the common name provided by our informant could not be confidently
associated with only one species. The validity of taxonomic names was cross-referenced
with The Plant List (2010), a collaborative effort of the Royal Botanic Gardens, Kew, and
Missouri Botanical Garden.

Socioeconomic Values
An important aspect of a study of this nature is to conduct a thorough review of what is
known about plant uses among modern Maya communities. Each species found in our
modern forest surveys and paleoethnobotanical study was evaluated for its value in the six
most commonly cited categories in the literature: food, medicine, construction, fuel, ritual,
and other, which includes fiber, fodder, dyes and tannins, adhesives, tools, household
goods, and poisons for fishing (Arvigo and Balick 1993; Atran et al. 2004; Balick et al.
2000; Campbell et al. 2006; Cannon and Cannon 1989; Henderson et al. 1995; Macbride
1951; Mutchnick and McCarthy 1997; Standley, Steyermark and collaborators 1946
1977; Rico-Gray et al. 1991; Roys 1931). We established two categories: high value (HV)
and low value (LV). High value was assigned if a species was considered a food resource
or had economic value in four or more of six categories. Relative dominance by basal area
of modern forest species was compared for these categories in SPSS (Version 21) using
the Mann-Whitney U Test, which compares distribution of a variable for two independent
groups (George and Mallory 2011; Mann and Whitney 1947). There were many tied ranks
for abundance of paleoethnobotanical remains. Therefore, a chi-square contingency test
was set up to evaluate differences in HV and LV groups on the basis of the presence or
absence of each species in the paleoethnobotanical record at Tikal.

Analysis
Our goal in the analytical portion of the study was statistically to compare data from the
modern forest, the ethnobotanical record, and the paleoethnobotanical study. Standard
diversity indices were calculated within individual plots and for the entire survey,
including species richness, Pielous index of evenness (J), the Shannon-Wiener Index (H),
and Simpsons Index (Hurlbert 1971; Pielou 1966; Shannon 1948; Simpson 1949).
Richness is a total count of species, and evenness describes the distribution of those
species in an area. The Shannon-Wiener Index incorporates both richness and relative
abundance for each species, and Simpsons Index measures dominance in a community.
Species accumulation curves (exact method), which pool average species richness values
across increasingly larger sample areas, were generated for each transect or grid and for a
larger subset of data encompassing 4.65 ha in 0.05 ha parcels across the bajos, transitional
forests, and upland forests (Arrhenius 1921; Kindt 2003). Species accumulation methods
are more informative than species area curves in characterizing plant communities because
the first method considers the identities of new species at increasingly larger areas while
the latter only measures changes in the numbers of species at each plot size (Ugland et al.
2003). Total species richness was extrapolated using first- and second-order Jackknife,
Chao, and Bootstrap methods (Burnham and Overton 1979; Chao 1984; Colwell and
Coddington 1994; Heltshe and Forrester 1983; Palmer 1990; Smith and van Belle 1984).
The second-order Jackknife estimator is sensitive to uncommon species, those that occur
only once or twice in a sample (Palmer 1995). These values, which were consistently the
highest, were reported since we identified a high proportion of uncommon species.
Species rank abundance, a plot of the rank order of a species against its abundance, was
accomplished for the entire survey area (Kindt and Coe 2005). Beta diversity, the variation
between sample units, was described with a Jaccard dissimilarity matrix for the modern
forest and paleoethnobotanical remains. The Jaccard matrix is created from presence or
absence of species in each context (Bray and Curtis 1957; Jaccard 1908). Each of these
measures was calculated by using Biodiversity (R Development Core Team 2005; Kindt
and Coe 2005). Comparisons in diversity between Tikal and other sites were tested using
chi-square analyses.
Proportional abundance was calculated for both the modern tree communities and the
paleoethnobotanical remains. Relative dominance, as a function of basal area, and relative
stem density were calculated for each species in each habitat of the modern forest. These
were summed to create an importance value (IV) for each species and averaged across
habitats to account for differences in sample area sizes (Cottam and Curtis 1956; Curtis
and McIntosh 1951). Proportional abundance for paleoethnobotanical remains was
calculated as the number of contexts in which a family, genus, or species appeared
compared to the total number of contexts. To test for differences in representation of
oligarchic species in the paleoethnobotanical remains, two groups were established:
oligarchic if the species was in the top ten for IV in any modern habitat and nonoligarchic
if the species was not. We tested for significant differences between these two groups in
the ancient wood assemblage using two methods: chi-square analyses and a Spearman
rank correlation. Using a chi-square analysis, the abundance of paleoethnobotanical
remains in each of the two groups, oligarchic and nonoligarchic according to modern
surveys, was compared to mean abundance for the total ancient assemblage at Tikal.
Therefore, the expected value was the mean number of contexts for the total group of
plant remains (both oligarchic and nonoligarchic). Then, using the total number of species
identified from the modern forest and the paleoethnobotanical remains, thereby including
species that were absent in ancient contexts, we compared abundance in the
paleoethnobotanical remains between the two groups, oligarchic and nonoligarchic, using
chi-square analyses as just described. Finally, a Spearman rank correlation was conducted
in SPSS to compare proportions of species found in the paleoethnobotanical record with
the average IV (relative abundance and dominance across all three habitats) of those
species also found in the modern forest at Tikal.

Results
Modern Forest
We documented a total of 124 species of trees and vines, with a DBH of at least 6 cm (115
at 10 cm threshold), in sampling plots that covered 5.95 ha. Of the 43 plant families
represented (40 at the 10 cm DBH threshold), Sapotaceae, Moraceae, Sapindaceae,
Meliaceae, and Arecaceae were dominant (Table 7.1). In the uplands, we identified 115
species with 26 unknowns (1.8 percent of total stems). We observed 54 tree species in the
transitional forest, with 4 unknowns (1.7 percent of total individuals), and there were 46
species of trees in the bajo with 5 unknowns (2.6 percent of total stems). Rare species,
defined as less than 1 individual per ha (Hubbell and Foster 1986), made up 15 percent of
the total number of species identified across our entire survey area at Tikal, and 24 percent
of species were not observed more than twice.
Table 7.1. Top ten plant families in our 5.95 ha forest survey in Tikal National
Park; N = number of stems for trees with a 6 cm (left) and 10 cm (right) or greater
diameter at breast height (DBH)

Family N Proportional Family N Proportional

Abundance Abundance

Arecaceae 954 0.185 Sapotaceae 522 0.186

Sapotaceae 730 0.142 Moraceae 491 0.175

Moraceae 686 0.133 Sapindaceae 400 0.143

Sapindaceae 591 0.115 Meliaceae 274 0.098

Meliaceae 542 0.105 Arecaceae 174 0.062

Euphorbiaceae 260 0.050 Fabaceae 148 0.053

Fabaceae 249 0.048 Apocynaceae 106 0.038

Apocynaceae 158 0.031 Euphorbiaceae 105 0.037

Myrtaceae 138 0.027 Myrtaceae 98 0.035

Anacardiaceae 120 0.023 Anacardiaceae 97 0.035

Total of top 0.860 Total of top 0.862

10 10

In the characterization of ecological communities, it has been accepted that counts of

species increase as sampling area increases. However, true species richness can be
estimated by subsampling (Arrhenius 1921; Palmer 1990; Smith and van Belle 1984).
Despite our extensive sampling efforts, which equaled or surpassed studies with similar
goals, species accumulation curves indicate that we did not capture the total richness of
species present in the forests we surveyed. When rare species were removed from our
analysis, the asymptote (the point on the curve where accumulation approaches 0)
occurred at 3 ha. The curves generated for a subset of the area at two different diameter
thresholds are provided, and these are representative of the curves in each habitat (Figure
7.1). For the subset of area for which species accumulation curves were generated,
richness was 118 (6 cm DBH threshold) and 106 trees (10 cm DBH). Estimates for
richness were 156 and 142 (second-order Jacknife) for the smaller and larger diameter
classes, respectively. The higher estimate corresponds to our richness value for the entire
area of 5.95 ha when unknowns are included. Schulze and Whitacre (1999) observed 185
tree species (minimum 7.5 cm DBH) across 12.1 ha at Tikal. The upland forest was the
focus of much of our sampling efforts, because it was an important zone for the ancient
Maya, in terms of both forest resources and agricultural land. We identified 91 and 107
species in the upland forest subsample covering 3.35 ha, with total richness in the upland
forest estimated as 131 and 152 (Jackknife 2), for trees at the 6 cm and 10 cm DBH
threshold, respectively.

Figure 7.1. Species accumulation curves (exact method) for trees and vines at the 10 cm
diameter (a, b) and 6 cm diameter (c, d) minimum threshold by 0.05 ha plot size with 95
percent confidence intervals over a 4.65 ha forest subsample at Tikal. Asymptote is
reached when rare species, occurring less than one time in the sample area (b: 21 percent;
d: 19 percent), are removed.
Species richness, although a useful indicator of diversity, is very sensitive to rare
species. Simpsons index (1-D), with a range of values from 0 to 1, predicts how likely it
is that two randomly encountered trees will be the same species. A value of 0.94 for our
entire survey area indicates high diversity and low dominance, predicting that 94 percent
of the time, if you sample two trees, they will be different individuals (Whittaker 1965). A
rank abundance curve illustrates which of these species you would most frequently
encounter and the shape illustrates a shared dominance (Figure 7.2). Cryosophila
stauracantha (Heynh.) R. J. Evans was the most frequently encountered tree across the
entire survey area at the 6 cm diameter limit. The top six trees made up 50 percent of the
community at Tikal. At the 10 cm DBH limit, Blomia prisca (Standl.) Lundell had the
highest frequency, followed by Brosimum alicastrum Sw., Pouteria reticulata (Engl.)
Eyma, and Trichilia minutiflora Standl.

Figure 7.2. Accumulated frequency of upland forest species for trees at least 6 cm DBH
in 0.05 ha plots across 4.65 ha at Tikal National Park.
The Shannon-Wiener index (H), which characterizes the relative frequency of species
in a community (MacArthur and MacArthur 1961), is more sensitive to rare species than
Simpsons index but not as sensitive as species richness. The Shannon-Weiner (H) index
is 0 when one species is present and approaches, at its maximum, the natural log of
species richness (approximately 4.8 for 124 species). H at Tikal ranges from 2.86 in the
bajo to 3.47 over our entire survey area (Table 7.2). To measure how evenly distributed are
the species in each habitat, Pielous J, an evenness index ranging in value from 0 to 1,
considers diversity proportional to a maximum diversity based on richness. Moderate to
high values of J (0.690.80), observed in our Tikal surveys, indicate that species are
distributed somewhat regularly within and among tree communities at Tikal.
Table 7.2. Diversity indices for all trees greater than 6 cm diameter at breast height
(DBH)

Habitat Area, Richness Stems Basal Shannons Simpsonb Jc

ha per Area, Ha
ha m2/ha

Bajo 0.35 46 1303 23.6 2.86 0.90 0.75

 Transitional 0.91 54 1068 28.5 3.18 0.94 0.80

Upland 4.69 115 818 38.8 3.29 0.93 0.69

Total 5.95 124 885 36.4 3.47 0.94 0.72

a H = p logp pi represents proportional abundance of species, I, in a population of indefinite size.

i i;
b 1p 2
i .
c J = H/H
max where Hmax is the log of estimated species richness.

Relative dominance, or the amount of basal area a species occupies in a community

compared to others, and stem density, or the relative number of trees each species
contributes to a community, together describe the ecological community at Tikal. Ten
species were identified as the most dominant and ten as the most abundant species. These
made up 6773 percent of the total basal area and total abundance (Table 7.3).
Table 7.3. Most abundant species by stem density and dominant species by basal
area (BA) for each habitat (6 cm DBH threshold) in our forest surveys at Tikal (5.95
ha)

Upland Proportional Upland Proportion,

Density BA

Cryosophila 16.6 Brosimum 30.4

stauracantha alicastrum

Blomia prisca 8.2

Blomia prisca 10.8 Ficus sp. 6.7

Trichilia minutiflora 10.1 Clusia sp. 5.6

Pouteria reticulata 10.0 Pouteria reticulata 5.4

Manilkara zapota 5.0

Brosimum alicastrum 7.2 Spondias mombin 4.5

Pseudolmedia sp. 5.0 Cedrela odorata 2.8

Manilkara zapota 2.2 Forchhammeria 2.7

trifoliata

Sabal mauritiiformis 2.1 Trichilia minutiflora 2.6

Forchhammeria 2.0
trifoliata

Protium copal 1.9

Total 67.9 Total 73.9

Transitional Transitional

Cryosophila stauracantha 14.7 Brosimum alicastrum 24.0

Manilkara zapota 8.1

Pseudolmedia sp. 12.0 Pseudolmedia sp. 7.1

Brosimum alicastrum 7.1 Blomia prisca 6.4

Pimenta dioica 6.7 Pimenta dioica 5.8

Trichilia minutiflora 6.7 Pouteria amygdalina 4.7

Gymnanthes lucida 6.2 Pouteria reticulata 3.3

Blomia prisca 3.7 Sabal mauritiiformis 2.9

Pouteria reticulata 3.7 Guarea glabra 2.9

Aspidosperma megalocarpon 3.5 Aspidosperma megalocarpon 2.9

Pouteria amygdalina 3.2

Total 67.5 Total 68.2

Bajo Bajo

Croton sp. 23.9 Haematoxylum campechianum 20.3

Haematoxylum 11.6 Sabal mauritiiformis 12.3

campechianum

Cryosophila stauracantha 9.9 Croton sp. 9.8

Cedrela odorata 8.1

Metopium brownei 6.6 Metopium brownei 6.4

Gymnanthes lucida 6.4 Cupania belizensis 5.0

 Manilkara zapota 5.2 Manilkara zapota 4.1

Sabal mauritiiformis 5.0 Simira salvadorensis 2.9

Margaritaria nobilis 3.5 Gymnanthes lucida 2.6

Lonchocarpus heptaphyllus 3.3 Cryosophila stauracantha 2.3

Nectandra sp. 2.8

Total 78.2 Total 73.7

Brosimum alicastrum was the dominant tree in the upland and transitional forests at
Tikal and was present to a lesser extent in the bajos. Haematoxylum campechianum L. was
the dominant species in the bajo, C. stauracantha was the most abundant tree in the
uplands and transitional forest, and Croton sp. was most prevalent in the bajo. When
importance values (the sum of relative dominance and relative density) were averaged
across habitats, B. alicastrum, C. stauracantha, H. campechianum, Croton sp., and B.
prisca had the highest scores, followed by Manilkara zapota (L.) P. Royen, Sabal
mauritiiformis (H. Karst.) Griseb., and H. Wendl., Pseudolmedia sp., P. reticulata, and T.
minutiflora. Manilkara zapota was the only species in the top ten for each of the three
habitats.
Species richness is heavily influenced by sample size so we compared our data from
Tikal to those of other studies in similar-sized plot areas across the broader Neotropical
landscape occupied by the Maya. At a location in the Maya Mountains in Belize with low-
level occupation by the ancient Maya approximately 1,000 years ago, species richness was
significantly greater than in our study. At the Maya Mountain location there were 89 and
91 species in each 1 ha plot (116 cumulative) for trees with 10 cm or greater DBH
(Brewer and Webb 2002). Two comparable plot sizes (1 ha) in our survey were found to
have 50 (Operation 1, subsample) and 49 (Operation 9) species at the 10 cm threshold (2
= 17, p < 0.0001). However, the Maya Mountains location has higher annual precipitation
than Tikal (ca. 2.5m y1 compared to ca. 1.5m y1). Similar levels of diversity are found
between the Maya Mountains and Los Tuxtlas, Mexico (88 species in 1 ha), a locale that
receives approximately 4.7 m rain per year (Bongers et al. 1988). Los Tuxtlas forests
underwent agricultural intensification in the Preclassic to Middle Classic era (1600600
BC), coinciding with landscape modification for agriculture in the Maya Lowlands
(Goman and Byrne 1998). The diversity observed in our surveys of Tikal forests more
closely resembled forests in densely settled Maya sites with similar precipitation regimes.
For example, a study area in the lowland forest of the El Pilar Archaeological Reserve,
also in Belize, was composed of 55 species in 0.54 ha (estimated from figure 1.2 in
Campbell et al. 2006) compared to 45 species each in 0.55 ha (Perdido) and 0.47 ha
(Operation 13) plots in our survey for the same diameter threshold (2 = 1.8, p = 0.18). A
list of richness values for each operation is provided in Table 7.4.
 Table 7.4. Sample areas at Tikal National Park (200910) and associated richness
of trees and vines with a 6 cm threshold of diameter at breast height; general location
information is based on a map of Tikal settlement (Carr and Hazard 1961)

Name of General Site Report Habitat Area, Species

Plot Location, Reference ha Richness
Decimal
Degrees

Operation 17.21068N Corriental Upland 1.55 70

1 Reservoir
089.61948W

Operation 17.20931 N El Xate Upland 1.03 49

9 089.61848
W

Operation 17.21751 N Perdido Upland 0.47 36

13 089.629799
W

Operation 17.013611 N El Jaguar Upland 0.81 36

14 089.007083
W

Operation 17.21077N Operation L Upland 0.10 21

L
089.62173W

Operation 17.21094 N Operation N Upland 0.10 13

N 089.61839
W

Operation 17.21056 N Operation P Upland 0.10 9

P 089.61996
W

Perdido 17.21660 N Perdido Upland 0.55 39

transects 089.63105 transects
W

Operation 17.219472N Terminos Transitional 0.81 45

15 089.57075
W
 Terminos 17.21821 N Terminos Bajo and 0.45 48
transects 089.56776 transects Transitional (0.35
W bajo)

Paleoethnobotanical Remains
The UCAPT recovered more than 756 g of carbonized plant remains, of which 79 percent
was identified at least to family. In addition, we included in our data set plant
macroremains collected by the Penn Project and identified by Lentz and Cavallaro. There
were a total of 77 distinct morphospecies in the ancient woods recovered by UCAPT and
the Penn Project at Tikal, representing 31 plant families. Sixty were identified to the level
of genus and 38 to species. The top ten families found among the archaeological plant
remains included seven that also dominated the modern forest, including Sapotaceae,
Moraceae, Arecaceae, Fabaceae, Apocynaceae, Euphorbiaceae, and Anacardiaceae
(Tables 7.1, 7.4). The dominant species among the paleoethnobotanical remains were M.
zapota (Figure 7.3) and H. campechianum, which were the source of timber for the lintels
in the temples and palaces of Tikal (Coe et al. 1986; Lentz and Hockaday 2009), but, even
when these latter contexts were not considered, they were among the top ten species.
Licaria sp. and Sebastiana sp. were among the most frequently encountered species in the
paleoethnobotanical remains, but they were not dominant in the modern forest. Pinus sp.
was the only wood species in the top ten that was not found in our surveys of the Tikal
forests but is known from savannas within the resource extraction zone predicted for
Tikal. The hardwoods identified among the paleoethnobotanical remains were mostly
upland forest species (Table 7.5), and correlations to oligarchic species of the modern
forest were revealing. Charred wood specimens were recovered from contexts that were
dated from Preclassic through Late Classic Maya periods.
Figure 7.3. Light micrographs of Brosimum alicastrum (Moraceae) transverse (a) and
tangential (b) sections at 100X and 500X, Manilkara zapota (Sapotaceae) transverse (c)
and tangential (d) sections each at 100X, and Protium copal (Burseraceae) transverse (e)
and tangential (f) at 100X.
Table 7.5. Most abundant tree species (a) and families (b) represented in the
paleoethnobotanical remains recovered at Tikal according to proportional abundance
by context; ties indicated by *

Taxon Proportional Family Proportional

Abundance Abundance

Manilkara zapota 0.113 Sapotaceae 0.216

(Sapotaceae)

Pinaceae 0.118

Haematoxylum 0.110 Fabaceae 0.098

campechianum (Fabaceae)

Lauraceae 0.093

Arecaceae 0.054

Licaria sp. (Lauraceae) 0.101 Moraceae 0.054

Pinus sp. (Pinaceae) 0.098 Apocynaceae 0.049

Pouteria spp. (Sapotaceae) 0.052 Rubiaceae 0.044

 Anacardiaceae 0.039

Nectandra sp. (Lauraceae) 0.031 Burseraceae 0.025*

Euphorbiaceae 0.025*

Protium copal 0.025

(Burseraceae)

Aspidosperma sp. 0.021

(Anacardiaceae)

Brosimum alicastrum 0.021

(Moraceae)

Sebastiana sp. 0.021

(Euphorbiaceae)

Total 0.595 Total 0.814

Correlations between Ancient Plant Remains and the

Modern Forest
Species designations as oligarchic or nonoligarchic were based on modern forest surveys.
The mean number of contexts in which paleoethnobotanical remains was found was
greater for oligarchic species than nonoligarchic species, and the representation of
oligarchs in paleoethnobotanical contexts was significantly different from the overall
mean (21,177 = 3.87, p = 0.049) while the mean abundance for nonoligarchs was not
significantly different (21,177 = 0.125, p = 0.72). When only species that occurred in both
modern and ancient contexts were considered, the relative species abundance in
paleoethnobotanical remains was positively correlated with IV of woods found in the
modern forest (Spearmans rho () = 0.292, p = 0.045, n = 46). This was driven by the
contribution of stem density ( = 0.330, p = 0.025) to IV rather than basal area ( = 0.280,
p = 0.059).
We did not observe a correlation between oligarchic species at Tikal and their
socioeconomic value. A high percentage of modern trees present at Tikal have value for
indigenous and contemporary cultures; only 5.9 percent have no known values and 44
percent have value in at least four categories (Table 7.6). The trees categorized as high
value (HV) made up 57 percent of the modern forest and 63 percent of the
paleoethnobotanical remains. There was not a significant difference in total basal area of
HV species compared to those identified as LV (p = 0.14, Mann-Whitney U) and HV
species were not better represented than LV in the paleoethnobotanical remains (21,177 =
1.09, p = 0.30).
 Table 7.6. Jaccard Dissimilarity Matrix comparing paleoethnobotanical remains
recovered at Tikal with the modern forest and each habitat, at the generic level; counts
are presence/absence; dissimilarity values range from to 1 (0 indicates identical
composition of genera; 1 indicates complete dissimilarity)

JACCARDa Paleoethnobotanical Upland Bajo Transitional

Wood Remains

Upland 0.648

Bajo 0.671 0.580

Transitional 0.729 0.483 0.500

Total Modern 0.580 0.133 0.592 0.520

Forestb

a Jaccard index: 2B/(1 + B) where B is Bray-Curtis dissimilarity (sum(abs(x x[ik]))/sum(x[ij]+x[ik])/(sum

(x[ij]+x[ik]) where i is individual species, x is species abundance, and j and k are sampling units).
b When a species was observed by Shulze and Whitacre (1999) and not in our survey, it was classified as
present.

Discussion
The hypothesis that modern forests can inform us about ancient subsistence practices in
the Neotropics has been investigated with a variety of approaches. Some researchers have
explored the relationship between oligarchic species and their economic value to modern
indigenous communities (Campbell et al. 2006; Erickson and Balee 2006). Others have
examined differences in species composition and clustering between areas with evidence
of dense ancient Maya settlement compared to areas of low settlement (Ross 2011; Ross
and Rangel 2011). Researchers have discovered evidence of selection by ancient cultures,
as phenological differences in species with higher yields of valued fruits, or in genetic
differences between cultivated and wild varieties of a single species (Gonzlez-Soberanis
and Casas 2004; Miller and Schaal 2006; Peters 2000). We approached the question with
an extensive investigation of modern forest composition coupled with identification of
carbonized remains of trees used by the ancient Maya and ethnographic records of plant
use by contemporary Maya communities (Table 7.7). We quantified the correlation
between ancient plant remains and species composition in the modern Tikal forest, as
functions of abundance and economic value, to learn more about how the ancient Maya
managed their tree resources at Tikal.
Table 7.7. Socioeconomic value of woods present at Tikal as trees in the modern
forest and as paleoethnobotanical remains; categorical uses for each sample compiled
from references1 and local informants
 Average Percentage of Species in Each Context for Each Use
Number of
Economic Construction Food Fuel Medicine Ritual Other
Uses* by
Context

Ancient 3.5 77% 51% 44% 78% 10% 83%

Modern 3.2 70% 49% 43% 71% 12% 81%

Both 3.7 79% 59% 51% 84% 14% 86%

* Six categories, shown on the right.

1 Alcorn 1984; Arvigo and Balick 1993; Atran et al. 2004; Balick et al. 2000; Campbell et al. 2006; Cannon and
Cannon 1989; Henderson et al. 1995; Macbride 1951; Mutchnick and McCarthy 1997; Rico-Gray et al. 1991;
Robyns 1963; Roys 1931; Standley et al. 194677 (Flora of Guatemala); Williams 1981.

We found high levels of tree species diversity in the modern forest and a positive
relationship between the paleoethnobotanical remains of the ancient Tikal residents and
the oligarchs of the modern forest. We did not detect a significant difference in the number
of economic species (as determined by ethnographic accounts) among oligarchic species
in the modern forest, undoubtedly because an exhaustive search of the literature on the
economic values of Neotropical tree species to modern cultures suggests that almost every
tree in the forest has value. This study demonstrates that employing analyses that include
paleoethnobotanical remains increases our ability to discern the impact of the ancient
Maya on Neotropical forests.

The Importance of Paleoethnobotanical Remains

Plant remains recovered from middens and other features associated with residential
structures provide physical evidence of past plant resources and are crucial to
understanding how the ancient Maya supported a complex culture (Hood 2012; Lentz
1999; Miksicek 1983, 1987). The correlation between abundance in paleoethnobotanical
samples and dominant trees at Tikal demonstrates that the modern forest may serve at least
as a partial proxy for our understanding of ancient Maya plant use. Modern oligarchic
species (the top ten species in abundance and dominance in our surveys) were recovered
from ancient contexts more frequently than would be randomly expected, suggesting a
correlation between the dominant tree species in the modern forest and the tree species
used by the ancient Maya. It further suggests that the ancient Maya were actively
managing their forest resources. Otherwise the overuse of these valued species would have
caused a decline in their dominance. Even low-impact extraction of forest resources has
been shown to have significant effects on forest structure. Forests that have undergone
timber extraction have been observed to have a significantly lower basal area of once-
dominant trees than forest reserves ten years later. A reduction of trees in the smallest size
classes in logged forests also disrupts the regeneration potential of those trees that were
removed with potential for long-term consequences on forests that are not well managed
(Gutirrez-Granados et al. 2011). The careful management of forest resources would have
been only one of the strategies of the ancient Maya in sustaining their community at Tikal.
Plant resources of the ancient Maya at Tikal included not only what are now modern
oligarchs, but also species, such as cacao and cashew, not native to the region. These may
be present as a result of the expansion of the natural resource base through trade (Lentz et
al. 2005; McKillop 2002) or may have been cultivated in specialized microenvironments
such as the perennially moist natural sinks where no evidence for reservoir development
was revealed by the UCAPT in this study (N. Dunning, personal communication, 2013).
This investigation shows that characterizing tree communities at ancient Maya sites can be
informative in interpreting ancient silvicultural practices particularly when carried out in
conjunction with paleoethnobotanical evidence.

Dynamics of the Modern Forest

We observed high levels of tree diversity at Tikal, comparable to other sites within ancient
Maya regions with similar rainfall (11.8 m y1) and extended dry seasons (Brokaw and
Mallory 1993; Campbell et al. 2006; Hernandez-Sefanoni and Ponce-Hernandez 2002;
Neisheim et al. 2010). Diversity is lower at Tikal than the Maya Mountains and Los
Tuxtlas forest in Mexico, but this difference may be explained by a lower diameter
threshold for tree sampling in the Maya Mountains and higher precipitation in both forests
(Bongers et al. 1988; Brewer and Webb 2002), which increases species diversity
(Clinebell et al. 1995). In other studies, species diversity and evenness are not
significantly different between areas of high and low settlement density by the ancient
Maya or between Maya Lowlands and other heterogeneous forests (Ross 2011; Wiseman
1978). Animal dispersal of seeds, a dominant mechanism in tropical forests, can advance
trees into new areas more than a kilometer away in one generation (Dick et al. 2008;
Hardesty et al. 2006; Howe and Smallwood 1982; Murray 1988), and pollen transfer over
a distance of 14 to 75 km is possible (Dawson et al. 1997; Dick et al. 2008). This offers
some reassurance that the rare species we identified are part of a widespread community
and not simply a vulnerable genetic resource (Gitzendanner and Soltis 2000).
What emerges is a complex picture that cautions us in interpreting how a culture,
however large and long-lasting, influenced a forested landscape that was largely
abandoned more than a millennium ago (Lambert and Arnason 1982; Miller and Nair
2006) without supporting data from the paleoethnobotanical record. Disturbances from
more recent natural or anthropogenic events are visible in the modern forest composition
of Maya sites where gap turnover and the ages of successional species can be estimated
(Brokaw and Mallory 1993), but the nature of disturbances extending from the time of the
last Maya occupation (ca. AD 900) would be challenging to identify in the modern forest.
The distribution of species at Tikal can be largely explained by environmental conditions
rather than settlement history, but this does not rule out the possibility that the overall
forest composition was influenced by the Maya. Rather, it suggests that ecological
processes over the past millennium may have obscured evidence of ancient Maya forests
as their structure evolved (Schulze and Whitacre 1999). Nevertheless, our study
demonstrated that many of the same species heavily utilized by the ancient Maya, as seen
in the paleoethnobotanical record, are today oligarchic species in the modern forest. If the
Maya were overexploiting useful trees, as occurred with mahogany extraction in Belize
during the nineteenth and twentieth centuries, there would have been substantial
reductions in the populations of these once-dominant trees (Dobson 1973; Downie 1959;
Thomson 2004). Thus, it seems likely that the ancient Maya at Tikal did not carelessly
strip away their essential forest resources without regard to future needs.

Socioeconomic Values of Trees at Tikal and Ancient Remains

We did not detect an association between high economic value, as determined by modern
ethnobotanical accounts, and dominance in the modern forest. This contrasts with other
studies that detected significant differences in the economic utility of dominant forest
species when forests in areas heavily populated by the ancient Maya were compared to
forests in sparsely populated areas (Campbell et al. 2006; Ross 2011; Ross and Rangel
2011). This may be due to the smaller sampling areas in previous studies or the methods
of assigning economic value. In our study, more than 94 percent of all species were found
to have at least one known economic use, as determined by an extensive literature review
of ethnographic studies and our own informants in the field. Any attempt at categorizing
trees as more or less useful to the ancient Maya is imperfect and, as more references are
cited, distinctions are more difficult to draw. Assigning economic status to forest trees is
often accomplished by using inventories of modern Maya gardens and orchards. This
offers an incomplete picture of Maya agroforestry practices, one that can be extended by
the paleoethnobotanical record. The diverse tree species that were valued and cultivated
by the ancient Maya, as indicated by ancient plant remains, are not all represented in the
modern mature forest. One explanation for this is that when the management of trees in
gardens and orchards in Neotropical forest environments ceases, the continued success of
each species is dependent on its ongoing regeneration through seedling establishment in a
competitive ecological system. Trees that were once actively cultivated within cleared or
managed forests, and later abandoned, are often replaced by native forest species (Peters
2000).
Modern indigenous systems of agroforestry mirror the natural forest ecosystem, and
these practices may have been inherited from ancient management systems so it is useful
to examine these traditions to help us understand why we do not see a correlation between
dominance and economic value (Miller and Nair 2006; Hawkes 1983). Could it be that the
ecological value of the forest as a whole and the maintenance of diversity were as
important to the Maya as the individual species on which they depended? Contemporary
Maya have managed their plant resources in diverse ways that include gardens, managed
fallows, and semicultivated forests. These settings contain high species diversity that
protects soil resources, promotes flexibility in uses of the plant resources, and increases
their capacity for responding to changing family or community needs (Diemont et al.
2006; Ford 2008; Levasseur and Olivier 2000; Peters 2000). Cultural preferences play a
large role in modern Maya and other indigenous communities in determining garden
species composition (Alcorn 1984; Hastorf 2006; Mercer et al. 2005; Rico-Gray et al.
1990). The large population of ancient Tikal may have encompassed an array of
subcultures and traditions due to class distinctions, immigration, and societal differences
(Haviland 1967; Scherer 2007; Willey 1982; Wright 2005), and this may have been
reflected in a diversity of plant preferences and locally distinct plant communities.
Traditional medicinal practices of modern Maya cultures illustrate the need for high
diversity in available plant species. Plant resources are subject to experimentation and the
development of new remedies based on knowledge of local resources and plant traits
(Pesek et al. 2010; Rico-Gray et al. 1991). The success of the ancient Maya surely
depended on their ability to avail themselves of the ecological, nutritional, medicinal, and
a plethora of other resources that would be present in a diverse forest (Pesek et al. 2010;
Milburn 2004; Paoletti 1995).
The Itza Maya, a community of Maya descendants currently living in the Petn region,
value and carry on the traditions of their ancestors (Atran et al. 2004; Morley 1947). They
recognize not only the socioeconomic value of trees but the ecological and spiritual
importance of the forest, as well. The Itza word ti-ikal means the place of the spirits, and
individuals are motivated to protect forest species according to their understanding of the
spirits preferences (Atran et al. 2004). A diverse forest may have been the result of the
ancient Mayas consideration of these same factors: valuing the forest as an ecological
entity and appeasing the spirits it housed. Recognition of what we now call ecosystem
services (shade, watershed protection) as well as the products provided by trees increases
the complexity in identifying one group of species as more economically valuable than
another in a holistic system (Costanza et al. 1997; Dumont 2005; Tundisi and Matsumura-
Tundisi 2008). Elite control over tracts of forest for pleasure, hunting, or religious
purposes combined with a tradition of fear and respect for spiritual forces that protect the
forest might have served as effective motivation for conserving forested areas for centuries
(Atran et al. 2004; Brown and Emery 2008; Masson 1999; McNeil et al. 2010). The scope
of the presence and activity of indigenous cultures when Spaniards arrived in the
Americas has been underestimated (Denevan 1992).

Distinctions between Ancient Plant Remains and the

Modern Forest at Tikal
Seven of the plant families that are most frequently observed in the paleoethnobotanical
remains are also dominant in the modern forest tree community. Lauraceae was found in
abundance in the plant remains, but was not a dominant family in the modern forest.
Lauraceae may have been a popular construction resource at Tikal, preserved when homes
were burned accidentally, ritually, or for reconstructive purposes. It may also have been
used as a fuel resource as it is by contemporary Huastec Maya (Alcorn 1984). Three
genera in the Lauraceae (Nectandra, Licaria, and Ocotea), each represented in the
archaeobotanical remains at Tikal, are described as construction resources by modern
Maya (Atran et al. 2004; Standley and Steyermark 1946). Nectandra and Ocotea are
utilized for posts and beams and Licaria (Figure 7.4) is a general construction resource
(Alcorn 1984; Lundell 1938; Williams 1981). Secondary regrowth of Nectandra in milpas
has been utilized for firewood by the Huastec Maya (Alcorn 1984). The abundance of
Lauraceae wood in the paleoethnobotanical record suggests that it was heavily exploited
by the ancient Maya at Tikal. Persea americana Mill., a tree valued for avocado fruits, in
the Lauraceae family, was also represented in the paleoethnobotanical remains from Tikal
as carbonized pits. The silvicultural practices of the ancient Maya undoubtedly included
the cultivation of this fruit tree.

Figure 7.4. Scanning electron micrograph of Licaria campechiana (Lauraceae)

recovered from a midden in El Jaguar (Operation 13). Images include transverse sections
at magnifications of 50X (a) and 100X (b) and tangential sections at 100X (c) and 500X
(d).
Burseraceae was another plant family that was more prevalent in ancient contexts than
in the modern forest at Tikal. Protium copal (Schltdl. and Cham.) Engl., an economically
important member of the Burseraceae family, has long been valued in Maya religious
ceremonies and for its medicinal properties (Lundell 1938). Although the Burseraceae
family was not dominant in the modern forest, Protium copal was one of the top ten
species in abundance in the modern upland forest (Table 7.3), and it also was one of the
ten most common species among the archaeological plant remains at Tikal.
Pinaceae is a family with high economic importance in the Maya area, noted as a source
of construction timber, resin, and firewood (Atran et al. 2004; Standley and Steyermark
1958; Williams 1981). Pinus caribaea Morelet is common to savannas in northeastern
Guatemala, where it grows in association with Attalea cohune Mart. palms, which were
also represented in the ancient plant remains at Tikal. Some of these savannas exist in the
Bajo de Santa Fe within the projected extraction zone of Tikal (Lentz et al., this volume;
Dunning et al. 2002; Fialko 2000). It was an important ceremonial resource for the ancient
Maya and may also have been a resource in the trade networks of the ancient Maya at
Tikal, as indicated for other Late Classic Maya sites (Morehart et al. 2005; Lentz et al.
2005).
The Rubiaceae family is more abundant in the paleoethnobotanical remains at Tikal
than in the modern forest and encompasses high diversity within the family in both
settings. Alseis sp., Morinda sp., Genipa americana L., Guettarda combsii Urb.,
Psychotria sp., and Zanthoxylum caribaeum Lam. were identified in the archaeobotanical
remains; Alibertia edulis (Rich.) A. Rich. ex DC., G. combsii, Psychotria carthagenensis
Jacq., Psychotria lundellii Standl., Psychotria mexiae Standl., and Simira salvadorensis
(Standl.) Steyerm. were in the modern forest. These species have a variety of uses
including fruit for wine making, thatch support or roofing, natural beehives, dyes, and
medicine (Atran et al. 2004; Balick et al. 2000; Campbell et al. 2006; Mutchnick and
McCarthy 1997; Nash 1976; Roys 1931). Their higher abundance in the
paleoethnobotanical record may be a result of their increased success under the
management practices of the ancient Maya.

Dominant Woods from Modern and Ancient Settings

There were many similarities between the makeup of the dominant families in the modern
forest and the paleoethnobotanical remains at Tikal, but we will focus on just two
examples: the dominant family for both settings, Sapotaceae, and the species with the
highest IV in the modern forest that was also abundant in the plant remains, B. alicastrum.
Sapotaceae was the most abundant family in both our modern and ancient surveys of Tikal
plants. It is a pan-tropical family known for its sweet fruits, durable timber, and useful sap
(Pennington 1991; Hill 1937). Pouteria reticulata (Engl.) Eyma and M. zapota were
among the top ten species in both abundance and dominance by basal area across habitats
of the modern forest, and Pouteria amygdalina (Standl.) Baehni was dominant in the
transitional forest. Manilkara zapota was the only species that was dominant in each
habitat we surveyed and also held a prominent place in the temple architecture (Coe et al.
1986; Lentz and Hockaday 2009). In addition to Manilkara and Pouteria, Chrysophyllum
and Sideroxylon were found in both the modern forest and as ancient plant remains at
Tikal, reinforcing the concept that the Sapotaceae was a diverse family valued by the
ancient Maya. The ubiquitous nature of the family in the modern forest and their ancient
plant remains strengthens our hypothesis that ancient Maya silvicultural practices are
reflected in the contemporary tree composition of Tikal.
Brosimum alicastrum had the highest overall importance value in the modern forest at
Tikal and was heavily represented as carbonized wood in ancient contexts. If the modern
forest and paleoethnobotanical remains together explain past practices of the Maya, then
B. alicastrum was a tree valued by the Maya at Tikal (Puleston 1968, 1973). It has been
suggested that changes in breeding systems and increases in reproductive output of B.
alicastrum at Tikal, compared to other parts of its range, may have been consequences of
artificial selection to increase yield (Peters 1983). The nutritional value of its seeds is well
known to the local people and they have been a food valued by the contemporary Maya,
who have protected B. alicastrum in forests and maintained them in gardens (Atran et al.
2004; Benjamin et al. 2001; Corzo Marquez and Schwartz 2008; Ford 2008; Gillespie et
al. 1993; Gomez-Pompa et al. 1987; Mercer et al. 2005; Thompson 1930). The
paleoethnobotanical evidence of its use as a food source is limited, having been previously
found at Cob (Beltrn Frias 1987), in a Postclassic burial at Tikal (Lentz, personal
communication, 2013; Moholy-Nagy 2003), and possibly in a Late Terminal context at
Tikal (Miksicek 1983). The paucity of seed remains may be due to their preparation in
food production; they are ground into meal or boiled and eaten as vegetables (Peters and
Pardo-Tejeda 1982; Thompson 1930). Such processing makes foods infrequent in the
archaeobotanical record (Miksicek 1987). Brosimum alicastrum charcoal remains were
found in multiple contexts at Tikal, and it was certainly at least a wood resource. This
debate serves to underscore the complexity in synthesizing the paleothnobotanical,
ecological, and ethnographic records to form a comprehensive picture of the ancient Maya
use of plants over two millennia.
Conclusions
Our study highlights the importance of coupling modern forest surveys with
paleoethnobotanical data to inform our understanding of plant use by the ancient Maya. A
highly diverse forest at Tikal, composed of many species with economic value, may be a
reflection of the type of forest that attracted the Maya to this area. Many of the tree species
they utilized are heavily represented in the modern forest, and some, found only in ancient
remains, may have been displaced with the cessation of active management or could have
been a commodity in economic trade networks between Tikal and other communities.
Forest management activities are likely to have evolved over time, as did other landscape
practices (Scarborough and Burnside 2010). A dynamic demographic, cultural, political,
and environmental climate throughout the long period of Maya occupation at Tikal must
have demanded changes in resource management over time. The influence of the tree
species native to the area and domesticated species introduced with migration, coupled
with the needs of the inhabitants, affected the evolving relationship with the forest. That
reality, in conjunction with long-range seed and pollen dispersal of Neotropical trees,
challenge interpretations of ancient Maya plant use based only on the ecology of forest
remnants surrounding ancient Maya settlements. The diversity of the ancient Maya
landscape is better understood when we include tree species represented in the ancient
plant remains. Nevertheless, our data comparing preserved wood from ancient contexts
with the modern forest structure affirm that there is a cooccurrence of the oligarchies of
the forest and the plants used by the ancient Maya. We hypothesize that this observed
cooccurrence is the result of ancient Maya silviculural practices that were essentially
conservative in nature. Those practices of the past may be similar to those observed
among contemporary Maya (Ford and Nigh 2009; McAnany 1995) so that the supply of
trees useful to them would not be depleted. The ancient Maya of Tikal did carefully
manage their forests and left a legacy in the trees.
Notes
1 Primers for DNA amplification were GGTTCAAGTCCCTCTATCCC (forward) and
ATTTGAACTGGTGACACGAG (reverse) for the transfer RNA intergenic spacer (trnL-
trnF) region in the choloropast (Taberlet et al. 1991); TCCGTAGGTGAACCTGCGG
(forward, ITS1) and GCTGCGTTCTTCATCGATGC (reverse, ITS2) for the nuclear
ribosomal internal transcribed spacer (nrITS) (White et al. 1990).
 8 Agroforestry and Agricultural Practices of the
Ancient Maya at Tikal
David L. Lentz, Kevin Magee, Eric Weaver, John G. Jones, Kenneth B. Tankersley,
Angela Hood, Gerald Islebe, Carmen E. Ramos Hernandez and Nicholas P. Dunning

Introduction
The Lowland Maya are a people whose culture and society have been shaped by
development within the Neotropical forests of Central America (Schwartz 1990; Taube
2003). The forest management and agricultural practices employed by the Precolumbian
Maya of Tikal were essential to the survival and prosperity of the polity over many
centuries. Ultimately, the bulk of food, fuel, and other critical materials necessary for the
functioning of the community were obtained from local forests and agricultural plots. This
study was designed to evaluate the fundamental resource base that enabled the Tikal Maya
to support a large population and complex social order during their zenith in the Late
Classic period. Moreover, we attempt to create a more complete understanding of the
basic economic underpinnings of one of the great Maya polities and how its support
system may have been overwhelmed by demographic, political, and climatic trends in the
late ninth century AD.
Models of Maya land use have evolved from a suggestion that the Maya used simple
slash-and-burn techniques (Thompson 1954) to those that suggested that they relied on
orchards (Folan et al. 1979; McKillop 1994), artificial rain forests (Wiseman 1978),
kitchen gardens (Ball and Kelsay 1992; McAnany 1992; Robin 1999), a managed
mosaic (Fedick 1996), or a forest resource management practiced in concentric zones of
intensity from a residential core outward (Killion 1990; Kunen 2004; Lentz et al. 2002).
Most of these models are theoretical and based on varying degrees of ethnographic and
archaeological data. In contrast, we offer a multicomponent landuse model that is based on
empirical evidence recovered from a combination of archaeological, paleoethnobotanical,
ecological, ethnographic and geographic information sources to support our
conceptualization.
From previous studies at Maya sites, such as Cob (Beltran Frias 1987), Cern (Lentz et
al. 1996), Dos Pilas (Lentz 1994), Aguateca (Lentz et al. 2014), Wild Cane Cay (McKillop
1994), Copn (Lentz 1991), Tikal (Lentz and Hockaday 2009), and Rio Azul (Hurst et al.
1989), we do have some background regarding the plant utilization practices employed
during the florescence of Late Classic Maya communities. To be sure, these studies have
shown that the ancient Maya used an assortment of field crops and tree species as
important components of both their domestic and political economies. How their crops
were grown and forests were managed, however, is a question that has been more elusive.
One of the principal objectives of this paper is to present a land use model that draws
on a variety of data sources and accurately reflects the interaction between the ancient
Maya of Tikal and their tropical forest environment. As a focus of this effort, the major
research question to be addressed will be: How was the landscape managed at Tikal to
provide the needed food, fuel, and structural material for the polity at the height of its
population? An important step in attempting to answer this question will involve an
assessment of the size, in terms of area, of the settlement of Tikal during various phases of
occupation. To narrow the scope of this broad question, we will focus on the Late
Preclassic and the Late Classic periods. Fortunately, the extent of the ancient built
environment at Tikal has been well studied and numerous scholars have attempted to
estimate the population of the polity during the various periods of occupation. As such, we
have a substantial database from which to recreate hypothesized zones of habitation and
resource extraction.

Methodology
During two field seasons at Tikal, we surveyed modern forests, analyzed satellite images,
cored reservoirs and aguadas for pollen evidence, and collected botanical macroremains
from archaeological excavations, mostly from middens, to evaluate changes in the
environment and plant use practices through time. Our goal was to assess the extent of the
landscape and determine how much of it was forested, cleared for agriculture, and
otherwise occupied by residential, ceremonial, and other constructed features common in
an ancient Maya city.

Paleoethnobotanical Studies
To learn about the vegetation of the past, we took numerous sediment samples from the
ancient reservoirs, aguadas, and other surfaces and extracted pollen from these samples.
Also, we analyzed archaeological soils for their chemical composition and conducted bulk
13C analyses of soil samples to look for traces of past maize agriculture. To help locate
middens, we conducted systematic phosphate tests in four plazuela groups to identify
areas of high phosphate content (Figure 8.1). These tests enabled us to find several
middens with significant quantities of carbonized plant material that were retrieved as
macroremains or through soil samples that were processed using water flotation (Figure
8.2). These plant remains inform us about the forest and agricultural products used by the
Maya inhabitants of Tikal and form the basis of a larger discussion about Maya resource
management activities.
Figure 8.1. Diagram of Grupo de Jaguar at Tikal including structures 7D20-44. Soil
samples collected at each grid corner were analyzed for extractable P. Test pits were
excavated in areas with high P levels. Using this method, we were able to locate three
middens, areas with high concentrations of archaeological plant remains. (Drawing by R.
Terry)
Figure 8.2. Flotation apparatus used to extract archaeological plant remains from soil.
Lake Petn Itza was used as the water source; the device is being operated by Angela
Hood and Santo Chico (Photograph by David Lentz).

Modern Plant Surveys

Ecologists have described the dominant vegetation in this section of Central America as
tropical rain forest (Wagner 1964), evergreen tropical forest (Rzedowski 1981; Greller
2000), moist semitropical forest (Holderidge et al. 1971), or tropical semideciduous forest
(Pennington and Sarukhan 1968). Floristic studies of the area have been conducted by
Millspaugh (1895, 1904), Standley (1930), Lundell (1937), and Schulze and Whitacre
(1999). Probably Holdridges definition is the most workable because it can be used to
compare the vegetation of the Tikal area to other parts of the Maya Lowlands and the
Neotropics in general.
The Tikal region is forested country of low, rolling hills with open drainage and fertile,
black soils. The dominant vegetation in the upland areas is tropical forest typical of the
Petn. These areas contain some of the most fertile soils in the region, and, if the forest
canopy is removed, the most highly productive land in terms of agricultural potential, a
circumstance of which the ancient Maya were apparently well aware. Another vegetation
type common in this region, as described by Andrews IV (1943), is the swamp forest
characterized by low headlands and hills with logwood swamps. Whether or not the hearts
of logwood swamps or bajos was arable (e.g., Fedick and Ford 1990; Ford 1986) has
been a topic of considerable debate (see chapters by Dunning et al. and Webster and
Murtha, this volume).
To gain a better understanding of the Maya agroforestry practices of the past, it was
important to acquire a firsthand understanding of the modern forests that now blanket the
ruins of Tikal to create a set of reference points for the estimation of the ancient forests.
During the 2009 field season, our forest transects (a series of 500 m rectangular plots)
were established in areas north and south of the Corriental Reservoir. Plots for the 2010
field season were located to the east of the Grupo de Trminos in the Bajo de Santa Fe
(www.cambridge.org/Tikal; See S. Fig. 6) and to the east and south of the Perdido
Reservoir. All plots were demarcated using a Magellan Platinum GPS unit. In addition to
the rectangular plots, four large survey plots of forested land were recorded (two were 100
m x 100 m and two were 50 m x 50 m). All trees and vines within each plot with a
diameter greater than 6 cm at breast height were measured for diameter, height, and exact
location (Figure 8.3). Herbarium voucher specimens from unknown trees were collected
for accurate identification. Plant collections, consisting of leaves, fruits, and flowers, were
obtained using pole pruners. Pressed specimens were dried in an electric plant drier
constructed out of locally available materials. Almost 6 hectares of forest were surveyed
during the two field seasons. Further details of the forest survey methods can be found in
project reports submitted to the Instituto de Antropologa e Historia de Guatemala (Lentz
et al. 2009, 2011).

Figure 8.3. Kim Thompson, Demetrio Crdova, and Marielos Corado conduct botanical
surveys in the modern tropical forest in Tikal National Park (Photograph by David Lentz).
Our vegetation surveys conducted in the Tikal National Park provide essential
background data for the quantitative analysis of satellite images of the entire catchment
area. Once analyzed, these images, when coupled with modern forest survey data, enable
us to prepare biomass estimates for the modern Tikal forests as well as the forests of the
ancient Maya. Complete vegetation maps of the modern area serve as baseline data for
reconstructing vegetation cover of the past.

Forest Biomass Estimation

GIS techniques and satellite images were employed to estimate the biomass of the various
forest types in the area around Tikal, determine the extent of those forest types, and
calculate the extent of the zone of extraction of the polity during the Late Preclassic and
Late Classic periods. To aid in the calculation of the aboveground biomass (AGB) of
modern Tikal forests, a March 27, 2003, Landsat 7 ETM+ image of the study area was
acquired from the USGS Global Visualization Viewer (GLOVIS; http://glovis.usgs.gov/)
to provide spectral data from which indices and data transforms could be derived. AGB
was calculated by superimposing biomass parcels from vegetation surveys onto the remote
sensing data sets and experimentally regressing AGB values against the remotely sensed
variables. Because many of the vegetation parcels overlapped multiple pixels, the mean
value was calculated for pixel values for each parcel (Figure 8.4).1 The imagery data set
mean values were tabulated with AGB values and imported into a statistical package
(SPSS v.16).

Figure 8.4. For upland biomass vegetation calculations, all possible combinations of
relationships between bands and variables were explored using multiple linear regression.
We found the strongest relationship of the biomass of upland tropical forest and forest on
the margins of wetlands to be the atmospherically corrected reflectance of Landsat Band 5.
The strongest correlation between the each parcels AGB and band 5 was found using a
cubic curve (Drawing by K. Magee).
Because the relationship between predicted and measured biomass derived from
vegetation indices, statistical bands, and linear decompositions (i.e., tasseled cap
transform) can differ greatly in both strength and direction between regions (Foody et al.
2003), separate models were experimentally determined for upland tropical forest and the
wetland, or bajo, vegetation parcels.2 Both equations were entered into ENVI 4.7 and
related back to the values for each vegetation parcel to determine mean AGB error.
Because vegetation parcels and pixels do not perfectly overlap, each parcel was compared
to an areally weighted sum of AGB values of its overlapping pixels to better reflect the
model response to each geophysical location. Each AGB modeled pixel was summed and
divided by the areal extent of its respective vegetation community and the respective error
bounds were calculated.

Ancient Boundaries of Tikal

Once the aboveground biomass was calculated for modern forests, we turned our attention
to the extent of forested land accessible to the Tikal Maya in the past. Determining the
territory controlled by an ancient polity is a challenging problem, and one for which only
an imperfect answer can be formulated (Garrison and Dunning 2009). Here we use the
term territory to connote the physical area under a sites control for the purpose of
resource extraction (Marcus 1993). For the Maya, hieroglyphic inscriptions sometimes
offer partial information about interpolity relationships including glimpses of political
hierarchies, warfare, and royal marriages (e.g., Martin and Grube 2008). Such texts,
however, even where preserved, seldom contain information relevant to the functional
economy of the ancient Maya; nor do they explain the nature and extent of a politys
effective resource extraction zone. Thus, we are left to reconstruct ancient economic
spheres with archaeological proxies.
One approach to modeling ancient interregional political and economic spheres of
influence is through the spatial analysis of archaeological settlement patterns. Specifically,
the relative size of ancient communities at various points in time offers information about
the labor and resources each controlled. Given the general similarity in residential density
at the majority of Maya sites, the areal extent of settlement can be used as a rough proxy
for community population size. Similarly, the volume of monumental architecture
constructed at a site during a given time can be used as a proxy for the amount of labor
and resources controlled by the sites rulers.
For ancient Tikal, we have attempted to reconstruct the size of its territory or primary
economic extraction zone for two points in time: circa AD 100, the apogee of the Late
Preclassic period, and circa AD 700, the high point of Tikals regional influence in the
Late Classic period. In both instances, we have used available archaeological and
epigraphic data to determine Tikals nearest neighbors that were large enough to be at least
partly autonomous and exert an appreciable degree of economic competition. Since
evidence indicates that none of these neighboring sites was either as large or as dominant
as Tikal during the respective periods, we posit that their zones of economic control did
not extend as far as that of Tikal, or some of their economic production went toward
supplying the needs of Tikal. Hence, for modeling purposes we have weighted the extent
of territories in favor of Tikal.
The amount of chronological and other information available for Tikals neighbors
varies tremendously. Data for Uaxactn (e.g., Ricketson 1937; Valdes 1986) and Yaxha
(e.g., Rice 1981; Cabrera 2000) have been available for many years. New excavation data
are now forthcoming for El Zotz and El Palmar (Houston 2011), Nakum (e.g., Hermes and
Caldern 2003), and Xultun (Saturno et al. 2012). The available data for other neighboring
sites are more preliminary, however, and are derived chiefly from early expeditions (e.g.,
Morley 1938; Bullard 1960), the University of Pennsylvania Tikal Project peripheral
surveys (e.g., Fry 1969, 2003; Puleston 1973, 1983), IDEAH site inventories (e.g.,
Quintana and Wurster 2001; Quintana and Noriega 2006), and surveys of the Rio Holmul
drainage and intersite transects between Tikal, Yaxha, and Nakum (e.g., Ford 1986; Lou
1997; Fialko 2000, 2005; Fialko et al. 2006). The Tikal Project surveys and those of Ford
and Fialko included some test pit excavations to obtain chronological data, whereas the
other projects examined only surface features, inscribed monuments, or looters trenches.
 For the Late Preclassic period (ca. AD 100), the following sites were Tikals nearest
competitive neighbors: Zocotzal, El Palmar, El Encanto, Jimbal, and Chalpate (also
known as Ramonal). In the Late Preclassic, each of these centers included between two
and four architectural attributes associated with major centers: triadic groups, pyramid
complexes, ball courts, and intrasite sacbeob (causeways). For the Late Classic (ca. AD
700), most of these sites show evidence of having been either abandoned or subsumed as
suburban or satellite centers of Tikal as it expanded its political and economic sway. For
this later period, Tikals likely nearest competitive neighbors lay at greater distances:
Motul de San Jose, El Zotz, Uaxactn, Xultun, Dos Aguadas, Nakum, Yaxha, and Ixlu.
While there is hieroglyphic evidence suggesting that several of these centers were either
allied with or under the political influence of Tikal, the large populations of these centers
would likely have required the use of a large portion of their own territorial resources.
The Voronoi diagram, also known as a Thiessen polygon, has been used as a tool in
settlement analysis to determine boundaries in other Lowland Maya territories (e.g., de
Montmollin 1988). Though far from precise, this method has been widely used. The
primary concern with earlier versions of Voronoi diagrams has been that the method
required all geographic centers to have equal influence, while in fact there were many
incidences where the relative political power of individual centers was unequal. Although
solutions to this obstacle have been suggested since 1980 (Boots), the application of a
weighted Voronoi polygon model, as presented here, has only recently begun to be
practical with increased processing speeds of computer systems and integration with
geographic software programs.
In our study, a weighted Voronoi diagram utilizing a Geographic Information System
(GIS) was employed to determine the approximate projected boundary of Tikal. In our
analysis, the weight of the Euclidean distance from each point in relationship to all other
points determines the boundary of each center. Tikal was given a two-thirds to one-third
weight over its neighbors. This allowed for Tikal to have a greater influence in defining
territorial boundaries than any of its neighbors on the basis of the rationale noted earlier.
The perimeter of the occupied area of Tikal, defined as the outer limits of the polity
where the remains of house structures could be observed, was described by Robert Fry in
a 2003 article that expanded upon the work of Dennis Puleston (1973) and others from the
University of Pennsylvania Project at Tikal. While this is a useful assessment because it
captures an occupational perimeter for Tikal that is both physically observable and
archaeologically defendable, it is our position that the Fry boundary does not fully
encompass the area from which inhabitants of Tikal could have drawn their vital
resources, in particular forest products and agricultural produce. It seems likely, and we
believe our data will demonstrate, that the extractive area of Tikal must have been larger
than the Fry boundary to have supported the population at its maximum during the Late
Classic period.

Results
Using the data from the forest inventories, we were able to calculate the average
aboveground biomass (AGB) for the modern upland forest areas as 28.9 2.6 million
kgkm2. This compares well to findings of other studies of Neotropical forest tracts in
Central America and South America (Brown et al. 1992; McWilliam 1993; Laurance et al.
1999; Drake et al. 2003; Steininger 2003). As anticipated, the AGB of the bajo areas
around Tikal (18.2 0.523 million kgkm2) was far lower than the upland forest because
of the lower density of tree species and the more diminutive canopy height of the swamp
forests.
Results of the Voronoi calculations can be seen in Figure 8.5. According to these
findings, during Late Preclassic times the Tikal Maya had a territory that included 256 km
of what is now upland forest and 40.0 km of bajo forest. The Late Classic boundaries
were much larger; they included 855 km of upland area and 256 km of bajo. Within the
Late Classic boundaries of Tikal, the built component occupied by the inhabitants, as
defined by Fry (2003), consisted of 165 km, which was located almost entirely within the
upland portion of their territory. From the Preclassic to the Late Classic period, the Tikal-
controlled upland area expanded to roughly 330 percent of its previous holdings, whereas
the controlled bajo area grew by 640 percent.

Figure 8.5. Voronoi diagram of hypothetical limits of extractive areas and boundaries
surrounding Tikal during the Late Preclassic and Late Classic periods. (Diagram from N.
Dunning and E. Weaver)
Pollen data were used to evaluate the extent of the Late Classic forests in the Tikal area
as compared to the modern forests. From our modern forest surveys, we have robust
estimates of the aboveground forest biomass, and by creating ratios from local pollen
profiles that compare the modern pollen rain to the pollen counts from the Late Classic
period, we can estimate the biomass of the Late Classic forest. One of the pollen profiles
collected as part of this study was from an aguada in the bajo east of Tikal (Aguada Vaca
de Monte; see Figure 6.1). The full profile can be seen in Figure 8.6. A second pollen
profile is from Lake Petn Itza, which is to the south of Tikal and was collected as part of
a previous study (Islebe et al. 1996). These pollen cores were selected because they
bracket the extractive zone of Tikal. The prevailing winds in the region are out of the
northwest (Dunning et al. 1998) so the use of these two pollen profiles is reasonable as
they are both downwind of critical areas we have defined as land used by Late Classic
Tikal residents.

Figure 8.6. Pollen profile from Aguada Vaca de Monte. An AMS radiocarbon date of
AD 700 was recorded from the base of this core (37 cm). (Drawing by J. Jones)
Using ratios of arboreal versus nonarboreal pollen from relevant periods compared to
modern pollen counts, the data from Vaca de Monte indicate that there was a 3253
percent reduction in arboreal pollen production (reflecting conditions in the Bajo de Santa
Fe), and from the Lake Petn Itza core there was a 6070 percent reduction (reflecting
principally upland areas) during the Late Classic period as compared to modern times.
Coupled with pollen data, paleoethnobotanical macroremains can inform us about the
kinds of crops and forest products the ancient occupants of Tikal were exploiting as well
as their forest management and agricultural practices.

Discussion
Through the use of our modern plant surveys, paleoethnobotanical data, and GIS/remote
sensing technology, we now have the tools and data set to assess the forests and fields of
the past and create a model of land use for the ancient Maya of Tikal. The landscape
defined by the Voronoi calculations offers a starting point from which to evaluate resource
exploitation during two critical periods of Tikals prehistory: the Late Preclassic and the
Late Classic. Tikal was a much smaller polity during the Late Preclassic than during the
Late Classic period when its population grew exponentially along with its influence in
terms of economic and political power. Because much more is known about the Late
Classic period at Tikal, particularly the extent of its built environment and iconographic
representations, and the events that took place during this period ultimately led to the
abandonment of the center, the remainder of this paper will focus on the Late Classic
period.
Our results described in the previous section estimate that the amount of land in the
upland areas available to the Late Classic Maya at Tikal was approximately 855 km. The
bajo area controlled by Tikal during the Late Classic period was about 256 km. In
general, it is thought that because of the energetic constraints faced by the Maya in terms
of transporting materials in bulk, particularly food, it was largely incumbent upon each
polity to produce its own food and most other essential supplies. Hence the question
before us is how this land was managed in a way that could have provided all the needed
fuel, raw materials, and food to sustain the polity.
First, we consider the extent of the Late Classic forests, both the upland forest and the
wetland forest (bajos) within the defined boundary. To provide an estimate of the forest
cover in Late Classic Tikal, we draw upon the pollen evidence derived from cores taken
from Aguada Vaca de Monte to the east of Tikal coupled with pollen cores taken from
Lake Petn Itza just to the south of the Tikal extraction zone. Using pollen counts of forest
species from these two cores, it seems the forest cover during the Late Classic period was
somewhere between 40 percent and 68 percent of the modern arboreal vegetation if we
use the most conservative numbers from each core. Following are three possible scenarios
that reflect different interpretations of the pollen cores from Vaca de Monte and Lake
Petn Itza.
In Scenario 1 (Table 8.1) using only the figures from Vaca de Monte, which suggest
that there would have been 68 percent of the forest remaining, about 581 km of land
vegetated with upland forest would have been present during the Late Classic period.
Using the biomass figures from the current forest at Tikal, we can calculate the standing
biomass of the Tikal upland forest during the Late Classic period. In Scenario 1, the
upland forest biomass was approximately 16.8 billion kg while the bajo forest was 3.17
billion kg. This is a substantial quantity of wood, but to use the forest sustainably, the
Tikal inhabitants could only extract the amount of growth of the forest each year. From
long-term studies of similar Central American seasonal forests we can calculate that the
Tikal forests have a biomass change rate of 0.76 Mgha1yr1 or 76,000 kgkm2yr1
(Phillips et al. 1998).3 Accordingly, the amount of wood that could be used sustainably in
a year would be around 44.2 million kg of wood in the uplands and 13.2 million kgyr1
in the bajos according to the Scenario 1 data set.
Table 8.1. Aboveground biomass (AGB) of modern Tikal forests and estimates for
forests during the Late Classic period (LCP) based on pollen records and modern AGB
calculations

Uplands Bajos

Modern AGB 28.9 million 18.2 million

kgkm-2 kgkm-2

Tikal area LCP (Voronoi calc.) 855 km 256 km

Scenario #1 LCP percentage forest 68% 68%

cover

Scenario #1 LCP forest area 581 km 174 km

Scenario #1 LCP forest AGB 16.8 billion kg 3.17 billion kg

Scenario #1 LCP usable AGB 84.1 million 15.8 million

kgyr-1 kgyr-1

Scenario #2 LCP percentage forest 40% 40%

cover

Scenario #2 LCP forest area 342 km 102 km

Scenario #2 LCP forest AGB 9.9 billion kg 1.9 billion kg

Scenario #2 LCP usable AGB 22.0 million 9.3 million kgyr-1

kgyr-1

Scenario #3 LCP percentage forest 40% 68%

cover

Scenario #3 LCP forest area 342 km 174 km

Scenario #3 LCP forest AGB 9.90 billion kg 3.17 billion kg

Scenario #3 LCP usable AGB 26.0 million 13.2 million

kgyr-1 kgyr-1

Scenario 2 draws on the pollen data from Lake Petn Itza. Using the conservative figure
of 60 percent reduction in arboreal pollen, these data suggest that 40 percent of the forest
remained during the Late Classic period. Thus, the extent of the upland forest would have
been 342 km with 26 million kgyr1 of wood available for sustainable use along with
102 km of bajo forest bearing 7.78 million kgyr1.
Scenario 3 recognizes that the two pollen cores referred to in this paper probably reflect
different aspects of the habitat surrounding Tikal. The Lake Petn Itza core reflects the
overall watershed surrounding it, which is predominantly upland forest, and includes a
sizable portion of the Tikal extractive zone. The Vaca de Monte core, on the other hand, is
from a small aguada deep in the Bajo de Santa Fe and seems to be more reflective of the
wetland forest as the predominance of wetland species pollen from that core attests. With
these concepts in mind, the third scenario bases the Late Classic upland forest evaluation
on the Lake Petn Itza pollen data and the extent of the bajo forest on the Vaca de Monte
data. Results from this scenario indicate that approximately 26 million kg of wood in the
upland forest and 13 million kg in the bajos were available for use on a sustainable basis
each year.

Wood Requirements
Up to this point we have focused on the amount of wood that the forest surrounding Tikal
might have produced. Now we consider how much wood would have been required to
meet the needs of a functioning Late Classic polity. Recent population estimates of eighth
century Tikal have ranged from a minimum of 45,000 (Haviland 1970) to 62,000 (Culbert
et al. 1990)4 and as high as 80,000 (Dickson 1980). Webster and Murtha (this volume)
estimate the Late Classic population as somewhere between 57,000 and 67,000.
Notwithstanding this wide range of population estimates, we will adhere to our
conservative approach and use the smallest estimation of 45,000 suggested by Haviland.
Firewood for cooking created a major demand on the forest of the ancient Maya
(Abrams and Rue 1988). Most of the principal foods of the Maya had to be cooked daily.
Beans, at least, are inedible if they are not cooked and maize is traditionally nixtamalized
and boiled (S. Coe 1994), as well, so the fundamental need for firewood for food
preparation was inexorable. Breedlove and Laughlin (2000) reported that a family of five
Tzotzil Maya used 16 kg of firewood per day in their hearths (or 3.2 kg per person).
Romero (2003) observed that traditional Petn farming families required from 2.3 to 2.8
kg/person/day of cooking fuel. If we take the most modest population estimates (45,000)
and multiply this figure by the minimum report of firewood consumed each year (2.3
kg/day x 365 = 839 kg/yr), the result is an estimate of 37.8 million kg/yr of wood needed
annually for cooking fuel at Late Classic Tikal (Table 8.2).
Table 8.2. Estimated fuel and timber needs of Tikal during the Late Classic period
(LCP)

Wood Requirement Quantity Needed (in millions of kgyr1)

LCP pop. 45,000 LCP pop. 62,000

 Firewood 37.8 52.1

Ceramics 3.31 4.55

Plaster production 1.59 1.59

Construction 0.12 0.12

TOTAL 42.8 58.4

Note: Total amount of wood available per year from upland and wetland forest = 39.2
million kgyr1 (as predicted in Scenario 3).
Other major needs included wood for the firing of ceramics; the construction of houses,
temples, palaces, and other structures; and the manufacture of plaster. Ceramics were a
huge part of the domestic, and probably the political, economy at Tikal. The polity was
famous throughout the Maya world for its polychomes. Leaving aside for the moment the
possibility of an export market, internally the need for ceramic wares for domestic and
ceremonial consumption was huge. Household 1 at the Late Classic Cern site, a small
Maya farming community in what is now El Salvador, had more than 70 ceramic vessels
within its compound (Sheets 2002). The House of Mirrors and the House of Axes, both
rapidly abandoned elite households at the Late Classic Maya center of Aguateca, showed
evidence of having 85 and 77 vessels, respectively, within each compound (Inomata et al.
2010: 182). If we apply the number of 70 vessels per household to Tikal (again using the
45,000 population figure) then approximately 630,000 ceramic vessels were in use at Tikal
at any given time during the Late Classic period. Studies have shown that the life of a
ceramic vessel is ephemeral (Arnold 1991; Longacre 1991); they last only about 12
months on average, so the Tikal occupants would have had to replace more than a half-
million ceramic vessels every year. This is undoubtedly a modest estimate, because the
occupants of all the palaces and elite households at Tikal likely would have required more
ceramic vessels than the humble occupants at Cern.
Nevertheless, if 3.75 kg of wood (Arnold 2008) were required to fire each vessel, then
approximately 2.36 million kg of firewood would have been needed to fire the pottery of
the Tikal inhabitants each year. Note that the data for the amount of fuel are taken from a
Yukatek Maya community that used beehive kilns. Although it has been hypothesized that
the Tikal Maya used trench kilns (Becker 2003) to make ceramics, the evidence for this
is inconclusive. Other scholars (e.g., Culbert 2003) believe they used open or heap kilns
that would have consumed 40 percent more fuel per vessel (Wernecke 2008). If the
ancient Maya were using heap kilns, and most likely they were, then the amount of fuel
needed to fire their ceramics should be revised upward to 3.31 million kg per year.
Construction materials also were derived from the forests of Tikal. The more common
dwellings of the polity were largely constructed of wood and thatch, but even the major
temples and palaces, composed mostly of cut stone, required large wooden beams for
windows, hallways, and doorways. A previous study (Abrams and Rue 1988) calculated
that it took approximately five moderate-sized trees to build a typical Maya house. The
average modern tree at Tikal has a diameter of 16.5 cm and a height of 9.2 m, including
both upland areas and bajos. This is about the size of tree a Maya commoner would want
to build a house; trees of greater girth might be too large and hard to manage and smaller
trees might not be strong enough to serve as uprights. To calculate the biomass of a tree,
we can use an equation published by Brown et al. (1989) for moist tropical forests:5
= 38.4908 11.7883 (D) + 1.1926 D, where = biomass and D = diameter
Using the mean DBH from trees in the Tikal forest, the average tree at Tikal has a biomass
of 169 kg. There are 1,778 residential structures in the central 9 km at Tikal (Arnold and
Ford 1980). If we put these numbers together (5 trees/house x 169 kg/tree x 1,778 houses),
then 1,502,410 kg of timber were required to build the residential portion of the Late
Classic polity. Among the modern Maya of Yucatan, Redfield and Villa Rojas (1934: 35)
recorded that traditional houses of this sort were built to last for your life, but more
realistically they had a life span of about 25 years (Vogt 1969: 90). Using the 25 year life
span figure, then the yearly construction need for timber would have been around 60,096
kg. Even if we doubled this number to include the temples (built mostly of stone),
scaffolding, and outlying residences, the amount of wood needed for construction would
have been a minor component of the overall annual wood budget.
The amount of firewood needed to make lime (a principal component in plaster),
however, was more substantial. Lime production, an essential part of the political
economy, was an element of elite prestige requiring that temples and palaces be
maintained to retain the status that their political and social positions required. Much of
the site core was covered with plaster, including all of the temples, the major plazas, the
palaces, and many of the reservoirs. The Great Plaza alone covered about 1.1 ha and was
covered with a 38 cm layer of plaster in four layers that averaged 9.5 cm per layer (Coe
1967). These four plaster layers were applied between 150 BC and AD 700. In our studies,
we observed plaster layers that ranged between 5 and 10 cm, so we will use an 8 cm layer
of plaster for our calculations. Using the digitized Carr and Hazard map (1961), we were
able to calculate the surface areas of the causeways (180,623 m), the major reservoirs
(172,347 m), the plazas (173,250 m), the temples (76,192 m), and the palaces (18,754
m). This total area (621,166 m) required 49,693 m of plaster. Typical Maya plaster
weighs 2,000 kg/m but only 16 percent is from lime, also known as quicklime or
calcium oxide, with the rest derived from sascab (decomposed limestone), crushed
limestone, and other fillers (Schreiner 2002). Lime is the binder that holds the other
aggregates together in traditional plaster preparation. Prepared by heating limestone, lime
is the only component of plaster that needs to be heated prior to mixing.
Around 15,901,849 kg of lime would have been required to plaster the exterior surfaces
of Late Classic Tikal. About 5 kg of wood was required to make 1 kg of lime using
traditional open kiln technology (Schreiner 2002). Therefore, it would have taken around
79.5 million kg of wood to fire enough lime for plaster to cover all of the surfaces at Tikal.
This is a huge amount of wood and would have consumed the entire annual sustainable
wood budget from the upland forest for Late Classic Tikal, but as with house construction,
the plastering for the entire polity would not have been completed in one year; rather it
would have been divided up over many years, even centuries (Schreiner 2002). Tozzer
(1941: 151) notes that the ancient Maya likely would have added new layers of lime wash
to their buildings to mark the annual renewal rites, but major replastering occurred only at
the end of a 52-year cycle. Abrams and Rue (1988) estimated that surfaces at Copn were
replastered every 50 years. If this were true at Tikal, then approximately 1.59 million kg of
wood would have been required every year to maintain the plaster surfaces. In the long
run these plaster requirements undoubtedly put heavy demands on the forests, the only
major source of fuel for the Maya. Other activities that involved the use of lime would
have been nixtamalization (a means of maize preparation) and plastering the inside of
some elite structures such as palaces and temples, but these would have been a relatively
small portion of the yearly lime budget compared to the needs of exterior surfaces.
According to these calculations, a prominent polity like Tikal had tremendous needs for
fuel and timber. In an average year, though, it seems likely that the available forested
lands, at least in Scenario 1, could have provided the necessary supplies of wood to keep
the polity functioning. Scenario 2 is a different story; it is highly unlikely that the forested
lands described in this case were adequate to provide the necessary forest products during
the Late Classic period. Scenario 3 represents a situation that would have been close to
sustainably supplying the amount of wood needed as presented in our analysis. To
enhance their timber productivity, there are indications that the Tikal occupants had
devised special agroforestry techniques in response to the pressing demand for wood and
could have functioned under Scenario 3 for a long time.
As discussed in the previous chapter (Thompson et al., this volume) there is substantial
paleoethnobotanical evidence that the Late Classic Maya obtained pine for fuel use,
probably from the pine/palmetto savanna in the Bajo de Santa Fe 18 km to the northeast of
Tikal (Lundell 1961; Fialko 2000; Dunning, et al. 2002). Some archaeologists have
suggested that this 180 ha stand of pine was planted by the ancient Maya (Fialko 2001),
but more recent molecular studies have pointed to the apparent genetic variablity of this
pine population as reflective of a natural stand, perhaps as a relict of a larger, more
extensive pine savanna from the pre-Maya era (Dvorak et al. 2005). In either case, it
seems likely that the ancient Maya were at least managing this stand of pine for timber
production if not actively cultivating it. If this were not so, they would have easily wiped
out such a small population from what was apparently heavy use. The active management
of this pine savanna may well have served as an important augmentation to the wood
productivity of the upland and bajo forests. Another, although probably less likely
possibility was that the Tikal Maya may have been carrying in pine charcoal, used for both
fuel and ceremonial purposes, from outside their extraction zone, as has been observed at
other Maya sites (Lentz et al. 2005). No doubt the needs of forest products were great, but
considering all of these possibilities together it seems reasonable to assert that the means
of meeting those needs were within the realm of capability of the defined extraction zone
of Tikal.

Agricultural Practices
Having discussed the forested upland and bajo areas in the Voronoi boundary under the
various scenarios, we now turn to the remaining lands available to the Late Classic Tikal
Maya where the native forest had been cleared, largely for agricultural purposes. A portion
of this land was occupied by the built environment in the center of the polity where the
temples, plazas, ball courts, palaces, and more humble dwellings were located. This is
similar in concept to what Webster and Murtha (this volume) refer to as the demographic
core of Tikal. According to Frys description (2003) of Late Classic Tikal, this area
occupied 165 km of what was once upland tropical forest. This would leave 143 km of
cleared upland area for agricultural fields under Scenario 1.
Various authors have calculated how much land is required for a nonindustrial farmer to
subsist. Harrison (1992) suggests that the minimum amount of land required is 0.2 ha per
person. In more regionally relevant studies, Palerm (1955) calculates that Totonac farmers
of southern Mexico in lowland forested environments required 0.24 ha/person. Maya
living in the Puuc area, a drier region north of the Petn, required about 0.45 ha of land per
person (Sanders 1957). In a data set collected by Cheetham (2010), 0.18 ha of reasonably
fertile land was needed to provide enough maize to feed the average modern Lowland
Maya person using traditional farming techniques. If we use Cheethams figure and extend
it to ancient Tikal, then the polity needed approximately 81 km (0.0018 km x 45,000
people) of fertile land, mostly upland and some colluvial bajo-margin soils, to feed the
populace. In any given year, only 28 km of land would have remained unaccounted for in
Scenario 1. If the Tikal Maya were reliant on a swidden or milpa system of agriculture that
required the renewing nutrients of ash from the plant cover, there would not have been
enough land for even one year of fallow cycle. This situation would not have been
sustainable given what we know about edaphic conditions and traditional Mesoamerican
farming techniques (e.g., Sanders 1967; Hammond 2000).
If we look at Scenario 3, however, with 342 km of forested land, it would have allowed
348 km for agriculture in the relatively fertile upland areas. This could provide three
years of fallow between each year of active cultivation, representing a more feasible
situation than Scenario 1.
While some scholars have stated that short fallow systems are not sustainable (e.g.,
Boserup 1965; Cowgill 1962), others have pointed to short fallow yield declines that occur
slowly and can be drawn out over long periods (Sanders 1979; Mertz 2002; Mertz et al.
2008). Steggerda (1937) observed in his maize experiments in the Maya Lowlands that if
the weeds are carefully removed before and after planting, then yields, at least for short
periods, can actually be increased. As McAnany (1995: 78) hypothesized, the settlement
density of the Late Classic Maya precluded the possibility of a long fallow type of
swidden agriculture; there simply was not enough land. This is exactly what we observe at
Tikal. Killion (1987) offers details of a fixed plot system of agriculture for farmers in
southern Vera Cruz with variable and often brief periods of fallow, a system that may have
been applicable to the Classic period Maya. Sanders (1973) describes what he calls a
bush fallow type of cultivation with one year of cultivation followed by three years of
fallow that produced modest yields. Similarly, a short fallow cycle with three years of
successive planting followed by five years of fallow as described by Griffin (2012) could
have been implemented in the high-quality agricultural land available to the Late Classic
Tikal farmers with results that could have minimized yield declines.
To render these short fallow techniques more effective, at least three species of
leguminous crops, namely, scarlet runner beans (Phaseolus coccineus L.), lima beans (P.
lunatus L.), and common beans (P. vulgaris L.), were cultivated by Tikal farmers (Table
8.3). As has been observed widely among modern Maya farmers, if these plants are
multicropped with maize, they can replenish accessible nitrogen in soil to help ameliorate
yield declines common in short fallow agricultural regimes.
Table 8.3. Comparison of climatic conditions of Tikal and Cern

Elevation Average Rainfall Average Temperature

Tikal 230 masl 1,945 mm 25 3 C

Cern 450 masl 1,700 mm 24 2 C

Phosphorus, however, is the soil nutrient in most critically short supply across the Maya
Lowlands (Dunning et al. 1998). With deforestation, the soil is progressively starved of
this critical element (Das et al. 2011; Lawrence et al. 2007), an eventuality with potentially
catastrophic consequences for the ancient Maya (Dunning et al. 2012; Turner and Sabloff
2012). Fortunately for regional farmers, phosphorus is naturally renewed in soils via the
capture of windblown soot, dust, and volcanic ash. This periodic input of phosphorus may
help to explain how the Maya at Tikal were able to continue intensive agriculture over
several centuries.
To gain a greater understanding of Late Classic Maya subsistence and create a model of
sustainable agriculture for that period, we turn to another contemporaneous Maya site
where the land use activities are reasonably well known: the Cern site in central El
Salvador. The material culture and plants used by the ancient Maya at Cern were
extraordinarily well preserved because the village was inundated by volcanic ash in AD
600, just at the beginning of the Late Classic period. Not only were seeds, fruit, cloth,
charcoal, and other typical kinds of plant remains found at Cern, but because of the
steamy ash that enveloped the village, standing plants were preserved as impressions and
rendered visible by means of a plaster casting process (Sheets 2006). As a result, at Cern
we not only find in situ seeds and fruits on the ground, but also the stems, twigs, and other
plant parts clearly outlined and located where they were at the time the village was last
occupied. Because of the plant casts, we can determine not only what plants the Cern
villagers were growing, but also how and where they were growing them.
In Household #1 at Cern, for example, the domiciliary structures (Figure 8.7) were
surrounded by orchards interspersed with small plots where annual crops, such as maize
(Zea mays L.), beans (Phaseolus spp.), squashes (Cucurbita spp.), peppers (Capsicum
annuum L.), and root crops, such as malanga (Xanthosoma sagittifolium L. Schott) and
manioc (Manihot esculenta Crantz), were planted adjacent to the structures (Lentz and
Ramrez-Sosa 2002). Extensive fields with furrowed rows of maize and manioc were
planted to the south of the village compound (Sheets et al. 2012). These extensive fields
probably provided most of the calories for the Cern villagers, whereas household gardens
provided nutritional diversity. Although there are slight differences in climate between the
two sites (Table 8.4) and the emphasis on the growth of various crops between Cern and
Tikal may have varied, the overall comparison of the two contemporaneous settlements is
informative.
Figure 8.7. Household 1 at Cern. This drawing shows the approximate layout of plants
adjacent to the structures in the house compound. The illustration is based on plant
remains, (e.g., leaves, fruits, twigs, and cast impressions) found on the exterior activity
surfaces of the compound. The plantings surrounding this Late Classic household were
probably similar to those of the Late Classic households of Tikal (Base map redrawn from
figure 4-1, Sheets 2006).
Table 8.4. Ancient plant remains identified at Tikal

Root Crops Trees (continued)

Canna cf. indica (LP)
Ipomoea batatas (MP)
Xanthosoma sagittifolium*
(LC)
Other Field Crops
Cucurbita moschata* (LC)
Cucurbita pepo* (EC)
Gossypium hirsutum* (EC)
Phaseolus coccineus (LP-EC)
Phaseolus lunatus* (EC)
Phaseolus vulgaris* (EC)
Zea mays* (EC-TC)
Tree Crops
Chrysophyllum sp. (LP-LC)
Clusia sp. (LP)
Croton sp. (LP-EC)
Cupania sp. (EC-TC)
Enterolobium cyclocarpum (TC)
Erythrina spp. (EC-LC)
Eugenia spp. (LP-LC)
Ficus sp.* (EC-LC)
Garcinia sp. (TC)
Gliricidia sepium (LC)
Guarea glabra (LP-LC)
cf. Guettarda combsii (?)
 Acrocomia aculeata* (LP-LC) Haematoxylum campechianum* (EC-
PoC)
Bactris major (EC)
Byrsonima crassifolia* (LP- Heliocarpus sp. (LP-EC)
PoC) Hirtella sp. (LP-LC)
Persea americana* (TC) cf. Lacmellea sp. (?)
Pouteria sapota (C) Licaria campechiana (LP-LC)
Spondias cf. purpurea* (LP- Lonchocarpus spp.(LP-PoC)
TC) Manilkara zapota* (LP-TC)
Theobroma cacao* (EC-TC) Metopium brownei (LC)
Other Useful Plants
Nectandra spp.* (LP-LC)
Cyperus canus* (LC)
Ocotea puberula (LP-LC)
cf. Morinda sp. (EC)
Pimenta dioica (TC)
Piper sp. (EC)
Pinus spp.* (LP-TC)
Tecoma stans (EC-LC)
Piscidia piscipula (LC)
Thevetia ahouai (LP) Pouteria spp. (LP-TC)
Trees
Protium copal (LP-LC)
Acacia sp. (?)
Pseudolmedia glabrata (LP-LC)
Acosmium panamense (EC)
Psychotria sp. (C)
Alvaradoa subovata (LC-TC)
Salix sp. (TC)
Ampelocera hottlei (LP-LC) Sebastiana sp. (LP-LC)
Aspidosperma spp.* (LP-LC)
Sideroxylon sp. (LP-LC)
Astronium graveolens (LP-LC) Stemmadenia sp. (LP-LC)
Brosimum alicastrum (LP-LC)
cf. Tabebuia sp. (LC)
Caesalpinia sp. (LC)
Tapirira sp. (LP-LC)
Cameraria latifolia (LP) Terminalia buceras (EC)
Casearia sp.* (LP) Trichilia hirta (EC)
cf. Carapa guianensis (LP-EC)
Zanthoxylum caribaeum (LP-LC)
Ceiba pentandra (TC) Zuelania guidonia (LP-LC)
Celtis iguanaea (LP-TC)

MP = Middle Preclassic, LP = Late Preclassic, EC = Early Classic, LC = Late Classic,

TC = Terminal Classic, PoC = Post Classic, ? indicates no known date. * indicates
plants also found at Late Classic Cern. seec Moholy-Nagy (2003). see Pohl et al.
(2000).
 When we examine the array of plants used by the Cern occupants and compare it to
the plants used by the people of Tikal (Table 8.3), there is a substantial overlap. Maize
clearly predominates at both sites in terms of the number of samples recovered as well the
ubiquity of contexts bearing maize remains. Two species of beans, two species of squash,
and numerous species of fruit trees (Figure 8.8) have been identified at both places. Root
crops are well represented, too. Although manioc remains have not been found at Tikal, it
has been identified by pollen in the Petn at the nearby San Bartolo site (Dunning et al.
2006) so in all likelihood the Tikal farmers had access to the root crop, as well. More
forest species were identified among the plant remains at Tikal than Cern, as would be
expected with the stronger emphasis on midden excavation at the former site.
Nevertheless, because the occupants of the two sites were culturally connected, had
somewhat similar environments, were synchronous in the time of occupation, and
exploited many of the same plants, both domesticated and wild, Cern serves as a
reasonable and useful model to help explain land use at Tikal.

Figure 8.8. Plant macroremains from Tikal (a) Cucurbita pepo seed (Reference no.
10030-001), (b) Spondias sp. pit (Reference no. 10094-001), (c) Acrocomia aculeata
endocarps (Reference no. 10152-001), (d) Zea mays cob fragment (Reference no. 10192-
001).
Assuming Cern serves as a valid comparison for Tikal land use interpretation, we
propose that the area defined by Fry, which accounted for 165 km2 of the total Tikal
extractive area, including temples, palaces, and other dwellings, would have been
analogous to the central village at Cern. Essentially all of the Tikal area within the Fry
boundary, with the exception of the main plazas, would have been planted much the same
as the Cern village with abundant orchards of broadleaf fruit trees and palms interspersed
with small plots of maize and other domesticated plants, notably root crops.
In addition to the extensive short fallow system that probably produced a major portion
of the food supply at Tikal and the system of dooryard gardens and orchards, there were
other, more intensive forms of primary production being utilized. Significantly, in addition
to the other approaches described, there were areas at Tikal where the Maya seem to have
developed intensive hydraulic agriculture, such as in the lands south of the Perdido
Reservoir. As evidence for this, a test pit excavated in the Pocket Bajo just below the
egress of Perdido Reservoir revealed that the soil formation abruptly shifts from typical
bajo soil to alluvium at the beginning of the Late Classic Period with subsequent alluvial
laminations (Figure 6.10), indicating that this area was repeatedly flooded, presumably for
agricultural purposes. The Perdido tank itself contained unusual amounts of artifactual
trash (see Scarborough and Grazioso Sierra chapter, this volume), indicating that the
reservoir had a purpose other than the storage of drinking water. Finally, the 13C
enrichment as seen in Figure 6.9 indicated the presence of C4 plants in the pocket bajo in
the past even though there were no C4 plants observed there in our modern plant surveys.
Also, maize was the only C4 plant found among the plant cultigens in the Tikal
paleoethnobotanical record. Taken together these reinforcing lines of evidence indicate
that the Late Classic Maya were using the Perdido Reservoir for irrigation purposes and
maize was likely one of the crops cultivated directly below the reservoir.
Furthermore, it seems plausible that the areas to the southeast of the Corriental
Reservoir could have been managed in the same way. Corriental Reservoir has a switching
station at the southeast end that could easily have facilitated irrigation of the fields below.
Areas to the east of the Tikal Reservoir and north of the Bejucal Reservoir were other
likely locations where irrigated agriculture might have taken place. With the input of water
from the reservoirs, these areas potentially could have been double-cropped, allowing
more than one harvest per year.
The Inscriptions Reservoir, because of its lack of ingress and egress features, is unusual
and may well have been used as a water impoundment to grow specialty crops, such as
cacao or root crops, in a unique microhabitat (Dunning, personal communication, 2013).
Finally, Tikal farmers also exploited the bajo margins for agriculture (see Dunning et al.;
Thompson et al., this volume) and timber as well (Lentz and Hockaday 2009). Scenario 3,
which seems to be the most appropriate interpretation of available pollen data, indicates
that about 32 percent of the bajo forest was removed and the Maya were practicing maize
and root crop agriculture in the Bajo de Santa Fe. In fact, the deep, colluvial soils along
the bajo margins probably were among the most agriculturally productive in the Petn.

MultiComponent Land Use Model for Tikal

Combining information derived from paleoethnobotanical studies, modern forest surveys,
satellite images, soil analyses, settlement patterns, isotopic profiles, hydrological studies,
and information from other contemporaneous sites, we are now in a position to piece
together a comprehensive multicomponent land use model for the Late Classic Maya at
Tikal. The components of this model include managed forests, dooryard gardens and
orchards, extensive agricultural plots, and zones of intensive agriculture irrigated from a
series of reservoirs. Overall, the resource extraction zone of Late Classic Tikal consisted
of approximately 1,111 km. Within this extractive zone, around 855 km was located in
upland areas and the other 256 km was seasonal swampland or bajo (as seen in Scenario
3). According to the pollen evidence discussed previously, somewhere around 342 km of
upland forest remained intact (about 40 percent of full forest cover) even into the height of
the population maximum of the Late Classic period. This upland forest was managed in
fixed plots in a way that would maintain diversity and provide a constant supply of
essential forest products, especially, but not only, firewood and timber.
Another 165 km of the upland area was occupied by the domiciles, palaces, ball courts,
temples, and sacbeob of the city of Tikal. The houses within this structural component of
the Tikal territory were widely enough spaced to allow for orchards and small gardens
with annual crops and medicinal plants in the areas between households. Also within this
built environment were agricultural plots with plantings of maize and other crops located
below some of the reservoirs that could be irrigated with stored water and double-cropped
each year. The remaining 349 km of the upland area in the Tikal extractive zone consisted
of land cleared for extensive agricultural fields, probably multicropped in some sort of
short fallow system of crop rotation.
The bajos also represented a vital component of the land use system at Tikal. During
the Late Classic period approximately 82 km of bajo margin was used for agriculture. The
soils of the bajo margins consisted, at least in part, of colluvium washed down from the
upland. These soils, evidently, were exploited for maize and root crop agriculture. The
remaining 174 km of bajo land in the Tikal extraction zone was used as a source of timber
and other wetland forest products. Included within this bajo area was a small (1.8 km),
but important, pine/palmetto savanna whose pine wood was intensively managed as a fuel
for ceremonial use and other purposes.
Several independent lines of evidence support the multicomponent model of land use at
Tikal. The first line of evidence is the pollen data from the Aguada Vaca de Monte and
Lago Pten Itza cores cited previously. These valuable data sources indicate reduced levels
of pollen production from arboreal species during the Late Classsic period. These pollen
data inform us that substantial portions of the forests had been removed by that time and
give us a framework to estimate the amount of land devoted to forest product production
and agricultural activities. In addition, the pollen information from the Aguada de
Terminos core (see Dunning et al., this volume) reveals that the bajo areas adjacent to the
built environment of Tikal were being used for maize and root crop agriculture.
Macroremains of maize and root crops found at Tikal reinforce this assertion (see Table
8.3; Pohl, Pope, and Jones 2000; Moholy-Nagy 2003).
A second line of support for the multicomponent model is from the temple lintels at
Tikal. Up until and including the construction of Temple 4, enormous beams that could
only have originated in old growth forest were used in construction, indicating that the
Maya had carefully managed their forests at least until the mid-eighth century, several
centuries after the onset of the dramatic Tikal population boom at the beginning of the
Late Classic period. Prior to the large temple construction boom of the eighth century, the
Tikal occupants appear to have been conserving large trees and protecting at least some
stands of old-growth upland forest (Lentz and Hockaday 2009). Quite possibly these
managed forests and fixed agricultural plots were controlled by royal functionaries or
lineage leaders (McAnany 1995: 645).
A third source of evidence that supports the concept of fixed plot, bush fallow
agriculture and fixed woodlots are the paleoethnobotanical remains themselves. We
examined thousands of wood charcoal fragments in our analysis and of the ones that were
large enough (421) to be subjected to SEM analysis, 377 (90 percent) were from large
trees. We know this because the rays of these samples appear to be parallel in transverse
(or cross) section. Ray cells transport water laterally from the center of a tree trunk to the
outer tissues. When they appear wedge shaped or convergent, it indicates that the trunk or
branch is of small diameter. When the rays appear to be parallel, however, then it is likely
that the wood fragment was from a large branch or tree trunk. Only 3 percent of the
samples (14) had rays that were clearly convergent and the other 7 percent were slightly
convergent (30). Our conclusion is that the vast majority of wood used at Tikal was from
large trees. This supports the idea that they had managed forests in more or less fixed plots
and established woodlots of relatively mature trees. Also, these data suggest that the Tikal
Maya were not getting their wood from long fallow agricultural plots. If they were, we
would expect to see more wood with convergent rays. In a bush fallow or grass fallow
system, trees would not have enough time to grow to a stage where they had accumulated
much biomass. A final point that supports the bush fallow, fixed plot agricultural system is
that pioneer trees, such as Cecropia spp., Trema micrantha (L.) Blume, and Miconia
argentea (Sw.) DC. (Brokaw 1987), which typically take hold after an agricultural plot is
allowed to fallow, were absent from the paleoethnobotanical record. Although this is
negative evidence, it suggests that these trees, characteristic of long to intermediate fallow
situations, were not common in the fallow systems of Tikal.
A fourth source of support for the model presented is the array of plants used by the
Cern and Tikal occupants; there is a remarkable and unmistakable overlap. The
agricultural and agroforestry systems are similar, though with some variation, probably
due to different edaphic and climate conditions and the great disparity in the overall
magnitude of need in the two communities, as Cern was a small farming village and
Tikal was one of the major centers of the Maya Lowlands. Otherwise, what we have
learned from Cern seems directly applicable to the interpretation of Late Classic
agricultural practices at Tikal
A fifth line of evidence that supports the multicomponent model emanates from the soil
profiles from Corriental and other Tikal reservoirs. These demonstrate that there was
relatively little erosion from the land surrounding the reservoirs. The soil profile from
Corriental illustrates this point (Figure 8.9). There were a number of catastrophic events
that preceded the Late Classic period, possibly hurricanes, but during the Late Classic we
see a distinct series of fine depositional sequences, probably amounting to only 12 mm of
sediment accumulating per year, and most of the sediment appears to be eolian. In central
Tikal, there were no terraces, check dams, or other physical features to contain lateral
water flow from the catchment areas even though the surrounding terrain was steep in
some cases (up to 10 percent slope), so the Late Classic Maya must have achieved erosion
control through vegetation cover. Plantings of fruit trees and useful understory shrubs in
the catchment basins, as proposed herein, would have protected the reservoirs and
provided useful products at the same time.
Figure 8.9. Excavation profile (north wall) from Corriental Reservoir. Note that the
depth of the Late Classic deposition was laid down in fine layers, mostly loess, and except
for the sandy layers, which reflect storm episodes, the indication is that there was very
little colluvium moving into the reservoir during the entire 250 year period. This provides
evidence that the ancient Maya were protecting the steep slopes of the watershed to
prevent erosion, probably with plantings of useful trees and perennial shrubss (Drawing by
N. Dunning).
Finally, the 13C isotopic data recorded from the site core reveal that there was little
isotopic enrichment from C4 plants in the reservoir catchment basins (see Chapter 6,
Dunning et al., this volume). One exception was the pocket bajo south of the Perdido
Reservoir, where irrigated maize agriculture likely was practiced. We do find significant
C4 enrichment in the Bajo Santa Fe, however. At Aguada de Terminos, where we found
maize and root crop pollen, we also observed elevated 13C levels. As at Cern (Sheets et
al. 2012), with evidence of maize and root crops, the extensive agricultural fields at Tikal
were relegated to the outskirts of the community and provided most of the calories.
This multicomponent land use model has, in many respects, drawn on the ideas of a
host of scholars who have preceded us (e.g., Bronson 1966; Wiseman 1978; Wilken 1971;
Tourtellot 1993; Fedick 1996), and these works have helped give direction to our
investigations. Using multiple channels of data input that inform our model building, we
attempted to test many of these ideas and present a revised and more nuanced view of
ancient Maya land use practices.
Climate Change
In an average year during the Late Classic period the rains would have arrived in sufficient
quantity and at the appropriate time, so the forests would have grown normally and
increased their net primary productivity. Well-tended fields provided adequate sustenance
to meet the needs of the Tikal occupants without rapid depletion of their basic resources.
What is not included in the calculations outlined previously is what might have happened
in the wake of a prolonged drought or a series of droughts, which has been hypothesized
for the Lowland Maya area at the end of the Classic period (e.g., Gill 2000; Hodell et al.
1995; Kennett et al. 2012). Interestingly, there is evidence that suggests that extensive
forest clearance may negatively impact rainfall patterns (Georgescu et al. 2013; Cook et
al. 2012), so it is conceivable that the ancient Maya, who cleared more than half of the
upland forest in the Tikal extractive zone, may have unwittingly contributed to the ruinous
droughts of the ninth century.
In the face of extended drought, forests would cease to grow or even shrink in terms of
their aboveground biomass. Tree mortality in a tropical forest rises dramatically during an
extended drought (Laurance et al. 2001). With the Late Classic Maya putting such heavy
demands on their forests, it is easy to imagine what might have happened if rainfall had
declined for a prolonged period: Trees added no extra mass to their trunks and limbs,
weaker trees may have died off, yet the need for fuel to run the polity continued. When a
drought of the projected magnitude occurred, however, it was not the lack of firewood that
created the social upheaval of the midninth century; shortages of food and water were by
far of more devastating impact.

Conclusions
The multicomponent land use model described for Late Classic Tikal is supported by
many lines of evidence that converge on the same reality; the Maya managed the land
intensively in a variety of ways to grow and extract the myriad products needed by an
urban center. They did clear a significant portion of their upland forest, but maintained
what was left (40 percent) in fixed plot woodlots. We fully agree with other scholars (e.g.,
Fedick 2010; McNeil 2010) who have asserted that the ancient Maya did not foolishly
clear away their forests; indeed they could not have done so because it would have
eliminated their major source of fuel and other necessities. The ancient Tikal Maya,
however, because their principal food crops, especially maize, needed full sunlight for
optimal growth, were forced to clear a significant portion of their upland forests. It
appears that they did carefully manage what forest remained, at least until the eighth
century, when they began to cut down their last surviving old-growth forests to build the
major temples that have become the hallmark of Late Classic Maya society (Lentz and
Hockaday 2009).
A major portion of the upland area (40 percent) was used in the Late Classic period for
extensive agriculture, probably under some sort of short-fallow cultivation of annual
crops. The remaining 20 percent of the cleared upland area was occupied by the built
environment with land between houses planted with orchards and door-yard gardens as at
Cern. Within the built environment there were intensive agricultural plots under
hydraulic management with water supplied from reservoirs. The reservoir watersheds
likely were protected by an assortment of perennial plants on the steep banks of the
catchment areas. The bajos, particularly the margins that were enriched with upland
colluvium, also were an important source of agricultural products and forest resources.
According to our calculations, it seems highly unlikely that the resource extraction zone
of Tikal could have supplied the needed food and fuel for populations larger than 45,000
without substantial input from the outside. Some scholars (e.g., Pyburn 1998) have
suggested that the importation of food into the large centers was a possibility, but others
(e.g., Drennan 1984; Webster 2002) have pointed to the prohibitive cost of overland
transport of large amounts of foodstuffs by human porters. In light of this, suggestions that
the extractive zone of Tikal could have supported numbers of people well in excess of
45,000 during the Late Classic period do not seem plausible vis--vis our calculations and
what we know about ancient Maya agricultural capabilities.
Unquestionably, the Maya developed an impressive system of agroforestry and
agriculture at Tikal that effectively provided needed sustenance and materials for a
substantial population for several centuries. When that resource exploitation strategy
combined with a period of low rainfall, however, the Tikal Maya may have reached a
tipping point where a system already operating at its maximum optimal potential relative
to available technology could not be further adapted to meet changing conditions. Overall
the amount of fuel needed was staggering and the heavy human demands on a finite forest
resource may have eventually outstripped the ability of the forest to rejuvenate itself
especially in the face of extended drought conditions. Although a shortage of fuel would
have been problematic, the effects of the ninth century droughts on agriculture and potable
water supplies from reservoirs reliant on rainwater for recharging would have represented
a much more immediate crisis. Considered together, the net effect of essential resource
depletion would have had a dramatic impact on social cohesion within the Tikal
community and would have contributed greatly to social unrest. Thus, environmental
overstretch combined with drought conditions, possibly exacerbated by anthropogenic
forces, may well have contributed to a loosening of the social bonds that held the society
of the Late Classic Tikal Maya together.
Notes
1 For bajo vegetation biomass calculations, all possible combinations of relationships
between bands and variables were explored using multiple linear regression. While initial
exploration indicated a strong linear relationship using both Band 3 (predominantly red
visible light; 0.630.69 m) and SAVI, it was not a statistically significant one. The
second strongest relationship, an inverse one, occurred with Band 3. This negative
relationship can be explained by the action of chlorophyll in absorbing red light for
photosynthesis; lower aggregate red responses correspond to either more concentrated or
more vigorous vegetation. This relationship was found to be strongest using a cubic curve
(r2 = 0.758, p = 0.029, n = 8, standard error = 1.98) with the equation as follows: 937.0139
+ (30.906)*Band3 + 0*Band32 + 0.005*Band33. A cubic relationship was selected over a
linear one because of the latters tendency to create anomalies (i.e., large negative
numbers for AGB) in a manner less predictable than the cubic curve.
For upland biomass vegetation calculations, all possible combinations of relationships
between bands and variables were explored using multiple linear regression. While
previous studies have found utility in the linear relationship between vegetation indices
and forest biomass (Zheng et al. 2004; Sader et al. 1989), this analysis found the strongest
relationship of the biomass of upland tropical forest and forest on the margins of wetlands
to be the atmospherically corrected reflectance of Landsat Band 5 (sensitive to 1.551.75
m wavelengths of the electromagnetic spectrum). This same relationship was also
observed by Steininger (2000) for AGB estimations in Tropical forests in Bolivia and
Brazil. The strongest correlation between the each parcels AGB and Band 5 was found
using a cubic curve (Figure 8.4; r2 = 0.818, p = 0.006, standard error = 12.02, n = 9) with
the equation summarized as follows: 7428.578 + (175.784)*Band5 + 1.041*Band5 2 +
0*Band53.
Band 5 has a marked inverse relationship to the water, including the internal moisture of a
plant stored in chloroplasts, thus giving strong rationale to the model. The negative
resultant of a cubic curve is that higher b5 values will erroneously report high AGB,
necessitating postprocessing modifications to the resulting AGB image. A systematic
inspection of such values revealed unrealistically high values of AGB were due to strong
sun glint on slope, which allowed for locally elevated Band 5 values in a manner not
reflective of vegetation. Negative AGB values were found in areas of modern
anthropogenic disturbance (e.g., recently cleared fields, burned areas, ash dumps, and
settled areas). These areas were removed using a land cover and land usebased mask
derived from Magee (2011). The masking process was vital given the low frequency of
samples from which the regression equations are derived.
2 We acknowledge that a variety of transitional forest types exist between upland and
bajo areas, and variation exists as well across both zones (Kunen et al. 2000; Dunning et
al. 2006). Because not all forest variants currently can be distinguished on widely
available satellite imagery, we have bifurcated the forest cover into two classes that can be
readily distinguished and have significant differences in biomass productivity.
3 These data were from a 10 year study of a 50 ha plot in the moist tropical forest of
Barro Colorado Island, Panama (BCI), where it was recorded that the biomass change rate
was 0.55 t ha1yr1. Because the modern Tikal forest had a larger basal area (38.8 m2 ha
1) than the BCI forest (28.1 m2 ha1), a proportional adjustment to 0.76 t ha1yr1was
performed. The BCI data were listed in the supplementary material section of the Phillips
et al. (1998) article.
4 Note that Culbert and associates (1990) projected a peak population of 62,000 for the
central core of Tikal (consisting of 120 km2). They estimated an extractive zone of 1,963
km2 surrounding Tikal that included a total of 425,000 inhabitants during the Late Classic
period.
5 The cited equation was developed for the same category of Neotropical forest found in
the Petn, although it was first applied to forests in South America.
9 Fire and Water: The Archaeological Significance
of Tikals Quaternary Sediments
Kenneth B. Tankersley, Nicholas P. Dunning, Vernon L. Scarborough, John G. Jones,
Christopher Carr and David L. Lentz

The Maya constructed reservoirs at Tikal to conserve water during the annual dry season
and to control and contain floodwaters during the rainy months. Six major reservoir
catchment areas drained the elevated precincts of Tikal (Figure 9.1), which covered an
area of approximately 300 ha with a total maximum reservoir capacity of more than
570,000 m3 (Scarborough and Gallopin 1991: 661). Surface water drained into the
reservoirs and culturally modified aguadas (i.e., depressions that contain ponds today) and
bajos (i.e., solution aldolines of varying size) (Scarborough 1993, 1994: 116; 2003: 51).
Quaternary sediments from abandoned and in-filling reservoirs, aguadas, and bajos at
Tikal provide a record of environmental change, human land use, and catastrophic
volcanic events. Unlike deeply inundated deposits from lake basins and ocean floors,
Maya reservoirs can be easily sampled with solid sediment coring devices and profile
excavations.
Figure 9.1. Location of Bajo la Justa, Tikal, and other sites mentioned in the text.
The chronostratigraphy of Tikals Quaternary sediments was determined with a series
of 45 AMS radiocarbon measurements made on carbonized plant remains and soil organic
matter (SOM) (Table 9.1) the former rich with airborne soot from the many cooking,
kiln, and forest clearing fires within the low-density urban setting (Scarborough et al.
2012). This work established that the Quaternary stratigraphy (i.e., soil horizons and
underlying sediments) extended from the prehabitation Late Pleistocene through the pre-
Hispanic human occupation. While PaleoIndian, Archaic, and Early Preclassic materials
and occupations have been found widely distributed within the Maya Lowlands, no
materials from these periods have thus far been recovered at Tikal. Archaeological studies
have demonstrated that some time prior to about 700 BC, small populations began to
reside at Tikal, and by 600 BC the first monumental architectural constructions appeared
(Laporte 2003). By the Late Preclassic period (ca. 350 BC), Tikal had developed into a
significant player in the emerging political landscape of the Maya Lowlands. Unlike
many larger centers in the nearby Mirador Basin, Tikal survived the turmoil of the second
century AD and emerged as a major center of the Early Classic period (Dunning et al.
2014). Tikal enjoyed variable prosperity during the Classic period, including a notable
downturn in its fortune or hiatus in the sixth century AD, but emerged as a paramount
center in the Late Classic period with a population estimated to have been around 60,000
in the eighth century (Martin and Grube 2008), though perhaps as low as 45,000 (seeLentz
et al., this volume). Tikals affluence declined dramatically in the ninth century and by 900
AD it was largely abandoned except for a small residual population that persisted for
another two hundred or so years.
Table 9.1. Chronostratigraphic data for Maya reservoirs and related contexts at
Tikal, Guatemala, including AMS radiocarbon sample composition, provenience, stable
carbon isotope analyses, context, measured radiocarbon years BP, calibrated age at 2 ,
and cultural period; samples were collected from excavation profiles and wet and dry
cores

Lab Composition Provenience Depth Measured Calibrated

Number (cm) 14C (yr Age (2 )
BP)

Beta- SOM Bajo de 60 2,850 40 13101040

266126a Santa Fe BC
(Op. 5C)

Beta- Charcoal Corriental 6580 990 + 40 AD 1010

258720a (Op. 1C) 1170

Beta- SOM Corriental 140180 2,010 + 34030 BC

280839b (Op. 1L 40
Core 8)

Beta- SOM Corriental 162194 2,110 + 40 19080 BC

266124a (Op. 1C)

Beta- SOM Corriental 180230 2,120 + 380170

280837b (Op. 1L 40 BC
Core 8)

Beta- SOM Corriental 265290 2,340 + 760400

258721a (Op. 1C) 40 BC

Beta- SOM Corriental 310312 8,960 + 82907970

270566b (Op. 1L 60 BC
Core 8)

Beta- SOM Corriental 90 2,110 + 40 19080 BC

266122a Arroyo (Op.
2A)

Beta- Charcoal Corriental Anthrosol 1,560 + AD 400

274990b Berm (Op. 40 570
1L Core 17)

Beta- SOM Corriental 60 1,930 + 90 BC-AD

266123a Pocket Bajo 40 80
1(Op. 2B)

88675b SOM Corriental 5060 6,250 + 53125076

Pocket Bajo 35 BC
2 (Op. 1L
Core 211)

88678b SOM Inscription 5060 4,170 + 28842632

(Op. 1L 35 BC
Core 201)

88678b SOM Inscription 8090 3,840 + 24622154

(Op. 1L 40 BC
Core 202)

88680b SOM Inscription 90100 3,000 + 14101049

(Op. 1L 65 BC
Core 202)

88681b SOM Inscription 120130 11,600 + 11761

(Op. 1L 100 11316 BC
Core 201)

Beta- Charcoal Palace (Op. Above 1,250 + AD 670

281750a 6Q) dam 40 880
collapse

Beta- Charcoal Palace (Op. Below 1,260 + AD 660

281751a 6Q) dam 40 880
collapse

Beta- Charcoal Palace (Op. 150 1,380 + AD 610

288914a 6L) 40 680

Beta- Charcoal Palace (Op. Dam fill 15,360 + 16860

281749a 6U) 50 16740 BC

Beta- SOM Palace (Op. Channel 3,310 + 18701850

281745a 6O) fill 40 BC

88638a SOM Palace (Op. 5060 3,360 + 17391535

6J-13 Core 30 BC
1-1)

88682a SOM Palace (Op. 100110 2,150 + 35855 BC

 6J-13 Core- 40
2)

Beta- Charcoal Perdido (Op. 4954 6,810 + 57405640

281747a 8K) 40 BC

Beta- SOM Perdido (Op. 5060 2,220 + 390180

289287b 8 Core 60 BC
N2E0)

Beta- SOM Perdido (Op. 110 1,540 + AD 350

280828a 8A) 40 540

Beta- SOM Perdido (Op. 180190 15,110 + 16860

289286b 8 Core 60 16740 BC
N2E0)

Beta- SOM Perdido (Op. 280290 15,480 + 16920

289285b 8 Core 60 16780 BC
N2E0)

Beta- SOM Perdido (Op. 390395 15,310 + 16820

289284b 8 Core 60 16670 BC
N2E0)

Beta- Charcoal Perdido 7679 1,570 + AD 400

281748a Pocket Bajo 40 570
(Op. 8D)

Beta- SOM Pucte (Op. 27 1,080 + AD 900

258723c 4A) 40 1300

Beta- Charcoal Temple 110 1,200 + AD 680

281746a Main Tank 40 890
(Op. 7C)

85584a SOM Temple 130140 1,230 + AD 721

Main Tank 25 839
(Op. 7C)

85585a SOM Temple 140162 1,830 + AD 143

Main Tank 25 215
(Op. 7C)

85583a SOM Temple 162194 1,250 + AD 701

Main Tank 35 811
 (Op. 7C)

88676b SOM Temple 7080 195 + 35 AD 1645

Silting Tank 1952
(Op. 7 Core
23-2)

88677b SOM Temple 110120 2,330 + 521216

Silting Tank 40 BC
(Op. 7 Core
23-2)

Beta- Charcoal Temple 130 1,370 + AD 640

298985a Silting Tank 30 680
(Op. 7A)

Beta- SOM Terminos 15 320 + 40 AD 1480

266125c (Op. 5A) 1660

Beta- SOM Terminos 33 2,000 + 170 BC-

258724c (Op. 5A) 40 AD 30

88684a SOM Terminos 40 950 + 30 AD 1024

(Op. 5F) 1156

88674a SOM Terminos 70 590 + 30 AD 1298

(Root?) (Op. 5F) 1413

Beta- SOM Terminos 100 1,940 + 50 BC-AD

279737a (Op. 5F) 40 120

Beta- SOM Tikal (Op. 37 3,000 + 14301260

258722c 3A) 40 BC

Beta- SOM Vaca del 27 400 + 40 AD 1440

279738c Monte (Op. 1640
11A)

Beta- SOM Vaca del 37 1,340 + AD 620

288915c Monte (Op. 40 690
11A)

Notes:
a Sample collected from an excavation profile.
 b Dry core with some compression.

c Wet core with compression.

Catastrophic Volcanic Events

While volcanic ash has been well documented as temper in ancient Maya ceramics, its
sources have remained in question (Ford 1991; Ford and Fedick 1992; Ford and Glicken
1987; West 2002). The plethora of ash temper in ceramics recovered from the limestone
lowlands of Guatemala has led some investigators to suggest that ash fell widely across
the region during the pre-Hispanic period of Maya occupation (Ford and Rose 1995: 149).
This theory has significant implications for understanding the prehistoric exploitation of
volcanic resources, landscape modification, and sustainability in the Maya Lowlands. If
significant quantities of volcanic ash fell on the limestone lowlands of Guatemala during
the pre-Hispanic occupation of the region, then we should expect to find direct positive
evidence in the numerous large reservoirs constructed in the Maya city of Tikal and in the
surrounding aguadas and bajos.
While it is possible for ash to survive intact in deep lakes, such as Yojoa in Honduras,
this is not the case in smaller and shallower reservoirs. In these settings, a significant
problem in sourcing ash from the Maya Lowlands is the fact that volcanic glass quickly
weathers (i.e., degrades) into smectite and kaolinite clay in moist, tropical, and alkaline
environments, which are characteristic of the region. This phenomenon is exemplified by
an ash fall from the El Chichn Volcano, which lasted from March 28 to April 4, 1982
(Robock 2002). Although Tikal was blanketed by several centimeters of ash during this
event, there is no visible evidence of the event in the soils today, and local residents report
that much of the ash had already been devitrified and incorporated into the soil within a
few months.
Smectite and kaolinite clays originate from the degradation of eruptive igneous rocks
(e.g., tuff) and volcanic ash (i.e., glass). Favorable physical and chemical conditions for
the formation of smectite and kaolinite include magnesium-rich environments with poor
drainage, which are characteristic of the reservoirs of Tikal (Laird et al. 1991). The
expandable nature of clays of this type is revealed by x-ray diffraction analysis of the
separated clay fraction (Moore and Reynolds 1997). Likely sources for the weathered ash
can be identified using bulk X-ray fluorescence (XRF) and electron microprobe data as
documented by Huff (2008) and Huff et al. (1999, 2000) for multiple Ordovician
bentonites.

Powder X-Ray Diffraction Analysis (XRD)

Powder X-ray diffraction (XRD) was used to identify the relative percentage mineral
composition of ancient reservoir, bajo, and arroyo sediments from Tikal. Samples (N =
115) included sediment from both solid cores and profile excavations. The minerals
calcite, kaolinite, quartz, and smectite were ubiquitous in the sediments of Tikal (Table
9.2). Zeolite was also identified in two reservoirs used for drinking water, Corriental and
Palace, and chlorite was identified in the Aguada de Terminos and Corriental Arroyo.
Calcite was the most abundant mineral (098 percent) and characterized by XRD peaks at
3.88 , 3.38 , 3.06 , 3.03 , 2.50 , 2.25 , 2.29 , 1.92 , and 1.88 . Smectite, the
second most abundant mineral (077 percent), was characterized by XRD peaks at 15.63
, 12.5215.46 , 5.98 , and 4.50 . The third most abundant mineral, quartz (037
percent), was characterized by XRD peaks at 4.204.25 , and 3.313.34 . Kaolinite (0
11 percent) was characterized by XRD peaks at 7.147.40 , and 3.533.56 . Chlorite
(010 percent) was characterized by XRD peaks at 14 , 7 , 4.72 , and 3.53.
Table 9.2. Relative percentage composition of Tikal sediments based on powder X-
ray diffraction (XRD) analysis; kaolinite, smectite, quartz, and zeolite are volcanogenic
minerals, which resulted from the decomposition of volcanic ash; AMS radiocarbon
dating suggests that ash falls at Tikal were frequent throughout the late Pleistocene and
Holocene. XRD samples were sieved through a 2 mm. Approximately 20 g of clay
sample was mixed with deionized water to make slurry in 100 ml beakers. Clay was
dispersed using a high-speed stirrer and gravity settling was used to obtain a fraction of
< 2 m. A 5 ml pipette was used to obtain a clay sample from the top of the slurry and
transferred to a glass slide and air-dried. A second oriented glass slide was prepared
from each sample and equilibrated overnight with ethylene glycol vapor. XRD patterns
were obtained for both the air-dried and glycolated samples. All slides were initially
scanned from 2o to 32o 2 at 0.5 increments and then broadened to 60o 2 on a Siemens
D-500 X-ray diffractometer using a Cu-K radiation source. The intensity threshold was
set at 1.6 and minerals were identified on the basis of peak position and peak intensity
as described by Chen (1977). Glycolated samples were prepared to test for the presence
of expandable clay minerals. Relative mineral percentages were calculated from the
total counts per second (cps), which were totaled for each 10 cm sample. The sum of all
peaks cps for each mineral was divided by the total cps for each 10 cm sample. The
relative percentage of each mineral was calculated from the total cps per mineral
divided by the total cps in the 10 cm sample.

Provenience Depth Calcite Chlorite Kaolinitea Smectitea Quartza

(cm) (%) (%) (%) (%) (%)

Aguada 20 0 8 7 48 37
deTerminos

Corriental 90 39 7 7 37 0
Arroyo

Corriental 010 31 0 0 31 37

Corriental 1020 66 0 0 17 10

Corriental 2030 68 0 0 10 11

Corriental 3040 76 0 0 12 12
Corriental 4050 80 0 0 9 7

Corriental 5060 70 0 0 15 13

Corriental 6070 86 0 0 8 6

Corriental 7080 75 0 0 11 7

Corriental 8090 47 0 0 38 36

Corriental 90 89 0 0 2 4
100

Corriental 100 81 0 0 15 5
110

Corriental 120 68 0 0 13 3
130

Corriental 130 52 0 0 9 4
140

Corriental 140 74 0 0 9 3
150

Corriental 150 71 0 0 14 7
160

Corriental 160 20 0 0 74 3
170

Corriental 170 18 0 0 40 6
180

Corriental 190 79 0 0 12 10
200

Corriental 200 68 0 0 5 4
210

Corriental 220 80 0 0 10 5
230

Corriental 230 70 0 0 18 11
240
Corriental 240 84 0 0 20 12
250

Corriental 250 84 0 0 4 4
260

Corriental 260 79 0 0 10 2
270

Corriental 270 92 0 0 7 5
280

Corriental 280 76 0 0 10 12
290

Corriental 290 64 0 0 19 12
300

Corriental 300 91 0 0 16 4
310

Corriental 010 0 0 0 77 23
Bajo 2

Corriental 1020 0 9 8 64 19
Bajo 2

Corriental 2030 0 10 10 63 17
Bajo 2

Corriental 3040 0 10 11 61 18
Bajo 2

Inscription 010 0 0 0 76 24

Inscription 1020 0 0 0 84 16

Inscription 2030 61 0 0 26 13

Inscription 3040 60 0 0 29 11

Inscription 4050 64 0 0 27 9

Inscription 6070 70 0 0 23 7

Inscription 7080 66 0 0 25 9
Inscription 100 80 0 0 12 8
110

Inscription 110 59 0 0 23 18
120

Palace Dam 73 0 0 0 0
Fill

Perdido 010 93 0 0 4 3

Perdido 1020 97 0 0 1 3

Perdido 2030 53 0 0 39 4

Perdido 3040 94 0 0 1 5

Perdido 4050 80 0 0 16 4

Perdido 5060 94 0 0 3 4

Perdido 6070 95 0 0 1 4

Perdido 7080 92 0 0 5 3

Perdido 8090 97 0 0 1 2

Perdido 90 74 0 0 21 5
100

Perdido 100 97 0 0 1 2
110

Perdido 110 69 0 0 25 6
120

Perdido 120 48 0 0 38 14
130

Perdido 130 98 0 0 1 1
140

Perdido 140 48 0 0 38 14
150

Perdido 150 98 0 0 1 1
160
Perdido 160 26 0 0 54 20
170

Perdido 170 38 0 0 54 8
180

Perdido 180 60 0 0 21 19
190

Perdido 190 67 0 0 23 10
200

Perdido 200 51 0 0 39 10
210

Perdido 210 71 0 0 23 6
220

Perdido 220 60 0 0 30 10
230

Perdido 230 42 0 0 52 6
240

Perdido 240 43 0 0 46 11
250

Perdido 250 34 0 0 54 12
260

Perdido 260 7 0 0 84 9
270

Perdido 270 25 0 0 59 16
280

Perdido 280 11 0 0 78 11
290

Perdido 290 37 0 0 48 15
300

Perdido 300 61 0 0 25 14
310
Perdido 310 36 0 0 41 23
320

Perdido 320 38 0 0 52 10
330

Perdido 330 35 0 0 34 31
340

Perdido 340 26 0 0 65 9
350

Perdido 350 33 0 0 62 5
360

Perdido 360 21 0 0 67 12
370

Temple 010 83 0 0 13 4
Main Tank

Temple 1020 81 0 0 14 5
Main Tank

Temple 2030 89 0 0 11 0
Main Tank

Temple 3040 85 0 0 9 3
Main Tank

Temple 4050 85 0 0 8 7
Main Tank

Temple 5060 89 0 0 9 2
Main Tank

Temple 010 82 0 0 18 0
Silting Tank

Temple 1020 79 0 0 21 0
Silting Tank

Temple 2030 90 0 0 10 0
Silting Tank

Temple 3040 84 0 0 16 0
 Silting Tank

Temple 4050 85 0 0 15 0
Silting Tank

Temple 5060 96 0 0 4 0
Silting Tank

Temple 6070 93 0 0 7 0
Silting Tank

Temple 7080 96 0 0 4 0
Silting Tank

Temple 8090 82 0 0 18 0
Silting Tank

Temple 90 80 0 0 20 0
Silting Tank 100

Temple 100 70 0 0 30 0
Silting Tank 110

Temple 110 79 0 0 21 0
Silting Tank 120

Temple 120 65 0 0 35 0
Silting Tank 129

a Volcanogenic mineral.

b Zeolite peak cooccurs with smectite on the X-ray diffractogram. Both minerals result from the weathering of
volcanic glass. Zeolite is universally used in water filtration (i.e., purification) systems.
c Natrolite is a zeolite found in amygdaloidal basalt and related igneous rock cavities. Zeolite is universally used
in water filtration (i.e., purification) systems.

Grin and Gven (1978: 128) found that smectite forms from weathering volcanic ash,
varying in composition from rhyolitic (quartz-rich) to basaltic (quartz-poor), and kaolinite
forms from the weathering of feldspar. Most bentonites (including smectite and kaolinite)
form from volcanic ash, ranging from rhyolitic to dacitic in composition. The association
of quartz and smectite in altered volcanic ashes is quite common. The abundance of calcite
is undoubtedly related to the weathering of Cretaceous and Tertiary limestone bedrock,
which underlies the terrain in the Maya Lowlands (Dunning et al. 1998), and from the
weathering of lime plaster widely used on Tikals architecture and plazas. Because quartz
is known to occur in carbonate rock (Chafetz and Zhang 1998), samples of local limestone
bedrock were subjected to powder XRD analysis. XRD peaks at 3.85,, 3.03,, 2.84,,
2.49,, 2.28,, 2.09,, 1.91,, 1.62,, 1.60,, and 1.52, characterized calcite. Peaks
typical of chlorite, kaolinite, quartz, smectite, and zeolite were completely absent from the
bedrock samples. This finding suggests that chlorite, kaolinite, quartz, and smectite were
deposited as aeolian minerals rather than through dissolution of the surrounding bedrock
and subsequent water transport (contra Cowgill and Hutchinson 1963: 41), though some
volcanogenic materials were likely redeposited locally by fluvial and colluvial processes.
While the zeolite may also have been deposited as an aeolian mineral, it is also possible
that it is anthropogenic in origin. The Maya may have used zeolite, as well as quartz sand,
for water filtration (see also chapters by Scarborough and Grazioso Sierra, and Lentz et
al., this volume).

Microscopy
Quartz and zircon grains were isolated from Tikal soil horizons and underlying sediments
using LST heavy liquid separation and identified under high magnification (>600x) with a
Leica MZ12 stereo microscope and an Environmental Scanning Electron Microscope
(ESEM). The quartz grains appear as dispersed slivers, fragments of euhedral crystals, and
complete bipyramidal crystals with scarce inclusions. Together, these characteristics
suggest they represent first quartz or an extrusive igneous or volcanic origin (Figure
9.2). Zirons cooccur with the euhedral bypyramidal quartz and appear as elongated
dipyramidal tetragonal crystals about 250 m in length with fluid inclusions (Figure 9.3).
Like quartz, zircons can withstand extreme changes in heat and pressure without
compromising their original composition and are direct evidence of catastrophic volcanic
events (Bindeman 2006). Zircons are common in felsic, granitic, and ultrapotassic igneous
rocks.

Figure 9.2. Environmental scanning electron micrograph of a bipyramidal euhedral

quartz crystal from Tikal reservoir sediments.
Figure 9.3. Environmental scanning electron micrograph of a bipyramidal euhedral a
zircon crystal from Tikal reservoir sediments.

X-Ray Fluorescence Spectrometry (XRF)

While chlorite, euhedral bipyramidal quartz crystals, kaolinite, smectite, and zircons
suggest a volcanic ash source, Tikal is located in the Sahara-Sahel Dust Corridor (Moreno
et al. 2006) and African dust has been identified as a major component of soils overlying
other carbonate landmasses in the Caribbean Basin, including Florida, the Bahamas, and
Barbados (Muhs et al. 2007). To determine whether the volcanogenic minerals from the
reservoirs of Tikal originated as airborne Sahara-Sahel dust or volcanic ash, the trace
element composition of the reservoir sediments from Tikal was analyzed using whole rock
X-Ray Fluorescence Spectrometry (XRF).
XRF was used to determine the trace element concentration of degraded volcanic ash.
Sediment samples, which contained high relative percentages of volcanogenic minerals,
were selected for XRF analysis from two reservoirs of Tikal, Corriental and Perdido. A
Rigaku 3070 X-ray fluorescence spectrometer was used to determine the intensity of the
trace elements Mo, Ba, Co, Cr, Cu, Nb, Pb, Rb, Sr, Th, U, V, Y, and Zn. Significant
differences between the trace element content of sediment from the reservoirs of Tikal and
Sahara-Sahel dust were found in the ratios of Ni, Cr, Zr, and Y (Tankersley et al. 2011).
Ratios are used rather than absolute amounts to eliminate the effect of the large amounts
of locally derived calcite. The trace element content of Sahara-Sahel dust has a
significantly lower range of Cr/Ni ratio and a higher Zr/Y ratio than do the samples from
Tikal (Figure 9.4).
Figure 9.4. The trace element content of Maya Lowland sediment samples compared to
volcanoes in Guatemala, Mexico, and El Salvador.
The Ilopango TBJ eruption is the largest and best-documented Holocene volcanic event
in Central America (Hart and Steen-McIntyre 1983; Sheets 2002). It occurred during the
Early Classic Period, some time between AD 408 and 536, and its ecological and cultural
impact would likely have been felt throughout the Maya region (Dull et al. 2001).
Professor Payson Sheets of the University of Colorado kindly provided our team with
samples of the TBJ Tephra for comparison. Data for other volcanic components are
available in the database developed by Carr (Carr et al. 2007). XRD analysis demonstrated
that the TBJ Tephra is composed of plagioclase feldspar with lesser amounts of quartz and
a large amount of glass. In other words, there are no clays in the TBJ Tephra, because the
glass is still intact, unlike the Tikal sediments, where all the glass has converted to
smectite. It is possible that the abundance of calcite in the Tikal reservoirscompared to the
slopes of the Ilopango Volcano accounts for some of this difference (Cowgill and
Hutchinson 1963). Weathering, however, should not affect the trace element ratios for
high-field-strength elements like Cr, Nb, Ti, Y, and Zr (Winchester and Floyd 1977; Floyd
and Winchester 1978; Maynard 1992). Nickel is more mobile than Cr in acidic soils under
tropical conditions (Maynard 1983) and accumulates lower in the profile, the mechanism
of genesis for lateritic nickel deposits. However, it should remain fixed in the high
carbonate environment of these deposits. The Ni/Cr ratios in the individual cores at Tikal
are constant with depth, indicating that vertical migration of Ni has been minimal. Our
XRF analysis of the TBJ Tephra sample found a higher ratio of Zr/Y than the sediments
from the reservoirs of Tikal. Similar results were obtained using the analyses in the Carr
database. Note that the position for Ilopango in Figure 9.4 is an average of our sample and
the Carr data.
Similar comparisons were made using the log Zr/TiO2and log Nb/Y ratios with volcanic
materials from Mexican, Salvadoran, and Guatemalan volcanoes. The trace element
content of Tikal sediment is comparable to that of volcanic rocks and Tephra from
Guatemalan or Salvadoran volcanoes (Table 9.3). Like the Zr/Y and Ni/Cr ratios, the log
Zr/TiO2and log Nb/Y ratios clearly show Tikal overlaps with Mexican and Guatemalan
sources (Figure 9.5).
Table 9.3. Trace element content of Tikal reservoir sediments and comparative
volcanic sources. Sample powders were pressed into briquettes at 2,000 psi. A separate
aliquot was heated to 1000 C for 1 hour to measure volatile content. Intensity data
were converted to parts per million (ppm) using bivariate and multiple variable
regressions applied to United States Geological Survey, National Institute of Standards,
and Japan Geological Survey rock standards

Location Reservoir Depth SiO2 Al2 Fe2O3t MnO MgO CaO

O3

Tikal Corriental 180 32.0 14.9 4.20 0.24 0.15 17.35

190

Tikal Corriental 190 28.4 12.3 3.55 0.25 0.15 23.10

200

Tikal Corriental 200 40.1 13.0 4.99 0.16 0.15 7.07

210

Tikal Corriental 210 36.2 10.8 2.41 0.06 0.14 17.39

220

Tikal Corriental 220 39.4 12.6 3.94 0.09 0.13 9.19

230

Tikal Perdido 180 42.4 17.8 6.62 0.34 1.53 9.68

190

Tikal Perdido 190 42.0 17.9 6.70 0.34 1.51 9.26

200

Tikal Perdido 200 42.9 18.5 7.21 0.28 1.54 8.73

210

Tikal Perdido 210 43.9 18.6 7.34 0.31 1.58 7.63

220

Tikal Perdido 220 43.6 18.6 7.09 0.38 1.56 7.86

230

Ilopango 68.5 15.3 3.37 0.13 1.32 3.01

Tephra

Tierra Average 77.5 13.0 1.25 0.07 0.20 1.15

Blanca
Joven

Tephra Average 60.7 11.0 4.90 0.07 1.56 5.78

Saharan
Dust

Antilles Average 57.4 17.3 7.74 0.16 3.90 8.44

Tephra

Mexico Average 58.4 17.8 6.48 0.15 2.70 6.58

Tephra

Guatemala Average 59.5 17.5 5.97 0.18 2.45 6.03

Tephra

El Average 57.4 17.3 7.98 0.16 2.90 7.15

Salvador
Tephra

Honduras Average 52.2 18.3 8.75 9.63 0.17 4.32

Tephra
Figure 9.5. The trace element content (log Zr/TiO2 and log Nb/Y ratios) of reservoir
sediment from Tikal.

Paleoenvironment and Land Use

While the archaeology of Tikal has been extensively researched, paleoenvironmental
changes and human modification of the vegetation are only partially understood. On the
basis of a synthesisof archaeological and ecological information, Turner and Sabloff
(2012) suggested that vegetation in the Maya Lowlands degraded as human population
increased over time. They attribute the economic chronocline that occurred during the
Late Classic period in part to deforestation related to an increase in urbanization and
agricultural intensification. In other words, vegetation at Tikal would have changed
through time as the Maya political and domestic economies grew and the demands on
locally available resources increased in the face of population expansion.
An increasing diet of C4 photosynthetic pathway plant foods such as maize and maize
eating animals is suggested by stable carbon isotope values obtained on Maya bone
collagen, bone apatite, and tooth enamel (Emery et al. 2000; Parker 2012; Repussard
2009; Scherer et al. 2007; Tykot 2002, 2004, 2006; Tykot and Staller 2002; Tykot et al.
1996; White et al. 1993). Note that of all of the food plants identified at Tikal, maize is the
only one employing the C4 photosynthetic pathway (see Lentz et al., this volume). If
increasing human populations at Tikal were dependent on maize agriculture, then we
should expect to find these changes reflected in the 13C isotope values of the Preclassic
and Classic period sediments preserved in the reservoirs, aguadas, and bajos. The isotopic
content of sediments at Tikal and the greater Maya Lowlands show that the 13C isotope
values are directly related to the sites vegetation composition (i.e., C3, C4, and CAM
photosynthetic pathway plants such as tropical trees, maize, and succulents, respectively
(Beach et al. 2006, 2008, 2011; Burnett 2009; Dunning et al. 2002; Lane et al. 2008;
Wright et al. 2009).
We examined the 13C isotope values of organic matter in Quaternary sediments from
Maya reservoirs, aguadas, and a diminutive pocket bajo. The objective of our analysis was
to compare stable carbon isotope values of organic matter in Quaternary sediments from
aguada contexts whose vegetation composition have been identified using palynology
with those from unknown archaeological contexts in urban center basins in order to
understand their relationship better.

Environmental Setting
Most of the site core at Tikal, composed of numerous temples, plazas, ball courts, palaces,
reservoirs, and most of the modest residential structures surrounding the polity center,
occupied areas that were once vegetated with upland tropical forest. Within the upland
forest there are many gradations such as ramonals as described by Lundell (1937)
located on the drier hill slopes, mesic upland forest in the low areas of ravines, cohune
palm forests in low areas with high clay content, hill base forests on the footslopes of hills,
sabal forest in areas similar to hill-base forests but dominated by sabal palms, and finally
transitional forests (Schulze and Whitacre 1999) adjacent to the wetlands. To the east of
the Tikal center is an extensive seasonal wetland called Bajo de Santa Fe, with its thick
black, clayey soils and low forest canopy (see Dunning et al., this volume, for a more
thorough description of the seasonal wetlands). Plant macroremains identified from Tikal
(Lentz et al., this volume; Lentz and Hockaday 2009) tell us that both the uplands and the
bajos were exploited heavily by the ancient Maya as sources of fuel, timber, and
agricultural products.
A pollen sequence was recovered from a 110-cm-deep pit excavated in the floor of
Aguada de Terminos, a reservoir pond on the margins of the Bajo de Santa Fe, but still
within the outer limits of Tikals sprawling residential zone. That excavation (Op. 5C) and
the pollen sequence are detailed in Chapter 6 of this volume (see Figures 6.6 and 6.7 and
associated text). The sediments within the aguada include a distinct unconformity at 90
cm, where Late Classic and later sediments directly overlie Late Preclassic material; this
disjunction likely represents dredging of the aguada to maintain its capacity. Preclassic
sediment contained pollen including Zea mays as well as other cultigens, as well as
Poaceae (grass) and other disturbance species indicative of significant forest clearance and
nearby agriculture. As will be discussed later, the maize and grasses are also reflected in
the carbon isotope composition of these early sediments. Nearby agricultural terraces also
reflect intensive cultivation in the area of the reservoir, though it is not know whether
these features date to the Late Preclassic, Late Classic, or both periods (Dunning et al., this
volume).
The pollen sequence associated with sediments between 90 and 40 cm appears to reflect
vegetation in the period from roughly AD 700 to 1100, representing the latter part of the
Late Classic, Terminal Classic, and first century of the Postclassic. The pollen record in
this zone shows significant forest clearance and maize cultivation in the Late Classic,
tapering through the Terminal Classic, and petering out in the Postclassic. Above 40 cm,
the pollen record is one of forest regrowth and persistence into the present day. As
expected, the carbon isotope record from organic matter within the sediments in these two
zones also reflects a transition from a landscape with abundant agricultural fields to
postabandonment forest dominance.

Vaca del Monte

The Vaca del Monte core contains 35 cm of organic-rich clays, representing 1,360 years of
sediment deposition (see Figure 8.6). A total of eight samples were examined. Pollen in
this sequence can be broken down into three zones. Overall, the core is dominated by
Cyperaceae (sedge family), a family known to produce copious quantities of readily
dispersed pollen grains. Sedges are currently found surrounding the Vaca del Monte bajo.
The basal zone, from 35 to 30 cm, reflects Classic to Late Classic occupation near the
coring location. Here, we see what is likely to be nearby cultivation/disturbance as
reflected by high concentrations of particulate charcoal and Asteraceae pollen, known to
reflect a disturbed rather than forested environment. Poaceae (grass family), Polygonaceae
(knotweed family), and Zea mays (maize) pollen are also somewhat elevated during this
basal zone. The occurrence of maize at this time reflects the cultivation of this plant near
the bajo.
The middle zone, occurring at 3020 cm, dates from some time prior to 400 BP. Here
we still see elevated concentrations of charcoal, but reduced occurrences of disturbance
taxons. Maize pollen is wholly lacking. Likely, the water-filled bajo was abandoned for
agricultural usage, though agriculture continued somewhere nearby as reflected by
elevated charcoal concentrations indicating that slash and burn clearing continued to take
place in the area. The abandonment of the Vaca del Monte bajo is also reflected by the
presence of Typha (cattail) pollen in this zone. This plant, although potentially
economically significant, does not appear to have been utilized by the ancient Maya and
likely signals a lack of maintenance in the wetlands.
The uppermost zone, undated and occurring from 20 to 0 cm, reflects a total
abandonment of the area for agricultural purposes. Charcoal concentrations are low,
indicating the local burning had ceased. Forest taxons become more plentiful at this time
and include Alchornea, Arecaceae (palm family), Combretaceae (probably Terminalia
buceras [bullet tree]), and Moraceae (mostly Brosimum-type). Disturbance taxons are
further reduced during this zone, although there is a slight increase in Poaceae pollen,
possibly representing an aquatic grass. The current mostly dry bajo is covered with aquatic
grasses (Olyra and Phragmites) and is probably the source of these grains, as the rest of
the pollen sequence reflects an infilling of the bajo. Spores from fern A (mostly or
exclusively Polypodiaceae) become abundant during this period. Elsewhere, there is a fern
spike following the total abandonment of a region, although this must remain speculative
for now. Salvinia spores also increase during this upper zone. This is an aquatic fern and
its presence in the Vaca del Monte bajo is consistent with the abandonment of the wetland
for agricultural use. This taxon is normally scarce from active wetlands, as it steals
nutrients from the system and is easy to clean from valued wetlands.
13C Isotope Samples
We examined the 13C isotope values of 80 bulk organic matter samples from seven
locations at Tikal and vicinity four Maya reservoirs (Corriental, Perdido, Temple Main
Tank, Temple Silting Tank) in central Tikal, two aguadas (Terminos, Vaca del Monte) in
peripheral Tikal, and a small depression (Corriental Pocket Bajo) used as a control. All of
the sediments were collected from AMS radiocarbon dated solid sediment cores or
excavation wall profiles. The 13C isotope values for the bulk organic matter from Tikals
reservoirs and the control depression are presented in Table 9.4. The 13C isotope values
for bulk organics from the Corriental Reservoir ranged from 18.5 to 28.1, with an
average of 21.9 (N = 20). The highest 13C isotope value was obtained from a
Postclassic stratum (<990 + 40 14C yr BP). The lowest 13C isotope value was obtained
from the upper 10 cm and is associated with present-day vegetation. The 13C isotope
values for bulk organic matter from the Perdido Reservoir ranged from 20.3 to
25.9, with an average of 23.2 (N = 13). The highest 13C isotope value was obtained
from a Late Preclassic stratum (2,220 + 6014C yr BP) and the lowest was obtained from an
Early Classic or older stratum (>1,540 + 4014C yr BP). The 13C isotope values for bulk
organic matter from the Temple Main Tank Reservoir ranged from 23.2 to 28.1,
with an average of 25.7 (N = 6). Both the highest and lowest 13C isotope values were
obtained from a Late Classic or more recent stratum (<1,200 + 40 14C yr BP). The 13C
isotope values for bulk organic matter from the Temple Silting Tank Reservoir ranged
from 22.8 to 28.9, with an average of 25.7 (N = 9). The highest 13C isotope
value was obtained from a Late to Middle Preclassic to Late Classic stratum (1,370 + 30 to
2,330 + 4014C yr BP), and the lowest was obtained from a Post-Conquest or more recent
stratum (195 + 35 14C yr BP). The 13C isotope values for bulk organic matter from the
Aguada de Terminos ranged from 12.9 to 30.3, with an average of 23.7 (N =
17). The highest 13C isotope value was obtained from a Late Preclassic stratum (1,940 +
4014C yr BP), and the lowest was obtained from a Post-Conquest stratum (< 320 + 40 14C
yr BP), likely representing present-day vegetation. The 13C isotope values for bulk
organic matter from the Vaca del Monte aguada ranged from 23.8 to 30.3, with an
average of 27.5 (N = 8). The highest 13C isotope value was obtained from a Late
Classic to Post-Conquest stratum (400 + 40 to 1,340 + 4014C yr BP), and the lowest was
obtained from a Late Postclassic to Post-Conquest stratum (<400 + 4014C yr BP). The
13C isotope values for bulk organics from the Corriental Pocket Bajo ranged from
19.4 to 22.9, with an average of 21.8 (N = 7). Both the highest and lowest 13C
isotope values were obtained from Archaic strata (6,250 + 3514C yr BP). However, it
should be noted that argilloturbation (shrinking, swelling, and inversion of clay) is likely a
significant pedogenic process in the pocket bajo, and lower/older horizons are likely
highly intermingled.
Table 9.4. Stable carbon isotope values for insoluble soil organic matter (SOM)
from the reservoirs of Tikal, Guatemala, with sample provenience, stratigraphic context,
distance to the city center (i.e., the Great Plaza), and cultural period. Samples were then
air-dried, hand reduced to a powder (<75 mm) in a mortar and pestle, and decalcified at
room temperature in a 1 N HCl bath overnight (i.e., >12 hours). The acid was rinsed out
of each sample with distilled, demineralized, deionized water until the pH was neutral.
While an hour-long 1 N HCl bath at room temperature is usually adequate to remove
carbonates from sediment samples, extended acidification was necessary to remove all
mineral calcite and prevent an artificial enrichment of the 13C values in the bulk
organic samples. Stable carbon isotope values on bulk organic matter was obtained on a
DELTAplusXP stable isotope ratio mass spectrometer (IRMS). The 13C isotopic values
of the bulk organic matter were measured with the IRMS relative to Vienna Pee Dee
Belemnite (VPDB) carbonate standards. A control blank (i.e., an empty tin capsule) was
run for each batch of samples (13C = 0). Cornstarch was used as a C4 control sample
(13C = 11.36 to 11.42, N = 2) and acetanilide was used as a C3 control sample
(13C = 29.31 to 29.75, N = 8). The precision of the standards was 0.1425 at 1-
sigma

Provenience Depth Distance and Cultural Period

13C (cm) Direction to Palace
() (m)

Corriental 10 1,300 SSE Post-Columbian

28.1

Corriental 20 1,300 SSE Post-Columbian

24.5

Corriental 30 1,300 SSE Post-Columbian

20.3

Corriental 40 1,300 SSE Postclassic

18.9

Corriental 50 1,300 SSE Postclassic

20.7

Corriental 60 1,300 SSE Postclassic

21.2

Corriental 70 1,300 SSE Early Postclassic

21.7

Corriental 80 1,300 SSE Early Postclassic

21.6

Corriental 90 1,300 SSE Late/Terminal

20.3 Classic
 Corriental 100 1,300 SSE Late/Terminal
20.4 Classic

Corriental 110 1,300 SSE Late/Terminal

20.5 Classic

Corriental 120 1,300 SSE Late/Terminal

20.4 Classic

Corriental 130 1,300 SSE Late/Terminal

20.5 Classic

Corriental 140 1,300 SSE Late/Terminal

20.4 Classic

Corriental 145 1,300 SSE Late Preclassic or

18.5 later

Corriental 150 1,300 SSE Late Preclassic or

27.0 later

Corriental 155 1,300 SSE Late Preclassic or

25.2 later

Corriental 165 1,300 SSE Late Preclassic or

21.9 later

Corriental 175 1,300 SSE Late Preclassic or

27.2 later

Corriental 185 1,300 SSE Late Preclassic or

18.9 later

Corriental 40 1,150 SSE Archaic

21.9 Pocket Bajo
2

Corriental 50 1,150 SSE Archaic

22.8 Pocket Bajo
2

Corriental 60 1,150 SSE Archaic

19.4 Pocket Bajo
2

Corriental 70 1,150 SSE Archaic

22.9 Pocket Bajo
2

Corriental 80 1,150 SSE Archaic

21.5 Pocket Bajo
2

Corriental 90 1,150 SSE Archaic

21.7 Pocket Bajo
2

Corriental 100 1,150 SSE Archaicorearlier

22.5 Pocket Bajo
2

Perdido 10 1,050 SW Late Preclassic or

20.3 later

Perdido 20 1,050 SW Late Preclassic or

21.5 later

Perdido 30 1,050 SW Late Preclassic or

20.6 later

Perdido 40 1,050 SW Late Preclassic or

21.9 later

Perdido 50 1,050 SW Late Preclassic or

20.7 later

Perdido 70 1,050 SW Late Preclassic or

23.6 later

Perdido 80 1,050 SW Late Preclassic or

25.7 later

Perdido 90 1,050 SW Late Preclassic or

24.1 later

Perdido 100 1,050 SW Late Preclassic or

23.6 later

Perdido 110 1,050 SW Late Preclassic or

23.6 later

Perdido 120 1,050 SW Early Classic or

25.9 earlier
 Perdido 130 1,050 SW Early Classic or
25.2 earlier

Perdido 140 1,050 SW Early Classic or

25.4 earlier

Temple 10 120 SW Post-Columbian

28.1 (Main Tank)

Temple 20 120 SW Post-Columbian

26.9 (Main Tank)

Temple 30 120 SW Post-Columbian

26.8 (Main Tank)

Temple 40 120 SW Late Classic or later

25.1 (Main Tank)

Temple 50 120 SW Late Classic or later

24.2 (Main Tank)

Temple 60 120 SW Late Classic or later

23.2 (Main Tank)

Temple 10 120 SW Post-Columbian

28.2 (Silting
Tank)

Temple 20 120 SW Post-Columbian

27.7 (Silting
Tank)

Temple 60 120 SW Post-Columbian

25.5 (Silting
Tank)

Temple 70 120 SW Post-Columbian

28.9 (Silting
Tank)

Temple 80 120 SW Post-Columbian

27.5 (Silting
Tank)

Temple 90 120 SW Middle Preclassic

24.0 (Silting to Late Classic
 Tank)

Temple 100 120 SW Middle Preclassic

23.7 (Silting to Late Classic
Tank)

Temple 110 120 SW Middle Preclassic

22.8 (Silting to Late Classic
Tank)

Temple 120 120 SW Middle Preclassic

23.4 (Silting to Late Classic
Tank)

Terminos 0 5,300 E Post-Columbian

30.0

Terminos 5 5,300 E Post-Columbian

29.2

Terminos 10 5,300 E Post-Columbian

29.3

Terminos 15 5,300 E Early Postclassic to

28.2 Post-Columbian

Terminos 25 5,300 E Early Postclassic to

27.1 Post-Columbian

Terminos 30 5,300 E Early Postclassic to

27.5 Post-Columbian

Terminos 35 5,300 E Early Postclassic to

26.6 Post-Columbian

Terminos 45 5,300 E Early Postclassic

25.9

Terminos 50 5,300 E Late/Terminal

25.9 Classic

Terminos 55 5,300 E Late/Terminal

25.3 Classic

Terminos 60 5,300 E Late/Terminal

25.7 Classic
 Terminos 65 5,300 E Late/Terminal
26.2 Classic

Terminos 75 5,300 E Late/Terminal

17.5b Classic

Terminos 80 5,300 E Late/Terminal

18.8b Classic

Terminos 90 5,300 E Late Preclassic

12.9b

Terminos 95 5,300 E Late Preclassic

12.9b

Terminos 100 5,300 E Late Preclassic

14.2b

Vaca del 1a 9,000 ENE Late Postclassic to

28.6 Monte Post-Columbian

Vaca del 7a 9,000 ENE Late Postclassic to

28.4 Monte Post-Columbian

Vaca del 13a 9,000 ENE Late Postclassic to

28.3 Monte Post-Columbian

Vaca del 19a 9,000 ENE Late Postclassic to

30.3 Monte Post-Columbian

Vaca del 25a 9,000 ENE Late Postclassic to

26.6 Monte Post-Columbian

Vaca del 28a 9,000 ENE Late Classic to

27.1 Monte Post-Columbian

Vaca del 31a 9,000 ENE Late Classic to

26.6 Monte Post-Columbian

Vaca del 34a 9,000 ENE Late Classic to

23.8 Monte Post-Columbian

a Wet core; depth does not represent the actual stratum because of compression.

b This stratum contained pollen from Steraceae, Poaceae, Polygonaceae, Solanaceae weeds associated with
nearby agriculture, disturbance, and clearing. Maize pollen is also common. It is associated with nearby Maya age
 settlement including agricultural terraces.

Vegetation Variation
The 13C isotope values on bulk organic matter from Tikal ranged from 12.9 to
30.3, and averaged 24.2 (see Table 9.4). As noted, both the highest and lowest 13C
isotope values were obtained from the Aguada de Terminos. Pearsons correlation
coefficient was calculated using 13C isotope values to determine whether vegetation
composition varied by the age of the sediment sampled, that is, regardless of the presence
or absence of human occupation. A strong positive relationship was found between the age
of the sediment and the 13C values (Pearsons r = 0.42, p = 0.0001). That is, the 13C
isotope values on bulk organic matter at Tikal increase with the age of the sediment.
Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied by cultural period (i.e., Archaic,
Preclassic, and Classic, Postclassic, and Post-Conquest). The average 13C isotope values
of bulk organic matter from Archaic age sediments (N1 = 7) was 21.8 (standard
deviation = 1.1950). The average 13C isotope values of bulk organic matter from
Preclassic Maya age sediments (N2 = 3) were 13.3 (standard deviation = 0.7506). The
difference between these means was significant t (2) = 5.578, p = < 0.031).
Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied between the Preclassic and Early
Classic Maya occupations of Tikal. The average 13C isotope values of bulk organic
matter from Early Classic Maya age sediments (N2 = 3) were 25.5 (standard deviation
= 0.3606). The difference between the Preclassic and Early Classic means was significant
t (2) = 23.224, p = < 0.002). In other words, there is a significant difference between the
vegetation composition of the Preclassic and Early Classic Maya occupations of Tikal.
Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied between the Early Classic and
Terminal Classic Maya occupations of Tikal. The average 13C isotope values of bulk
organic matterfrom Terminal Classic Maya age sediments (N2 = 12) were 21.8
(standard deviation 3.0544). The difference between the Preclassic and Early Classic
means was significant t (2) = 20.265, p = < 0.002). In other words, there is a significant
difference between the vegetation composition of the Early Classic and Terminal Classic
Maya occupations of Tikal.
Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied between the Terminal Classic and
Post Classic Maya occupations of Tikal. The average 13C isotope values of bulk organic
matter from Post Classic Maya age sediments (N2 = 6) were 21.7 (standard deviation =
2.3123). The difference between the Terminal Classic and Postclassic means was not
significant t (2) = 1.335, p = < 0.240). In other words, there is no significant difference
between the vegetation composition of the Terminal Classic and Postclassic Maya
occupations of Tikal.
 Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied between the Postclassic and the
Post-Conquest cultural periods. The average 13C isotope values of bulk organic matter
from Post-Conquest age sediments (N2 = 14) were 27.2 (standard deviation = 2.4828,
standard error = 1.9359). The difference between these means was significant t (5) =
2.733, p = < 0.041). There is a significant difference between the vegetation composition
of the Postclassic and Post-Conquest cultural periods of Tikal.
Independent t-tests of the 13C isotope values of bulk organic matter were used to
determine whether the vegetation composition varied by its proximity to the city center,
that is, the plaza (urban < 2,000 m vs. rural > 5,000 m). The average 13C isotope value on
bulk organic matter from sediments within 2,000 m of the plaza (N1 = 65) was 23.3
(standard deviation = 2.78978, standard error = 0.37617). The average on sediments more
than 5,000 m from the plaza (N2 = 15) was 24.9 (standard deviation = 5.27049,
standard error = 1.05410). The difference between these means is highly significant (t (78)
= 5.55934, p = < 0.000001). Stated differently, the difference between the vegetation
composition of urban (cleared upland forest) and rural (bajo) areas of Tikal is highly
significant. This finding, however, likely reflects an age bias at least as much as showing
spatial variation. Because neither Terminos nor Vaca del Monte is older than the Late
Preclassic, and the majority of samples are from preabandonment times, the rural values
should trend higher than the urban ones.
While these statistical tests demonstrate that the Maya had a profound effect on the
vegetation composition of Tikal, it is also possible that the results may be biased by
anthroturbation, that is, Quaternary deposits that were disturbed by ancient dredging
activities. An additional independent t-test was run in order to determine whether the 13C
isotope values on bulk organic matter obtained from basins with evidence of dredging
activities (i.e., Temple Main Tank, Temple Silting Tank, and Terminos) were significantly
different from those of basins with less evidence of dredging (i.e., Corriental, Corriental
Pocket Bajo, Perdido, and Vaca del Monte). It is important to note that Perdido shows
some evidence of dredging, though not much. Also, a single compressed core from Vaca
del Monte is not sufficient to say one way or another whether it was dredged.
The average 13C isotope value on bulk organics from sediments from dredged basins
(N1 = 32) was 24.7 (standard deviation = 4.60860, standard error = 0.81469). The
average on sediments from basins with less evidence of dredging (N2 = 48) was 23.2
(standard deviation = 3.03824, standard error = 0.43853). The difference between these
means is highly significant t (78) = 3.56977, p = < 0.000001). The difference between the
vegetation record of basins that were dredged and those that were not dredged or less
clearly dredged may reflect the removal of sediment with high 13C values from some
reservoirs.
Independent t-tests for the 13C isotope values on bulk organic matter obtained from the
Quaternary sediments in reservoirs, aguadas, and pocket bajos were rerun with samples
from clearly dredged basins removed. Independent t-tests on 13C isotope values on bulk
organics between nondredged or less dredged urban basins (N1 = 40) and nondredged or
less dredged rural basins (N2 = 8) were highly significant (t (46) = 4.4662, p = <
0.00291). Independent t-tests on 13C isotope values of bulk organic matter from
nondredged or less dredged sediments that predate the Maya occupation of Tikal (N1 = 7)
and nondredged or less dredged sediments that date to the Maya occupation (N2 = 33)
were highly significant (t (38) = 8.41997, p = < 0.000001). Independent t-tests on 13C
isotope values of bulk organic matter from nondredged or less dredged sediments that date
to the Maya occupation of Tikal (N1 = 33) and nondredged or less dredged sediments that
postdate to the Maya occupation (N2 = 20) were highly significant (t (51) = 4.69943, p =
< 0.00003).

Discussion
The 13C isotope values on bulk organic matter from the Quaternary deposits of Tikals
reservoirs and aguadas demonstrate that the Maya had a direct impact on the vegetation
composition of Tikal. The 13C isotope values also show, however, that C3 photosynthetic
pathway plants were widespread throughout time. There is no isotopic evidence from
reservoir sediments of widespread intensification of maize agriculture, though such
evidence may be found in the plant-rooting zone of some bajo soils (see chapter by
Dunning et al., this volume). Seasonal drought and dry climatic conditions can cause the
13C values of C3 photosynthetic pathway plants to increase by several mils and into the
range of 19.0 13C (Bowen and Beerling 2004; Difendorf et al. 2012; Farquhar et al.
1982; Freeman et al. 2011; Shim et al. 2009; Tieszen 1991; Wynn 2007; Wynn and Bird
2008; Wynn et al. 2005). Given this caveat, the isotopic evidence of C4 (and/or CAM)
photosynthetic pathway plants is confined to the Postclassic sediments of Corriental, the
Late Preclassic sediments of Corriental (18.9 to 18.5), and Terminos (18.8 to 12.9).
At Terminos, the highest 13C isotope values occur in Late Preclassic strata, which
contained pollen from C4 plants (e.g., Poaceae, Polygonaceae, and maize). There is also
archaeological evidence of nearby agricultural terraces and ancient settlement. At
Corriental there is an apparent decrease in the presence of C4 plants in the watershed
between the Late Preclassic and Early Classic. This decrease may reflect urbanization
trends at Tikal, where maize cropping was shifting farther away from the site center.
Rather than widespread intensive maize agriculture, variation in the 13C isotope values
on bulk organic matter from the Quaternary deposits in Tikals reservoirs and aguadas are
more likely related to a declining forest resource base as described by Lentz and
Hockaday (2009). Within central Tikal, residents likely grew C4 plant crops only in garden
plots likely comparable to those found at Cern, El Salvador (Lentz et al. 1996; Lentz and
Ramirez 2002; Linares et al. 1975; Sheets 2002; Lentz et al., this volume), with the
probable exception of pocket bajos like Perdido where intensive maize cultivation likely
took place on a larger scale (see chapter by Dunning et al., this volume). If the occupants
of Tikal consumed maize in great quantities, then it was likely produced more widely in
outlying areas (including along the margins of the sprawling Bajo de Santa Fe), and it is
possible that trade supplemented local garden produce. It is equally likely that C3 produce,
such as manioc, achira (Canna cf. indica L., malanga (Xanthosoma sagittifolium L.
Schott), and possibly other root crops, played a greater role in the Maya subsistence
economy of Tikal than previously thought, comparable to that described by Sheets and
Dixon (2011), Sheets et al. (2011), and Sheets et al. (2012) for Cern and vicinity.
Finally, during the Postclassic, the isotopic signature for maize cultivation increases
again at Corriental and is high at Terminos (where maize pollen also persists). This spatial
pattern may indicate that remnant populations at Tikal clustered around water sources,
where they both lived and farmed.

Conclusion
Quaternary sediments from reservoirs, aguadas, and bajos are composed largely of
degraded volcanic minerals. Indeed, a significant portion of the sediments is degraded ash
from explosive eruptions of volcanoes in Guatemala and Mexico. The sediments from
Tikals reservoirs, aguadas, and bajos provide the first unequivocal evidence for long-term
explosive volcanic ejecta deposition in the southern Maya Lowlands. Without ash fall
contributions soils on limestone surfaces would have likely remained skeletal and posed
significant problems for the long-term subsistence by ancient Maya populations.
Variation in the 13C isotope values on bulk organic matter from the Quaternary
deposits of Tikals reservoirs, aguadas, and bajos shows varying degrees of vegetation
change across time and space. The Maya clearly had a significant impact on the vegetation
composition of Tikal. In addition to erosion, soils suffered from nutrient depletion. The
regular, but unpredictable input of volcanic ash would have helped to offset the
degradation processes. For the Maya, whose cultural system was largely dependent on
local food production within both urban centers and their hinterlands, maintaining soil
productivity was critical.
 10 Fractious Farmers at Tikal
David Webster and Timothy Murtha

Introduction
Our chapter presents both data and a dilemma. It explores two fundamental issues
concerning the Classic Maya how kings, lords, and common farmers asserted rights to
basic agrarian resources and how these rights and resources were managed as conditions
changed (Webster 2002b, 2005; Murtha 2002). The dilemma is that we have almost no
information about these topics from inscriptions, art, and archaeology.1 What is clear is
that Maya people contended through time with increasingly complex, risky, and
demographically saturated agrarian environments. Such changes stimulated competitive
processes and status rivalry, which led to reduced sociopolitical resilience and contributed
heavily to eventual collapse.2 J. E. S. Thompson (1954) argued that peasant revolts
undermined Classic polities. We do not subscribe to this view, but our perspective
resembles Thompsons in emphasizing internal discord.
Our main focus is on niche construction and niche inheritance, concepts that have
recently entered the Maya literature (see Webster 2013). Niche construction is the
process whereby organisms, through their metabolism, their activities and their choices,
modify their own and/or each others niches (Odling-Smee et al. 2003: 419). As Hardesty
(1972) observed, culture is the human ecological niche. Humans, like other organisms,
engage in inceptive and counteractive perturbation and relocation, but they uniquely
evolve transgenerational cultural niches or traditions. Such niches are literally inherited
or otherwise culturally allocated through time. The dynamic cultural and sociological
dimensions of agrarian niche construction and inheritance are central to understanding
how Classic Maya polities were structured and their eventual fates.
Dennis Pulestons Tikal Sustaining Area project (Puleston 1973, 1983) was a superb
effort to understand how the Maya supported (apparently) large populations on Tikals
landscape. It stands out for its synthesis of landscape features, settlement character and
distribution, hinterland demarcation, demographic reconstruction, and agricultural
strategies. Much new information about Tikal has since accumulated, some of it our own
(Webster et al. 2004, 2008; Straight 2012; Burnett 2009). Our chapter is in part a critical
reevaluation of the agrarian implications of Tikals settlement and demographic patterns. It
also raises questions of agency how did Maya individuals or groups assert and enforce
rights to resources, and what were the consequences of such assertion? How did they
manage, successfully or not, transgenerational inheritance (or other) transfers of agrarian
capital resources, particularly land and soil?
Mayanists and behavioral ecologists typically build models of ancient food production
systems by emphasizing basic technologies, soil, rainfall and hydrological variables, crop
complexes, labor inputs and intensification, storage capacity, risk factors, and other
abstract components. A recent paper by Winterhalder and Puleston (2012) presents an
excellent Malthusian simulation of the strategies whereby the governing apparatus of
agrarian states (including the Maya) might have extracted resources from farmers in the
form of food or labor, and the tolerances and stresses linked to such strategies.3 As with
most such simulations, the governing/consuming sectors and the producing sectors are
treated monolithically, as is the environment, for any given interval. Equally important but
less often acknowledged are the socioeconomic intricacies and competitive processes
whereby individuals and groups acquired, maintained, transmitted, and sometimes lost
access to agrarian assets, particularly to soil. Ancient Maya landscapes were increasingly
fragmented into complicated mosaics of differential risk and productivity, because of both
niche construction and population growth. They also became the objects of intense passive
and active competition by Late Classic times and probably earlier. Such competitive
maneuvering is not directly accessible to us, but it played a critical role in how farmers
adapted to their resources, to each other, and to the demands and prerogatives of great
kings and lords. Here we explore how abstract, rational, formal models might be
augmented with elements of real-world messiness (see Webster 2005 for a similar analysis
of Copn).
Linkage between Tikals subsistence economy and its institutional or political economy
offers many insights into the stresses experienced by the Tikal Maya and has important
implications for the so-called Classic collapse. Whatever else it was, the collapse was a
rejection of the ancient institution of kingship, and particularly its pretensions to the
maintenance of cosmic, agrarian, and sociopolitical order (Webster 2002a, b). Collateral
casualties were the royal and noble courts that provided whatever overall administrative
and managerial capacities were deployed by political elites. We believe the Tikal Maya
were less canny managers of sustainable and resilient agricultural ecosystems than
reasonably effective copers with a cascade of changes that began in Preclassic times and
became especially challenging during the eighth and ninth centuries. If the collapse
interval coincided with unusually intense droughts, their effects were greatly compounded
by centuries of environmental transformation and the competition it stimulated. In the end
the coping capacity of the Tikal Maya proved insufficient, and the old system fell apart
(see Dunning and Beach 2010 for a multiregion overview).

Tikal as a Test Case

Tikal is unfortunately not the best region to analyze using the model we sketch out here.
Copn would be far better (Wingard 1993, 1996; Webster et al. 2000; Webster 2005). We
work with what we have, but it is best to be candid about deficiencies in the
archaeological record:
1. There are good data on recent soils and their distributions, both from around Tikal
itself and from the general region of the northeast Petn (Olson 1977, 1981;
Beach et. al 2005). We also have some specific data on Tikal soils, both ancient
and modern, from recent research, including our own (Burnett 2009; Burnett et
al. 2012; Dunning et al. this volume). Still lacking are detailed data for soil
transformations through time. Nor are there really long sediment cores available
 for Tikal such as those from Copn (Webster et al. 2005; Rue 1987), Lake Petn
Itza (Islebe et al. 1996), or the Petexbatun region (Dunning, Beach and Rue
1997; but see Dunning et al. this volume for shorter ones).
2. Remote sensing data for Tikal are not as sophisticated as those available for
nearby regions, such as that of San Bartolo (Griffin 2012). Particularly lacking is
LiDAR imagery.
3. The rough spatial extent of Tikals three most fundamental landforms, the
uplands, the bajos or bottomlands/wetlands, and the bajo margins, is known.
Unfortunately the size of its core agrarian territory is not. Until recently
archaeologists believed that the celebrated earthworks delimited this Late
Classic hinterland (Puleston and Callender 1967; Haviland 2008), but our own
research suggests otherwise (Webster et al. 2004, 2008). Proximity of other
major centers (Uaxactn, El Zotz, Yaxha, etc.) allows, however, reasonable
hinterland estimates (see Lentz et al., Chapter 8 this volume). More important
than the specific size or extent of Tikals hinterland is determining what was a
reasonable and representative sample of the character of the agrarian landscape.
4. The main outlines of Tikals culture history are clear, mainly from work in the
core architecture and its monuments (Martin and Grube 2008; Harrison 1999;
Sabloff 2003). Several studies document wider settlement distributions (Puleston
1973, 1983; Fry 1969; Webster et al. 2004, 2008; Straight 2012) that allow
reasonable reconstructions of the eighth century population. Earlier population
dynamics remain problematical.
5. The few extensive exposures of Tikal residences are all near the site center (see
Haviland 2003). Especially rare are excavations of subroyal elite residences.
6. We have direct isotopic evidence for cultivation of C4 plants, probably maize, in
some soil zones. Maize contribution to the diet is unclear because there has been
no isotopic study of a large collection of skeletons from different social ranks
and periods comparable, say, to that available for Copn (Reed 1998).
7. We do find some evidence for landesque-type agricultural features around Tikal
(e.g., agricultural terracing and irrigation systems see Dunning et al., Chapter
6: Scarborough and Grazioso Sierra, Chapter 2; and Lentz et al. Chapter 8 this
volume).
8. It is reasonable to conclude that Tikals basic agrarian economy was highly self-
contained. Tikal is not on a major waterway that might have facilitated bulk
commodity transport; nor were riverine or marine resources locally available in
appreciable quantities.

The Actors, Their Organization, and Their Interests

By the eighth century three general subgroups had economic and political interests in
Tikals agrarian landscape: farmers, kings and their immediate families, and titled lords
(who might be related to royalty).4 Many fine gradations of rank and status existed in
Maya society, but these three broad categories capture an essential dichotomy. Kings and
lords were consumers who did not labor to produce food, while commoners were basic
producers (see Webster 1985). All these groups had very different histories and adaptive
challenges in the evolution of the Tikal polity.

Farming Households
This ancient folk component of the Tikal population was present on the landscape in
small numbers at least by Early and Middle Preclassic times and had well-developed
agrarian traditions long before the first kings appeared. Displaced people from the Mirador
Basin polities might have arrived at the end of the Preclassic, one source of early
competition. Whatever their origins, farmers were the basic producers of agrarian products
and, eventually, the source of labor for elites. Farmers were organized on the nuclear or
extended family levels and probably practiced ancestor veneration (McAnany 1995). We
do not know, however, whether they had larger corporate descent groups or other
community identities. Early farmers lived in essentially egalitarian political communities,
but by the time of the first kings they had assumed the identity of peasants. Whether any
households were thereafter demoted to client or serflike status is conjecture, as is whether
there were slaves.

Kings
Tikals first dynastic rulers appeared around AD 63, but royal institutions were probably in
place two or three centuries earlier. Kings and their families were organized into royal
patrilineages and linked by marriage or alliance to the dynasties of other polities. Royal
factions from the earliest times probably competed for political dominance.

Subroyal Lords
Grandees bearing various titles were conspicuous in the inscriptions at many Maya courts
after about AD 650. Unfortunately Tikals monuments convey very little information
about them, although large palatial compounds indicate their presence (see Haviland
2008). Whenever they first appeared, these Tikal lords by Late Classic times were almost
certainly also organized into elite patrilineages. So far as we can tell, these subroyal lords
emerged later than the other two components, so their agrarian prerogatives had to be
carved out of earlier patterns of resource distribution. To the extent that kings and lesser
elites perceived themselves as separate, corporate interest groups they were natural
competitors for agrarian assets.
The basic concern of farmers was having a sufficient stake in land, water, and good soil
to feed themselves, to produce marginal surpluses to store against shortfalls, to accumulate
seed, and to fund elite households. Griffin (2012) calculates that such surpluses would
ideally be about 50 percent above household subsistence needs. The main concern of elites
was to extract both the products of agrarian labor and a portion of that labor itself to
support their large households and to fund ritual events, political dominance, and alliance.
We do not understand how kings, elites, or other nonfarmers were funded, especially
with food.
Rights to Agrarian Assets
How did these various kinds of agents map onto Tikals agrarian resources? Here we do
not mean ownership as we understand it today, but simply assertion of socially
recognized economic interests in property by multiple individuals, groups, or institutions.
As an example, during the eighteenth and nineteenth centuries rulers granted Egyptian
grandees called beys or pashas the rights to tax communities of peasant farmers. No doubt
rulers believed they had graciously given the pashas rights to lands they still ultimately
controlled, the pashas thought they in some sense owned the farming villages and their
lands, and the farmers thought the pashas were oppressive and predatory tax collectors
taking a cut from products of the lands the peasants owned and on which their ancestors
had lived from time immemorial. All these individuals or groups had customary interests
in land and its products. At Tikal similar overlapping claims probably emerged even
during the folk-farming phase but became enormously more complicated as kings and
elites dominated the polity.

Tikals Agrarian Landscape

Most projects carried out at Tikal identify two related landscape elements that influenced
productivity, settlement, and land use: 1) soil (type and properties, especially drainage
capabilities and texture) and 2) topographic position. Nevertheless, models of ancient
Maya agriculture at Tikal are primarily derived from conceptions of settlement and social
organization, political authority and population history, or studies of monumental
architecture rather than from these two key landscape elements. Often these conceptions
are misinformed, are tied to old interpretations about Tikal, and are stubbornly
maintained.5 Glassman and Annaya, for example, describe the Tikal settlement as follows:

The core area of the city, occupied by great temples and other elite constructions,
took up six square miles. An additional 19 square miles, 25 miles altogether, were
occupied by buildings. Another 22 square miles 47 square miles altogether was
passively defended by walls, moats and natural boundaries. These 22 square miles
were farmed intensively. They were the granary that the city made sure was never
destroyed by enemies and may indeed have been another of Tikals many reasons for
rising to prominence.
(Glassman and Annaya 2011: 6970)

In fact, almost all the major core architecture at Tikal is scattered over a core area of only
about 4 km2 we say scattered because there is a lot of empty space in this core and if all
the buildings were aggregated they would cover roughly 2 km2. Around this core and
contiguous with it is Central Tikal, the 89 km2 zone of densest urban residential
concentration what we call the demographic core (Webster et al. 2008: 369). All other
settlement mapped before 2006 is in the residual part of the 16 km2 Carr and Hazard
(1961) map and in the transects mapped by Puleston (1973), which together amount to
about 2930 km2. Glassmans and Annayas 47 mi2 figure is what archaeologists have
routinely called residential Tikal, the roughly 120 km2 region long presumed to be
bounded by the earthworks discovered by Puleston and Callender in 1967. Our own work
(Webster et al. 2004, 2007, 2008) shows that this earthwork did not serve as any kind of
boundary for Late Classic Tikal. There is nothing like 47 mi2 of mapped settlement at
Tikal, although unmapped settlement extends over a very much larger area.6
We believe the landscape of the Tikal polity was a fragmented and negotiated space
whose boundaries were dynamic, and that distinctions between residential areas and
agrarian areas were not clearly defined. This is not to imply complex urban agriculture or
an exquisite garden city pattern such as attributed to other Maya centers, just a lot of open
space at Tikal, much of which would have been productive, especially given intensified
labor investment. We have long championed this perspective, which Scarborough and
Grazioso Sierra (this volume) call low density urbanism.
While Tikal is well known for its monumental architecture, extensive outlying
settlement, and defensive earthworks, it also resembles many other Maya polities in that
there is limited evidence of extensive landesque agricultural intensification. Many
archaeologists find this surprising. Some suggest that the evidence is present, just not yet
identified, but as Brookfield (1972) and others have demonstrated, the archaeological
evidence for intensification can be misleading. Terraces, for example, are products of an
unfolding process of intensification that includes fallow shortening, weeding, mulching,
and double cropping, all of which are relatively invisible to archaeologists. Although
landesque features have been identified by studies described in this volume, evidence of
extensive agricultural landscape modifications remain largely unidentified. By evaluating
the landscape as a whole, we can begin to identify potential changes and opportunities for
intensification

Tikals Soils and Topography

Buy land theyre not making any more of it. So goes the hoary adage of Will Rogers
(or maybe it was Mark Twain no one knows for sure). Whoever said it viewed land as a
fixed resource, but this is not really true. Within limits new agricultural land can be
created and lost through both natural and human processes. More to the point, particular
spatial segments of the landscape (territory) are less important to farmers than the thin
skin of soil that constitutes (along with water and labor) their principal natural capital. The
soilscape is much more dynamic than land. It can be augmented and enriched by
fertilization, it can lose its nutrients or texture (and sometimes regain them), and it can be
transported by wind, water, and gravity. Farmers can mark out sections of the landscape to
which they lay claim, but soil processes might be beyond their control or even
unobserved. Having access to good land is not the same as having access to good soil over
the long run.
The most comprehensive and useful study of Tikal soils is that of Olson (1977, 1981),
who sampled, classified, and mapped soils in the 16 km2 Carr and Hazard map. Among
his various conclusions are several central to analysis of the agrarian landscape of Tikal
(Olson (1981: 261262):
1. There is an important distinction between upland and lowland soil varieties.
 Upland soils (or Mollisols) are naturally fertile. Lowland (often wetland) soils
would have presented substantial challenges to Maya farmers.
2. Though fertile, upland soils are extremely vulnerable to erosion and damage. The
Maya would have been able to manage the visible aspects of soil, but unable
to manage the invisible or less obvious characteristics, primarily erosion and
declines in soil fertility.
3. The Tikal landscape was probably stable from 40,000 years ago until the first
settlers began clearing the forest (roughly 2,000 years before the present),
thereby developing a fertile soil base.7 He believed that modern soils have not
recovered from the Maya occupation even after more than 1,000 years of
abandonment and noted farming in Petn is still challenging today with
pesticides, erosion control, and fertilizers.
Newer studies indicate soil distributions similar to those noted by Olson, as do test pit
profiles from our settlement work (Burnett 2009; Burnett et al. 2012; Dunning et al. this
volume; Straight 2012). Tikals upland soils are well drained and fertile, but fragile and
susceptible to erosion. Buried isotopic evidence of maize production is found in footslope
and toeslope soils near bajos and particularly around the regional center of Ramonal. All
these characteristics reinforce the emphasis placed on upland soils by Maya farmers and
begin to illustrate some of the niche construction and inheritance issues they encountered.
Our current, if as yet oversimplified interpretation is that early farmers interrupted a
relatively stable soil system just as Olson thought, relying mainly on upland soils for
agricultural production (see Lentz et al., Chapter 8 this volume). Farmers cleared forest in
sloping areas and hillsides eroded, depositing deeper pockets of fertile soils along
footslopes and toeslopes, especially around the margins of upland depressions and the
much larger, low-lying bajos. Our shallow hillside profiles suggest that many upland soils
have not yet recovered from the effects of Maya land use.8 Ultimately the fertile soil base
not only shifted, but shrank, leaving pockets of soil unevenly distributed throughout the
region. In addition to these challenges, Maya farmers had to contend with diminishing
returns and a decline of fertility in remaining upland soils.
Several interacting key processes affected Tikals soils over the long run:
Soil diminishment: This process had both pedological and spatial dimensions. Upland
soil mass and quality were diminished by erosion, loss of nutrients, and reduced
capacity to hold moisture. This large-scale process affected uplands in general. The
spatial extent of good soils also diminished.
Soil transport: Some soils were spatially displaced, particularly by erosion, from
uplands down to the toeslopes or footslopes of large bajos and smaller upland
depressions. Transport stimulated spatial diminishment in the sense that productive
upland soils became more limited in area.
Soil concentration: Displaced soils concentrated in spatially restricted parts of the
landscape, particularly along margins of large low-lying bajos. Such erosion buried
old topsoils, but probably sufficiently slowly to be unnoticed on an intragenerational
level.
 Soil enhancement: Some soils were enhanced in qualities attractive to farmers, either
by purposeful human action or otherwise. Transport and concentration along bajo
margins created deeper soils that were more nutrient-rich, stable, and moisture-
retentive. Soils in upland house-lot gardens were probably also improved through
direct human effort such as mulching and fertilization. Other possibly enhanced soils
were those associated with landesque features such as terraced and irrigated fields at
Tikal.
The soil base was not completely spent by Late or Terminal Classic times; nor were
soils destroyed by overuse. Rather, a pattern of environmental transformation observed in
other Maya regions played out at Tikal. Ultimately these transformations increased the
circumscription of productive soils and the likelihood of crop failure due to extreme
seasonal shifts in climate or productivity declines. They also created new constructed
niches and inheritance issues. Before erosion and transport became marked, most upland
soil was either good or at least acceptable for farming. As soils were eroded and
redeposited they became more spatially circumscribed, resulting in a much starker farmer
perception of good and inferior lands.
In order to map out the size and distribution of these key processes affecting Tikal we
utilized a 452 km2 zone within a 12 km radius of the Central Acropolis. Within it we
classified the landscape into five zones with respect to productivity and erodibility. All of
our own population figures that follow refer to this zone.
Several key patterns emerge from our simple classification of land zones (Table 10.1).
The first and most important is revealed by our various land zones. While bajo and largely
unusable sections of the landscape are prominent, their distribution only represents 15
percent of the total hectares in the region, much less than a simple upland (33 percent)
characterization. Conversely, upland areas shrink to 56 percent of the region with 11
percent in areas that are highly susceptible to erosion. These areas, with slopes greater
than 15 percent, would erode rapidly if cleared. Land zone 2, which bridges lowlands and
uplands, makes up 30 percent of the landscape. While its soils cannot be considered as
fertile as land zone 3, they probably represent the most sensitive areas associated with
niche construction and inheritance. These margins, footslopes, and toeslopes today exhibit
evidence of erosion from past land use, but they also have the clearest isotopic signatures
that indicate probable maize production. Because of their slope and topographic position
they represent some of the more stable resource pockets during the Classic period,
especially as upland soils in land zone 3 began to erode and decline in fertility.
Table 10.1. Tikal landscape (slope + topographic position)

Zone Hectares Percentage Topographic Description

of
Landscape

1 6,695 15% Bajo areas with standing water during the

rainy season, typically < 1% slope

2 13,462 30% Palm bajos, bajo margins, toeslopes, and

 footslopes with a slope of 13%

3 20,433 45% Upland areas with well-drained soil not

sloping more than 8%

4 3,964 9% Steeply sloping upland areas that are

highly susceptible to erosion, 815%
slope

5 684 2% Severely sloping uplands > 15%

Total 45,238

These landscape elements offer what we argue are the key physical landscape scenarios
or niche inheritance issues at play. Initially, productive land had rather even spatial
distributions in the Tikal region. As a result of human-induced changes or niche
construction, by at least AD 250 high quality soil was becoming concentrated in distinct
patches. Patches were both differentially productive and differentially risky, probably
causing differences in the perception of productive land quality and greater variation in
household well-being, ultimately initiating competing economic interests. Eroded soils
accumulated in bajo margins and around their edges, creating new, distinctive ecological
niches, primarily associated with zone 2. These changes opened up new niches, but they
were also part of a larger process whereby high quality soil resources were fragmented and
distributed in patches that did not correspond to earlier perceptions of the distribution of
high quality land. Through time these physical shifts influenced the availability of
resources and the distribution of risk from year to year. For example, during Preclassic
times, when productive land was widely distributed, the effects of severe climactic events
(e.g., droughts or hurricanes) or social changes (e.g. influxes of migrants) would have
been broadly shared and buffered across the landscape. Households would have adapted to
them through spatial relocation and adjustment of agricultural strategies. But by Classic
times droughts would have differentially affected varied resource patches. Moreover,
Classic period households were less mobile and more directly tied to particular pieces of
the landscape. Beginning in the Classic period, cultural niches (or land-use traditions)
were probably as important as ecological niches in creating the potential for internal
competition and stress.

Tikal Settlement and Population

Our competition analysis requires a reasonable grasp of Tikals settlement and population
dynamics, but both are very poorly understood. So dominant were Pulestons Sustaining
Area data that, with the exception of Anabel Fords Tikal-Yaxha transect (Ford 1986), no
large-scale settlement survey was done at Tikal for almost forty years.9 Such neglect is
unfortunate because the region has been well protected as a national park so that small
sites are well preserved and accessible on the surface (visibility aside). Nor have ancient
settlement remains been as heavily affected by erosion or hydrological processes as at
Copn. If Puleston could revisit his survey landscape today he would find it almost
unchanged. In sharp contrast, Webster observed major and disorienting transformations of
the Copn archaeological landscape within a decade.

Settlement
Puleston surveyed settlement on four 500 m wide survey transects that radiated out about
1011 km from Tikals monumental core. Total coverage was not quite 25 km2, of which
Puleston calculated that 10.7 km2 was uninhabitable bajo. Some 2,192 structures were
recorded in his survey zones. From this sample, plus the previously mapped 16 km2 zone
of Central Tikal, Puleston extrapolated a total of 13,948 structures for the site of Tikal,
by which he meant the 120 km2 region enclosed by earthworks and bajos (Puleston 1973:
229). Fry (1969, 2003) concurrently excavated 97 test excavations in Pulestons north and
south transects, each one near a structure, and acquired two artifact collections from the
outlying minor centers of Jimbal and Navajuelal. His sample amounts to about 4.5 percent
of all structures located by the Sustaining Area Project, and about 0.7 percent of the
projected total for the sites landscape. Haviland, Becker, and others carried out similar
tests, along with more ambitious household excavations, in sites in the zone mapped by
Carr and Hazard (see Haviland 2003: 113 for a list). As a general rule, Tikals Late Classic
household remains are rather evenly distributed within fragmented clusters except in
uninhabitable areas such as the Santa Fe Bajo and have respectable amounts of land
around them that could have been used for orchards or gardens. Moderately sized
complexes such as Ramonal (which includes a ball court) are rather evenly distributed
(except to the southwest) at distances between 4 and 8 km from the Central Acropolis.
Our own 20032006 research remapped the earthwork and found several previously
unknown sections of it. We concluded that it was not an effective fortification, that it was
probably incomplete, and that it was much longer and differently laid out than expected.
Our most important conclusion is that the earthwork did not function in any meaningful
sense as a boundary or territorial marker for the Late Classic Tikal polity in the eighth
century. More importantly, we did settlement surveys over three block segments and a
corridor along the earthwork, amounting to about 7 km2 in all. Puleston (1973: 227)
estimated structure densities on the order of 49/ km2 in his intersite areas. We found
more about 67/km2 probably because our eastern survey block encroached on the large
satellite centers of Ramonal and Chalpate, where settlement is virtually continuous. As
matters now stand, only about 48 km2 of the Tikal landscape within our 12 km radius
(including the 16 km2 zone area mapped by Carr and Hazard) has reasonably continuous
survey coverage. Our residential test-pit sample (Straight 2012), added to Frys, amounts
to only 159 such excavations beyond the Tikal epicenter. The scale of both rural survey
and test pitting at Tikal is currently woefully inadequate for population simulations
compared to similar data from Copn (Webster et al. 2000) and a few other Maya polities.

Population
Unfortunately almost all systematic estimates of Tikals population since Pulestons time
focus on the now-suspect earthwork/bajo territory. Most also refer to the presumed eighth
century demographic peak of the polity (AD 700) for which the best settlement data exist.
We thus have an improperly constrained spatial frame and an overly narrow diachronic
perspective. The two most influential and durable eighth century population estimates are
those of Haviland (2008) and Culbert et al. (1990). Haviland thinks there were about
45,000 people in the 120 km2 zone at AD 700, while Culbert et al. prefer the higher figure
of 62,000 people. These estimates translate into respective overall densities of 375 and
517 people/km2 (Webster et al 2008). If we use Pulestons survey figures and discount 43
percent of the landscape as uninhabitable bajo, these upland densities rise to 599 and 771.
We find these estimates astoundingly high and unrealistic for a landscape already
cultivated for 2,000 years or more, especially given the apparent lack of bajo cultivation
and upland terracing.
Even more obscure is Tikals population history. Our own household excavations at 52
(mostly small) rural sites produced overwhelmingly Late Classic material (Classic Ik and
Imix phase AD 550825). Late Preclassic and Early Classic ceramics were extremely
sparse. We take this to mean that preLate Classic populations were heavily concentrated
in and around the Late Classic site core or at outlying minor centers such as Ramonal.
Settlement expansion over the wider landscape thus seems quite late.10 Similar explosive
expansion is seen at Copn and is one of the most intriguing and unsolved issues of Maya
archaeology.
Both Haviland and Culbert recognize that there is scant evidence for reliable estimates
of Early Classic and Preclassic populations. Culbert et al. nevertheless suggest that the
Manik Phase (Early Classic ca. AD 550) populations of Central Tikal and its sustaining
area were respectively 33 percent and 50 percent of their Late Classic counterparts. Even
such proportions result in Early Classic overall densities in the 170235 people/km2 range,
or 272376 for upland densities.
Robert Griffin (2012) produced a lengthy simulation of long-term productive capacity
around three sites in the San Bartolo region using sophisticated remote sensing data and
modeling of soils, vegetation, and agrarian production strategies. San Bartolo is about 50
km northeast of Tikal, and many of its environmental features are very similar. As at Tikal
there are few signs of agricultural terracing or bajo cultivation. Griffin found that various
fallow regimes of maize-based swidden agriculture could plausibly support reasonably
dense populations for considerable periods, although locally there must have been some
contribution from bajo cultivation. We hope to be able to apply some of Griffins methods
directly at Tikal during our future research, but for the present we accept his overall
density range of 117150 people/ km2 for AD 800 (Griffin 2012: 235). Applied to Tikal
this range yields a Late Classic population range of 52,88467,800 for our estimated
hinterland. These figures compare well with Havilands and Culberts. They preserve
long-standing conceptions of the size of Tikals core population, but they apply to a region
almost four times the area of the original earthwork-defined boundary. Upland densities
using Griffins model are thus reduced to 204263 people/km2. Such densities seem much
more reasonable than those of Haviland and Culbert, especially if they were maintained
for a very short time at the demographic peak of the polity.11 According to Griffins
model, short-fallow upland swidden supplemented by some bajo production and house-lot
gardens supported these densities. A core Late Classic Tikal population in the 53,000
67,000 range provides plenty of labor for civic/elite construction and goes far to
eliminating the enduring problem of how Maya agriculture supported so many people. The
answer might simply be that there were far fewer people than we think, and they were not
on the landscape in great densities for very long.12
This takes us back to the thorny problem of Tikals earlier population levels. Only two
points are clear: (1) There was huge population growth after AD 550600, and (2) Early
Classic and Preclassic populations were heavily concentrated, especially near Central
Tikal. A reasonable seat-of-the pants estimate for the whole kingdom in AD 400, just after
the Teotihuacan entrada, would be 5,00010,000 people. For the Late Preclassic 2,000
3,000 people seems about right. Although some archaeologists will find these numbers
heretical, such reduced demographic scale is no disrespect to the Maya quite the
opposite.13 They were clever and industrious enough to accomplish all they did with far
fewer people than we imagine.
The overall demographic picture is one of very slow growth beginning with the first
farmers sometime before 1000 BC, a slight uptick in numbers in the Late Preclassic and
Early Classic, then rapid increases after AD 550600, a pattern evident at other sites such
as Copn that has been difficult to explain. Recently Kennett et al. (2012) have presented a
detailed climatic reconstruction based on speleothem analysis from southern Belize. They
detect comparatively high rainfall during the Early Classic. There followed a drying
interval after AD 640. If these data are widely valid, regional populations might have been
responding to intervals of differential drought risk. Unpredictable drought has been a
perennial problem for traditional Maya maize farmers. The Yucatecan villages studied by
Kramer (2005: 69) report crop failures due to drought in 46 years during every 10.14
Although human niche construction, traditions of cultural niche inheritance, and
probably some competition all originated centuries earlier, the interval of AD 600800
was the crucible of internal discord. We suspect that as populations rose, sometime in the
seventh century there occurred a critical inflection point when the Tikal agrarian system
changed from land rich and labor poor to land poor and labor rich. That Tikals core
architecture burgeoned during just these centuries is thus no surprise. As Winterhalder and
Puleston point out, it is precisely when the landscape is demographically saturated that
labor is cheapest and most abundant.

Management
Maya kings and lords managed many activities, including ritual, construction, war,
diplomacy, and alliance. Some archaeologists think they also managed water in ways that
provided political leverage, and hydraulic features are well documented in Central Tikal
(e.g., Scarborough et al. 2012, Scarborough and Grazioso Sierra this volume).15 Whether
they exerted any managerial control over food production is unknown. Lentz et al. (this
volume) think agroforestry resources were managed by a combination of royal
administrators, lineage heads and local agriculturalists. Many years ago Milton Altschuler
remarked that

Mayan agriculture practices, by their very nature, did not involve the necessary
 development of central control for their ordinary operation. Such administration as
was needed was probably done on a kinship basis, either through the family or
lineage heads. (Altschuler 1958: 194)

Altschuler conveys the prevailing view of his time that the Classic Maya practiced
some form of swidden that could be easily managed by households, as among the Contact
period Maya. He could not take into account arboriculture, terracing, drained fields, or
other more intensive kinds of agriculture. Nevertheless, his remark does capture a basic
truth: So long as good land was plentiful and rain fell adequately, Maya farmers could
manage their own domestic economies perfectly well. They knew how to acquire and
maintain the tools and raw materials they needed. They could decide what to plant, and
how much, how to recruit and deploy labor, how to negotiate with other households, and
how to move and establish new ones. Drought or storms sometimes caused bad years, but
marginal household surpluses were adequate and could be stockpiled or exchanged to tide
people over situational shortfalls. General productive capacity and material well-being
early on did not vary greatly from one household to the next; nor were there strong social
distinctions based on differential access to agrarian capital land and labor. Shifting
cultivation was not locked into permanent fields, but instead allowed farmers to adjust
flexibly to dynamic soil conditions.
By at least EarlyLate Classic times Tikals kings and lords somehow managed their
income derived from the subsistence exchequers of their polities, as Winterhalder and
Puleston put it. But did they meddle by telling farmers what to grow, where to grow it, and
what agricultural strategies to use?16 Did they promote landesque improvements such as
terracing? Some archaeologists think they did so at Caracol and elsewhere (Chase and
Chase 1998), but others think development of such features was perfectly within the
capacities of individual households or household clusters (Murtha 2002). In any case,
while modest and scattered evidence has been presented for landesque capital at Tikal, the
modifications do not appear to be on the same scale as at Caracol.
The Classic Maya generally linked agricultural, plant, and earthy metaphors to courtly
governance (Jackson 2013). Whether such presentation reflects any pragmatic concerns is
unknown. Our own opinion is that there was little or no direct, top-down management of
household production. What kings and lords did have to manage eventually, however, was
increasingly fractious squabbling over dynamic agrarian soil capital and its allocation
niche construction and inheritance. They also had to manage labor for private and public
elite projects (although labor costs were much lower for farmers than usually envisioned).
All this said, we come up against our dilemma: We do not know how these exactions or
transfers were structured. One possibility is direct household extraction each household
was required to turn over part of its production and available labor to the royal/elite sector
at a fixed rate. A second possibility, more in line with general Mesoamerican patterns, is
that kings and lords possessed estates that were worked by commoner (and perhaps
client or slave) labor that is, kings and elites were landlords. Winterhalder and Puleston
argue that several strategies of taxation were variably productive and that these strategies
ideally shifted in ancient complex societies according to demographic and agrarian
conditions.
 How food transfers worked is additionally obscured by our ignorance about how any of
Tikals interest groups was organized above the nuclear/extended family level. Most
Mayanists would probably agree that Classic kings and elites were organized into
patrilineages, as among the Postclassic Maya. Both royal and subroyal elite lineages
plausibly formed the cores of noble houses (Gillespie 2001), part of whose patrimony
were land and attached labor. Most would also probably agree that commoners had some
kind of corporate lineage organization, probably patrilineal, above the level of the
household, as attested both ethnographically and ethnohistorically. Convincing as this last
conclusion is, remember that in some other ancient complex societies lineage/clan
organization was retained by elites but broke down to the nuclear family level among
commoners Hawaii is a case in point (see Webster 1998). All this matters because
organization affects the management of resource flow, how claims to access were made
and enforced, and how they were eventually contested. If, for example, elites simply
extracted a per capita tax on household labor and production, the system was reasonably
flexible and expandable. If the royal establishment grew, then so did demand on individual
households, unless they themselves also became more numerous. If, on the other hand,
kings and nobles had estates, enlargement of their social components meant that more land
(and labor) somehow had to be acquired and allocated.
Whether elites managed production, collection, or simply squabbles over agrarian
assets, their effectiveness depended in part upon how politically centralized the Tikal
polity was. Winterhalder and Puleston (2012) modeled the strategies and efficiencies of
revenue extraction in ancient agrarian states from the perspective of highly centralized
exchequers and bureaucracies. Tikal by the eighth century was probably one of the most
centralized of Maya polities, and it is tempting to apply the same perspective. An
alternative is diffuse elite management, in which a variety of high-ranked people,
including kings, made independent management decisions (i.e., the polity was less
statelike). We suspect that Tikals Late Classic contemporary Copn functioned this
way, but for Tikal itself we really have no idea.
Complicating this load of overlapping claims to agrarian resources are the details of
what we generally call inheritance or reallocation of agrarian cultural niches. If kings were
polygynous (as seems likely), did they partition their resource rights to estates or
otherwise among their offspring and lesser kin? Did they spin off landless lordlings by
primogeniture or other sorts of restrictions? Did kings provide dowries in land as part of
the complex spouse exchange so crucial to alliances? Did they reward successful warriors
or courtiers with rights to land and labor? How did growing peasant families find land for
their offspring? Did commoner wives convey rights in land to their husbands families?
All of these practices would have contributed heavily to increased culturally determined
patchiness and internal discord, but we know nothing about them.

Competition, Status Rivalry, and Stress

Winterhalders and Pulestons abstract simulation revealed several highly punctuated
temporal benchmarks when shifts in Malthusian well-being occurred and competitive
processes emerged or intensified (Griffin detected similar thresholds in his model,
depending upon local soil and topographic patterns). At real-world Tikal we can only
roughly piece together key physical landscape and demographic changes that created
internal management problems, stresses, and eventual competition. Although our main
concern is food, we are mindful that certain other essential household resources, notably
fuel, fiber, and building materials, all became more patchy, restricted, and expensive,
and the objects of competition.
During Preclassic times the principal forms of agrarian capital soil and water were
rather evenly distributed and population densities were low, probably in the range of 10
12 people/km2 in the uplands. Productivity was quite predictable, household well-being
reasonably uniform, and vulnerability to uncontrollable risks widely shared. Long-fallow
cultivation allowed farmers to adjust to the dynamics of soil changes through mobility.
Small populations were maximally buffered by adequate marginal surpluses against the
effects of severe climactic events such as droughts. Local populations adapted to natural
and human stresses through spatial relocation and adjustment of agricultural strategies.
Because land use was not highly fixed, there were probably flexible and comparatively
noncontentious patterns of niche inheritance or allocation.17 By at least Late Preclassic
times even light populations caused niche construction as soils were eroded from steeper
uplands and redeposited in bajos and around their margins.
At the very end of the Preclassic, royal dynasties were established at Tikal, and for the
first time political economy began to make significant demands on farmers, although
surpluses were produced so efficiently that these demands were light. If there were royal
estates, they were first created at this time, restricting choices of farmers and creating new
forms of niche inheritance. Farmers began to take on the identities of peasants. Labor
demands by elites increased, but even a few thousand farmers were able to meet them
without undue pressure on their household economies.
By AD 600 the population had risen to perhaps 10,00015,000 people, with densities of
roughly 4060 people/km2. Fallow periods were systematically reduced as the landscape
filled up and as uplands were degraded. Swidden farmers accustomed to using uplands
had to work harder up to the point that their lands became either too restricted or too
unproductive. Farmers confined to uplands became poorer and had reduced capacity both
to feed their households and to create their marginal surpluses. Palm-bajo margins (land
zone 2), partly created by unintentional niche construction, were probably then, as they are
today, favorite zones for cultivation. People who had long claimed traditional rights to
such lands were further advantaged by soil transport and enrichment. They enjoyed higher
well-being than upland farmers because their production was higher, more predictable,
and less vulnerable to risk. This component of farmers, however, cultivated only a small
portion of the landscape that could not absorb the influx of many increasingly poor
neighbors. These changes affected farming households in a very patchwork fashion
depending upon the nature of traditional claims to sectors of the landscape. Variance in
well-being became more marked (Wood 1998), and overall household resilience might
have been lowered if a drying trend set in after about AD 640.
During the seventh and eighth centuries the elite component of Tikals population
peaked in size and political dominance even as the farming population reached
unprecedented size. Newly powerful noble consumers required support, and perhaps
estates, but had no long history of claims to such lands. We do not know how they were
provided for, but any estates they acquired had to be carved out of the lands already
claimed by farmers or by earlier elites. One possible solution early on was
political/territorial expansion that provided new resources, as Webster long ago argued
(1975). By the seventh and eighth centuries, if not earlier, such expansion was probably
not a viable solution, despite suggestions by some archaeologists that Late Classic warfare
was motivated by the prospect of territorial gain. Other solutions were possible. Royal and
elite factions might have simply expropriated lands from farmers, who were thus demoted
to some form of landless clientage (i.e., they were in no sense smallholders as we use the
term today). Another is that elites competed among themselves over estates and
commoner labor on an intrapolity basis a potent expression of status rivalry. Elites could
have directly intervened in agrarian production, dictating what people planted and how,
especially if they owned estates, but we have no evidence they did so. They also might
have instituted various forms of landesque investments, such as the Perdido Reservoir
irrigation system at Tikal.18
One way or another, the elites own impressive establishments had to be funded.
However this was accomplished, it is reasonable to assume that they took steps to insulate
themselves from shortfalls in ways that caused increasing stress and resentment. Elite and
commoner well-being further diverged. Reallocating agrarian capital within or between
generations (cultural niche inheritance) became more complicated and contentious. The
management issue from the elite perspective was to collect resources from increasingly
stressed producers whose marginal surpluses were shrinking or even disappearing, and to
protect their prerogatives from one another.
How farmers reacted to these pressures depended on how they were organized, as
discussed earlier. The issue is what forms of aggregate agency operated. If farmers had
little formal organization above the household level (nuclear or extended families), then
they were politically marginalized. Some would simply have died, others were reduced to
a bare living, some defected to other regions, and still others became clients of richer
households all outcomes abundantly documented in the ethnographic record of complex
agrarian societies. If farmers were organized into large corporate lineages attached to
kings and nobles (Sanders 1989; Webster 2002), they were members of potent political
factions. They exerted great pressure on their elite senior kin to intercede on their behalf.
In this case, leadership of competitive maneuvering was deflected to at least the subroyal
nobles of the kingdom, who were already natural competitors with each other and with the
royal establishment for other reasons. The management challenge from the elite
perspective would be to keep their own kin factions advantaged at the probable expense
other similar groups more intense status rivalry. While kings and elites might never have
managed the year-by-year production decisions of households or instituted landesque
improvements, they certainly had to try to manage ever-more-contentious claims to
resources from people of all ranks. And it is worth remarking here that making plausible
claims to resources is not the same as enforcing those claims. At some point force
overrides custom.
With demographic saturation there occurred what Winterhalder and Puleston call
competitive inefficiencies essentially fiscal mismanagement that amplifies rather than
solves problems. Winterhalder and Puleston seem to think of such mismanagement as a
destructive syndrome of state bureaucracies. We doubt that Tikal was administered in such
centralized, statelike ways. Instead, by the eighth century there were numerous foci of
royal and elite interests and administration, and any such inefficiencies probably derived
more from competition than cooperation among them a potent arena for status rivalry
and confused and contested cultural inheritance. Agrarian disarray stimulated not only
social dissonance at the resource level, but also cultural dissonance that ate away at the
ideological underpinnings of Maya royal authority, devoted as it was to maintaining
cosmic order. Although we will never discern them directly, the prophetic Maya voices so
prominent in historic times probably operated at the end of the Classic as well, implicating
kings and kingship itself in apocalyptic rhetoric (Landes 2012).
Whether or not internal violence was part of this Malthusian process at Tikal is
unknown. Late Classic kings certainly gloried in military prestige, although their
conquests probably had little economic effect. To the extent that kings had arisen by
arrogating to themselves ritual responsibility for cosmic order and the well-being of their
subjects, they became devalued, as some ambitious elite competitors were doubtless only
too glad to point out (this is not, we hasten to add, an ideological explanation for the
collapse). Tikals royal dynasty came to an abrupt end in the late ninth century. Severe
droughts, if they had anything to do with this collapse, were unusually disruptive because
of a long legacy of deleterious niche construction, and contentious niche inheritance and
status rivalry. Agrarian conditions at Tikal had entered a high-risk, zero-sum stage, which
for ordinary farmers was encapsulated by this more recent metaphor:

There are districts in which the position of the rural population is that of a man
standing permanently up to the neck in water, so that even a ripple is sufficient to
drown him.19

Summary
We began this chapter by championing the ancient Maya, at Tikal and elsewhere, as
effective copers. In their influential overview of bad year economics, Paul Halstead and
John OShea (1989) emphasize four main coping strategies: mobility, diversity, storage,
and exchange. At Tikal, ironically, all of these options were most feasible in Preclassic and
Early Classic times, just when they were least needed. By the Late Classic long-distance
mobility was constrained externally by the social circumscription of neighboring high-
density polities, and internally by restrictions of resource location and inherited rights to
soil and land. Alterations of the soilscape inhibited diversity. Food storage, never effective
for much more than a year or so in Tikals tropical environment, was ever-more
problematical as marginal surpluses shrank, and exchanges of cheap, bulk goods over any
distance were always inefficient, even when surpluses were available. If drought was a
problem, it added to these stresses.
Much less obtrusive in Halsteads and OSheas analysis is a fifth strategy
competition. This strategy became more intense and unmanageable in the two centuries
before the collapse. There are two methods of curing the mischiefs of faction: the one, by
removing its causes; the other, by controlling its effects.20 So noted James Madison at a
critical juncture of our own political system. For the Classic Maya at Tikal and elsewhere
the causes inherited niches and dwindling resources could not be removed; nor were
Maya kings or institutions sufficiently powerful and centralized to control their effects.
Many archaeologists attribute to the Maya a high capacity for resilience. If that means
demographic and political durability in a challenging tropical environment affected by
many uncontrollable risks, then the Tikal Maya were indeed resilient for roughly twenty-
five hundred years.21 From our perspective, however, most of the interval of this resilience
was characterized by low population densities and considerable latitude to adjust to
changing conditions. Whatever caused the surge in Tikals population after about AD 550
600, the polity was propelled into an unsustainable demographic and sociopolitical state
characterized not only by human niche construction, but by irresolvable levels of internal
niche inheritance. It was this stage of development at Tikal, Copn, and elsewhere that
lacked resilience. Even as Tikal reached its apogee of artistic and architectural brilliance,
its sociopolitical system was undermined by unprecedented real-world messiness
expressed as competition among farmers and status rivalry among elites. What lacked
resilience at Tikal and most other Maya centers were the final Late Classic expressions of
Maya civilization the very ones that impress us most.
Finally, we would like to resurrect a good old reductionist remark from the very dawn
of Western historical political commentary. Plato attributes to Socrates the following
observation: No man is qualified to become a statesman who is entirely ignorant of the
problems of wheat. At Tikal the politics of maize stimulated competition among
commoners and elites alike, aggravated by long traditions of niche construction and niche
inheritance.
There is a lesson here about our own anthropological status rivalries as well. Mayanists
are fortunate in that we can often perceive individual kings, queens, and sometimes elites
as actors. At Tikal we can begin to perceive the environmental conditions and historical
processes that plausibly conditioned the actions of lesser, anonymous individuals or
factions as well, whose interests sometimes coincided and sometimes did not, and
determined who cooperated or competed with one another as conditions dictated.
Presumably both agency theorists and cultural ecologists will find something congenial in
this interpretation. Perhaps our views are not as divergent as our rhetoric often seems to
imply.
Notes
1 The work by Payson Sheets and his colleagues at Cern provides very provisional
insights into such patterns (Sheets and Dixon 2011), as do stone field boundaries in other
parts of the Maya Lowlands. Often such features delimit house lots or orchards rather than
outlying field systems.
2 For discussions of status rivalry see Webster (1998) and Pohl and Pohl (1994).
3 Cedric Puleston is, fittingly enough, Dennis Pulestons son.
4 Our focus here is on food, fiber, fuel, and other basic necessities produced by farmers;
we ignore the possibility that rulers or other elites possessed estates that produced
specialized or commercial crops, as we know occurred among the Conquest period
northern Maya. Nor are we concerned with seasonally limited resources such as water,
which some archaeologists believe elites used to leverage political power (Scarborough et
al 2012; Scarborough and Grazioso Sierra this volume). That being said, addition of such
variables, themselves subject to niche construction and cultural inheritance, would
exacerbate the competitive processes we envision.
5 One might say we cherish for far too long inherited intellectual niches such as the
earthwork/hinterland model.
6 Our tally does not include sites in Anabel Fords (1986) Tikal/Yaxha transect, or in our
own recent surveys.
7 We would now adjust Olsons chronological estimates. Tropical forest conditions
similar to modern ones date back only to about 9,000 years ago, and early farmers were
present on the landscape some 2,0003,000 years earlier than he imagined.
8 Kirk Straight, who dug our residential test pits, estimates that today A horizons are less
than 25 cm, and often a bit thinner 18 to 19 cm on average (personal communication
2012).
9 Vilma Fialko (2005) did survey work in the region but results have not been widely
published in any detail.
10 During 2005 and 2006 we made a vegetation survey and retrieved 844 soil samples
from about 200 soil profiles in excavations both at or near the earthworks and from the
surrounding landscape. Analysis by Terry and his colleagues of our soil profiles and from
old buried horizons beneath the earthwork embankment shows stable carbon isotope
enrichment consistent with the cultivation of C4 plants probably maize.
11 We think that even 200 people/km2 is still too high, but for our analysis we utilize this
figure, assuming that there is some invisible food input from house-lot gardens, orchards,
and nonmaize staple crops.
12 Zachary Nelson (2005) drew very similar conclusions for eighth century Piedras
Negras.
13 If Tikal had a very large Late Preclassic or Early Classic population, then it was by no
means the dynamic, successful polity usually imagined and had remained demographically
quite stagnant through time. If on the other hand its Late Classic florescence (cultural and
demographic) is real, then there must have been far fewer people in earlier times. One
cannot have it both ways.
14 Admittedly this region, at 1,100 mm of rain per year, is rather drier than Tikal. The
drought problem Kramer reports is not total rainfall, however, but the amount of rain at
the beginning of the growing season. This episodic shortfall is much exacerbated during
years that are generally dry.
15 Long ago Edward Higbee noted that the landscape of southern Yucatan, particularly
the area about Tikal, has a feature that made it especially attractive to primitive man. A
profusion of little ponds, aguadas, dot the countryside. They are seldom more than a few
miles apart, and in between them most of the soils are well drained. This splendid mosaic
of aguadas and arable land made the region suitable for concentrated settlement. To
recognize the significance of aguada distribution around Tikal is to appreciate why it was
so desirable a site for the development of the first great city of the Old Mayan Empire. A
broad hinterland could be occupied! (Higbee 1948: 459). Puleston (1973) also noted this
profusion of water sources during his surveys. Their abundance is one reason why we
doubt that management of dry season drinking water gave rulers much political leverage.
Certainly it would be a puny political lever elsewhere, as along the Usumacinta or at
Tikals nearby lakeside neighbor Yaxha.
16 Some ancient elites dictated inheritance patterns. For example, a long tradition of
state-enforced partible inheritance helped to maintain elite dominance in China (Fairbank
and Goldman 2006: 21). We have no idea whether Maya rulers did anything comparable,
or even meddled much in customary law.
17 Complicating Maya niche inheritance is the association among the living inhabitants
of households and their buried dead, who still retained great social, political, and probably
economic significance (Fitzsimmons 2012). In essence, ancestors embodied (among other
things) political and economic identity and claims. How these affected resource use and
allocation is unknown, but they probably were very important. We cannot address such
symbolic dimensions here, but they would have created additional social tensions and
pretexts for competition.
18 By the eighth century the locations of Tikals mapped residences probably are very
poor guides to where main milpas were located. Exceptions are household gardens and
orchards, which must have been close to residences.
19 Scott (1976: 1) citing Tawney (1966) on famine among Chinese farmers.
20 From The Federalist Papers, No. 10 (1787) by James Madison.
21 Resilience here means the capacity of a system to remain sufficiently flexible to
ensure long-term survival to adapt to what R. M. Adams (1978: 329) characterized as
profound, unexpected, continuing perturbations.
 11 Material Culture of Tikal
Palma J. Buttles, Carmen E. Ramos Hernandez and Fred Valdez Jr.

Introduction
Among the products of past societies that are left to us today are the material remains or
the material aspects of culture. It is our perspective that a common unifying principle
holds that material culture is not culture, but is the product of cultural and social behavior.
The analysis of material remains is, thus, a starting point for understanding prehistoric
cultures and their society. Various material artifacts (ceramics, lithics, etc.) from the
University of Cincinnati Archaeological Project at Tikal (UCAPT) were analyzed to
describe morphology and assess chronology. Each artifact category studied is reported
with a brief description as well as interpretations for particular contexts and the collection
as a whole. While many of the contexts are from within reservoirs at Tikal and areas
outside the ceremonial center, the range of the material culture represented is surprisingly
broad and impressive.
Archaeologists often seek to identify intangible aspects of culture by analysis of the
material remains, including nonportable artifacts. In the Maya area as in many arenas,
material culture has often been used to establish site chronologies and social hierarchies
and to identify exchange networks and technological systems (Adams 1982; Buttles 2002;
Carpenter et al. 2012; Hester and Shafer 1994). Recent analytical approaches include
those focused on the reconstruction of social communities by viewing material culture as
symbols that served to construct, illustrate, and reinforce individual and community
identities and ideologies (Buttles 1992, 2002; Canuto and Yaeger 2000; Hester and Shafer
1994; Meadows 2001). This latter aspect would certainly have been a valuable component
at Tikal.

Aspects of cultural behavior may be elucidated when viewed from a contextual

perspective. Additionally, patterns of material culture consumption and disposal may
be isolated when considering contexts and when viewed in combination, material
culture may reveal specific practices that serve to reinforce individual and
community identities, affiliations, and ideologies. (Buttles 2002)

Represented in the Tikal collection are artifacts that were used to perform a variety of
tasks and activities that encompass both practical and prestige technologies. Such potential
activities have been defined by Hayden (1998). Examples of these tasks and activities
include spinning thread, weaving, and performing other technological processes. Defining
the artifacts that aided in these activities and tasks allows a glimpse into a part of the Maya
world not often addressed in site reports and articles (Buttles 2002). The objectives for this
study required the establishment of a typology based on raw material, technology, and
morphology. Archaeologists through various theoretical and practical methods seek to
understand the relationship between prehistoric cultures and their material remains (e.g.,
Chilton 1999). While the best method or path leading to this information is and may
always be debatable (see Chilton 1999; Dietler and Herbich 1998; Stark 1999) we must
continue to look at material culture from various perspectives.
What is often missing in most assemblages is the multitude of artifacts that were made
of perishable materials. Artifacts derived from plants, animal parts, and cloth must always
be in consideration when discussing the material culture of any prehistoric society. It has
been estimated that up to 95 percent of prehistoric material culture was made of perishable
materials (Drooker 2001: 4). Unfortunately, while Tikal collections may indicate the use
of some of these perishable materials, in most cases the perishables themselves are not
present. We are often left only with the implied use of the perishables, but these
implications are nevertheless important in understanding some of the craft production
activities of the ancient Maya at Tikal.

Artifacts: Types and Descriptive Comments

The artifacts documented and interpreted in this study are from several locations in and
around Tikal. Most artifacts are commented upon here in terms of material type,
morphology, and chronology while a few selected items are highlighted within particular
contexts and in terms of possible function. The extent of excavation at each context
(operation) varied, but what information is available provides valuable insights into the
activities of the ancient Maya at Tikal. Metrics for artifacts are presented in table form and
discussed in the text. It is, however, the form and function of the material culture that hold
significant interest. Specific metric data for each category have been reported elsewhere
(e.g., Lentz et al. 2009, 2011). Each artifact grouping, such as chipped stone or obsidian, is
discussed in terms of a descriptive composite of the entire subassemblage. Interpretations
for the groupings, in terms of function and perhaps representation, are provided for the
entire assemblage.

Obsidian
Though relatively rare in the lowlands as compared to chipped stone (chert), obsidian is
found in many of the Tikal excavations. Nearly all sites in the Maya Lowland regions have
reported the recovery of obsidian, though the quality and quantity do vary. The analysis
undertaken for the Tikal obsidian has followed traditional designations (Trachman 2011).
Although the majority of recovered obsidian is in the form of blade fragments (proximal,
medial, and distal), there are also core fragments and bifaces (Table 11.1). Because of the
somewhat brittle nature of obsidian and bioturbation of the last 1,000 or more years, it is
often difficult to ascertain the extent of human use- wear on the lateral edges of the blade
fragments. There are, however, numerous pieces that exhibit intentional retouch and/or
extensive utilization (Figure 11.1).
Table 11.1. Obsidian artifacts from various excavations

Operation. Number Comment(s) L W Th Period*

 of (ranges)
Artifacts

1 10 Blade .72 .86 .17.36 EC, LC

fragments 2.66 1.53

5 1 Blade 3.01 1.15 .36 LC

fragment

6 82 2 Cores, 4 .65 .62 .181.21 MPLC

green blades, 6.17 2.79
5 complete
blades

7 14 Blade 1.25 .79 .19.40 MPLC

fragments 2.45 1.45

8 16 1 Core, 1 1.81 .85 .16.50 LPLC

green blade 4.61 1.42
fragment

9 4 1 Core 1.24 .88 .23.50 LPLC

2.75 1.72

13 6 1 Biface 1.54 .95 .19.36 LPLC

2.74 1.19

14 196 1 Biface, 1 .74 .29 .12.90 LPLC

biface stem, 3 3.83 1.46
cores, 4
complete
blades, 14
green blades

16 7 Blade 1.36 .77 .23.37 LPLC

fragments 3.59 1.35

131 18 Blade 1.60 .99 .22.36 LPLC

fragments 3.62 1.45

Notes: The length (L), width (W), and thickness (Th) measurements represent the
range measured for blades, smallest to largest.
* Chronological assessment per operation is abbreviated as follows: MP = Middle Preclassic, LP = Late
Preclassic, EC = Early Classic, LC = Late Classic; a range of periods may be indicated, as LPLC indicates Late
Preclassic through Late Classic.
Figure 11.1. Lithic artifacts of the UCAPT: a, obsidian edge altered blade; b, corner
modified obsidian blade; and c, a carved greenstone fragment (jade?). The lower
illustrations are d, ovoid chert scraper and e, chert agricultural tool (original figures
redrawn by Rachel Egan).
One interesting flake resulted from the truncation of an obsidian core as determined by
blade scars on the flake edge. The noteworthy observation is the extent of utilization and
the reworking of a core, to get the most cutting edge possible from the material. Several
obsidian bifaces were recovered from Ops. 13 (Perdido) and 14 (El Jaguar). Excavations
at El Jaguar produced a medial biface fragment of green obsidian in the form of a
proximal fragment or stem, reminiscent ofTeotihuacan material in form. Green obsidian
found from Ops. 6, 8, and 14 attests to access of the occupants to material from Central
Mexico during the Early Classic period. Most blades and several of the biface fragments
are some shade of gray in color and likely derived from the Guatemala Highlands. Many
range in macroscopic observation from gray to striped black and smoky gray, but cannot
be attributed to a specific source without detailed testing.

Chipped Stone
Nearly all contexts produced chipped stone flakes made of chert (sometimes called flint).
Some of the more interesting and telling tools are listed here by operation and
suboperation along with a chronological assessments (see Table 11.2). While detailed
descriptions and metrics have been documented elsewhere (Lentz et al. 2009, 2011), it is
the distribution of formal tools that is of interest in this study. Only formal tools are
considered here from the varied contexts. While expedient tools, such as modified flakes,
were observed, they have been left for other venues of detailed discussion. The formal
tools are discussed in terms of possible functional activities at or near their points of
discovery.
Table 11.2. Chert artifacts from various operations
Operations. Suboperations Morphology/Type Comments L x Period*
W x
Th

2 A Ovate scraper Brown 6.1 EC, LC

chert x
6.1
x
1.2

6 A Resharpened flake Edge 6.1 EC, LC

trimmed x
3.6
x
0.5

B Exhausted core White 6.5 EC, LC

chert x
4.5
x
2.5

B Unifacial core 5.3 EC, LC

x
5.3
x
1.5

C Limestone With flake 6.7 EC, LC

scars x
4.0
x
1.1

C Diamond cross- White 5.2 EC, LC

section tool chert x
2.0
x
1.5

C Rough biface Chalky 4.7 EC, LC

chert x
4.8
x
1.7
 F Ovate biface Pink-red 5.3 EC, LC
chert x
4.6
x
1.6

J Biface distal 5.4 EC, LC

fragment x
2.4
x
0.7

J Eccentric flake Tan chert 7.1 EC, LC

x
6.5
x
1.1

L Long biface 20.3 LP, EC,

x LC
5.7
x
4.1

L Biface medial Dark gray 8.2 LP, EC,

fragment chert x LC
6.1
x
3.4

Q Biface medial Gray chert 5.9 LP, EC,

fragment x LC
5.2
x
2.4

7 A Biface medial Tan chert 5.1 EC, LC

fragment x
4.1
x
1.6

8 A Biface medial 5.2 LC

fragment x
4.7
x
 1.8

D Biface medial 3.3 LC

fragment x
4.8
x
2.1

M Ovate biface 8.3 LC

x
6.1
x
2.6

M Biface medial Impact 10.5 LC

fragment scars x
6.5
x
3.0

13 J Worn chert tool Heavily W = nd

ground 4.9

14 A Biface proximal 4.3 LP, EC,

fragment x LC
2.8
x
1.1

A Biface distal Tip 2.6 LP, EC,

fragment fragment x LC
2.7
x
0.7

A Biface distal 5.1 LP, EC,

fragment x LC
5.4
x
2.1

A Biface (complete) 6.3 LP, EC,

x LC
3.0
x
0.8
A Biface (complete) Dark 7.1 LP, EC,
w/stem x LC
(1.4) 4.0
x
1.1

A Biface (complete) With stem 7.0 LP, EC,

(1.8) x LC
2.2
x
1.4

I Biface (complete) Dark 5.3 LC

w/stem x
(1.8) 3.1
x
0.9

L Biface medial 4.1 LP, EC,

fragment x LC
5.1
x
2.0

L Biface distal Granular 4.6 LP, EC,

fragment chert x LC
4.6
x
2.2

L Biface distal Purple 4.1 LP, EC,

fragment chert x LC
5.5
x
1.1

L Biface distal 5.8 LP, EC,

fragment x LC
5.3
x
2.5

L Biface distal 10.1 LP, EC,

fragment x LC
5.0
x
 3.0

L Triangular cross- 9.5 LP, EC,

section biface x LC
4.2
x
3.0

N Biface medial Burned 2.5 LP, EC,

fragment chert x LC
2.8
x
1.0

15 G Biface medial White/gray 4.4 LP, LC

fragment chert x
4.9
x
2.6

G Utilized blade Utilized 5.0 LP, LC

edges x
1.6
x
0.2

Notes: The length (L), width (W), and thickness (Th) measurements are provided per
formal tool. *Chronological assessment per operation and subop is abbreviated as
follows: LP = Late Preclassic, EC = Early Classic, LC = Late Classic, nd = no date.
Arroyo Corriental (Op. 2) produced a fascinating ovate scraper (Figure 11.1) made of a
high-quality brown chert. The scraper has lots of polish on the ventral surface representing
the effects of extensive utilization. The Palace Reservoir (Op. 6) produced a dozen
significant tools including a fragment of flaked limestone, one extensively used core of a
white (possibly patinated) chert, as well as a unifacial core. Op. 6 excavated the greatest
range of chert artifact forms, especially biface fragments, in addition to the items
mentioned previously. Both were medial and distal fragments including several with
impact scars. One fragment of a possible agricultural implement from the UCAPT Tikal
collection is similar to specimens observed in other areas of the Petn and NW Belize.
This agricultural tool is diamond-shaped in cross section and extensively worn and
polished along the central flake scars (Figure 11.1). Another very interesting artifact is an
eccentric made on a large flake (Meadows 2001). The eccentric can be described as a tool
with a stem and a five pronged distal form (Figure 11.2; Moholy-Nagy 2008).
Figure 11.2. Miscellaneous UCAPT artifacts: a, incised bone awl fragment; b, carved
chert eccentric; c, shell button/ear flare segment (dorsal side); d, shell button/ear flare
segment (ventral side); e, perforated shell pendant. All scale bars = 1 cm.
The Perdido Pocket Bajo (Op. 8) produced several medial biface fragments, one distal
fragment, and a rough ovate biface. One of the medial fragments displayed impact scars at
both ends, implying a secondary use (or recycling) after initial fracturing of the biface. Op.
13 provided a heavily worn or ground biface fragment. Whether the tool was first
fractured and then reused in a capacity that caused the ground smooth surface or the
original activity caused the smoothed wear and polish is uncertain. The flake scars on this
tool were worn in a manner that was barely visible. The Op. 13 chert tool is the only such
artifact in the UCAPT collection indicating an activity not likely replicated at other
locations investigated by the project.
Excavations at El Jaguar (Op. 14) recovered the greatest number of chert tools with
fourteen biface fragments (proximal, medial, and distal), a dart point stem, and three
complete dart point examples. The exact nature of activities requiring the tools represented
from Op. 14 remains uncertain, but obviously there was a need for bifacial points along
with other bifaces. Six of the biface fragments are distal sections reflecting stress activities
placed on the tools that snapped the distal portion, but the specific process is varied and
unfortunately not definable for the current analysis.

Ground and Polished Stone

Two operations produced evidence of ground-and-polished stone in the form of jade items.
Excavations at Corriental Reservoir (Op. 1) unearthed an interesting fragment that is
plano-convex in profile with a lunate cut (Figure 11.1). Op. 14 investigations recovered a
small chip of apple-green jade. Access to this high-quality material is represented at these
two operations, but the complete form of either fragment remains undetermined.

Ground Stone
This artifact category is represented by mano (n = 2) and metate (n = 5) fragments. Op. 6
as excavated by the UCAPT had only metate fragments. Op. 8 and Op. 14 investigations
recovered one mano fragment each. Op. 14 also produced two metate segments. In all
cases, the mano and metate artifacts are relatively small fragments and difficult to assess.
Most fragments are macroscopically familiar as possibly derived from the Guatemala
Highlands. The difficulty in this collection is whether the fragments are local products,
recycled artifacts, or gathered as part of fill for construction activities.

Bone Artifacts
An awl fragment was recovered from excavations at Corriental Reservoir (Op. 1). A
second awl fragment, but with an incised decoration at the proximal end, was from the
Palace Reservoir (Op. 6; Figure 11.2). A bone tube was uncovered in the Perdido
Reservoir (Op. 8) that may have served as a large bead, but possibly as a feather fan
holder or handle. Two other bone artifacts were from Op. 14, an undefined bone fragment
displaying a few incised lines and another bone fragment with blue pigment.

Shell Artifacts
A fascinating shell button was recovered from Op. 1 (Figure 11.2). It is termed a
button from a lack of knowing its true function, but was likely part of an ear ornament
(Moholy-Nagy 2008) and demonstrates a very high level of technological ability. Op. 14
investigations recovered three interesting shell artifacts: a fragment with an incised line,
an olive shell with a modified edge, and a small perforated shell pendant (Figure 11.2).

Ceramic Artifacts
Ceramic artifacts include sherd disks (n = 5), perforated sherds/sherd disks (n = 12), an ear
flare fragment (n = 1), whistle fragments (n = 2), and figurine fragments (n = 21). Of the
five sherd disks, three were recovered from the Palace Reservoir excavations (Op. 6). The
function of sherd disks has been identified as possible gaming items, but specific details
are vague. Perforated sherds/disks were located in significant numbers at Palace Reservoir
and El Jaguar and these probably were used as spindle whorls. This type of artifact is often
associated with weaving activities and may indeed reflect an aspect of household
production.
The Op. 14 excavations at El Jaguar, in addition to sherd disks and perforated sherds,
revealed a ceramic ear flare fragment (Figure 11.3), two ceramic whistle fragments, a
ceramic stamp (Figure 11.3), and ten figurine fragments. Certainly, the occupants at El
Jaguar had access to many artifact types not held at many other locations. Op. 6, the
Palace Reservoir investigations, also interestingly produced numerous (n = 13) ceramic
figurine fragments. Although the question of access to and availability of some of the
special artifacts is not directly connected to the Palace Reservoir and El Jaguar occupants,
they quite likely utilized these items in daily or ritual activities. The fragmentary nature of
the recovered artifacts may be related to excavation contexts, but reflect the activities and
beliefs of the locale occupants.
Figure 11.3. Miscellaneous UCAPT artifacts: a, ceramic ear flare fragment; b, ceramic
stamp; c, polychrome sherd with figure. All scale bars = 1 cm.

Ceramics (Sherds and Vessels)

Ceramic fragments, or sherds, represent the largest artifact category by far with more than
twenty-nine thousand sherds collected from the 2009 and 2010 seasons. The data as
presented here are a reflection of the preservation properties of each individual context.
Sherds were retrieved from every operation investigated and while most were quite
eroded, many were wonderfully preserved. The reservoir contexts as well as the plazuela
El Jaguar were particularly interesting contexts from the ceramic preservation perspective.
The most significant period represented by the entire collection was the Late Classic, in
terms of both quantity and preservation. While sherds dating to the Preclassic and Early
Classic periods were located in various contexts (Table 11.3), none was from a pure or
unmixed context. Almost all lots showed a Late to Terminal Classic component (Table
11.4), thus dating nearly all lots to the Late Classic period.
Table 11.3. Major temporal periods* and associated types in the UCAPT collection

Period Approximate Dates Major Types Observed

Postclassic AD 9001200(?) Augustine Red (?)

Late Classic AD 600900 Cambio Unslipped

Encanto Striated

Tinaja Red

Cameron Incised

Pantano Impressed
 Maquina Brown

Achote Black

Yuhactal Black-on-red

Palmar Orange-polychrome

Zacatel Cream-polychrome

Early Classic AD 250600 Triunfo Striated

Aguila Orange

Balanza Black

Paradero Fluted

Dos Arroyos Orange-polychrome

Late Preclassic 400 BCAD 250 Achiotes Unslipped

Sapote Striated

Sierra Red

Laguna Verde Incised

Flor Cream

Polvero Black

Unnamed Buff

Middle Preclassic 900 BC400 BC Consejo Red

Joventud Red

Guitara Incised

Chunhinta Black

* The major periods reported for this chart follow broad Maya designations and not the defined complexes of
Culbert (1993). The quality of the recovered pottery did not allow for a refining of the periods as documented by
Culbert.

Table 11.4. Quantity of sherds and chronological assessment per operation and
subop

Operation Suboperation Number Periods Represented

of Sherds

1 A 1 nd

1 C 155 Early Classic, Late Classic

1 D 131 Early Classic, Late Classic

1 E 1 Late Classic

1 G 2 nd

1 I 1 nd

1 J 23 Classic

1 K 77 Early Classic, Late Classic(?)

1 L 90 Classic

1 P 12 nd

2 A 196 Early Classic, Late Classic

2 B 28 Early Classic, Late Classic

4 B 22 Late Classic

5 B 1 nd

5 C 28 Late Classic

5 D 15 Late Classic

5 E 1 nd

5 F 19 nd

6 A 461 Early Classic, Late Classic

6 B 268 Early Classic, Late Classic

6 C 675 Early Classic, Late Classic

6 D 90 Early Classic, Late Classic(?)

6 E 614 Late Preclassic, Early Classic,

Late Classic

6 F 85 Early Classic, Late Classic

6 G 49 Late Classic

6 H 9 Early Classic(?)

6 I 320 Middle Preclassic(?), Late

Preclassic, Late Classic

6 J 532 Early Classic, Late Classic

6 K 1,963 Late Classic

6 L 1,311 Late Preclassic, Early Classic(?),

Late Classic

6 M 243 Late Classic

6 N 70 Late Classic

6 O 614 Middle & Late Preclassic, Early

Classic(?), Late Classic

6 P 287 Late Classic

6 Q 944 Late Preclassic, Early Classic,

Late Classic

6 R 40 Late Classic

6 S 219 Late Preclassic, Late Classic

6 T 225 Late Preclassic, Late Classic

6 U 1,769 Late Classic

6 V 584 Late Classic

6 X 73 Late Classic

7 A 314 Early Classic, Late Classic

7 B 539 Middle Preclassic, Early Classic,
Late Classic

7 C 26 Late Classic

7 D 146 Late Classic

7 E 7 nd

7 F 216 Early Classic, Late Classic

7 G 384 Late Classic

7 H 44 Late Preclassic, Late Classic

7 I 365 Late Preclassic, Early Classic,

Late Classic

7 J 743 Late Preclassic, Early Classic,

Late Classic

7 K 437 Late Preclassic, Late Classic

7 L 42 Late Classic

7 N 2 nd

8 A 212 Late Classic

8 B 1,134 Late Preclassic, Early Classic(?),

Late Classic

8 C 105 Classic

8 D 193 Late Classic

8 E 250 Late Classic

8 G 81 Late Classic

8 H 58 Late Classic

8 I 144 Late Preclassic, Early Classic,

Late Classic

8 J 8 nd
8 K 369 Late Preclassic, Late Classic

8 L 140 Late Classic

8 M 67 Late Classic

8 N 587 Late Preclassic, Early Classic,

Late Classic

8 O 48 Late Preclassic, Late Classic

8 P 106 Early Classic(?), Late Classic

8 Q 105 Late Classic

9 A 300 Late Preclassic, Early Classic,

Late Classic

9 B 23 Late Classic

9 F 51 Late Classic

9 G 4 nd

9 H 147 Late Preclassic, Early

Classic(?),Late Classic

10 A 63 nd

10 B 5 nd

11 C 11 Late Preclassic, Late Classic

11 D 2 nd

12 A 56 Late Classic(?)

13 A 41 Late Classic

13 B 102 Late Classic

13 C 32 Late Classic

13 D 4 Late Classic(?)
13 E 79 Preclassic(?), Late Classic

13 F 67 Late Classic

13 G 460 Late Preclassic, Early Classic,

Late Classic

13 H 14 Late Preclassic, Late Classic

13 I 696 Late Classic

14 A 1,596 Late Preclassic, Early Classic,

Late Classic

14 B 283 Late Classic

14 C 12 Late Classic(?)

14 E 38 Late Classic

14 G 85 Late Classic

14 I 140 Late Classic

14 J 94 Late Classic

14 K 61 Early Classic, Late Classic

14 L 2,909 Late Preclassic, Early Classic,

Late Classic

14 M 107 Late Classic

14 N 273 Late Preclassic, Early Classic,

Late Classic

15 C 15 Classic(?)

15 G 198 Late Preclassic, Late Classic

15 H 56 Early Classic, Late Classic,

Postclassic(?)

15 I 6 Early Classic

15 K 4 Classic(?)
16 A 1,272 Late Preclassic, Early Classic,
Late Classic

16 C 87 early Classic, Late Classic

131 A 167 Early Classic, Late Classic

131 B 254 Late Preclassic, Late Classic

131 C 282 Late Classic

Partial vessels:

7-B-6 Late Classic cylinder/bowl represented by two rim sherds and one body
sherd; Palmar Orange-polychrome with figural face and two possible glyphs

7-B-8 Partial cylindrical vessel weathered and fragmented; unnamed cream-

polychrome.

Op. 13 (Perdido) produced four (4) complete vessels from this context:

Vessel #1 Unnamed Red Bowl; rounded base

Diameter = 12.49 cm

Height = 8.5 cm

Thickness = .54 cm

Vessel #2 Unnamed Orange-polychrome; cylinder, flat

base

Diameter = 9.05 cm

Height = 14.03 cm

Thickness = .50 cm

Vessel #3 Unnamed Orange-polychrome; tall bowl, flat

base

Diameter = 15.90 cm

Height = 9.10 cm

Thickness = .68 cm
 Vessel #4 Unnamed Orange-polychrome; tripod dish.

Diameter = undetermined

Height = 7.14 cm

Thickness = .96 cm (at midbody)

Feet (avg.) = 4.58 cm height, 3.93 cm diameter

Note: The period designation of nd = no date.

A few near-complete vessels from Temple Reservoir (Op. 7) were identified. A Late
Classic jar from Op. 7 is represented by two rim sherds and one body sherd. The vessel is
an incomplete flat-base polychrome cylinder. It is typed as a Palmar Orange-polychrome
with a figure face and two possible glyphs (Figure 11.3). The second partial pot is a
cylindrical vessel. It is an unnamed cream-polychrome with no design and very weathered
as well as very fragmented. Four complete pots were recovered from an interment near
Perdido (Op. 13; Figures 11.411.7) and are good examples of Late Classic Maya pottery.

Figure 11.4. Complete vessel from Operation 13: unnamed red bowl with rounded base
(photo by Maurcio Diaz and illustration by Blanca Mijangos).
Figure 11.5. Complete vessel from Operation 13: unnamed orange-polychrome cylinder
(photo by Mauricio Diaz and illustration by Blanca Mijangos).

Figure 11.6. Complete vessel from Operation 13: unnamed orange-polychrome tall bowl
(photo and illustration by Mauricio Diaz).
Figure 11.7. Nearly complete vessel from Operation 13: unnamed orange-polychrome
tripod dish (photo and illustration by Mauricio Diaz).
The major ceramic types identified for each chronological period are presented in Table
11.3. While it is not a complete list of types per period, these are identified with
confidence as described by Culbert (1993) among others (Adams 1971; Valdez 1987). As
indicated previously, much of the material was too small and/or too eroded to identify by
type. Most sherds, however, are clearly of local production given the vast similarity of
paste.
The ceramic sherds and whole vessels are discussed later along with interpretations for
material utilization at Tikal. Rather than a simple discussion for each artifact category, the
approach here is to view several excavation areas (operations) from an assemblage
perspective to understand and interpret ancient Maya activities in the studied zones better.

Analytical Approach
Among the significant interests of the UCAPT was the examination of ancient Maya
adaptation to Tikal environments (Lentz et al. 2009, 2011). Part of the successful
adaptation included landscape modification and utilization. The material culture reviewed
as part of the UCAPT represents ancient Maya behavior at several levels. A brief review is
considered for material culture and the Maya behaviors described here as The Practical,
The Externally Derived, The Decorative and/or Leisure, and The Ritual. While these
aspects can be applied to and are part of most ancient Maya societies, our application here
is specific to material culture clues from the Tikal data. In addition are the considerations
of craft, occupational activities and interpretative (implied) data.
 In the Practical material remains represent the Maya attempts to harness the local
environment. These are artifacts that help secure a successful adaptation. Many of the
bifaces and biface fragments are the very items the Maya were using to cut down trees,
clear fields, modify what became the reservoirs, shape stones for construction, and so
forth. The current collection includes bifaces that have significant impact scars and/or
breakage patterns consistent with hard utilization. Reviewing the general inventory of
stone tools, a clear distinction is seen between the Palace Reservoir and El Jaguar
collections. The Op. 6 collection includes cores, biface fragments, and a significant
diamond cross-section agricultural tool while the Op. 14 inventory is significantly
composed of biface distal fragments and biface dart points. The subtle differences in each
collection provide insight demonstrating varying primary activities for each location.
Mano and metate fragments provide clues to the processing of plants, seeds, minerals, and
other necessities of life. The smoothed and polished surfaces on many of these tools are
the hard proof of Maya practical uses of this type of material culture.
Locally produced ceramics include bowls, jars, and plates among other forms. There
are clear if not obvious uses for many of the pottery forms, even if the specific tasks may
have varied from across the site. Some sherd disks have central perforations that have
often been described as spindle whorls and discussed as belonging to weaving activities.
The dilemma for interpreting specific function of any particular artifact category is that the
focus or reason behind the use of material may vary from one location to another given the
makeup of the occupants and/or users of the material culture. As Carpenter et al. (2012)
have stated, Even high-status households participated in this activity (spinning), although
their economic focus was slightly different.
The Externally Derived designation is applied to those items/artifacts that are not of
local material. There are artifacts or raw material traded into the area and then
incorporated into everyday use or activities in a variety of capacities (Practical, Ritual,
etc.). Some of the best examples are obsidian blades and bifaces that function not in the
same capacity as chert artifacts at Tikal, but as a complement to chipped stone functions.
The brittle nature and the rarity of obsidian in the lowlands make the use of this natural
glass a top option for fine-line incising, engraving, and slicing. Most of the Tikal obsidian
is in the form of blades, and many show edge wear related to slicing and soft-cutting
activities. Several of the blades, however, show more intensive use indicated by extensive
reworked edges, perhaps for engraving harder material such as bone.
Other materials in the Externally Derived category include manos and metates made
from stone acquired from distant sources. Most of the ground stone fragments are of stone
reminiscent of the Guatemala Highlands. The fragments excavated as part of this project
are very small and likely represent recycled segments of the utilized tools. A reuse or
recycling of the mano and metate fragments is likely tied in part to the artifacts being
made of exotic material. Greenstone (sometimes called jade or jadeite) is represented by
items that served a decorative or ornamental function (overlapping the following
behavioral use of material culture). Some ceramics, few in this collection, are also from
external sources and may represent either special function items or perhaps vessels gifted
into Tikal.
The Decorative and/or Leisure designation refers to items that do not function in a
subsistence capacity, but are important to daily life. Included under this category are
artifacts that serve as adornments or perhaps indicate an association with a leisurely
activity. Items recovered by the UCAPT include jade artifacts that may have been
adornments. A ceramic ear-spool fragment and a perforated shell pendant also served as
adornments likely emphasizing the wearers position and status at Tikal. A bone awl
fragment, decorated with incised lines, is included as a decorative element from material
culture although it may have functioned as a practical implement as well. Artifacts
indicating a leisure function include the multitude of sherd disks that are often interpreted
as gaming chips or whistles associated with recreational activities.
The functional designation of Ritual here includes figurines. They may be exclusively
ritual in use or possibly multifunctional. The figurines (as represented in this collection by
fragments only) as well as whistles can serve in ritual activities and may have been moved
back and orth as required for social, political, and religious activities. Ceramic vessels,
including some of the types in the current collection, can serve a ritual activity and/or
purpose. Interestingly, ceramic whistle fragments and many of the figurine fragments were
from El Jaguar, a plazuela group with a significant courtyard. The Palace Reservoir
excavations also produced a significant percentage of the figurine fragments. Examples of
Ritual use for ceramics include their placement as containers in caches, burials, and so on,
a common practice across the Maya Lowlands.
The understanding of Craft Occupational Activities, either as specialized craft
production or as a cottage industry, can be taken from the quality and quantity of material
culture observed. This means that the context of material production may help determine
the type or level of production. All of the artifacts discussed fit into some kind of
production scheme implemented by the ancient Maya. Defining whether a productive
activity is specialized or not depends upon the archaeological analysts ability to
differentiate between the two. There is a tendency to consider more prestige-oriented
items as specialized and the more mundane materials (stone tools, common ceramics, etc.)
as the result of cottage industry production. The distinction is not always so simple and
context should be a significant consideration for any interpretation.
For the moment, it is sufficient to acknowledge an overlap of Craft Occupational
Activities with the Practical, Externally Derived, Decorative and Leisure, and Ritual
designations. Chert from local sources may not be of highest quality, but serve to produce
the required tools. Finer quality chert may be a limited local resource or imported.
Obsidian is certainly imported, as is the stone utilized in mano and metate production. To
this latter point, Tikal received final products and did not necessarily manufacture the
observed artifacts as part of some form of craft production. On the other hand, the
recycling of these and other items may require some crafting ability, but not likely as a
specialized activity. The recycled or reused artifact most often has a different function in
its new life, but that new function may not always be clear.
Interpretative Data refers to material culture as it allows for additional interpretation
beyond the descriptive component. While one task is the description of material culture as
a basic unit (e.g., chipped stone tools), the second requirement that applies here
specifically is the consideration of interpreting a particular function and whether other
data can be brought to bear on the interpretation. Ceramic vessels, stone tools, figurines,
manos and metates, and so on, all imply certain functions. Beyond a premise of
determined or implied function is the realization that these materials as cultural products
enable human societies, in this case, the Tikal Maya, to adapt to and modify their
environments (physical and social). Figurines and scene-painted vessels also may provide
insight into Maya dress and beliefs. All material culture will have some connection to
perishables and should be given that particular acknowledgment as implied, for example,
by spindle whorls and awls for the weaving of textiles, even though we do not see or
recover the textiles themselves.

Summary
We have outlined the various material culture categories represented in the excavations
across the site. The material and their contexts provide a fascinating look at the use of
these areas and hold interesting implications for interpreting just how these artifacts
made their way specifically into or close to the context in question.
Although many of the contexts are from within reservoirs at Tikal as well as more
distant contexts near Tikal, the range of the material culture represented is surprisingly
broad and impressive. While this might be expected from the large/grand site of Tikal, the
specific areas of excavation might not normally be anticipated to produce the range and
quality of many of the finds.
The time range for artifacts recovered includes material from the Middle and Late
Preclassic, the Early Classic, and the Late to Terminal Classic. While we know Tikal was
occupied and thrived through many of these periods, the material culture from these
periods in the many reservoir contexts calls into question the potential range of activities
that may have taken place at and/or near these water systems as well as the now-forested
zones. The Maya would certainly have been careful about keeping their water supplies and
the reservoirs relatively clean, yet the material culture implies access to the reservoir
environs throughout Tikals history. That access though, perhaps by specialized attendants
of the precious water supplies, may produce an interesting model for how scarce resources
were managed in downtown Tikal.
Contexts for most of the artifacts uncovered by the UCAPT indicate widespread access
at Tikal to a broad array of exotic and locally obtained materials or perhaps the location(s)
for the end of the line disposal of those materials. How do some of these artifacts end up,
for example, at the various reservoirs? Are the material remains recovered during the
UCAPT excavations from postabandonment activity or remnants of ritual offerings? The
latter possibility may be difficult to ascertain in many cases; however, it remains clear that
the Tikal Maya across the site had access to a wide variety of materials.
The artifacts located at specific household groups or at possible agricultural fields
represent the kinds of activities, production to ritual, that took place at those specific
locations. For areas of agricultural production, the material culture tends to be scant,
usually limited to a few ceramic fragments, but still provides indications of ancient
activities and chronology. Lithic remains may be relatively rare as these items even when
broken may be taken back to the house/group and recycled.
 The various artifacts produced from local materials thus indicate part of the successful
adaptation at Tikal to the local environment(s). External contacts as indicated by obsidian
(gray and green), fine cherts, jade, some pottery, and other finds, speak to access and
economic successes. The Tikal material culture recovered by the UCAPT has served as an
opportunity to discuss the kinds of artifacts associated with myriad contexts. This form of
study allows us to document, in a broad sense, the artifacts recovered and the range of
expressions for material culture.
 12 A Neighborly View: Water and Environmental
History of the El Zotz Region
Timothy Beach, Sheryl Luzzadder-Beach, Jonathan Flood, Stephen Houston, Thomas G.
Garrison, Edwin Romn, Steve Bozarth and James Doyle

Introduction
Given the long dry season in much of the Maya Lowlands, water management was a key
part of habitation for urban and rural communities (Luzzadder-Beach, 2000; Johnstone,
2004; Dunning et al. 2012; Luzzadder-Beach, Beach, and Dunning, 2012; Scarborough et
al. 2012). Our focus in this chapter is on reservoirs, which are an important part of both
communities. These features range from dammed-off local depressions (Beach and
Dunning 1997) to large-scale hydrological storage and water treatment systems
(Scarborough et al. 2012). The term aguada generally refers to both natural and
anthropogenic reservoirs. In the Central Maya Lowlands, where Maya sites lie far above
the perennial water table, aguadas stood with chultunes as the two crucial water sources,
especially during the dry season from January to June. This chapter considers the size,
water-holding capacities, and paleoecological evidence from aguadas of the El Zotz region
beyond the western edge of Tikal. Such neighborly views offer an indirect portrait of
Tikal, particularly its western margins and areas beyond. Beset by instabilities and
competing political affiliations, El Zotz and its environs contrast, at times, with conditions
at Tikal. In other respects, as Tikals nearest neighbor to the west, El Zotz confirms a
scholarly need to balance a shared ecological trajectory against highly local patterns.

Environments
El Zotz lies in the midst of the Petn Karst Plateau composed of Cretaceous-Paleogene
carbonate rocks that rise to nearly 400 m (Figure 12.1). Normal faulting has formed a
landscape of ridges and valleys represented here by a higher plateau with its steep and
complex escarpment near the site of El Zotz, which descends into the large structural
valley occupied by the Preclassic site of El Palmar and its large cival (grassy wetland
depression). This structural valley of the El Zotz region runs southwest to northeast,
producing a low-lying corridor that splits the higher karst plateau, a north-south spine
running through the northern Petn of Guatemala. The corridor is a nexus between the
lowland wetlands of the northwest Petn, which drains through the San Pedro River
system to the Gulf of Campeche, and the lowlands that empty northeastward through Ro
Azul and northern Belize and adjacent Mexico to the Caribbean.
Figure 12.1. Location map of El Zotz and surroundings.
Karst dissolution has weathered the landscape into a rich assortment of karst features
such as mogotes, dry valleys, sinkholes, cave systems, and other solution features (Miller
1996; Alvarado et al. 2001). In most cases, aguadas lie in naturally low-lying karst sinks,
generally in dissolution dolines, often in larger bajos or along fault lines where karst
weathering has preferentially dissolved depressions. Some of these depressions naturally
clog with sediment during periods of erosion such as the Maya Preclassic (Beach et al.
2003; Anselmetti et al. 2007). Thus, aguadas may start to fill with water because of the
unintended consequences of sedimentation by clays (Dunning et al. 2002), including the
case of the Aguada Catolina in the Petexbatun area, which filled with a meter deep clay
plug around the Middle Preclassic about 3000 BP (Dunning and Beach 1994). Other
aguadas may have started as quarries that the Maya adapted logically into reservoirs
(Weiss-Krejci and Sabbas 2002): A pronounced cavity represents a natural place to collect
water. Our main study site here, the El Zotz aguada, is a broad karst sink, solution doline,
lying at the western margin of the site of El Zotz. It is a roughly circular depression with a
surface that lies slightly below 230 masl. Assuming the highest elevation of the water to
be 234 masl, then the aguada has a larger and less symmetrical dimension of about 280 m
northeast to southwest and 200 m northwest to southeast in diameter.

Background
There is a century or more of scholarly work on aguadas in the Maya world (Akpinar-
Ferrand et al. 2012; Dunning et al., this volume). Such studies corroborate their ability to
store water and supplement ground and surface water, where available, or provide water in
dry periods. In many cases the ancient Maya lined their aguadas with plaster, clay, and/or
stone to improve water-holding capacity (Adams 1981; Akpinar-Ferrand et al. 2012).
Aguadas, like all reservoirs, have a trap efficiency (Brune 1953), meaning the
percentage of all sediments washed into the reservoir that remains trapped in the reservoir.
For aguadas, virtually the only losses are through swallow holes and overtopping of the
upper limits. As a consequence, trap efficiency should vary greatly, depending on the size
of outlets, the capacity of outlets to take in water, the sites sedimentation history, and the
capacity for storage versus flood, or other depositional agent, magnitude.
 Aguadas often contain important repositories of paleoecological information. The
sediment within was often deposited in standing water with low oxygen conditions, thus
leading to preservation of pollen and other microfossils. Also, aguadas can provide
information about highly local conditions: rural areas near agricultural production, sites
with mixed environments, and urban zones. A few studies have taken advantage of this
broader potential for aguadas in the Maya Lowlands (including Rue 2002; Dunning et al.
2003; Akpinar-Ferrand et al. 2012; and Wahl et al. 2007). We have only begun to study the
rich information aguadas hold for historical, geomorphological, ecological, and climatic
change. This chapter introduces aguada studies in the Biotopo San Miguel la Palotada-El
Zotz, an area of 354.68 km2. We focus here primarily on the aguada at El Zotz, the main
site in the region. The evidence for these wetlands derives from multiple excavations and
cores and a battery of chemical analyses, including AMS dating, pollen, phytoliths,
artifacts, and geochemistry.
The El Zotz region includes the main site of El Zotz and its satellites on an escarpment
north of the lowland corridor. Nearby is the separate site of El Palmar, which spreads
across a rise at the edge of a large cival in the midst of this corridor. Recent excavations
indicate that El Palmar began by the Middle Preclassic, with multiple phases of
occupation, and then continued, albeit with an interruption, into the Early Classic (Doyle
2012: table 1; Doyle et al. 2011). Monumental construction ceased at El Palmar during the
Late Preclassic, and occupation appears to have shifted over the Late Preclassic and Early
Classic to the upland terrain around El Zotz, an area with more defensible, high ground
(Garrison and Garrido 2012; Doyle, Garrison, and Houston 2012; Doyle et al. 2012). El
Zotz next experienced curious surges or pulses of population or construction, followed by
signal gaps, when macro-antagonisms with sites such as El Peru and Tikal buffeted the
region. Recent excavations also show that the El Zotz Maya expanded throughout the
Classic Period, again with multiple if highly episodic phases, and persisted into the Petns
broader Terminal Classic collapse and beyond to the Postclassic (Arredondo Leiva and
Houston 2008; Prez Robles, Romn, and Houston 2009; Garrido Lpez et al. 2011;
Garrido Lpez et al. 2012). Whatever the causes for the broader collapse evidence for
drought has only increased the Maya in this region somehow endured, perhaps because
of the success of their large and reliable aguada.

Methods
We investigated the aguadas with a series of excavations aimed at gathering
paleoenvironmental information and asking how or whether the ancient Maya altered the
aguada for water management. The following articles and chapters discuss in detail the
methods we have used here, including Beach et al. (2002, 2003, 2006, 2008, 2009, 2011)
for soil methods, Luzzadder-Beach and Beach (2008, 2009) for water chemistry, and
Bozarth and Guderjan (2004) for phytolith methods and pollen methods (see also Jones
1994). Our range of methods includes geochemical, physical, stratigraphic, and
chronological techniques to characterize and date sediment layers. In the field we
identified field texture, structure, pH, consistence, HCl reaction, Munsell color, magnetic
susceptibility, preliminary stratigraphy and horizons (adjusted after laboratory analyses),
and additional soil characteristics (SCS 1993; SSS 2003). We sampled all potential
radiocarbon materials and artifacts by soil horizon changes and systematically by depths.
A GF Instruments Magnetic Susceptibility Meter SM-20 was used to measure magnetic
susceptibility at 50 mm intervals on soil profiles. Brigham Young Universitys Soil
Laboratory analyzed samples from soil profiles for elements using inductively coupled
plasma atomic emission spectrometry (ICP AES) (Johnson et al. 2007). The lab also
measured 13C, stable carbon isotope ratios, of the residual soil humin with the Finnigan
Delta Plus isotope-ratio mass spectrometer along with a Costech elemental analyzer
(EAIRMS) (Webb et al. 2007; Sweetwood et al. 2009; Wright et al. 2009; Beach et al.
2011). Beta Analytic employed AMS procedures on all radiocarbon samples, which we
report here as 2 (95 percent probability) dates as calibrated by INTCAL98 Radiocarbon
Age Calibration (Stuiver et al. 1998). Our radiocarbon samples were from young wood,
peat, or charcoal in distinct strata.

Water Chemistry Methods

We collected water samples at stations in the Biotopo San Miguel la Palotada-El Zotz of
Guatemala during the 2009 and 2010 field seasons, at the end of the dry season, which
simulates the most extreme conditions for water quality and availability in this region
(Luzzadder-Beach and Beach, 2008). Sample sites included surface water sources and
springs: the El Palmar Cival and a spring feeding the Cival, the Aguada Palmar, Aguada
Pucte, and the Cecon Aguada next to the El Zotz Camp. We measured electrical
conductivity (EC), total dissolved solids (TDS), salinity, and temperature in the field using
an HACH Conductivity meter, then collected water samples in triple-rinsed plastic bottles
and took them to the George Mason University (GMU) Water Quality Laboratory for
further testing. Before analyzing the samples for dissolved minerals, we followed the
standard procedure of filtering the water samples through 0.45 m/47 mm Gelman
Scientific Metricel membrane filters to remove any remaining undissolved solids. In the
GMU Lab, we retested the EC and TDS before and after filtering, and then tested the
samples for sulfate (SO42), calcium (Ca+2), magnesium (Mg+2), total hardness (as
CaCO3), chloride (Cl), and nitrate (NO3). The methods and instruments that we used to
analyze the water chemistry are detailed in Beach et al. (2008) and Luzzadder-Beach and
Beach (2008, 2009).

Stratigraph.5y and Chronology

El Zotz Aguada Units
There were five excavations (Figures 12.2 and 12.3) in the El Zotz aguada, ranging from
the berm unit on the east side of the aguada to the unit nearest the aguadas center. The El
Zotz aguada had four main zones: a Late Classic layer around a gravel and plaster lining at
105115 cm with sediment accumulation into the Postclassic; an Early to Late Classic
depositional zone; an Early Classic floor made of flagstone, clay, and ceramic at 230 cm;
and the underlying paleosol beneath the floor.
Figure 12.2. Zotz Aguada excavation map.

Figure 12.3. Zotz Aguada stratigraphy.

We excavated one unit (EZ-13A-5) across part of the prominent berm along the west
side of the aguada (Figures 12.2 and 12.3). The berm area is a preexisting higher edge of
the karst sink with a terrace or line of boulders that anchor a ridge of heaped-up sediment
nearly 1 m thick. There were no datable artifacts in the berm. The surface soil is a
Rendoll, typical of young, regional uplands (Beach et al. 2003; see Dunning et al., this
volume). This sequence has a thick (> 30 cm), dark (10YR2/1) A horizon overlying a Ck
horizon (10YR4/1) over the berm boulders. The Maya built the berm wall above the old
topsoil now demelanized (10YR 3/1) but ~ 25 cm thick, which formed above the 2Ck
horizons, largely light-colored (10YR 7/1) sascab.
Aguada unit 13A-1 (Figures 12.2 and 12.3), near the aguadas berm, showed a highly
disturbed soil, both with a faint A horizon in the SW corner and with an overly thick A
horizon in SE side. This was probably created by a large tree fall that toppled top soil from
the SW corner and onto the SE corner. The fall occurred recently enough for the A horizon
in the SW corner to have only faint melanization (10YR 4/2 rather than 10YR 2/1 in
surrounding soils) and thin development (7 cm). The rest of the excavation displayed the
normal profile of a footslope soil with a thickened A horizon (> 50 cm) created by
deposition of the tree fall.
Aguada unit 13A-4 (Figures 12.2 and 12.3) provided a window into both floors of the
aguada with a thick cobble, boulder floor from 90 to 120 cm depth and with an AMS date
just above the lower floor as Cal BP 1820 to 1640. This corresponds to the Early Classic
floor in 13A-2 and 3. Corresponding to the Late Classic, this upper floor is a packed
limestone gravel and densely packed clay zone; two ceramic sherds here were Late Classic
in date. This unit exposed a thin (10 cm) O and A horizon (10 YR 2/1) sequence and deep
(>100 cm), gray (10YR 4/1) Cg horizon that indicates continuously moist conditions.
The next two aguada units, 13A-2 and 3 (Figures 12.2 and 12.3), were near the
aguadas center and exposed deep profiles of sediment and both floor levels. Each
excavation continued below the lowest floor, 13A-2 (Figure 12.4) mainly through the floor
at 250 cm and 13A-3 down to 360 cm, 180 cm below the floor, to understand how the
sediments changed above and below the floor.

Figure 12.4. Zotz Aguada 13A-2 Photo.

In 13A-2, the Lower Floors Early Classic construction is of stone layers with clay,
plaster, and ceramic, composition that is similar to another recently discovered aguada
floor at Uxul, Mexico (Grube and Paap 2010), and a floor at the Palace Reservoir at Tikal
(Scarborough and Grazioso Sierra, this volume). The El Zotz aguadas lower floor is
securely Early Classic on the basis of the following evidence. First, Sierra Red ceramics
lay on the floor, along with an unambiguously Early Classic jar within the floor matrix.
Second, there was one AMS date of Cal BP 2680 to 2340 on charcoal at 250 cm near the
floor, but this probably washed in from reservoir construction; this last point can be
inferred because two other AMS dates in parallel stratigraphy were Cal BP 1820 to 1640
just above the floor in 13A-4 and Cal BP 1520 to 1340 20 cm above the floor in 13A-3.
Since the artifacts and two of three radiocarbon dates point to an Early Classic date (the
third AMS date is much older than the preponderance of evidence), we conclude the lower
floor was an Early Classic construction. Indeed, all of the identified ceramics below 125
cm in the unit were Early Classic with the exception of one Late Preclassic ceramic at 160
cm. Such mixings are common in early deposits at El Zotz and, in all cases, logically cue a
later date, not one in the Preclassic.
The Upper Floor is even more securely dated to the Late Classic. All of the ceramics
here and above this level are Late Classic. Two AMS dates from just below and above the
upper floor date to Cal BP 1290 to 1140 and to Cal BP 1300 to 1080, respectively. This 10
cm thick floor is composed of limestone gravel compacted together, and not the cut stone
of the Early Classic floor. The upper floor required significantly less labor for building and
upkeep, but it was also wavy and discontinuous because of bio- and argilliturbation.
Akpinar-Ferrand et al. (2012) discussed a similar plaster floor in an aguada dated to cal BP
1030870 (p = .95, 2) at the Aguada Los Tambos at the Late Classic site of Xultun in
northeast Petn.

Soil Chemistry and Physical Data

In the aguada units clay dominates all the soil textures, ranging from 90 to 93 percent of
the total below the ancient Maya floors and above 85 cm. However, during the Maya
period clay drops to 4393 percent of total, which, although clays in terms of soil texture,
indicates an increase in coarseness. In turn, this signals a higher energy deposition during
this time, with the proviso that some sediment was volcanic ash, to judge from its clay
mineralogy and identification of ash shards in the pollen analysis (see also Tankersley et
al., this volume). The berm soils were also clays but ranged from 58 to 80 percent clay,
usually much lower than the aguada soils, a pattern to be expected from low-energy
deposition.

Carbon Isotope Ratios

Soil carbon isotopic ratios from the soil units at EZ-13A-5, 2, and 3 provide further
evidence for environmental change (see also Tankersley et al., this volume). The first unit
EZ-13A-2 provides a 250 cm profile depth, over which the carbon isotopic ratios rise from
29.1 near the surface to 25.4 in Late Classicdated floor sediments and persist near
this level (25.2) to just below the Early Classic Floor at 250 cm depth. There is a slight
decrease between the Early Classic floor and the Late Classic floor down to 26.2. The
second unit EZ-13A-3 demonstrated similar levels of soil carbon isotopic ratios, rising
5.89 from the surface to as high as 21.03 and producing levels of 23.21 near the
Late Classic floor and 22.88 near the Early Classic floor. The 13C evidence from
13A-3 also shows that the prefloor aguada environment had elevated organic inputs from
C4 vegetation with carbon isotopic ratios as high as 23.75 before the earliest date in
this profile.
The berm unit indicated a rise through the terrace material from 30 to 22.92,
though we have only artifactual evidence for the date for this structure. Further
archaeometric data may not refine the chronology because it remains strongly subject to
contamination from bioturbation. The surface of both of the latter units have high ratios,
near 24 and 27, where the usual surface is 29, suggesting that these surfaces
have been affected by root throw and other forms of bioturbation, all certain to be active
processes in this environment. Indeed, the surface of EZ-13A-3, and EZ-13A-1, has clear
stratigraphic evidence of bioturbation in a greatly thickened A horizon along one side.
Beach et al. (2011) used the following equation to estimate the percentage of soil
organic carbon (SOC) derived from C4 plants (Nordt 2001: 423; Wright et al., 2009):
% SOC obtained from C4 vegetation (CC4) =

100* (13Csoc 13CC3)/ (13CC4 13CC3)

On the basis of Wright et al. (2009) in the same region, we used 27 for 13CC3 and
12 for 13CC4 to calculate the percentage of the soils SOC from C4. We should note
that David Lentz et al. (this volume, Chapter 8) write that maize is the only C4 species of
all the food plants at Tikal, though there are other disturbance C4 plants. Using this
equation, the berm area around the terrace had as much as 27 percent of its SOC from C4
plants; the aguada units had up to 40 percent of its SOC from C4 vegetation. Both the Late
Classic and Early Classic floor areas ranged from about 10 to 27 percent SOC obtained
from C4 vegetation, but the percentage of SOC obtained from C4 vegetation dropped to as
low as 5 percent between the floors. Below the floor in aguada unit 13A-3, the soils still
had up to 22 percent of their SOC from C4 vegetation, perhaps indicating the earliest
agricultural impacts at El Zotz. This could indicate C4 enrichment from grasses and sedges
growing in the aguada. These lines of evidence suggest that the regions organic matter
through the last two millennia had more than half of its source from C3 or forest species,
registering strong C3 plant inputs through Maya times. The sequence of decreased C4
vegetation between the floors and increased C4 vegetation near the floors may result from
cyclical high and low clearance, sampling error, or simply site evolution (see also
Tankersley et al., this volume). The latter point of site evolution would indicate more C4
vegetation input into organic matter around the Early and Late Classic floors, perhaps for
C4 crops in the Early Classic, and more runoff from maize use and processing or more C4
species like grasses in the Late Classic. The evidence of Zea mays, which we discuss later,
together with higher 13C indicative of C4 species near the Preclassic floor corroborates
the nearness of maize agriculture at the Early Classic site. But we only have the higher
13C with no maize pollen at the higher Late Classic floor, though squash pollen appears
at one Late Classic level. Interestingly, we have copal throughout.

Bejucal and El Diablo

Studies of two aguadas around Bejucal and one at El Diablo (Figure 12.1) produced more
subtle data about ecological change. Around Bejucal we studied two excavation units, one
in the Los Bocutes Aguada, located 1.42 km NNE of the ruins, and one in the large,
sprawling Bejucal Cival, 0.5 km ESE of the ruins. The first excavation (MA5A1) was a
1.5 by 1.5 m window near the center of the aguada into the top 165 cm, which we
extended to 234 cm with a thin corer. At 234 cm the corer struck hard limestone in three
different locations that lie near the same level as the stone paved floor at the El Zotz
aguada discussed earlier. Indeed, the sediment preserved many of the same characteristics
as at El Zotz, though the corer did not reveal an upper floor, so clearly visible in the El
Zotz excavations. The soil profile is typical of moist Rendoll soil, with a thin O horizon
that drapes over a thick A horizon (10YR 2/1), reaching a depth of 55 cm, which then tops
a very dark gray (7.5YY 3/0), organic rich Cg horizon. Again, the sediments preserve
visible organics and have textures that are nearly 100 percent clay; the exception is from
165 to 235 cm, which has preserved laminations and is about 60 percent sand and silt. At
similar depths through the El Zotz aguada, higher energy layers like these are associated
with the Early Classic period deposition above a floor. The stable carbon isotope evidence
is also revealing because, although it produces no increase in 13C ratios, all measures
through the profile have elevated 13C from 15 to 225 cm, significantly higher than any
other site in the region, about 22. This produced an estimate of about 29 percent C4
vegetation in sediments below vegetation that is today largely C3, and may evidence for
C4enrichment. Intensive bioturbation may have mixed and expunged the influence of
tropical forest C3 vegetation, which have 13C levels of ~ 27. This confirms the need
for further testing, excavation, and additional proxy evidence.
For similar reasons, the large cival next to Bejucal also produced incomplete but
promising results, with 13C as high as ~ 22 producing an estimated ~29 percent C4
vegetation. This Vertisol soil profile with its diagonal horizons, slickensides, and all-clay
textures would probably neither have good paleoproxy preservation nor reliable
stratigraphy for dating. The soils highly contorted horizons (argilliturbation by wetting
and drying cycles of 2:1 clays) confuses relative dating, and its orange mottling indicates
oxidation down to 160 cm. We hope to find smaller, wetter aguadas that preserve better
records in this cival.
We also analyzed two excavations in El Diablos aguada, one on the margins and one in
the middle. These units showed evidence for quarrying before the sediments started to
build up. The bedrock steps abruptly down by more than 1 m in the margin pit, and the
sediment sequence in both units offers no evidence for preexisting soils. Most depositional
soil sequences, as at El Zotz, contain some evidence of the original soil level. For
example, the magnetic susceptibility profiles and the organic carbon contents for these
units drop steadily throughout the profiles. They do not rise again, in contrast to other
buried soils (Beach et al. 2008). The 13C rises as high as ~ 26, reflecting an estimated
6 percent C4 vegetation, well within the range of natural fractionation and not significant.
We found no evidence of a floor in this aguada, though the clay rich sascab would have
produced a very slow infiltration rate and high trap efficiency. It is possible, however, that
the excavation, which still found artifacts at 160 cm, did not reach deeply enough to find a
floor. The sediments appeared to be hardened sascab, but they may have been lime plaster
washed in from the temples uphill at El Diablo; this deposit solidified as water dripped
into it over the millennium and a half. The sand fraction ranges from 7 to 13 percent and
SiO2 is 1737 percent, both high for the region and suggestive of accumulation in this
sink.
To judge from the two AMS dates from El Diablos aguada central unit, Early Classic
sedimentation was rapid. The lowest date from 130 cm (Cal BP 1830 to 1560) overlaps
with the date from 95 cm (Cal BP 1690 to 1520), and all the ceramics from 160 cm
upward were Early Classic. This evidence indicates that much of the sediment was
deposited within the Early Classic period, the period of greatest building activity or just
after abandonment.

Paleoecology Proxies of the El Zotz Aguada

We obtained a suite of paleoenvironmental proxies (Figure 12.5) for our deepest
excavation unit in the aguada (13A-2). Below the Early Classic floor there was no clear
evidence for anthropogenic use. In both aguada central units (13A-2 and 13A-3) below the
aguadas constructed floor the sediments were clays with lenticular peds and had layers of
gravels with faintly oxidized colors, from very pale to pale brown Munsell colors of 10YR
7/3 to 6/3. There are only faint visual clues to a buried soil surface faintly oxidized
colors like 10YR6/3 that remain after demelanization below the aguadas floor (Figure
12.2). But chemical evidence in the form of high amounts of organic carbon, highest
levels of several metals such as Fe and Ti, evidence for up to 22 percent C4 vegetation,
and increased magnetic susceptibility all point to the subfloor sediments as being
paleosols (Beach et al. 2008). The principal change from below to above the floor is the
changeover from oxidizing to reducing conditions, thus underscoring the efficacy of the
floor in changing this from a seasonally dry to a perennially wet environment.

Figure 12.5. Diagrams of pollen morphs (above) and biosilicates (below) from El Zotz.
The lowest pollen and phytolith sample (Figure 12.5) is from 210 cm, just above the
Early Classic aguada floor. At this level, disturbance was already near its peak, as
indicated by multiple lines of evidence. First, the highest 13C levels occurred in both
units through this zone (22.88 in 13A-3) and (25.2 in 13A-2). Second, a high
concentration of small particulate charcoal and the highest concentration of large
particulate charcoal (> 80 m) both reflect high amounts of regional and local burning.
The charcoal concentrations are relatively robust at these sample sites, though not as high
as in areas of intensive agriculture such as the wetland field-canal sections that date to the
Late Classic period in Belize (Beach et al. 2009). Third, for the first time, higher-energy
sand particles show up in these sediments, with sand rising from 0 to 10 percent.
Elemental analysis (ICP AES) shows the aguada has 2040 percent SiO2, which is higher
than most of this carbonate dominant region. The sand could have originated in
allochthonous sources or eroded from bedrock (but see Tankersley et al., this volume).
Fourth, arboreal phytoliths represent only 6.8 percent of all phytoliths, while Asteraceae
(the sunflower family) phytoliths represent 31 percent of all phytoliths and Asteraceae
pollen account for 93.1 percent of all pollen. This zone also has four Zea mays pollen
grains, high amounts of grass pollen, and volcanic ash; the dominant tree pollen are Pinus
sp. (pine) and Protium sp. (copal). Pachira aquatica Aubl. (provision tree), which
produces large edible fruits, appeared for the only time in the record, as did Combretaceae
pollen, probably from bullet tree (Terminalia buceras (L.) C. Wright). Paleoecologists
have often associated the presence of Pinus pollen with disturbance, despite the case of El
Zotz, its distant source is today at least 30 km away. This is because of Pinus pollens
capacity for distant transport, and their greater likelihood of deposition when the
surrounding trees do not intervene in that movement. This is also the only zone in the
aguada that does not have fern spores.
Above 210 cm (Figure 12.5), at 150 and 110 cm, the layers range from Early Classic,
on the basis of consistent ceramic evidence, to near the Late Classic AMS date at 105 cm
of Cal BP 1290 to 1140. Disturbance indicators all decline through this 100 cm of infill;
small charcoal declines 3040 percent but 13C decreases only a small amount. Another
countervailing trend is that high-energy sands increase through this zone, reaching 33
percent of the sediment in a profile dominated by nearly 100 percent clay above and below
this zone.
At 150 cm smooth sphere phytoliths (< 10 m) increase from 4 percent to 30.1 percent,
showing that at least one woody species became much more common at this level. The
taxon may also be a disturbance indicator in view of its parallel with a slight increase in
Asteraceae pollen, another strong index of cultural disturbance. Copal pollen is still
present. The regional and local forests continued to be burned, as shown by an increase in
the number of charred phytoliths from 8 to 11 percent, a relatively large concentration of
small particulate charcoal, and the presence of large charcoal, though it drops by 8090
percent. At 110 cm this assemblage is very similar to the 150 cm sample except for the
presence of Cucurbita (squash) pollen (0.2 percent). There is still abundant evidence of
cultural disturbance in that Asteraceae still dominates (92 percent) the pollen. Copal
continued to grow near the aquada. Regional and in situ burning declined slightly, as
reflected in decreases in charcoal concentrations and charred phytoliths.
The second large spike in economic and disturbance taxa, along with charcoal and the
highest 13C, occurs around the Late Classic floor at 100 to 90 cm (Figure 12.5); this has
an AMS date of Cal BP 1300 to 1080. Native grasses, including Panicoids, Chloridoids,
and Pooids, dominate the phytolith assemblage, and Asteraceae abound in the pollen and
are second in the phytolith assemblage. Low-spine Asteraceae pollen, generally dispersed
by the wind, is especially high. Although researchers generally consider native grasses and
Asteraceae as disturbance indicators, this large family of 45 species in the northern Petn
includes at least one herb (Eleutheranthera ruderalis [Sw.] Sch. Bip.), cultivated for
medicinal purposes in the recent past (Lundell 1937). Some of the Asteraceae may have
originated from this or other unknown medicinal plants.
At 80 cm Pinus was the most common arboreal pollen, reaching its highest frequency
(3.3 percent) at this Late Classic level, 7 cm below the AMS date of Cal BP 1290 to 1140.
Copal (Protium) is the only other arboreal pollen present, suggesting the El Zotz Maya
may have grown it for the ritual use of its resin or as medicine (Balick et al. 2000). The
only evidence of food-producing plants is the presence of cheno-ams, which make up 7.7
percent of the assemblage. These plants are also potent indicators of disturbance. The
Maya were burning local vegetation at this level, a point emphasized by the presence of 11
charred phytoliths. Regional burning also increased, as indicated by a rise by nearly a half
of small particulate charcoal (1080 m).
The amount of woody plants decreased at the 80 cm level, with a decline in smooth
sphere phytoliths from 36.4 to just 4.2 percent. Three Cucurbita phytoliths occurred at this
level, providing additional evidence of squash. The frequency of cheno-am pollen
increased from 3 percent to 22.9 percent, suggesting possible Maya use of
chenopodium/amaranthus for edible greens. After evaluating many other samples, we
surmise that, if the Maya used these plants for seeds, then larger aggregates of pollen
would have been present. Copal pollen continues as before. Charcoal concentrations
demonstrate that there was more regional burning at this level than any other. The
preservation of five volcanic ash shards also indicates rapid deposition in water since
otherwise glass would break down rapidly into smectite and kaolinite clays in this wet,
tropical, and alkaline environment (see Tankersley et al., this volume).
We have no dates directly from the 50 cm level, but it is ca. 900 BP, in the Early
Postclassic, according to a second-order polynomial regression of the sequence of AMS
dates. Disturbance taxa are still significant and 13C has actually increased at this level,
though charcoal drops abruptly. Asteraceae continue to be the dominant in pollen and
phytoliths. As in all the other samples, copal pollen is present, continuing to implicate
management/cultivation of this important species. There was less regional, local, and in
situ burning, as inferred from concentrations of particulate charcoal and charred
phytoliths. Overall, this level reflects somewhat less but continued disturbance, and the
archaeological evidence indicates that Early Postclassic El Zotz continued to be occupied,
indeed, in the South Group, not far from the aguada. Moreover, Tankersley et al. (this
volume) draw a similar conclusion from isotopic, charcoal, and pollen evidence around
Tikal. In the 50 cm above this we have only 13C evidence, which shows a steep decrease
by 5.89 ppt in one unit and 4.3 ppt in another, both significant proxies for a changeover
from a mix of C4 to mainly C3 vegetation.

Future Study: Water Management and Climate Drying

Hypotheses
The El Zotz aguadas floors imply high labor inputs in the Early and Late Classic, which
allowed the inhabitants to use this as an aguada. But there are several more logical steps
ancient water managers could carry out to increase the magnitude and quality of the water.
First, there are two main logical positions to dam for better water management. One is to
the southeast of the aguada, where maps suggest a higher ridge that extends from the
urban site to a natural high point (Figure 12.6). With relatively small effort, a dam at this
constriction could have increased reservoir capacity by 23 meters. Second, the berm
excavated to the north of the aguada is partly anthropogenic. The berm continues around
to lower points in the topography, detaching the main aguada from smaller depressions to
the northwest and possibly the northeast. These areas lie closer to the steepest parts of the
site, a slope that rises to the site of El Diablo, an area with the highest erosion rates,
especially during episodes of construction. As Scarborough et al. (2012) have suggested
for the Temple Reservoir at Tikal, these smaller depressions may be settling tank features
that separated sediment laden water from the main aguada, or, perhaps, they arise from
natural karst features. We plan to test these hypotheses in a future season. It is also
possible that the sand that increases first in the Early Classic and then again in the Late
Classic is associated with water filtration, as Scarborough et al. (2012) suggest for Tikals
reservoirs. There is no sand in earlier or later levels, and SiO2 levels are higher in the
aguada sediments than in the El Palmar sediments (Beach et al. in press). Moreover,
Tankersley et al., in this volume, found no quartz in crushed local limestone samples but
plenty in reservoirs.
Figure 12.6. Cross sections of Zotz Aguada.
Many scholars have suggested the Maya cleaned out their aguadas (Dunning et al., this
volume; Akpinar-Ferrand et al. 2012). Possibly the examples at El Zotz were cleaned:
After all, the lowest sediments show abundant disturbance proxies, though there is no
clear evidence of truncation in the sediments (evidence for disturbance in the reservoirs
stratigraphic sequence starts with a floor itself). We interpret the berms wall construction,
with its stone terrace covered by clay, not as a two-stage plan but as one: a stone terrace
wall and a second berm of clay piled from dredging.
Climatic changes could have had significant impacts on the El Zotz region. Recent
paleoclimatic change evidence indicates 3050 percent reduction in the summer
precipitation in the later Late Classic period (Medina-Elizalde and Rohling 2012), which
produced an estimated 30 percent lower lake level at Chichancanab in the Yucatan.
Similar evidence is taken from southern Belize (Webster, J., et al. 2007). There is also
strong evidence for Late Preclassic drying in those areas, and El Zotz resides in between
it may well have been similarly affected. Indeed, the Late Preclassic drying potentially
negated the advantages to living near the El Palmar cival, leading to a compelling reason
for this sites early abandonment and an equally compelling motivation to engineer a
reservoir at El Zotz during the Early Classic. Without question, El Zotz offered a more
defensible location as a presumed Pax Preclasica turned to times of insecurity and stiff
competition. Further research on paleoecology at these sites should clarify such
interactions.

Water Holding Capacity

Brewer (2007) and Akpinar-Ferrand et al. (2012) estimated water storage capacity in
aguadas by modifying an elliptical cone-volume formula
Volume = H(1/2) * * (L/2) * (W/2)
where H, , L, W are defined as H = height, = pi, L = length, W = width.
The El Zotz aguada (Figure 12.6) could today hold water from its base at about 230.5 m
to about 234.5 m in elevation, at which point it would spill out to the southeast over the
proposed impoundment. The Brewer equation, with a maximum depth of 4 m, and length
of 280 m, and width of 200 m, yields 87,920 m3 (87,920,000 liters or 23,226,006.84
gallons). Several authors have used different assumptions about water usage per person
per day (see Scarborough and Gazioso Sierra, this volume), ranging from 1.9 L (Brewer
2007; Akpinar-Ferrand et al. 2012) to 2.44.8 L (McAnany 1990: 269) in northern
Yucatan. Assuming a broad range from 25 L of water consumption per person per day,
along with a volume of 87,920,000 liters without considering the water balance and trap
efficiency the aguada could provide for the drinking water requirements for 48,000 to
120,000 people today. This capacity far exceeds any credible estimate of population at the
site, though it does not include the unknown quantities for washing, bathing, irrigating,
and sundry rituals of ablution. In the Early Classic, the aguada was about 2 m lower at its
center and would have supplied ca. 131,880,000 L, enough for ca. 70, 000 to 180,000
people (far more than enough water for its population). We should note that
evapotanspiration would vary tremendously from year to year and under forested and
urbanized conditions. Indeed, the urbanized landscape would produce much more runoff
and lower evapotranspiration, enhancing water quantity.
The El Diablo aguada was a smaller, oval shaped aguada, about 30 m east-west and 23
m north-south. Its original depth would have been about 2 m deeper than todays 362
masl, and its upper limit would be 364 masl, yielding a maximum depth of about 4 m. The
preceding equation produces a maximum of 1,083 m3 or 1,083,000 liters and 286,098.33
gallons, which yields enough annual water for ca. 590 to 1,500 people, again according to
the same assumptions as earlier. We plan further modeling to develop more precise
estimates of the water balance of these aguadas during normal, wet, and dry periods after
we can confirm the actual dam sites and reservoir heights.

Water Quality
We sampled and analyzed water from all available sources (n = 6) to ascertain their
provenance, that is, whether they derived from a perched groundwater aquifer or from
rainfall. These samples all represent the surface water supply at the end of a dry season.
All water sources have low to medium electrical conductivity (EC) readings and low to
medium total dissolved solids (TDS), indicating only short residence time in the aquifer.
Preliminary analyses of water sources divided the sources into two distinct groups; cival
or lake sources with medium electrical conductivities and medium TDS values (median
EC = 532 S and median TDS = 253 mg/L), and aguada sources with low conductivities
and low TDS values: (Median EC = 286 S and median TDS = 137 mg/L). The only
exception to this characterization was Aguada Pucte, about 5 km north of El Zotz, which
had the highest values of all the samples (EC = 538 S and TDS = 256 mg/L). Salinities
all were in the low range, from 0.1 ppt to 0.3 ppt, most likely representing evaporation
taking place in these water bodies during the dry season, rather than long residence time in
a geological source. The low EC and low TDS group initially matches other water sources
sampled farther northeast in the watershed in 2008: the Rio Tikal and the Ixcanrio (EC =
236 and 307 S and TDS = 112 and 146 mg/L, respectively), near San Bartolo (samples
courtesy of N. Dunning). Further chemical analysis during summer 2009 and 2010 in the
Water Quality Lab at GMU determined the mineral content and more precise ionic profiles
of these water sources (see Beach et al. forthcoming, for more detail). Clearly, the El Zotz
Aguada would have been crucial and convenient as a rainwater capture and storage feature
for the urban site of El Zotz, yet, at the same time, readily subject to contamination. We
have yet to find springs at El Zotz, though the springs elsewhere in the region have high
SO4, while, as usual, surface water has very low levels.

Conclusions
Although water resources and paleoecology at El Zotz need further study especially to
test hypotheses about water dams, ecological and climatic change, and water quality this
chapter provides a substantial body of new evidence about the ecological history of the El
Zotz area, Tikals nearest neighbor to the west. We estimated water quantities at El Zotz
and El Diablo, evidence for their use chronology, modern water quality, reservoir floor
construction at El Zotz, the first paleoecology record at El Zotz, ranging from the Early to
Postclassic, and evidence for broad ecological change in the form of carbon isotopic ratios
at El Zotz, El Diablo, and Bejucal.
One significant finding at El Zotz was that a Late Preclassic floor in the aguada
changed this feature from a moderately drained pocket bajo with Vertisol soils to a wet
site, in large part by sealing water infiltration with clay, dressed stone, and ceramics. The
Late Classic floor was a far more modest endeavor, with lower impacts on water storage.
In terms of todays aguada floor, the reservoir could hold enough water for the yearly
drinking needs of 48,000 to 120,000 people, well more than the sites estimated
populations. In the Early Classic, before siltation, when the aguadas floor was 2 m lower,
there was enough water for ca. 70,000 to 180,000 per year.
The paleoecology record demonstrated twin peaks in disturbance in the Early Classic
and Late Classic, and disturbance continued from multiple proxies well beyond (30 cm
above in the sediment profile) the highest Late Classic AMS date. Economic taxa show
maize, squash, copal, and many disturbance species that dovetail with high charcoal
inputs, burned phytoliths, and carbon isotopic evidence. Maize only appears at the lowest
sample from the sites Early Classic origins, whereas copal shows up throughout the
sequence in high amounts. The only other clearly economic taxon was squash, which
shows up only in the Late Classic. In contrast, the core from the El Palmar cival had much
greater amounts of maize from the Archaic to the Late Classic. As the El Palmar
population declined, maize agriculture apparently flourished (Beach et al. in press).
Perhaps, El Zotz had an Early Classic agricultural component that evolved with
urbanization into mainly ritual and medicinal associations based on copal. Furthermore,
we hypothesize that El Zotz residents would have been growing corn in the valley
between El Zotz and El Palmar but living in the defensible uplands in the EarlyLate
Classic.
These findings fit broadly with the El Zotz regional survey and with extensive
excavations in the site and its vicinity. First, the El Zotz aguada and the main site El Zotz
arose from what we take to be a Pax Preclasica and its diffuse settlement along the valley
floor. Large Preclassic settlements like El Palmar exhibit little evident concern for defense.
The El Zotz Maya built their aguada as the population shifted into the higher zones,
focusing on palaces, situated on steep hills, at El Diablo and Bejucal, along with expansive
facilities at lower levels. The reservoir sediments and floors with their evidence of wide
fluctuations in erosion, vegetation, and human manipulation may even reflect El Zotzs
surges and gaps of population and construction as the site was buffeted by antagonisms
between El Peru and Tikal. Known conflicts between these large sites erupted during the
Late Classic period in particular, and one encounter resulted in Tikals defeat of El Peru in
AD 743 (Martin and Grube 2008: 49). El Zotz lay in a straight line between these
kingdoms and is attested in glyphic texts to have had good relations with El Peru; a wall
(albeit a low one) between El Zotz and Tikal suggests that their interaction was less than
amicable (Webster, D., et al. 2007). Moreover, the paroxysms of aguada disturbance in the
Early and Late Classic periods sandwich smaller disturbances. The El Zotz record also
includes disturbance indicators that persist 30 cm above the highest Late Classic AMS
date, correlating with the sites surprisingly substantial Postclassic occupation and the
continued reliance on its reservoir (Kingsley and Cambranes 2011; Kingsley and Rivas
2012). El Zotz and environs reveal an environmental pattern of surges, curtailments,
continuities, and resiliencies that allowed it to exist in uneasy relation with its giant
neighbor. Some changes were nonhuman in origin, such as cival depletion; others,
doubtless, arose from the restless frictions of being so close, too close, to Tikal.
 13 Defining the Constructed Niche of Tikal: A
Summary View
David L. Lentz, Nicholas P. Dunning and Vernon L. Scarborough

For many people, mention of the ancient Maya immediately conjures up an image of
Tikals towering temple pyramids emerging from the rain forest canopy. Yet, despite
decades of investigations at the site, relatively little is known about why the Maya chose
this location to found a city, or how its burgeoning Late Classic population was supplied
with water, food, fuel, and timber. Our recent studies of Tikal and adjacent communities
have expanded the understanding of the relationship between the ancient Maya and their
environment. As a result of the various studies described in the preceding chapters, we
now have a much clearer picture of how the ancient occupants of the Tikal area managed
their water, forest, agricultural, and soil resources and how they were able to do so in a
manner that would sustain large populations.
Previous studies at Tikal have revealed the history of palace and pyramid architecture
and a nuanced view of political intrigue both within and between polities drawn from the
epigraphic record. Our work, based on the analysis of archaeolgical plant remains, stable
isotopic data and hydrological features, suggests an ecologically enduring city for well
over a millennium. Of special merit throughout our expanded coverage and assessment of
past environmental factors is the complementary role that forested and agricultural
landscapes played. Although limits to growth are explicitly articulated by our studies
(Chapters 7 and 8), the ingenious land use adaptations made by the Maya of Tikal
sustained their coupled human-nature dynamic for many centuries in an otherwise fragile
tropical ecosystem.
There can be no doubt of the sociopolitical and socioeconomic turbulence that
generationally impacted all decision making for a city of the scalar complexity of Tikal.
The devastating Maya drought of the ninth century surely upended the previously
unbroken trajectory of sustainable environmental use as well as the sociopolitical and
socioeconomic resilience of the city. Notwithstanding the final abandonment of the city,
the fact that it was ecologically resilient for more than a millennium is one of the strongest
testaments to Tikals legacy for human history.

Methodology
Among the underpinnings of the resource management and data analyses presented in this
volume are the several innovative methodological aspects of this study that contributed
greatly to the final results. Several chapters were devoted to discussions of the research
design used by the University of Cincinnati Archaeological Project at Tikal (UCAPT).
Because these methods proved to be of great utility, we encourage others to consider using
them, as well. We are aware of our unique problem set and the methodologies that have
contextually evolved; nonetheless, to introduce and facilitate comparable and accessible
data for future regional analyses, our approaches may be of aid.
Lane and associates (Chapter 3) describe the percussion coring techniques employed in
the reservoirs, aguadas, and other areas. Although the reservoirs were used as water
retention devices in the past, for the most part, they no longer store water and today all are
at least partially filled with sediment. Accordingly, what we recovered were deeply buried
samples from dry sediment. These cores often descended deep into reservoir fill,
providing a broad cultural, geomorphic, compositional, and pedological history of the
reservoirs. The cores were tied by elevation into survey data from the surrounding
landscape and excavation pits, providing both vertical and horizontal controls for
volumetric capacity estimations of the reservoirs as well as the slope gradient into and out
of the tanks. The cores also permitted contextual dating because of the regular input of
charcoal into the system from ancient Maya plaster and ceramic manufacture as well as
cooking fires. Furthermore, study of these cores produced a fine-grained analysis of
volcanic tephra fall in the Maya area during the Preclassic and Classic periods (Tankersley
et al., Chapter 9).
The digitization of the Carr and Hazard maps, created by the University of
Pennsylvania Project at Tikal (Carr and Hazard 1961), made the maps especially useful in
all aspects of our study and will enhance the accessibility of these excellent conventional
survey maps to scientists of all stripes who have an interest in research at Tikal. Chris Carr
(no relation to Robert Carr, who helped create the original survey maps) and his associates
(Chapter 4) were able to digitize and georeference the Carr and Hazard maps to a global
coordinate system.
Building on the digitized and georeferenced Carr and Hazard maps, Weaver et al.
(Chapter 5) applied GIS technology to the problem of hydrologic and ecological modeling
for the site core of Tikal. Using this approach, they created a three-dimensional model of
Tikal and were able to suggest the hydrologic flow direction, flow accumulation rates,
agricultural modeling, and delineation of the drainage areas of Tikal. In addition to
elucidating the catchment areas of ancient Tikal, this study has implications for the
sustainable reoccupation of the Maya Lowlands.

Water
The procurement of adequate water supplies was an essential need for the Tikal occupants
to survive in the seasonal desert of the interior Maya Lowlands. The elucidation of the
water management practices and how those practices evolved at Tikal were particularly
critical for understanding the development of the ancient city. Why Tikal is located where
it is has long been a question for Maya scholars. The absence of permanently flowing
rivers, lakes, cenotes, or any other apparent year-round water sources nearby that would
sustain a community through the dry season or offer obvious trade and commercial
opportunities has intrigued scholars for decades (e.g., Dunning et al. 1999; Scarborough
1994, 1998; Scarborough and Burnside 2010; Scarborough and Gallopin 1991). The role
of paved and slightly canted surfaces designed to shed abundant seasonal runoff from the
great plazas and courtyards into the several large reservoirs throughout and below the
uptown center of Tikal continues to be argued. Nevertheless, new data generated by
members of the UCAPT team allowed Scarborough and Grazioso (Chapter 2) to offer
intriguing details about the development of the water system at Tikal including the role of
architecture and aspects of kingship. It seems that the great ring of water reservoirs
evident from Late Classic times was preceded by at least one spring that emanated from
the area underlying the Temple Reservoir (Scarborough et al. 2012). For a time, this
source would have provided clean water for a founding community near a high point in
the regional topography.
The water retention system that evolved by the Late Classic period was replete with a
cofferdam in the Temple Reservoir, a massive retention dam in the Palace Reservoir,
several other reservoirs constructed for various purposes, and a sand filtration capability
designed, presumably, to keep the water clean as it flowed into the low-lying reservoirs.
Ultimately, the extensive construction built around each reservoir designed to redirect
water during the rainy seasons limited the recharge capacity of the spring(s). This made
the Tikal Maya even more dependent on yearly rainfall to keep the reservoirs full.
Comparison between the water supply systems of Tikal and its near neighbor, El Zotz,
is informative. The later polity, a smaller community than Tikal and one that suffered
uneasy relations with its colossal neighbor to the east, pursued a slightly different strategy
in terms of water conservation. Instead of investing in several large and many smaller
reservoirs, as was practiced at Tikal, El Zotz constructed one enormous reservoir that was
carefully sealed at the base to minimize leakage (Beach et al., Chapter 12). The El Zotz
reservoir was expanded over the centuries so that by the Early Classic period, their
reservoir was able to provide clean water for potentially 70,000 to 180,000 occupants. By
the Late Classic period, the land use pattern at El Zotz had shifted to accommodate the
precarious nature of its political existence, being located between two warring groups,
Tikal and El Peru. At that point the El Zotz inhabitants moved to the defensible uplands
and constructed a low wall, presumably for military purposes, on the Tikal side of their
territory. This parallels other Late Classic centers in the Maya realm such as Dos Pilas
and, to a lesser degree, Tikal. The Zotz fields of maize and squash were located between
El Zotz and El Palmar, another small neighboring community. Copal (Protium copal
[Schltdl. & Cham.] Engl.) wood was a common archaeological plant remain at the site, a
commodity undoubtedly used as a principal component in ceremonial activities. Although
the variety of foodstuffs and forest species found among the paleoethnobotanical remains
at Tikal is greater than at El Zotz, the basic pattern of agriculture and forest use appears
similar.
In the final analysis, their land use strategy and stable water supply served the El Zotz
occupants well because it enabled them to endure into the Postclassic period, long after the
site core of Tikal was abandoned. Undoubtedly, the smaller population of El Zotz
combined with their oversized reservoir in all likelihood added to their resiliency. These
were key factors that facilitated their survival during the ninth century droughts.
Forest Management
While a visitor to Tikal today looks out over a seemingly endless forest, that visitor need
only cross the southern edge of the National Park to witness how quickly the forest can
disappear. As human populations grew in ancient times, maintaining forest resources
would have been of critical importance for the sustainability of each Maya community.
UCAPT investigators offered a comprehensive approach to understanding Maya
paleoecology through intensive studies that combined both paleoethnobotanical and
modern forest ecological data. Thompson and her colleagues (Chapter 7) observed that
there was a decided cooccurrence between the diverse archaeological wood remains found
among the ruins of ancient Tikal and the oligarchic tree species of the modern forest.
These results suggest some type of low-impact forest management practice likely was
utilized by the Tikal Maya to preserve the most highly valued tree species and overall
species diversity. For example, sapodilla (Manilkara zapota [L.] P. Royen) was heavily
exploited by the ancient Maya at Tikal (Lentz and Hockaday 2009; Thompson et al.,
Chapter 7, this volume), yet the tree continues as one of the dominant species in the forest
of today. True, the forests of Tikal have been largely undisturbed in the last 1,200 years
since the Late Classic Maya departed and numerous changes could have occurred in a
tropical forest during that period, but sapodilla is a long-lived tree species (two hundred
years and more) so from the trees perspective it has only been six or seven generations
since the Maya were erecting temples and carving stelae. Compare the Maya forest use to
European logging practices involving mahogany (Swietenia macrophylla King) during the
seventeenth century in neighboring Belize, where forest utilization practices were largely
focused on the intensive extraction of a few species, unlike our Tikal example. As a result
of those single-minded logging practices, mahogany, once a forest dominant, is today an
infrequent or even rare species in Belize. In contrast, the evidence presented by Thompson
and her colleagues reveals that the forest management practices of the ancient Maya at
Tikal were more strongly oriented toward sustainability.
The comparison of oligarchic to the nonoligarchic forest species in terms of their
economic utility as described in the ethnographic record is especially insightful. Almost
all the tree species found in the modern forest are useful to one degree or another
according to ethnographic accounts (Thompson et al., Chapter 7). Indeed, the forest
contains a considerable pharmacopoeia as well as timber and fuel supplies. Evidently, the
ancient Maya found it in their best interests to maintain forest diversity to retain all the
useful timber and nontimber forest products.
Exactly how the Maya managed their forests is another question entirely and one at
which we can only guess, but fortunately the paleoethnobotanical record combined with
evidence from ethnographic accounts and the modern forest surveys provide helpful clues.
The first bit of evidence is from the large Tikal temple lintels. These huge beams, many of
which were cut from trees more than forty years old and some even older than one
hundred years, were extracted from what was clearly old growth forest. These enormous
trees were felled to build Temples I, II, and IV sometime between the late seventh and
early eighth centuries (Lentz and Hockaday 2009), long after the onset of the population
surge that began around AD 550 (Webster and Murtha, Chapter 10). These old growth
forests sat astride some of the best agricultural land in the region yet were protected from
overharvesting in spite of intense population growth that began more than a century
earlier. There must have been some societal control that preserved these forest tracts and
their giant floral denizens, otherwise the forests would have been quickly cleared and the
soils exploited for agriculture. Possibly these old growth forests were royal dynastic lands
that were drawn upon for the massive temple construction projects that were designed to
aggrandize the politys greatest lords and appease the most powerful deities. In any case,
the Tikal Maya exerted some degree of centralized control over their forests because
rampant slash-and-burn land use practices evidently were not permitted, at least until the
seventh century.
Another argument for active forest management is taken from the paleoethnobotanical
data. The pollen record (Islebe et al. 1996) from Lake Petn Itza indicated a forest
clearance in the upland areas that reached a removal plateau of around 60 percent of the
forest cover. The Maya cleared more than half of their upland forest but appeared to
maintain what was left. Examination of the paleoethnobotanical charcoal remains from
Tikal revealed that 90 percent of the charred wood samples collected had parallel rays.
This indicates that the wood was from the trunks of mature trees, not branches or the boles
of young saplings that would have been found in an intermediate term swidden situation.
The evidence suggests they were managing mature forests and extracting their wood from
large trees. The ethnographic literature offers some explanations as to how this may have
been accomplished.
Our observations from the paleoethnobotanical record at Tikal suggest that they used
long-term strategies by way of harvesting mostly mature trees and did so in a way that
would preserve diversity. One possible model of how this could have been accomplished
occurred in medieval Europe, where they employed a technique called partial felling or
umbrella felling to maintain high forest canopies (hochwald). The main principle of this
technique is that trees would only be harvested if there was another tree in the area that
could serve as the seed source or umbrella tree to provide the germplasm for
replacements (Schulze et al. 2002). This strategy allows continual harvesting but
conserves biodiversity. The challenge of cutting down a tall forest tree with a stone ax also
seems daunting, but a culture that managed to build 50 m tall stone temples also could fell
trees using a combination of girdling, cutting, and burning. Why they managed these
forests so carefully is not hard to understand: They needed the wood and the other forest
products. Their forests were their main fuel source, a major habitat for forest animals, and
an important source of construction material, gathered foods, and medicinal plants.

Agriculture
Perhaps no topic has generated as much debate among generations of scholars studying
the ancient Maya as the manner in which they fed themselves. With population growth
over time, the ancient agricultural system needed to expand accordingly and gradually
intensify. One of the most significant aspects of the niche construction activities of the
ancient Maya at Tikal relates to their agricultural methods. Lentz and colleagues (Chapter
8) described a Voronoi diagram (sometimes referred to as a Thiessen polygon) that
approximates the extractive zone of the Late Classic Maya inhabitants vis--vis
surrounding contemporaneous polities. This extractive zone included 1,111 km of mostly
upland area that would have represented the landscape from which the Tikal residents
extracted their basic necessities, outside of a few assorted trade items that were carried in
by human porters. According to the pollen data, approximately 350 km of this extractive
zone was in managed upland forest. Another 165 km of the Voronoi diagram was a
residential area likely planted in orchards and dooryard gardens much as observed at the
Cern site. The rest of the upland portion of the Tikal extractive zone, around 350 km,
was devoted to extensive agriculture in some sort of short fallow cycle, probably with
ridge and furrow plantings (once again, as practiced at Cern) and other strategies to
reduce soil erosion, admittedly with mixed success.
Major crop production at Tikal included maize, three species of beans (Phaseolus spp.),
two species of squash (Cucurbita spp.), cotton (Gossypium hirsutum L.), and several
species of root crops including sweet potato (Ipomoea batatas (L.) Lam.), achira (Canna
indica L.), malanga (Xanthosoma sagittifolium (L.) Schott), and probably manioc
(Manihot esculenta Crantz). A variety of tree crops were cultivated, including cacao
(Theobroma cacao L.). It was believed that cacao could not have been cultivated at Tikal
because of the long, hot dry season. Paleoethnobotanical wood remains of cacao at Tikal,
however, clearly show that they did grow cacao at Tikal. The Tikaleos may have wanted
to import dried cacao seeds for consumption, but there is no reason why they would
import the wood (see Lentz et al., Chapter 8). Most likely they planted cacao groves in
special, protected niches with deep soil, such as the Inscriptions Reservoir, where the trees
could be watered in the dry season (see Dunnings note in Chapter 2).
In a major landesque capital investment at Tikal, the Perdido Reservoir and the land
below it evidently were developed as an irrigation system during the Late Classic period.
This system allowed for a second crop during the dry season and would have dramatically
expanded the productive potential of this sector of the Tikal landscape. Carbon isotopic
results indicate that maize was at least one of the crops cultivated in the fields below the
Perdido Reservoir.
The bajos also played a role in agriculture. Approximately 80 km of the bajo forest
was cleared for agriculture, mostly along the western fringes of the Bajo de Santa Fe. The
remaining 175 km of the bajo area likely was used for intensive forest product extraction.
It has been widely observed that many Late Classic Maya communities have been located
on the edges of bajos, but why this is so, until recently, was largely a matter of conjecture.
From findings reported in Chapters 6, 7, and 8 it seems clear that the bajos were not just
useful; they were, in fact, indispensible components of the food and forest product supply
chains of Tikal. The pollen cores and geoarchaeological data from the aguadas at Vaca del
Monte and Aguada de Terminos, located in the western portion of the Bajo de Santa Fe,
demonstrate that these areas were used for the cultivation of maize and probably the root
crop achira, as well.
At the Terminos settlement, chert debris was used to construct terraces in an effort to
retain moisture and conserve soil resources in an intensified agricultural context (see
Dunning et al., Chapter 6). Notably, this humble settlement on the bajo fringes of Tikal
seems to have outlasted the great civic ceremonial center. The settlement continued well
into the Postclassic period, long after the site core of Tikal was abandoned. Quite probably
their little aguada did not run dry (it remains a fairly dependable water source even today),
allowing the inhabitants to survive while others at Tikal perished or moved on.
A final note about Bajo de Santa Fe: In addition to the small water features (aguadas),
there are hints of possible canals or navigation channels that may have been used to
connect the islands of higher ground within the bajo and the Rio Holmul, a seasonal
stream that winds through the bajo during the rainy season. These canals, if confirmed,
could have been used to facilitate the transshipment of trade items between Rio Azul and
Tikal, at least during the rainy season (Chapter 6). During dry seasons, however, these
hypothesized canals seemingly would have been of little use for transportation purposes.

Soils
The lush vegetation that mantles Tikal today might suggest an underlying soilscape of
equal fecundity. While the upland soils of the northern Petn are indeed more fertile than
many of their tropical counterparts, these upland soils are typically thin and fragile while
those in the bajos are often beset with drainage problems. How the Maya adapted to and
maintained this essential resource is only partly understood.
One of the unanticipated discoveries from the recent analyses of Tikal arose from the
soil studies described by Tankersley and others (Chapter 9). They determined that many of
the sediments from the bajos, reservoirs, and aguadas were composed largely of degraded
volcanic ejecta. This influx of volcanic ash on a sporadic, but consistent, basis likely
contributed to the long-term subsistence success of the ancient Maya at Tikal. As pointed
out by Lentz et al. (Chapter 8), the Late Classic Maya likely operated under some kind of
a short-fallow agricultural regime in their outfields. Owing to the variety of leguminous
crops used by the Tikal farmers (see Lentz et al., Chapter 8), the problem of nitrogen
deficiency seemingly was addressed. The depletion of other kinds of minerals, especially
phosphorus, would have been much more problematic without any evident means of
amending the soil to compensate. The apparently semiregular influx of minerals from
volcanic ash helps to explain how the ancient Maya were able to farm their lands
intensively over many centuries without a precipitous decline in yields.
In the seasonal wetlands, most agricultural activity occurred in the cumulic soils along
the bajo margins, as reported by Dunning and collaborators (Chapter 6). Ironically, the
colluvial soils that became the focus of Classic period intensive agriculture were at least in
part created by accelerated soil erosion from adjacent uplands in the Preclassic when the
land was first cleared of forest and cultivated. In excavations near the Aguada de Terminos
on the fringe of the Bajo de Santa Fe, ancient topsoils were buried by Late Classic
colluvium and offered a clear sign of ancient forest clearance and extensive agriculture.
These observations coincide with other studies that indicate bouts of erosion in some
sectors of the Tikal area (Hodell et al. 2000). The evidence of soil erosion during the Late
Classic period is not all uniform at Tikal; however, we do find indications that the Maya
found ways to address the erosion problem. For example, if we look at the sediment core
from Corriental Reservoir, it seems that the runoff into this feature, which has side wall
slopes of 10 in some areas, had a sediment buildup of less than 2 mm yr-1 during the Late
Classic period and this was primarily aeolian, or wind-blown sediment. There had to have
been some mechanism to control erosion to prevent runoff from ruining the utility of the
reservoir. Quite probably the Late Classic Maya at Tikal were able to control erosion, at
least in the areas surrounding the reservoir watersheds, by using plantings of perennial
shrubs and cultivated trees observed in the paleoethnobotanical record (Lentz et al.,
Chapter 8).

Social Aspects of Land Management and Resource

Availability
Although our studies focus primarily on natural resource utilization practices at Tikal,
explaining how these resources were managed has occupied our attention, as well. We
have sought examples from the ethnographic literature to help interpret our findings and
place those explanations of ancient Maya land management in a plausible social context.
The practice of first occupancy among the Maya has been recorded by many observers
from the early contact period (Tozzer 1941) to more modern times (Redfield and Villa
Rojas 1962; Roys 1939). This principle of Maya land tenure, as described by McAnany
(1995), prescribes that the earliest occupant of a tract of land will lay claim to it and that
claim is inheritable. Thus, the earliest families or lineages to settle at Tikal likely claimed
the most fertile agricultural land and the most productive forest tracts. Through time, these
early claimants, made prosperous by the productivity of their landholdings and easy access
to scarce water resources, could have become the ruling elite of Classic times. Land within
each lineage would be passed down unevenly, creating a significant power and wealth
disparity within lineages. These lineages, particularly the most influential members of
those lineages, would be expected to exert control over the way the agricultural plots and
forests were managed. If these early stakeholders had protected some of their lands as
ancestral or inherited forests (Roys 1939), this would explain where the old-growth
timbers used to construct the great temples of the Late Classic period were obtained.
These forests likely were managed by royal administrators and were part of the inherited
niches discussed by Webster and Murtha (Chapter 10). Why the change in management
of these ancestral groves took place in the seventh century is an essential question.
Perhaps the great military victories of the late seventh and early eighth centuries were
sufficient cause or justification to disturb the old ancestors and honor the more recent
ones.
As Tikal populations grew, parvenus from existing lineages or more recent immigrants
could claim what land was unoccupied, most likely the less desirable tracts. During the
population expansion of the Classic period, the bajo areas may have represented the last
remaining lands to be claimed. This certainly represents a heterarchical arrangement as
power differentials existed within each lineage and between lineages, based on access to
agricultural land and forest tracts and probably many other factors, as well. The royal
lineage possibly controlled some of the most valuable land, but judging from the
settlement pattern at Tikal, lands appear to have been divided among many powerful
lineages. As Webster and Murtha note (Chapter 10), the social structure at Tikal seemed to
foster competition among farmers and rivalry for status among the lineage heads or elites.
As the Classic period developed, demographic pressure for agricultural land
undoubtedly increased. Some have argued that land management likely was taken over by
the ruling elite (Haviland 2003). The organization of Tikal has been viewed by some as
hierarchical (McAnany and Gallareta Negrn 2010); because there were many levels of
social strata that could be differentiated by domestic architecture, grave goods, access to
trade items, and dietary differences (Haviland and Moholy-Nagy 2003. We observed that
the land in different sectors of Tikal was managed in different ways, suggesting a broader,
more diffuse power scheme in terms of administration. For example, the pocket bajo south
of the Perdido Reservoir appears to have been associated with the central hydraulic system
of the polity, as a sector where irrigation was practiced, seemingly under control of the
ruling elite.
At the other end of the power spectrum, agriculture at places like Terminos in the Bajo
de Santa Fe with its crude chert terraces were more likely under control of local
inhabitants associated with low-status lineages. The home gardens surrounding the many
plazuela groups of Tikal were probably under more localized control, as well. The
intensive agricultural fields in the upland areas, the lifeblood of the Tikal economy,
however, would have required more coordinated management to avoid boundary disputes
and other internecine conflicts. Lineage heads likely would have been involved to help
make this part of agricultural production operate more efficiently. In consequence, the
complex layers of social stratification at Tikal, with lineage heads retaining influence in
some but not all sectors of the economy, would have crosscut the power base of the ruling
elite. Thus, the collective social organization at Tikal is perhaps best described as having
heterarchies of both power and scale (Crumley 1995; Scarborough et al. 2003).
Population indications from the assessment of available forest resources and
agricultural land clearly show that the 1,111 km within the Voronoi diagram defined for
the Late Classic period could have sustainably supported 40,000 to 45,000 inhabitants in
the entire extractive zone, if all lands were exploited to their maximum potential. This
finding suggests that other studies (e.g., Culbert et al. 1990) examining population
parameters at Tikal may have overestimated the number of Late Classic inhabitants. If the
larger population estimates for Tikal were in fact correct, then the Late Classic Maya were
living well past the point of sustainability. Alternatively, a more populous Tikal would
have needed to import food in tribute or in trade (cf. Pyburn 1998). Some scholars,
however (e.g., Drennan 1984; Webster 2002), have asserted that this kind of support,
powered largely by human porters, was not feasible in this context for any length of time.

Drought
There are those (e.g., McAnany and Gallareta Negrn 2010; Demarest 2004) who have
dismissed the drought hypothesis for various reasons, but recent speleothem data (e.g.,
Medina-Elizade et al. 2010; Kennett et al. 2012) strongly indicate that there was an
extensive, multidecadal drought in the Maya Lowlands in the middle ninth century (AD
820870). This defined period of drought strikingly coincides with the dramatic
population decline at Tikal and the construction of the last dated monument, Stela 11, in
AD 869 (Houston and Inomata 2009). Curiously, there is also evidence that extensive
forest clearance may negatively impact rainfall patterns (Georgescu et al. 2013; Cook et
al. 2012). Forest clearance in the tropics dramatically affects transpiration rates and
hydrologic cycling so it is quite possible that the Tikal Maya, who cleared approximately
60 percent of their upland forest, may have unknowingly contributed to the devastating
droughts of the Late Classic period.
When the multiyear droughts of the mid-ninth century commenced, the polity of Tikal
must have been subjected to considerable strain. Water levels in the reservoirs surely
dropped. Maize and other crops likely failed and, as described by Webster and Murtha
(Chapter 10), the ancient Maya had no major granaries or other collective means of storing
large quantities of food. Food storage was likely on a local level with each farmer storing
his own surplus (cf. Scarborough and Valdez 2014). This was a tenuous arrangement, at
best, with a probable storage technology that could preserve food for around three years
(Smyth 1991). Even if the Tikal Maya were fully prepared for a drought, their food supply
could have lasted no longer than three years, assuming that the reservoirs retained water
that long. A drought longer than three years, as hypothesized, would have resulted in
repeated crop failures, food shortages, and, ultimately, exodus from Tikal.
Resupplying a city like Tikal from an area that was not drought stricken would have
been functionally impossible, not only because of the energy requirements of human
porters, which limit the distance they can carry a load without eating all the cargo, but also
because any sort of food importation effort would have been overwhelmed by famished
inhabitants along the way (Gill 2000: 79).
Notably, interpretation of the archaeological deposits defining the final Terminal
Classic activities in epicentral Tikal has varied, with some scholars seeing a quick and
violent end (e.g., Suhler and Freidel 2003) and others a lingering death (e.g., Coe 1990).
Our findings from the Aguada de Terminos indicate that, in some peripheral areas,
occupation persisted for a time into the Postclassic period, but these were largely
exceptional cases. We note also that the occupants of El Zotz, the smaller polity to the
west of Tikal, were able to survive the mid-ninth century droughts, probably because of
their oversized reservoir and a greater resiliency in an agricultural and silvicultural system
that was not heavily overtaxed by a dense population. El Zotz was eventually abandoned,
but this phenomenon probably had less to do with immediate environmental concerns and
more to do with the general collapse of communities around them and the disruption of
trade routes across the Petn.
Undoubtedly, the calamity of water and food shortage at Tikal had a deleterious effect
on social order, and this societal breakdown may have been the immediate cause of the
abandonment. Indeed, what kind of leadership could have managed an overpopulated city,
already at or exceeding its agricultural potential and then confronted with an extensive
drought? It was the ruling elites responsibility to supplicate the deities to ensure that
catastrophic events, such as droughts, did not occur. Obviously, by Terminal Classic times
it was beyond their power to ameliorate the effects of drought and public confidence
surely waned accordingly. In any case, the city was experiencing abandonment by the time
Stela 11 was erected, and soon thereafter the once-glorious palaces of the Central
Acropolis were occupied by squatters (Martin 2003).
Our data strongly indicate that rapid depopulation of Tikal resulted from the
intersection of a subsistence strategy that was already at, or perhaps even beyond, the edge
of sustainability coinciding with a multidecadal drought. The Late Classic Maya had
maximized the productive potential of their constructed niche in the heart of the Petn rain
forest during the relatively stable periods of rainfall in the seventh and eighth centuries.
When the periods of low rainfall became manifest in the ninth century, their systems of
water, land, and forest management were unable to meet the needs of a large populace.
This was the root cause of the collapse of the Late Classic occupation at Tikal: People
experienced dramatic shortages of basic necessities, they died or they left, with only vastly
reduced numbers persisting in places with more dependable water sources.
This is not a simple tale of environmental determinism; on the contrary, it is a complex
record of human agency that unfolded over several centuries. The ancient Maya of Tikal
engineered a constructed niche that was designed to provide sustenance and fuel for as
many people as possible using existing technology. When prolonged drought occurred in
the ninth century, however, they had wrung all of the resilience out of their anthropogenic
ecosystem and reached a tipping point. Moreover, it is quite possible that their forest
clearance activities even helped to bring about the period of attenuated rainfall. The
environmental changes that transpired exceeded their ability to compensate, and this
impasse brought about the end of the Late Classic splendor at Tikal.
One cannot think about the collapse of the great city of Tikal without considering the
fate of our own culture and how this may represent a harbinger of our future prospects. We
see some of the same signs all around us, with widespread forest clearance, erosion of soil
resources, exponential population growth, and anthropogenically induced climate change.
We take solace in our technological prowess and the resilience it provides; we can move
food and fuel from one part of the globe to another when droughts and other natural
disasters occur. But what will happen if our current world population of seven billion
continues to expand? Odds are that the resiliency of the environment will be far less than
it is today. There may have been those among the seventh century inhabitants of Tikal who
foresaw that their Malthusian trajectory of an increasing population operating on a finite
resource base could not be sustained, but their society as a whole was unable to respond in
any meaningful way. We have shown how clever and adaptive the Maya were at Tikal.
Indeed, they sustained an expanding population for many centuries through the changes
they made to their engineered landscape to accommodate civilization and its many
resilient institutions through time. Given the longitudinal successes at Tikal, why would
the Maya believe that a drought could undo their intergenerationally confirmed
accomplishments? Unlike the ancient Maya, we hope the spectacular achievements of our
modern society can avoid their hubris and we can learn valuable lessons from the not-so-
distant past.
 References

Abrams, E. M., and Rue, D. J. 1988. The Causes and Consequences of Deforestation
among the Prehistoric Maya. Human Ecology 16(4): 37795.
Acevedo-Rodrguez, P. 2003. Flora Neotropica Monograph 87. Melicocceae
(Sapindaceae): Melicoccus and Talisia. New York: New York Botanical Garden.
Adams, R. E. W. 1971. The Ceramics of Altar de Sacrificios. Paper of the Peabody
Museum 63(1), Cambridge, MA: Harvard University.
Adams, R. E. W. 1980. Swamps, Canals, and the Location of Ancient Maya Cities.
Antiquity 54: 20614.
Adams, R. E. W. 1981. Settlement Patterns of the Central Yucatan and Southern
Campeche Regions. In Ashmore, W. (ed.), Lowland Maya Settlement Patterns, pp. 211
59. Albuquerque: University of New Mexico Press.
Adams, R. E. W. 1982. Rank Size Analysis of Northern Belize Maya Sites. In Hester, T.
R. (ed.), Archaeology at Colha, Belize: The 1981 Interim Report. San Antonio:
University of Texas.
Adams, R. E. W. 1993. Rebuttal to Pope and Dahlin: 2. Journal of Field Archaeology 20:
383.
Adams, R. M. 1978. Strategies of Maximization, Stability, and Resilience in
Mesopotamian Society, Settlement, and Agriculture. Proceedings of the American
Philosophical Society 12:(5).
gren, G. I., Bosatta, E., Jrme, B. 1996. Isotope Discrimination during Decomposition
of Organic Matter: A Theoretical Analysis, Soil Science Society of America Journal 60:
11216.
AIRSAR. 2008. Jet Propulsion Laboratory Airborne Synthetic Aperture Radar.
Available on the Internet, http://airsar.jpl.nasa.gov/, accessed August 21, 2012.
Aiuvalasit, M. J., Neely, J. A., and Bateman, M. D. 2010. New Radiometric Dating of
Water Management Features at the Prehistoric Purrn Dam Complex, Tehuacn Valley,
Puebla, Mxico. Journal of Archaeological Science 37: 120713.
Akpinar-Ferrand, E., Dunning, N., Lentz, D., and Jones, J. 2012. Aguadas as Water
Sources at Southern Maya Lowland Sites. Ancient Mesoamerica 23: 85101.
Alcorn, J. B. 1984. Huastec Mayan Ethnobotany. Austin: University of Texas Press.
Altschuler, M. 1958. On the Environmental Limitations on Mayan Cultural Development.
Southwestern Journal of Anthropology 14(2): 18998.
Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. 1990. Basic Local
 Alignment Search Tool. Journal of Molecular Biology 215: 40310.
Alvarado, G. D., and Herrera, I. R. 2001. Mapa Fisiogrfico-Geomorfolgico de la
Repblica de Guatemala a escala 1:250,000, Memoria Tcnica: Ministerio de
Agricultura, Ganadera, y Alimentacin. Guatemala City: Programa de Emergencia por
Desastres Naturales.
Alves, E. S., and Angyalossy-Alfonso, V. 2000. Ecological Trends in the Wood Anatomy
of Some Brazilian Species. 1. Growth Rings and Vessels. IAWA Journal 21(1): 330.
Andrieu, C. 2013. Late Classic Maya Lithic Production and Exchange at Rio Bec and
Calakmul, Mexico. Journal of Field Archaeology 38(1): 2137.
Anselmetti, F. S., Hodell, D. A., Ariztegui, D., Brenner, M., and Rosenmeier, M. 2007.
Quantification of Soil Erosion Rates Related to Ancient Maya Deforestation. Geology
35(10): 91518.
Arnold, D. E. 2008. Social Change and the Evolution of Ceramic Production and
Distribution in a Maya Community. Boulder: Colorado University Press.
Arnold, J. E., and Ford, A. 1980. A Statistical Examination of Settlement Patterns at Tikal,
Guatemala. American Antiquity 45(4): 71326.
Arnold, J. G., and Allen, P. M. 1996. Estimating Hydrologic Budgets for Three Illinois
Watersheds. Journal of Hydrology 176(14): 5777.
Arnold, P. J. III 1991. Domestic Ceramic Production and Apatial Organization: A
Mexican Case Study in Ethnoarchaeology. New York: Cambridge University Press.
Arredondo Leiva, E., and Houston, S. (eds.) 2008. Proyecto Arqueolgico El Zotz
Informe No. 1: Temporada de Campo. Guatemala: Institute of Anthropology and
History. Mesoweb. Available on the Internet, http://mesoweb.com/zotz/El-Zotz-
2008.pdf
Arrhenius, O. 1921. Species and Area. Journal of Ecology 9: 959.
Arvigo, R., and Balick, M. 1993. Rainforest Remedies: 100 Healing Herbs of Belize. Twin
Lakes, WI: Lotus Press.
Atran, S. 1999. Classification of Useful Plants by the Northern Petn Maya. In C. White
(ed.), Reconstructing Ancient Maya Diet, pp. 1959. Salt Lake City: University of Utah
Press.
Atran, S., Lois, X., Ek, E. U. 2004. Plants of the Petn Itza Maya: Plantas de Los Maya
Itza del Petn. Memoirs of the Museum of Anthropology No. 38. Ann Arbor:
University of Michigan.
Balesdent, J., and Balabane, M. 1996. Major Contribution of Roots to Soil Carbon Storage
inferred from Maize Cultivated Soils. Soil Biology and Biochemistry 28: 12613.
Balick, M. J., Nee, M. H., and Atha, D. E. 2000. Checklist of the Vascular Plants of Belize
with Common Names and Uses. Bronx: New York Botanical Garden Press.
Ballantyne, M. R. 2009. Miami Fort: An Ancient Hydraulic Structure. M. A. thesis.
Department of Anthropology. University of Cincinnati.
Ball, J. W., and Kelsay, R. G. 1992. Prehistoric intrasettlement land use and residual soil
phosphate levels in the Upper Belize Valley, Central America. In Killion, T. W. (ed.),
Gardens of Prehistory: The Archaeology of Settlement Agriculture in Greater
Mesoamerica, pp. 23462. Tuscaloosa: University of Alabama Press.
Beach, T., and Dunning, N. 1997. An Ancient Maya Reservoir and Dam at Tamarindito,
Petn, Guatemala. Latin American Antiquity 8(1): 209.
Beach, T., Luzzadder-Beach, S., Hageman, J., and Lohse, J. C. 2002. Upland Agriculture
in the Maya Lowlands: Ancient Conservation in Northwestern Belize. The
Geographical Review 92(3): 37297.
Beach, T., Dunning, N., Luzzadder-Beach, S., and Scarborough, V. L. 2003. Depression
Soils in the Lowland Tropics of Northwestern Belize: Anthropogenic and Natural
Origins. In Gomez-Pompa, A., Allen, M., and Fedick, S. (eds.), Lowland Maya Area:
Three Millennia at the Human-Wildland Interface, pp. 13974. Binghamton: Haworth
Press.
Beach, T., Dunning, N., Luzzadder-Beach, S., Cook, D. E., and Lohse, J. 2006. Impacts of
the Ancient Maya on Soils and Soil Erosion in the Central Maya Lowlands. Catena
65(2): 16678. doi:10.1016/j.catena.2005.11.007.
Beach, T., Luzzadder-Beach, S., Dunning, N., and Cook, D. 2008. Human and Natural
Impacts on Fluvial and Karst Systems in the Maya Lowlands. Geomorphology 101(1
2): 30131.
Beach, T., Luzzadder-Beach, S., Dunning, N., Jones, J. G., Lohse, J. C., Guderjan, T. H.,
Bozarth, S., Millspaugh, S., and Bhattacharya, T. 2009. A Review of Human and
Natural Changes in Maya Lowlands Wetlands over the Holocene. Quaternary Science
Reviews 28(1718): 171024.
Beach, T., Luzzadder-Beach, S., Dunning, N., Terry, R., Houston, S., and Garrison, T.
2011. Carbon Isotopic Ratios of Wetland and Terrace Soil Sequences in the Maya
Lowlands of Belize and Guatemala. Catena 85(2): 10918.
Beach, T., Luzzadder-Beach, S., Houston, S., Garrison, T., Bozarth, S., and Doyle, J.
Submitted. Soils and Water Chemistry and Paleoecology at El Zotz and Its
Surroundings. In Garrison, T. G., and Houston, S (eds.), An Inconstant Landscape: The
Archaeology of El Zotz, Guatemala. Boulder: University Press of Colorado.
Becker, M. J. 2003. A Classic-Period Barrio Producing Fine Polychrome Ceramic at Tikal,
Guatemala. Ancient Mesoamerica 14: 95112.
Beltrn Frias, L. 1987. Subsistencia y Aprovechamiento del Medio. In L. Manzanilla (ed.),
Coba, Quintana Roo: Anlisis de Dos Unidades Habitacionales Mayas, pp. 21332.
Mexico City: Universidad Nacional Autnoma de Mxico.
Bender, M. M. 1968. Mass Spectrometric Studies of Carbon-13 Variations in Corn and
Other Grasses. Radiocarbon 10: 46872.
Benjamin, T. J., Montaez, P. I., Jimnez, J. J. M., and Gillespie, A. R. 2001. Carbon,
Water and Nutrient Flux in Maya Homegardens in the Yucatn Peninsula of Mxico.
Agroforestry Systems 53: 10311.
Becquelin, P., and Michelet, D. 1994. Demografa en la Zona Puuc: el recurso del mtodo.
Latin American Antiquity 5: 289311.
Bindeman, I. N. 2006. The Secrets of Super Volcanoes. Scientific American 294(6): 36
43.
Blair, N., Leu, A., Munos, E., Olsen, J., Kwong, E., Des Marais, D. 1985. Carbon Isotopic
Fractionation in Heterotrophic Microbial Metabolism. Applied and Environmental
Microbiology 50: 9961001.
Bongers, F., Popma, J., del Castillo, M., and Carabias, J. 1988. Structure and Floristic
Composition of the Lowland Rain Forest of Los Tuxtlas, Mexico. Vegetatio 74: 5580.
Boots, B. N. 1980. Weighting Thiessen Polygons. Economic Geography 56(3): 24859.
Boserup, E. 1965. The Conditions of Agricultural Growth: The Economics of Agarian
Change Under Population Pressure. London: Earthscan.
Boutton, T. 1996. Stable Carbon Isotope Ratios of Soil Organic Matter and Their Uses as
Indicators of Vegetation and Climate Change. In Boutton, T. W., and Yamasaki, S.
(eds.), Mass Spectrometry of Soils, pp. 4782. New York: Marcel Dekker.
Boutton, T. W., Archer, S. R., Midwood, A. J., Zitzer, S. F., Bol, R. 1998. d13C Values of
Soil Organic Carbon and their Use in Documenting Vegetation Change in a Subtropical
Savanna Ecosystem. Geoderma 82: 541.
Bowen, G. J., and Beerling, D. J. 2004. An Integrated Model for Soil Organic Carbon and
CO2: Implications for Paleosol Carbonate pCO2 Paleobarometry. Global
Biogeochemical Cycles 18(1): GB1026. doi:10.1029/2003GB002117.
Bozarth, S. R., and Guderjan, T. H. 2004. Biosilicate Analysis of Residue in Maya
Dedicatory Cache Vessels from Blue Creek, Belize. Journal of Archaeological Science
31(2): 20515.
Bray, J. R., and Curtis, J. T. 1957. An Ordination of the Upland Forest Communities of
Southern Wisconsin. Ecological Monographs 27: 32549.
Breedlove, D. E., and Laughlin, R. M. 2000. The Flowering of Man: A Tzotzil Botany of
Zinacantan. Washington, DC Smithsonian Institution Press.
Brenner, M., Rosenmeier, M. F., Hodell, D., and Curtis, J. 2002. Paleolimnology of the
Maya Lowlands. Ancient Mesoamerica 13: 14157.
Brewer, J. L. 2007. Understanding the Role of a Small Depression in Ancient Maya Water
Management at the Medicinal Trail Site, Northwest Belize. Unpublished Masters
thesis. University of Cincinnati.
Brewer, S. W., and Webb, M. A. H. 2002. A Seasonal Evergreen Forest in Belize:
Unusually Hightree Species Richness for Northern Central America. Biological Journal
of the Linnaean Society 138: 27596.
Broda, J., Carrasco, D., and Matos, M. E. 1982. The Great Temple of Tenochtitlan, Center
and Periphery in the Aztec World. Berkeley: University of California Press.
Brokaw, N. V. 1987. Gap-Phase Regeneration of Three Pioneer Species in a Tropical
 Forest. Journal of Ecology 75: 919.
Brokaw, N. V. L., and Mallory, E. P. 1993. Vegetation of the Rio Bravo Conservation and
Management Area, Belize. Manomet: Manomet Bird Observatory.
Bronson, B. 1966. Roots and the Subsistence of the Ancient Maya. Southwestern Journal
of Anthropology 22(3): 25179.
Brookfield, H. C. 1972. Intensification and Disintensification in Pacific Agriculture: A
Theoretical Perspective. Pacific Viewpoint 13(1): 3048.
Brookfield, H. C. 1984. Intensification Revisited. Pacific Viewpoint 25: 1544.
Brown, S., Gillespie, A. J. R., and Lugo, A. E. 1989. Biomass Estimation Methods for
Tropical Forests with Applications to Forest Inventory Data. Forest Science 35(4): 881
902.
Brown, S., and Lugo, A. E. 1992. Above Ground Biomass Estimates for Tropical Moist
Forests of the Brazilian Amazon. Interciencia 17(1): 818.
Brown, L. A., and Emery, K. F. 2008. Negotiations with the Animate Forest: Hunting
Shrines in the Guatemalan Highlands. Journal of Archaeological Method and Theory
15: 30037.
Brune, G. M. 1953. Trap Efficiency of Reservoirs. Transactions of American Geophysical
Union 34(3): 40717.
Buckley, B. M., Anchukaitis, K. J., Penny, D., Fletcher, R., Cook, E. R., Sano, M., Nam,
L. C., Wichienkeeo, A., Minh, T. T., and Hong, T. M. 2010. Climate as a contributing
factor in the demise of Angkor, Cambodia. Proceedings of the National Academy of
Science, USA 107(15) 67486752.
Bullard, W. 1960. The Maya Settlement Pattern in Northeastern Petn, Guatemala.
American Antiquity 25: 35572.
Burnett, R. 2009. Stable Carbon Isotope Evidence of Ancient Maya Agriculture at Tikal,
Guatemala. M. A. Thesis, Department of Plant and Wildlife Sciences. Salt Lake City:
Brigham Young University.
Burnett, R. L., Terry, R. E., Alvarez, M., Balzotti, C., Murtha, T., Webster, D., Silverstein,
J. 2012. The Ancient Agricultural Landscape of the Satellite Settlement of Ramonal
near Tikal, Guatemala. Quarternary International 265: 10115.
doi:10.1016/j.quaint.2011.03.002.
Burnham, K. P., and Overton, W. S. 1979. Robust Estimation of Population Size When
Capture Probabilities Vary among Animals. Ecology 60: 92736.
Buttles, P. J. 1992. Small Finds in Context: The Preclassic Miscellaneous Artifact of
Colha. Unpublished MA Thesis, Department of Anthropology. Austin: University of
Texas.
Buttles, P. J. 2002. Material and Meaning: A Contextual Examination of Select portable
Material Culture from Colha, Belize. Unpublished Ph. D. dissertation, Department of
Anthropology. Austin: University of Texas.
Cabrera, T. 2000. Resultados preliminares de las investigaciones realizadas en el Grupo
Maler, Yaxha, Petn. In Laporte, J. P., Escobedo, H., Arroyo B., and de Suasnvar A. C.
(eds.), XIII Simposio de Investigaciones Arqueolgicas en Guatemala, pp. 35473.
Guatemala: Museo Nacional de Arqueologa y Etnologa.
Campbell, D. G., Ford, A., Lowell, K. S., Walker, J. Lake, J. K., Ocampo-Raeder, C.,
Townesmith, A., and Balick, M. 2006. The Feral Forests of the Eastern Petn. In Bale
W., and Erickson C. L. (eds.), Time and Complexity in Historical Ecology, pp. 2156.
New York: Columbia University Press.
Campbell, D. G., Guittar, J., and Lowell, K. S. 2008. Are Colonial Pastures the Ancestors
of the Contemporary Maya Forest? Journal of Ethnobiology 28(2): 27889.
Cannon, A. 2000. Settlement and Sea-Levels on the Central Coast of British Colombia:
Evidence from Shell Midden Cores. American Antiquity 65(1): 6777.
Cannon, M. J., and Cannon, J. F. M. 1989. Central American Araliaceae: A Precursory
Study for the Flora Mesoamericana. Bulletin of the British Museum (Natural History).
Botany 19: 561.
Canuto, M. A., and Yaeger, J. 2000. The Archaeology of Communities: A New World
Perspective. London: Routledge.
Carlquist, S. 2001. Comparative Wood Anatomy: Systematic, Ecological and Evolutionary
Aspects of Dicotyledon Wood, 2nd ed. New York: Springer.
Carpenter, L. B., Feinman, G. M., and Nicholas, L. M. 2012. Spindle Whorls from El
Palmillo: Economic Implications. Latin American Antiquity 23(4): 381400.
Carr, C. 2013. Map of the Ruins of Tikal, El Petn, Guatemala and Georeferenced
Versions of the Maps Therein. Tikal Report No. 11 (tDAR ID: 390922). Available on the
Internet, http://core.tdar.org/project/390922, accessed June 3rd, 2013.
Carr, C., Weaver, E., Dunning, N., and Scarborough, V. 2011. Evaluacin de la Exactitud
de Los Mapas de Tikal de La Universidad de Pennsylvania, por GPS y Estacin Total
(Accuracy assessment of the Penn Project maps of Tikal, by GPS and Total Station). In
Lentz, D., Ramos, C., Dunning, N., Scarborough, V., and Grazioso, L. (eds.), Proyecto
de Silvicultura y Manejo de Aguas de Los Antiguos Mayas de Tikal. Report Submitted
to the Instituto de Antropologia y Historia, Guatemala.
Carr, D. L. 2008. Farm Households and Land Use in a Core Conservation Zone of the
Maya Biosphere Reserve, Guatemala. Human Ecology 36: 23148.
Carr, M. J., Feigenson, M. D., and Patino, L. C. 2007. Petrology and Geochemistry of
Lavas. In Bundschuh, J., and Alvarado, G. E. Central America: Geology, Resources and
Hazards, pp. 56590. London: Taylor and Francis.
Carr, R. F., and Hazard, J. E. 1961. Map of the Ruins of Tikal, El Petn, Guatemala. Tikal
Report 11. University Museum Monographs. Philadelphia: University Museum,
University of Pennsylvania.
Cassens, D. L., and Miller, R. B. 1981. Wood Anatomy of the New World Pithecellobium
(sensu lato). Journal of the Arnold Arboretum 62: 14.
Chafetz, H. S., and Zhang, J. 1998. Authigenic Euhedral Mega Quartz Crystals in a
Quaternary Dolomite. Journal of Sedimentary Research 68(5): 9941000.
Chao, A. 1984. Nonparametric Estimation of the Number of Classes in a Population.
Scandinavian Journal of Statistics 11: 26570.
Chase, A. F., and Chase, D. Z. 1998. Scale and Intensity in Classic Period Maya
Agriculture: Terracing and Settlement at the Garden City of Caracol, Belize. Culture
& Agriculture 20(23): 6077. doi:10.1525/cag.1998.20.2-3.60.
Chase, A. F., Chase, D. Z., Fisher, C. T., Leisz, S. J., and Weishampel, J. F. 2012.
Geospatial Revolution and Remote Sensing LiDAR in Mesoamerican Archaeology.
Proceedings of the National Academ of Science USA 109(32): 1291621.
Chase, A. F., and Scarborough, V. L. (eds.). 2014. The Resilience and Vulnerability of
Ancient Landscapes: Transforming Maya Archaeology through IHOPE. Archeological
Papers of the American Anthropological Association, No. 22. Hoboken, NJ: Wiley
Periodicals.
Cheetham, D. 2010. Corn, Calanders, and Cooking: Early Maize Processing in the Maya
Lowlands and its Implications. In Staller, J. E., and Carrasco, M. (eds.), Pre-Columbian
Foodways: Interdisciplinary Approaches to Food, Culture, and Markets in Ancient
Mesoamerica, pp. 34568. Berlin: Springer.
Chilton, E. S. 1999. Material Meanings: Critical Approaches to the Interpretation of
Material Culture. Foundations of Archaeological Inquiry. Salt Lake City: The
University of Utah Press.
Cliff, M. B., and Crane, C. J. 1989. Changing Subsistence Economy at a Late Preclassic
Community. In McAnany P., and Isaac B. (eds.), Prehistoric Maya Economies of Belize,
pp. 295324. Greenwich, CT: JAI Press.
Clinebell, R. R. II, Phillips, O. L., Gentry, A. H., Starks, N., and Zuuring, H. 1995.
Prediction of Neotropical Tree and Liana Species Richness from Soil and Climatic
Data. Biodiversity and Conservation 4: 5690.
Coe, M. D. 1992. Breaking the Maya Code. New York: Thames and Hudson.
Coe, S. D. 1994. Americas First Cuisines. Austin: University of Texas Press.
Coe, W. R. 1962. A summary of excavation and research at Tikal, Guatemala: 195661.
American Antiquity 27(4): 479507.
Coe, W. R. 1963. A Summary of Excavation and Research at Tikal, Guatemala: 1962.
Estudio de Cultura Maya, Vol. III, pp. 4164. Mexico.
Coe, W. R. 1965. Tikal, Guatemala and Emergent Maya Civilization. Science 147: 1401
19.
Coe, W. R. 1967. Tikal: A Handbook of the Ancient Maya Ruins. Philadelphia: University
Museum, University of Pennsylvania.
Coe, W. R., and Haviland, W. 1982. Tikal Report No. 12: Introduction to the Archaeology
of Tikal, Guatemala. Philadelphia: The University Museum.
Coe, W. R., Shook, E. M., and Satterthwaite, L. 1986. Tikal Report No. 6: The Carved
Wooden Lintels of Tikal. In Tikal Reports 511. The University Museum. Philadelphia:
The University of Pennsylvania.
Coe, W. R. 1990. Excavations in the Great Plaza, North Terrace, and North Acropololis of
Tikal. Tikal Reports No. 14. Univesity Museum. Philadelphia: University of
Pennsylvania.
Colwell, R. K., and Coddington, J. A. 1994. Estimating Terrestrial Biodiversity through
Extrapolation. Philosophical Transactions: Biological Sciences 345: 10118.
Cooke, C. W. 1931. Why the Mayan Cities of the Peten District, Guatemala, Were
Abandoned. Journal of the Washington Academy of Sciences 21: 2837.
Cook, B. I. Anchukaitis, K. J. Kaplan, J. O., Puma, M. J., and Kelley, M. 2012. Pre-
Columbian Deforestation as an Amplifier of Drought in Mesoamerica. Geophysical
Research Letters 39. doi:10.1029/2012GL052565.
Corrado, K. 2014. An Analysis of the Corriental Reservoir Sediments in Relation to
Ancient Maya Land Management and Sustainability Practices at Tikal, Guatemala.
M.A. Thesis, Department of Anthropology. University of Cincinnati.
Corzo Mrquez, A. R., and Schwartz, N. B. 2008. Traditional Home Gardens of Petn,
Guatemala: Resource Management, Food Security, and Conservation. Journal of
Ethnobiology 28: 30517.
Costanza, R., dArge, R., Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K.,
Naeem, S. T., ONeil, R., Paruelo, J., Raskin, R., Sutton, P., and van den Belt, M. 1997.
The Value of the Worlds Ecosystems, Services and Natural Capital. Nature 387: 253
60.
Cottam, G., and Curtis, J. T. 1956. Measures in Phytosociological Sampling. Ecology 37:
45160.
Cowgill, U. M. 1962. An Agricultural Study of the Southern Maya Lowlands. American
Anthropologist 64(2): 27386.
Cowgill, U. M., and Hutchinson, G. E. 1963. El Bajo de Santa Fe. Transactions of the
American Philosophical Society 53(7): 344.
Crumley, C. L. 1995. Heterarchy and the Analysis of Complex Societies. Archaeological
Papers of the American Anthropological Association 6(1): 15.
Culbert, T. P. 1977. Maya Development and Collapse: An Economic Perspective. In
Hammond, N. (ed.), Social Process in Maya Prehistory, pp. 50930. New York:
Academic Press.
Culbert, T. P., Kosakowsky, L. J., Fry, R. E., and Haviland, W. A. 1990. The Population of
Tikal, Guatemala. In Culbert, T. P., and Rice, D. S. (eds.), Precolumbian Population
History in the Maya Lowlands, pp. 10321. Albuquerque: New Mexico University
Press.
Culbert, T. P., and Rice, D. S. 1990. Precolumbian Population History in the Maya
Lowlands. Albuquerque: University of New Mexico Press.
Culbert, T. P., Levi, L. J., and Cruz, L. 1990. Lowland Maya Wetland Agriculture: The Rio
Azul Agronomy Program. In Clancy, F. S., and Harrison, P. D. (eds.), Vision and
Revision in Maya Studies, pp. 11524. Albuquerque: University of New Mexico Press.
Culbert, T. P. 1993. The Ceramic of Tikal: Vessels from the Burials, Caches, and
Problematic Deposits.Tikal Report No. 14. Philadelphia: The University Museum,
University of Pennsylvania.
Culbert, T. P. 2003. The Ceramics of Tikal. In Sabloff, J. A. (ed.), Tikal: Dynasties,
Foreigners, and Affairs of state Advancing Maya Archaeology, pp. 4782. Santa Fe:
School of American Research Press.
Culbert, T. P. 2009. Los Materiales Arqueolgicos. In Lentz, D. L., Grazioso L. S.,
Dunning, N. P., and Scarborough, V. L. (eds.), Proyecto de Silvacultura y Manejo de
Aguas de los Antiguos Mayas de Tikal: Temporada de 2009. Direccin Patrimonio
Cultural y Natural de Guatemala.
Curtis, J. T., and McIntosh, R. P. 1951. An Upland Forest Continuum in the Prairie-Forest
Border Region of Wisconsin. Ecology 32: 47696.
Dahlin, B. H., and Litzinger, W. J. 1986. Old Bottles, New Wine: The Function of
Chultuns in the Maya Lowlands. American Antiquity 51: 72136.
Dahlin, B. H. 1989. Notes on the Paleoecology of the El Mirador Bajo. Unpublished
manuscript in possession of the author.
Dahlin, B. H., Dahlin, A. 1994. Platforms in the Aklache at El Mirador, Peten, Guatemala
and Their Implications. Geoarchaeology 9: 20337.
Dahlin, B. H., and Chase, A. F. 2014. A Tale of Three Cities: Effects of the AD 536 Event
in the Maya Heartland. In Iannone, G. (ed.), The Great Maya Droughts in Cultural
Context, pp. 127-156. Boulder: University of Colorado Press.
Das, R., Lawrence, D., DOdorico, DeLonge, M. 2011. Impact of Land Use Change on
Atmospheric Inputs in a Tropical Dry Forest. Journal of Geophysical Research 116,
G01027. doi:10.1029/2010JG001403,2011
Dawson, I. K., Waugh, R., Simons, A. J., and Powell, W. 1997. Simple Sequence Repeats
Provide a Direct Estimate of Pollen-Mediated Gene Dispersal in the Tropical Tree
Gliricidia sepium. Molecular Ecology 6: 17983.
Deevey, E. S., Rice, D. S., Rice, P. M., Vaughn, H. H., Brenner, M., and Flannery, M. S.
1979. Maya Urbanism: Impact on a Tropical Karst Environment. Science 206: 298306.
Demarest, A. 2004. Ancient Maya: Rise and Fall of a Rainforest. New York: Cambridge
University Press.
Denevan, W. M. 1992. The Pristine Myth: The Landscape of the Americas in 1492.
Annals of the Association of American Geographers 82(3): 36985.
Denevan, W. 2006. Pre-European Forest Cultivation in Amazonia. In W. Bale and C. L.
Erickson (eds.), Time and Complexity in Historical Ecology, pp. 15363. New York:
Columbia University Press.
Dick, C. W., Jones, F. A., Hardy, O. J., and Petit, R. 2008. Spatial Scales of Seed and
 Pollen-Mediated Gene Flow in Tropical Forest Trees. Tropical Plant Biology 1: 2033.
Dickson, B. D. 1980. Ancient agriculture and population at Tikal, Guatemala: an
application of linear programming to the simulation of an archaeological problem.
American Antiquity 45: 697712.
Diefendorf, A. F., Freeman, K. H., and Wing, S. L. 2012. Distribution and Carbon Isotope
Patterns of Diterpenoids and Triterpenoids in Modern Temperate C3 Trees and Their
Geochemical Significance. Geochimica et Cosmochimica Acta 85: 342356
Diemont, S. A., Martin, J. F., Levy-Tacher, S. I., Nigh, R. B., Lopez, P. R., and Golicher,
J. D. 2006. Lacandon Maya Forest Management: Restoration of Soil Fertility Using
Native TreeSpecies. Ecological Engineering 28: 20512.
Dietler, M., and Herbich, I. 1998. Habitus, Techniques, Style: An Integrated Approach to
the Social Understanding of Material Culture Boundaries. In Stark, M. T. (ed.), The
Archaeology of Social Boundaries, Smithsonian Series in Archaeological Inquiry,
Smithsonian University Press.
Dobson, N. 1973. A History of Belize. London: Longman Caribbean.
Downie, J. 1959. An Economic Policy for British Honduras. Belize: H.M. Government.
Doyle, J. A., Houston, S. D., Garrison, T. G., and Romn, E. 2011. Al alcance de la vista
de Mundo Perdido? La planificacin urbana y el abandono abrupto de El Palmar, Petn,
Guatemala. In Arroyo, B., Aragn, L. P., Palma, A. L., and Arroyave, A. L. (eds.), XXIV
Simposio de Investigaciones Arqueolgicas de Guatemala, pp. 4556. Guatemala:
Ministerio de Cultura y Deportes.
Doyle, J. A. 2012. Re-Group on E-Groups: Monumentality and Early Centers in the
Middle Preclassic Maya Lowlands. Latin American Antiquity 23(4).
Doyle, J., Garrison, T., and Houston, S. 2012. Watchful Realms: Integrating GIS Analysis
and Political History in the Southern Maya Lowlands. Antiquity 86(333): 792807.
Doyle, J. A., Houston, S. D., Romn, E., Garrison, T. G., Beach, T., and Luzzadder-Beach,
S. 2012. El Palmar, Petn, Guatemala y las vas de la poca preclsica en el lmite sur
de la Meseta Krstica Central. In Arroyo, B., Paiz, L., y Meja, H. (eds.), XXV Simposio
de Investigaciones Arqueolgicas en Guatemala, pp. 83744. Guatemala: Museo
Nacional de Arqueologa y Etnologa.
Drake, J. B., Knox, R. G., Dubaya, R. O., Clark, D. B., Condit, R., and Blair, B. J.,
Hofton, M. 2003. Above-Ground Biomass Estimation in Closed Canopy Neotropical
Forests Using Lidar Remote Sensing: Factors Affecting the Generality of Relationships.
Global Ecology & Biogeography 12(2): 14759.
Drennan, R. D. 1984. Long-Distance Movement of Goods in the Mesoamerican Formative
and Classic. American Antiquity 49: 2743.
Drooker, P. B. 2001. Material Culture and Perishability. In Drooker, P. B. (ed.), Fleeting
Identities: Perishable Material Culture in Archaeological Research, Occasional Papers
No. 28, pp. 315. Center for Archaeological Investigations, Carbondale: Southern
Illinois University.
Dull, R. A., Southon, J. R., and Sheets, P. 2001. Volcanism, Ecology, and Culture: A
Reassessment of the Volcano Ilopango TBJ Eruption in the Southern Maya Relm. Latin
American Antiquity 12: 2544.
Dumont, H. J. 2005. Biodiversity: A Resource with a Monetary Value? Hydrobiologia
542: 1114.
Dunning, N. P. 1992. Lords of the Hills: Ancient Maya Settlement in the Puuc Region,
Yucatn, Mexico. Madison, WI: Prehistory Press.
Dunning, N. P., and Beach, T. 1994. Soil Erosion, Slope Management, and Ancient
Terracing in the Maya Lowlands. Latin American Antiquity 5(1): 5169.
Dunning, N., Beach, T., and Rue, D. 1997. The Paleoecology and Ancient Settlement of
the Petexbatun Region, Guatemala. Ancient Mesoamerica 8(02): 25566.
doi:10.1017/S0956536100001711.
Dunning, N. P., Beach, T. P., Farrell, P., Luzzadder-Beach, S. 1998. Pre-Hispanic
Agriculture and Adaptive Systems in the Maya Lowlands. Culture and Agriculture
20(23): 87101.
Dunning, N. P., Scarborough, V. L., Valdez, F. Jr., Luzzadder-Beach, S., Beach, T., and
Jones, J. G. 1999. Temple Mountains, Sacred Lakes, and Fertile Fields: Ancient Maya
Landscapes in Northwestern Belize. Antiquity 73(281): 65060.
Dunning, N. P., and Beach, T. 2000. Stability and Instability in Pre-Hispanic Maya
Landscapes. In Lentz, D (ed.), An Imperfect Balance: Landscape Transformations in the
Precolumbian Americas, pp. 179202. New York: Columbia University Press.
Dunning, N. P., Luzzadder-Beach, S., Beach, T., Jones, J. G., Scarborough, V., and
Culbert, T. P. 2002. Arising from the Bajos: The Evolution of a Neotropical Landscape
and the Rise of Maya Civilization. Annals of the Association of American Geographers
92(2): 26783.
Dunning, N. P., Culbert, T. P., Fialko, V., Sever, T., Jones, J. G., Grazoiso, L., and Irwin,
D. 2002. The Bajo de Santa Fe Revisited. Paper presented at the 67th Annual Meeting
of the Society for American Archaeology, Denver.
Dunning, N., Beach, T., Luzzadder-Beach, S., and Jones, J. G. 2003. Physiography,
Habitats, and Landscapes of the Three Rivers Region. In Scarborough, V., Valdez, F. Jr.,
and Dunning, N. P. (eds.), Heterarchy, Political Economy, and the Ancient Maya: The
Three Rivers Region of the East-Central Yucatan Peninsula, pp. 1424. Tucson:
University of Arizona Press.
Dunning, N. P., and T. Beach. 2004. Fruit of the Luum: Lowland Maya Soil Knowledge
and Agricultural Practices. Mono y Conejo 2: 125.
Dunning, N., Beach, T., and Luzzadder-Beach, S. 2006. Environmental Variability among
Bajos in the Southern Maya Lowlands and Its Implications for Ancient Maya
Civilization and Archaeology. In Lucero, L., and Fash, B. (eds.), Pre-Columbian Water
Management, pp. 11133. Tempe: University of Arizona.
Dunning, N. P., Jones, J., Chmilar, J., and Blevins, J. 2006. Investigaciones ambientales y
 geoarqueologicas in las alrededoras de San Bartolo. In Saturno, W. and Urquizu, M.
(eds.), Proyecto Arqueolgico Regional San Bartolo, Informe Preliminar #3, Tecera
Temporada 2005. Guatemala: Instituto Nacional de Antropologia e Historia.
Dunning, N. P., and Griffin, R. 2008. Investigaciones de canales lineares in el Bajo de
Azcar. In Urquiz, Mnica and Saturno, William (eds.) Proyecto Arqueolgico San
Bartolo: Informe Preliminar No. 7, Sptima Temporada 2008. Report to Guatemala:
Instituto de Antropologa e Historia de Guatemala.
Dunning, N. P., and Beach, T. P. 2010. Farms and Forests: Spatial and Temporal
Perspectives on Ancient Maya Landscapes. In Martini, I. P., and Chesworth, W. (eds.),
Landscapes and Societies, pp. 36989. New York: Springer.
Dunning, N. P., Griffin, R., Jones, J. G., Carr, C., Lane, B., Thomas, B., and Weaver, E.
2009. Investigaciones de Geoarqueloga y Paleoambeinte en la Zona Mayor de Tikal. In
Lentz, D. L., Grazioso L. S., Dunning, N. P., and Scarborough, V. L. (eds.), Proyecto de
Silvacultura y Manejo de Aguas de los Antiguos Mayas de Tikal: Temporada de 2009,
pp. 5065. Direccin Patrimonio Cultural y Natural de Guatemala.
Dunning, N. P., and Houston, S. 2011. Chan Ik: Hurricanes as a Disruptive Force in the
Maya Lowlands. In Persson, B. L., and Isendahl, C. (eds.), Ecology, Power, and
Religion in Maya Landscapes, pp. 4959. Mckmhl: Verlag Anton Saurwein.
Dunning, N. P., Jones, J. G., Carr, C., and Magee, K. 2011. Investigaciones de
Geoarqueloga y Paleoambeinte en la Zona Mayor de Tikal. In Lentz, D., Ramos, C.,
Dunning, N., Scarborough, V., and Grazioso, L. Sierra, Proyecto de Silvacultura y
Manejo de Aguas de los Antiguos Mayas de Tikal: Temporada de 2010, pp. 10112.
Direccin Patrimonio Cultural y Natural de Guatemala.
Dunning, N. P., Beach, T., Luzzadder-Beach, S. 2012. Kax and Kol: Collapse and
Resilience in Lowland Maya Civilization. Proceedings of the National Academ of
Science USA 109: 36527.
Dunning, N. P., Wahl, D., Beach, T. P., Jones, J. G., Luzzadder-Beach, S., and
McCormick, C. 2014. The End of the Beginning: Environmental Instability and Human
Response in the Late Preclassic East-Central Yucatan Peninsula. In Iannone, G. (ed.),
The Great Maya Droughts in Cultural Context, pp. 107-126. Boulder: University of
Colorado Press.
Dvorak, W. S., Hamrick, J. L., and Gutirrez, J. L. 2005. The Origin of Caribbean Pine in
the Seasonal Swamps of the Yucatn. International Journal of Plant Sciences 166: 985
94.
Eberl, M., lvarez, M., and Terry, R. E. 2012. Chemical Signatures of Middens at a Late
Classic Maya Residential Complex, Guatemala. Geoarchaeology 27: 42640.
Ehleringer, J. R. 1991. 13C/12C Fractionation and Its Utility in Terrestrial Plant Studies.
In Coleman, D. C., and Fry, B. (eds.), Carbon Isotope Techniques, pp. 187200. San
Diego: Academic Press.
Emery K. F., Wright, L. E., and Schwarcz, H. P. 2000. Isotopic Analysis of Ancient Deer
Bone: Biotic Stability in Collapse Period Maya Land-Use. Journal of Archaeological
 Science 27(5): 53750.
Environmental Systems Research Institute. 2009. ArcGIS 9.3 Help: What Is a Functional
Surface? Available on the Internet,
http://webhelp.esri.com/arcgisdesktop/9.3/index.cfm?
TopicName=What%20is%20a%20functional%20surface%3F
Environmental Systems Research Institute. 2012. ArcGIS 9.3 Help: Hydrologically
Correct Surfaces: Topo to Raster. Available on the Internet,
http://webhelp.esri.com/arcgisdesktop/9.3/index.cfm?TopicName=
Hydrologically%20correct%20surfaces%3A%20Topo%20to%20Raster
Erickson, C. L. 2006. The Domesticated Landscapes of the Bolivian Amazon. In Bale,
W., and Erickson, C. L. (eds.), Time and Complexity in Historical Ecology, pp. 23678.
New York: Columbia University Press.
Erickson, C. L., and Balee, W. 2006. The Historical Ecology of a Complex Landscape in
Bolivia. In Bale, W., and Erickson, C. L. (eds.), Time and Complexity in Historical
Ecology, pp. 187234. New York: Columbia University Press.
Fairbank, J. K., and Goldman, M. 2006. China: A New History. Cambridge, MA: Harvard
University Press.
Farquhar, G. D., O Leary, M. H., and Berry, J. A. 1982. On the Relationship between
Carbon Isotope Discrimination and the Intercellular Carbon Dioxide Concentration in
Leaves. Australian Journal of Plant Physiology 9(2): 12137.
Faust, B. B. 2001. Maya Environmental Successes and Failures in the Yucatan Peninsula.
Environmental Science and Policy 4(45): 15369.
Fedick, S. L., and Ford, A. 1990. The Prehistoric Agricultural Landscape of the Central
Maya Lowlands: An Examination of Local Variability in a Regional Context. World
Archaeology 22(01): 1833
Fedick, S. L. 1996. The Managed Mosaic: Ancient Maya Agriculture and Resource Use.
Salt Lake City: University of Utah Press.
Fedick S. L., Delgadillo, M. L. F., Sedov, S., Rebolledo, E. S., and Mayorga, S. P. 2008.
Adaptation of Maya Homegardens by Container Gardening in Limestone Bedrock
Cavities. Journal of Ethnobiology 28: 290304.
Fedick, S. L. 2010. The Maya Forest: Destroyed or Cultivated by the Ancient Maya?
Proceedings of the National Academ of Science USA 107: 9534.
Fernandez, F. G., Johnson, K. D., Terry, R. E., Nelson, S., Webster, D. 2005. Soil
Resources of the Ancient Maya at Piedras Negras, Guatemala. Soil Science Society of
America Journal 69: 202032.
Fialko, V. 1988. Mundo Perdido, Tikal: Un Ejemplo de Complejos de Conmemoracin
Astronmica. Mayab 4: 1324.
Fialko, V. 2000. Recursos Hidraulicos en Tikal y sus Periferias. In LaPorte, J. P.,
Escobeda, H., Arroyo, B., and, Suasnavar, A. C. (eds.), XIII Simposio de
Investigaciones Arqueolgicas en Guatemala, pp. 55665. Guatemala: Museo Nacional
 de Arqueologa.
Fialko, V. 2001. Investigaciones Arqueolgicas en el Bajo de Santa Fe y la Cuenca del
Ro Holmul, Petn. II. Guatemala City: Programa National Tikal-Tringulo.
Fialko, V. 2005. Diez aos de investigaciones arqueolgicas en la Cuenca del Ro Holmul,
region norteste del Petn. In LaPorte, J. P., Arroyo, B., and Meja, H. (eds.), XVIII
Simposio de Investigacioies Arqueolgicas en Guatemala, pp. 24460. Guatemala:
Museo Nacional de Arqueologa e Etnologa.
Fialko, V., Folan, W., and, Gunn, J. 2006. Investigations in the Intersite Areas between
Yaxha, Nakum, and Tikal. In Los Investigadores de la Cultura Maya 13(2): 487501.
Campeche: Universidad Autnoma de Campeche.
Figueiredo, E., Paiva, J., Stvart, T., Oliveira, F. & Smith, G. F. 2011. Annotated
Catalogue of the Flowering Plants of So Tom and Prncipe. Bothalia 41: 4182.
Fitzsimmons, J. 2012. The Living and the Dead. In Nichols, D., and Pool, C. (eds.), The
Oxford Handbook of Mesoamerican Archaeology, pp. 77684. New York: Oxford
University Press.
Fletcher, R. 2009. Low-Density Agrarian-Based Urbanism: A Comparative View. Institute
of Advanced Study, Durham University. Insights 2(4): 119.
Floyd, P. A., and Winchester, J. A. 1978. Identification and Discrimination of Altered and
Metamorphosed Volcanic Rocks Using Immobile Elements. Chemical Geology 21(3):
291306.
Folan, W. J., Fletcher, L. A., and Kintz, E. R. 1979. Fruit, Fiber, Bark, and Resin: Social
Organization of a Maya Urban Center. Science 204(4394): 697701.
Folorunso, A. E. 2011. Diversity in the Stem Anatomy and Tissues of Several Species of
Annona (Annonaceae) in Nigeria. Notulae Scientia Biologicae 3(3): 2032.
Foody, G. M., Boyd, D. S., and Cutler, M. E. J. 2003. Predictive relations of tropical forest
biomass from Landsat TM data and their transferability between regions, Remote
Sensing of Environment 85: 46374.
Ford, A. 1986. Population Growth and Social Complexity: An Examination of Settlement
and Environment in the Central Maya Lowlands. Anthropological Research Papers, p.
35. Tempe: Arizona State University.
Ford, A., and Glicken, H. 1987. The Significance of Volcanic Ash Tempering in the
Ceramics of the Central Maya Lowlands. In Rice, P. M., and Sharer, R. J. (eds.), Maya
Ceramic. Papers from the 1985 Maya Ceramics Conference. Oxford: BAR International
Series.
Ford, A. 1991. Economic Variation of Ancient Maya Residential Settlement in the Upper
Belize River Area. Ancient Mesoamerica 2(1): 3546.
Ford, A., and Fedick, S. L. 1992. Prehistoric Settlement Patterns in the Upper Belize River
Area: Initial Results of the Belize River Archaeological Settlement Survey. Journal of
Field Archaeology 19(1): 3549.
Ford, A., and Rose, W. I. 1995. Volcanic Ash in Ancient Maya Ceramics of the Limestone
 Lowlands: Implications for Prehistoric Volcanic Activity in the Guatemala Highlands.
Journal of Volcanology and Geothermal Research 66(1): 14962.
Ford, A. 2008. Dominant Plants of the Maya Forest and Gardens of El Pilar: Implications
for Paleoenvironmental Reconstructions. Journal of Ethnobiology 28: 17999.
Ford, A., and Emery, K. 2008. Exploring the Legacy of the Maya Forest. Journal of
Ethnobiology 28: 14753.
Ford, A., and Nigh, R. 2009. Origins of the Maya Forest Garden: Maya Resource
Management. Journal of Ethnobiology 29: 21336.
Freeman K. H., Mueller K. E., Diefendorf, A. F., Wing, S. L. and Koch P. L. 2011.
Clarifying the Influence of Water Availability and Plant Types on Carbon Isotope
Discrimination by C3 Plants. Letter. Proceedings of the National Academies of Science
108: E59E60.
French, K. D., and Duffy, C. J. 2010. Prehispanic Water Pressure: A New World First.
Journal of Archaeological Science 37(5): 102732.
French, K. D., Duffy, C. J., and Bhatt, G. 2012. The Hydroarchaeological Method: A Case
Study at the Maya Site of Palenque. Latin American Antiquity 23(1): 2950.
Frias, B. L. 1987. Subsistencia y aprovechamiento del medio. In Manzanilla, L. (ed.),
Coba, Quinana Roo Anlisis de Dos Unidades Habitacionales Mayas, pp. 21332.
Mexico City: Universidad Nacional Autonoma.
Fry, R. E. 1969. Ceramics and Settlement in the Periphery of Tikal. Ph.D. dissertation,
Department of Anthropology. Tucson: University of Arizona.
Fry, R. E. 2003. The Peripheries of Tikal. In Sabloff, J. A. (ed.), Tikal: Dynasties,
Foreigners and Affairs of State, pp. 14370. Santa Fe: SAR Press.
Gallopin, G. G. 1990. Water Storage Technology at Tikal, Guatemala. M.A. Thesis,
Department of Anthropology. University of Cincinnati.
Garber, J. F. 1981. Material Culture and Patterns of Artifact Consumption and Disposal
atthe Maya Site of Cerros in Northern Belize. Unpublished Ph.D. dissertation, Southern
Methodist University. Dallas: Texas.
Garmin. 2007. Garmin GPSMAP 60CSx Owners Manual. Olathe, Kansas, Garmin
International.
Garrido Lpez, J. L., Houston, S., and Romn, E. (eds.) 2011. Proyecto Arqueolgico El
Zotz Informe No. 5 Temporada. Report Submitted to Guatemala: The Institute of
Anthropology and History. Mesoweb. Available on the Internet,
http://mesoweb.com/zotz/El-Zotz-2010.pdf
Garrido Lpez, J. L., Houston, S., Romn, E., and Garrison, T. (eds.) 2012. Proyecto
Arqueolgico El Zotz Informe No. 6 Temporada 2011. Report Submitted to
Guatemala: The Institute of Anthropology and History. Mesoweb. Available on the
Internet, http://mesoweb.com/zotz/El-Zotz-2011.pdf
Garrison, T., and Dunning, N. 2009. Settlement, Environment, and Politics in the San
Bartolo Xultun Territory, El Petn, Guatemala. Latin American Antiquity 20: 52552.
Garrison, T. G., and Garrido Lpez, J. L. 2012 Investigaciones intersitios en el
reconocimiento de las Tierras Bajas. In Arroyo, B., Paiz, L., and Meja, H. (eds.), XXV
Simposio de Investigaciones Arqueologicas en Guatemala, pp. 102944. Guatemala:
Museo Nacional de Arqueologa y Etnologa.
Gasson, P., Warner, K., and Lewis, G. 2009. Wood Anatomy of Caesalpinia s.s.,
Coulteria, Erythrostemon, Guilandina, Libidibia, Mezoneuron, Poincianella, Pomaria
and Tara (Leguminosae, Caesalpinioideae, Caesalpinieae). IAWA Journal 3: 24776.
Gee, G. W., and Bauder, J. W. 1986. Particle-Size Analysis. In Klute, A (ed.), Methods of
Soil Analysis. Part 1, 2nd ed., pp. 383411. Madison, WI: Soil Science Society of
America.
George, D., and Mallery, P. 2011. SPSS for Windows Step by Step: A Simple Guide and
Reference 18.0 Update. Boston, MA: Allyn & Bacon/Pearson.
Georgescu, M., Lobell, D. B., Field, C. B., and Mahalov, A. 2013. Simulated
Hydroclimatic Impacts of Projected Brazilian Sugarcane Expansion. Geophysical
Research Letters 40(5): 9727
Gill, R. B. 2000. The Great Maya Droughts: Water, Life, and Death. Albuquerque:
University of New Mexico Press.
Gillespie, A. R., Knudson, D. M., and Geilfus, F. 1993. The Structure of Four Home
Gardens in the Petn, Guatemala. Agroforestry Systems 24: 15770.
Gillespie, S. 2001. Rethinking Ancient Maya Social Organization: Replacing Lineage
with House. American Anthropologist 102(3): 46784.
Gitzendanner, M. A., and Soltis, P. S. 2000. Patterns of Genetic Variation in Rare and
Widespread Plant Congeners. American Journal of Botany 87: 78392.
Glassman, S., and Annaya, A. 2011. Cities of the Maya in Seven Epochs, 1250 B.C. to
A.D. 1903. Jefferson: McFarland.
Goman, M., and Byrne, R. 1998. A 5000-Year Record of Agriculture and Tropical Forest
Clearance in the Tuxtlas, Veracruz, Mexico. The Holocene 8: 839.
Gmez-Pompa, A. 1987. On Maya Silviculture. Mexican Studies 3(1): 117.
Gmez-Pompa, A., Flores, S., and Sosa, V. 1987. The Pet Kot: A Man-Made Tropical
Forest of the Maya. Interciencia 12: 1015.
Gonzlez-Soberanis, C., and Casas, A. 2004. Traditional Management and Domestication
of Tempesquistle, Sideroxylon palmeri (Sapotaceae) in the Tehuacn-Cuicatln Valley,
Central Mexico. Journal of Arid Environments 59: 24558.
Greller, A. M. 2000. Vegetation in the Floristic Regions of North and Central America. In
Lentz, D. L. (ed.), Imperfect Balance: Landscape Transformations in the Precolumbian
Americas, pp. 3987. New York: Columbian University Press.
Griffin, R. E. 2012. The Carrying Capacity of Ancient Maya Swidden Maize Cultivation:
A Case Study in the Region around San Bartolo, Petn, Guatemala. Ph.D. dissertation,
Department of Anthropology. University Park: Pennsylvania State University.
Grim, R. E. and Gven, N. 1978. Bentonite Geology, Mineralogy, Properties and Uses.
New York: Elsevier.
Groark, K. P. 1997. To Warm the Blood, to Warm the Flesh: The Role of the Steambath in
Highland Maya (Tzeltal-Tzotzil) Ethnomedicine. Journal of Latin American Lore 20(1):
396.
Grube, N., and Paap, I. 2010. Uxul, Petn Campechano: Primera Temporada de Campo.
Los Investigadores de la Cultura Maya 18(2): 724. Campeche: Universidad de
Campeche.
Gunawardana, R. A. L. H. 1978. Hydraulic Engineering in Ancient Sri Lanka: The Cistern
Sluice. In Prematilleke, L., and Indrapala, K. (eds.), Senarat Paranavitana
Commemoration Volume, pp. 6174. Leiden: E. J. Brill.
Gunn, J. D., Folan, W. J., and Robichaux, H. R. 1995. A Landscape Analysis of the
Candelaria Watershed in Mexico: Insights into Paleoclimates Affecting Upland
Horticulture in the Southern Yucatan Peninsular Semi-Karst. Geoarchaeology 10:342.
Gunn, J. D., Gunn, W. J. Folan, C. Isendahl, M. R. Domnguez Carrasco, B. B. Faust, and
B. Volta. 2014. Calakmul: Agent Risk and Sustainability in the Western Maya
Lowlands. In Chase, A. F., and Scarborough, V. L. (eds.), The Resilience and
Vulnerability of Ancient Landscapes: Transforming Maya Archaeology through IHOPE,
Archeological Papers of the American Anthropological Association. Malden, MA:
Wiley Periodicals.
Gutirrez, M., San Miguel-Chavez, R., and Terrazas, T. 2009. Xylem Conductivity and
Anatomical Traits in Diverse Lianas and Small Tree Species from a Tropical Forest of
Southwest Mexico. International Journal of Botany 5: 27986.
Gutirrez-Granados, G., Prez-Salicrup, D. R., and Dirzo, R. 2011. Differential Diameter-
Size Effects of Forest Management on Tree Species Richness and Community
Structure: Implications for Conservation. Biodiversity Conservation 20: 157185.
Halstead, P., and OShea, J. (eds.) 1989. Bad Year Economics: Cultural Responses to Risk
and Uncertainty. Cambridge: Cambridge University Press.
Hammond, N. 2000. The Maya Lowlands: Pioneer Farmers to Merchant Princes. In
Adams, R. E. W., and, Macleod, M. J. (eds.), The Cambridge History of Native Peoples
of the Americas: Mesoamerica, 2 (part 1), pp. 197249. Cambridge: Cambridge
University Press.
Hammond, N., Rose, P., Staneko, J., and Muyskens, D. 1988. Excavation and Survey at
Nohmul, Belize, 1986. Journal of Field Archaeology, V15: 115.
Hansen, R. D., Bosarth, S., Jacob, J., Wahl, D., and Schreiner, T. 2002. Climatic and
Environmental Variability in the Rise of Maya Civilization: A Preliminary Perspective
from the Northern Peten. Ancient Mesoamerica 13: 27395.
Hardesty, D. L. 1972. The Human Ecological Niche. American Anthropologist 74(3):
45866. doi:10.1525/aa.1972.74.3.02a00150.
Hardesty, B. D., Hubbell, S. P., and Bermingham, E. 2006. Genetic Evidence of Frequent
 Long-Distance Recruitment in a Vertebrate-dispersed Tree. Ecology Letters 9: 51625.
Harrison, P. D. 1977. The Rise of the Bajos and the Fall of the Maya. In Hammond, N.
(ed.), Social Process in Maya Prehistory, pp. 469508. New York: Academic Press.
Harrison, P. D. 1978. Bajos Revisited: Visual Evidence for One System of Agriculture. In
Harrison, P. D., and Turner II, B. L. (eds.), Pre-Hispanic Maya Agriculture, pp. 24754.
Albuquerque: University of New Mexico Press.
Harrison, P. D., and Turner, B. L. II. 1978. Pre-Hispanic Maya Agriculture. Albuquerque:
University of New Mexico Press.
Harrison, P. D. 1990. The Revolution in Ancient Maya Subsistence. In Clancy, F., and
Harrison, P. D. (eds.), Vision and Revision in Maya Studies, pp. 99113. Albuquerque:
University of New Mexico Press.
Harrison, P. D. 1992. The Third Revolution: Environment, Population and a Sustainable
World. London: I. B. Tauris.
Harrison, P. D. 1999. The Lords of Tikal: Rulers of an Ancient Maya City. London:
Thames and Hudson.
Harrison, P. D. 2012. The Marvel of Maya Engineering: Water Management at Tikal.
Expedition 54(2): 1926.
Hart, W. J. E., and Steen-McIntyre, V. 1983. Tierra Blanca Joven Tephra from the AD 260
Eruption of the Ilopango Caldera. In Sheets, P. D. (ed.), The Zapotitan Valley of El
Salvador, Austin: University of Texas Press.
Hastorf, C. A. Domesticated Food and Society in Early Coastal Peru. 2006. In Bale, W.,
and Erickson, C. (eds.), Time and Complexity in Historical Ecology, pp. 87126. New
York: Columbia University Press.
Haug, G. H., Gunther, D., Perterson, L. C., Sigman, D. M., Hughen, K. A., and
Aeschlimann, B. 2003. Climate and the Collapse of Maya Civilization. Science 299:
17315
Haviland, W. A. 1967. Stature at Tikal, Guatemala: Implications for Ancient Maya
Demography and Social Organization. American Antiquity 32: 31625.
Haviland, W. A. 1969. A New Population Estimate for Tikal, Guatemala. American
Antiquity 34: 42433.
Haviland, W. A. 1970. Tikal, Guatemala and Mesoamerican Urbanism. World
Archaeology 2: 18698.
Haviland, W. A. 1972. Estimates of Maya Population: Comment on Thompsons
Comments. American Antiquity 37: 26162.
Haviland, W. A. 1985. Excavations in Small Residential Groups of Tikal: Groups 4 F-1
and 4 F-2. Tikal Report No. 19. Philadelphia: University Museum, University of
Pennsylvania.
Haviland, W. 2003. Settlement, Society, and Demography at Tikal. In Sabloff, J. A. (ed.),
Tikal: Dynasties, Foreigners and Affairs of State: Advancing Maya Archaeology, pp.
 11142. Santa Fe: School of American Research Press.
Haviland, W. 2008. Tikal, Guatemala: A Maya Way to Urbanism. In Mustache, A.,
Cobean, R., Cook, G. A., and Hirth, K. (eds.), Urbanism in Mesoamerica, pp. 25984.
Mexico City and University Park: Instituto Nacional de Antropologia e
Historia/Pennsylvania State University.
Haviland, W., and H. Moholy-Nagy. 2003. Distinguishing the High and Mighty from the
Hoi Polloi at Tikal, Guatemala. In Chase, D. Z. and A. F. Chase (eds.) Mesoamerican
Elites: An Archaeological Assessment, pp. 5060. Norman: University of Oklahoma
Press.
Hawkes, J. G. 1983. The Diversity of Crop Plants. Cambridge, MA: Harvard University
Press.
Hayden, B. 1998. Practical and Prestige Technologies: The Evolution of Material Systems.
Journal of Archaeological Method and Theory 5(1): 155.
Heintzelman, C. E., Jr., and Howard, R. A. 1948. The Comparative Morphology of the
Icacinaceae. V. The Pubescence and the Crystals. American Journal of Botany 35: 42
52.
Heltshe, J. F., and Forrester, N. E. 1983. Estimating Species Richness Using the Jackknife
Procedure. Biometrics 39: 112.
Henderson, A., Galeano, G., and Bernal, R. 1995. Field Guide to the Palms of the
Americas. Princeton, NJ: Princeton University Press.
Hermes, B., and, Caldern, Z. 2003. La secuencia de ocupacin prehispnica en Nakum:
Una visin preliminar. In Laporte, J. P., Arroyo, B., Escobedo, H., and Meja, H. (eds.),
XVI Simposio de Investigaciones Arqueolgicas en Guatemala, pp. 30519. Guatemala:
Museo Nacional de Arqueologa y Etnologa.
Hernandez-Sefanoni, J. L., and Ponce-Hernandez, R. 2002. Mapping the Spatial
Distribution of Plant Diversity Indices in a Tropical Forest using Multi-Spectral
Satellite Image Classification and Field Measurements. Biodiversity and Conservation
13: 25992621.
Hester, T. R. 1985. Late Classic Early Postclassic Transitions: Archaeological
Investigations at Colha, Belize. Final Performance Report to the National Endowment
to the Humanities, Grants RO 2053483 and RO-20755. San Antonio: The University
of Texas, Center for Archaeological Research.
Hester, T. R., and Shafer, H. J. 1989. The Ancient Maya Craft Community at Colha,
Belize and Its External Relationships. Texas Papers on Latin America, pp. 8911.
Institute of Latin American Studies. Austin: University of Texas.
Hester, T. R., and Shafer, H. J. 1994. The Ancient Maya Craft Community at Colha,
Belize and Its External Relationships. In Schwartz, G. M., and Falconer, E. S. (eds.),
Archaeological Views from the Countryside: Village Communities in Early Complex
Societies, pp. 4863. Washington, DC: Smithsonian Institution Press.
Higbee, E. 1948. Agriculture in the Maya Homeland. Geographical Review 38(3): 457
 64.
Hill, A. F. 1937. Economic Botany: A Textbook of Useful Plants and Plant Products.
McGraw-Hill.
Hodell, D. A., J. H. Curtis, and Brenner, M. 1995. Possible Role of Climate in the
Collapse of Classic Maya Civilization. Nature 375: 3914.
Hodell, D. A., Brenner, M., and Curtis, J. H. 2000. Climate Change in the Northern
American Tropics and Subtropics since the Last Ice Age: Implications for Environment
and Culture. In Lentz, D. L. (ed.), Imperfect Balance: Landscape Transformations in
the Precolumbian Americas, pp. 1338. New York: Columbia University Press.
Hodell, D. A., Brenner, M., Curtis, J. H., and Guilderson, T. 2001. Solar Forcing of
Drought Frequency in the Maya Lowlands. Science 292: 136770.
Hodell, D. A., Brenner, M., and Curtis, J. H. 2005. Terminal Classic Drought in the
Northern Maya Lowlands inferred from Multiple Sediment Cores in Lake
Chichancanab (Mexico). Quaternary Science Review 24: 141327
Hodell, D. A., and Guilderson, T. 2001. Solar Forcing of Drought Frequency in the Maya
Lowlands. Science 292: 136770.
Hodge, T. A. 2002. Roman Aqueducts and Water Supply. London: Duckworth
Archaeology.
Holderidge, L. R., Genke, W. C., Hatheway, W. H., Laing, T., and, Tosi, W. A., Jr. 1971.
Forest Environments in Tropical Life Zones: A Pilot Study. Oxford: Pergamon Press.
Hood, A. N. 2012. Testing the Veracity of Paleoethnobotanical Macroremain Data: A
Case Study from the Cern Site, El Salvador. Thesis. Department of Anthropology.
University of Cincinnati.
Hosler, D., Sabloff, J. A., and Runge, D. 1977. Simulation Model Development: A Case
Study of the Maya Collapse. In Hammond, N. (ed.), Social Process in Maya Prehistory,
pp. 55390. New York: Academic Press.
House, P. R., Lagos-Witte, S., Ochoa, L., Torres, C., Meja, T., and Rivas, M. 1995.
Plantas Medicinales Comunes de Honduras. Tegucigalpa: Universidad Nacional
Autnoma de Honduras.
Houston, S. D., and Inomata, T. 2009. The Classic Maya. New York: Cambridge
University Press.
Houston, S. D. 2011. In the Shadows of a Giant: Archaeology of El Zotz, Guatemala.
Final Report to the National Endowment for the Humanities on Grant RZ-50680-07.
Howe, H. F., and Smallwood, J. 1982. Ecology of Seed Dispersal. Annual Review of
Ecology and Systematics 13: 20128.
Hubbell, S. P., and Foster, R. B. 1986. Commonness and Rarity in a Neotropical Forest:
Implications for Tropical Tree Conservation. In M. E. Soule (ed.), Conservation
Biology: The Science of Scarcity and Diversity, pp. 20531. Sunderland, MA: Sinauer.
Huff, W. D., Bergstrm, S. M., Kolata, D. R., and Muftuoglu, E. 1999. K-bentonite Bed
 Preservation and Its Event Stratigraphic Significance. Acta-Universitatis Carolinae
Geologica 43: 4913.
Huff, W. D., Bergstrm, S. M., and Kolata, D. R. 2000. Silurian K-bentonites of the
Dnestr Basin, Podolia, Ukraine, Journal of the Geological Society, London 157(2): 493
504.
Huff, W. D. 2008. Ordovician K-bentonites: Issues in Interpreting and Correlating Ancient
Tephras. Quaternary International 178(1): 27687.
Hurlbert, S. R. 1971. The Nonconcept of Species Diversity: A Critique and Alternative
Parameters. Ecology 52: 57786.
Hurst, W. J., Martin, R. A., Jr., Tarka, S. M., Jr., and Hall, G. D. 1989. Authentication of
Cacao in Ancient Mayan Vessels Using High-Performance Liquid Chromatographic.
Journal of Chromatography 466: 27989.
IAWA Committee.1989. IAWA List of Microscropic Features for Hardwood Identification.
IAWA Journal 10: 219332.
Iglesias Ponce de Len, M. J. 1987. Excavaciones en el grupo habitacional 6D-V, Tikal,
Guatemala. Ph.D. dissertation. Madrid: Universidad Complutense.
Inomata, T., Triadan, D., and Pinto, E. 2010. Complete, Reconstructible, and Partial
Vessels. In Inomata, T., and Triadan, D (eds.), Burned Palaces and Elite Residences of
Aguateca: Excavations and Ceramics. Salt Lake City: University of Utah Press.
InsideWood. 2004-onwards. Available on the Internet,
http://insidewood.lib.ncsu.edu/search.
Isendahl, C., Dunning, N. P., Fedick, S. L., Gunn, J. D., Iannone, G., Lucero, L., and
Scarborough, V. L. In Press. Applied Perspectives on Prehispanic Water Management in
the Maya Lowlands: Which Are the Insights for Future Water Security? In Isendahl, C.,
and Stump, D. (eds.), Handbook of Historical Ecology and Applied Archaeology.
Oxford: Oxford University Press.
Islebe, G. A., Hooghiemstra, H., Brenner, M., Curtis, J. H., and, Hodell, D. A. 1996. A
Holocene Vegetation History from Lowland Guatemala. Holocene 6(3): 26571.
Jaccard, P. 1908. Nouvelles recherches sur la distribution florale. Bulletin de la Socite
Vaudense des Sciences Naturelles 44: 22370.
Jackson, S. E. 2013. Politics of the Maya Court: Hierarchy and Change in the Late
Classic Period. Norman: University of Oklahoma Press.
Jacob, J. S. 1995. Archaeological Pedology in the Maya Lowlands. In Pedological
Perspectives, Archaeological Research. Soil Science Society of America. Special
Publication No. 44, pp. 5180. Madison, WI: Soil Science Society of America.
Johnson, K. D., Wright, D. R., and Terry, R. E. 2007a. Application of Carbon Isotope
Analysis to Ancient Maize Agriculture in the Petn Region of Guatemala.
Geoarchaeology: An International Journal 22: 31336.
Johnson, K. D., Terry, R. E., Jackson, M. W., Golden, C. 2007b. Ancient Soil Resources
of the Usumacinta River Region, Guatemala. Journal of Archaeological Science 34:
 111729.
Johnson, S. 2006. The Ghost Map:The Story of Londons Most Terrifying Epidemic and
How It Changed Science, Cities and the Modern World. London: Riverhead Books.
Johnston, K. J., and Gonlin, N. 1998. What Do Houses Mean? Approaches to the Analysis
of Classic Maya Commoner Residence. In Houston, S. D. (ed.), Function and Meaning
in Classic Maya Architecture. Washington, DC: Dumbarton Oaks Research Library and
Collection.
Johnstone, K. J. 2004. Lowland Maya Water Management Practices: the Household
Exploitation of Rural Wells. Geoarchaeology 19(3): 26592.
Jones, C. 1996. Excavations in the East Plaza of Tikal. Tikal Report 16. Philadelphia:
University Museum Publications, University of Pennsylvania.
Jones, J. G. 1994. Pollen Evidence for Early Settlement and Agriculture in Northern
Belize. Palynology 18(1): 20511.
Jones, M. 1952. Current Reports No. 1: Map of the Ruins of Mayapn, Yucatan, Mexico.
In Weeks, J (2009). The Carnegie Maya II, Current Reports 19521957. Boulder:
University Press of Colorado.
Junker, L. L. 1999 Raiding, Trading, and Feasting: the Political Economy of Philippine
Chiefdoms. Honolulu: University of Hawaii Press.
Kendal, Jeremy, Tehrani, J. J., and Laland, Kevin N. 2011. Human Niche Construction in
Interdisciplinary Focus. Philosophical Transactions of the Royal Society (Series B)
366(1566): 78592.
Kennett, D. J., Breitenbach, S. F. M., Aquino, V. V., Asmerom, Y., Awe, J., Baldini, J. U.
L., Bartlein, P., Culleton, B. J., Ebert, C., Jazwa, C., Macri, M. J., Marwan, N., Polyak,
V., Prufer, K. M., Ridley, H. E., Sodemann, H., Winterhalder, B., Haug, G. H. 2012.
Development and Disintegration of Maya Political Systems in Response to Climate
Change. Science 338(6108): 78891.
Killion, T. 1987. Agriculture and Residential Site Structure among Campesions in
Southern Vera Cruz, Mexico: A Foundation for Archaeological Inference. Ph.D.
dissertation, Department of Anthropology, University of New Mexico. Ann Arbor:
University Microfilms.
Killion, T. W. 1990. Cultivation Intensity and Residential Site Structure: An
Ethnoarchaeological Examination of Peasant Agriculture in the Sierra De Los Tuxtlas,
Veracruz, Mexico. Latin American Antiquity 1(3): 191215.
Kilmartin, J. O. 1924. Year Book 23, Report 38.2. Report on the survey and base-map at
Chichen Itza, Mexico. In Weeks, J., and Hill J. (eds.), The Carnegie Maya Research
Program, 19131957 (2006). Boulder: University Press of Colorado.
Kindt, R. 2003. Exact Species Richness for Sample-based Accumulation Curves.
Manuscript.
Kindt, R., and Coe, R. 2005. Tree Diversity Analysis. A Manual and Software for Common
Statistical Methods for Ecological and Biodiversity Studies. Nairobi: World
 Agroforestry Centre (ICRAF).
Kingsley, M., and Cambranes, R. 2011. Excavaciones en el Grupo Sur (Operacin 6). In
Garrido Lpez, J. L., Houston, S., Romn, E., and Garrison, T. (eds.), Proyecto
Arqueolgico El Zotz Informe No. 5: Temporada de Campo, pp. 16398. Report
submitted to the Guatemala Instituto de Antropologa e Historia. Mesoweb. Available
on the Internet, http://mesoweb.com/zotz/El-Zotz-2010.pdf
Kingsley, M., and Rivas, A. 2012. Excavaciones en el Grupo Sur (Operacin 6). In
Garrido Lpez, J. L., Houston, S., Romn, E., and Garrison, T. (eds.), Proyecto
Arqueolgico El Zotz Informe No. 6: Temporada de Campo., pp. 10738. Report
submitted to the Guatemala Instituto de Antropologa e Historia. Mesoweb. Available
on the Internet, http://mesoweb.com/zotz/El-Zotz-2011.pdf.
Kramer, K. 2005. Maya Children: Helpers at the Farm. Cambridge, MA: Harvard
University Press.
Kribs, D. A. 1959. Commercial Foreign Woods on the American Market. University Park,
PA: Buckhout Laboratory, Department of Botany, Pennsylvania State University.
Kukachka, B. F. 1978. Wood Anatomy of the Neotropical Sapotaceae. Vols. 1, 2, 7, 20, 31.
Madison, WI: Forest Products Laboratory, Forest Service, U.S. Department of
Agriculture.
Kunen, J. L. 2004. Ancient Maya Life in the Far West Bajo: Social and Environmental
Change in the Wetlands of Belize. Anthropological Papers. Tucson: University of
Arizona Press.
Kunen, J. L., Culbert, T. P., Fialko, V., McKee, B. R., and Grazioso, L. 2000 Bajo
Communities: A Case Study from the Central Peten. Culture & Agriculture 22(3): 15
31.
La Torre-Cuadros, M. L. A., and Islebe, G. A. 2003. Traditional Ecological Knowledge
and Use of Vegetation in Southeastern Mexico: A Case Study from Solferino, Quintana
Roo. Biodiversity and Conservation 12: 245576.
Laird, D. A., Barak, P. E., Nater, E. A., and Dowdy, R. H. 1991. Chemistry of Smectitic
and Illitic Phases in Interstratified Soil Smectite. Soil Science Society of America
Journal 55(5): 14991504.
Lambert, J. D. H., and Arnason, J. T. 1982. Ramn and Maya Ruins: An Ecological, Not
an Economic, Relation. Science 216: 2989.
Lane, C. S., Horn, S. P., Mora, C. I. 2004. Stable Carbon Isotope Ratios in Lake and
Swamp Sediments as a Proxy for Prehistoric Forest Clearance and Crop Cultivation in
the Neotropics. Journal of Paleolimnology 32: 37581.
Lane, C. S., Mora, C. I., Horn, S. P., Orvis, K. H. 2008. Sensitivity of Bulk Sedimentary
Stable Carbon Isotopes to Prehistoric Forest Clearance and Maize Agriculture, Journal
of Archaeological Science 35: 211932.
Landes, R. 2012. From Counting Down to Counting Out: On the Relationship between
Apocalyptic and Normal Time in the Western Passion for Precise Time Measurement.
 Paper presented at the Dumbarton Oaks Symposium. The Measure and Meaning of
Time in the Americas, Antony Aveni, Organizer.
Laporte, J. P., and Fialko, V. 1985. Reporte Arqueolgico (19791984): Mundo Perdido y
Zonas de Habitacin, Tikal, Petn. Guatemala City: Ministerios de Educacin y
Comunicaciones, Transporte y Obras Pblicas, Guatemala.
Laporte, J. P. 1987. El Grupo 6C-XVI, Tikal, Petn: un centro habitacional del Clsico
Temprano. In Memorias del Primer Coloquio Internacional de Mayistas, pp. 22144.
Universidad Autnoma de Mxico, Mexico City: Centro de Estudios Mayas.
Laporte, J. P. 1990 New Perspectives on Old Problems: Dynastic References for the Early
Classic at Tikal. In Clancy, Flora, and Harrison, Peter D. (eds.), Vision and Revision in
Maya Studies, pp. 3366 Albuquerque: University of New Mexico Press.
Laporte, J. P. 1993. Architecture and Social Change in Late Classic Maya Society: The
Evidence from Mundo Perdido, Tikal. In Sabloff, J. A., and Henderson, J. S. (eds.),
Lowland Maya Civilization in the Eighth Century A.D., pp. 299317. Washington, DC:
Dumbarton Oaks Research Library and Collection.
Laporte, J. P. 1995 Un Reencuentro con Mundo Perdido, Tikal, Guatemala. Ancient
Mesoamerica 6: 4194.
Laporte, J. P. 1995. Preclsico a Clsico en Tikal: Proceso de Transformacin en Mundo
Perdido. In Nikolai Grube, ed. The Emergence of Lowland Maya Civilization: The
Transition from the Preclassic to the Early Classic. Acta Mesoamericana, Vol. 8, pp.
1733. Verlag Anton Sauerwein: Mockmuhl, Germany, pp. 10316.
Laporte, J. P. 1997 Exploracin y Restauracin en la Gran Pirmide de Mundo Perdido,
Tikal (Estructura 5C-54). In X Simposio de Investigaciones Arqueolgicas en
Guatemala (1996), pp. 32550. Guatemala: Museo Nacional de Arqueologa y
Etnologa.
Laporte, J. P. 2003. Thirty Years Later: Some Results of Recent Investigations in Tikal. In
Jeremy A. Sabloff, ed., Tikal: Dynasties, Foreigners, and Affairs of State, pp. 281318.
Santa Fe, NM: School of American Research, Advanced Seminar Series.
Laurance, W. F., Fearnside, P. M., Laurence, S. G., Delamonica, P., Lovejoy, T. E.,
Chambers, J. Q., Merona, J. M. R., and Gascon, C. 1999. Relationship between Soils
and Amazon Forest Biomass: A Landscape-Scale Study. Forest Ecology and
Management 118(13): 12738.
Laurance, W. F., Williamson, G. B., Delamnica, P., Oliveira, A., Lovejoy, T. E., Gascon,
C., Pohl, L. 2001. Effects of a Strong Drought on Amazonian Forest Fragments and
Edges. Journal of Tropical Ecology 17(6): 77185.
Lawrence, D., DOdorico, P., Diekmann, L., DeLonge, M., Das, R., & Eaton, J. 2007.
Ecological Feedbacks Following Deforestation Create the Potential for a Catastrophic
Ecosystem Shift in a Tropical Dry Forest. PNAS. 104: 20696701.
Laws, W. D. 1961. Investigations of Swamp Soils from the Tintal and Pinal Associations
of Peten, Guatemala. Wrightia 2: 12732.
Lee, D. 2004. Longtime Mason Employee Morris R. Jones to Retire. The Mason Gazette,
May 25, 2004. George Mason University.
Lens, F., Jansen, S., Caris, P., Serlet, L., and Smets, E. 2005. Comparative Wood Anatomy
of the Primuloid Clade (Ericales s.1.) Systematic Botany 30: 16383.
Lens, F., Endress, M. E., Baas, P., Jansen, S., and Smets, E. 2008. Wood Anatomy of
Rauvolfioideae (Apocynaceae): A Search for Meaningful Non-DNA Characters at the
Tribal Level. American Journal of Botany 95: 1199215.
Lentz, D. L. 1991. Maya Diets of the Rich and Poor: Paleoethnobotanical Evidence from
Copn. Latin American Antiquity 2(3): 26987.
Lentz, D. L. 1994. Paleoethnobotanical Evidence for Subsistence Practices and Other
Economic Activities in the Petexbatun Region during the Classic Period. Paper
presented at the 93rd American Anthropological Association Meeting, Atlanta, GA.
Lentz, D. L., Beaudry-Corbett, M., Reyna de Aguilar, M. L., and Kaplan, M. L. 1996.
Foodstuffs, Forests, Fields and Shelter: A Paleoethnobotanical Analysis of Vessel
Contents from the Cern Site, El Salvador. Latin American Antiquity 7(3): 24762.
Lentz, D. L. 1999. Plant Resources of the Ancient Maya: The Paleoethnobotanical
Evidence. In C. White (ed.), Reconstructing Ancient Maya Diets, pp. 318. Salt Lake
City: University of Utah Press.
Lentz, D. L., and Ramrez-Sosa, C. R. 2002. Cern plant resources: Abundance and
diversity. In Sheets, P. D. (ed.), Before the Volcano Erupted: The Cern Village in
Central America, pp. 3342. Austin: University of Texas Press.
Lentz, D. L., Haddad, L., Cherpelis, S., Joo, H. J. M., and Potter, M. 2002. Long-Term
Influences of Ancient Maya Agroforestry Practices on Tropical Forest Biodiversity in
Northwestern Belize. In Stepp, J. R., Wyndham, F. S., Zarger, R. (eds.), Ethnobiology
and Biocultural Diversity: Proceedings of the Seventh International Congress of
Ethnobiology, pp. 43142. Athens: University of Georgia Press.
Lentz, D. L., Yaeger, J., Robin, C., and Ashmore, W. 2005. Pine, Prestige and Politics of
the Late Classic Maya at Xunantunich, Belize. Antiquity 79: 57385.
Lentz, D. L., and Hockaday, B. 2009. Tikal Timbers and Temples: Ancient Maya
Agroforestry and the End of Time. Journal of Archaeological Science 36(7): 134253.
Lentz, D. L., Grazioso, S. L., Dunning, N. P., and Scarborough, V. L. 2009. Proyecto de
Silvicultura y Manejo de Aguas de los Antiguos Mayas de Tikal: Temporada de 2009.
Guatemala City: Report submitted to Direccion de Patrimonio Cultural y Natural.
Lentz, D. L., Ramos, C. E., Dunning, N. P., Scarborough, V. L., and Graziosa Sierra, L.
2011. Proyecto de Silvicultura y Manejo de Aguas de los Antiguos Mayas de Tikal:
Temporada de 2010. Guatemala City: Report submitted to Direccin de Patrimonio
Cultural y Natural.
Lentz, D. L., B. Lane, and K. Thompson. 2014. Food, Farming and Forest Management at
Aguateca. In Inomata, T., and Triadan, D., (eds.), Life and Politics at the Royal Court of
Aguateca: Artifacts, Analytical Data, and Synthesis, pp. 20317. Monographs of the
 Aguateca Archaeological Project First Phase, Vol. 3. Provo: University of Utah Press.
Len H., W. J. 2003. Anatoma de la Madera de 7 Especies del Gnero Annona L.
(Annonaceae). Revista Pittieria 32: 2736.
Len H., W. J. 2005. Ecological Anatomy of Secondary Xylem in a Tropical Dry Forest of
Venezuela. Acta Botnica Venezuelica 28: 25774.
Levasseur, V., and Olivier, A. 2000. The Farming System and Traditional Agroforestry
Systems in the Maya Community of San Jose, Belize. Agroforestry Systems 49: 27588.
Leyden, B. W. 1987. Man and Climate in Maya Lowlands. Quaternary Research 28: 407
14.
Liendo, R. 1999. The Organization of Agricultural Production at a Maya Center: The
Settlement Patterns in the Palenque Region, Chiapas, Mexico. Mexico City: Instituto
Nacional de Antropologa e Historia.
Linares, O. F., Sheets, P. D., and Rosenthal, E. J. 1975. Prehistoric Agriculture in Tropical
Highlands. Science 187(4172): 13745.
Liu, R., Clapp, C. E., Cheng, H. H. 1997. Usefulness of the Carbon-13 Tracer Technique
for Characterizing Terrestrial Carbon Pools. Nutrient Cycling in Agroecosystems 49:
2616
Longacre, W. A. 1991. Ceramic Frequency and Use-Life. Ceramic Ethnoarchaeology, pp.
16281. Tucson: University of Arizona Press.
Lou, B. 1997. Chalpate: anlisis del asentamiento y orientacin de una centro satellite de
Tikal. In LaPorte, J. P., and Escobeda, H. (eds.), X Simposio de Investigaciones
Arqueolgicas en Guatemala, pp. 37380. Guatemala: Museo Nacional de Arqueologa.
Love, W. A. 1927. On Magnetic Work and on Determination of Geographic Positions of
Certain Maya Ruins in Guatemala, Feburary to April 1923. In H. Fisk, Land Magnetic
Observations 19141926, pp. 18489, p. 259. Washington, DC: Carnegie Institution.
Love, M. 2000. In Memorium Edwin M. Shook 19112000. Ancient Mesoamerica (13):
12.
Low, J. W. 1952. Plane Table Mapping. New York: Harper.
Luchi, A E. 2011. Quantitative Features of Cedrela odorata L. Wood (Meliaceae). Revista
Brasil Botany 34: 40310.
Lundell, C. L. 1937. The Vegetation of the Petn. Carnegie Institution of Washington Pub.
478. Washington, DC: Carnegie Institution.
Lundell, C. L. 1938. Plants Probably Utilized by the Old Empire Maya of Petn and
Adjacent Lowlands. Papers of the Michigan Academy of Science, Arts, and Letters 24:
3756.
Lundell, C. L. 1961. Plantae Mayanae II: Collections from Peten and Belice. Wrightia
2(3): 11126.
Luzzadder-Beach, S. 2000. Water Resources of the Chunchucmil Maya. The Geographical
Review 90(4): 493510.
Luzzadder-Beach, S., and Beach, T. 2008 Water Chemistry Constraints and Possibilities
for Ancient and Contemporary Maya Wetlands. The Journal of Ethnobiology 28(2):
21130.
Luzzadder-Beach, S., and Beach, T. 2009. Arising from the Wetlands: Mechanisms and
Chronology of Landscape Aggradation in the Northern Coastal Plain of Belize. Annals
of the Association of American Geographers 99(1): 126.
Luzzadder-Beach, S., Beach, T., and Dunning, N. 2012. Wetland Fields as Mirrors of
Drought and the Maya Abandonment. Proceedings of the National Academy of Science
109(10): 364651.
MacArthur, R., and MacArthur, J. 1961. On Bird Species Diversity. Ecology 42: 5948.
Macbride, J. F. 1951. Flora of Peru. Part IIIA(1). Chicago: Field Museum of Natural
History Botanical Series.
Madison, J. 1787. The Federalist Papers, No. 10.
Magee, K. S. 2011. Segmentation, Object-Oriented Applications for Remote Sensing Land
Cover and Land Use Classification. Ph.D. dissertation, University of Cincinnati.
Maidment, D. R. 2002. Arc Hydro: GIS for Water Resources. Redlands: ESRI Press.
Maler, T. 1901. Researches in the Central Portion of the Usumatsintla Valley: Report of
Explorations for the Museum, 18981900, Vol. II, No. 1. Cambridge, MA: Peabody
Museum of American Archaeology and Ethnology, Harvard University. (Piedras Negras
site map plate XXXIII).
Maler, T. 1911. Explorations in the Department of Peten, Guatemala-Tikal, Vol. 5, No.1.
Cambridge, MA: Memoirs, Peabody Museum, Harvard University.
Mann, H. B., and Whitney, D. R. 1947. On a Test of Whether One of Two Random
Variables is Stochastically Larger than the Other. The Annals of Mathematical Statistics
18(1): 5060.
Marchiori, J. N. C., and Brum, E. T. 1997. Wood Anatomy of Calyptranthes concinna DC.
(Myrtaceae). Ciencia Rural 27: 21722.
Marcus, J. 1993. Ancient Maya Political Organization. In Sabloff, J. A., and Henderson, J.
S. (eds.), Lowland Maya Civilization in the Eighth Century A.D. 111183. Washington,
DC: Dumbarton Oaks.
Marcus, J., and Sabloff, J. 2008. The Ancient City: New Perspectives on Urbanism in the
Old and New World. Santa Fe: SAR Press.
Martin, S. 2003. In the Line of the Founder: A View of Dynastic Politics at Tikal. In
Sabloff, Jeremy A., (ed.), Tikal: Dynasties, Foreigners, and Affairs of State Advancing
Maya Archaeology, pp. 345. Santa Fe: School of American Research Press.
Martin, S., and Grube, N. 2008. Chronicle of the Maya Kings and Queens: Deciphering
the Dynasties of the Ancient Maya. 2nd ed. London: Thames & Hudson.
Martindale, A., Letham, B., McLaren, D., Archer, D., Burchell, M., and Schone, B. R.
2009. Mapping of Subsurface Shell Midden Components through Percussion Coring:
 Examples from the Dundas Islands. Journal of Archaeological Science 36(7): 156575.
Martinelli, L. A., Pessenda, L. C. R., Espinoza, E., Camargo, P. B., Telles, E. C., Cerri, C.
C., Victoria, R. L., Aravina, R., Richey, J., and Trumbore, S. 1996. Carbon-13 Variation
with Depth in Soils of Brazil and Climate Change during the Quaternary. Oecologia
106: 37681.
Masson, M. A. 1999. Animal Resource Manipulation in Ritual and Domestic Contexts at
Postclassic Maya Communities. World Archaeology 31(1): 93120.
Maynard, J. B. 1983. Geochemistry of Sedimentary Ore Deposits. New York: Springer-
Verlag Press.
Maynard, J. B. 1992. Chemistry of Modern Soils as a Guide to Interpreting Precambrian
Paleosols. Journal of Geology 100: 27989.
McAnany, P. A. 1990 Water Storage in the Puuc Region of the Northern Maya Lowlands:
A Key to Population Estimates and Architectural Variability. In Culbert, T. P., and Rice,
D. S. (eds.), Precolumbian Population History in the Maya Lowlands, pp. 26384.
Albuquerque: University of New Mexico Press.
McAnany, P. 1992. Agricultural Tasks and Tools: Patterns of Stone Tool Discard Near
Prehistoric Maya Residences Bordering Pulltrouser Swamp, Belize. In Killion, T. W.
(ed.), Gardens of Prehistory: The Archaeology of Settlement Agriculture in Greater
Mesoamerica, 184213. Tuscaloosa: University of Alabama Press.
McAnany, P. 1995. Living with the Ancestors: Kinship and Kingship in Ancient Maya
Society. Austin: University of Texas Press.
McAnany, P. A., and Gallareta Negrn, T. 2010. Belicose Rulers and Climatological Peril?
In McAnany, P. A., and Yoffee, N. (eds.), Questioning Collapse: Human Resilience,
Ecological Vulnerability and the Aftermath of Empire, pp. 14275. New York:
Cambridge University Press.
McKillop, H. 1994. Ancient Maya Tree Cropping, a Viable Subsistence Adaptation for the
Island Maya. Ancient Mesoamerica 5: 12940.
McKillop, H. 1996. Prehistoric Maya Use of Native Palms: Archaeobotanical and
Ethnobotanical Evidence. In S. L. Fedick (ed.), The Managed Mosaic: Ancient Maya
Agriculture and Resource Use, pp. 27894. Salt Lake City: University of Utah Press.
McKillop, H. 2002. Salt, White Gold of the Ancient Maya. Gainesville: University Press of
Florida.
McNeil, C. L., D. A. Burney and L. P. Burney. 2010. Evidence Disputing Deforestation as
the Cause for the Collapse of the Ancient Maya. Proceedings of the National Academy
of Sciences USA 107: 101722.
McNeill, W. H. 1976. Plagues and Peoples. Garden City, NY: Anchor Press.
McVaugh, R. 1963. Flora of Guatemala 24, Part 7(3). Chicago: Fieldiana Botany.
McWilliam, A., Roberts, J. M., Cabral, O. M. R., Leitao, B. R., de Costa, A. C. L.,
Maitelli, G. T., and Zamparoni, C. A. G. P. 1993. Leaf Area Index and Above-Ground
Biomass of Terra Firme Rain Forest and Adjacent Clearings in Amazonia. Functional
 Ecology 7(3): 31017.
Meadows, R. 2001. Crafting Kawil: A Comparative Analysis of Maya Symbolic Flaked
Stone Assemblages from Three Sites in Northern Belize. Ph.D. dissertation, Department
of Anthropology. Austin: University of Texas.
Medina-Elizalde, M., Burns, S. J., Lea, D. W., Asmerom, Y., von Gunten, L., Polyak, V.,
Vuille, M., Karmalkar, A. 2010. High Resolution Stalagmite Climate Record from the
Yucatn Peninsula Spanning Themaya Terminal Classic Period. Earth and Planetary
Science Letters 298(2010) 25562
Medina-Elizalde, M., and Rohling, E. J. 2012. Collapse of Classic Maya Civilization
Related to Modest Reduction in Precipitation. Science 335(6071): 9569.
Mercer, E. D., Haggar, J., Snook, A., and Sosa, M. 2005. Agroforestry Adoption in the
Calakmul Biosphere Reserve, Campeche, Mexico. Small-Scale Forest Economics,
Management and Policy 4: 16384.
Mertz, O. 2002. The Relationship between Length of Fallow and Crop Yields in Shifting
Cultivation: A Rethinking. Agroforestry Systems 55(2) 14959.
Mertz, O., Wadley, R. L., Nielsen, U., Bruun, T. B., Colfer, C. J. P., Neerguard, A., Jepsen,
M. R., et al. 2008. A Fresh Look at Shifting Cultivation: Fallow Length an Uncertain
Indicator of Productivity. Agricultural Systems 96(13): 7584.
Miksicek, C. H. 1983. Macrofloral Remains of the Pulltrouser Area: Settlements and
Fields. In Turner, B. L., and Harrison, P. D. (eds.), Pulltrouser Swamp: Ancient Maya
Habitat, Agriculture and Settlement in Northern Belize: 94104. Austin: University of
Texas Press.
Miksicek, C. H. 1986. Paleobotanical Identifications. In A. A. Demarest (ed.), The
Archaeology of Santa Leticia and the Rise of Maya Civilization, pp. 199200. New
Orleans: Tulane University Press.
Miksicek, C. H. 1987. Formation Processes of the Archaeobotanical Record. Advances in
Archaeological Method and Theory 10: 21147.
Miksicek, C. H. 1991. The Ecology and Economy of Cuello: The Natural and Cultural
Landscape of Preclassic Cuello. In Hammond, N. (ed.), Cuello: An Early Maya
Community in Belize, pp. 7084. Cambridge, MA: Cambridge University Press.
Milburn, M. P. 2004. Indigenous Nutrition: Using Traditional Food Knowledge to Solve
Contemporary Health Problems. American Indian Quarterly 28(34): 41134.
Miller, A., and Schaal, B. 2006. Domestication and the Distribution of Genetic Variation
in Wild and Cultivated Populations of the Mesoamerican Fruit Tree Spondias purpurea
L. (Anacardiaceae). Molecular Ecology 15: 146780.
Miller, R. P., and Nair P. K. R. 2006. Indigenous Agroforestry Systems in Amazonia:
From Prehistory to Today. Agroforestry Systems 66: 15164.
Miller, T. 1996. Geologic and Hydrologic Controls on Karst and Cave Development in
Belize. Journal of Cave and Karst Studies 58(2): 10020.
Millspaugh, C. F. 18951904. Contribution [I]-III to the Flora of Yucatan. Chicago: Field
 Museum, Botany Series.
Mitchell, J. D. 1987. The Cashew and Its Relatives (Anacardium: Anacardiaceae).
Memoirs of the New York Botanical Garden, Vol. 42.
Moholy-Nagy, H. 2003. The Hiatus at Tikal, Guatemala. Ancient Mesoamerica 14: 7783.
Moholy-Nagy, H. 2003. Tikal Report No. 27(B): The Artifacts of Tikal: Utilitarian
Artifacts and Unworked Material. Philadelphia: University of Pennsylvania Museum of
Archaeology and Anthropology.
Moholy-Nagy, H. 2008. The Artifacts of Tikal: Ornamental and Ceremonial Artifacts and
Unworked Material. Tikal Report No. 27, Part A. Philadelphia: University of
Pennsylvania Museum of Archaeology and Anthropology.
Molina, J. A. E., Clapp, C. E., Linden, D. R., Allmaras, R. R., Layese, M. F., Dowdy, R.
H., et al. 2001. Modeling the Incorporation of Corn (Zea mays L.) Carbon from Roots
and Rhizodeposition into Soil Organic Matter. Soil Biology and Biochemistry 33: 83
92.
Montmollin, O. D. 1988. Scales of Settlement Study for Complex Societies: Analytical
Issues from the Classic Maya Area. Journal of Field Archaeology 15(2): 151.
Moore, D., and Reynolds, R. C. 1997. X-Ray Diffraction and the Identification and
Analysis of Clay Minerals, 2nd ed. New York: Oxford University Press.
Morehart, C. T., Lentz, D. L., and Prufer, K. M. 2005. Wood of the Gods: The Ritual Use
of Pine (Pinus spp.) by the Ancient Lowland Maya. Latin American Antiquity 16: 255
74.
Moreno, T., Querol, X., Castillo, S., Alastuey, A., Cuevas, E., Herrmann, L., Mounkaila,
M., Elvira, J., and Gibbons, W. 2006. Geochemical Variations in Aeolian Mineral
Particles from the SaharaSahel Dust Corridor. Chemosphere 65(2): 26170.
Morley, S. G. 1938. Inscriptions of the Peten. 1. Publication 437. Washington, DC:
Carnegie Institution of Washington.
Morley, S. G. 1947. The Ancient Maya. Stanford, CA: Stanford University Press.
Moyes, H. 2006. The Sacred Landscape as a Political Resource: A Case Study of Ancient
Maya Cave Use at Checham Ha, Belize, Central America. Ph.D. dissertation,
Department of Anthropology. Buffalo: State University of New York.
Mueller, A. D., Islebe, G. A., Hillesheim, M. B., Grzesik, D. A., Anselmetti, F. S.,
Ariztegui, D., Brenner, M., Curtis, J. H., Hodell, D. A., and Venz, K. A. 2009. Climate
Drying and Associated Forest Decline in the Lowlands of Northern Guatemala during
the Late Holocene. Quaternary Research 71: 13341.
Muhs, D. R., Budahn, J. R., Prospero, J. M., and Carey, S. N. 2007. Geochemical
Evidence for African Dust Inputs to Soils of Western Atlantic Islands: Barbados, the
Bahamas, and Florida. Journal of Geophysical Research 112, FO2009.
doi:10.1029/2005JF000445.
Murray, K. G. 1988. Avian Seed Dispersal of Three Neotropical Gap Dependent Plants.
Ecological Monographs 58: 27198.
Murtha, T. M. 2002. Land and Labor: Classic Maya Terraced Agriculture at Caracol,
Belize. Ph.D. dissertation, Department of Anthropology. University Park: Penn State
University.
Mutchnick, P. A., and McCarthy, B. C. 1997. An Ethnobotanical Analysis of the Tree
Species Common to the Subtropical Moist Forest of the Petn, Guatemala. Economic
Botany 51: 15883.
NGDC. 2010. Estimated Value of Magnetic Declination On-Line Calculator. NOAA
National Geophysical Data Center. Available on the Internet,
http://www.ngdc.noaa.gov/geomagmodels/Declination.jsp, accessed July 5, 2010.
NGS. 2010. Universal Transverse Mercator Coordinate On-Line Calculator. National
Geodetic Survey. Available on the Internet,
http://www.ngs.noaa.gov/TOOLS/utm.shtml, accessed July 5, 2010.
Nagy, A. 2012. Water Management Strategies in an Engineered Neotropical Landscape.
M.A. Thesis, Department of Anthropology. University of Cincinnati.
Nash, D. 1976. Flora of Guatemala 24, Part 11(4). Chicago: Fieldiana Botany.
Nash, D., and Williams, L. O. 1976. Flora of Guatemala 24, Part 12. Chicago: Fieldiana
Botany.
Nations, J. D. 1992. Xateros, Chicleros, and Pimenteros: Harvesting Renewable Tropical
Forest Resources in the Guatemalan Petn. In Redford K. H., and Padoch C. (eds.),
Conservation of Neotropical Forests: Working from Traditional Resource Use, pp. 208
19. New York: Columbia University Press.
Neisheim, I., Halvorsen, R., and Nordal, I. 2010. Plant Composition in the Maya
Biosphere Reserve: Natural and Anthropogenic Influences. Plant Ecology 208: 93122.
Nelson, Z. 2005. Settlement and Population at Piedras Negras, Guatemala. Ph.D.
dissertation, Department of Anthropology. University Park: Penn State University.
Nordt, L. C. 2001. Stable Carbon and Oxygen Isotopes in Soils. In Goldberg, P., Holliday,
V., and Ferring, C. R. (eds.), Earth Sciences and Archaeology, pp. 41948. New York:
Plenum.
OBrien, M. J., Mason, R. D., Lewarch, D. E., and Neely, J. A. 1982. A Late Formative
Irrigation Settlement below Monte Albn. Austin: University of Texas Press.
OBrien, M. J., and Laland, K. N. 2012. Genes, Culture, and Agriculture: An Example of
Human Niche Construction. Current Anthropology 53(4): 43470.
Odling-Smee, F. J., Laland, K. N., and Feldman, M. W. 2003. Niche Construction: The
Neglected Process in Evolution (MPB-37). Princeton, NJ: Princeton University Press.
Olson, G. W. 1977. The Soil Survey of Tikal. Department of Agronomy. Ithaca, NY:
Cornell University. Unpublished manuscript.
Olson, G. W. 1981. Soils and the Environment. New York: Chapman and Hall.
Olson, G. W. 1981. Archaeology: Lessons on Future Soil Use. Journal of Soil and Water
Conservation 36(5): 2614.
Palerm, A. 1955. The Agricultural Bases of Urban Civilization in Mesoamerica. In
Steward, J. H. (ed.), Irrigation Civilization: A Comparative Study. Washington, DC:
Pan America Union Social Science Monographs 1.
Palmer, M. W. 1990. The Estimation of Species Richness by Extrapolation. Ecology 71:
11958.
Palmer, M. W. 1995. How Should One Count Species? Natural Areas Journal 15: 12435.
Paoletti, M. G. 1995. Biodiversity, Traditional Landscapes and Agroecosystem
Management. Landscape and Urban Planning 31: 11728.
Parker, D. Y. R. 2012. Late and Terminal Classic Maya Subsistence: Stable Isotope
Analysis at ChacBalam and San Juan on Northern Ambergris Caye, Belize. Ph.D.
dissertation, Department of Anthropology. Ann Arbor: University of Michigan.
Parnell, J. J., Terry, R. E., and P. Sheets. 2002. Soil Chemical Analysis of Ancient
Activities in Cern, El Salvador: A Case Study of a Rapidly Abandoned Site. Latin
American Antiquity 13: 33142.
Paton, T. R. 1973. Origins and Terminology for Gilgai in Australia. Geoderma 11: 221
42.
Pearsall, D. M. 1989. Paleoethnobotany: A Handbook of Procedures. San Diego:
Academic Press.
Pennington, T. D., and Sarukhan, J. 1968. Arboles tropicales de Mexico. Mexico City:
Inst. Nac. Investig. Forestales.
Pennington, T. D. 1990. Sapotaceae. Flora Neotropica Monograph, Vol. 52. Bronx: New
York Botanical Garden.
Pennington, T. D. 1991. The Genera of Sapotaceae. Bronx: New York Botanical Garden.
Prez Robles, G., Romn, E., and Houston, S. (eds.) 2009. Proyecto Arqueolgico El
Zotz Informe No. 4 Temporada. Mesoweb. Available on the Internet,
http://mesoweb.com/zotz/El-Zotz-2009.pdf.
Perna, N. E. D., and Melandri, J. L. 2006. Wood Anatomy of the Tribe Caesalpinieae
(Leguminosae, Caesalpinioideae) in Venezuela. IAWA Journal 27: 99114.
Pesek, T., Abramiuk, M. Fini, N., Otarola Rojas, M., Collins, S., Cal, V., Sanchez, P.,
Poveda, L., Arnason, J. 2010. Qeqchi Maya Healers Traditional Knowledge in
Prioritizing Conservation of Medicinal Plants: Culturally Relative Conservation in
SustainingTraditional Holistic Health Promotion. Biodiversity Conservation 19: 120.
Peters, C. M., and Pardo-Tejeda, E. 1982. Brosimum alicastrum (Moraceae): Uses and
Potential in Mexico. Economic Botany 36: 16675.
Peters, C. M. 1983. Observations on Maya Subsistence and the Ecology of a Tropical
Tree. American Antiquity 48: 61015.
Peters, C. M. 2000. Precolumbian Silviculture and Indigenous Management of
Neotropical Forests. In D. L. Lentz (ed.), Imperfect Balance: Landscape
Transformations in the Precolumbian Americas, pp. 179202. New York: Columbia
 University Press.
Phillips, O. L., Malhi, Y., Higuchi, N., Laurance, W. F., Nuez, P. V., Vsquez, R. M.,
Laurence, S. G., Ferreira, L. V., Stern, M., Brown, S., Grace, J. 1998. Changes in the
Carbon Balance of Tropical Forests: Evidence from Long-Term Plots. Science 282:
43942.
Pielou, E. C. 1966. The Measurement of Diversity in Different Types of Biological
Collections. Journal of Theoretical Biology 13: 13144.
Pitman, N. C. A., Terborgh, J. W., Silman, M. R., Nez V. P., Neill, D. A., Cern, C. E.,
Palacios, W. A., and Aulestia, M. 2001. Dominance and Distribution of Tree Species in
Upper Amazonian Terra Firme Forests. Ecology 82: 210117.
The Plant List. 2010. Version 1. Available on the Internet, http://www.theplantlist.org/.
Podobnikar, T. 2009. Georeferencing and Quality Assessment of Josephine Survey Maps
for the Mountainous Region in the Triglav National Park. Acta Geodaetica et
Geophysica Hungarica 44(1): 4966.
Pohl, M., and Pohl, J. 1994. Cycles of Conflict: Political Factionalism in the Maya
Lowlands. In Brumfiel, E., and Fox, J. (eds.), Factional Competition and Political
Development in the New World, pp. 13857. Cambridge: Cambridge University Press.
Pohl, M., Pope, K., and Jones, J. 2000. Base Agrcola de la Civilizacin Maya de las
Tierras Bajas. In Laporte, J. P., Escobedo, H., Arroyo, B., and de Suasnvar, A. C.
(eds.), XIII Simposio de Investigaciones Arqueolgicas en Guatemala, pp. 25867.
Guatemala City: Museo Nacional de Arqueologa y Etnologa.
Pollock, H. E. D., Roys, R., Proskouriakoff, T., and Smith, A. L. 1962. Mayapan, Yucatan,
Mexico, Publication 619. Washington, DC: Carnegie Institution.
Pope, K. O., and Dahlin, B. H. 1993. Radar Detection and Ecology of Ancient Maya
Canal Systems Reply to Adams et al. Journal of Field Archaeology 20: 37983.
Pope, K. O., Pohl, M. D., and Jacob, J. S. 1996 Formation of Ancient Maya Wetland
Fields: Natural and Anthropogenic Processes. In Fedick S. L. (ed.), The Managed
Mosaic: Ancient Maya Agriculture and Resource Use, pp. 16576. Salt Lake City:
University of Utah Press.
Proskouriakoff, T. 1946. An Album of Maya Architecture. Publication 558. Washington,
DC: Carnegie Institution.
Proskouriakoff, T. 1963. An Album of Maya Architecture. Norman: University of
Oklahoma Press.
Puleston, D. E. 1967. Proable Field Report following the 1967 Tikal Sustaining Area
Project. Available on the Internet, http://www.puleston.org/
documents/Probable%20Field%20Report%20Following%20the%201967%20Tikal%20Sustain
accessed May 18, 2013.
Puleston, D., and Callender, D. W. 1967. Defensive Earthworks at Tikal. Expedition 9(30):
408.
Puleston, D. 1968. Brosimum alicastrum as a Subsistence Alternative for the Classic Maya
 of the Central Southern Lowlands. M.S. Thesis. Philadelphia: University of
Pennsylvania.
Puleston, D. E. 1973. Ancient Maya Settlement Patterns and Environment at Tikal,
Guatemala: Implications for Subsistence Models. Ph.D. dissertation, University of
Pennsylvania. Ann Arbor: University Microfilms.
Puleston, D. E. 1977. The Art and Archaeology of Hydraulic Agriculture in the Maya
Lowlands. In Hammond, N. (ed.), Social Process in Maya Prehistory, pp. 44968.
Academic Press, New York.
Puleston, D. E. 1983. The Settlement Survey of Tikal. Tikal Reports 13. Philadelphia:
University Museum Publications.
Pyburn, K. A. 1998. The Albion Island Settlement Pattern Project: Domination and
Resistance in Early Classic Northern Belize. Journal of Field Archaeology 25: 3762.
Quintana, O., and Noriega, R. 2006. Tikal y sus vecinos: complejodad cultural en el
tringulo Yaxha-Nakum-Naranjo. In LaPorte, J. P., Arroyo, B., and Meja, H. (eds.),
XIX Simposio de Investigacioies Arqueolgicas en Guatemala, pp. 3339. Guatemala:
Museo Nacional de Arqueologa e Etnologa.
Quintana, O., and Wurster, W. 2001. Ciudades Mayas del Noreste del Petn, Guatemala:
Un estudio urbanstico comparativo. AVA Materialien 59. Germany: Verlag Philipp
von Zabern, Mainz am Rhein.
R Development Core Team. 2005. R: A Language and Environment for Statistical
Computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-
07-0. Available on the Internet, http://www.R-project.org.
Ramos, C. E., Valdez, F., Buttles, P. J., Mijangos, B., and Daz, M. 2011. Los Materiales
Arqueolgicos del Proyecto de Silvacultura de los Antiguos Mayas de Tikal. In Lentz,
D., Ramos, C., Dunning, N., Scarborough, V., and Grazioso, L. Sierra, Proyecto de
Silvacultura y Manejo de Aguas de los Antiguos Mayas de Tikal: Temporada de 2010,
pp. 1829. Direccin Patrimonio Cultural y Natural de Guatemala.
Rebollar, S., and Quintanar, A. 2000. Anatoma y Usos de la Madera de Siete rboles
Tropicales de Mxico. Revista de Biologa Tropical 48(23): 569578.
Redfield, R., and Villa Rojas, A. 1934. Chan Kom: A Maya Village. Publication 488.
Washington, DC: Carnegie Institution of Washington.
Reed, D. M. 1998. Ancient Maya Diet at Copn, Honduras. Ph.D. dissertation,
Department of Anthropology. University Park: Penn State University.
Repussard, A. 2009. Stable Carbon And Oxygen Isotopes In Bone Tracing Droughts
during the Maya Era Using Archaeological Deer Remains. Open Access Dissertations
and Theses. Paper 4332. Available on the Internet,
http://digitalcommons.mcmaster.ca/opendissertations/4332.
Restall, M. 1997. The Maya World. Stanford, CA: Stanford University Press.
Rice, D. S., and Culbert, T. P. 1990. Historical Contexts for Population Reconstructions in
the Maya Lowlands. In Culbert, T. P., and Rice, D. S. (eds.), Precolumbian Population
 History in the Maya Lowlands, pp. 136. Albuquerque: University of New Mexico
Press.
Rice, D. S. 1996. Paleolimnological Analysis in the Central Peten, Guatemala. In Fedick,
S. L. (ed.), The Managed Mosaic: Ancient Maya Agriculture and Resource Use, pp.
193206. Salt Lake City: University of Utah Press.
Rice, P. M. 1981. Evolution of Specialized Pottery Production: A Tikal Model. Current
Anthropology 22(3): 21940.
Richter 1981. Wood and Bark Anatomy of Lauraceae I. Aniba Aublet. IAWA Bulletin New
Series 2: 7987.
Richter, H. G. 1985. Wood and Bark Anatomy of Lauraceae. II. Licaria Aublet. IAWA
Bulletin New Series 6(3).
Richter, H. G., and Dallwitz, M. J. 2000 onwards. Commercial Timbers: Descriptions,
Illustrations, Identification, and Information Retrieval. In English, French, German,
Portuguese, and Spanish. Version: 25th June 2009. Available on the Internet,
http://delta-intkey.com.
Ricketson, O. G., Jr., and Ricketson, E. 1937. Uaxactun, Guatemala, Group E 1926
1931. Part 1: The Excavations. Publication 477. Carnegie: Institution of Washington.
Rico-Gray, V., Chems, A., and Mandujano, S. 1991. Uses of Tropical Deciduous Forest
Species by the Yucatecan Maya. Agroforestry Systems 14: 14961.
Rico-Gray, V., Garcia-Franco, J. G., Chem s, A., Puch, A., and Sima, P. 1990. Species
Composition, Similarity, and Structure of Mayan Homegardens in Tixpeual and
Tixcacaltuyub, Yucatan, Mexico. Economic Botany 44: 47087.
Robin, C. 1999 Towards an Archaeology of Everyday Life: Maya Farmers of Chan
Nohol, Belize. Ph.D. dissertation, Department of Anthropology. University of
Pennsylvania.
Robin, C. (ed.). 2012. Chan: An Ancient Maya Farming Community. Gainesville:
University Press of Florida.
Robock, A. 2002. Volcanic Eruption, El Chichn. In MacCracken, M. C., and Perry, J. S.
(eds.), Encyclopedia of Global Environmental Change. Chichester: John Wiley & Sons.
Robyns, A. 1963. Bombacaceae Pseudobombax ellipticoideum. Bulletin du Jardin
Botanique de ltat Bruxelles 33: 48.
Romero, J. J. 2003. Lea y carbn. Report 9140777. Centro Universitario de Petn.
Guatemala: Universidad de San Carlos de Guatemala.
Ross, N. J. 2011. Modern Tree Species Composition Reflects Ancient Maya Forest
Gardens in Northwest Belize. Ecological Applications 21: 584.
Ross, N. J., and Rangel, T. F. 2011. Ancient Maya Agroforestry Echoing through Spatial
Relationships in the Extant Forest of NW Belize. Biotropica 43: 1418.
Roys, R. L. 1931. The Ethno-Botany of the Maya. Philadelphia: Institute for the Study of
Human Issues.
Roys, R. L. 1939. The Titles of Ebtun. Publication 505. Washington, DC: Carnegie
Institution.
Rue, D. 1987. Early Agriculture and Early Postclassic Maya Occupation at Copn,
Honduras. Nature 326: 2856.
Rue, D., Webster, D., and Traverse, A. 2002. Late Holocene Fire and Agriculture in the
Copn Valley, Honduras. Ancient Mesoamerica 13: 26772.
Rumsey, D., and Williams, M. 2002. Historical maps in GIS. In Knowles, A. (ed.), Past
Time, Past Place GIS for History. Redlands: ESRI Press.
Ruppert, K. 1935. The Caracol at Chichen Itza, Yucatan, Mexico. Publication no. 454. Site
map illustration 350. Washington, DC: Carnegie Institution.
Ruppert, K., and Smith, A. L. 1951. Year Book 50, Report 40.3, Mayapn, Yucatn. In
Weeks, J., and Hill J. (eds.) 2006, The Carnegie Maya, Maya Research Program, 1913
1957. Boulder: University Press of Colorado.
Ruppert, K. 1952. Chichen Itza, Architectural Notes and Plans, Publication 595. Site map
illustration 151. Washington, DC: Carnegie Institution.
Rzedowski, J. 1981. Vegetacin de Mxico. Mexico, D. F.: Editorial Limusa.
Sabloff, J. A. (ed.) 2003. Tikal: Dynasties, Foreigners, and Affairs of State: Advancing
Maya Archaeology. Santa Fe: School of American Research Press.
Sader, S. A., Waide, R. B., Lawrence, W. T., and Joyce, A. T. 1989, Tropical Forest
Biomass and Successional Age Class Relationships to a Vegetation Index Derived from
Landsat TM Data. Remote Sensing of Environment 28: 14398.
Sanders, W. 1957. Tierra y Agua (Soil and Water): A Study of the Ecological Factors in
the Development of Meso-American Civilizations. Ph.D. dissertation, Department of
Anthropology, Harvard University. Ann Arbor: University Microfilms.
Sanders, W. 1973. The Cultural Ecology of the Lowland Maya: A Reevaluation. In
Culbert T. P. (ed.), The Classic Maya Collapse, pp. 32536. Albuquerque: University of
New Mexico Press.
Sanders, W. 1979. The Fon of Bafut and the Classic Maya. In Prehistoric Maya
Agriculture, pp. 389400. Paris: 42nd Internatonal Congress of Americanists.
Sanders, W., Parsons, J. R., and Santley, R. S. 1979. The Basin of Mexico: Ecological
Processes in the Evolution of a Civilization. New York: Academic Press.
Sanders, W. T. 1989. Household, Lineage, and State at Eighth Century Copn, Honduras.
In The House of the Bacabs, Copn, Honduras, pp. 89105. Studies in Pre-Columbian
Art & Archaeology 29. Washington, DC: Dumbarton Oaks Research Library and
Collection.
Sandstrom, A. R. 2014. Water and the Sacred in Mesoamerica. In Scarborough, V. L. (ed.),
Water and Humanity: Historical Overview. Paris: UNESCO.
Sapper, K. 1896. La geografa fisica y la geologa de la peninsula de Yucatn. Boletin del
Instituto de Geologa de Mxico 3: 157.
Satterthwaite, L. 1933. Piedras Negras Preliminary Papers: Description of the Site, with
Short Notes on the Excavations of 19311932. In Weeks, J., Hill, J., and Golden, C.
(eds.) Piedras Negras Archaeology, 19311939 (2005). Philadelphia: University of
Pennsylvania museum of Archaeology and Anthropology.
Satterthwaite, L. 1943. Piedras Negras Archaeology: Architecture: Introduction. In Weeks,
J., Hill, J., and Golden, C. (eds.), Piedras Negras Archaeology, 19311939 (2005).
Parris site map in appendix 6. Philadelphia: University of Pennsylvania museum of
Archaeology and Anthropology.
Saturno, W. D., Stuart, A.F. Aveni, and F. Rossi. 2012. Ancient Maya Astronomical Tables
from Xultun, Guatemala. Science 336: 714717.
Scarborough, V. L. 1983. A Preclassic Maya Water System. American Antiquity 48(4):
72044.
Scarborough, V. L., and Gallopin, G. G. 1991. A Water Storage Adaptation in the Maya
Lowlands. Science 251(4994): 65862.
Scarborough, V. L. 1993. Water Management in the Southern Maya Lowlands: An
Accretive Model for the Engineered Landscape. In Scarborough, V. L., and Isaac, B. L.
(eds.), Economic Aspects of Water Management in the Prehispanic New World, pp.17
69, Research in Economic Anthropology, Supplement 7. Greenwich: JAI Press.
Scarborough, V. L. 1994. Maya Water Management. Research and Exploration 10(2):
18499.
Scarborough, V. L. 1994. Water Management as a Function of Locational and
Appropriational Movements and the Case of the Classic Maya of Tikal. In Colin, A. M.,
and Tandy, D. W. (eds.), From Political Economy to Anthropology: Situating Economic
Life in Past Societies. New York: Black Rose Books.
Scarborough, V. L., Connelly, R., and Ross, S. 1994. The Prehispanic Maya Reservoir
System at Kinal, Peten, Guatemala. Ancient Mesoamerica 5(1): 97106.
Scarborough, V. L. 1998. Ecology and Ritual: Water Management and the Maya. Latin
American Antiquity 8(2): 13559.
Scarborough, V. L., Schoenfelder, J. W., and Lansing, J. S. 1999. Early Statecraft on Bali:
The water temple complex and the decentralization of political economy. Research in
the Economic Anthropology 20: 299330.
Scarborough, V. L., Schoenfelder, J. W., and Lansing, J. S. 2000. Water Management and
Landscape Transformation at Sebatu, Bali. Indo-Pacific Prehistory Association Bulletin
20(4): 7992.
Scarborough, V.L. 2003a. Ballcourts and Reservoirs: The Social Construction of a
Tropical Landscape. In Breton, A., Monod Becquelin, A., and Ruz, M.H. (eds.), Maya
Space: Representations, Uses and Beliefs, pp. 7792. Mexico: Centro de Estudios
Mayas (UNAM)/CEMCA.
Scarborough, V. L. 2003b. The Flow of Power: Ancient Water Systems and Landscapes.
Santa Fe: School of American Research Press.
Scarborough, V. L., Valdez, F., and Dunning, N. (eds.) 2003. Heterarchy, Political
Economy, and the Ancient Maya: The Three Rivers Region of the East-Central Yucatan
Peninsula. Tucson: University of Arizona Press.
Scarborough, V. L. 2006 An Overview of Mesoamerican Water Systems. In Lucero, L. J.,
and Fash, W. B. (eds.), Precolumbian Water Management, pp. 22335. Tucson:
University of Arizona Press.
Scarborough, V. L. 2010. The Archaeology of Sustainability: Mesoamerica. Ancient
Mesoamerica 20(2): 197203
Scarborough, V. L., and Burnside, W. R. 2010. Complexity and Sustainability:
Perspectives from the Ancient Maya and the Modern Balinese. American Antiquity
75(2): 32763.
Scarborough, Vernon L., and Lisa Lucero 2011. The Non-Hierarchical Development of
Complexity in the Semitropics: Water and Cooperation. Water History 2(2): 185205.
Scarborough, V. L., Dunning, N. P., Tankersley, K. B., Carr, C., Weaver, E., Grazioso, L.,
Lane, B., Jones, J. G., Buttles, P., Valdez, F., and Lentz, D. L. 2012. Water and
Sustainable Land Use in an Ancient Tropical City: Tikal, Guatemala. Proceedings of the
National Academy of SciencesUSA 109(31): 1240813.
Scarborough, V. L., and Valdez, F. 2014. The Alternative Economy: Resilience in the Face
of Complexity from the Eastern Lowlands. In Chase, A. F., and Scarborough, V. L.
(eds.), The Resilience and Vulnerability of Ancient Landscapes: Transforming Maya
Archaeology through IHOPE, Archeological Papers of the American Anthropological
Association. Malden, MA: Wiley Periodicals.
Scherer, A. K. 2007. Population Structure of the Classic Period Maya. American Journal
of Physical Anthropology 132: 36780.
Scherer, A., Wright, L. E., and Yoder, C. J. 2007. Bioarchaeological Evidence for Social
and Temporal Differences in Diet at Piedras Negras, Guatemala. Latin American
Antiquity 18(1): 85104.
Schofield, W., and Breach, M. 2007. Engineering Surveying. 6th ed. Oxford: Elsevier.
Schreiner, T. P. 2002. Traditional Maya Lime Production: Environmental and Cultural
Implications of a Native American Technology. Ph.D. dissertation. Berkeley: University
of California.
Schulze, E. D., Beck, E., and Muller-Hohenstein, K. 2002. Plant Ecology. New York:
Springer.
Schulze, M. D., and Whitacre, D. F. 1999. A Classification and Ordination of the Tree
Community of Tikal National Park, Petn, Guatemala. Bulletin of the Florida Museum
of Natural History 41(3): 169297.
Schwartz, N. B. 1990. Forest Society: A Social History of Petn, Guatemala. Philadelphia:
University of Pennsylvania Press.
Scott, J. 1976. The Moral Economy of the Peasant Rebellion and Subsistence in Southeast
Asia. New Haven, CT: Yale University Press.
Sever, T., and D. Irwin. 2003. Landscape Archaeology: Remote-Sensing Investigation of
the Ancient Maya in the Peten Rainforest of Northern Guatemala. Ancient Mesoamerica
14: 11322.
Shafer, H. J., and Hester, T. R. 1986. Maya Stone-Tool Craft Specialization and
Production at Colha, Belize: A Reply to Mallory. American Antiquity 51: 14866.
Shannon, C. E. 1948. A Mathematical Theory of Communication. Bell System Technical
Journal 27: 379423.
Sharer, Robert J. 1977. The Maya Collapse Revisited: Internal and External Perspectives.
In Hammond, N. (ed.), Social Process in Maya Prehistory, pp. 53152. Academic
Press, New York.
Sharer, R. J., and Traxler, L. P. 2006. The Ancient Maya, 6th ed. Stanford, CA: Stanford
University Press.
Sheets, P. 2002. Before the Volcano Erupted. Austin: University of Texas Press.
Sheets, P. 2002. Ceramics and Their Use at Cern. In Sheets, P., (ed.), Before the Volcano
Erupted, pp. 11730. Austin: University of Texas Press.
Sheets, P. 2006. The Cern Site: A Prehistoric Village Buried by Volcanic Ash in Central
America, 2nd ed. Ft. Worth: Harcourt, Brace, Jovanovich.
Sheets, P., Dixon, C., Guerra, M., and Blanford, A. 2011. Manioc Cultivation at Cern, El
Salvador: Occasional Kitchen Garden Plant or Staple Crop? Ancient Mesoamerica
22(1): 111.
Sheets, P., and Dixon, C. 2011. Maya Agriculture South of the Cern Site, El Salvador. In
Sheets, P. (ed.), Archeology and Volcanism in Central America: The Zapotitan Valley of
El Salvador, pp. 16194. Austin: University of Texas Press.
Sheets, P., Lentz, D., Piperno, D., Jones, J., Dixon, C., Maloof, G., and Hood, A. 2012.
Ancient Manioc Agriculture South of the Cern Village, El Salvador. Latin American
Antiquity 23(3): 25981.
Sheets, T. P., and Dahlin, B. H. 1978. Artifacts and Figurines. In The Prehistory of
Chalchuapa, El Salvador, Vol. 2. Philadelphia: Penn Press.
Shim, J. H., Pendall, E., Morgan, J. A., and Ojima, D. S. 2009. Wetting and Drying Cycles
Drive Variations in the Stable Carbon Isotope Ratio of Respired Carbon Dioxide in
Semi-Arid Grassland. Oecologia 160(2): 32133. doi:10.1007/s00442-009-1302-4.
Shook, E. 1950. Year Book 49, Report 23.7; Guatemala. In Weeks, J. and Hill, J. (2006)
The Carnegie Maya, Maya Research Program, 19131957. Boulder, CO: University
Press of Colorado.
Shook, E. M. 1951. Investigaciones arqueolgicas en las ruinas de Tikal, Departamento de
El Petn, Guatemala. Antropologia e historia 3(1): 932.
Shook, E. M. 1958. Field Directors Report: The 1956 and 1957 Seasons. Tikal Report No.
1. In Shook, E., Coe, W., Broman, and Satterthwaite, L. (eds.), Tikal Reports numbers
14. Philadelphia: The University Museum,.
Shook, E. M. 1962. Tikal: Problems of a Field Director. Expedition Magazine 4: 1126.
Siemens, A. 1978 Karst and the Pre-Hispanic Maya in the Southern Lowlands. In
Harrison, P. D., and Turner II, B. L. (eds.), Pre-Hispanic Maya Agriculture, pp. 11744.
Albuquerque: University of New Mexico Press.
Simmons, C. S., Tarano, J. M., and Pinto Z. J. H. 1959. Clasificacin de Reconocimiento
de los Suelos de la Repblica de Guatemala. Guatemala: Instituto Agropecuario
Nacional, Servicio Cooperativa Inter-Americana de Agricultura.
Simpson, E. H. 1949. Measurement of Diversity. Nature 163: 688.
Smith E. P., and van Belle, G. 1984. Nonparametric Estimation of Species Richness.
Biometrics 40: 11929.
Smyth, M. P. 1991. Modern Maya Storage Behavior: Ethnoarchaeological Case Examples
from the Puuc Region of Yucatan. Pittsburgh: University of Pittsburgh Memoirs in Latin
American Archaeology.
Soil Conservation Service (SCS). 1993. Soil Survey Manual. U.S. Department of
Agriculture Handbook 18.
Soil Survey Staff (SSS). 2003. Keys to Soil Taxonomy. 9th ed. Blacksburg, VA:
Pocahontas Press.
Soil Survey Staff. 2006. Keys to Soil Taxonomy, 10 ed. Washington, DC: United States
Department of Agriculture, Natural Resources Conservation Service.
Standley, P. C. 1930. Flora of Yucatan. Field Museum of National History, Botanical
Series 3: 157492.
Standley, P. C., Steyermark, J. A., and collaborators (McVaugh, Nash, Williams). 1946
1977. Flora of Guatemala. Fieldiana. Botany 24(113).
Standley, P. C., and Williams, L. O. 19611975. Flora of Guatemala 24, Parts 79,11.
Fieldiana Botany.
Stanton, T. W., Brown, M. K., and Pagliaro, J. B. 2008. Garbage of the Gods? Squatters,
Refuse Disposal, and Termination Rituals among the Ancient Maya. Latin American
Antiquity 19: 22747.
Stark, M. T. 1999. Social Dimensions of Technical Choice in Kalinga Ceramic Traditions.
In Chilton, E. (ed.), Material Meanings: Critical Approaches to the Interpretation of
Material Culture, pp. 2443. Foundations of Archaeological Inquiry, Salt Lake City:
University of Utah Press.
Steggerda, M. 1941. Maya Indians of Yucatan. Carnegie Institute Publication. Washington,
DC: Carnegie Institute.
Stein, J. K. 1986. Coring Archaeological Sites. American Antiquity 51(3): 50527.
Steininger, M. K. 2000. Satellite Estimation of Tropical Secondary Forest Above-Ground
Biomass: Data from Brazil and Bolivia. International Journal of Remote Sensing 21(6
7): 113957.
Stern, W. L. 1954. Comparative Anatomy of Xylem and Phylogeny of Lauraceae. Tropical
 Woods 100: 173.
Stone, G. D. 1996. Settlement Ecology: The Social and Spatial Organization of Kofyar
Agriculture. Tucson: University of Arizona Press.
Straight, K. 2012. Consumption of Pottery and Lithics in the Periphery of Tikal,
Guatemala. Ph.D. dissertation, Department of Anthropology. University Park: Penn
State University.
Stuart, D. 2000. The Arrival of the Strangers: Teotihuacan and Tollan in Classic Maya
History. In Carrasco, D., Jones, L., and Sessions, S. (eds.), Mesoamericas Classic
Heritage: From Teotihuacan to the Aztecs, pp. 465513. Boulder, CO: University Press
of Colorado.
Stuiver, M., Reimer, P. J., Braziunas, T. F. 1998. High-Precision Radiocarbon Age
Calibration for Terrestrial and Marine Samples. Radiocarbon 40(3): 112751.
Suhler, C., and Freidel, D. 2003. The Tale End of Two Cities: Tikal, Yaxuna, and
Abandonment Contexts in the Lowland Maya Archaeological Record. In Inomata, T.,
and Webb, R. W. (eds.), The Archaeology of Settlement Abandonment in Middle
America, pp. 13547. Salt Lake City: University of Utah Press.
Sullivan, A. P. III, Magee, K. S., Mink, P. B., and Forste, K. M. 2011. Remote Sensing of
Heritage Resources for Research and Management. Park Science 28(3): 938.
Sweetwood, R., Terry, R., Beach, T., and Dahlin, B. 2009. The Maya Footprint: Soil
Resources of Chunchucmil, Yucatn, Mexico. Soil Science Society of America Journal
73(4): 120920.
Taberlet, P., Gielly, L., Pautou, G., and Bouvet, J. 1991. Universal Primers for
Amplification of Three Non-Coding Regions of Chloroplast DNA. Plant Molecular
Biology 17: 11059.
Tankersley, K. B., and Balantyne, M. R. 2010. X-ray Powder Diffraction Analysis of Late
Holocene Reservoir Sediments. Journal of Archaeological Science 37(1): 1338.
Tankersley, K. B., Scarborough, V. L., Dunning, N., Huff, W., Maynard, B., and Gerke, T.
L. 2011. Evidence for Volcanic Ash Fall in the Maya Lowlands from a Reservoir at
Tikal, Guatemala. Journal of Archaeological Science 38: 292538.
Taube, K. 2003. Ancient and Contemporary Maya Conceptions about Field and Forest. In
Gmez-Pompa, A., Allen, M. F., Fedick, S. L., and Jimnez-Osornio, J. (eds.), The
Lowland Maya Area: Three Millennia at the Human-Wildland Interface, pp. 46192.
New York: Hawthorn Press.
Tawney, R. H. 1966. Land and Labor in China. Boston: Beacon Press.
Tedlock, D. 1985. Popol Vuh: The Mayan Book of the Dawn of Life. New York: Simon &
Schuster.
Terry, R. E., Hardin, P. J., Houston, S. D., Nelson, S. D., Jackson, M. W., Carr, J., and
Parnell, J. 2000. Quantitative Phosphorus Measurement: A Field Test Procedure for
Archaeological Site Analysis at Piedras Negras, Guatemala. Geoarchaeology: An
International Journal 15: 15166.
Thomas, B. 2010, Locating Aguadas in Northern Guatemala Using Remote Sensing. M.A.
Thesis, Department of Geography. University of Cincinnati.
Thompson, J. E. 1930. Ethnology of the Mayas of Southern and Central British Honduras.
Laufer B. (ed.) Chicago: Field Museum of Natural History XVII. Publication 274.
Thompson, J. E. S. 1954. The Rise and Fall of Maya Civilization. The Civilization of the
American Indian Series. Norman: University of Oklahoma Press.
Tieszen, L. L. 1991. Natural Variations in the Carbon Isotope Values of Plants:
Implications for Archaeology, Ecology, and Paleoecology. Journal of Archaeological
Science 18(3): 22748.
Thomson, P. 2004. Belize: A Concise History. Oxford, UK: Macmillan.
Tikal, C. 1990. Tikal, Guatemala 2267I 1:50,000 Topographic Map (declassified). Instituto
Geogrfico Militar, USGS product number 113357.
Tikal Reports No. 11. 1958. Museum Monographs, University Museum & Jay I. Kislak
Reference Collection. Philadelphia: University of Pennsylvania.
Tikal Reports.1986. Tikal Reports: Numbers 111. Philadelphia: University Museum
University of Pennsylvania.
Tourtellot, G., III 1993. A View of Ancient Maya Settlement in the Eighth Century. In
Sabloff, J. A., and Henderson, J. S. (eds.), Lowland Maya Civilization in the Eighth
Century A.D., pp. 21942. Washington, DC: Dumbarton Oaks.
Tourtellot, G., Belli, F., Rose, J. J., and Hammond, N. 2003. Late Classic Maya
Heterarchy, Hierarchy, and Landscape at La Milpa, Belize. In Scarborough, V. L.,
Valdez, F., Jr., and Dunning, N. (eds.), Heterarchy, Political Economy, and the Ancient
Maya, pp. 3751. Tucson: University of Arizona Press.
Tozzer, A. M. 1911. Preliminary Study of the Ruins of Tikal, Guatemala. Memoirs,
Peabody Museum, Harvard University 5(2).
Tozzer, A. M. 1941. Landas Relacin de las Cosa de Yucatan, XVIII. Peabody Museum of
American Archaeology and Ethnology. Cambridge, MA: Harvard University.
Trachman, R. M. 2011. PfBAP 20072010. Analysis of Obsidian, Select Operations: La
Milpa Group B (Op. B1), La Milpa los pisos Courtyard (Op. A2), and Medicinal Trail
(Ops. 7 and 12). In Research Reports from the Programme for Belize Archaeological
Project, Vol. 5. Austin: University of Texas.
Tsukata, M., and Deevey, E. S. 1967. Pollen Analyses from Four Lakes in the Southern
Maya Area of Guatemala and El Salvador. In Cushing, E. J., and Wright, E. (eds.),
Quarternary Paleoecology, pp. 30331. New Haven, CT: Yale University Press.
Tundisi, J. G., and Matsumura-Tundisi, T. 2008. Biodiversity in the Neotropics:
Ecological, Economic and Social Values. Brazilian Journal of Biology 68(4): 91315.
Turner B. L., and Miksicek, C. H. 1984. Economic Plant Species Associated with
Prehistoric Agriculture in the Maya Lowlands. Economic Botany 38: 17993.
Turner, B. L., and Sabloff, J. A. 2012. Classic Period Collapse of the Central Maya
 Lowlands: Insights about HumanEnvironment Relationships for Sustainability.
Proceedings of the National Academy of SciencesUSA 109(35): 1390814.
Tykot, R. H., Van der Merwe1, N. J., and Hammond, N. 1996. Stable Isotope Analysis of
Bone Collagen, Bone Apatite, and Tooth Enamel in the Reconstruction of Human Diet:
A Case Study from Cuello, Belize. Archaeological Chemistry, ACS Symposium Series
625(25): 35565.
Tykot, R. H. 2002. Contribution of Stable Isotope Analysis to Understanding Dietary
Variation among the Maya. In Archaeological Chemistry, ACS Symposium Series
831(14): 21430.
Tykot, R. H., and Staller, J. E. 2002. On the Earliest Introduction of Maize Agriculture in
Coastal Ecuador: New Data from La Emerenciana. Current Anthropology 43(4): 666
77.
Tykot, R. H. 2004. Stable Isotopes and Diet: You Are What You Eat. In Martini, M.,
Milazzo, M., and Piacentini, M. (eds.), Proceedings of the International School of
Physics Enrico Fermi Course CLIV. Amsterdam: IOS Press.
Tykot, R. H. 2006. Isotope Analyses and the Histories of Maize. In Staller, J. E., Tykot, R.
H., and Benz, B. F. (eds.), Histories of Maize: Multidisciplinary Approaches to the
Prehistory, Linguistics, Biogeography, Domestication, and Evolution of Maize, pp. 131
42. Elsevier: Academic Press.
Ugland, K. I., Gray, J. S., and Ellingsen, K. E. 2003. The Species-Accumulation Curve
and Estimation of Species Richness. Journal of Animal Ecology 72: 88897.
UNAVCO. 2012. Geoid Height Calculator, University NAVSTAR Consortium. Available
on the Internet, http://www.unavco.org/community_science/science-
support/geoid/geoid.html, accessed October 18, 2012.
United Nations Development Programme. 2006. Beyond Scarcity: Power, Poverty and the
Global Water Crisis. Paris: United Nations Development Programme.
United States Geological Survey. 2005. Elevation Derivatives for National Applications
(EDNA): EDNA Stage 1 Processing Steps. Available on the Internet,
http://edna.usgs.gov/Edna/stages/stage1.asp
Uribe, D. C. 1988. La Madera: Estudio Anatmico y Catlogo de Especies Mexicanas.
Mexico City: Instituto Nacional de Antropologa e Historia.
Valdes, J. A. 1986. Uaxactun: Recientes Investigaciones. Mexicon 6: 1258.
Valdez, F. 1987. The Ceramics of Colha. Unpublished Ph.D. dissertation. Cambridge, MA:
Harvard University.
Van der Werff, H. 2002. A Synopsis of Ocotea (Lauraceae) in Central America and
Southern Mexico. Annals of the Missouri Botanical Garden 89(3).
Vaughn, H. H., Deevey, H. S., and Garrett-Jones, S. E. 1985. Pollen Stratigraphy of Two
Cores from the Petn Lake District, with An Appendix on Two Deep-Water Cores. In
Pohl, M. D. (ed.), Prehistoric Lowland Maya Environment and Subsistence Economy,
pp. 7390. Papers of the Peabody Museum. Cambridge, MA: Harvard University Press.
Veronda, Winifred 1998. Edwin M. Shook: Incidents in the life of a Maya Archaeologist.
San Marino, CA: Southwestern Academy Press.
Villaseor, I. 2010. Building Materials of the Ancient Maya: A Study of Archaeological
Plasters. Saarbcken: Lambert Academic.
Vogel, A. 2004. History of Dam Engineering. In Wieland, M., Ren, Q., and Tan, J. S.
(eds.), New Developments in Dam Engineering. Proceedings of the 4th International
Conference on Dam Engineering, October 1820, Nanjing, China, Chapter 106. Oxford:
Taylor & Francis.
Vogt, E. Z. 1969. Zinacantan: A Maya Community in the Maya Highlands of Chiapas.
Cambridge, MA: Belknap Press.
WAAS. 2010. Wide-Area Augmentation System Performance Analysis Report No. 32.
Available on the Internet, www.nstb.tc.faa.gov, accessed May 25, 2010.
Wagner, P. L. 1964. Natural Vegetation of Middle America. In West, R. C. (ed.), Natural
Environment and Early Cultures. Handbook of Middle American Indians, Vol. 1.
Austin: University of Texas Press.
Wahl, D., Byrne, R., Schreiner, T., and Hansen, R. 2006. Holocene Vegetation Change in
the Northern Petn and its Implications for Maya Prehistory. Quaternary Research 65:
3809.
Wahl, D., Schreiner, T., Byrne, R., and Hansen, R. 2007. A Paleoecological Record from a
Late Classic Maya Reservoir in the North Petn. Latin American Antiquity 18(2): 212
22.
Webb, E. A., Schwarcz, H. P., Healy, P. F. 2004. Carbon Isotope Evidence for Ancient
Maize Agriculture in the Maya Lowlands. Journal of Archaeological Science 31: 1039
52.
Webb, E., Schwarcz, H., Jensen, C., Terry, R., Moriarty, M., and Emery, K. 2007. Stable
carbon isotopes signature of ancient maize agriculture in the soils of Motul De San Jose
, Guatemala. Geoarchaeology: An International Journal 22(3): 291312.
Webster, D. 1975. Warfare and the Evolution of the State: A Reconsideration. American
Antiquity 40(4): 46470.
Webster, D. 1985. Surplus, Labor, and Stress in Late Classic Maya Society. Journal of
Anthropological Research 41(4): 37599.
Webster, D. 1998. Status Rivalry Warfare: Some Maya-Polynesian Comparisons. In
Feinman, G., and Marcus, J. (eds.), Archaic States, pp. 31152. Santa Fe: School of
American Research.
Webster, D., Freter, A. C., and Gonlin, N. 2000. Copn: The Rise and Fall of an Ancient
Maya Kingdom. Fort Worth, TX: Harcourt College.
Webster, D. 2002a. Groundhogs and Kings: Issues of Divine Kingship among the Classic
Maya. In Love, M., Hatch, M. P., and Escobedo, H. (eds.), Incidents of Archaeology in
Central America and Yucatan, pp. 43358. New York: University Press of America.
Webster, D. 2002b. The Fall of the Ancient Maya: Solving the Mystery of the Maya
 Collapse. Thames & Hudson.
Webster, D. 2002. The Fall of Ancient Maya: Solving the Mystery of the Maya Collapse.
New York: Thames and Hudson.
Webster, D., Silverstein, J., Murtha, T., Martinez, H., and Straight, K. 2004. The Tikal
Earthworks Revisited, Occasional Papers in Anthropology No. 28. Department of
Anthropology, University Park: Penn State University.
Webster, D. 2005. Political Ecology, Political Economy, and the Culture History of
Resource Management at Copn. In Andrews, E. W., and Fash, W. L. (eds.),Copn: The
History of an Ancient Maya Kingdom, pp. 3372. Albuquerque: School of American
Research.
Webster, D., Rue, D., and Traverse, A. 2005. Early Zea Cultivation in Honduras:
Implications for the Iltis Hypothesis. Economic Botany 22(2): 16673.
Webster, D., Murtha, T., Silverstein, J., Martnez, H. R., Terry, R., and Burnett, R. 2007.
The Great Tikal Earthworks Revisited. Journal of Field Archaeology 32(1): 4164.
Webster, D., Silverstein, J., Murtha, T., Martinez, H., and Straight, K. 2008. Political
Ecology of the Tikal Earthworks: A Maya Altepetl Boundary? In Urbanism in
Mesoamerica 1: 34976. Mexico City and University Park: Instituto Nacional de
Antropologia e Historia and Pennsylvania State University.
Webster, D. 2013. Maya Drought and Niche Inheritance. In Gyles Iannone (ed.), The
Great Maya Droughts in Cultural Context. pp. 333-357. Boulder: University of
Colorado Press.
Webster, J. W., Brook, G. A., Railsback, L. B., Cheng, H., Edwards, R. L., Alexander, C.,
and Reeder, P. P. 2007. Stalagmite Evidence from Belize Indicating Significant
Droughts at the Time of Preclassic Abandonment, the Maya Hiatus, and the Classic
Maya Collapse. Palaeogeography, Palaeoclimatology, Palaeoecology 250(14): 117.
Weiss-Krejci, E., and Sabbas, T. 2002. The Potential Role of Small Depressions as Water
Storage Features in the Central Maya Lowlands. Latin American Antiquity 13(3): 343
57.
Weller, E. T. 2006. Satellites, Survey, and Settlement: The Late Classic Maya Utilization
of Bajos (Seasonal Swamps) at Tikal and Yaxha, Guatemala. In Campana, S., and Forte,
M. (eds.), From Space to Place: 2nd International Workshop, CNR, pp. 316. Oxford,
UK: British Archaeological Reports, International Series 1568.
Wernecke, D. C. 2008. A Burning Question: Maya Lime Technology and the Maya Forest.
Journal of Ethnobiology 28(2): 20010.
West, G. 2002. Ceramic Analysis in the Late Classic and Post Classic Maya Lowlands: A
Diachronic Approach. In Masson, M. A., and Friedel, D. A. (eds.), Ancient Maya
Political Economies, pp. 14096. Oxford: AltaMira Press.
Wheeler, E. A. 2011. InsideWood a Web Resource for Hardwood Anatomy. IAWA
Journal 32(2): 199211.
White, C. D., Healy P. F., and Schwarcz, H. P. 1993. Intensive Agriculture, Social Status
 and Maya Diet at Pacbitun, Belize. Journal Anthropological Research 49: 34775.
White, T. J., Bruns, T., Lee, S., and Taylor, J. 1990. Amplification and Direct Sequencing
of Fungal Ribosomal RNA Genes for Phylogenetics. In Innis, M. A., Gelfand, D. H.,
Sninsky J. J., and White T. J. (eds.), PCR Protocols: A Guide to Methods and
Applications, pp. 31522. San Diego: Academic Press.
Whittaker, R. H. 1965. Dominance and Diversity in Land Plant Communities. Science
147: 25060.
Wiersum, K. F. 1997. Indigenous Exploitation and Management of Tropical Forest
Resources: An Evolutionary Continuum in Forest-People Interactions. Agriculture,
Ecosystems and Environment 63: 16.
Wilken, G. C. 1971. Food-Producing Systems Available to the Ancient Maya. American
Antiquity 36(4): 43248.
Wilkinson, T. J. 2014. Comparative Landscape Analysis: Contrasting the Near East and
the Maya. In Chase, A. F. and Scarborough, V. L. (eds.), The Resilience and
Vulnerability of Ancient Landscapes: Transforming Maya Archaeology through IHOPE.
Archaeological Papers of the American Anthropological Association. Malden. MA:
Wiley Periodicals.
Wilkinson, T. J., Boucharlat, Rmy, Ertsen, Maurits W., Gillmore, Gavin, Kennet, Derek,
Magee, Peter, Rezakhani, Khodadad, and De Schacht, Tijs 2012. From Human Niche
Construction to Imperial Power: Long-Term Trends in Ancient Iranian Water Systems.
Water History 4(2): 15576.
Willey, G. R. 1982. Maya Archeology. Science 215: 26067.
Williams, L. O. 1981. The Useful Plants of Central America. Ceiba 24(12).
Williams, T. P. 1977. Comprehensive Index to the Flora of Guatemala. Fieldiana Botany
24(13).
Wilson, S. 2012. Bob Carr in His Own Words. Smoking Mirror 19(11): 35. Newsletter of
the Pre-Columbian Society of Washington, DC Available on the Internet,
www.pcswdc.org
Winchester, J. A., and Floyd, P. A. 1977. Geochemical Discrimination of Different
Magma Series and Their Differentiation Products Using Immobile Elements. Chemical
Geology 20: 32543.
Wingard, J. 1993. The Role of Soils in the Development and Collapse of Classic Maya
Civilization at Copn, Honduras. Ph.D. dissertation, Department of Anthropology.
University Park: Penn State University.
Wingard, J. 1996. Interactions between Demographic Processes and Soil Resources in the
Copn Valley, Honduras. In Fedick, Scott (ed.), The Managed Mosaic Ancient Maya
Agriculture and Resource Use, pp. 20735. Salt Lake City: University of Utah Press.
Winterhalder, B., and Puleston, C. 2012. The Exchequers Guide to Revenue in the
Agrarian State. UC Davis Contribution to the Collaborative Research Project,
Development and Resilience of Complex Socioeconomic Systems: A Theoretical Model
 and Case Study from the Maya Lowlands. Davis: University of California.
Wisdom, C. 1940. The Chorti Indians of Guatemala. Chicago: The University of Chicago
Press.
Wiseman, F. M. 1978. Agricultural and Historical Ecology of the Maya Lowlands. In
Harrison, P. D., and Turner, B. L. II (eds.), Pre-Hispanic Maya Agriculture, pp. 63115.
Albuquerque: University of New Mexico Press.
Wolf, P. R., and Ghilani, C. D. 2006. Elementary Surveying: An Introduction to
Geomatics. Upper Saddle River, NJ: Prentice Hall.
Wood, J. 1998. A Theory of Preindustrial Population Dynamics. Current Anthropology
391(1): 99135.
Woodbury, R. B., and Neely, J. A. 1972. Water Control Systems of the Tehuacan Valley.
In Johnson, F. (ed.), Chronology and Irrigation: The Prehistory of the Tehuacan Valley,
Vol. 4, pp. 81153. Austin: University of Texas Press.
Wright, A. C. S., Romney, D. H., Arbuckle, R. H., and Vial, V. E. 1959. Land in British
Honduras: Report of the British Honduras Land Survey Team. Colonial Research
Publications No. 24. London: Her Majestys Stationary Office.
Wright, D. R., Terry, R. E., Eberl, M. 2009. Soil Properties and Stable Carbon Isotope
Analysis of Landscape Features in the Petexbatn Region of Guatemala.
Geoarchaeology 24(4): 46691.
Wright, H. E. 1993. Core Compression. Limnology and Oceanography 38(3): 699701.
Wright, L. E. 2005. Identifying Immigrants to Tikal, Guatemala: Defining Local
Variability in Strontium Isotope Ratios of Human Tooth Enamel. Journal of
Archaeological Science 32: 55566.
Wynn, J. G., Bird, M. I., and Wong, V. N. L. 2005. Rayleigh Distillation and the Depth
Profile of 13C/12C Ratios of Soil Organic Carbon from Soils of Disparate Texture in
Iron Range National Park, Far North Queensland, Australia. Geochimicaet
Cosmochimica Acta 69(8): 196173.
Wynn, J. G. 2007. Carbon Isotope Fractionation During Decomposition of Organic Matter
in Soils and Paleosols: Implications for Paleoecological Interpretations of Paleosols.
Palaeo 251(3): 43748.
Wynn, J. G., and Bird, M. I. 2008. Environmental Controls on the Stable Carbon Isotopic
Composition of Soil Organic Carbon: Implications for Modelling the Distribution of C3
and C4 Plants, Australia. Tellus B 60(4): 60421.
Yaeger, J., and Hodell, D. A. 2008. The Collapse of Maya Civilization: Assessing the
Interaction of Culture, Climate, and Environment. In Sandweiss, D. H., and Quilter, J.
(eds.), El Nio, Catastrophism, and Culture Change in Ancient America, pp. 187242.
Cambridge, MA: Harvard University Press.
Zheng, D., Rademacher, J., Chen, J., Crow, T., Bresee, M., Le Moine, J., and Ryua, S. R.
2004. Estimating Aboveground Biomass Using Landsat 7 ETM+ Data across a
Managed Landscape in Northern Wisconsin, USA. Remote Sensing of Environment 93:
40211.
 Index
above ground biomass 157, 158, 161, 182
Acacia sp. 172
Acosmium panamense 172
Acrocomia aculeata 172, 174
activity 4, 9, 16, 18, 21, 28, 29, 88, 100, 111, 120, 146, 243, 253, 254, 255, 256, 270
AGB. See above ground biomass
agriculture 1, 4, 10, 18, 21, 30, 33, 34, 35, 40, 41, 45n6, 101, 104, 107, 109, 110, 111, 113,
115, 116, 117, 118, 120, 121, 122, 123, 133, 137, 152, 154, 155, 161, 164, 171, 172,
173, 176, 178, 179, 180, 181, 183, 199, 200, 201, 202, 210, 211, 213, 214, 215, 218,
219, 221, 242, 253, 257, 260, 267, 268, 270, 278
aguada 105, 107, 108, 109, 110, 162, 166, 200, 203, 258, 260, 261, 263, 264, 265, 266,
267, 268, 269, 270, 273, 274, 276, 278, 279
Aguada Benito 105
Aguada de Carlos 105
Aguada de Elmer 14, 105
Aguada de Terminos 13, 14, 98, 100, 101, 105, 107, 108, 112, 116, 119, 120, 121, 122,
179, 181, 191, 200, 203, 204, 288, 289, 293
Aguada La Presa 105
Aguada La Sarteneja 105
Aguada Pital 13, 14
Aguade Terminos 108
Aguateca 153, 167, 283
Akpinar-Ferrand, E. 266
Alibertia edulis 148
Alseis sp. 148
Alvaradoa subovata 172
Amaranthus sp. 273
Ampelocera hottlei 172
AMS 27, 35, 111, 127, 162, 186, 191, 202, 260, 262, 264, 265, 266, 269, 272, 273, 278,
279
AMS dating 48, 56
Anacardiaceae 132, 138, 139, 322
Annaya, A. 218
Apocynaceae 132, 138, 139, 318
Archaic 12, 186, 203, 204, 278
architecture 1, 21, 28, 38, 65, 85, 87, 88, 90, 125, 149, 159, 191, 215, 218, 219, 280
Arecaceae 132, 138, 202
Arroyo Corriental 13, 118, 191
Aspidosperma megalocarpon 135
Aspidosperma sp. 139, 172
Asteraceae 110, 201, 271, 272, 273
Astronium graveolens 172
Attalea cohune 148
axis 80
Bactris major 172
bajo 1, 4, 7, 8, 9, 10, 11, 14, 19, 26, 31, 33, 34, 36, 38, 41, 42, 43, 44n5, 45n6, 95, 96, 97,
98, 99, 100, 101, 102, 104, 105, 110, 112, 113, 116, 117, 118, 119, 120, 121, 122,
123, 128, 131, 132, 134, 137, 156, 157, 161, 162, 164, 165, 166, 168, 171, 177, 178,
179, 181, 183, 184n1, 185n2, 186, 190, 191, 199, 200, 201, 202, 208, 209, 210, 211,
215, 220, 221, 222, 259
Bajo de Azcar 7, 99
Bajo de Santa Fe 14, 50, 96, 97, 98, 99, 100, 102, 104, 105, 110, 111, 112, 116, 119, 120,
121, 122, 148, 164, 170, 200, 210, 287
basalt 191
Beach, T. 261, 262
Bejucal 7, 60, 268, 269, 278, 279
Belize 64, 95, 101, 125, 137, 144, 242, 259, 270, 275, 284
biface 240, 242, 244, 253
biomass 14, 157, 162, 165, 168, 180, 184n1, 185n2, 185n3
blade 240, 254
Blomia prisca 11, 134, 135, 137
Bozarth, Steve 261
Brady, C.B. 66
Breedlove, D. 167
Brookfield, H. 219
Brosimum alicastrum 11, 134, 135, 137, 138, 139, 149, 150, 172
Brosimum sp. 110
burial 44n4, 150, 255
Burseraceae 139, 147
Byrsonima crassifolia 172
cacao. See Theobroma cacao
cache 255
Caesalpinia sp. 172
Calakmul 12, 42, 44n5
calcite 191
Calcite 191
Callender, D. 219
Cameraria latifolia 172
Campbell, D. 125
Canna indica 172, 210, 287
Canna sp. 109
Capsicum annuum 173
Caracol 43, 44n5
Carapa guianensis 172
carbon isotope 112, 186, 199, 200, 262, 268
Carr, Christopher 27, 282
Carr, Robert 65, 66, 73, 89, 169
Casearia sp. 172
causeway 21, 23, 25, 27, 34, 37, 44n3, 67, 88, 90, 99, 160, 169
Cavallaro, D. 127
cave 6, 87, 259
Cecropia spp. 180
Cedrela odorata 135
Ceiba pentandra 172
Celtis iguanaea 172
ceramic 5, 35, 37, 42, 107, 120, 127, 167, 168, 190, 238, 245, 246, 247, 250, 252, 253,
254, 255, 257, 263, 264, 265, 266, 269, 272, 278
ceremonial 4, 126, 148, 167, 178, 238
Cern 153, 167, 173, 174, 176, 180, 181, 183, 210, 236n1
charcoal 11, 30, 32, 36, 106, 108, 110, 117, 150, 173, 179, 201, 262, 265, 270, 272, 273,
278
Cheetham, D. 171
Chenopodium sp. 273
chert 107, 119, 122, 240, 242, 243, 244, 253, 255
Chichancanab 275
Chichn Itz 60, 64, 66
Chloridoids 272
Chlorite 191
chocolate. See Theobroma cacao
chlorite 191
Chorti 125
chronostratigraphy 186
Chrysophyllum sp. 172
Chultun 77
climate 4, 10, 18, 94, 150, 174, 180, 222, 274
climate change 182, 260, 275, 278
Clusia sp. 135, 172
Cob 149, 153
Coe, William 65, 66
collapse 3, 4, 10, 30, 41, 63, 96, 122, 212, 214, 261
commoner 63, 87, 168, 216
community 16, 19, 21, 43, 125, 127, 128, 130, 134, 135, 143, 145, 146, 152, 158, 159,
167, 168, 181, 184, 216, 238, 239
constriction 274
construction 10, 18, 21, 22, 27, 28, 32, 33, 35, 37, 38, 39, 40, 41, 42, 43, 44n4, 44n5,
45n9, 74, 88, 93, 126, 130, 146, 148, 167, 169, 179, 212, 213, 220, 223, 236n4, 245,
253, 261, 265, 274, 278, 279
Copn 29, 153, 170, 214, 215
coring 13, 18, 25, 31, 33, 34, 46, 47, 48, 49, 53, 55, 186, 201
Corriental 7, 13, 14, 30, 31, 32, 36, 39, 40, 43, 45n9, 49, 58, 80, 89, 96, 116, 118, 119,
122, 156, 177, 180, 186, 191, 196, 202, 209, 210, 211, 242, 244, 245
Corriental Reservoir 30, 32, 52, 53, 55, 56, 57, 118, 289
court 214, 217
courtyard 35, 254
Cowgill, U.M. 98
Croton billbergianus 11
Croton sp. 135, 137, 172
Cryosophila stauracantha 11, 134, 135, 137
Cucurbita moschata 172
Cucurbita pepo 172, 174
Cucurbita sp. 173, 174, 268, 272, 273, 278, 287
Culbert, T. 250
Cupania belizensis 135
Cyperus canus 172
deforestation 38, 67, 173, 199
DEM 81, 85, 90, 91
diversity 125, 126, 128, 130, 133, 135, 137, 142, 143, 145, 148, 150, 174, 178
Dixon, C. 211
DNA 129, 151n1
Dos Aguadas 160
Dos Pilas 153
drought 1, 10, 19, 21, 22, 32, 39, 44n1, 121, 182, 184, 210, 261, 292, 293
dry cores 186
Dunning, Nick 78, 289
dwellings 168, 171, 176
Early Classic 12, 27, 32, 39, 40, 42, 43, 120, 187, 196, 203, 204, 210, 242, 247, 256, 261,
263, 264, 265, 266, 267, 268, 269, 270, 272, 274, 276, 278
Early Postclassic 12, 109
Early Preclassic 12, 35, 186
East Brecha 13, 97, 99, 102, 105, 107, 110, 112, 117
economy
economic 158, 167, 169, 210, 214, 215
EDM 50
El Diablo 268, 269, 274, 276, 278, 279
El Encanto 160
El Jaguar 240, 244, 245, 246, 253, 254
El Palmar 159, 160, 259, 261, 262, 274, 275, 278, 279, 283
El Peru 261, 279, 283
El Salvador 167, 173, 196, 210
El Zotz 15, 159, 160, 215, 258, 260, 261, 262, 263, 265, 267, 268, 269, 270, 271, 273,
274, 275, 276, 277, 278, 279, 283, 293
elite 2, 16, 22, 146, 167, 169, 170, 214, 215, 216, 217, 218, 236n4
Enterolobium cyclocarpum 172
Environmentalists Subsoil Probe. See ESP
erosion 31, 32, 36, 41, 98, 104, 117, 119, 121, 122, 180, 211, 220, 221, 222, 259, 274, 279
Erythrina spp. 172
ESP 46, 50, 51, 54
Eugenia spp. 172
Euphorbiaceae 132, 138, 139
Fabaceae 132, 138, 139
Fialko, Vilma 3, 5, 29, 44n5, 99, 102, 119, 120, 121
Ficus sp. 135, 172
firewood 147, 148, 167, 168, 169, 178, 182
flake 240, 242, 243
Forchhammeria trifoliata 135
forest cover 164, 178, 185n2
Fry, Robert 161
fuel 104, 130, 146, 152, 153, 164, 167, 168, 170, 178, 182, 183, 184, 200, 236n4
Gallopin, G. 57
Garcinia sp. 172
Genipa americana 148
GIS 3, 14, 56, 59, 71, 76, 89, 94, 100, 157, 160, 164, 282
Glassman, A. 218
Gliricidia sepium 172
Global Positioning System. See GPS
Gossypium hirsutum 172, 287
GPS 59, 69, 71, 72, 73, 74, 77, 78, 79, 84, 86, 98, 99, 128, 156
Grazioso, Liwy 219
Griffin, R. 217
group
groups 16, 43, 69, 113, 132, 146, 254, 257, 277
Guarea glabra 135, 172
Guderjan, T. 261
Guettarda combsii 148, 172
Gymnanthes lucida 135
Haematoxylum campechianum 11, 121, 135, 137, 138, 139, 172
Harrison, P.D. 27
Haviland, W. 21, 45n10, 166
Hazard, James 65, 66, 89, 169
hectares 157, 222
Heliocarpus sp. 172
hiatus 12, 41, 42, 43, 187
Hirtella sp. 172
Hockaday, B. 210
Holmul River 5, 7, 10, 37, 44n5, 97, 99, 122, 159
Holocene 9, 36, 191, 196
Honduras 190
Hood, Angela 154
household 21, 87, 90, 112, 113, 115, 120, 130, 167, 168, 173, 174, 178, 216, 217, 245,
253, 257
Huastec 125, 146
Huff, W. 191
hunting 146
Hutchinson, G.E. 98
hydraulic systems 46
hydrologic modeling 14, 94
Ilopango 196
impact scars 242, 243, 253
inscription 35, 36, 42, 45n8, 60, 158, 177, 191, 212, 217
Inscriptions Reservoir 287
Ipomoea batatas 172, 287
Itza 64, 123n1, 125, 146, 164, 165, 166
Ixlu 160
Jaccard, P. 131
jade 35, 240, 244, 254, 257
Jasaw Chan Kawiil 41
Jimbal 160
Jones, Morris 66
kaolinite 190, 191, 196, 273
Killion, T. 172
Kilmartin, Jerome 64, 66
Lacandon 125
Lacmellea sp. 172
Lake Peten Itza 285
Late Classic 3, 11, 12, 19, 22, 23, 27, 28, 32, 33, 34, 35, 37, 40, 41, 42, 43, 44n4, 44n5,
45n10, 89, 90, 94, 101, 106, 108, 109, 110, 120, 139, 148, 152, 153, 157, 159, 160,
161, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 178, 179, 180, 182, 183, 184,
185n4, 187, 199, 200, 201, 203, 213, 215, 217, 219, 247, 263, 264, 266, 267, 270,
272, 274, 275, 278, 279, 280, 282, 283, 284, 288, 289, 292, 294
Late Postclassic 12, 203
Late Preclassic 3, 12, 18, 27, 28, 32, 35, 37, 38, 39, 40, 41, 43, 44n2, 44n5, 45n9, 106,
107, 109, 117, 119, 153, 157, 159, 160, 161, 164, 187, 200, 203, 208, 210, 256, 261,
265, 275, 278
Laughlin, R. 167
Lauraceae 139, 146, 147
Lentz, David 127, 138, 154, 210, 286, 289
Levine, Newton 65
Licaria campechiana 147, 172
Licaria sp. 138, 139, 146
limestone 18, 26, 32, 36, 38, 96, 100, 105, 112, 113, 169, 190, 191, 211, 242, 264, 266,
268, 274
lithics 5, 120, 125, 238, 257
Lonchocarpus heptaphyllus 135
Lonchocarpus sp. 172
lowlands 1, 4, 5, 9, 10, 11, 12, 19, 26, 29, 32, 86, 87, 94, 95, 103, 105, 119, 143, 173, 180,
186, 187, 190, 191, 211, 236n1, 258, 260, 282
Luzzadder-Beach, S. 261, 262
maize 40, 109, 110, 111, 112, 113, 116, 117, 118, 120, 121, 154, 167, 170, 171, 172, 173,
174, 176, 177, 178, 179, 181, 183, 199, 201, 210, 211, 215, 220, 223, 267, 268, 271,
278, 279, 283, 288
Maler Convention 64, 69
Maler, Theobert 4
Manihot esculenta 174, 287
Manilkara zapota 11, 135, 137, 138, 139, 149, 172, 284
manioc 174, 210
mano 244, 253, 254, 255
Margaritaria nobilis 135
marine 215
Martinez, Eduardo 65
Maudsley, Alfred 4
Mayapn 60, 63, 64, 65, 66, 67, 69
McAnany, P. 172, 290
Meliaceae 132, 320
Merida 74
metate 244, 253, 254, 255
Metopium brownei 135, 172
Mexico 27, 65, 74, 95, 137, 143, 152, 171, 196, 211, 259, 265
Miconia argentea 180
Middle Classic 120, 137
Middle Preclassic 12, 37, 39, 119, 203, 216, 259, 261
Mirador Basin 6, 12, 187, 216
Missouri Botanical Garden Herbarium 129
moats 218
modern forest 125, 127, 129, 130, 131, 138, 139, 140, 141, 142, 144, 146, 147, 148, 149,
150, 162, 177
monument
statue 74, 84
Moraceae 110, 132, 138, 139, 202, 325
Morinda sp. 148, 172
Morley, Sylvanus 4, 65
morphology 238, 239, 240
Murtha, Timothy 15, 39, 121, 166, 212, 290
Myrtaceae 110, 132, 320
Nagy, A. 53
Nakum 159, 160
NASA 80, 83, 98
National Center for Biotechnology Information 129
natrolite 191
Nectandra sp. 135, 139, 146, 147, 172
noble 214
Nohmul 64
ONeill, John P. 64
obsidian 240, 253, 257
Ocotea puberula 172
Ocotea sp. 146
Olson, G. 98, 220
organization 216, 218
oxidized
oxidizing 270
Pachira aquatica 271
palace 1, 2, 12, 14, 23, 25, 26, 27, 28, 30, 34, 35, 36, 37, 38, 40, 41, 42, 43, 88, 92, 93,
191, 242, 245, 246, 253, 255, 265
Palace Reservoir 283
Palenque 25, 42
PaleoIndian 12, 186
Panicoids 272
Parris Convention 64, 67, 69
Parris, Fred 63, 64
pedon 111, 112, 115, 116
Perdido 7, 12, 13, 14, 30, 31, 33, 34, 36, 39, 40, 42, 45n6, 45n9, 49, 56, 58, 80, 82, 89, 96,
 111, 112, 113, 116, 117, 118, 120, 122, 137, 156, 176, 181, 191, 196, 202, 209, 210,
240, 242, 245, 250
Perdido Reservoir 52, 56, 57, 291
Persea americana 147, 172
Petn 3, 29, 67, 71, 73, 74, 87, 95, 96, 100, 101, 105, 111, 124, 146, 155, 185n5, 258, 261,
266, 272, 293
Peten Itza 14, 162, 165, 166, 178, 214
Petexbatun 214, 259
Petty Benchmark 60, 73, 74, 84
Phaseolus coccineus 172
Phaseolus lunatus 172
Phaseolus vulgaris 172
Piedras Negras 60, 63, 64, 65, 67
Pielou 130, 135
Pimenta dioica 135, 172
Pinaceae 139, 148
Pinus sp. 139, 148, 172, 271, 272
Piper sp. 172
Piscidia piscipula 172
plaster 21, 30, 33, 39, 40, 88, 167, 169, 173, 191, 260, 263, 265, 266, 269
plaza 27, 33, 45n7, 60, 67, 69, 80, 81, 82, 87, 88, 97, 107, 113, 117, 169, 208
Pleistocene 36, 100, 186, 191
Poaceae 109, 201, 202, 210
pocket bajo 36, 96, 115, 118, 200, 203, 278
politics 4, 12, 22, 42, 87, 88, 120, 150, 152, 153, 158, 159, 160, 164, 167, 169, 187, 199,
214, 216, 217, 218, 236n4, 254, 258
pollen 14, 108, 109, 110, 121, 143, 150, 153, 154, 162, 164, 166, 174, 177, 178, 181, 200,
201, 202, 210, 211, 260, 261, 266, 268, 270, 272, 273
Polygonaceae 109, 201, 210
Pooids 272
Postclassic 12, 30, 110, 121, 150, 201, 203, 204, 208, 210, 211, 261, 279, 284
Pouteria amygdalina 135, 149
Pouteria reticulata 11, 134, 135, 137, 149
Pouteria sapota 172
Pouteria sp. 139, 172
power 12, 160, 164, 236n4
Preclassic 12, 19, 28, 29, 32, 37, 38, 39, 40, 41, 45n9, 101, 107, 109, 119, 122, 137, 139,
160, 161, 164, 199, 201, 203, 204, 210, 214, 216, 247, 259, 261, 265, 268, 275, 279
production 1, 21, 65, 66, 90, 112, 113, 120, 121, 150, 159, 164, 169, 170, 176, 179, 211,
213, 220, 223, 239, 245, 252, 255, 257, 260
Proskouriakoff, Tatiana 4, 67
Protium copal 135, 139, 147, 172, 271, 273, 278
Protium Copal 138
provenience 186
Pseudolmedia glabrata 172
Pseudolmedia sp. 134, 135, 137
Psychotria carthagenensis 148
Psychotria lundellii 148
Psychotria mexiae 148
Psychotria sp. 148, 172
Pucte 13, 14, 105, 262, 277
Puleston, Cedric 213
Puleston, Dennis 4, 97, 98, 99, 107, 161, 213, 219
purification 191
Purrn Dam 27
Qeqchi 125
quartz 30, 191, 194, 195, 196, 274
radiocarbon 100, 103, 105, 108, 110, 117, 119, 127, 162, 186, 202, 262, 265
rainfall 1, 9, 10, 19, 21, 22, 25, 32, 40, 41, 143, 182, 184, 213, 277
Ramonal 116, 160, 220
regions 12, 111, 129, 143, 157, 215, 222, 240
relationships 158, 184n1
relative dating 269
religious 87, 88, 146, 147, 254
reservoir 2, 4, 5, 14, 19, 21, 22, 23, 25, 27, 28, 30, 31, 34, 52, 88, 91, 93, 94, 107, 116,
120, 121, 122, 124, 153, 154, 169, 177, 178, 180, 181, 183, 184, 186, 190, 191, 196,
 197, 199, 200, 202, 203, 209, 210, 211, 238, 253, 256, 258, 259, 260, 274, 281, 289
Rio Azul 7, 153, 259, 288
Rio Holmul 288
ritual 35, 126, 130, 217, 246, 254, 256, 257, 273, 278
river 5, 10, 22, 99, 215
root crops 121, 173, 176, 177, 179, 181, 210
Rubiaceae 139, 148
rulers 1, 2, 12, 159, 216, 217, 236n4
Sabal mauritiiformis 11, 135, 137
Salix sp. 172
Salvinia sp. 109, 202
San Bartolo 99, 174, 215, 277
San Pedro River 259
Sapindaceae 132, 297
Sapotaceae 110, 132, 138, 139, 149, 311, 316, 325
sascab 169, 264, 269
Scarborough, Vernon 17, 219, 274
Schulze, M.D. 133
Scleria bracteata 11
scraper 240, 242
scribes 16
Sebastiana sp. 138, 139, 172
Serrana Macanche 1
Shannon-Wiener Index 130, 134
Sheets, Payson 196, 211, 236n1
sherds 34, 101, 105, 107, 245, 246, 247, 252, 253, 254, 264, 266, 273
Shook, Edwin 4, 65, 66, 67, 68
Sideroxylon sp. 149, 172
silting tank 27, 29, 31, 37, 42, 44n4, 202, 209
Simira salvadorensis 135, 148
Simmons, C.S. 97
Simpsons Index 130
smectite 100, 190, 191, 196, 273
socioeconomic values 130, 140, 144, 280
soil organic matter. See SOM
SOM 111, 116, 123n2, 186
Spondias cf. purpurea 172
Spondias mombin 135
Spondias sp. 174
squash. See Cucurbita sp.
status 109, 145, 169, 212, 216, 236n2, 253, 254
Steggerda, M. 172
Stemmadenia sp. 172
stratigraphic 31, 36, 37, 39, 261, 267, 274
Stuart, David 29
subsistence 4, 5, 125, 141, 173, 210, 211, 214, 217, 254
summit 19, 21, 22, 23, 25, 28, 36, 38, 40, 41, 42, 43, 112
summit-ridge 21
Swietenia macrophylla 284
Tabebuia sp. 172
Tankersley, Kenneth 273, 289
Tapachula 74
Tapirira sp. 172
Tayasal 30
Tecoma stans 172
Temple Reservoir 53, 55, 58, 283
Teotihuacan 12, 29, 242
Terminal Classic 12, 109, 110, 121, 201, 204, 208, 221, 247, 256, 261, 293
Terminal Preclassic 12
Terminalia buceras 172, 271
textiles 256
Theobroma cacao 143, 172, 287
Thevetia ahouai 172
Thompson, J. 212
Thompson, Kim 156
Tikal National Park 67, 86, 97, 99, 102, 110, 111, 128, 134, 157
Tikal Reservoir 13, 14, 34, 35, 177
time period 65, 159, 164, 173
toeslopes 116, 120, 221, 223
tomb 2
tools 69, 93, 130, 164, 242, 244, 253, 254, 255
topographic 14, 64, 84, 85, 89, 92, 99, 112, 123, 218, 223, 274
Topoxte 30
Tozzer, Alfred 4
trace elements 196
transitional forest 11, 132, 137, 149, 185n2
Trema micrantha 180
Trichilia hirta 172
Trichilia minutiflora 11, 134, 135, 137
Trophis spp. 172
Tutankhamen 2
Uaxactn 65, 159, 160, 215
UCAPT 2, 3, 127, 128, 138, 143, 238, 240, 242, 244, 246, 252, 254, 256, 257, 281, 284
Universal Transverse Mercator. See UTM
University of Cincinnati Herbarium 129
upland 6, 7, 8, 10, 19, 41, 95, 96, 97, 100, 102, 110, 117, 119, 121, 122, 123, 128, 131,
132, 133, 134, 137, 139, 147, 155, 157, 161, 164, 165, 166, 168, 169, 171, 172, 178,
179, 182, 183, 184n1, 185n2, 200, 208, 215, 220, 221, 222, 261, 263, 279
use-wear 240
UTM 60, 71, 72, 74, 75, 76
Vaca del Monte 13, 14, 105, 108, 110, 116, 120, 162, 165, 166, 178, 201, 202, 208, 209,
288
vessels 167, 246, 247, 252, 254, 255
volcano 196
volcanogenic 58, 191, 194, 196, 289
Voronoi 160, 161, 164, 171
warfare 158
weaving 239, 245, 253, 256
Webster, David 15, 39, 121, 166, 212, 290
Weller, E.T. 99
wet cores 48, 54
wetland 50
wetlands 4, 7, 29, 95, 98, 103, 104, 122, 157, 184n1, 200, 202, 215, 221, 259, 260
Whitacre, D.F. 133
Wild Cane Cay 153
Winterhalder, B. 213
Wright, D. 267
Wright, H. E. 54
Wurman, Richard 65
Xanthosoma sagittifolium 172, 174, 210, 287
Xculoc 21
Xoxocotln 27
X-ray diffraction. See XRD
X-ray Fluorescence Spectrometry. See XRF
XRD 57, 190, 191, 196
XRF 57, 191, 196, 197
Xultun 159, 160, 266
Yaxha 159, 160, 215, 236n6
Yikin Chan Kawiil 41
Yojoa 190
Yukatek 125, 168
Zanthoxylum caribaeum 148, 172
Zea mays. See maize
zeolite 191
Zocotzal 160
zone 11, 14, 21, 23, 41, 42, 45n10, 74, 75, 76, 80, 96, 97, 108, 109, 110, 111, 112, 113,
116, 120, 133, 139, 148, 157, 159, 162, 165, 166, 178, 182, 183, 185n4, 200, 201,
202, 210, 218, 222, 263, 264, 270, 272
Zuelania guidonia 172