Surface Albedo and Desertification

Author(s): Ray D. Jackson, Sherwood B. Idso and J. Otterman

Source: Science, New Series, Vol. 189, No. 4207 (Sep. 19, 1975), pp. 1012-1015
Published by: American Association for the Advancement of Science
Stable URL: http://www.jstor.org/stable/1740662
Accessed: 01-11-2017 13:01 UTC

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/1740662?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

American Association for the Advancement of Science is collaborating with JSTOR to

digitize, preserve and extend access to Science

This content downloaded from 195.43.3.137 on Wed, 01 Nov 2017 13:01:55 UTC
All use subject to http://about.jstor.org/terms
 Because
Becausethe
theresults
results
reported
reported
by Fioren-
by Fioren- monocular
monocularparalysis,
paralysis,
many
many
simple
simple
cells in
cells in
tini and Maffei were limited to cells whose - 50 the
the visual
visualcortex
cortex
may
may
be driven
be driven
by Y-cells
by Y-cells
receptive fields fell within 5? of the area of the LGNd.
centralis, we were initially concerned pri- DAWN L. BROWN

marily with units whose receptive fields 30

WALTER L. SALINGER

were in the central area of visual space. Al- Department of Psychology,

though cells with more peripheral receptiveS 20 University of North Carolina,
X (16) Greensboro 27412
fields were sampled, only those within the 10 (40)

binocular segment were included in this re- References and Notes

port. Recordings were made at either 1 to 3 0?-10? 10?-20? 20?-M
1. T. N. Wiesel and D. H. Hubel, J. Neurophysiol. 26,
days after surgery (acute monocular paral- Eccentricity (degrees)
1003 (1963).
ysis) or at periods of more than 14 days Fig. 2. Same data as in 2. D. H. Hubel and T. N. Wiesel, J. Physiol. (Lond.)
Fig,
206, 419 (1970).1, expressed as a
(chronic monocular paralysis). These in- function of eccentricity of
3. T. N. Wieselreceptive field loca-
and D. H. Hubel, J. Neurophysiol. 28,
tervals were chosen to place our observa-tion. Eccentricities run from the vertical meri- 1029 (1965); S. M. Sherman, K.-P. Hoffmann, J.
S. Stone, ibid. 35, 532 (1972); B. G. Wickelgren
tions clearly on either side of the period re-dian (0?) to the beginning of the monocular seg- and P. Sterling, Nature (Lond.) 224, 1032 (1969);
ment (M). The frequency of X-units is expressed P. Sterling and B. G. Wickelgren, Brain Behav.
ported previously to be critical for changes Evol. 3, 1210 (1970); B. G. Wickelgren and P. Ster-
as a function of the total population (frequency
in simple cortical cells (5). ling, J. Neurophysiol. 32, 16 (1969).
of X fields plus frequency of Y fields) at each lo- 4. H. A. Buchtel, G. Berlucchi, G. G. Mascetti, Brain
Of the 124 LGNd units which received cation. The numbers in parentheses refer to the Res. 37, 355 (1972).
total number of fields from which the percentage 5. A. Fiorentini and L. Maffei, Vision Res. 14, 217
visual information from the paralyzed eye,
of X fields was calculated at each point. Follow- (1974).
119 could be classified with confidence as 6. J. S. Stone, Invest. Ophthalmol. 11, 338 (1972).
ing chronic monocular paralysis, X-cells in the 7. J. Orem, M. Schlag-Rey, J. Schlag, J. Exp. Neurol.
X or Y. The few units which could not be
LGNd were encountered with markedly dimin- 40, 784 (1973).
classified occurred with roughly equal fre-
ished frequency at all retinal eccentricities in the 8. R. Fernald and R. Chase, Vision Res. 11, 95
(1971).
quency in both chronic and acute condi- binocular segment.
9. C. Enroth-Cugell and J. G. Robson, J. Physiol.
tions and were excluded from the analysis. (Lond.) 187, 517 (1966).
10. B. G. Cleland, M. W. Dubin, W. R. Levick, ibid.
Figure 1 summarizes the data obtained 217, 473(1971).
from both conditions. As in other reportssidered in the parallel processing model 11. Y. Fukada, Vision Res. 11, 209 (1971).
12. __ and H. A. Saito, ibid., p. 227.
(6). According to Fiorentini and Maffei 13. B. G. Cleland, W. R. Levick, K. J. Sanderson, J.
(16), in acutely paralyzed animals, X-units
(5), a certain proportion of simple cells 14. J.Physiol. (Lond.) 228, 649 (1973).
represent approximately 51 percent of the S. Stone and K.-P. Hoffmann, Brain Res. 32,
continue to be driven by the paralyzed eye
cells, while Y-cells constitute the remain- 454(1971).
15. M. W. Dubin, paper presented at the meeting of
ing 49 percent. When recordings wereafter chronic monocular paralysis. It the Association for Research in Vision and Oph-
seems unlikely that these cells receive their
made at more than 14 days after paralysis, thalmology, Inc., Sarasota, Florida, 28 April to 2
May 1974.
input from LGNd X-cells, since so few X- 16. K.-P. Hoffmann, J. S. Stone, S. M. Sherman, J.
these percentages were drastically altered.
Only 7 percent of the units encountered cells remain responsive to stimulation via Neurophysiol. 35, 518 (1972).

were X-cells, while the remaining 93 per-the paralyzed eye. This suggests that after 25 March 1975; revised 18 April 1975
cent were Y-cells.
In Fig. 2, the fraction of recorded units
which were X-cells is displayed as a func-
tion of eccentricity. Although the X-cellSurface Albedo and Desertification
loss appears to be present in all areas of the
binocular segment, the small sample size in Otterman
Otterman(1) (1)has
hasproposed
proposed that
thatdeserti- Phoenix,
Phoenix,Arizona.
deserti- Arizona. ForForabout
about
a week,
a week,
we we
the most peripheral areas limits confidencefication
fication in
inregions
regionsofofmarginal
marginal rainfall
rainfall
kept
kept four
fourdifferent
differentsoils
soils
in 25-m2
in 25-m2
plotsplots
very very
with respect to our estimate of the loss inmaymay be
be due
duetotoan
anincrease
increase in in
surface
surface wet
wet by
al- al- byspraying
spraying them
them withwith
water
water
everyevery
this region. Our estimate of the magnitudebedo
bedo caused
causedby
bythetheremoval
removal ofofvegetation morning,
morning,noon,
vegetation noon,and andnight.
night.
During
During
this this
of the overall loss of X-cells is probably by by overgrazing.
overgrazing.He Hehypothesizes
hypothesizes that, time,
time,air
that, airtemperature
temperature (TA),
(TA),
surface
surface
soil soil
conservative because, in the chronicallywhen when high-albedo
high-albedosoils
soils
are
are
denuded,
denuded, thethe temperatures
temperatures(Ts),
re- re- (Ts),
andandalbedos
albedos
werewere
mea-mea-
paralyzed cats, we sampled proportion- sultant increase in surface albedo causes sured
suredevery
every2020
minutes
minutes
with
with
fine-wire
fine-wire
cop- cop-
per-constantan
per-constantan
ately more from those regions of thelower surface temperatures, which in turn thermocouples
thermocouples
and and
upright
upright
LGNd which normally have a higher con-reduce the heat input to the lower atmo- and inverted solarimeters. The soils were

centration of X-cells (16). then allowed to dry. When the volumetric

sphere, decrease its temperature lapse rate,
These results indicate that the number ofand hence somewhat reduce convective ac- water contents of their upper 2 cm were be-
functional X-cells in the LGNd was re- tween 2 and 4 percent, similar data were
tivity leading to rainfall. Over a period of
duced by chronic monocular paralysis. Theseveral years we have measured albedosobtained for another week. Details of this
and surface temperatures of soils and
attrition of the X-cell population provides work and closely allied experiments have
a new demonstration of neural plasticity plants
in been reported elsewhere (3).
in the Sonoran Desert climate of the
the adult visual system. The data are in southwestern United States, and, from the Figure 1 shows the average difference
between Ts and TA for the four different
results of our investigations, we would pre-
large part consistent with the parallel pro-
cessing model for connectivity within thedict that the denuded surface would be soils when wet and when dry at 1400 local
warmer than the vegetated one. Since the time as a function of their mean albedo at
visual system (6). Monocular paralysis si-
multaneously disrupts binocularity among that time. When the soils are wet, Ts is
importance of correctly identifying the cli-
matological mechanisms of desertification
simple cells in the visual cortex and pro- nearly equal to TA and is insensitive to al-
duces loss of X-cells in the LGNd, while cannot
at be overemphasized in light of thebedo variations. When the soils are dry,
the same time apparently leaving unaf- devastation and human suffering caused however,
by T, rises far above TA and is very
fected complex cells in the visual cortex sensitive to albedo. Indeed, the predicted
these processes in the Sahel (2), we believe
and Y-cells in LGNd. that further analysis is warranted. drop in Ts relative to TA in traversing the
At the same time, however, the resultsThe primary data to be used in this con- postulated albedo range 0.25 to 0.37, sug-
text were obtained in May-June 1974 at
raise a possibility which has not been con- gested by Otterman (1) as representative of
1012 SCIENCE, VOL. 189

This content downloaded from 195.43.3.137 on Wed, 01 Nov 2017 13:01:55 UTC
All use subject to http://about.jstor.org/terms
 the vegetated and bare soils he investi- (o 0.98). Using these values
significantly in thealong with
form of latent heat, re- a
gated, is more than enough (13?C) to ac- ducingthe
representative value of the temperature,
background but it can differ
sky
count for the apparent temperature differ- radiation (9), we calculate that
sharply between Otterman's
types of plants.
ences he measured (3.5? to 5?C). However, approach could have When
introduced relative
the plant canopy covers only a
Fig. 1 applies only to bare soil surfaces. temperature errors as great
portion as 8?C,
of the ground, consid-
quite logically an in-
When the land is covered by vegetation, erably larger than the
creased differences
infrared reflectance of thehe re-
vegetat-
relationships are very different. For non- ported for the bare and
ed area isvegetated areas.
expected with increasing ground
stressed plants, leaf temperature tends to In light of these considerations, we
cover fraction, plant projection, be-
height,
differ from air temperature by only a few lieve that there is little justification
and leaf for the
surface index of vegetation (4).
degrees-leaf temperature is slightly above Accompanying
desertification mechanism this increased reflectance
proposed by Ot-
air temperature for air temperatures below terman. Indeed, ourin analysis tends
the infrared is a decreased to in-
reflectance in
330C and slightly below for air temper- dicate that the denuding of
the red portion of soil may have
the spectrum.
atures above 33?C (4). Even if water is se- The sharp contrast
thermal and climatic effects just between
the theoppo-
Sinai
verely limited, leaf temperatures generally site of those that he
andhas postulated.
the Negev Until
was one of the preliminary
rise less than 10?C above air temperature actual on-site measurements
"quick look" results of
of therainfall
Israeli ERTS-1
(4, 5), particularly those of small-leaved and of surface temperatures based upon
(Earth Resources Technology Satellite)
plants (6) of the type Otterman mentions. correct surface emissivities show
program (5). The observed low other-
reflectance
Thus, since our data indicate that surface wise, the validity of inthe rainfall
the MSS-7 reduction
(multispectral scanner) in-
soil temperatures would still be 10?C above mechanism Otterman postulates should
frared band, 0.8 to 1.1 Im, of the area withbe
air temperature even at the highest albedo seriously questioned. "an appreciable vegetation cover" (1) in
value he postulates, it is difficult to see how RAY D. JACKSON the western Negev was totally unexpected,
an area with an "appreciable vegetation SHERWOOD B. IDSO and indeed can be referred to as the Negev
cover" could become warmer than an adja- U.S. Water Conservation Laboratory, infrared reflectance paradox. The mea-
cent bare soil area during midday. Phoenix, Arizona 85040 surements of the contrast between the
Nevertheless, Otterman hasopresented Sinai and the Negev made with a micro-
References and Notes
temperature measurements indicating that densitometer scan were reported in Science
the vegetated surface he studied was 1. J. Otterman, Science 186, 531 (1974).
2. N. Wade, ibid. 185, 234 (1974). (6). More accurate computation was re-
warmer than the adjacent bare soil. These 3. S. B. Idso, R. J. Reginato, R. D. Jackson, B. A. cently carried out (7) directly from the
Kimball, F. S. Nakayama, Soil Sci. Soc. Am. Proc.
temperature measurements were obtained 38, 831 (1974); S. B. Idso, R. D. Jackson, R. J. ERTS-1 digital tapes, from pairs of points
by infrared thermometry. To use this tech- Reginato, B. A. Kimball, F. S. Nakayama, J. Appl.
or small sampling areas on the two sides of
Meteorol. 14, 109(1975).
nique successfully, the emissivity of the4. E. T. Linacre, Agric. Meteorol. 1, 66 (1964). the demarcation line [see figure 1 in (6)].
surface viewed over the spectral sensitivity 5. W. L. Ehrler and C. H. M. van Bavel, Agron. J. 59, The contrast between the Sinai and the
243 (1967); J. H. Palmer, Agric. Meteorol. 4, 39
region of the infrared thermometer must (1967).
Negev in the MSS-7 infrared band ranges
be known. Otterman reports that he as- 18,B.171
6. S. Idso, D. G. Baker, D. M. Gates, Adv. Agron.
(1966). from 1.73 for a pair of points some 15 km
sumed a value of 0.9 for both the bare and 7. S. B. Idso and R. D. Jackson, J. Appl. Meteorol. from the Mediterranean to 1.27 for an in-
vegetated areas he compared. Emissivity 8, 168 (1969); S. B. Idso, R. D. Jackson, W. L.
Ehrler, S. T. Mitchell, Ecology 50, 899 (1969). land pair where precipitation is less and the
measurements made by us (7) and others 8. K. J. K. Buettner and C. D. Kern, J. Geophys. Res. soil is more sandy. In the MSS-5 band, or-
(8), however, indicate that for a desert soil 70, 1329 (1965); M. Fuchs and C. B. Tanner,
Agron. J. 58, 597 (1966). ange-red, the appropriate contrasts are
of albedo 0.37 the emissivity could well be 9. S. B. ldso and R. D. Jackson, Agron. J. 60, 388 1.88 and 1.30, somewhat higher than in the
(1968).
lower than 0.9 (o 0.88), and that for vege-10. Contribution from the Agricultural Research Serv- infrared but not drastically so, as might be
ice, U.S. Department of Agriculture.
tated surfaces it is considerably larger expected (4).
4 December 1974
The studies, which began in the winter of
1972-1973, of temperature relationships
30 - Jackson and Idso's presentation of leaf on both sides of the demarcation line and
temperatures versus air and bare soil tem- of the possible meteorological-climatolog-
26 -
\ peratures (1) is consistent with a concise ical effects were initiated only as a result of
22 -
statement by Gates: "A dark, dry loam in observing the Negev infrared reflectance
\ dry full sunlight may have surface temper- paradox. I agree with Jackson and Idso
_ i c
18I _
o Navajo clay
0\ atures as high as 50?C when the air temper- that "it is difficult to see how an area with
C-)
ature at a height of 2 m is 30?C. A stand of an 'appreciable vegetation cover' could be-
* Avondale loam
t- 14
o Gran sandy loam vegetation nearby will have leaf temper-
A Cashion silty clay atures near or below air temperature" (2).
- 10
Indeed, these typical temperature charac-
teristics of healthy leaves are a direct result
6
of two factors. One of these is high albedo
wet
2 - ?
in the infrared: "the very abrupt increase in
* u reflectance near 0.7 # and the fairly abrupt
-2 decrease near 1.5 i are present for all ma-
0.1 0.22 00 0.14 .18 0.38
ture, healthy, green leaves" (2). The in-
Albedo
frared radiation is 50 percent or more of
Fig. 1. Mean values of the the solarsurface soil
insolation at the temper-
ground (3), and
ature minus the air temperature (T,
the high albedo in this part of -
the TA) at
spectrum
1400 local time for four different bare soils kept
reduces
very wet (near saturation) the radiation
for 1 week absorbed
and by the
very
leaves.
dry (2 to 4 percent water The second by
content factor, which can befor
volume)
Fig. 1. Ground view of the western Negev (22
another week plotted as a function of the mean
dominant, is evapotranspiration. This fac- June 1974), showing vegetation growing in
soil albedos for the same times. clumps. Tamarisk trees are only a rare occur-
tor effectively causes the heat output to be rence.
19 SEPTEMBER 1975
1013

This content downloaded from 195.43.3.137 on Wed, 01 Nov 2017 13:01:55 UTC
All use subject to http://about.jstor.org/terms
 MEDITERRANEAN NORTH WESTERN NEGEV
-
*
ei w

a, 0 S S 9 0 - - - $ _a , ,C9 9 - o.o.o CI- .o faru-Nuto

oaoo--rvru

-'-?-?-?

u P\)-?rVM--?
I ~ ~ ~ ~ ~ ~ ~ O o, 0o .0 .0 .OO O O O O O O O O O O O OO O O O -
"'"'

9--N

--? --?

^""'ir s " ? I Dc 9 \11\ O 0 0?

. 0 9 9 99 9 99 9o -9 9__-?"" " ^ " ^

l^C,~~DDUO .Q U~~~^ ^'''.Z - - _ /^o ^
--?

^ or 9
o 9rso
9 9 4^ as?- .:o
9 9.S/
9- -.'- -. -O-
^9 <o.^?.o.o.o.oos
9 9 - l- '.-$- ^- o^o
116 9 a Vlh 0

o^? ^ s s^o ? ? < ^sao.o.s? ^o ^ ^ .^I s s s O6Ua ht

,I-- a a u 9

rlu u 6 0? 0? 610
'"'^F^'S A10 'o o 'o <? 'o - ZOD' c
a a a a ~~n a N w a a P a a a Ln 4 O? a ~~~~~n ~o 4 a u o~~: 9 ~~~ `D Q ~C ?Iv, O? a cn 0 U1P

09~~999999999~~9949999999 . . . Llvl r- U V O? cne

~~~u~~u~u~~w uuo~~~~u~uouruaI u
-~~~~~~~~^ ^- - - Cl - ucnug?O?oru

-?

q )90?-?IVQOQ
W o0

NILE , 0 a9 99 a9 O, O,QQ ~ ~ ~ N JaoWE999

)VOZ?0?(TO?

DELTA Z u, a a 9 N a

iia900099

c
a uuu9uO?Pw

C, u D - 0- 0? I a. 0 cl , Cr uaI pumuu990,
999999999~~~~~~~~~~~ , 1 -?

II0?UQNWO
II

CcnurvruOaw

SAUDI-
II

nfrooo

:: :: ~ :: :: ; :ARA
oa0?c)oe-

Uicsouss

C09069v,

-J

Eeooo

CFEULIIVO?
^ ^ 1'. 1 : ^ ^"'^^^ 1; 5
Ua6PUUaa

n
?

90--rWPW-

-?

0?V090

I -..- > , . .

WESTERN EGYPT F TF SE SINAI GULF OF- AQABA "

GULF OF SUEZ GULF OF AQABA

DESERT
Fig. 2. Radiation temperature map of the area made by the scannin
tion temperatures are given as two-digit numbers either above 200

come warmer than anomalousan adjacent

reflectance and temperature bare because soil
re- of the multiple reflections in the
area during midday" (1),
lations of the but,
western for
Negev-Sinai an area
area canopy,
is which produce a cavity effect cor-
that does not exhibit the infrared reflec- that the vegetation covers only 25 to 35 responding to nearly ideal blackbody emis-
tance paradox, it is downright impossible. percent of the ground surface in the Negev sion. It is an open question to what extent
In view of the paradoxically low reflec- (as compared with some 10 percent or less the cavity effect applies to the clumps of
tance of the Negev, the predominance in ofthe Sinai) and that it is the interstices vegetation in the Negev and the Sinai. As-
interstices between vegetation clumps im- between the vegetation clumps that effec- suming an emissivity of 0.95 for the vege-
posed by the limitations in the available tively control the reflectances and temper- tative fraction of the ground cover and av-
moisture, and the dry appearance of the atures (8). In the Negev, the interstices eraging linearly in emissivity in proportion
vegetation (see Fig. 1), the question aroseshow dark-gray plant debris littering the to the ground cover, one obtains an emis-
whether the usual temperature relations surface; in the Sinai the scant plant debris sivity of 0.905 for the Sinai with a 10 per-
prevail for such dark areas "with an appre-are trampled into the ground and covered cent ground cover and 0.9175 for the Ne-
ciable vegetative cover." by dust in the unstable soil environment gev of with a 35 percent ground cover. Under
The existence of the temperature anom- the overgrazed area. Thus, the high albedo this assumption, the actually measured ra-
aly has been confirmed in a fragmentary of the Sinai is attributable to the bare un- diation temperatures of 45?C for the Neg-
way by one overflight (6) and the radiationstabilized soil whereas in the Negev theev and 40?C for the Sinai would corre-
temperature maps from satellites (see Fig. more stable soil is covered by the plant spond to sensible temperatures of 51.9?C
2). The temperature anomaly shows that debris. An additional effect is that the un- for the Negev and 47.9?C for the Sinai,
the radiation temperatures of the western stabilized soil has a higher reflectance thanin other words, a 4?C temperature dif-
Negev are higher by 4.1?K than those the of more stable, partially crusted soil offerential.
the northern Sinai (the areas in the box thein vegetation interstices (9). Even assuming an emissivity of 1.0 for
the upper center of Fig. 2), whereas the ra-There is an uncertainty in the emissivitythe plant fraction of the ground cover, we
diation temperatures of the cultivated which must be assumed in interpreting the would still find that the Negev is warmer
areas of the Nile Delta are lower by some radiation temperature data. I agree withthan the Sinai by 3?C. The 8?C emissivity
10?K. Nighttime data for the Negev-Sinai Jackson and Idso (1) that the emissivity ofdifference correction suggested by Jackson
area do not show any observable differ- the bare soil can be about 0.9 for the and Idso would apply, at most, if the Neg-
ences in radiation temperature. The fact PRT-5 radiometer (Barnes precision radi- ev had a thick, 100 percent vegetative cov-
that the thermal mountain of the western ation thermometer) measurements from er and
8 the Sinai a 0 percent vegetative cov-
Negev (6) shows up in contrast to the Sinai to 16 um. The low emissivity value is, er, to awhich is not the case. The scanning
only in the daytime strongly corroborates large extent, due to very low emissivity radiometer of the National Oceanic and
the viewpoint that the effect is caused by around 9 jam, the reststrahlen (residual Atmospheric Administration (NOAA)
sensible temperatures due to albedo dif- satellite operates in a narrower band
rays) effect. The emissivity of the vegeta-
ferences and not to emissivity differences. tion approaches unity, both because aroundof a 11 Mm, effectively outside the
Apparently, the explanation for the higher emissivity of the leaf surfacereststrahlen
and band. A higher emissivity for
1014 SCIENCE, VOL. 189

This content downloaded from 195.43.3.137 on Wed, 01 Nov 2017 13:01:55 UTC
All use subject to http://about.jstor.org/terms
soil
soil would
wouldapply
applyforfor
Fig.Fig.
2, with
2, with
probably flectance
probably
a a anomalies have been found to ex- argues
argues that
that
a time
a time
series regression
series regression
should should
negligible
negligibledifference
difference of emissivity
of emissivity
between ist in many parts of the world (7), including
between be
beused
used
andand
asserts,
asserts,
on the basis
on theof a basis
ther- of a ther-
the
the Sinai
Sinaiand
andthe
theNegev.
Negev. the Sahel. modynamic
modynamic argument,
argument,
that suchthat
a regres-
such a regres-
The multifaceted study of possible J. OTTERMAN* sion sionshould
should
yieldyield
almostalmost
direct proportion-
direct proportion-
desertification in this region is continuing Department of Environmental Sciences, ality
alitybetween
betweenthe daily
the demand
daily fordemand
heatingfor heating
and should in no way be regarded as de- Tel-A viv University, oil and the degree-day variable. We con-
finitive, not even for any of the subtopics.Ramat-A viv, Israel tend that the thermodynamic solution,
In spite of the efforts since the winter of References and Notes while relevant on a day-to-day basis, is not
1972-1973 to study and analyze these 1. R. D. Jackson and S. B. Idso, Science 189, 1012 appropriate for forecasting the impact of a
(1975).
phenomena, many uncertainties and ques- 2. D. M. Gates, Physical and Physiological Proper-
season-long policy.
tions remain. Specifically, appropriate ties of Plants in Remote Sensing, with Special Ref- A time series elasticity of unity implies a
erence to Agriculture and Forestry (National
mapping of radiation temperature dif- Academy of Sciences, Washington, D.C., 1970), constant short-run return to that input. It
ferences from satellites will be available pp. 224-252. is unrealistic to expect this constant uni-
3. H. H. Lamb, Climate, Present, Past and Future
only in 1977, when LANDSAT-C (Land (Methuen, London, 1972), vol. 1. tary elasticity to prevail for longer than an
Satellite), with a thermal band in its scan- 4. J. E. Colwell, Remote Sensing Environ. 3, 175 extremely brief interval within a social sys-
(1974).
ner will offer a relatively high resolution 5. J. Otterman, Rass. 20th Int. Elettron. Nucl. tem. The purpose of our study was to de-
and the coregistration of the thermal radi- (1973), pp. 199-203.
6. .., Science 186, 531 (1974).
termine what one could reasonably expect
ation measurements with the multispectral 7. __ , L. Walter, T. J. Schmugge, paper VI.3.3 from a policy implemented for a major
presented at the 18th plenary meeting of the Com-
reflectance measurements. However, for mittee on Space Research (COSPAR), Varna, portion of the heating season, during which
most pertinent information, the operating Bulgaria, 1975. time other social and economic forces were
8. J. Otterman, Y. Waisel, E. Rosenberg, Agro Eco-
time should be changed from 9:30 a.m. to syst. 2(No. 1), 47 (1975). effective. The thermodynamics of the situ-
about 10:30 a.m. (10). Further studies in a 9. N. H. MacLeod, personal communication. ation, while centrally relevant to the de-
10. J. Otterman, J. Br. Interplanet. Soc., in press.
variety of disciplines are needed, and spe- 11. I thank T. Gray and J. Winston of the National mand for heating fuel, should not be inter-
cifically I personally hope that Jackson and Oceanic and Atmospheric Administration for thepreted as the overriding effect in assessing
scanning radiometer data (Fig. 2).
Idso will carry out detailed temperature * Presently resident research associate, National an aggregate system demand.
Academy of Sciences-National Research Council,
measurements in an area of natural vegeta- Goddard Space Flight Center, Greenbelt, Mary- At present we know of no time series
tion in the United States that does exhibit land 20771. studies which would alter the conclusions
an infrared reflectance anomaly. Such re-
14 July 1975; revised 25 July 1975 reached in our report. Moreover, in a time
series study by Strout (1) on heating fuels,
a degree-day elasticity of 0.468 was estab-
lished. Similarly, Miller (2), employing
Fuel Savings by Lowering Thermostats data from December 1973 to May 1974
and September 1974 to May 1975, ob-
In criticizing Federal Energy Office Comparison
Comparisonofof perper
capita
capita
consumption
consumption tained results substantially similar to those
(FEO) estimates of the fuel savings that between
betweentwo twoareas
areasshows
shows onlyonly
thatthat
archi-
archi-
presented in our work.
would result from lowering thermostats tectural
tecturaland andlife
life
styles
styleschange
change
in response
in response In the fuel oil heating industry a degree-
6?F, Ferrar and Nelson (1) appear to have to to the
thelocal
localclimatology
climatology of degree-days.
of degree-days.A dayA elasticity of unity is widely used as a
erred. Their double-log regression showed resident
residentofofthe theSouth
SouthAtlantic
Atlantic
region
region
who who
base point, but correction factors are regu-
fuel demand about proportional to the moves
movesto toNew
New England,
England, where
where
more more
than than
larly applied, not only for geographic re-
square root of heating degree-days, as de- twice as many degree-days are encoun- gions (Reed's point) but also for seasonal
rived from averages of individual state sta- tered, would not double his fuel consump-combustion efficiency changes. In attempt-
tistics. This relationship should not be usedtion, because his new home would be bettering to forecast demand over a heating sea-
to estimate degree-day effects within a re-insulated and designed against this sever- son, one must make sure that these correc-
gion. A time series regression, against ob-ity. The square root effect would apply. tion factors are taken into account. Sim-
served degree-days within a state or region, But if a particularly cold winter in the ilarly, it is common in the heating industry
must be used, and it should yield almost di- South Atlantic area caused doubled de- to calculate relationships between con-
rect proportionality. Conduction heat gree-days to be recorded, his fuel con- sumption and degree-days for commercial
losses from residences are proportional to sumption there would indeed be doubled inestablishments on a building-by-building
the difference in temperature between in- response. basis. Again, this industry behavior calls
side and outside. Radiant heat losses, pro- JACK W. REED into question the reliability of a unitary
portional to the fourth power of absolute Applied Fluid Mechanics Division, elasticity for forecasts of a seasonal dura-
temperature, also turn out to be nearly di- Sandia Laboratories, tion.
rectly proportional to this temperature dif- Albuquerque, New Mexico 87115 TERRY A. FERRAR, JON P. NELSON
ference for small absolute reference shifts Environmental Policy Center and
of 3?K (6?F). Using Ferrar and Nelson's References Department of Economics,
tabulated data, a revised calculation yields 1. T. A. Ferrar and J. P. Nelson, Science 187, 644 Pennsylvania State University,
a fuel saving of 24 percent of 1973 esti- (1975). University Park 16802
13 March 1975
mated demand; that is, 1.7 x 1014 Btu, or
References
840,000 barrels per day, which is close to Reed questions the use of cross-sectional
the FEO estimate (900,000 barrels per day) data to estimate relationships between 1.perA. M. Strout, Rev. Econ. Stat. 43, 185 (May
1961).
that they quoted. capita fuel demand and degree-days. He 2. K. D. Miller, unpublished results.

19 SEPTEMBER 1975 1015

This content downloaded from 195.43.3.137 on Wed, 01 Nov 2017 13:01:55 UTC
All use subject to http://about.jstor.org/terms