water

Article
Influence of Meander Confinement on
Hydro-Morphodynamics of a Cohesive
Meandering Channel
Parna Parsapour-Moghaddam * and Colin D. Rennie * ID

Department of Civil Engineering, University of Ottawa, 161 Louis Pasteur, Ottawa, ON K1N 6N5, Canada
* Correspondence: p.parsapour@uottawa.ca (P.P.-M.); Colin.Rennie@uottawa.ca (C.D.R.);
Tel: +1-613-562-5800 (ext. 6161) (P.P.-M.); +1-613-562-5800 (ext. 6124) (C.D.R.)




Received: 6 January 2018; Accepted: 19 March 2018; Published: 22 March 2018


Abstract: Despite several decades of intensive study of the morphological changes in meandering
rivers, less attention has been paid to confined meanders. This paper studies the hydro-
morphodynamics of two adjacent sub-reaches of a meandering creek, located in the City of Ottawa,
Canada. Both of these sub-reaches are meandering channels with cohesive bed and banks, but
one is confined by a railway embankment. Field reconnaissance revealed distinct differences in
the morphological characteristics of the sub-reaches. To further study this, channel migration and
morphological changes of the channel banks along each of these sub-reaches were analyzed by
comparing the historical aerial photography (2004, 2014), light detection and ranging (LIDAR) data
(2006), bathymetric data obtained from a total station survey (2014), and field examination. Moreover,
two different spatially intensive acoustic Doppler current profiler (ADCP) surveys were conducted
in the study area to find the linkage between the hydrodynamics and morphological changes in the
two different sub-reaches. The unconfined sub-reach is shown to have a typical channel migration
pattern with deposition on the inner bank and erosion on the outer bank of the meander bend.
The confined sub-reach, on the other hand, experienced greater bank instabilities than the unconfined
sub-reach. The average rate of bank retreat was 0.2 m/year in the confined sub-reach whereas it
was lower (0.08 m/year) in the unconfined sampling reach. In the confined sub-reach, an irregular
meandering pattern occurred by the evolution of a concave-bank bench, which was caused by reverse
flow eddies. The sinuosity of the confined sub-reach decreased from 1.55 to 1.49 in the 10-year study
period. The results of the present study demonstrate the physical mechanisms by which meander
confinement can change the meandering pattern and morphological characteristics of a cohesive clay
bed creek.

Keywords: meandering rivers; meander confinement; cohesive bed rivers; hydro-morphodynmics;

spatial ADCP survey

1. Introduction

1.1. Background
Many river scientists and engineers have studied meandering rivers over the past few centuries
(e.g., [1–16]). Despite the fact that river meandering is an erudite topic with a long literature history,
there are still some uncertainties on the source and initiation of the meandering pattern and its
migration. It is known that river meandering is associated with the bank erosion mechanism and can
be influenced by the spatial progression of bars [17].
Bank instability and erosion are intrinsic characteristics of meandering rivers [18]. Lanzoni and
Seminara illustrated how the morphodynamics of meandering rivers can be impacted by meander

Water 2018, 10, 354; doi:10.3390/w10040354 www.mdpi.com/journal/water

Water 2018, 10, 354 2 of 18

instability [8]. River meandering may also lead to a flow separation which could impact bank erosion
patterns [11,19–23]. Bank erosion processes influence a wide range of social, environmental and
economic factors [24]. Consequently, bank erosion predictions are of essential importance in sustainable
river management [25]. Despite numerous well-documented previous studies devoted to the bank
retreat mechanism, the erosion of cohesive river substrates is not completely understood [26–28].
Cohesive properties of river banks can impact the rate of bank erosion [29]. Unlike non-cohesive river
bed sediments for which resistance to entrainment is merely mechanical, cohesive material interactions
depend on the electro-chemical bonds amongst the particles [26]. Furthermore, several studies have
recognized the significance of pore water pressure in the erosion and removal of cohesive bed river
particles or assemblages e.g., [30–33]. However, a general predictive theory for entrainment of cohesive
river boundary sediments is yet to be developed; thus, prediction of bank erosion in a cohesive river
remains challenging [27]. Bank erosion and subsequent meander migration are important natural
processes for both unconfined and partially confined rivers [34].
Although meander morphodynamics have been widely studied, less attention has been paid to
confined meandering. The morphological development of a confined meander is impeded by a natural
or manmade restriction. A confined meander cannot freely develop, which results in a distinctive
meander pattern that differs from those presented by freely meandering rivers. Nicoll noted that the
dynamics of meander migration may be affected in confined meandering rivers [35], and Ghinassi et al.
observed that meander confinement can lead to the downstream migration of fluvial point bars [36].
Lewin and Brindle showed that free meandering could be hindered by bedrock or anthropogenic
structures [37], and this can result in a square-wave shape meandering configuration. They defined
three degrees of confinement: first-degree confinement takes place in wide valleys whereby the stream
impinges irregularly against the confining valley wall. Lane called this type of meander a “restrained”
meander [38]. In second-degree confinement, the outer side of each meander bend impinges upon
the confining wall, and in third-degree confinement, the meandering stream does not have space
to progress.
Most river meanders do not undergo purely downstream translation. Nicoll observed downstream
translation without substantial distortion only in confined meanders with limited amplitude and low
curvature [35]. Lewin and Brindle noted that downstream translation without meander deformation is
prevalent in second-degree confined rivers [37]. Nicoll and Hickinstudied the planform geometry and
migration pattern of several second-degree confined meandering rivers on the Canadian prairies [39].
They related the channel-migration rate of the studied rivers to basic geomorphic and hydrologic
variables. The authors indicated that the planform relations were generally consistent with those
manifested by freely meandering rivers, with small yet significant differences due to the distinctive
meander behavior of confined meanders.
Previous studies have shown that meander confinement can lead to development of a concave-
bank bench [40–44]. A concave-bank bench is a crescent-shaped accretion on the upstream portion
of the outer bank (concave side) of a meander bend. These deposits are generally observed in
meandering rivers that migrate down valley [45]; however, they may also occur in unconfined meander
belts [41,44,46]. Page and Nanson demonstrated that when the channel flow has the power to erode
the channel banks [41], regular meandering is developed with a point bar and overbank deposition.
However, in cases where the stream power hardly surpasses the shear strength of the bank material,
then irregular meanders may be generated with long straight reaches and sharp bends. They showed
that concave-bank benches are commonly developed when sharp meander bends migrate downstream.
The effect of manmade features on meandering river morphodynamics has been studied,
particularly for river stabilization works [47]. Nevertheless, less studied has been the morphodynamics
of meandering rivers confined by structures such as railway and road embankments [37]. Furthermore,
the migration behavior of confined bends has been studied, most previous research has focused on
second-degree confinement where the river impinges against the confining edge in every meander
wavelength [39].
Water 2018, 10, 354 3 of 18

1.2. Objectives and Structure

As outlined above, there are still some uncertainties on the source and initiation of the meandering
pattern and its migration, particularly for cohesive bed rivers. Moreover, the morphology and dynamics
of confined meandering rivers are relatively poorly studied. Due to natural or manmade restrictions,
confined meanders cannot freely develop, and this makes them have a meander pattern distinct
from those presented by freely meandering rivers. Previous studies of confined meandering rivers
have been mostly limited to second-degree confinement. Moreover, to the best of our knowledge,
the hydrodynamics of a confined meandering channel have not previously been studied in terms
of measuring the spatial distribution of the velocity flow field. Given the prevalence of confined
meandering rivers, enhancing the understanding of their behavior in the landscape is of essential and
practical importance for sustainable river management.
In this study, we evaluate the meander migration dynamics in a first-degree confining medium, i.e.,
in which the river makes irregular contact against a confining railway embankment, in a meandering
cohesive bed river. We employ a paired sub-reach study approach, wherein one sub-reach is freely
meandering and the second adjacent sub-reach is first degree confined by the railway embankment.
Morphodynamics of each sub-reach are measured by repeat surveying over a multi-year period.
Specifically, channel migration and morphological changes of the channel banks along each of these
sub-reaches are analyzed by comparing historical aerial photography, light detection and ranging
(LIDAR) data and bathymetric data obtained from a total station survey. Furthermore, channel
hydrodynamics are measured in both sub-reaches by spatially intensive acoustic Doppler current
profiler (ADCP) surveying, which, to the best of our knowledge, has not previously been conducted
in a confined meander. This allows for a direct comparison of the hydrodynamic characteristics of
unconfined and confined sub-reaches. Furthermore, the measured velocity fields are used to find
linkages between the channel hydrodynamics and morphological changes. We are thereby able to
explore how local meander confinement leads to different hydro-morphodynamic characteristics and
corresponding meander morphology. The methodology employed in this study is comprehensively
explained in the next section (Section 2). Results are shown in Section 3, followed by the discussion
(Section 4) and conclusions (Section 5).

2. Methodology

2.1. Study Site

Watts Creek flows into the Ottawa River at Shirley’s Bay in the Kanata region of the City of Ottawa,
Canada (Figure 1). Two upstream branches (Watts Creek and Kizell Drain) join at a confluence situated
at 45.340172◦ latitude, −75.880610◦ longitude (UTM 431006 m E, 5021119 m N). Watts Creek has been
recognized as providing important coolwater fish habitat [48]. Rates of channel degradation are still
not clear. In this study, we mainly focus on two adjacent sub-reaches. The two sub-reaches are M3 and
M4 (Figure 1c). Both of these sub-reaches are meandering channels with a cohesive bed and banks.
M4 is partially confined by the City of Ottawa rail line, and thereby is undergoing enhanced erosion.
Watts Creek is a semi-alluvial channel offering a diverse habitat, with a mixture of runs, pools, and
riffles. While fine clay and silt substrate is prevalent in Watts Creek, some coarse gravel is also present.
Watts Creek has a bed that consists of a high (>30%) percentage of clay, i.e., soil particles that are smaller
than two to four micrometers [49]. Clay particle properties cannot be as straightforwardly predicted
because cohesion is dictated by electrostatic forces, which depend on the chemical composition of the
clay particles. In addition, the natural consolidation state and hydraulic conductivity, organic matter,
and a few other properties influence particle cohesion. The results of the bed sediment core samples
collected from the entire reach indicated that all median grain sizes were in the silt or clay range which
characterized them as fine grained, cohesive soils. The results of a piston-flume critical bed shear stress
test suggest that the cohesive clay bed sediments in Watts Creek watershed are not heavily consolidated.
Water 2018, 10, 354 4 of 18
Water 2018, 10, x FOR PEER REVIEW 4 of 17

a b

c
M4 sub-reach
2
1 2
1
M3 sub-reach

Figure 1. (a) Location of the study area within Canada (adopted from https://www12.statcan.gc.ca);
Figure 1. (a) Location of the study area within Canada (adopted from https://www12.statcan.gc.ca);
(b) Study reach location in the Ottawa Study reach location in the Ottawa area (adopted from
(b) Study reach location in the Ottawa Study reach location in the Ottawa area (adopted from http:
http://data.ottawa.ca/en/dataset/water); (c) Surveyed sections in Watts Creek sampling reaches M4
//data.ottawa.ca/en/dataset/water); (c) Surveyed sections in Watts Creek sampling reaches M4 (to
(to left) and M3 (to right) including total station surveyed points, flow from west (left) to east (right)
left) and M3 (to right) including total station surveyed points, flow from west (left) to east (right)
(adopted from Google earth). Squares show the meander locations and their referred number within
(adopted from Google earth). Squares show the meander locations and their referred number within
each sub-reach. Please note that the study area is situated at 45.340172° latitude, −75.880610°
each sub-reach. Please note that the study area is situated at 45.340172◦ latitude, −75.880610◦ longitude.
longitude.

2.2.2.2.
SiteSite
Reconnaissance
Reconnaissance
Channel
Channelcondition
condition was assessed by
was assessed byvisual
visualobservation
observation during
during different
different flow regimes.
flow regimes. The
Theobservations
observations areare categorized
categorized by location
by location within
within eacheach sampling
sampling sub-reach
sub-reach specified
specified in 1c.
in Figure Figure
As is1c.
As shown, M4M4
is shown, sub-reach
sub-reachof Watts
of WattsCreek
Creekis confined by the
is confined by City of Ottawa
the City rail line.
of Ottawa M3 and
rail line. M3M4 andare
M4
otherwise similar; they are meandering channels with cohesive bed and banks.
are otherwise similar; they are meandering channels with cohesive bed and banks. They convey the They convey the same
discharge
same (based
discharge (basedon on
thethe
ADCP
ADCP measurements),
measurements), and they
and theyhave
havesimilar
similarbed
bedsubstrate
substrate and
andriparian
riparian
vegetation.
vegetation. Figures
Figures 2–52–5
showshowthethe field
field conditions
conditions immediatelyafter
immediately aftera aspring
springfreshet
freshetflood.
flood.Locations
Locationsof
of erosion
erosion and channel
and channel incisionincision were identified
were identified by exposed
by exposed tree
tree roots and roots and
steep, steep, undercut
undercut or
or collapsing
collapsing river banks. Observations for each sampling
river banks. Observations for each sampling reach are provided below. reach are provided below.

2.2.1.
2.2.1. M4M4 Sampling
Sampling Reach
Reach
This
This part
part of of Watts
Watts Creek
Creek meandersisisadjacent
meanders adjacentto
tothe
the City
City of
of Ottawa
Ottawarail
railline.
line.Overbank
Overbanksediment
sediment
deposits throughout this reach indicated that the previous spring freshet had overtopped the banks
deposits throughout this reach indicated that the previous spring freshet had overtopped the banks in
in the Main Creek. Figure 2a shows the development of a concave-bank bench on the upstream limb
the Main Creek. Figure 2a shows the development of a concave-bank bench on the upstream limb of
of the outer bend at the last sharp meander. There were indications of instabilities on the downstream
the outer bend at the last sharp meander. There were indications of instabilities on the downstream
portion of the outer bends in M4 (Figure 2b). Inner bank instability was also observed at the upstream
portion of the outer bends in M4 (Figure 2b). Inner bank instability was also observed at the upstream
bend in M4. As shown in Figure 3a, an erosion pathway cut through the inner bank point bar during
bend in M4. As shown in Figure 3a, an erosion pathway cut through the inner bank point bar during
the freshet flow; vertical stakes in the channel suggest failure of a previous attempt to stabilize the
theinner
freshet flow;
bank vertical
(Figure 3b). stakes
Figure in
4athe channelthe
illustrates suggest failure
evolution of longitudinal-shaped
of the a previous attempt bar
to stabilize the
as well as
Water 2018, 10, 354 5 of 18
Water 2018, 10, x FOR PEER REVIEW 5 of 17
Water 2018, 10, x FOR PEER REVIEW 5 of 17
inner
thebank (Figure
secondary 3b). Figure
channel 4a illustrates
adjacent thebank
to the outer evolution
of the of the
last longitudinal-shaped
meander bar as well
bend. These features as the
suggest
the secondary
secondary
that channel
M4 has channel
a very adjacent
adjacent
active, to thetoouter
unstable the outer bank
of theoflast
bankpresumably
channel, the due
last to
meander
meander bend.
the bend. These
These
meander features
features
confinement by suggest
suggest that
the rail
that
M4line. M4 has a very active, unstable channel, presumably due to the meander confinement
has a very active, unstable channel, presumably due to the meander confinement by the rail line. by the rail
line.

a b
a b

Figure 2. M4 sampling reach, meander bend restrained by the City of Ottawa rail, facing downstream.
Figure 2. M4 sampling reach, meander bend restrained by the City of Ottawa rail, facing downstream.
Figure
Note 2. M4
the Citysampling
of Ottawareach,
rail meander bend restrained
line immediately adjacentbytothe
theCity of Ottawa
south rail, facing
of the river: downstream.
(a) Concave-bank
Note the City of Ottawa rail line immediately adjacent to the south of the river: (a) Concave-bank
Note the
bench City of on
formation Ottawa rail line of
the upstream immediately adjacent
the bend apex; to failure
(b) The the south of thebank,
of outer river:downstream
(a) Concave-bank
of the
bench formation on the upstream of the bend apex; (b) The failure of outer bank, downstream of the
benchapex.
bend formation on were
Pictures the upstream
taken by of thethe bend apex;
authors on 30 (b) The2014.
April failure of outer bank, downstream of the
bend apex. Pictures were taken by the authors on 30 April 2014.
bend apex. Pictures were taken by the authors on 30 April 2014.

a b
a b

Figure 3. M4 sampling reach, meander bend restrained by the City of Ottawa rail, facing downstream:
Figure
(a) Inner3.bank
M4 sampling reach, meander
was overtopped bend
during the restrained
previous highbyfreshet
the City of Ottawa
flow, rail, facing
to an erosion downstream:
pathway through
Figure 3. M4 sampling reach, meander bend restrained by the City of Ottawa rail, facing downstream:
(a) Inner
the bank was
inner bank; overtopped
(b) Vertical metalduring
stakesthe previous
used for bankhigh freshet flow,
stabilization. to an were
Pictures erosion pathway
taken by the through
authors
(a) Inner bank was overtopped during the previous high freshet flow, to an erosion pathway through
the30
on inner bank;
April (b) Vertical metal stakes used for bank stabilization. Pictures were taken by the authors
2014.
the inner bank; (b) Vertical metal stakes used for bank stabilization. Pictures were taken by the authors
on 30 April 2014.
on 30 April 2014.
Water 2018, 10,
Water 2018, 10, 354
x FOR PEER REVIEW 66 of
of 18
17
Water 2018, 10, x FOR PEER REVIEW 6 of 17

aa b
b

Figure
Figure 4.
Figure M4
4. 4.
M4M4sampling reach:
samplingreach:
sampling (a)
reach:(a) meander
(a)meander
meanderbendbend restrained
restrainedby
restrained bythe
by theCity
the Cityofof
City ofOttawa
Ottawa
Ottawa rail,
rail, facing
facing
rail, facing upstream.
upstream.
upstream.
Formation
Formation of the
of longitudinal
the longitudinal bar
bar and
and the
the secondary
secondary channel
channel on
on the
the upstream
upstream
Formation of the longitudinal bar and the secondary channel on the upstream of the outer bend ofof the
the outer
outer bend
bend apex;
apex;
apex;
(b)(b) ADCP
ADCP spatial
spatial survey
survey inin the
the study
study reach.
reach. The City
City of
of Ottawa
Ottawa rail
railline
linecan
can
(b) ADCP spatial survey in the study reach. The City of Ottawa rail line can be seen immediatelybe beseen
seen immediately
immediately
adjacent
adjacent
adjacent toto
to the
the
the south
south
south ofofthe
of theriver.
the river.Pictures
river. Pictureswere
Pictures weretaken
were taken by
taken by the authors
authorson
by theauthors on14
on 14April
14 April2015.
April 2015.
2015.

2.2.2. M3
2.2.2. Sampling Reach
2.2.2. M3
M3 Sampling
Sampling Reach
Reach
Sampling
Sampling reach M3 encompassestwo two unconfined meander bends. The
Thechannel is reasonably
Sampling reach
reach M3
M3 encompasses
encompasses two unconfined
unconfined meander
meander bends.
bends. The channel
channel isis reasonably
reasonably
stable
stable in M3, displaying only modest bank slumping on the outer bank downstream of a bend apex.
stable in M3, displaying only modest bank slumping on the outer bank downstream of aa bend
in M3, displaying only modest bank slumping on the outer bank downstream of bend apex.
apex.
This unconfined section of the creek represents a regular meandering pattern with moderate erosion
This
This unconfined section of
of the
the creek represents aa regular meandering pattern
pattern with
with moderate
moderate erosion
onunconfined section
the outer bank creek
and deposition onrepresents regular
the inner bend meandering
(Figure 5b). erosion
on the outer bank and deposition on the inner bend (Figure
on the outer bank and deposition on the inner bend (Figure 5b). 5b).

Figure 5. M3 facing downstream: (a) In the middle of M3, between the two major bends; (b) Last
meander bend manifests regular meandering pattern with erosion on the outer band and deposition
Figure 5. M3 facing downstream: (a) In the middle of M3, between the two major bends; (b) Last
Figure M3 bank.
in the5.inner facingPictures
downstream: (a) In
were taken bythe
the middle of M3,
authors on between
30 April 2014. the two major bends; (b) Last
meander bend manifests regular meandering pattern with erosion on the outer band and deposition
meander bend manifests regular meandering pattern with erosion on the outer band and deposition in
inData
2.3.thethe inner
inner bank.
Collection
bank. Pictures
and were
Analysis
Pictures were taken
taken by by
thethe authors
authors on on 30 April
30 April 2014.
2014.

Morphological
2.3. Data Collection andstudies
Analysisof the two sub-reaches were done based on the results of: (a) bathymetric
2.3. Data Collection and Analysis
survey of each section (b) available LIDAR data (c) historical aerial photographs. A topographic and
Morphological
bathymetric studies
studystudies of the two sub-reaches
was conducted were doneemploying
based on the results of: (a) survey
bathymetric
Morphological of the two during summer
sub-reaches were2014,
done based on the a results
total station to
of: (a) bathymetric
survey of each section (b) available LIDAR data (c) historical aerial photographs. A topographic and
survey of each section (b) available LIDAR data (c) historical aerial photographs. A topographic in
accurately distinguish the bank and bed topography. Over 4065 bathymetric points were collected and
bathymetric study
both studystudy was conducted
sub-reaches with an duringduring
average summer
spacing of 1.22014, employing
and 0.3 m aintotal a total and
streamwise station survey to
cross-stream
bathymetric was conducted summer 2014, employing station survey to accurately
accurately distinguish the
directions,therespectively bank
(Figureand bed topography.
1c). TheOver
collected Over 4065 bathymetric
points werepoints
then were points
interpolatedwere collected in
distinguish bank and bed topography. 4065 bathymetric collectedby the TIN
in both study
both study sub-reaches with an average spacing of 1.2 and 0.3 m in streamwise and cross-stream
sub-reaches with an average spacing of 1.2 and 0.3 m in streamwise and cross-stream directions,
directions, respectively (Figure 1c). The collected points were then interpolated by the TIN
Water 2018, 10, 354 7 of 18
Water 2018, 10, x FOR PEER REVIEW 7 of 17

respectively
interpolation(Figure
method 1c).
in The collected(Esri,
ArcGIS10.2 points were then
Redlands, interpolated
CA, USA) to obtainby thetheTINDEMinterpolation method
(Digital Elevation
in ArcGIS10.2 (Esri, Redlands, CA, USA) to obtain the DEM
Model) of 2014. Slope and hillshade maps were then obtained from the DEM. (Digital Elevation Model) of 2014. Slope
and hillshade
The expense mapsof were
river then obtained from thefield
hydro-morphological DEM. studies has increased utilization of aerial survey
The expense of river hydro-morphological
techniques [50]. Relatively recent development of LIDAR field studies has increased
technologyutilization
allowsof aerial survey
for accurate
techniques [50]. Relatively recent development of LIDAR technology allows
measurements of bank locations and elevations [51]. The City of Ottawa has collected LIDAR datafor accurate measurements
of
forbank locations
various parts and elevations
of the [51].different
city during The Cityyears.
of Ottawa
Thishas collected
study employs LIDAR data for
available various
LIDAR dataparts of
of the
the city during different years. This study employs available LIDAR data
study creek from 2006 obtained from the National Capital Commission (NCC). We converted the of the study creek from 2006
obtained
LIDAR data fromtothe National
DEMs usingCapital
ArcGIS.Commission (NCC).
Historical aerial We converted
photographs were theacquired
LIDAR data usingtoGoogle
DEMs Earth.
using
ArcGIS.
The photos Historical aerial photographs
were registered were acquired
using georeferencing using
tools Google
available Earth. The
in ArcGIS. photos were
Stationary groundregistered
control
using
pointsgeoreferencing
for rectificationtools available
were obtainedin ArcGIS. Stationary
using corners ground
of the controlroad
buildings, points for rectification
intersections, rail were
trail,
obtained usingand
solitary trees corners
large of the nearby
rocks buildings,theroad
studyintersections,
site. Figure 6rail trail,the
shows solitary
aerialtrees andoflarge
images rocks
the study
nearby the study
reach from 2004 site.
and Figure
2014. It6 should
shows thebe aerial
notedimages
that bothof the study
aerial reachshow
photos from the2004site
andcondition
2014. It should
in the
be noted that both aerial photos show the site condition in the summer
summer (June) which represents the low flow regime (discharge ~0.15 m /s based on (June)
3 which represents
2014 ADCP the
low flow regimein (discharge 3
measurements) the study~0.15
creek.m /s based on 2014 ADCP measurements) in the study creek.

Figure 6.
Figure 6. Aerial
Aerialphoto
photoofofthe
thestudy
studyreach
reach adopted
adopted from
from Google
Google Earth:
Earth: (a) (a)
20142014
(b) (b) 2004.
2004. Location
Location of
of the
the each sampling reach is shown with the
each sampling reach is shown with the square. square.

A comparison was then made between the aerial photographs (2004 and 2014), DEMs obtained
A comparison was then made between the aerial photographs (2004 and 2014), DEMs obtained
from the LIDAR data (2006) and total station surveys (2014). The river DEM maps from 2004, 2006
from the LIDAR data (2006) and total station surveys (2014). The river DEM maps from 2004, 2006 and
and 2014 were overlapped in ArcGIS, and polygons were created to define the active channel
2014 were overlapped in ArcGIS, and polygons were created to define the active channel boundary
boundary and the bank retreat. To distinguish the location of the base and top of the river banks
and the bank retreat. To distinguish the location of the base and top of the river banks during
during different times, slope classification maps were overlapped with the hillshade maps for both
different times, slope classification maps were overlapped with the hillshade maps for both LIDAR
LIDAR and surveyed DEMs. This allowed for assessment of the meander behavior and bank retreat
and surveyed DEMs. This allowed for assessment of the meander behavior and bank retreat of each
of each sub-reach. Having DEMs of the study creek at different periods also allows for measurements
sub-reach. Having DEMs of the study creek at different periods also allows for measurements of
of elevation changes. Accordingly, the DEM from 2006 was subtracted from that of 2014 through the
elevation changes. Accordingly, the DEM from 2006 was subtracted from that of 2014 through the
raster calculator tool in ArcGIS to produce a DEM difference raster to explore the amount of erosion
raster calculator[52].
and deposition tool in ArcGIS to produce a DEM difference raster to explore the amount of erosion
and deposition [52].
In order to evaluate the effect of the velocity field on the channel morphology, spatially intensive
In order to evaluate the effect of the velocity field on the channel morphology, spatially intensive
ADCP surveys were conducted in both study reaches during August (low flow) and October
ADCP surveys were conducted in both study reaches during August (low flow) and October (moderate
(moderate flow) 2014. According to the ADCP measurements, flow discharge was 0.17 and 0.45 m/s
flow)
under2014.
low According
and moderate to theflow,
ADCP measurements,
respectively. flow discharge
It should be noted was
that 0.17
the and 0.45 m/s
bankfull under was
discharge low
and moderate flow,
approximately respectively.
1.4 m/s [53], and theIt should be noted
return flow of thethat
highthe bankfull
flow dischargetowas
was estimated approximately
be 1.4 years [54]. A
1.4 m/s [53], and the return flow of the high flow was estimated to be 1.4 years [54]. A Sontek M9
Sontek M9 River Surveyor ADCP was deployed on an Ocean Sciences trimaran riverboat (Figure 4b).
River Surveyor
Standing ADCP was
on opposite banks deployed on an we
of the creek, Ocean Sciences
operated thetrimaran
trimaranriverboat
boat with(Figure
ropes4b).
andStanding on
moved the
opposite banks of the
boat downstream increek, we operated
narrowly spaced the trimaran
transects in boat with array.
a zigzag ropes and
Themoved the boat
measured downstream
depth-averaged
in narrowly spaced transects in a zigzag array. The measured depth-averaged velocity
velocity data were then post processed using in-house Matlab codes [55]. Parsapour-Moghaddam data were then
post
and processed using in-house
Rennie provide Matlaboncodes
further details [55]. Parsapour-Moghaddam
conducting and Rennie
a spatially intensive ADCP surveyprovide further
in a clay-bed
details on conducting a
meandering river [16,56,57].spatially intensive ADCP survey in a clay-bed meandering river [16,56,57].
Water 2018, 10, 354 8 of 18

3. Results
Table 1 illustrates the channel and flow characteristics of both sub-reaches during ADCP
Water 2018, 10, x FOR PEER REVIEW 8 of 17
measurements at the moderate flow in 2014. As can be seen, the channel and averaged values of
the flow characteristics were quite similar in both sub-reaches.
3. Results
Table
Table 1. 1 illustrates
Channel and flowthe channel andinflow
characteristics characteristics
the study area basedof on
both
thesub-reaches during ADCP
ADCP measurements at the
measurements at the moderate flow in 2014. As can be seen, the channel and averaged values of the
moderate flow in 2014.
flow characteristics were quite similar in both sub-reaches.

Table 1. Channel andParameter M4 Sub-Reach

flow characteristics in the study M3 Sub-Reach
area based on the ADCP measurements at the
moderate flow in 2014.
Channel slope 0.0017 0.0016
Averaged width (m) 3.4 3.6
Parameter M4 Sub-Reach M3 Sub-Reach
Averaged depth (m) 0.4 0.6
Channel slope 3 0.0017
0.45 0.0016
0.45
Averaged discharge (m /s)
Averaged width (m)
Averaged flow velocity (m/s) 3.4
0.2 3.6
0.17
Averaged depth (m)
Average Froude number 0.4
0.1 0.6
0.07
Averaged discharge (m3/s) 0.45 0.45
Averaged flow velocity (m/s) 0.2 0.17
We compared the position
Average of number
Froude the active channel mapped
0.1 from 20040.07
aerial photography with
that of 2014. This is shown for both M4 and M3 sampling reaches in Figures 7 and 8, respectively.
We compared
The boundaries the position
were obtained of the active
through channelthe
digitizing mapped frommargin
channel 2004 aerial
basedphotography with that
on the aerial images in
of 2014. This is shown for both M4 and M3 sampling reaches in Figures 7 and 8, respectively. The
ArcGIS. A sufficiently large number of vertices was employed to avoid digitization errors. Figure 9
boundaries were obtained through digitizing the channel margin based on the aerial images in
shows the cumulative erosion and sedimentation that occurred in the channel margins between
ArcGIS. A sufficiently large number of vertices was employed to avoid digitization errors. Figure 9
2006 (based on cumulative
shows the the LIDARerosion
data) and
andsedimentation
2014 (basedthaton the totalinstation
occurred survey).
the channel This
margins was obtained
between 2006 by
subtracting the 2006 DEM of the study area from the corresponding 2014 DEM.
(based on the LIDAR data) and 2014 (based on the total station survey). This was obtained by
Figure 8 shows
subtracting thatDEM
the 2006 the of
M3thesampling
study areareach exhibited
from the a typical
corresponding 2014 meandering
DEM. pattern in which
erosion and Figure 8 showsoccur
deposition that the M3 outer
at the sampling
andreach exhibited
inner a typical meandering
bank, respectively. This canpattern in in
be seen which
Figure 9b
erosion and deposition occur at the outer and inner bank, respectively. This can be seen in
where negative and positive values show erosion (outer bend) and deposition (inner bend), respectively. Figure 9b
where negative and positive values show erosion (outer bend) and deposition (inner bend),
This confirms our field observations (Figure 5). Table 2 illustrates that the sinuosity of this sub-reach
respectively. This confirms our field observations (Figure 5). Table 2 illustrates that the sinuosity of
was more-or-less consistent, with only a slight increase from 2004 to 2014.
this sub-reach was more-or-less consistent, with only a slight increase from 2004 to 2014.

Figure 7. Channel margin migration along the M4 sampling reach. Flow from left (west) to right (east).
Figure 7. Channel margin migration along the M4 sampling reach. Flow from left (west) to right (east).
Background image adopted from Google Earth image (2004).
Background image adopted from Google Earth image (2004).
Water 2018, 10, 354 9 of 18
Water 2018,
Water 10,10,
2018, x FOR PEER
x FOR REVIEW
PEER REVIEW 9 of917
of 17

Figure 8. Channel margin migration along the M3 sampling reach. Flow from left (west) to right (east).
Figure 8. Channel margin migration along the M3 sampling reach. Flow from left (west) to right (east).
Figure 8. Channel
Background margin
image migration
adopted along Earth
form Google the M3 sampling
image (2004).reach. Flow from left (west) to right (east).
Background image adopted form Google Earth image (2004).
Background image adopted form Google Earth image (2004).

a b
a
Figure 9. Erosion (negative) and deposition (positive) (m) from 2006 to 2014 in: (a) M4 and (b) M3 b
sampling reaches. Background image adopted form google earth image (2004).
Figure 9. Erosion (negative) and deposition (positive) (m) from 2006 to 2014 in: (a) M4 and (b) M3
Figure 9. Erosion 2.(negative) and deposition (positive)
of (m) fromcreek
2006within
to 2014 in: (a) M4 and (b) M3
samplingTable
reaches.Channel geometry
Background of each
image sub-reaches
adopted the study
form google earth image (2004).different years.
sampling reaches. Background image adopted form google earth image (2004).
Sampling Reach Date Length of the Channel Path (m) Sinuosity
Table 2. Channel geometry of2004
each sub-reaches of the study creek within1.55
187.88 different years.
M4
Table 2. Channel geometry of 2014
each sub-reaches of the study creek within
175.27 1.49 different years.
Sampling Reach Date Length of the Channel Path (m) Sinuosity
2004 106.26 1.44
M3 2004 187.88 1.55
Sampling ReachM4 Date
2014 Length of107.1
the Channel Path (m)1.46 Sinuosity
2014 175.27 1.49
2004
2004 187.88
106.26 1.44 1.55
M4 the
On the contrary, M3M4 sampling reach (Figure 7) showed
2014
2014 175.27irregular meandering
107.1
behavior. From
1.46 1.49
Table 2, it can be inferred that the sinuosity of the channel decreased from 1.55 to 1.49 in the 10-year
period. As shown M3 in both Figures20047 and 9a, a concave-bank 106.26
bench can be observed in 1.44
the upstream
On the contrary, the M4 sampling 2014reach (Figure 7) showed 107.1 irregular meandering behavior. From
1.46
limb of the outer bank of the first meander bend. Furthermore, a longitudinal bar is developing in
Table 2, it can be inferred that the sinuosity of the channel decreased from 1.55
the second meander bend, which may be a precursor to concave bank bench. These results are
to 1.49 in the 10-year
period. As
consistent shown in both
with the the
resultsFigures 7 and 9a, a concave-bank bench can be observed in the Figures
upstream
On the contrary, M4of sampling
our field reconnaissance
reach (Figure (Figures 2a and 4a).
7) showed On the other
irregular hand,
meandering behavior.
limb of the outer
7 and2,9aitindicate bank of the first
the occurrence meander
of sinuositybend.
bank erosion Furthermore,
on channel
the downstreama longitudinal
of the outer bar is
bend developing
apexinatthe in
the10-year
From Table can be inferred that the of the decreased from 1.55 to 1.49
the second meander bend, which may be a precursor to concave bank bench. These results are
period. As shown
consistent within
theboth Figures
results of our7 field
and 9a, a concave-bank
reconnaissance bench
(Figures can be
2a and 4a).observed in the
On the other upstream
hand, Figureslimb
of the outer
7 and 9abank of the
indicate thefirst meander
occurrence ofbend. Furthermore,
bank erosion on the adownstream
longitudinalofbar
theisouter
developing in the
bend apex second
at the
meander bend, which may be a precursor to concave bank bench. These results are consistent with the
results of our field reconnaissance (Figures 2a and 4a). On the other hand, Figures 7 and 9a indicate the
occurrence of bank erosion on the downstream of the outer bend apex at the confined meander bends.
Water 2018, 10, 354 10 of 18

Water 2018, 10, x FOR PEER REVIEW 10 of 17

This was also observed during the field site examination (Figure 2b). Apparently, the sharp meander
bends of M4 sampling
confined meander bends.reach migrated downstream
This was also whereas
observed during thethe straight
field portion seemed
site examination to 2b).
(Figure be more
stable. The average
Apparently, the rate
sharpofmeander
bank retreat
bendsin of
theM4M4sampling
sub-reach wasmigrated
reach 0.2 m/year, while it was
downstream 0.08the
whereas m/year
in thestraight
M3 sub-reach.
portion seemed to be more stable. The average rate of bank retreat in the M4 sub-reach was
0.2
To m/year, while itthe
study further washydro-morphodynamics
0.08 m/year in the M3 sub-reach.
of the irregular meandering development observed
To study further the hydro-morphodynamics
in the confined sampling reach in the study creek, spatially of distributed
the irregular meandering
ADCP development
depth-averaged velocities
observed in the confined sampling reach in the study creek, spatially
were employed. Figures 10 and 11 show the interpolated measured depth-averaged velocitiesdistributed ADCP depth-
during
averaged velocities were employed. Figures 10 and 11 show the interpolated measured depth-
both low flow and moderate flow in the meander bends of the M4 and M3 sub-reaches, respectively.
averaged velocities during both low flow and moderate flow in the meander bends of the M4 and M3
As shown in Figure 10, reverse flow occurred in the confined meander bends while the unconfined
sub-reaches, respectively. As shown in Figure 10, reverse flow occurred in the confined meander
meanders
bends had
whilea the
regular flow pattern
unconfined meandersin had
the streamwise
a regular flowdirection
pattern in(Figure 11). Thedirection
the streamwise reversing flow was
(Figure
associated with
11). The the irregular
reversing flow wasbend geometry
associated with and the flowbend
the irregular obstruction
geometryimposed on the
and the flow bend flow by
obstruction
the concave-bank bench.
imposed on the bend flow by the concave-bank bench.

0.5

0.5

Figure 10. Cont.

Water 2018, 10, 354 11 of 18
Water 2018, 10, x FOR PEER REVIEW 11 of 17

0.5

0.5

Figure 10. Measured depth-averaged ADCP velocities in the meander bends of the M4 sub-reach at:
Figure 10. Measured depth-averaged ADCP velocities in the meander bends of the M4 sub-reach at:
(a) first bend during August 2014 (low flow); (b) second bend during August 2014 (low flow); (c) fist
(a) first bend during August 2014 (low flow); (b) second bend during August 2014 (low flow); (c) fist
bend during October 2014 (moderate flow); (d) second bend during October 2014 (moderate flow).
bendRefer
during October 2014 (moderate flow); (d) second bend during October 2014 (moderate flow).
to Figure 1c for location of these bends. Surveyed bathymetric data (2014) are shown in the
Referbackground.
to Figure 1c forthat
Note location of these
the circles showbends. Surveyed
the induced reversebathymetric
eddies. data (2014) are shown in the
background. Note that the circles show the induced reverse eddies.
Water 2018, 10, 354 12 of 18
Water 2018, 10, x FOR PEER REVIEW 12 of 17

A
0.5

0.5

Figure 11. Cont.

Water 2018, 10, 354 13 of 18
Water 2018, 10, x FOR PEER REVIEW 13 of 17

C 0.5

D
0.5

Figure Figure 11. Measured

11. Measured depth-averagedADCP
depth-averaged ADCPvelocities
velocities in
inthe
themeander
meander bends
bendsof the M3 M3
of the sub-reach at: at:
sub-reach
(a) first bend during August 2014 (low flow) (b) second bend during August 2014 (low
(a) first bend during August 2014 (low flow) (b) second bend during August 2014 (low flow) (c) fist flow) (c) fist
bend bend
during during October
October 20142014 (moderate
(moderate flow)
flow) (d) second
(d) second bend bend during
during October
October 2014
2014 (moderateflow).
(moderate flow).Refer
Refer to Figure 1c for location of these bends. Surveyed bathymetric data (2014) are shown
to Figure 1c for location of these bends. Surveyed bathymetric data (2014) are shown in the background. in the
background.

4. Discussion
The morphology of meandering cohesive bed rivers is yet not fully understood, particularly
when they are confined. The results of this study illustrated an irregular meandering pattern in the
Water 2018, 10, 354 14 of 18

confined sub-reach of the study creek. This irregular meandering pattern included development of a
concave-bank bench in the upstream portion of outer bank in the confined sampling reach.
Page and Nanson developed a conceptual model for the formation of a concave-bank bench [40],
which begins with cut-bank erosion that enlarges the channel. The possible growth of a point bar
upstream of a sharp bend would deflect the flow and would leave a region of flow expansion and
separation downstream and beside the upstream limb of the concave bank, which generates large
eddies and induces reverse flow. The developed reverse flow causes the upstream margin of the inner
bend to erode and thus widen the channel. Consequently, compensating sedimentation may occur
in the downstream portion of the inner meander bend, which leads to downstream migration of the
meander. The vacated zone at the upstream limb of the outer bend could be filled with eddy accretion
consisting mostly of fine grains, leading to development of a concave-bank bench. In coherence with
the Page and Nanson conceptual model for development of a concave-bank bench, the encroachment
of the railway embankment on the M4 sampling reach in the study area confined the lateral erosion
and caused the meander to migrate downstream, which provided room for fine-grained accretion
on the outer meander bends. On the other hand, a lower ratio of the bedload to the suspended
load in this semi-alluvial cohesive-bed river may have limited downstream development of the
point bar [43,47]. This may have created an open space for separation zone and flow expansion.
The measured ADCP velocities in the confined meander (Figure 10) confirmed the creation of the
reverse flows caused by the flow expansion which would favor the development of the concave-bank
bench. Spatially distributed ADCP velocity measurements employed in the present study provide
the first corroborative proof for the theory developed by [40], which was not previously supported by
detailed velocity field measurements.
Nanson and Page indicated that during the process of eddy accretion, a longitudinal-shaped bar
could be developed [41], close to the concave bank and upstream of the meander apex, prior to the full
formation of the bench. As a result, a secondary channel may occur since the generated longitudinal
bar bench may fail to migrate completely to the concave bank. This longitudinal bar serves as a core
for further deposition. Gradual deposition and aggradation of the bar leads to complete formation of
the concave-bank bench. Figure 4a shows the creation of a longitudinal-shaped bar and an induced
secondary channel in the confined M4 sampling reach. This can be caused by the sedimentation in the
flow separation zone at the upstream of the outer bank.
Bank instability, which was mostly observed in the confined section, can be attributed to induced
reverse eddy currents, which facilitate the undercut erosion and bank failure on the upstream of the
outer banks. Another possible reason for the observed bank instability in the confined setting could be
linked to elevated pore water pressure during flood drawdown, which cannot rapidly drain due to
the cohesive nature of the bank. With a decline of the spring freshet flow stage and removal of the
confining river pressure, the elevated pore water pressure could reduce the frictional shear strength
and increase the unit weight of the bank material, which may have contributed to the observed bank
failures. Furthermore, construction of the railway embankment may have consolidated the river bank,
further hindering dissipation and drainage of elevated pore water pressures in the river bank.

5. Conclusions
Despite previous research on meander migration patterns, both the impact of channel confinement
and the detailed hydro-morphodynamics in a cohesive meandering clay bed river are not yet fully
understood. The present study examined the meandering behavior of a cohesive clay bed river over a
10-year period. Two sub-reaches of the same meandering cohesive clay bed river were shown to have
different morphodynamic characteristics and migration pattern. The unconfined sampling reach had
a typical meandering pattern with erosion on the outer banks and deposition on the inner banks of
meander bends. The sinuosity of the reach remained more-or-less constant over the ten-year period.
On the other hand, analysis of aerial images along with LIDAR data, total station survey, and field
examination revealed an irregular meandering pattern in the confined sub-reach. The sinuosity of this
Water 2018, 10, 354 15 of 18

part of the creek decreased from 1.55 to 1.49. The average rate of bank retreat was 0.2 and 0.08 m/year
in the confined and unconfined sub-reaches, respectively. The results showed an evolution of the
concave-bank bench on the upstream limb of the outer banks of the sharp menders in the confined
reach, whereas bank instability was observed downstream of the bend apices. It was shown that
different locations along a river, depending on degree of channel confinement, could have distinctly
different morphological characteristics. To explore how the morphodynamics of each sub-reach
could be linked to its hydrodynamics, we employed spatially intensive ADCP surveying. ADCP
measurements confirmed that the averaged values of the flow characteristics such as depth, velocity
and Froude number were quantitatively quite similar in both sub-reach; however, the results of
spatially distributed ADCP depth-averaged velocities confirmed the occurrence of reverse flow on
the upstream limb of the outer meander bends in the confined sub-reach, which could be linked to
the irregular meandering pattern and generation of the concave-bank bench. The results of this study
shed light on the potential impacts of channel confinement on the bank retreat and river migration in
comparable case studies.

Acknowledgments: The authors wish to thank Bina Chakraburtty at the National Capital Commission for project
support and funding, as well as provision of the LIDAR data.
Author Contributions: Parna Parsapour-Moghaddam and Colin D. Rennie conceived and designed the study,
performed the field work, analyzed the data, and wrote the paper. Parsapour-Moghaddam performed the bulk of
the initial data analysis and wrote the first draft of the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Thomson, J. On the origin of windings of rivers in alluvial plains, with remarks on the flow of water round
bends in pipes. Proc. R. Soc. Lond. 1876, 25, 5–8. [CrossRef]
2. Einstein, A. The Cause of the Formation of Meanders in the Courses of Rivers and of the So-Called Baer’s
Law. Die Naturwiss. 1926, 14, 223–224. [CrossRef]
3. Bridge, J.S.; Jarvis, J. Flow and sedimentary processes in the meandering river South Esk, Glen Clova,
Scotland. Earth Surf. Process. Landf. 1976, 1, 303–336. [CrossRef]
4. Thompson, A. Secondary flows and the pool-riffle unit: A case study of the processes of meander
development. Earth Surf. Process. Landf. 1986, 11, 631–641. [CrossRef]
5. Ikeda, S.; Parker, G. (Eds.) River Meandering; American Geophysical Union: Washington, DC, USA, 1989.
6. Demuren, A.O. A numerical model for flow in meandering channels with natural bed topography.
Water Resour. Res. 1993, 29, 1269–1277. [CrossRef]
7. Blanckaert, K.; de Vriend, H.J. Nonlinear modeling of mean flow redistribution in curved open channels.
Water Resour. Res. 2003, 39. [CrossRef]
8. Lanzoni, S.; Seminara, G. On the nature of meander instability. J. Geophys. Res. Earth Surf. 2006, 111.
[CrossRef]
9. Alam, J.B.; Uddin, M.; Ahmed, U.J.; Cacovean, H.; Rahman, H.M.; Banik, B.K.; Yesmin, N. Study of
morphological change of river old Brahmaputra and its social impacts by remote sensing. Geogr. Tech. 2007,
2, 1–11.
10. Alam, J.B.; Uddin, M.; Ahmed, J.U.; Rahman, M.H.; Banik, B.K.; Yesmin, N.; Islam, M. Study of the
morphological change of the river old Brahmaputra and its impacts. Asian J. Water Environ. Pollut. 2009, 6,
11–19.
11. Blanckaert, K. Hydrodynamic processes in sharp meander bends and their morphological implications.
J. Geophys. Res. Earth Surf. 2011, 116. [CrossRef]
12. Kasvi, E.; Vaaja, M.; Alho, P.; Hyyppä, H.; Hyyppä, J.; Kaartinen, H.; Kukko, A. Morphological changes on
meander point bars associated with flow structure at different discharges. Earth Surf. Process. Landf. 2013, 38,
577–590. [CrossRef]
13. Asahi, K.; Shimizu, Y.; Nelson, J.; Parker, G. Numerical simulation of river meandering with self-evolving
banks. J. Geophys. Res. Earth Surf. 2013, 118, 2208–2229. [CrossRef]
Water 2018, 10, 354 16 of 18

14. Kasvi, E.; Vaaja, M.; Kaartinen, H.; Kukko, A.; Jaakkola, A.; Flener, C.; Hyyppä, H.; Hyyppä, J.; Alho, P.
Sub-bend scale flow–sediment interaction of meander bends—A combined approach of field observations,
close-range remote sensing and computational modelling. Geomorphology 2015, 238, 119–134. [CrossRef]
15. Parsapour-Moghaddam, P.; Rennie, C.D. Hydrostatic versus nonhydrostatic hydrodynamic modelling
of secondary flow in a tortuously Meandering River: Application of Delft3D. River Res. Appl. 2017, 33,
1400–1410. [CrossRef]
16. Parsapour-Moghaddam, P.; Rennie, C.D. Calibration of a 3D Hydrodynamic Meandering River Model Using
Fully Spatially Distributed 3D ADCP Velocity Data. J. Hydraul. Eng. 2018, 144, 04018010.
17. Schuurman, F.; Shimizu, Y.; Iwasaki, T.; Kleinhans, M.G. Dynamic meandering in response to upstream
perturbations and floodplain formation. Geomorphology 2016, 253, 94–109. [CrossRef]
18. Hooke, J. River meander behaviour and instability: A framework for analysis. Trans. Inst. Br. Geogr. 2003, 28,
238–253. [CrossRef]
19. Hodskinson, A.; Ferguson, R.I. Numerical modelling of separated flow in river bends: Model testing and
experimental investigation of geometric controls on the extent of flow separation at the concave bank.
Hydrol. Process. 1998, 12, 1323–1338. [CrossRef]
20. Ferguson, R.I.; Parsons, D.R.; Lane, S.N.; Hardy, R.J. Flow in meander bends with recirculation at the inner
bank. Water Resour. Res. 2003, 39. [CrossRef]
21. Blanckaert, K.; Kleinhans, M.G.; McLelland, S.J.; Uijttewaal, W.S.; Murphy, B.J.; van de Kruijs, A.;
Parsons, D.R. Flow Separation and Morphology in Sharp Meander Bends. In Proceedings of the AGU
Fall Meeting, San Francisco, CA, USA, 13–17 December 2010; American Geophysical Union: Washington,
DC, USA, 2010.
22. Constantinescu, G.; Kashyap, S.; Tokyay, T.; Rennie, C.D.; Townsend, R.D. Hydrodynamic processes and
sediment erosion mechanisms in an open channel bend of strong curvature with deformed bathymetry.
J. Geophys. Res. Earth Surf. 2013, 118, 480–496. [CrossRef]
23. Blanckaert, K.; Kleinhans, M.G.; McLelland, S.J.; Uijttewaal, W.S.; Murphy, B.J.; Kruijs, A.; Parsons, A.;
Chen, Q. Flow separation at the inner (convex) and outer (concave) banks of constant-width and widening
open-channel bends. Earth Surf. Process. Landf. 2013, 38, 696–716. [CrossRef]
24. Nardi, L.; Campo, L.; Rinaldi, M. Quantification of riverbank erosion and application in risk analysis.
Nat. Hazards 2013, 69, 869–887. [CrossRef]
25. Rinaldi, M.; Casagli, N.; Dapporto, S.; Gargini, A. Monitoring and modelling of pore water pressure changes
and riverbank stability during flow events. Earth Surf. Process. Landf. 2004, 29, 237–254. [CrossRef]
26. Simon, A.; Collison, A.J. Pore-water pressure effects on the detachment of cohesive streambeds: Seepage
forces and matric suction. Earth Surf. Process. Landf. 2001, 26, 1421–1442. [CrossRef]
27. Pizzuto, J. An empirical model of event scale cohesive bank profile evolution. Earth Surf. Process. Landf. 2009,
34, 1234–1244. [CrossRef]
28. Salem, H.; Rennie, C.D. Practical determination of critical shear in cohesive soils. J. Hydraul. Eng. (ASCE)
2017, 143, 04017045. [CrossRef]
29. Eke, E.; Parker, G.; Shimizu, Y. Numerical modeling of erosional and depositional bank processes in migrating
river bends with self-formed width: Morphodynamics of bar push and bank pull. J. Geophys. Res. Earth Surf.
2014, 119, 1455–1483. [CrossRef]
30. Casagli, N.; Curini, A.; Gargini, A.; Rinaldi, M.; Simon, A. Effects of pore pressure on the stability of
streambanks: Preliminary results from the Sieve River, Italy. In Proceedings of the Management of
Landscapes Disturbed by Channel Incision, Oxford, MS, USA, 19–23 May 1997; pp. 243–248.
31. Simon, A.; Curini, A. Pore pressure and bank stability: The influence of matric suction. In Proceedings of the
Water Resources Engineering 98, Memphis, TN, USA, 3–7 August 1998; pp. 358–363.
32. Rinaldi, M.; Casagli, N. Stability of streambanks formed in partially saturated soils and effects of negative
pore water pressures: The Sieve River (Italy). Geomorphology 1999, 26, 253–277. [CrossRef]
33. Simon, A.; Curini, A.; Darby, S.E.; Langendoen, E.J. Bank and near-bank processes in an incised channel.
Geomorphology 2000, 35, 193–217. [CrossRef]
34. Choné, G.; Biron, P.M. Assessing the relationship between river mobility and habitat. River Res. Appl. 2016,
32, 528–539. [CrossRef]
35. Nicoll, T.J.A. Planform Geometry and Kinematics of Confined Meandering Rivers on the Canadian Prairies.
Master’s Thesis, Simon Fraser University, Burnaby, BC, Canada, 2008.
Water 2018, 10, 354 17 of 18

36. Ghinassi, M.; Ielpi, A.; Aldinucci, M.; Fustic, M. Downstream-migrating fluvial point bars in the rock record.
Sediment. Geol. 2016, 334, 66–96. [CrossRef]
37. Lewin, J.; Brindle, B.J. Confined meanders. In River Channel Changes; Gregory, K.J., Ed.; John Wiley and Sons:
Chichester, UK, 1977; pp. 221–233.
38. Lane, E.W. A Study of the Shape of Channels Formed by Natural Streams Flowing in Erodible material; US Army
Engineer Division, Sediment Series, 9; Mississippi River Corps of Engineers: Omaha, NE, USA, 1957.
39. Nicoll, T.J.; Hickin, E.J. Planform geometry and channel migration of confined meandering rivers on the
Canadian prairies. Geomorphology 2010, 116, 37–47. [CrossRef]
40. Page, K.; Nanson, G. Concave-bank benches and associated floodplain formation. Earth Surf. Process. Landf.
1982, 7, 529–543. [CrossRef]
41. Nanson, G.C.; Page, K. Lateral accretion of fine-grained concave benches on Meandering Rivers. In Modern
and Ancient Fluvial Systems; Collinson, J., Lewin, J., Eds.; International Association of Sedimentologists
Special Publication: Gent, Belgium, 1983; Volume 6, pp. 133–143.
42. Burge, L.M.; Smith, D.G. Confined meandering river eddy accretions: sedimentology, channel geometry
and depositional processes. In Fluvial Sedimentology VI; Smith, N.D., Rogers, J., Eds.; Special Publication of
Internat, the International Association of Sedimentologists: Gent, Belgium, 1999; Volume 28, pp. 113–130.
43. Makaske, B.; Weerts, H.J. Muddy lateral accretion and low stream power in a sub-recent confined channel
belt, Rhine-Meuse Delta, central Netherlands. Sedimentology 2005, 52, 651–668. [CrossRef]
44. Smith, D.G.; Hubbard, S.M.; Leckie, D.A.; Fustic, M. Counter point bar deposits: lithofacies and reservoir
significance in the meandering modern Peace River and ancient McMurray Formation, Alberta, Canada.
Sedimentology 2009, 56, 1655–1669. [CrossRef]
45. Daniel, J.F. Channel Movement of Meandering Indiana Streams (No. 732); United States Geological Survey
Professional Paper 732-A; Geological Survey: Reston, VA, USA, 1971; 18p.
46. Lewin, J. Meander development and floodplain sedimentation: A case study from mid-Wales. Geol. J. 1978,
13, 25–36. [CrossRef]
47. Jamieson, E.C.; Ruta, M.A.; Rennie, C.D.; Townsend, R.D. Monitoring stream barb performance in a
semi-alluvial meandering channel: Flow field dynamics and morphology. Ecohydrology 2013, 6, 611–626.
[CrossRef]
48. Maarschalk-Bliss, S. Seasonal Variation in Assemblage Structure and Movement of Small Stream Fish in an
Urban Environment. Master’s Thesis, Biology, Carleton University, Ottawa, ON, Canada, 2014.
49. Salem, H.; Rennie, C.D.; Custodio, C.Z. Influence of Pore Pressure on Clay Erosion. In Proceedings of the
River Flow Proceedings, Lausanne, Switzerland, 3–5 September 2014.
50. Biron, P.M.; Choné, G.; Buffin-Bélanger, T.; Demers, S.; Olsen, T. Improvement of streams hydro-
geomorphological assessment using LIDAR DEMs. Earth Surf. Process. Landf. 2013, 38, 1808–1821. [CrossRef]
51. De Rose, R.C.; Basher, L.R. Measurement of river bank and cliff erosion from sequential LIDAR and historical
aerial photography. Geomorphology 2011, 126, 132–147. [CrossRef]
52. Williams, R.D.; Rennie, C.D.; Brasington, J.; Hicks, D.M.; Vericat, D. Linking the spatial distribution of bed
load transport to morphological change during high-flow events in a shallow braided river. J. Geophys. Res.
Earth Surf. 2015, 120, 604–622. [CrossRef]
53. Parsapour-moghaddam, P.; Rennie, C.D.; Midwood, J.; Cvetkovic, M.; Cooke, S.J. Influence of Channel
Erosion on Fish Habitat Utilization. In Proceedings of the 36th IAHR World Congress, Hague,
The Netherlands, 28 June–3 July 2015.
54. Brennan, C.P.; Parsapour-moghaddam, P.; Rennie, C.D.; Seidou, O. Continuous prediction of clay-bed stream
erosion in response to climate model output for a small urban watershed. J. Hydrol. Process. 2018, in press.
[CrossRef]
55. Rennie, C.D.; Church, M. Mapping spatial distributions and uncertainty of water and sediment flux in a
large gravel bed river reach using an acoustic Doppler current profiler. J. Geophys. Res. Earth Surf. 2010, 115,
F03035. [CrossRef]
Water 2018, 10, 354 18 of 18

56. Parsapour-Moghaddam, P.; Rennie, C.D. ADCP Validation of 3D Morphodynamic Modelling in Clay-Bed
Meandering Rivers. In Proceedings of the 36th IAHR World Congress, Hague, The Netherlands,
28 June–3 July 2015.
57. Parsapour-Moghaddam, P.; Rennie, C.D. 3D versus 2D calibration of a 3D hydrodynamic model.
In Proceedings of the 37th IAHR World Congress, Kuala Lumpur, Malaysia, 13–18 August 2017.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).