Content uploaded by Carl J. Legleiter
Author content
All content in this area was uploaded by Carl J. Legleiter on Jan 02, 2015
Content may be subject to copyright.
LETTERS
PUBLISHED ONLINE: 2 NOVEMBER 2014 | DOI: 10.1038/NGEO2282
Sediment supply as a driver of river meandering
and floodplain evolution in the Amazon Basin
José Antonio Constantine1*, Thomas Dunne2,3*, Joshua Ahmed1, Carl Legleiter4and Eli D. Lazarus1
The role of externally imposed sediment supplies on the
evolution of meandering rivers and their floodplains is poorly
understood, despite analytical advances in our physical under-
standing of river meandering1,2. The Amazon river basin hosts
tributaries that are largely unaected by engineering controls
and hold a range of sediment loads, allowing us to explore the
influence that sediment supply has on river evolution. Here we
calculate average annual rates of meander migration within 20
reaches in the Amazon Basin from Landsat imagery spanning
1985–2013. We find that rivers with high sediment loads
experience annual migration rates that are higher than those of
rivers with lower sediment loads. Meander cuto also occurs
more frequently along rivers with higher sediment loads.
Dierences in meander migration and cuto rates between
the study reaches are not explained by dierences in channel
slope or river discharge. Because faster meander migration
and higher cuto rates lead to increased sediment-storage
space in the resulting oxbows, we suggest that sediment
supply modulates the reshaping of floodplain environments
by meandering rivers. We conclude that imposed sediment
loads influence planform changes in lowland rivers across
the Amazon.
Decades of fluvial research have yielded physical insight
into river behaviours associated with various channel patterns,
from planform controls on meander migration, to mechanisms
responsible for dispersing sediment into the floodplain2,3 and
the in-channel dynamics that produce sediment supply from
local bank erosion4. By contrast, the role of externally imposed
sediment supplies in channel planform dynamics—that is, when
sediment supplies are an imposed environmental variable to which
the channel planform must adjust—remains poorly understood.
External sources of sediment may derive from mountain building,
glacial and other climatic changes, volcanism, and land uses that
accelerate landscape-scale soil loss. They may also be interrupted
by water resource developments such as the large reservoirs under
consideration for many Amazonian tributaries5.
The Amazon Basin remains a unique opportunity to assess
the role of externally imposed sediment supplies in river
evolution because of the generally undeveloped nature of its
fluvial environments and the range of sediment supplies in its
various tributary basins. Rivers within the 6.9 ×106km2basin
are generally free to modify their forms, and they drain distinct
physiographic provinces, each of which supplies sediment to the
channel network at different rates6,7. Rivers draining the highly
erodible Andean Cordillera yield the greatest amount of sediment
to the Amazon lowlands, on the order of 100 Mt of suspended
material per river each year8–10. Rivers with headwaters draining the
fine-grained Neogene sedimentary rocks of the Central Amazon
Trough yield an order of magnitude less sediment per river each
year9, and those that drain the Guiana Shield to the north and the
Brazil Shield to the south of the Trough yield the least7.
As sedimentary deposits attached to the inner banks (convex
relative to the channel centreline) of meander bends, point bars
seem to be the link between changes in externally imposed
sediment supplies and adjustments in river behaviour. Modern
theory attributes the meandering behaviour of alluvial rivers to
instabilities in river flow11 that arise either inherently12 or from
disruption of the flow field by bedforms13. Recent lab experiments
suggest that this latter influence may not be ubiquitous14, but the
capacity for point bars to promote meandering by altering the flow
field15 has been observed both in the field16 and the laboratory17 .
A view of the Rio Mamoré, flowing through the Andean foreland
basin of Bolivia, offers an illustration. The Rio Grande, a tributary
of the Rio Mamoré, drains 7.2×104km2of the Bolivian Andes and
transports about 136 Mt yr−1of suspended sediment as it enters the
foreland basin10. The substantial sediment load of the Rio Grande
has resulted in the enhanced development of point bars on the Rio
Mamoré downstream of the confluence (Fig. 1a) that has instigated
a nearly 1.7-fold increase in rates of meander migration (Fig. 1b).
Observations from the Sacramento River of California, USA, further
Mamoré River
5 km
Flow
La Satísima
Trinidad
Rio Grande
(∼69 Mt yr
−1
)
M0
0.039 ch-w yr
−1
M1
0.066 ch-w yr
−1
M2
0.047 ch-w yr
−1
35 km
Puerto Ganadero
(64 Mt yr
−1
)
Flow
Rio Grande
N
Input into M1
(∼82 Mt yr
−1
)
a
b
Flow N
Puerto Villaroel
(9 Mt yr−1)
Rio Chimoré
(∼4 Mt yr−1)
Mamoré River
Figure 1 | Enhanced development of point bars on the Rio Mamoré
downstream of the confluence with the Rio Grande. a, Landsat image
from 2013 of the Rio Mamoré and its confluence with the Rio Grande,
which drains the Bolivian Andes from the southeast. b, Study reaches
of the Mamoré (M0, green; M1, red; M2, blue), with average annual
meander migration rates provided for each (in units of channel widths
per year) for the time period 1986–2010. The box shows the location
of the image in a. Estimates of total suspended sediment (TSS) flux are
shown in brackets. Details of TSS flux estimates can be found in the
Supplementary Information.
1School of Earth and Ocean Sciences, Cardi University, Cardi CF10 3AT, UK. 2Bren School of Environmental Science and Management, University of
California–Santa Barbara, Santa Barbara, California 93106, USA. 3Department of Earth Science, University of California–Santa Barbara, Santa Barbara,
California 93106, USA. 4Department of Geography, University of Wyoming, Laramie, Wyoming 82071, USA. *e-mail: constantineja@cardi.ac.uk;
tdunne@bren.ucsb.edu
NATURE GEOSCIENCE | VOL 7 | DECEMBER 2014 | www.nature.com/naturegeoscience 899
© 2014 Macmillan Publishers Limited. All rights reserved.
LETTERS NATURE GEOSCIENCE DOI: 10.1038/NGEO2282
M0
M1
M2
Ma
Hu
Ar, Araguaia; Va, Vaupés; Br, Branco; Ir, Iriri; Xi, Xingu; P1, Purus1;
P2, Purus2; Ju, Juruá; Jt, Jutai; It , Ituí; Cu, Curuca; Na, Nanay;
Pt, Putumayo; M0, Mamoré0; M1, Mamoré1; M2, Mamoré2; Be, Beni;
Uc, Ucayali; Hu, Huallaga; Ma, Madre de Díos
Shield
Central Trough
Andes–Foreland Basin
80° W70° W60° W50° W
Be
Uc
Br
Va
Ir
Xi Ar
P2
Ju
P1
Cu It Jt
Pt
0°
10° S
0.0
0.1
0.3
0.2
0.4
0.000
0.001
0.002
0.0
0.1
0.3
0.2
0.4
0.00
0.03
0.06
Mean TSS
(Mt yr−1 m−1)
Mean MR
(ch-w yr−1)
Mean CR
(no. yr−1 ch-w−1)
Mean NL
(no. ch-w−1)
0.0001 0.0010 0.0100
MR (ch-w yr−1)
CR (no. yr−1 ch-w−1)
TSS (Mt yr−1 m−1)TSS (Mt yr−1 m−1)
0.1000 1.0000 0.0001 0.0010 0.0100 0.1000 1.0000
0.005
0.010
P1
Uc
M1
Be
M2
Ma
P2
0.020
0.050
0.100
M0 Hu
Ir Jt Pt
Xi
Ar
Ju
Va Na
Br
MR = 0.043TSS0.28
= 0.73, B = 0.53
< 0.001
It
Cu
0.00005
0.00020 P1
Uc
M1
Be
M2
Ma
P2
0.00050
0.00200
0.00500
M0
Hu
Ir
Jt Pt
Xi
Ar Ju
Va
Na
Br
CR = 0.0016TSS0.43
= 0.66, B = 0.51
< 0.002
It
Cu
0.00100
0.00010
Shield Trough Andes
Shield Trough Andes
Shield Trough Andes
Shield Trough Andes
Na
ρ
ρ
τ
τ
α
α
a
cd
b
Figure 2 | Study sites, river characteristics, and sediment fluxes. a, Locations of study reaches within the physiographic provinces with reach
abbreviations defined. b, Mean values for total suspended sediment (TSS) flux (megatonnes per year per metre of channel width), average annual meander
migration rate (MR; channel widths per year), average annual meander cuto rate (CR) (number per year per channel width of down-valley distance), and
total number of observable oxbow lakes (NL; number per channel width of down-valley distance). Error bars represent ±1σabout the means. c,MRplotted
against TSS flux. d,CRplotted against TSS flux.
highlight the role of point bars in facilitating bank erosion in reaches
that experience a net accumulation of bed material18.
Within a given meander, curvature and shoaling of river flow
onto and over a point bar force high-momentum fluid to be
advected outwards, increasing boundary shear stress along the outer
(concave) riverbank and decreasing boundary shear stress over
the inner-bank bar15. Elevated shear stresses near the outer bank
promote the fluvial entrainment of the bank material primarily
responsible for bank erosion19, and the displacement of the zone
of maximum shear stress from the inner to the outer bank shifts
the zones of maximum bed-material transport and bank scour
accordingly15. As a result, the bar aggrades, growing laterally and
vertically20 until its surface is infrequently inundated, promoting the
continued advection of high-momentum fluid by increasing flow
curvature and asymmetric shoaling.
To assess the importance of sediment supply in meandering river
evolution, we used sequences of Landsat images for twenty reaches
from across three physiographic provinces of the Amazon Basin to
calculate average annual rates of meander migration and meander
cutoff (Fig. 2a), which were related to reported estimates of the
average annual total suspended sediment (TSS) supply to each reach.
Meander migration rates (MR) were normalized by average bankfull
channel width (RW) to account for the influence of channel size on
rates of lateral shifting. Cutoff rates (CR) were normalized by reach
valley-length, expressed in channel-width units (ch-w). TSS flux
estimates were normalized by RWto account for the ability of wider
channels with greater flows to convey more sediment. We used TSS
data from the most extensive and systematically collected survey of
sediment supplies currently available for the Amazon Basin7. Details
of values from other sources are provided in the Supplementary
Information. TSS flux is the only widely (if sparsely) distributed
measure of sediment supplies in the Amazon and other large river
systems. It includes both sandy bed material, which is the material of
bar formation, and silt–clay washload, which is transported through
the channel and into floodplains without being sequestered onto
bars. We used TSS as a proxy for bed-material load (including
sediment transported as bed load or suspended load), which will
be required until direct measurements of bed-material flux become
available. Such measurements are available for only one season in a
single reach of an Amazonian tributary21. For the Amazon mainstem
and seven of its tributaries, it has been estimated that bedload
transport rates are <1% of total load, the sandy bed-material
transport is dominated by suspension and annual suspended sand
loads comprise an average of 20% of TSS for both the mainstem
Amazon and its tributaries (17% for tributaries with no or very small
Andean sources)3. In our analysis, we grouped rivers according
900 NATURE GEOSCIENCE | VOL 7 | DECEMBER 2014 | www.nature.com/naturegeoscience
© 2014 Macmillan Publishers Limited. All rights reserved.
NATURE GEOSCIENCE DOI: 10.1038/NGEO2282 LETTERS
Uc
0.100
MR (ch-w yr−1)
MR (ch-w yr−1)TSS (Mt yr−1 m−1)
P1
M1
Be
M2
Ma
Br
0.005 0.010 0.020 0.050 0.100
MR (ch-w yr−1)
0.005 0.010 0.020 0.050
0.0050
CR (no. yr−1 ch-w−1)
NL (no. ch-w−1)
M0
0.0020
0.0010
0.0005
0.0001
0.0005
Hu
P2
CR = 0.07MR1.29
= 0.77, B = 0.62
< 0.001
0.0002
Ju
Va
Na
Ar
Pt
Xi
Jt
Ir
Cu It
Change in sinuosity (yr−1)
0.000
−0.004
−0.008
−0.012
−0.016
M1
Uc
Be
M2
Ma
Ju
P1
P2
0.004
0.008
M0
Hu
Cu
Br
−0.020
Jt
Ir
Pt
Xi
Na
Ar
Va
It
Be
M0
Uc
M2
P2
Br Ar
Xi Va
Cu
It Jt
Ir
0.010 0.100 0.001 0.010 0.100 1.000
P1
Ma
Na
Pt
0.001
0.002
0.050
0.100
0.500
0.005
0.010
0.020
0.200
Be
Hu
M0
Uc
M2 M1
P1
Ma
P2
Br
Na
Jt
Pt
Xi Va
Ar
Ju
Cu
Ir
M1
Hu
Ju
ρ τ
α
NL = 5.1MR0.98
= 0.75, B = 0.58
< 0.001
ρ τ
α
NL = 0.48TSS0.46
= 0.69, B = 0.51
< 0.001
ρ τ
α
ab
c
Figure 3 | Meander migration and cuto in the maintenance of channel sinuosity. a, Average annual rate of meander cuto (CR; number per year per
channel-width down-valley distance) for the 20 study reaches shown in Fig. 2a plotted against average annual meander migration rate (MR; channel widths
per year). b, Average annual rate of change in channel sinuosity plotted against MRfor each reach. c, Observable oxbow lakes counted within the
floodplains of the study reaches (NL; number per channel width of down-valley distance) plotted against MRand reported total suspended sediment (TSS)
flux (megatonnes per year per metre of channel width).
to their dominant sediment-source regions: the Andes (Andes–
Foreland Basin), the Central Amazon Trough, and the Guiana
Shield and Brazil Shield (shields). We used two-tailed t-tests and
Kruskal–Wallis tests (KW) to quantify the significance of differences
in our measurements among these populations. Spearman’s rank
correlation coefficients (ρ) and Kendall rank correlation coefficients
(τB) provide measures of significance for correlations.
Our results reveal statistically significant differences between the
20 study reaches based on physiography (Fig. 2b). Reported TSS
fluxes for rivers from the Andes–Foreland Basin are significantly
greater than the fluxes for rivers of the Central Trough and
shields (t-tests: α < 0.001; KW: α < 0.005). The sediment-laden
rivers of the Andes–Foreland Basin also experience the highest
rates of meander migration and cutoff—rates significantly greater
than those in rivers of the Central Trough and shields (t-tests:
α < 0.001; KW: α < 0.001). Meander migration and cutoff rates
for the study reaches positively correlate with estimates of TSS
flux (Fig. 2c,d). The systematic differences observed here are not
associated with differences in average channel gradient, at least
within the uncertainties of available Shuttle Radar Topography
Mission data (Supplementary Fig. 2), nor associated with differences
in river discharge, which was accounted for through normalization
by RW, which scales with bankfull discharge22.
Although migration rates increase with TSS flux, channel
sinuosity has been maintained within a small range owing to the
corresponding increase in the frequency of cutoffs (Fig. 3a). Within
the time period of investigation, sinuosity for 18 of the study reaches
varied by only ±6%, with the sinuosity for the remaining two
reaches reduced by roughly 17% owing to a temporal clustering of
cutoffs (Fig. 3b). Maintenance of sinuosity through this mechanism
results in floodplains of the rapidly migrating, sediment-laden
rivers of the Andes–Foreland Basin being populated by the greatest
density of open-water oxbow lakes among the sediment-source
regions (Fig. 2b) (t-tests: α < 0.001; KW: α< 0.005). Oxbow density
also positively and significantly correlates with rates of meander
migration and TSS flux (Fig. 3c).
The production of oxbow lakes by meander cutoff creates
negative relief23 or accommodation space for sediment storage
in floodplains, important to the development of lentic (that is,
relatively still water) sedimentary environments, landforms and
ecosystems, as well as to the ability of floodplains to entrap
carbon and pollutants in developed regions. If channel sinuosity
(T) is constant over time, then the rate of production of valley
accommodation space (SV) per channel width of down-valley
distance (ch-w3yr−1ch-w−1) is given by
SV=CRLRWRD(1)
as long as a central tendency (L) exists for oxbow lengths, and
the bankfull width (RW) and cross-sectionally averaged bankfull
depth (RD) of oxbows approximate those of the channel at the time
of cutoff. Most of this accommodation space will be occupied by
wash load in river environments where neck cutoff is commonplace
and by variable amounts of bed material where chute cutoff
occurs24. Because the planform deformation of meanders is an
important driver of the cutoff process, the role of the average annual
rate of meander migration in determining SVcan be examined
using the empirical relation identified here (Fig. 3a), where CRis
NATURE GEOSCIENCE | VOL 7 | DECEMBER 2014 | www.nature.com/naturegeoscience 901
© 2014 Macmillan Publishers Limited. All rights reserved.
LETTERS NATURE GEOSCIENCE DOI: 10.1038/NGEO2282
1.1
1.5
1.9
2.3
2.7
2.5
2.1
1.7
1.3
0.00 0.01 0.02 0.03 0.04
MR (ch-w yr−1)
0.05 0.06 0.07
SV (ch-w3 yr−1 ch-w−1)
SV (ch-w3 yr−1 ch-w−1)
0.003
0.002
0.001
0.000
Temporally invariable sinuosities
0.0001 0.0010 0.0100
TSS (Mt yr−1 m−1)
0.1000 1.0000
0.00002
0.00010
P1
Uc
M1
Be
M2
Ma
P2
0.00020
0.00100
0.00200
M0
Hu
Ir
Jt
Pt
Xi
Ar
Ju
Va
Na
Br
SV = 0.0010TSS0.40
= 0.76, B = 0.55
< 0.001
It
Cu
0.00050
0.00005
ρτ
α
a
b
Figure 4 | Meander cuto in the production of valley accommodation
space for sediment storage. a, Average annual rate of production of valley
accommodation space for sediment storage (SV) generated by meander
cuto (cubic channel widths per year per channel width of down-valley
distance) plotted against average annual meander migration rates (MR;
channel widths per year) of rivers with temporally invariable sinuosities.
b, Predictions of SVplotted against reported values of total suspended
sediment (TSS) flux (megatonnes per year per metre of channel width) for
the 20 study reaches shown in Fig. 2a.
approximated as 0.07M1.29
R. A worldwide survey of 911 oxbows25
observed that L(in channel-width units) can be approximated
as 3.0e0.82T, and after approximating RDas 0.05RWbased on
observations26, solutions to equation (1) reveal the sensitivity of SV
to rates of meander migration and sinuosity (Fig. 4a). SVincreases
with migration rate because the river produces oxbows more rapidly
while maintaining its gradient and channel length. SValso increases
with sinuosity because cutoffs on more sinuous channels produce
longer oxbows that provide more space for storing sediment.
There is no indication of a secular trend in oxbow density, and
so a steady state in SVseems to develop over a sufficiently long
period, defined in part by the time required for the removal of
oxbows by meander migration and alluviation27,28. The important
implication here is that externally imposed sediment supplies that
promote bar building and meander migration lead to an increase
in valley-wide sediment-storage potential due to the indirect
consequence of increasing cutoff rates, as reflected by solutions to
equation (1) for the 20 study reaches (Fig. 4b). Predictions of SV
positively correlate with TSS flux and are greatest within reaches
of the Andes–Foreland Basin (t-tests: α < 0.001; KW: α < 0.005;
Supplementary Table 2). Oxbow alluviation can significantly
enhance valley-wide sediment storage29,30, and our findings suggest
that the addition of bed material to the channel should also increase
the topographic and sedimentological complexity of the floodplain
as a dynamic but steady-state planform develops and is maintained.
The dynamism of the planforms within the sediment-rich rivers of
the Andes–Foreland Basin indicates the great sensitivity of these
river systems to the disruption of sediment supplies by large water-
storage reservoirs proposed for the Amazon Basin5.
Methods
We used geo-referenced Landsat images recorded between 1985 and 2013 (three
to five per river) to digitize channel centrelines from points delineating vegetated
bank lines at a spacing of approximately one channel width in the downstream
direction. Average bankfull channel width was calculated from at least 15
measurements along straight channel segments between meanders using the most
recent image available. Sequential images of a given reach yielded different
centrelines, and intervals between crossings in two sequential centrelines define
polygons from which minimum average annual migration rates were calculated18.
The migration rates are minima because the direction of channel shift may have
reversed for some portion of time between images. Sinuosity was calculated as
the ratio of channel centreline length to the sum of the channel-belt axis lengths.
Meander cutoffs were recorded between consecutive image dates, and the average
annual meander cutoff rate was calculated as (no. +1)/N, where no. is the
number of cutoff events observed and Nis the number of years between recorded
images. This calculation is consistent with the fact that satellite records are of
finite length, allowing only approximations to the true frequencies of counted
events. The number of visible oxbows was tabulated from the most recent images,
with oxbows recognized as arcuate open-water bodies. Details of the TSS flux
estimates can be found in the Supplementary Information.
Received 23 May 2014; accepted 3 October 2014;
published online 2 November 2014;
corrected online 7 November 2014
References
1. Pittaluga, M. B. & Seminara, G. Nonlinearity and unsteadiness in river
meandering: A review of progress in theory and modelling. Earth Surf. Process.
Landf. 36, 20–38 (2011).
2. Asahi, K., Shimizu, Y., Nelson, J. & Parker, G. Numerical simulation of river
meandering with self-evolving banks. J. Geophys. Res. 118, 2208–2229 (2013).
3. Dunne, T., Mertes, L. A. K., Meade, R. H., Richey, J. E. & Forsberg, B. R.
Exchanges of sediment between the flood plain and channel of the Amazon
River in Brazil. Geol. Soc. Am. Bull. 110, 450–467 (1998).
4. Engel, F. L. & Rhoads, B. L. Interaction among mean flow, turbulence, bed
morphology, bank failures and channel planform in an evolving compound
meander loop. Geomorphology 163–164, 70–83 (2012).
5. Finer, M. & Jenkins, C. N. Proliferation of hydroelectric dams in the Andean
Amazon and implications for Andes-Amazon connectivity. PLoS ONE 7,
e35126 (2012).
6. Baker, V. R. Fluvial Sedimentology—Modern Rivers: Geomorphology and
Sedimentology Vol. 5, 211–230 (Can. Soc. Petrol. Geol., 1977).
7. Filizola, N. & Guyot, J. L. Suspended sediment yields in the Amazon basin: An
assessment using the Brazilian national data set. Hydrol. Process. 23,
3207–3215 (2009).
8. Guyot, J. L. et al. Proc. 4th Int. Symp. River Sedimentation (International
Network on Erosion and Sedimentation, 1989).
9. Latrubesse, E. M., Stevaux, J. C. & Sinha, R. Tropical rivers. Geomorphology 70,
187–206 (2005).
10. Guyot, J. L., Bourges, J. & Cortez, J. Variability in Stream Erosion and Sediment
Transport Vol. 224, 223–231 (IAHS Publ., 1994).
11. Callander, R. A. Instability and river channels. J. Fluid Mech. 36,
465–480 (1969).
12. Gorycki, M. A. Hydraulic drag: A meander-initiating mechanism. Bull. Geol.
Soc. Am. 84, 175–186 (1973).
13. Ikeda, S., Parker, G. & Sawai, K. Bend theory of river meanders. Part 1. Linear
development. J. Fluid Mech. 112, 363–377 (1981).
14. Van de Lageweg, W. I., van Dijk, W. M., Baar, A. W., Rutten, J. &
Kleinhans, M. G. Bank pull or bar push: What drives scroll-bar formation on
meandering rivers. Geology 42, 319–322 (2014).
15. Dietrich, W. E., Smith, J. D. & Dunne, T. Flow and sediment transport in a sand
bedded meander. J. Geol. 87, 305–315 (1979).
16. Lewin, J. Initiation of bed forms and meanders in coarse-grained sediment.
Geol. Soc. Am. Bull. 87, 281–285 (1976).
17. Braudrick, C. A., Dietrich, W. E., Leverich, G. T. & Sklar, L. Experimental
evidence for the conditions necessary to sustain meandering in coarse-bedded
rivers. Proc. Natl Acad. Sci. USA 106, 16936–16941 (2009).
18. Constantine, C. R. Quantifying the Connections Between Flow, Bar Deposition,
and Meander Migration in Large Gravel-Bed Rivers PhD thesis,
Univ. California (2006).
902 NATURE GEOSCIENCE | VOL 7 | DECEMBER 2014 | www.nature.com/naturegeoscience
© 2014 Macmillan Publishers Limited. All rights reserved.
NATURE GEOSCIENCE DOI: 10.1038/NGEO2282 LETTERS
19. Pizzuto, J. E. & Meckelnburg, T. S. Evaluation of a linear bank erosion equation.
Wat. Resour. Res. 25, 1005–1013 (1989).
20. Legleiter, C. J., Harrison, L. R. & Dunne, T. Effect of point bar development on
the local force balance governing flow in a simple, meandering gravel-bed river.
J. Geophys. Res. 116, http://dx.doi.org/10.1029/2010JF001838 (2011).
21. Latrubesse, E. M., Amsler, M. L., de Morais, R. P. & Aquino, S. The
geomorphologic response of a large pristine alluvial river to tremendous
deforestation in the South American tropics: The case of the Araguaia River.
Geomorphology 113, 239–252 (2009).
22. Beighley, R. E. & Gummadi, V. Developing channel and floodplain dimensions
with limited data: A case study in the Amazon Basin. Earth Surf. Process. Landf.
36, 1059–1071 (2011).
23. Lewin, J. & Ashworth, P. J. The negative relief of large river floodplains. Earth
Sci. Rev. 129, 1–23 (2014).
24. Constantine, J. A., Dunne, T., Piégay, H. & Kondolf, M. Controls on the
alluviation of oxbow lakes by bed-material load along the Sacramento River,
California. Sedimentology 57, 389–407 (2010).
25. Constantine, J. A. & Dunne, T. Meander cutoff and the controls on the
production of oxbow lakes. Geology 36, 23–26 (2008).
26. Mertes, L. A. K., Dunne, T. & Martinelli, L. A. Channel-floodplain
geomorphology along the Solimões-Amazon River, Brazil. Geol. Soc. Am. Bull.
108, 1089–1107 (1996).
27. Lauer, J. W. & Parker, G. Net local removal of floodplain sediment by river
migration. Geomorphology 96, 123–149 (2008).
28. Aalto, R., Lauer, J. W. & Dietrich, W. E. Spatial and temporal dynamics of
sediment accumulation and exchange along Strickland River floodplains
(Papua New Guinea) over decadal-to-centennial timescales. J. Geophys. Res.
113, http://dx.doi.org/10.1029/2006JF000627 (2008).
29. Räsänen, M., Neller, R., Salo, J. & Jungner, H. Recent and ancient fluvial
deposition systems in the Amazonian foreland basin, Peru. Geol. Mag. 129,
293–306 (1992).
30. Miller, J. et al. Effects of the 1997 flood on the transport and storage of
sediment and mercury within the Carson River Valley, West-Central Nevada.
J. Geol. 107, 313–327 (1999).
Acknowledgements
The study was supported by NERC grant NE/I002081/1, NASA grant NAG5-6120, the
Don J. Easterbrook Award of the Geological Society of America, and by the National
Research Foundation Singapore and the Singapore Ministry of Education under the
Research Centres of Excellence initiative (EOS contribution 78). The UK Leverhulme
Trust, the US Fulbright Commission, and the Earth Observatory of Singapore also
supported T. Dunne. We thank N. Morales, J. Teng and C. Constantine for assistance in
data collection and for discussions relating to hypothesis development.
Author contributions
J.A.C. and T.D. conceived the study. J.A. conducted calculations of channel change, and
T.D., J.A.C. and J.A. compiled data on sediment loads. C.L. and E.D.L. provided technical
expertise in analysing channel change from remote sensing data. J.A.C. and T.D. led data
analysis and interpretation, assisted by all co-authors. J.A.C. drafted the paper, which was
then reviewed by all co-authors.
Additional information
Supplementary information is available in the online version of the paper. Reprints and
permissions information is available online at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to J.A.C. or T.D.
Competing financial interests
The authors declare no competing financial interests.
NATURE GEOSCIENCE | VOL 7 | DECEMBER 2014 | www.nature.com/naturegeoscience 903
© 2014 Macmillan Publishers Limited. All rights reserved.
In the version of this Letter originally published online, the contributions of the author Eli D. Lazarus were missing, they should have
been included in the following sentence: ‘C.L. and E.D.L. provided technical expertise in analysing channel change from remote sensing
data.’ is error has been corrected in all versions of the Letter.
Sediment supply as a driver of river meandering and floodplain evolution in the
Amazon Basin
José Antonio Constantine, Thomas Dunne, Joshua Ahmed, Carl Legleiter and Eli D. Lazarus
Nature Geoscience http://dx.doi.org/10.1038/ngeo2282 (2014); published online 2 November 2014; corrected online 7 November 2014.
CORRIGENDUM
© 2014 Macmillan Publishers Limited. All rights reserved
