Millennial lag times in the Himalayan sediment routing system

Abstract

Any understanding of sediment routing from mountain belts to their forelands and offshore sinks remains
incomplete without estimates of intermediate storage that decisively buffers sediment yields from erosion
rates, attenuates water and sediment fluxes, and protects underlying bedrock from incision. We quantify
for the first time the sediment stored in > 38 000 mainly postglacial Himalayan valley fills, based on
an empirical volume-area scaling of valley-fill outlines automatically extracted from digital topographic
data. The estimated total volume of 690(+ 452/− 242) km³ is mostly contained in few large valley fills > 1 km³, while catastrophic mass wasting adds another 177( ± 31) km³. Sediment storage volumes are
highly disparate along the strike of the orogen. Much of the Himalaya’s stock of sediment is sequestered
in glacially scoured valleys that provide accommodation space for ~ 44% of the total volume upstream of
the rapidly exhuming and incising syntaxes. Conversely, the step-like long-wave topography of the central
Himalayas limits glacier extent, and thus any significant glacier-derived storage of sediment away from
tectonic basins. We show that exclusive removal of Himalayan valley fills could nourish contemporary
sediment flux from the Indus and Brahmaputra basins for > 1 kyr, though individual fills may attain
residence times of > 100 kyr. These millennial lag times in the Himalayan sediment routing system may
sufficiently buffer signals of short-term seismic as well as climatic disturbances, thus complicating simple
correlation and interpretation of sedimentary archives from the Himalayan orogen, its foreland, and its
submarine fan systems.

Full text

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights

Author's personal copy
Earth and Planetary Science Letters 382 (2013) 38–46
Contents lists available at ScienceDirect
Earth and Planetary Science Letters
www.elsevier.com/locate/epsl
Millennial lag times in the Himalayan sediment routing system
Jan Henrik Blöthe ∗,OliverKorup
Institut für Erd- und Umweltwissenschaften, Universität Potsdam, Karl-Liebknecht-Str. 24, 14476 Potsdam, Germany
article info abstract
Article history:
Received 6 June 2013
Received in revised form 20 August 2013
Accepted 23 August 2013
Available online 26 September 2013
Editor: T.M. Harrison
Keywords:
sediment storage
Himalayas
sediment budget
tectonic geomorphology
geomorphometry
Any understanding of sediment routing from mountain belts to their forelands and offshore sinks remains
incomplete without estimates of intermediate storage that decisively buffers sediment yields from erosion
rates, attenuates water and sediment fluxes, and protects underlying bedrock from incision. We quantify
for the first time the sediment stored in >38000 mainly postglacial Himalayan valley fills, based on
an empirical volume-area scaling of valley-fill outlines automatically extracted from digital topographic
data. The estimated total volume of 690(+452 /−242 )km3is mostly contained in few large valley fills
>1km
3, while catastrophic mass wasting adds another 177(±31)km3. Sediment storage volumes are
highly disparate along the strike of the orogen. Much of the Himalaya’s stock of sediment is sequestered
in glacially scoured valleys that provide accommodation space for ∼44% of the total volume upstream of
the rapidly exhuming and incising syntaxes. Conversely, the step-like long-wave topography of the central
Himalayas limits glacier extent, and thus any significant glacier-derived storage of sediment away from
tectonic basins. We show that exclusive removal of Himalayan valley fills could nourish contemporary
sediment flux from the Indus and Brahmaputra basins for >1 kyr, though individual fills may attain
residence times of >100 kyr. These millennial lag times in the Himalayan sediment routing system may
sufficiently buffer signals of short-term seismic as well as climatic disturbances, thus complicating simple
correlation and interpretation of sedimentary archives from the Himalayan orogen, its foreland, and its
submarine fan systems.
©2013 Elsevier B.V. All rights reserved.
1. Introduction
The Indus and Ganges–Brahmaputra Rivers rank amongst
Earth’s largest river systems, and drain the Himalayas, one of
the planet’s premier mountain belts, featuring active tectonic
shortening, extreme relief, highly seasonal precipitation, and com-
mensurate erosion rates. Sediments flushed from the orogen are
deposited in the foreland basin of the Indo-Gangetic Plain, and, ul-
timately, in the Indus and Bengal submarine fan systems, which
have attained sediment piles >9 and >16 km thick, respectively
(Clift et al., 2001; Curray, 1994). The Ganges–Brahmaputra sys-
tem delivers by far the largest amount of terrestrial sediment to
the ocean, at an annual flux ∼103Mt yr−1(e.g. Curray, 1994;
Goodbred and Kuehl, 2000; Milliman and Meade, 1983). Be-
sides analytical errors associated with measurement procedures,
large uncertainties in these estimates (Table A.1) derive from elu-
sive data on the build-up and removal of intermediate sediment
storage. This critical term in the sediment budget is potentially
governed by stochastic internal system dynamics that introduce
significant variability to short-term measurements of sediment
flux, likely to be amplified by the reworking of stored sedi-
*Corresponding author. Tel.: +49 331 9776272.
E-mail address: jan.bloethe@geo.uni-potsdam.de (J.H. Blöthe).
ments (Jerolmack and Paola, 2010; Simpson and Castelltort, 2012;
Van de Wiel and Coulthard, 2010). Particularly intermontane
valley fills such as floodplains, fans, and terraces, are impor-
tant landforms, decoupling hillslopes from river-channel processes
and buffering sediment sources from sinks (Castelltort and Van
Den Driessche, 2003; Fryirs et al., 2007; Straumann and Korup,
2009); sequestering biogeochemical constituents including nutri-
ents and pathogens alike; containing archives of environmental
change; modulating natural hazards by either attenuating or am-
plifying water-sediment fluxes as well as seismic shear velocities
(Wald and Allen, 2007); and ultimately providing the amenity of
flat ground for tens of millions of people and their agricultural
livelihood in otherwise steep mountainous terrain.
Storage is fundamental to any sediment budget, but remains
ablack box for many large drainage basins, spawning large uncer-
tainties about reported sediment yields and potentially introducing
long-term stability of sediment yields by buffering signals of envi-
ronmental change (e.g. Allen, 2008; Métivier and Gaudemer, 1999;
Milliman and Syvitski, 1992; Phillips, 2003). Distinct research gaps
concern the spatial distribution, residence times, and resulting lag
times between rates of erosion and sediment yields that only a
quantification of sediment storage can elucidate (Castelltort and
Van Den Driessche, 2003; Hinderer, 2012). Until recently, system-
atic analyses and quantification of sediment storage focused on
smaller drainage basins or individual landforms (e.g. Schrott et al.,
0012-821X/$ – see front matter ©2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.epsl.2013.08.044

Author's personal copy
J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46 39
Fig. 1. Study area of the Himalayas and adjacent areas. (a) Topography with rivers, lakes, and contemporary glacier cover. Black triangles are major peaks: NP=Nanga Parbat;
ND =Nanda Devi; AP =Annapurna; ME =Mount Everest; NB =Namche Barwa. Labels indicate rivers referred to in text and tables: Chi =Chitral; Ind =Indus; Gil =
Gilgit; Che =Chenab; Hun =Hunza; Bra =Braldu; Sut =Sutlej; Nub =Nubra; Shy =Shyok; Kar =Karnali; Nar =Narayani; Kos =Kosi; Yig =Yigong Tsangpo; Sia
=Siang; Par =Parlung Tsangpo. (b) Mean annual precipitation from APHRODITE dataset (Yatagai et al., 2009) with major contour lines. (c) Mean local relief, expressed as
maximum elevation difference in 10-km radius on SRTM90 data. (d) Long-wave topographic gradient (LWT), calculated from mean elevation in a 100-km radius based on
SRTM data resampled to 270-m resolution. Black dashed lines are major tectonic lineaments: KF =Karakorum Fault, ITSZ =Indus-Tsangpo Suture Zone, STDZ =Southern
Tibetan Detachment Zone. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
2003). Efforts to integrate up to the mountain-belt scale (Hinderer,
2001; Straumann and Korup, 2009; Wasson, 2003), as well as to
quantify sediment budgets on million-year timescales (Métivier
and Gaudemer, 1999), have been rare. Yet estimates of the sed-
iment storage in the vast floodplains of the Brahmaputra River
(Allison et al., 1998; Goodbred and Kuehl, 1998) have underscored
the unique opportunity to contributing regional jigsaw pieces to
completing our understanding of Earth’s largest sediment routing
system.
Here we estimate Himalayan sediment storage by extracting
and analyzing the size, regional distribution, and minimum life
span of intermontane valley fills from digital topography. We eval-
uate their pattern with respect to the variability of litho-tectonic
units, local and long-wave topographic relief as proxies of ero-
sion rates (e.g. Montgomery and Brandon, 2002), precipitation
patterns, glacier cover, and river-channel steepness along the en-
tire Himalayan orogen and its adjacent ranges over an area of
∼438 780 km2(Figs. 1 and 2a). We automatically extracted the
outlines of major valley fills along the Himalayan arc from a digital
elevation model (DEM), and used an empirical volume-area scaling
relationship with Monte Carlo-based error propagation to conser-
vatively estimate the minimum volume contained in >38 000 val-
ley fills.
2. Study area
Our study area encompasses the entire Himalayan orogen as
defined by Yin (2006) together with the southernmost parts of the
Karakorum, the Gangdese Shan, and those parts of the Tibetan
Plateau that are drained by the Indus and Ganges–Brahmaputra
river systems. We simplistically refer to this area (∼995 000 km2)
as the Himalayas (Figs. 1a and 2a). We distinguish between three
major hydrological compartments, i.e. the Western, Central, and
Eastern Himalayas, which are drained by the Indus, Ganges, and
Brahmaputra River systems, respectively. The elevation in the study
area rises from <500 m to >8000 m asl within a 250–500 km
horizontal distance. This pronounced topographic gradient be-
tween the Greater Himalayas and the Trans-Himalaya is steepest
in the Central Himalaya, and coincides with a sharp precipitation
gradient, although precipitation is by no means uniform along the
strike of the orogen (Bookhagen and Burbank 2010, 2006)(Fig. 1b).
Mean annual rainfall is dominated by the South Asian summer
monsoon (SASM), whereas the influence of the westerlies circu-
lation, mainly bringing winter precipitation, decreases towards the
East (Bookhagen and Burbank, 2010). Oscillations in SASM inten-
sity have been reported on various timescales, though the overall
regional climatic pattern appears to have remained largely un-
changed since the Early Miocene (Clift et al., 2008). Mean local
relief, computed as the maximum elevation range in a 10-km ra-
dius, exceeds 3000 m in the Central Himalayas, the Karakorum,
and the Nyainqentanglha mountains; it is highest at the core of
the Himalayan syntaxes (e.g. Korup et al., 2010)(Fig. 1c). The sharp
break in topography in the vicinity of the Main Central Thrust
(MCT) (e.g. Wobus et al., 2003)iswellcapturedbythelongwave-
length topographic gradient (LWT) that we calculated from mean
elevation in a 100-km radius based on DEM data that we resam-
pled to a 270-m grid-cell resolution (Fig. 1d).
3. Methods
3.1. Digital topography
We analyzed digital topographic data from the SRTM90 DEM
with gaps filled by topographic map data (www.
viewfinderpanoramas.org,srtm.csi.cgiar.org). Hydrologic correction
was done using a Matlab TopoToolbox carving routine (Schwanghart
and Kuhn, 2010), followed by a fill calculation using the ArcMap
Spatial Analyst fill algorithm; DEM tiles were merged for the entire
Indus and Ganges–Brahmaputra drainage networks, excluding ar-
eas below a smoothed 500-m contour line in order to restrict our
analyses to the mountain range.

Author's personal copy
40 J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46
Fig. 2. Extracted valley fills along the Himalayan arc. (a) Blue polygons are contributing to volumetric estimate, areas in red were excluded from scaling. Yellow triangles are
locations of volumetric estimates from published data (Dill et al., 2001; Dimri et al., 1983; Fort, 2010, 1987; Fort et al., 2010; Hewitt et al., 2011; Montgomery et al., 2004;
Pandey and Kazama, 2010); KB =Kashmir Basin, TK =Tso Kar, ZB =Zhada Basin, TG =Thakkhola Graben, PB =Pokhara, KA =Kathmandu Basin. (b) Left panel shows
exemplary swath profile of the central Himalayas. Thick line is mean elevation, polygon outlines are minimum and maximum elevation. Color-coding indicates weighting
factor for volume estimation of individual valley fills based on fuzzy membership rule, shown in right panel. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
3.2. Delineating valley fills
A number of algorithms, involving differing levels of complex-
ity and spatial resolution, have been designed for automatically
delineating and extracting from DEMs landforms or landform el-
ements such as terraces (Bowles and Cowgill, 2012; Demoulin et
al., 2007), valley bottoms (Gallant and Dowling, 2003; Straumann
and Korup, 2009; Williams et al., 2000), or general landform clas-
sification (Dr˘
agu¸t and Blaschke, 2006; Giles and Franklin, 1998;
Klingseisen et al., 2008). Despite this diversity of approaches, com-
prehensive assessments of such algorithms are rare. We compare
three approaches of unsupervised extraction of valley fills from
the DEM in order to gauge both variability and reliability of dif-
ferent techniques at the regional scale. We selected three differ-
ent algorithms, i.e. (A) the Region Growing Algorithm (RGA) (Graff
and Usery, 1993; Straumann and Korup, 2009); (B) the Multi-
Resolution Valley Bottom Flatness index (MRVBF) (Gallant and
Dowling, 2003); and (C) the Surface Classification Model (SCM)
(Bowles and Cowgill, 2012), to delineate from DEM data three
sets of abundances and areas of Himalayan valley fills (Table A.2).
We defined valley fills as either flat or gently sloping valley bot-
toms, elevated terrace surfaces, or alluvial fans covering a mini-
mum area of 40 500 m 2(i.e. five SRTM pixels), but excluded sedi-
ment storage on hillslopes (colluvium), in talus, debris-flow cones,
scree slopes, and mass-wasting deposits (which we estimated sep-
arately from a dataset of >200 Himalayan landslides; Korup et al.,
2007) from our algorithm-based extraction analyses. We validated
these automatically delineated valley fills with n=34 valley-fill
polygons that were mapped manually from SRTM90 and optical
data (ETM+) by an independent trained operator. Results between
DEM-derived and independently mapped areas are consistent, and
in highest agreement for the SCM algorithm, which we therefore
used for all further analyses (R2for MRVBF: 0.92; SCM: 0.94; RGA:
0.94).
3.3. Estimating accommodation space
In order to quantify the maximum accommodation volume for
sediment for a given valley-fill planform area and valley topogra-
phy, we generated hypothetical valley fills from DEM manipulation
in order to obtain a robust empirical volume-area scaling relation-
ship for our study area. This approach works on the assumption
that the bedrock topography beneath existing valley fills does not
differ significantly from the dissected topography elsewhere in the
Himalayas. To cover as many topographic bedrock conditions as
possible, we selected n=3687 randomly distributed points along
the Himalayan drainage network, corrected for the skewed size
distribution of contributing catchment area. At each point we ma-
nipulated the DEM such that we emplaced a uniform dam of con-
stant height normal to the local drainage direction. We found that
the size distribution of natural dam height can be approximated
by a log-normal model, based on n=240 world-wide cases. In
order to avoid spill-over effects that lead to significant overes-
timates of accommodation spaces we randomly sampled from a
relief-corrected log-normal distribution with μ=log10(hrel), and
σ=1 [a.u.]; where hrel ishalfthelocalreliefasmeasuredina
10-km radius. The hypothetical backwater accommodation space
was filled using the ArcMap Spatial Analyst fill algorithm; the
resulting volume [m3],andplanformarea[m
2] were computed
by a cut-and-fill routine. We further corrected for the displace-
ment effects of these hypothetic dams (Kuo et al., 2011), given
a number of assumptions regarding the dam type and its geom-
etry. Finally, our artificial valley-fill database was corrected for
glaciers and broad alluviated valleys. These were excluded be-
cause bedrock topography submerged below ice and sediment
would lead to an underestimation of storage potential per unit
area. We used a digital glacier inventory to mask out any glaciers
from our volumetric estimates, before analyzing the distribution of
valley-fill volumes with respect to contemporary glacier cover. The

Author's personal copy
J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46 41
glacier outlines used for this analyses were mainly taken from the
GLIMS database (www.glims.org), based on the Chinese and ICI-
MOD inventory (Mool et al., 2001; Yafeng et al., 2010), and other
regional sources for the northwest Himalaya (Frey et al., 2012;
Bhambri et al., 2012). Our artificial valley fills cover the full range
of major lithologic units (Fig. A.1), local channel slopes, and dam
heights, i.e. factors that potentially exert a first order control the
storage capacity behind natural dams.
We used quantile regression (Koenker and Hallock, 2001)tode-
rive scaling relationships between volume and area of our artificial
valley-fill dataset for quantiles in 5%-steps that can be expressed
by a power-law of the form V=bAs, where bis the intercept
[m3–2s], sis the scaling exponent, Ais area [m2], and Vis vol-
ume [m3]. Quantile regression allows for a more complete picture
of the entire data distribution, as multiple solutions are estimated
for different proportions of the data (Cade and Noon, 2003), and
helps detect size-dependent differences in the volume-area scal-
ing, given that slight differences in scaling exponents may lead
to large deviations in volumetric estimates (Larsen et al., 2010).
The range among all exponents sfor the 5th to 95th percentile
is from 1.23 to 1.44. We compiled data on 14 valley fills reported
fromtheliterature(Figs. 2a and 3a) in order to cross-check the
range of our predictions, and validate our scaling relationship (Dill
et al., 2001; Dimri et al., 1983; Fort 2010, 1987; Fort et al., 2010;
Hewitt et al., 2011; Montgomery et al., 2004; Pandey and Kazama,
2010). We find that median scaling exponents derived for our data
(s=1.29±0.01) are not significantly different from median scaling
exponents obtained from the published data (s=1.30 ±0.10).
3.4. Fuzzy membership
Our simplistic assumption of v-oru-shaped cross-sectional
bedrock valleys beneath Himalayan sediment fills might not hold
true for large foreland and Subhimalayan piggy-back basins as well
as tectonic graben systems along the southern Tibetan Plateau
margin, i.e. at the southern and northern fringes of our study area.
In order to prevent potential overestimates of sediment volumes
from these areas, we used a fuzzy classification rule to assess
the degree of membership of each individual valley fill to a fuzzy
set (Zadeh, 1965) comprising large basins, foreland-basin fills, and
plateau surfaces that we wished to exclude from our study. Fuzzy
classification assesses an object’s degree of membership instead of
using a binary membership, thus providing a measure of member-
ship uncertainty. Plateaus are high-elevation-low-relief areas, while
intermontane basins are characterized by low relief. Thus we com-
bined published threshold values on elevation and relief used to
define the Tibetan Plateau (Fielding et al., 1994; van der Beek et
al., 2009). From these, we built two sigmoidal membership func-
tions, one assessing the degree of membership to a fuzzy set based
on elevation, the other based on local relief:
εhigh elevation(xe)=1
1+e(−xe−me
k)(1)
εlow relief (xr)=1−1
1+e(−xr−mr
k)∗
(2)
where εis the degree of membership to the fuzzy set, xeis
elevation, xris local relief, and k=250 m, me=4500 m, and
mr=1750 m, are constants determining the shape of the function.
Final membership was calculated using a non-interactive union of
the results of Eq. (1) and Eq. (2):
εfinal =maxεhigh elevation(xe), εlow relief (xr)(3)
We used the inverse fuzzy membership degree as a weighting
factor for the areas obtained from the SCM algorithm. We used
Fig. 3. Scaling method to estimate volumes of Himalayan valley fills. (a) Empirical
volume-area scaling derived from storage capacity assessment of digital topogra-
phy (blue dots), and published volumetric estimates (yellow triangles, see Fig. 2
for refs.). Lines are median quantile regression fit to published (red dashed line;
log b=−0.37 ±0.78 with units [V(3–2s)]) and modeled data (black solid line;
log b=−0.39 ±0.04 with units [V(3–2s)]); scaling exponents given in rectangles
(±bootstrapped standard errors). Black dot-dashed lines are 5th and 95th Per-
centile regression for modeled data. (b) Probability density estimate (blue) and
cumulative distribution function (red) of >38000 valley-fill volumes. Red bubbles
are 20 largest valley fills, storing >50% of the total volume; bubble size scaled to
fraction of total volume. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
εfinal =0.35 as a cut-off value, and set all valley fills having a lower
degree of membership to null, resulting in a contributing study
area of ∼439 000 km2(Figs. 2a, b). This was necessary as some ex-
tensive storage areas on the Tibetan Plateau otherwise would yield
overestimated volumes despite their low weight factor.
3.5. Application of volume-area scaling
We applied the volume-area scaling relationships to extrapo-
late the volumetric budgets of low-gradient valley fills (Straumann
and Korup, 2009), and landslides (Larsen et al., 2010)inorderto
augment the few available data with known volume and area. For
the Himalayas, published volumetric data of sediment storage for
calibrating this relationship is sparse, and underlying bedrock to-
pography is largely unknown. Using the fuzzy classification rule

Author's personal copy
42 J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46
Fig. 4. Size and distribution of Himalayan valley fills. (a) Estimated volumes of individual fills (n=38 197; open circles color coded by volume), inferred mean valley-fill
thickness (volumes normalized by valley-fill area), minimum residence times assuming mean aggradation and removal rates of 1 mm yr−1. (b) Catchment-wide estimates of
total sediment storage volume per unit study area; absolute sediment volumes coded to numbers in rectangles. (c) Location of study area. (For interpretation of the references
to color in this figure legend, the reader is referred to the web version of this article.)
and the valley-fill outlines derived from the SCM algorithm, we
obtained n=38 197 valley fills, covering an area of 18 248 km2
(Figs. 2a, b). We quantified error propagation by repeating volume
calculations for each polygon n=1000 times with random sam-
ples of normally distributed values of band s, applying the inverse
degree of membership as a weight for the individual contribution
of each storage unit.
4. Results
We estimate that 90% of the total volumes of Himalayan val-
ley fills are between 448 and 1142 km3, with a median volume of
∼690 km3. Our estimates of Himalayan valley-fill abundance vary
by up to 30% with algorithm type, with valley fills covering 11–16%
of the total study area (Fig. 2a and Table A.2). Our volumetric
scaling of valley-fill volumes with area also varies significantly
between most of the major litho-tectonic units, with power-law
scaling exponents s=1.27–1.35. Median scaling exponents are sta-
tistically different for more than half of the different lithologies,
likely reflecting a rock mass-dependent susceptibility to erosion
and sediment storage (Fig. A.1). Quantile regression yields higher
scaling exponents for upper percentiles (Fig. 3a), indicating that
larger valley fills store disproportionately more volume per unit
area than smaller ones. Our volume-area scaling exponents from
a global median regression s=1.29 ±0.01 (bootstrapped standard
errors) approximate previously reported values from the European
Alps (Straumann and Korup, 2009). The volumetric scaling for Hi-
malayan valley fills is largely consistent over nearly four orders of
magnitude with estimates obtained from published data (Fig. 3a).
Our volumetric estimates are clearly conservative, as we excluded
from our calculations most of the large structural sedimentary
basins such as the Thakkhola Graben or Zhada Basin along the
southern Tibetan Plateau margin, together with those in the Hi-
malayan foreland (Kashmir basin and dun-type piggy-back basins)
that host substantial amounts of pre-Quaternary fills (Fig. 2a and
Table A.3).
The overall spatial pattern of sediment volumes stored per
unit study area shows a distinct regional tri-partitioning, with
most sediment sequestered in the Western and Eastern Hi-
malayas, i.e. 381(+300/−151)km3, and 214(+138/−81)km3,respec-
tively (Figs. 4, 5). More than half of the Himalaya’s total sedi-
ment volume is stored in the 20 largest valley fills (Fig. 3b) that
primarily straddle major tectonic structures such as the Indus-
Tsangpo Suture Zone, the Southern Tibetan Detachment Zone or
the Karakorum Fault (Figs. 1 and 4), highlighting the contribution
of crustal deformation to creating preferential pathways for fluvial
and glacial erosion. Large (>1km
3) fills also cluster in the West-
ern and Eastern Himalayas, especially upstream of the syntaxes,
where 86% and 93% of the total volume contained in the Indus
and Brahmaputra catchments are stored, respectively (Fig. 6). In
contrast, the Central Himalaya is strikingly devoid of large valley
fills, except for the lowermost parts of orogen-normal graben sys-
tems in the Karnali, Narayani, and Kosi basins.
5. Discussion
5.1. Spatial pattern of Himalayan valley fills
Clusters of large valley fills occur in regions with extensive con-
temporary glacier cover and Quaternary glaciation history, such
as the Karakorum and Nyainqentanglha mountains (Owen et al.,
2008). The Himalayan and Karakorum mountains are the most

Author's personal copy
J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46 43
Fig. 5. Sediment storage [km3]andfluxes[km
3yr−1] in the Himalayan sediment routing system. Blue arrows thickness scaled to sediment flux; labels denote range of pub-
lished data converted to [km3yr−1] (Table A.1). Grey lines are rivers without sediment-flux data; blue lines show Himalayan high-order drainage network. Major hydrological
basins delineated by red, green, and orange catchment boundaries; estimated sediment storage errors encompass 90% of simulated valley-fill volumes. Floodplain storage
(FP?) remains largely unquantified. River courses and submarine fans are not to scale. Brown triangles indicate submarine fans of Indus and Ganges–Brahmaputra systems
with total volumes estimated. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 6. Disproportionate sediment storage upstream of Himalayan syntaxes. Cumulative valley-fill volumes along the Indus (red line and circles), and Yarlung Tsangpo/Siang
(blue line and circles) catchments. Shaded area is position of Himalayan syntaxes. Few large valley fills dominate the sharp increase in and total of catchment-wide sediment
volumes upstream of both syntaxes. Mountain front refers to lowest pour points of our study area. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
heavily glaciated regions on Earth outside the polar regions (Qiu,
2008). Glacial extent was larger during Pleistocene glaciations,
forming high local relief and scouring broad u-shaped valleys pro-
viding ample accommodation space for postglacial sediment fills
(Bolch et al., 2012; Owen et al., 2008). Especially upstream of
the Himalayan syntaxes, where river steepness decreases substan-
tially (Korup et al., 2010), major valley floors are as wide as
10 km. The volume of large valley fills increases with contem-
porary glacier cover such that we infer that these sediment fills
are largely glacigenic (Fig. 7a). Glacial widening and overdeepen-
ing of valleys has contributed to enlarging accommodation space
along structurally controlled valleys occupied by, among others, the
Shyok, Nubra, Braldu, Gilgit, Indus, Zanskar, Chitral, and Parlung
Rivers in the Western and Eastern Himalayas. However, the Central
Himalayas, though heavily glaciated, are comparatively sediment-
starved. Glaciers in the Central Himalayas, though abundant, are
steep and limited in their length by the sharp Tibetan Plateau
margin that, accentuated by aggressive fluvial incision, constrains
glacial provision of accommodation space to high elevations. This
distinct step-like gradient in long-wave topography (LWT) explains
the polarity in the distribution of large valley fills along the orogen
(Figs. 1d, 7d). In the Central Himalayas, where LWT rises sharply
near the Main Central Thrust, only 14% of the Himalaya’s total
sediment volume is sequestered in valley fills. The Eastern and
Western Himalayas have a less conspicuous LWT, although feature
the highest local topographic relief around the syntaxes (e.g. Korup
et al., 2010; Zeitler et al., 2001).
Superimposed on this first-order topographic constraint on
glacier-induced sediment storage is a significant decline of individ-
ual valley-fill volumes with mean annual precipitation, supporting
the intuitive notion of lower sediment storage potential in areas
of higher erosion (Fig. 7b). This is further augmented by a strong
decline in total volume with increasing mean local relief, whereas
individual valley-fill volumes increase (Fig. 7c). Moreover, the spa-
tial abundance of Himalayan valley-fills along the strike of the
orogen appears to be inversely correlated with that of long-term
crustal exhumation and denudation rates (Burbank et al., 2003;
Finnegan et al., 2008; Grujic et al., 2006; Thiede and Ehlers, 2013;
Thiede et al., 2004; Zeitler, 1985).

Author's personal copy
44 J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46
Fig. 7. Estimated valley-fill volumes versus (a) Fraction of glaciers measured in a
25-km radius; (b) Mean annual precipitation from APHRODITE data (Yata gai et al .,
2009); (c) Mean local relief from SRTM90 measured in 10-km radius; and (d) Long
wavelength topographic (LWT) gradient. Grey circles are estimated volumes for in-
dividual valley fills, blue dotted lines are total volumes per bin (bin widths: (a) 0.1,
(b) 0.5 m, (c) 0.5 km, and (d) 1%). Lines are solutions for linear quantile regression:
solid for median; large dashed for 75th; small dashed for 90th; and dotted for 99th
percentile. Red lines have slopes that are significantly different from zero (99% con-
fidence interval); slopes of grey lines are not significantly different from zero. For
map view of different influencing factors presented here see Fig. 1. (For interpre-
tation of the references to color in this figure legend, the reader is referred to the
web version of this article.)
While our method has captured low-gradient fluvial and allu-
vial valley fills, we estimate the additional volumetric contribution
of more hummocky and irregular catastrophic mass-wasting de-
posits to valley filling to be at least 177(±31)km3for the whole
mountain range (Korup et al., 2007, 2010). Nearly 87% of this vol-
ume is stored in only 57 landslide deposits containing >1km
3
each, which proliferate in the Indus Basin, mostly upstream of the
western Himalayan syntaxis (Korup et al., 2010). The overall obser-
vation that ∼44% of the mountain belt’s sediments are trapped
upstream of the Himalayan syntaxes supports previous specula-
tions about the role of rapid tectonic uplift (Zeitler et al., 2001)in
spatially distorting the orogen’s sediment budget in favor of head-
water reaches of major rivers. Here, large volumes of sediments
are retained given the relatively shallow river gradients in concert
with pronounced aridity (Korup et al., 2010).
5.2. Estimating mean residence times of valley fills
Our volumetric estimates of Himalayan valley fills elucidate not
only the spatial distribution, but also allow assessing the resi-
dence times, of intermontane sediment storage in the Indus and
Ganges–Brahmaputra catchments. We estimated the regional mean
residence time of Himalayan valley fills in a three-fold manner.
First, we computed the mean thickness of each individual valley
fill. We then simplistically assumed, based on the literature on Hi-
malayan erosion rates, mean aggradation rates of 0.5 to 5 mm yr−1
to form, and the same mean erosion rates to fully remove, this
fill (Figs. 4a, 8b). Second, we assessed how long it would take to
build up the total volume of 690 km3we quantified in our analyses
for trapping efficiencies (Et) between 0.5% and 100%, i.e. the frac-
tion of density-corrected sediment retained in a given mountain
Fig. 8. Cumulative distribution functions of proxies of sediment-storage longevity.
(a) Cumulative distribution function of Himalayan valley fill dates compiled from
the literature. (b) Residence time of Himalayan valley fills (this study) estimated
for different mean aggradation and removal rates. Dashed vertical lines are medians
of the distributions; colored rectangles are interquartile range from 25th to 75th
percentile. (c) Trapping efficiency versus estimated time needed to build up total
valley-fill volume of ∼690 km3for varying denudation rates, given a total area of
A=994 656 km2,arockdensityofρr=2.6tm
−3and a bulk density of sedimen-
tary fills, ρs=1.8tm
−3. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
river reach (Fig. 8c). Third, we divided our estimates of Himalayan
valley-fill volumes for major drainage basins (Fig. 4b) by published
estimates of sediment flux from the orogen (Fig. 5 and Table A.1)
in order to arrive at the average time span that sole erosion of val-
ley fills would be able to completely nourish sediment export from
the mountain belt with no additional contribution from hillslope,
glacial, or fluvial bedrock erosion.
Estimates of mean denudation rates across the Himalayas based
on multiple independent proxies mostly fit within the range of
1–3 mm yr−1, depending on catchment location, method, and in-
terpolation timescale (Finnegan et al., 2008; Galy and France-
Lanord, 2001; Garzanti et al., 2005; Lupker et al., 2012; Vance
et al., 2003). Thus our mean rate estimates cover most of the
spectrum of documented rates (Fig. 8). Mean regional denudation
rates derived from cosmogenic nuclides of 1–1.5mmyr
−1(Galy
and France-Lanord, 2001; Lupker et al., 2012) are consistent with
sediment yields estimated from sedimentary archives of the In-
dus and Bengal submarine fans (∼106Mtyr−1)on10
2to 107yr
timescales (Table A.1). Sole erosion of Himalayan valley fills ex-
cluding any input from bedrock erosion could sustain these rates
for >1000 yr in the Indus and Brahmaputra, and >300 yr in the
Ganges drainage basin; these figures would increase by at least
another ∼800 and ∼100 yr in Indus and Ganges basins, were
mass-wasting deposits to be included, respectively. These figures
highlight the substantial lag times introduced by intermediate sed-

Author's personal copy
J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46 45
iment storage in valley fills along the Himalayan arc that may
complicate interpretations and correlations of high-resolution in-
tramontane sedimentary archives with those in the foreland sinks.
Assuming that build-up and full removal of sediment storage
would occur at constant average rates representative for the Hi-
malayas, we estimate median residence times between ∼3 and
∼32 kyr for Himalayan valley fills (Fig. 8b). The size distribution
and pattern of valley fills predicts that several dozen of large valley
fills along the southern Tibetan Plateau margin and in the Subhi-
malayas may persist for >100 kyr. Moreover, 75% of all fills with
much smaller volumes would remain in storage for less than 4,
10, 20, and 41 kyr, assuming mean rates of aggradation and re-
movalof0.5,1,2,and5mmyr
−1, respectively (Fig. 8b). These
first-order estimates are consistent with the distribution of pub-
lished dates on fluvial terraces throughout the Himalayas (Fig. 8a),
which indicates that Etof Himalayan sediment fills is on average
of between 1% and 10% (Fig. 8c). This orogen-scale assessment of
sediment storage neglects any contribution from pre-Quaternary
sediments stored in large tectonic basins: The Kathmandu Basin
alone contains half of the sediment volume that we estimated for
the Central Himalayas, i.e. 95(+15/−10)km3. Larger tectonic basin
fills such as in the Zhada Basin, Kashmir Basin, or the Thakkhola
Graben are the end members of (Trans-)Himalayan sediment stor-
age that we roughly estimate to have trapped >2000 km3of sed-
iment since Miocene times (Table A.3). The volumes of such long-
lived basin fills rival our volumetric estimates of much younger
sediments, and contribute to emphasizing the substantial lag times
that possibly buffer signals of climatic disturbances in the Hi-
malayan sediment routing system.
6. Conclusions
In conclusion, we present the first comprehensive estimate
of millennial-scale sediment storage in large valley fills for the
entire Himalayas. Future work on the orogen’s sediment budget
may refine the volumetric accuracy, though the implications of
the spatial distribution of valley fills remain. The striking prolif-
eration of large valley fills upstream of the Himalayan syntaxes
(Fig. 6) stands out against the relatively sediment-starved Central
Himalaya, where a distinct step in long-wave topography limits
significant glacier extent and postglacial sediment accumulation.
We further find that 15–25% of Himalayan valley fills are tied
to a catastrophic mass-wasting origin. While glacial overdeepen-
ing and widening along major tectonic fault zones provide the
largest accommodation spaces, this partly tectonics-, partly mass-
wasting driven trapping of mainly postglacial sediment dampens
fluvial bedrock incision via a pronounced cover effect (Lague, 2010;
Sklar and Dietrich, 2001). Moreover, the spatially disjunct pattern
of Himalayan sediment storage with larger and older fills pref-
erentially located along the southern Tibetan Plateau margin is
quantitatively corroborated by a declining trend of an old refrac-
tory organic carbon component released from such storage, and
measurable in bulk ages of river samples during biospheric carbon
export from the Central Himalayas (Galy and Eglinton, 2011). Fi-
nally, the millennial residence times of sediment storage in the
orogen reverberate on the Himalayan sediment routing system,
as they warrant sufficiently long lag times that may complicate
the interpretation of offshore sedimentary archives, their reliable
correlation with coeval terrestrial counterparts, and any resulting
attribution of shorter-term seismic or climatic disturbance signals
of the Himalayan mass balance.
Acknowledgements
Funded by the German Research Foundation (DFG Grants
KO3937/1 and 2), and the Potsdam Research Cluster for Georisk
Analysis, Environmental Change and Sustainability (PROGRESS). We
computed statistics using SAGA-GIS (www.saga-gis.org), and the R
software environment (www.r-project.org).
Appendix A. Supplementary material
Supplementary material related to this article can be found on-
line at http://dx.doi.org/10.1016/j.epsl.2013.08.044.
References
Allen, P.A., 2008. Time scales of tectonic landscapes and their sediment routing sys-
tems. Geol. Soc. Spec. Publ. 296, 7–28.
Allison, M.A., Kuehl, S.A., Martin, T.C., Hassan, A., 1998. Importance of flood-plain
sedimentation for river sediment budgets and terrigenous input to the oceans:
Insights from the Brahmaputra–Jamuna River. Geology 26, 175–178.
Bhambri,R.,Bolch,T.,Kawishwar,P.,Dobhal,D.P.,Srivastava,D.,Pratap,B.,2012.
Heterogeneity in Glacier response from 1973 to 2011 in the Shyok valley,
Karakoram, India. The Cryosphere Discuss. 6, 3049–3078.
Bolch, T., Kulkarni, A., Kääb, A., Huggel, C., Paul, F., Cogley, J.G., Frey, H., Kargel, J.S.,
Fujita, K., Scheel, M., Bajracharya, S.R., Stoffel, M., 2012. The state and fate of
Himalayan glaciers. Science 336, 310–314.
Bookhagen, B., Burbank, D.W., 2006. Topography, relief, and TRMM-derived rainfall
variations along the Himalaya. Geophys. Res. Lett. 33, 1–5.
Bookhagen, B., Burbank, D.W., 2010. Toward a complete Himalayan hydrological
budget: Spatiotemporal distribution of snowmelt and rainfall and their impact
on river discharge. J. Geophys. Res. 115, 1–25.
Bowles, C.J., Cowgill, E., 2012. Discovering marine terraces using airborne LiDAR
along the Mendocino-Sonoma coast, northern California. Geosphere 8, 386–402.
Burbank, D.W., Blythe, A.E., Putkonen, J.K., Pratt-Sitaula, B.A., Gabet, E.J., Oskin, M.,
Barros, A., Ojha, T.P., 2003. Decoupling of erosion and precipitation in the Hi-
malayas. Nature 426, 652–655.
Cade, B.S., Noon, B.R., 2003. A gentle introduction to quantile regression for ecolo-
gists. Front. Ecol. Environ. 1, 412–420.
Castelltort, S., Van Den Driessche, J., 2003. How plausible are high-frequency sed-
iment supply-driven cycles in the stratigraphic record?. Sediment. Geol. 157,
3–13.
Clift,P.D.,Hodges,K.V.,Heslop,D.,Hannigan,R.,VanLong,H.,Calves,G.,2008.
Correlation of Himalayan exhumation rates and Asian monsoon intensity. Nat.
Geosci. 1, 875–880.
Clift, P.D., Shimizu, N., Layne, G.D., Blusztajn, J.S., Gaedicke, C., Schlüter, H.-U., Clark,
M.K., Amjad, S., 2001. Development of the Indus Fan and its significance for
the erosional history of the Western Himalaya and Karakoram. Geol. Soc. Am.
Bull. 113, 1039–1051.
Curray, J.R., 1994. Sediment volume and mass beneath the Bay of Bengal. Earth
Planet. Sci. Lett. 125, 371–383.
Demoulin, A., Bovy, B., Rixhon, G., Cornet, Y., 2007. An automated method to extract
fluvial terraces from digital elevation models: The Vesdre valley, a case study in
eastern Belgium. Geomorphology 91, 51–64.
Dill, H., Kharel, B., Singh, V., Piya, B., Busch, K., Geyh, M., 2001. Sedimentology and
paleogeographic evolution of the intermontane Kathmandu basin, Nepal, during
the Pliocene and Quaternary. Implications for formation of deposits of economic
interest. J. Asian Earth Sci. 19, 777–804.
Dimri, D.B., Baranwal, M., Biswas, U.K., 1983. Integrated geophysical studies in
Tsokar Basin, District Ladakh, J & K, India. Geol. Surv. India 37 (2), 39–46.
Dr˘
agu¸t, L., Blaschke, T., 2006. Automated classification of landform elements using
object-based image analysis. Geomorphology 81, 330–344.
Fielding, E., Isacks, B., Barazangi, M., Duncan, C., 1994. How flat is Tibet?. Geol-
ogy 22, 163–167.
Finnegan, N.J., Hallet, B., Montgomery, D.R., Zeitler, P.K., Stone, J.O., Anders, A.M.,
Yuping, L., 2008. Coupling of rock uplift and river incision in the Namche
Barwa–Gyala Peri massif, Tibet. Geol. Soc. Am. Bull. 120, 142–155.
Fort, M., 1987. Sporadic morphogonesis in a continental subduction setting: An ex-
ample from the Anapurna Range, Nepal Himalaya. Z. Geomorph. N.F. Suppl. 63,
9–36.
Fort, M., 2010. The Pokhara valley: A product of a natural catastrophe. In: Migon,
P. (Ed.), Geomorphological Landscales of the World. Springer, Netherlands, Dor-
drecht, pp. 265–274.
Fort, M., Cossart, É., Arnaud-Fassetta, G., 2010. Hillslope-channel coupling in the
Nepal Himalayas and threat to man-made structures: The middle Kali Gandaki
valley. Geomorphology 124, 178–199.
Frey, H., Paul, F., Strozzi, T., 2012. Compilation of a glacier inventory for the western
Himalayas from satellite data: methods, challenges, and results. Remote Sens.
Environ. 124, 832–843.
Fryirs, K.A., Brierley, G.J., Preston, N.J., Kasai, M., 2007. Buffers, barriers and blankets:
The (dis)connectivity of catchment-scale sediment cascades. Catena 70, 49–67.
Gallant, J.C., Dowling, T.I., 2003. A multiresolution index of valley bottom flatness
for mapping depositional areas. Water Resour. Res. 39.

Author's personal copy
46 J.H. Blöthe, O. Korup / Earth and Planetary Science Letters 382 (2013) 38–46
Galy, V., Eglinton, T., 2011. Protracted storage of biospheric carbon in the Ganges–
Brahmaputra basin. Nat. Geosci. 4, 843–847.
Galy, A., France-Lanord, C., 2001. Higher erosion rates in the Himalaya: Geochemical
constraints on riverine fluxes. Geology 29, 23–26.
Garzanti, E., Vezzoli, G., Andò, S., Paparella, P., Clift, P.D., 2005. Petrology of Indus
River sands: a key to interpret erosion history of the Western Himalayan Syn-
taxis. Earth Planet. Sci. Lett. 229, 287–302.
Giles, P.T., Franklin, S.E., 1998. An automated approach to the classification of the
slope units using digital data. Geomorphology 21, 251–264.
Goodbred, S.L.J., Kuehl, S.A., 1998. Floodplain processes in the Bengal Basin and the
storage of Ganges–Brahmaputra river sediment: an accretion study using 137Cs
and 210Pb geochronology. Sediment. Geol. 121, 239–258.
Goodbred, S.L.J., Kuehl, S.A., 2000. Enormous Ganges–Brahmaputra sediment dis-
charge during strengthened early Holocene monsoon. Geology 28, 1083–1086.
Graff, L.H., Usery, E.L., 1993. Automated classification of generic terrain features in
digital elevation models. Photogramm. Eng. Remote Sens. 59, 1409–1417.
Grujic, D., Coutand, I., Bookhagen, B., Bonnet, S., Blythe, A., Duncan, C., 2006. Cli-
matic forcing of erosion, landscape, and tectonics in the Bhutan Himalayas.
Geology 34, 801–804.
Hewitt, K., Gosse, J., Clague, J.J., 2011. Rock avalanches and the pace of late Quater-
nary development of river valleys in the Karakoram Himalaya. Geol. Soc. Am.
Bull. 123, 1836–1850.
Hinderer, M., 2001. Late Quaternary denudation of the Alps, valley and lake fillings
and modern river loads. Geodin. Acta 14, 231–263.
Hinderer, M., 2012. From gullies to mountain belts: A review of sediment budgets
at various scales. Sediment. Geol. 280, 21–59.
Jerolmack, D.J., Paola, C., 2010. Shredding of environmental signals by sediment
transport. Geophys. Res. Lett. 37, L19401.
Klingseisen, B., Metternicht, G., Paulus, G., 2008. Geomorphometric landscape anal-
ysis using a semi-automated GIS-approach. Environ. Model. Softw. 23, 109–121.
Koenker, R., Hallock, K.F., 2001. Quantile regression. J. Econ. Perspect. 15, 143–156.
Korup, O., Clague, J.J., Hermanns, R.L., Hewitt, K., Strom, A.L., Weidinger, J.T., 2007.
Giant landslides, topography, and erosion. Earth Planet. Sci. Lett. 261, 578–589.
Korup, O., Montgomery, D.R., Hewitt, K., 2010. Glacier and landslide feedbacks to
topographic relief in the Himalayan syntaxes. Proc. Natl. Acad. Sci. USA 107,
5317–5322.
Kuo, Y.-S., Tsang, Y.-C., Chen, K.-T., Shieh, C.-L., 2011. Analysis of landslide dam ge-
ometries. J. Mt. Sci. 8, 544–550.
Lague, D., 2010. Reduction of long-term bedrock incision efficiency by short-term
alluvial cover intermittency. J. Geophys. Res. 115, F02011.
Larsen, I.J., Montgomery, D.R., Korup, O., 2010. Landslide erosion controlled by hill-
slope material. Nat. Geosci. 3, 247–251.
Lupker, M., Blard, P.-H., Lavé, J., France-Lanord, C., Leanni, L., Puchol, N., Charreau, J.,
Bourlès, D., 2012. 10Be-derived Himalayan denudation rates and sediment bud-
gets in the Ganga basin. Earth Planet. Sci. Lett. 333–334, 146–156.
Métivier, F., Gaudemer, Y., 1999. Stability of output fluxes of large rivers in South
and East Asia during the last 2 million years: implications on floodplain pro-
cesses. Basin Res. 11, 293–303.
Milliman, J.D., Meade, R.H., 1983. World-wide delivery of river sediment to the
Oceans. J. Geol. 91, 1–21.
Milliman, J.D., Syvitski, J.P.M., 1992. Geomorphic/tectonic control of sediment dis-
charge to the Ocean: The importance of small mountainous rivers. J. Geol. 100,
525–544.
Montgomery, D.R., Brandon, M.T., 2002. Topographic controls on erosion rates in
tectonically active mountain ranges. Earth Planet. Sci. Lett. 201, 481–489.
Montgomery, D.R., Hallet, B., Yuping, L., Finnegan, N.J., Anders, A.M., Gillespie, A.,
Greenberg, H.M., 2004. Evidence for Holocene megafloods down the Tsangpo
River gorge, southeastern Tibet. Quat. Res. 62, 201–207.
Mool, P.K., Bajracharya, S.R., Joshi, S.P., 2001. Inventory of Glaciers, Glacial Lakes,
Glacial Lake Outburst Floods: Monitoring and Early Warning Systems in the
Hindu Kush-Himalayan Region, Nepal. International Center for Integrated Moun-
tain Development, Kathmandu.
Owen, L.A., Caffee, M.W., Finkel, R.C., Seong, Y.B., 2008. Quaternary glaciation of the
Himalayan–Tibetan orogen. J. Quat. Sci. 23, 513–531.
Pandey, V.P., Kazama, F., 2010. Hydrogeologic characteristics of groundwater aquifers
in Kathmandu Valley, Nepal. Environ. Earth Sci. 62, 1723–1732.
Phillips, J., 2003. Alluvial storage and the long-term stability of sediment yields.
Basin Res. 15, 153–163.
Qiu, J., 2008. China: The third pole. Nature 454, 393–396.
Schrott, L., Hufschmidt, G., Hankammer, M., Hoffmann, T., Dikau, R., 2003. Spatial
distribution of sediment storage types and quantification of valley fill deposits
in an alpine basin, Reintal, Bavarian Alps, Germany. Geomorphology 55, 45–63.
Schwanghart, W., Kuhn, N.J., 2010. TopoToolbox: A set of Matlab functions for topo-
graphic analysis. Environ. Model. Softw. 25, 770–781.
Simpson, G., Castelltort, S., 2012. Model shows that rivers transmit high-frequency
climate cycles to the sedimentary record. Geology 40, 1131–1134.
Sklar, L.S., Dietrich, W.E., 2001. Sediment and rock strength controls on river incision
into bedrock. Geology 29, 1087–1090.
Straumann, R.K., Korup, O., 2009. Quantifying postglacial sediment storage at the
mountain-belt scale. Geology 37, 1079–1082.
Thiede, R.C., Ehlers, T.A., 2013. Large spatial and temporal variations in Himalayan
denudation. Earth Planet. Sci. Lett. 371–372, 278–293.
Thiede, R.C., Bookhagen, B., Arrowsmith, J.R., Sobel, E.R., Strecker, M.R., 2004. Cli-
matic control on rapid exhumation along the Southern Himalayan Front. Earth
Planet. Sci. Lett. 222, 791–806.
Van De Wiel, M.J., Coulthard, T.J., 2010. Self-organized criticality in river basins:
Challenging sedimentary records of environmental change. Geology 38, 87–90.
Van der Beek, P., Van Melle, J., Guillot, S., Pêcher, A., Reiners, P.W., Nicolescu, S.,
Latif, M., 2009. Eocene Tibetan plateau remnants preserved in the northwest
Himalaya. Nat. Geosci. 2, 364–368.
Vance, D., Bickle, M., Ivy-Ochs, S., Kubik, P.W., 2003. Erosion and exhumation in the
Himalaya from cosmogenic isotope inventories of river sediments. Earth Planet.
Sci. Lett. 206, 273–288.
Wald, D., Allen, T., 2007. Topographic Slope as a proxy for seismic site conditions
and amplification. Bull. Seismol. Soc. Am. 97, 1379–1395.
Wasson, R.J., 2003. A sediment budget for the Ganga–Brahmaputra catchment. Curr.
Sci. 84, 1041–1047.
Williams, W.A., Jensen, M.E., Winne, J.C., Redmond, R.L., 2000. An automated tech-
nique for delineating and characterizing valley-bottom settings. Environ. Monit.
Assess. 64, 105–114.
Wobus, C.W., Hodges, K.V., Whipple, K.X., 2003. Has focused denudation sustained
active thrusting at the Himalayan topographic front?. Geology 31, 861–864.
Yafeng, S., Chaohai, L., Ersi, K., 2010. The glacier inventory of China. Ann. Glaciol. 50,
1–4.
Yatagai,A.,Arakawa,O.,Kamiguchi,K.,Kawamoto,H.,Nodzu,M.I.,Hamada,A.,2009.
A44-year daily gridded precipitation dataset for Asia based on a dense network
of rain gauges. SOLA 5, 137–140.
Yin, A., 2006. Cenozoic tectonic evolution of the Himalayan orogen as constrained by
along-strike variation of structural geometry, exhumation history, and foreland
sedimentation. Earth-Sci. Rev. 76, 1–131.
Zadeh, L.A., 1965. Fuzzy sets. Inf. Control 8, 338–353.
Zeitler, P.K., 1985. Cooling history of the NW Himalaya, Pakistan. Tectonics 4,
127–151.
Zeitler, P.K., Meltzer, A.S., Koons, P.O., Craw, D., Hallet, B., Chamberlain, C.P., Kidd,
W.S.F., Park, S.K., Seeber, L., Bishop, M.P., Shroder, J., 2001. Erosion, Himalayan
geodynamics, and the geomorphology of metamorphism. GSA Today 11, 4–9.