TY - JOUR A1 - Stolle, Amelie A1 - Langer, Maria A1 - Blöthe, Jan Henrik A1 - Korup, Oliver T1 - On predicting debris flows in arid mountain belts JF - Global and planetary change N2 - The use of topographic metrics for estimating the susceptibility to, and reconstructing the characteristics of, debris flows has a long research tradition, although largely devoted to humid mountainous terrain. The exceptional 2010 monsoonal rainstorms in the high-altitude mountain desert of Ladakh and Zanskar, NW India, were a painful reminder of how susceptible arid regions are to rainfall-triggered flash floods, landslides, and debris flows. The rainstorms of August 4-6 triggered numerous debris flows, killing 182 people, devastating 607 houses, and more than 10 bridges around Ladakh's capital of Leh. The lessons from this disaster motivated us to revisit methods of predicting (a) flow parameters such as peak discharge and maximum velocity from field and remote sensing data, and (b) the susceptibility to debris flows from catchment morphometry. We focus on quantifying uncertainties tied to these approaches. Comparison of high-resolution satellite images pre- and post-dating the 2010 rainstorm reveals the extent of damage and catastrophic channel widening. Computations based on these geomorphic markers indicate maximum flow velocities of 1.6-6.7 m s(-1) with runout of up to similar to 10 km on several alluvial fans that sustain most of the region's settlements. We estimate median peak discharges of 310-610 m(3) s(-1), which are largely consistent with previous estimates. Monte Carlo-based error propagation for a single given flow-reconstruction method returns a variance in discharge similar to one derived from juxtaposing several different flow reconstruction methods. We further compare discriminant analysis, classification tree modelling, and Bayesian logistic regression to predict debris-flow susceptibility from morphometric variables of 171 catchments in the Ladakh Range. These methods distinguish between fluvial and debris flow-prone catchments at similar success rates, but Bayesian logistic regression allows quantifying uncertainties and relationships between potential predictors. We conclude that, in order to be robust and reliable, morphometric reconstruction of debris-flow properties and susceptibility requires careful assessment and reporting of errors and uncertainties. (C) 2015 Elsevier B.V. All rights reserved. KW - debris flow KW - peak discharge KW - channel geometry KW - geomorphometry KW - Bayesian logistic regression KW - Transhimalaya Y1 - 2015 U6 - https://doi.org/10.1016/j.gloplacha.2014.12.005 SN - 0921-8181 SN - 1872-6364 VL - 126 SP - 1 EP - 13 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Mey, Jürgen A1 - Scherler, Dirk A1 - Zeilinger, Gerold A1 - Strecker, Manfred T1 - Estimating the fill thickness and bedrock topography in intermontane valleys using artificial neural networks JF - Journal of geophysical research : Earth surface N2 - Thick sedimentary fills in intermontane valleys are common in formerly glaciated mountain ranges but difficult to quantify. Yet knowledge of the fill thickness distribution could help to estimate sediment budgets of mountain belts and to decipher the role of stored material in modulating sediment flux from the orogen to the foreland. Here we present a new approach to estimate valley fill thickness and bedrock topography based on the geometric properties of a landscape using artificial neural networks. We test the potential of this approach following a four-tiered procedure. First, experiments with synthetic, idealized landscapes show that increasing variability in surface slopes requires successively more complex network configurations. Second, in experiments with artificially filled natural landscapes, we find that fill volumes can be estimated with an error below 20%. Third, in natural examples with valley fill surfaces that have steeply inclined slopes, such as the Unteraar and the Rhone Glaciers in the Swiss Alps, for example, the average deviation of cross-sectional area between the measured and the modeled valley fill is 26% and 27%, respectively. Finally, application of the method to the Rhone Valley, an overdeepened glacial valley in the Swiss Alps, yields a total estimated sediment volume of 9711km(3) and an average deviation of cross-sectional area between measurements and model estimates of 21.5%. Our new method allows for rapid assessment of sediment volumes in intermontane valleys while eliminating most of the subjectivity that is typically inherent in other methods where bedrock reconstructions are based on digital elevation models. KW - sediment storage KW - sediment thickness KW - intermontane valleys KW - geomorphometry KW - artificial neural networks Y1 - 2015 U6 - https://doi.org/10.1002/2014JF003270 SN - 2169-9003 SN - 2169-9011 VL - 120 IS - 7 SP - 1301 EP - 1320 PB - American Geophysical Union CY - Washington ER -