Abstract. Large boulders, ca. 10 m in diameter or more, commonly linger in Himalayan river channels. In many cases, their lithology is consistent with source areas located more than 10 km upstream, suggesting long transport distances. The mechanisms and timing of “exotic” boulder emplacement are poorly constrained, but their presence hints at processes that are relevant for landscape evolution and geohazard assessments in mountainous regions. We surveyed river reaches of the Trishuli and Sunkoshi, two trans-Himalayan rivers in central Nepal, to improve our understanding of the processes responsible for exotic boulder transport and the timing of emplacement. Boulder size and channel hydraulic geometry were used to constrain paleo-flood discharge assuming turbulent, Newtonian fluid flow conditions, and boulder exposure ages were determined using cosmogenic nuclide exposure dating. Modeled discharges required for boulder transport of ca. 103 to 105 m3 s−1 exceed typical monsoonal floods in these river reaches. Exposure ages range between ca. 1.5 and 13.5 ka with a clustering of ages around 4.5 and 5.5 ka in both studied valleys. This later period is coeval with a broader weakening of the Indian summer monsoon and glacial retreat after the Early Holocene Climatic Optimum (EHCO), suggesting glacial lake outburst floods (GLOFs) as a possible cause for boulder transport. We, therefore, propose that exceptional outburst events in the central Himalayan range could be modulated by climate and occur in the wake of transitions to drier climates leading to glacier retreat rather than during wetter periods. Furthermore, the old ages and prolonged preservation of these large boulders in or near the active channels shows that these infrequent events have long-lasting consequences on valley bottoms and channel morphology. Overall, this study sheds light on the possible coupling between large and infrequent events and bedrock incision patterns in Himalayan rivers with broader implications for landscape evolution.

Abstract. Large boulders, ca. 10 m in diameter or more, commonly linger Himalayan river channels. In many cases, their lithology is only compatible with source areas located > 10 km upstream suggesting long transport distances. The mechanisms and timing of exotic boulder emplacement are poorly constrained, but their presence hints at processes that are significant for landscape evolution and geohazard assessments in mountainous regions. We surveyed river reaches of the Trishuli and Sunkoshi, two trans-Himalayan rivers in central Nepal to improve understanding of the processes responsible for exotic boulder transport and the timing of emplacement. Boulder size and channel hydraulic geometry were used to constrain paleo-discharges and boulder emplacement ages were determined using cosmogenic nuclide exposure dating. Modelled discharges required for boulder transport, of ca. 103 to 105 m3/s, exceed typical monsoonal floods in these river reaches. Exposure ages range between ca. 1.5 and 13.5 kyrs BP with clustering of ages around 4.5–5 kyrs BP in both studied valleys. This later period is coeval with a broader weakening of the Indian summer monsoon and glacial retreat after the Early Holocene Climatic Optimum (EHCO), suggesting Glacial Lake Outburst Floods (GLOFs) as a possible cause for boulder transport. We, therefore, propose that these exceptional events are climate-driven, but counter-intuitively occur in the wake of transitions to drier and warmer climates leading to glacier retreat rather than during wetter periods. Furthermore, the old ages and prolonged preservation of these large boulders in or near the active channels shows that these infrequent events have long-lasting consequences on valley bottoms and channel morphology. Overall this study sheds light on the possible coupling between large-infrequent events and bedrock incision patterns in Himalayan rivers with broader implications on landscape evolution.

Abstract. Increasingly advanced and affordable close-range sensing techniques are employed by an ever-broadening range of users, with varying competence and experience. In this context a method was tested that uses photogrammetry and classification by machine learning to divide a point cloud into different surface type classes. The study site is a peat scarp 20 metres long in the actively eroding river bank of the Rotmoos valley near Obergurgl, Austria. Imagery from near-infra red (NIR) and conventional (RGB) sensors, georeferenced with coordinates of targets surveyed with a total station, was used to create a point cloud using structure from motion and dense image matching. NIR and RGB information were merged into a single point cloud and 18 geometric features were extracted using three different radii (0.02 m, 0.05 m and 0.1 m) totalling 58 variables on which to apply the machine learning classification. Segments representing six classes, dry grass, green grass, peat, rock, snow and target, were extracted from the point cloud and split into a training set and a testing set. A Random Forest machine learning model was trained using machine learning packages in the R-CRAN environment. The overall classification accuracy and Kappa Index were 98% and 97% respectively. Rock, snow and target classes had the highest producer and user accuracies. Dry and green grass had the highest omission (1.9% and 5.6% respectively) and commission errors (3.3% and 3.4% respectively). Analysis of feature importance revealed that the spectral descriptors (NIR, R, G, B) were by far the most important determinants followed by verticality at 0.1 m radius.

No abstract