China: Imerys – calcium carbonate

Abstract

Episodic and catastrophic landslides are considered to be one of the main sources of sediment in the steep, mountainous landscapes of the Himalayas. However, the evolution of a single landslide through time and its contribution to erosional processes remain poorly constrained. In this study, we focus on a single, large (0.5 km2) landslide in a small catchment on the southern flank of the Annapurnas in Nepal (the Khudi valley) in order to quantify its importance in the overall erosion of this steep Himalayan catchment.The evolution of the Saituti landslide has been continuously monitored by remote sensing for the past 46 years. During that period, the Saituti landslide displayed sustained activity, such that the area of the landslide scar increased by a factor of 4. This retrogressive failure, a consequence of several sporadic flank and crown collapses, has not been continuous. Rather, acceleration phases have alternated with more quiescent periods. Simultaneously, the upper edge moved upward by 900 m. Based on field evidence from recent activity (such as scarps and open tension cracks above the landslide) and on an analysis of slope angles, at least the next 500 m is expected to fail.Volume losses within the landslide were estimated from differences between digital elevation models (DEMs) and from changes in landslide area, using a calibrated power law relationship between landslide area and volume. Corresponding landslide-induced erosion rates at the scale of the whole Khudi catchment were found to be 2.6 ± 0.9 mm/y for the past half-century. Those rates are similar to denudation rates obtained from sediment load measurements between 1999 and 2004. Those results, along with the lack of other major landslides in the valley for the last 46 years, suggest that the Saituti landslide plays a dominant role in the modern erosion of the High Himalayan Khudi catchment for the last years and possibly for the past few decades.We propose that continuous and sustained activity of a few major landslides over the past few decades might represent a significant contribution to the erosion of the High Himalayan range. This long-lasting, landslide-induced erosion should be taken into account when interpreting suspended load measurements, results from provenance analysis, or cosmogenic nuclides in river sand. Such processes should also be included in landscape evolution models when annual to secular problems are explored.