Detection and analysis of landslide-induced river course changes and lake formation (2017-2020)

A better understanding of landslide-river interaction is crucial to assess and predict the resulting natural hazards. The objective of RiCoLa is to develop novel methods to semi-automatically detect locations of abrupt landslide-river interference in mountain regions and to analyse the derived results with respect to possible triggers and causes of such events.

The project contributes to understanding past and predicting future land surface change due to the interaction of the hillslope and channel systems, and helps improving management and mitigation of resulting natural hazard cascades such as dam breach and flooding. Multi-temporal optical satellite images (e.g. Landsat, Sentinel-2) are employed together with digital elevation data to extract landslides, river courses and lakes using object-based image analysis, and to detect and quantify river course changes by performing change detection and time-series analysis. Four study areas in different mountain regions (European Alps, Southern Alps of New Zealand, Central Mountains of Taiwan, Himalayas of Nepal) serve to validate the developed methods and to evaluate and demonstrate their applicability and transferability. The study areas have a wide range of tectonic, climatic and lithological settings which make them ideally suited to extensively test our techniques. Results are fed into a comprehensive database, which is further enhanced with information on possible triggers and causes of the observed events. Statistical analyses are performed to determine spatio-temporal hotspots of landslide-induced river course changes and lake formation and to evaluate their linkage to predisposing, preparatory and triggering factors.

Instrument: Austrian Academy of Sciences (ÖAW)
Role: Partner
Project volume: 297,000 EUR, Z_GIS share: 135,750 EUR
Contact person: Daniel Hölbling
Researchers involved: Lorena Abad, Zahra Dabiri, Barbara Friedl
Project website: