Changing post-landslide cascade hazard intensity due to increasing temperature: a process-based simulation of the 2021 Melamchi disaster in Nepal

Abstract Background Large, devastating hazards persist in the Himalayas, primarily driven by hydrological and tectonic processes. Recent observations, however, indicate that climate warming may subtly modify geomaterial properties, altering the behaviour of landslide hazards and the associated process chains. Building on evidence from geotechnical tests and climate projections, this study investigates the influence of a temperature-dependent soil shear strength on cascading hazard dynamics under a set of boundary conditions using a multi-hazard modelling framework. The simulation was performed using the process chain observed in June 2021 in Melamchi (central Nepal), and a comparison was made with a projected late-century climate scenario corresponding to a worst-case temperature increase of 4.9 °C. Temperature-controlled ring-shear experiments offered insights into how soil behaviour varies with temperature, especially concerning fines content. These insights were integrated into the model to project their effects under future warming, while remaining within a process-based simulation. Results Although the thermally induced reduction in internal friction angle was modest (< 0.6°), the multihazard simulation revealed pronounced changes in hazard behaviour. Warmer conditions led to longer, faster landslide runouts, resulting in higher landslide dams. Ultimately, process-chain modifications delayed the peak discharge while increasing its magnitude, leading to higher debris-flow heights, amplified sediment fluxes to elements at risk, altered entrainment patterns, and increased impact pressures at critical infrastructure. Conclusions The results demonstrate that temperature-driven changes in soil friction alone can measurably intensify hazard extremes in high-mountain environments. This highlights the importance of incorporating the thermo-hydro-mechanical response of soil into advanced modelling frameworks to prevent underestimating future risks and to support effective, climate-resilient countermeasures in mountainous regions.