Assessing RAMMS for glacial lake outburst floods: a comparative study across diverse geographical regions

Glacial lake outburst floods (GLOFs) pose significant threats to downstream infrastructure and communities and are considered the most hazardous glacial phenomena globally. GLOFs can occur with minimal forewarning and reach extreme magnitudes due to their capacity to entrain sediments, thereby altering their flow behavior. Since the 2000s, several models have been developed to address the dynamic and complex behavior of GLOFs in terms of inundation extents and runout distances. This thesis investigates the capability of the numerical single-phase model RAMMS to replicate these characteristics. To assess the inundation extent and flow heights of GLOF simulations, the study sites of Weingarten Lake in Switzerland and South Lhonak Lake in India were analyzed. The findings demonstrate that RAMMS can accurately replicate GLOF inundation extents, performing comparably with other models such as HEC-RAS or r.avaflow. Furthermore, the study underscores the critical influence of applied Digital Elevation Model (DEM) and input volumes on modeling outcomes. The scope of the simulations in this study was constrained by limited computational power and data availability. Although the single-phase RAMMS version applied in this research successfully models GLOF characteristics, it remains limited to a single process within the cascading event chain and cannot independently replicate flow transitions. However, the upcoming two-phase version of RAMMS Is expected to address this limitation. Future research should focus on examining this elevated GLOF model. Additionally, a comprehensive literature review of friction parameters used in previous GLOF studies with RAMMS could help modelers select appropriate values for predictive simulations.