Three-dimensional simulations of future GLOF events in High Mountain Asia under different SSP scenarios

In High Mountain Asia (HMA), rising temperatures and retreating glaciers are leading to the formation of new glacial lakes and the expansion of existing ones. The sudden release of water from such lakes can lead to devastating glacial lake outburst floods (GLOF) threatening people and infrastructure for many kilometers downstream. Therefore, it is important to obtain information on future glacial lakes, e.g., their location, area, and volume as well as the timing of their development. This data can in turn be used to estimate the range and destructive potential of future GLOF events, which is crucial for the sustainable development of settlements and infrastructures.

Dam failures at glacial lakes and the subsequent flooding events are often investigated using two-dimensional models (e.g. HEC-RAS). These 2D models are based on the solution of the shallow water equations, which assume that the vertical velocity of the water is always much lower than the horizontal velocity. In case of a moraine failure, however, high vertical accelerations are observed in the behavior of the dam break wave in mountainous terrain, which violates the shallow water equations. To overcome these shortcomings of 2D models, fully three-dimensional Computational Fluid Dynamic (CFD) models can be used, which are based on the solution of the Navier-Stokes equations along with the volume of fluid method to locate the interface between water and air.

In our research project, we use the 3D open-source CFD model OpenFOAM to determine the possible range of GLOF events at future glacial lakes in HMA. To capture the different climate pathways and the corresponding differences in lake volume, we use previously published data on the evolution of future glacial lakes over the course of the 21st century under different SSP scenarios. To account for the uncertainties of future moraine height or composition, we simulate different moraine failure scenarios resulting in different magnitudes of GLOFs. These simulations allow us to determine the velocity of the initial break wave and identify potential inundation areas. By intersecting the resulting flood map with land-use and land-cover maps (while taking into consideration the potential changes in this data during the coming decades), we can estimate affected agricultural land and potentially damaged infrastructures. Our findings can contribute to helping local communities adapt to emerging challenges, implement risk minimization measures, and enhance sustainable development in HMA.