Numerical simulation of long-distance debris flows (lahars) on glacier-clad volcanoes: the case of Cotopaxi, Ecuador

Abstract

Volcanic debris flows (lahars) are highly destructive volcanic phenomena and present significant challenges in numerical simulation. This manuscript tackles the three fundamental requirements for modelling gravitational flows: determining plausible source configurations; selecting suitable topographic data; and employing appropriate mathematical models to assess the current hazard posed by long-distance lahars at Cotopaxi volcano. After incorporating these elements, we successfully simulated the characteristics of a future 1877-type lahar under current conditions, accounting for glacier size and topography. For the source conditions, or “scenario”, we identified 27 equidistant source locations along the lower edge of the current glacier’s extent. Each source was assigned a hydrograph based on the weighted volume of water available on Cotopaxi’s current glacier. Additionally, we introduced a methodology for quantifying channel width when high-resolution digital elevation models (DEMs) are available. This method enabled us to determine the minimum pixel size required for accurate representation of ravine shapes. While higher resolution DEMs demand robust computational resources and extended computational timeframes, we upscaled Cotopaxi’s DEM from 3 m to 15 m to balance accuracy and efficiency, as a 15-m DEM capture over 90% of the topography and reduces computing time significantly. Optimizing DEM selection is crucial, especially when contemplating future ensemble approaches. After employing the dynamic-based model Kestrel, parameterised for large lahars, we obtained predictions closely aligned with field observations, historical flow conditions inferred for the 1877 lahar-event, and results from previous simulation studies. Notably, we observed higher depths and speeds in canyons compared to plains, consistent with historical reports and previous studies. Minor discrepancies in the inundation area, when compared with existing hazard maps, emphasize the importance of understanding flow dynamics and lahar trajectories for effective hazard assessment and mitigation strategies. Furthermore, our results contribute valuable information to current hazard maps and can aid in damage quantification and cost/benefit analyses, particularly when planning the construction of mitigation infrastructure.