Understanding the 2004 Glacier Detachment in the Amney Machen Mountains, Northeastern Tibetan Plateau, Via Multi-Phase Modeling

Glacial detachment is a newly emerged glacial hazard type, and more than 20 cases have been identified worldwide. The Xiao Ma Glacier in the Amney Machen Mountains, northeastern Tibetan Plateau, experienced an intensive detachment of the terminal low-slope glacier in 2004, forming a large dammed lake downstream and burying 2.27 km2 of grassland. An advanced multi-phase r.avaflow model was used to simulate this glacier detachment event to reconstruct its slide mechanism and dynamic process. In total, 9.77 × 106 m3 mass detached were categorized into three different phases according to their properties: the old ice remaining in the middle layer following the 1988–1997 glacier surge (47% of total volume), new ice in the upper layer originating from the 2003–2004 surge, which directly caused the 2004 detachment (48%), and subglacial water trapped in the bedrock (5%). The modeling results showed that the glacier detachment–induced debris flow reached a peak discharge of 0.25 × 106 m3/s at 40 s at the mouth of the Xiao Ma Valley, and the entire propagation process was completed within 140 s. The initially trapped meltwater dominated the debris flow state, with a maximum velocity of 103 m/s. The underlying old glacial ice was pushed and deposited to the front of the fan, and the new ice in the upper layer was primarily deposited over the middle fan. Additionally, several important parameters in the r.avaflow model were recommended for the optimal interval of values via optimality analysis, and the potential of the model to be extensively applied for large-scale risk assessment was discussed under relatively simple settings. The modeling of the 2004 Amney Machen glacier detachment has resulted in a comprehensive understanding of the dynamic processes of such events, which is significant for future glacier disaster simulation and assessment thereby improving regional strategies in hazard early warning, prevention, and mitigation.