EVALUATIONS FOR POTENTIAL OF GLACIAL LAKE OUTBURST FLOODS (GLOFS) IN THE BHUTAN HIMALAYA USING PRISM AND AVNIR-2 ONBOARD ALOS

摘要

The objectives of this study are to validate generated Digital Surface Models (DSMs) by PRISM onboard the Advanced Land Observing Satellite (ALOS, nicknamed "Daichi") in snow and glacier regions to contribute to the evaluation of potential Glacial Lake Outburst Floods (GLOFs) and their mitigation in the Bhutan Himalaya. Glacial lake inventory will also be generated to investigate the expansion process of glacial lakes using historical satellite imageries, including those by ALOS, as current conditions. Failures of glacial lake dams terminated by natural moraines can cause outburst floods and represent a serious hazard to downstream regions. The development and expansion of glacial lakes is sometimes said to be due to recent global warming, while other researchers say no correlation between them exists. The fact is, people living in such regions are exposed to the risk of GLOFs. To mitigate the damage of GLOFs, investigation into the melting process of glaciers and the development and expansion processes of it and evaluations of outburst potential of glacial lakes and triggers of GLOF are necessary. Preparations for risk management, such as operational monitoring of glaciers and glacial lakes, hazard maps, early warning systems, etc., are equally as important. Earth observation satellites have the potential to investigate existing conditions and operational monitoring of glaciers and glacial lakes. This paper introduces a study to validate generated PRISM DSMs in general terrains without snow as well as those covered by snow in mountainous areas. Airborne Lidar data and field-based GPS measurements have been used to validate PRISM DSMs as reference data. The height accuracies of PRISM DSMs have achieved 2.88 m (RMS) at the Alaska test site, and 6.31 m (RMS) at Mt. Tateyama, Japan. Both sites were covered by snow. These results are sufficient to analyze glaciers and glacial lakes in the Bhutan Himalaya. The three-dimensional glacial lake inventory is currently being generated by PRISM and AVNIR-2 as the latest condition.