Abstract: Affected by global climate change, most glaciers in southeastern Tibet have deteriorated more and more frequently. The risk of mountain disasters, such as floods, surges and debris flows, has increased sharply in recent years. This paper implements a research on glacial lakes classification by using multi-source synthetic aperture radar (SAR) images. According to the statistical analysis of intensity difference between signals coming from water and non-water objects, a method of glacial lake extraction and dynamic monitoring based on the intensity standardization ratio of sequential SAR images is proposed. Within a typical experimental area in Gongba Glacier basin, the image series of ALOS/PALSAR-1 of Japan Space Agency and Sentinel-1A of European Space Agency are selected to carry out dynamic extraction of glacial lakes and long-term change analysis over 11 years'time, just for validation purpose. The temporal and spatial variations of glacial lakes at the end of Gongba Glacier from 2007 to 2018 are successfully obtained. And the further analysis finds out that the volume of glacial lakes increase rapidly in the past ten years. This evidence can also confirm the current situation of aggravated melting of Gongba Glacier. In addition, the recent monitoring results in 2018 show that the expansion of local burst gushes has broken the inherent life cycle of glacial lakes, and even trigger secondary disasters of floods and debris flows. It is necessary to strengthen monitoring and prevention.