High Mountain Asia (HMA) hosts the largest glacier concentration outside of polar regions. This region also represents the second major surge cluster, globally. Despite numerous sub-regional and individual glacier – scale studies of surging, our understanding of this phenomenon remains largely ambiguous. This study sought to identify and quantify trends in the characteristics and distribution of glacier surging synoptically across HMA using remote sensing techniques. This was with the intention enhancing the state of knowledge of glacier surging in this highly dynamic region and demonstrating the capabilities of remote sensing for this purpose.

Terminus positions of surging glaciers identified in this study using satellite imagery were delimited using Google Earth Engine Digitisation Tool (GEEDiT). Subsequently, using terminus change profiles, a 6-part classification scheme was developed to characterise surge phenomenology. Surging variables, including duration, terminus advance distance, terminus advance rate, and surge recurrence intervals were also quantified using terminus change profiles. A statistical assessment was performed on the data to test surging relationships both intra- and extra- sub-regionally.

In total, 139 glaciers were identified as surging in HMA between 1987-2019. Of this 139, 15 surged multiple times, and 52 were newly identified as surging in this study. Overall, surging across HMA is heterogeneous in distribution and behaviour. Surge incidence is greater along a northwest-southeast transect, stretching from Pamir through Karakoram, to West Kunlun Shan. In Tian Shan and Himalaya surge incidence was isolated. Pamir glacier surge variables support a dominant hydrologically-regulated surge mechanism, in contrast to Karakoram and West Kunlun Shan glaciers, characterised by long durations, and likely controlled by a thermal switch mechanism. This supports the idea that surging occurs along a spectrum of flow instability, validating earlier surge studies.

The synoptic nature of this study provides a degree of representation of the surge phenomenon not attainable by smaller-scale studies. The results highlight the necessity to update the RGI, and the continued need to monitor the surging phenomenon in HMA. It would be possible to automate the methods adopted in this study, facilitating near real-time monitoring of surging across HMA.