Abstract ID: 3.11780 | Accepted as Talk | Requested as: Talk | TBA | TBA

Snowline on a glacier marks the transition between snow and bare ice surfaces and is particularly suitable for mapping during the melt season using remote sensing. Traditional approaches of monitoring the snowline rely on Landsat or Sentinel-2 missions; however, the long revisit time and short observation periods of these missions limit their glaciological applications. This is particularly crucial in data-sparse regions, such as Central Asia. We designed a novel method to retrieve snowlines from the MODIS surface reflectance products, covering the period since the beginning of the 21st century. To bridge the coarse spatial resolution of MODIS, we established a statistical relationship between its reflectance in NIR band and high-resolution glacier snow cover data of Sentinel-2. This integration produces a time series with both high spatial and temporal resolution. We validated our results against manually derived snowlines from Landsat imagery. Our results provide unique insight into the glacier snowline dynamics. The high density of snowline observations enabled us to capture the snow depletion throughout the melt season, including the end-of-summer snowline. By applying this method to selected glaciers representing different mountain regions of the Pamir and Tien Shan, we analysed the snowline variability over the last 24 years. The annual rate of snowline change reveals an accelerating trend in annual snowline retreat. Our results provide insight into the onset and duration of the melt season, intra-seasonal snow cover changes and sub-seasonal to monthly snowline variability. Our work also investigate the relationship between seasonal snowline evolution and air temperature. To link snowlines to glacier meltwater contribution, we used time series to constrain a surface mass balance model at sub-seasonal to daily scale and found that the model appears to deplete the glacier snow cover faster than indicated by MODIS-based snowlines. This has significant implications for estimating glacier runoff. The presented snowline time series provide a unique tool to improve melt modelling at unprecedented temporal resolution, especially relevant during the growing season for Central Asia. It ultimately has the potential to be integrated into operational, near real-time glacier monitoring, which is thus of great benefit for water resource management in the region.