Snow accumulation dynamics and its contribution to the hydrology of a glacierized catchment in the Northern Pamirs

Mountain glaciers are shrinking at accelerating rates due to enhanced ablation and reduced accumulation. In High Mountain Asia (HMA), recent glacier and snow changes have been highly heterogeneous, due to differences in accumulation regimes and sensitivity of glacier mass balances to temperature increases. The Pamir-Karakoram region is well known for hosting some of the only glaciers featuring neutral or even positive mass balance since the 2000, yet the causes for this anomaly are not fully understood, neither how long it will persist in the future nor its hydrological implications. In the semi-arid basins of Central Asia, snow- and glacier melt sustains most of the annual streamflow, with glacier melt being especially important towards the end of the dry summers. However, very few direct observations exist at high elevation, hindering the quantification of glacier mass inputs which is essential to estimate the long-term sensitivity of glaciers to warming.  In this study, we combine in-situ hydro-meteorological observations with remote sensing observations to constrain a land-surface model and understand snow accumulation dynamics at a glacierized catchment in the Pamir mountains of Tajikistan. In-situ snow height and mass changes have been collected since 2021 from automatic weather stations, time-lapse cameras and pressure loggers in seasonally frozen lakes, providing a uniquely rich dataset for this region. We use MODIS, Landsat-8 and Sentinel-2 satellite images to derive snow cover dynamics at high spatial and temporal resolutions, and very high-resolution (2m) optical stereo imagery (Pleiades) to derive spatially resolved snow depths. These in-situ and remote-sensing observations are then used to inform a land-surface model that we force with statistically downscaled and bias-corrected reanalysis data (ERA5-Land) at 100m spatial and hourly temporal resolution, from 2015 to 2023. We use our model to dissect the glacier mass balance seasonal dynamics, to quantify how much mass is gained through snowfall and avalanches, and how much mass is lost through melting and sublimation. We find that glaciers in our catchment receive a large part (58 %) of their annual mass input (1081 mm w.e.) from March to July, suggesting that spring and early summer precipitation events are key to control accumulation and therefore dictate glacier mass balances. Importantly, 11% of the annual snowfall is returned to the atmosphere via sublimation. At the catchment scale, snowmelt contributes to 67% of the annual runoff (625 mm), followed by glacier melt (24%) and rain (9%). When most of the seasonal snowpack has melted out (usually in August), glacier melt becomes the dominant contribution (with 55% in September). In most of the study period years, the glacier mass balance is close to neutral, but it turned negative in the last three years, where warmer conditions have led to more rapid seasonal snowpack melt-out and higher glacier ELAs, deteriorating the health of these previously spared glaciers and casting doubts on their ability to provide fresh water during the dry summers in the longer term.