Decreasing trends of mean and extreme snowfall in High Mountain Asia

Intense warming profoundly alters precipitation phase patterns and intensity in High Mountain Asia (HMA). While snowfall climatology and precipitation extremes have been studied, there is a lack of understanding of snowfall extremes within HMA. Here, we investigate the spatial and temporal variability of non-extreme and extreme snowfall in hydrological years 1979-2020 using multi-source meteorological data, compare weather systems during extreme and non-extreme snowfall events, and identify key circulation factors that influence fluctuations in mean annual snowfall and extreme snowfall. The snowfall amount (-0.13 d/mm), days (-0.56 d/a), and fraction (-0.0012) were significantly reduced in HMA, with a shorter snowfall season (-0.52 d/a). Some extreme snowfall metrics (maximum 1-day snowfall and maximum 3-day snowfall) were insensitive to climate change, whereas the maximum consecutive snowfall days (-0.007 d/a), snowfall amount (-0.0023 mm/a), heavy snowfall days (S95pD; 0.0087 d/a), and extremely heavy snowfall days (S99pD; -0.1019 d/a) showed significant decreases. Synthetic analyses show that extreme snowfall events were more likely to occur within a narrow temperature range (-5 °C to 3 °C) with higher relative humidity and precipitation compared to non-extreme events. A stepwise regression method was used to determine that the fluctuation in the average annual snowfall was closely related to the Atlantic Multidecadal Oscillation, whereas the variation in extreme snowfall was mainly influenced by the Southern Oscillation Index. Our research provides a reference for assessing the potential impacts of climate change on a regional scale for risk management and disaster adaptation.