Abstract
Under the context of climate change, significant variations in snow density have been observed in the High Mountain Asia, however, its spatiotemporal patterns and underlying drivers remain incompletely understood. By integrating ERA5 and ERA5-Land reanalysis datasets with large-scale atmospheric circulation data, combined with advanced statistical methods, this study systematically analyzes the spatiotemporal patterns and driving factors of snow density across multiple scales in the High Mountain Asia. The results indicate that: The snow density exhibited a significant decreasing trend at a rate of -0.4 kg/m3·per decade (p < 0.01) from 1960 to 2023. Spatially, snow density consistently demonstrated a "high in mountains, low in plateaus" distribution pattern, which is closely associated with snow depth and snow accumulation. Significant decreases in snow density were concentrated in areas with relatively low snow accumulation, such as the southwestern (S2) and southeastern (S3) Tibetan Plateau, where snowpack exhibits higher sensitivity to temperature variations. Snow depth and air temperature serves as key geographical factor influencing snow density, the latter primarily affects snow density by modulating the proportion of snowfall in total precipitation and altering snow phenology. The East Atlantic/Western Russia (EA/WR) teleconnection pattern indirectly influences snow density through its control on temperature. A weakened EA/WR pattern facilitates increased advection of warm air from the southeast into the Asian High Mountain region, thereby elevating summer temperatures and contributing to reduced snow density.