Spatial and temporal variability in snow density across the Northern Hemisphere

As a fundamental physical property of snowpack, snow density is used to describe many essential features of snowpack behaviour. However, the variability in snow density across the Northern Hemisphere (NH) is largely unknown. Here, we investigate snow density variability in conjunction with snow classes and geographic elements in the NH based on 6,954 snow sites from 1909 to 2019. Precipitation, air temperature, and snowfall based on meteorological sites, as well as the aridity index (AI) and wind speed from reanalysis data, are also applied to describe the effect of climate on snow density. The results present that the long-term mean snow density is 246 & PLUSMN; 70 kg/m3 considering all in-situ measurement sites. Considerable spatial heterogeneity in snow density exists with contrasting snow densities among differing snow classes. The values range from 198 & PLUSMN; 79 kg/m3 for ephemeral snow to 363 & PLUSMN; 63 kg/m3 for maritime snow. For the seasonal evolution of snow density, the different snow classes share a general characteristic with the overall NH, a slight decrease from October to September, followed by a sustained increase. Moreover, the densification rate in the snow stable period varies over a much smaller range than that during the snowmelt period. Furthermore, the longitudinal trends in the variability of snow density are more pronounced compared to altitudinal and latitudinal trends. High snow densities are typically associated with adequate precipitation, warm air temperature, large aridity index, a long snow season, and heavy snowfall for different snow classes. The results will deepen the understanding of the snow density distribution at hemispherical scale, and provide basic data for the remote sensing of snow water equivalent and parameterization of snow models.