Changes in March mean snow water equivalent since the mid-20th century and the contributing factors in reanalyses and CMIP6 climate models

Trends in March mean snow water equivalent (SWE) in the Northern Hemisphere are attributed to changes in three main factors: total precipitation (P), fraction of precipitation as snowfall (F), and fraction of accumulated snowfall remaining on the ground (G). This trend attribution is repeated for two reanalyses (ERA5-Land from March 1951 to 2022 and MERRA2 - Modern-Era Retrospective analysis for Research and Applications, Version 2 - from 1981 to 2022) and simulations by 22 climate models from the 6th phase of the Coupled Model Intercomparison Project (CMIP6). The results reveal a decrease in SWE in most of the Northern Hemisphere, as decreases in F and G dominate over mostly positive trends in P. However, there is spatial variability in both the magnitude and sign of these trends. There is substantial variation between the individual CMIP6 models, but the agreement between the CMIP6 multi-model mean and ERA5-Land is reasonable for both the area means and the geographical distribution of the trends from 1951 to 2022, with a spatial correlation of 0.51 for the total SWE trend. The agreement for the trends from 1981 to 2022 is worse, probably partly due to internal climate variability but also due to the overestimation of the recent warming in the CMIP6 models. Over this shorter period for which ERA5-Land can be compared with MERRA2, there are also marked trend differences between these two reanalyses. However, the SWE decreases associated with reduced snowfall fraction (F) are more consistent between the different data sets than the trends resulting from changes in P and G.