Relative Contributions of Global Warming, ENSO and the Arctic Oscillation to Snow Cover and Their Projection on Surface Air Temperature and Snowfall in Sensitive Regions

Abstract

In this study, we explored the reasons for variability in monthly snow cover extent (SCE) in November–April during 1972–2021 in the Northern Hemisphere and clarified the relative contributions of global warming (GW), the El Niño–Southern Oscillation (ENSO), and the Arctic Oscillation (AO), as well as the relative contributions of local surface air temperature (SAT) and snowfall projected upon them. The study focused on regions where SCE showed a large standard deviation. Application of singular value decomposition and partial correlation methods showed that monthly SCE variability was affected predominantly by three modes: GW, ENSO and the AO. Based on multivariate information flow analysis, the impact of GW on SCE was found most pronounced on the Tibetan Plateau and in central Asia with a percentage contribution of > 73% and > 49%, respectively. The impact of the AO on SCE was found most pronounced in Europe, central-eastern North America, and the region east of Lake Baikal with a percentage contribution of > 75%, > 73% and > 55%, respectively. The impact of ENSO on SCE was found most pronounced in western North America with a percentage contribution of > 51%. The increased SCE related to GW in the eastern Tibetan Plateau can be attributed to southerly flows that transport moisture from the Indian Ocean and converge in the eastern Tibetan Plateau. Anomalous anticyclones related to positive phases of the AO and ENSO caused a reduction in snowfall and an increase in SAT, then a substantial decline in SCE.