Towards prediction of snowmelt onset over Arctic sea ice

Snowmelt onset signifies an important transitional point in the Arctic’s surface energy budget, when albedo decreases and energy absorption increases rapidly. Spatial and temporal variations in the timing of snowmelt onset are relatively large, and they are significantly related to variations in low-frequency atmospheric patterns, such as the Arctic Oscillation (AO) and the North Pacific (NP) anomaly. Furthermore, there are significant relationships between snowmelt onset dates occurring from March through June and the February atmospheric teleconnection indices. The objective of this paper is therefore to examine statistical relationships between snowmelt onset and winter through early spring (November to February) atmospheric teleconnection indices. A stepwise regression is used to predict snowmelt onset based on the atmospheric teleconnection indices, while an investigation of the atmospheric manifestations of the indices (e.g. near surface air temperatures, mean sea level pressures) is utilized to give a physical interpretation of the observed statistical relationships.