Warm air intrusions and surface melt over sea ice from MOSAiC during spring 2020

Abstract:

Onset of surface melt over sea ice at the MOSAiC Central Observatory (CO) occurred at 83.3ºN on 25 May 2020. This date is defined thermodynamically following Persson (2012), but represents the point at which the transition in the surface energy budget from wintertime net cooling to summertime net warming begins to melt the snow rather than just warm it. It is the seasonal genesis of persistent top-side melt of the sea ice and in the central Arctic onset dates have trended earlier by ~7 d/da since 1979 (Bliss and Anderson 2018). Onset at MOSAiC was associated with a southerly air intrusion of cyclonic origin and melt initiation was dominated locally by longwave radiative processes. Melt was sustained nearly un-interrupted for ~128 hours. These characteristics are similar to the onset at both SHEBA in the Beaufort at 76.4ºN on 28 May 1998 (Persson 2012) and N-ICE2015 at ~80ºN near Svalbard on 1 June 2015 (Walden et al. 2017, Cohen et al. 2017). At MOSAiC, both shortwave and longwave processes worked in concert to sustain melt. In 2020, the breakdown of the polar vortex beginning in mid-April allowed for such air masses to reach the CO from the North Atlantic with synoptic regularity thereafter. Indeed, the air temperature was rarely above -20ºC before and rarely below -20ºC after. The bulk temperature of the ice responded similarly. Four unique pre-onset events having the potential to melt the surface occurred after 13 April. While episodic melt was observed conclusively on 19 April, evidence of heterogenous melt was observed earlier on the 16th, highlighting the importance of spatial variability in snow optical and/or physical properties. These events were followed by near-neutrality in the surface radiation budget through April before it turned persistently positive in early May. Pre-onset events reduced the (clear-sky) surface albedo by ~2% and were followed by colder near-surface air and, when melt occurred, refreezing of the upper snow layer. Conversely, melt onset was associated with an abrupt reduction of diffuse-regime surface albedo of >5% and subsequent episodes of cold air became less frequent. In this study, we examine the surface energy budget in more detail to elucidate the role of springtime warm air intrusions on melt and the reasons that events prior to onset produced melt that was brief and episodic rather than persistent.