The sensitivity of Arctic Sea ice to cloud radiative conditions in spring and early summer
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The rapid decline in Arctic sea ice is a key driver of the amplified warming signal observed in the Arctic region, making this a critical phenomenon in climate science. Accurate seasonal sea ice projections, however, remain challenging due to a large degree of interannual variability in sea ice extent. This study analyzed the role of clouds in the early melt season, and their associated surface radiative effects, in modulating the magnitude of sea ice loss. A combination of observed and modeled sea ice thickness data was used to track temporal and spatial patterns of sea ice volume loss. A stepwise multiple linear regression analysis revealed that variants of Arctic cloud radiative fluxes in March and June were valuable in predicting the total volume of sea ice loss during the melt season. This study then explored the causalities behind the particular variable selection by the regression model, which yielded an adjusted R2 value of 0.88. Downwelling longwave cloud radiative fluxes in March were found to be negatively correlated with melt onset, with enhanced downward fluxes initiating earlier melt. Downwelling longwave fluxes in June were interpreted to be significant due to the large volume of ice volume lost in June, as well as the heightened effect of clouds on the surface radiative budget during periods of maximum insolation. Sea ice loss can also be influenced by the spatial patterns and magnitude of sea ice advection. Anomalous surface wind conditions and resulting anomalies in sea ice advection, were found to be critical in 2013, a year that fell outside the confidence interval of the regression model.