The importance of spring atmospheric conditions for predictions of the Arctic summer sea ice extent

Recent studies have shown that atmospheric processes in spring play an important role for the initiation of the summer ice melt and therefore may strongly influence the September sea ice concentration (SSIC). Here a simple statistical regression model based on only atmospheric spring parameters is applied in order to predict the SSIC over the major part of the Arctic Ocean. By using spring anomalies of downwelling longwave radiation or atmospheric water vapor as predictor variables, correlation coefficients between observed and predicted SSIC of up to 0.5 are found. These skills of seasonal SSIC predictions are similar to those obtained using more complex dynamical forecast systems, despite the fact that the simple model applied here takes neither information of the sea ice state, oceanic conditions nor feedback mechanisms during summer into account. The results indicate that a realistic representation of spring atmospheric conditions in the prediction system plays an important role for the predictive skills of a model system.