Seasonal Sensitivity of the Jet-Stream to Arctic Temperatures

Near-surface Arctic warming (aka “Arctic amplification”) has been shown in previous modeling studies to impact the midlatitude jet-streams. While both the future Arctic temperature and the circulation responses are projected to vary with season, an open question is how much of the seasonality in the circulation response is due to seasonality in the Arctic temperatures and how much is due to seasonality in the dynamics itself. Furthermore, climate models vary substantially in their ability to simulate the seasonal cycle of the mid-latitude circulation and also vary substantially in their projections of its future response, raising the question of whether biases in the mean-state seasonality (and thus seasonal dynamics) might be partially to blame for differences in the seasonal responses across models. 

In this work, we take a Granger-causality regression approach to quantify the part of the zonal winds that respond to changes in near-surface Arctic temperatures in the CMIP5 models. We demonstrate that the sensitivity of the jet position to Arctic warming is a function of season, being largest in summer and weakest in winter. We argue that this seasonal sensitivity is due to the induced wind anomalies remaining relatively fixed in latitude, while the jet migrates in and out of the anomalies throughout the annual cycle. We then demonstrate that model biases in the jet-stream position lead to potential biases in the jet-stream’s sensitivity to Arctic warming.