A modeling and observational investigation of North Atlantic SST anomalies and their effects on Eurasian snow cover

Both observational and modeling approaches are used to examine the role of a North Atlantic SST anomaly dipole-type pattern in relation to snow cover in three different regions of Eurasia: (1) west of the Ural Mountain range, (2) West Siberian Plain, and (3) Central Siberian Plateau region. In the observational study, both the North Atlantic Oscillation (NAO) and the Arctic Oscillation (AO) are shown to be highly correlated with the North Atlantic SST anomalies of this dipole pattern and also with snow cover variability in the three Eurasian regions, especially during late winter and spring months. Two model sensitivity cases are simulated by specifying two opposite dipoles of SST anomalies in the model's oceanic boundary forcing conditions. The model used for these two experiments is the NCAR Community Climate Model, version 3 (CCM3), which is an atmospheric global climate model coupled to a Land Surface Model (LSM). The first experiment consists of a positive SST anomaly in the western Atlantic Ocean, just east of the U.S., and a negative SST anomaly farther north which is located to the east of Newfoundland. This experiment reflects real-world ocean conditions during 1989-94, which also corresponds to a very positive phase of the NAO/AO. The second experiment has the SST anomalies of the opposite sign; it reflects the time period, 1964-69, and coincides with a very negative NAO/AO phase.

The results of the two experiments show that the model produces a somewhat realistic atmospheric response to the two SST anomaly dipoles that resembles the spatial pattern of the NAO and AO over the North Atlantic and Europe. The model shows a lag in the response when compared to the observations. The observations reveal a maximum difference in sea level pressure between the two time periods in the early winter months, but the model has a maximum in the late winter to spring months. The model compares well with the observations for the snow cover variability west of the Ural Mountain range, but does not capture any realistic variability east of the Urals. This suggests that CCM3 does not show a central and eastern Eurasian atmospheric signal in response to the SST anomaly forcing over the North Atlantic. These results have been found in other modeling studies as well. Model biases that contribute to the lack of an atmospheric signal farther downstream in Eurasia are presented.