An analysis of late twentieth and twenty-first century predictions of Arctic circulation patterns in a ten-model ensemble of global climate system models, using the method of self-organizing maps (SOMs), is presented. The model simulations were conducted in support of the fourth assessment report of the intergovernmental panel on climate change (IPCC). The analysis demonstrates the utility of SOMs for climate analysis, both as a tool to evaluate the accuracy of climate model predictions, and to provide a useful alternative view of future climate change.
It is found that not all models accurately simulate the frequency of occurrence of Arctic circulation patterns. Some of the models tend to overpredict strong high-pressure patterns while other models overpredict the intensity of cyclonic circulation regimes. In general, the ensemble of models predicts an increase in cyclonically dominated circulation patterns during both the winter and summer seasons, with the largest changes occurring during the first half of the twenty-first century. Analysis of temperature and precipitation anomalies associated with the different circulation patterns reveals coherent patterns that are consistent with the different circulation regimes and highlight the dependence of local changes in these quantities to changes in the synoptic scale circulation patterns. Copyright © 2006 Royal Meteorological Society.