Predicted and observed long night and day temperature trends

We performed a principal component analysis of observed winter and summer temperature north and south of 55° latitude, in order to examine the spatial and temporal history of long day and night temperatures. We used the component weightings calculated from the observed record to calculate an analogous set of component scores from an infrared-transient coupled ocean-atmosphere general circulation model (GCM). We found no evidence for the increase in the observed night temperatures that is observed at lower latitudes. Comparison of the observed and GCM-calculated climate revealed that in only one of forty possible components were the patterns of observed and calculated climate change similar. This behavior is disturbing, because it is at these latitudes that the greatest climatic changes have been predicted to have already occurred.