Winter 2008 in North America - post-mortem analysis
The distribution of SAT anomalies during the winter (DJF) of 2008 (Fig. 1) was in general agreement with our forecasts issued in June and October 2007. Negative temperature anomalies were observed over much of western and central North America, and positive SAT anomalies were in the southern and eastern parts of the continent, stretching from Texas to the Canadian Maritime provinces (Fig. 1 and Fig. 2 that shows it in more details for the United States). As predicted, frequent outbreaks of cold Canadian air were observed in the Midwestern US, particularly in February, when mean monthly temperatures were 6°C to 8°C below normal. The average temperature for the contiguous United States was near its long-term mean (Fig. 3).
The atmospheric circulation was characterized by a frequent upper-level toughing over the West and ridging over the East (Fig. 4). The jet stream position was close to what is typically observed during La Niña winters (Fig. 5), but shifted about 15 degrees west. As expected, the PNA index was close to zero. The TNH index was positive (0.8), which is typical for La Niña winters (rule 108).
The jet stream steered frequent storms from the Pacific Northwest toward Utah and Colorado and then northeast, toward the Great Lakes. These storms brought record snow to the Colorado Rockies and the Midwest. The amount of winter precipitation in parts of the Colorado Rockies exceeded 300% of normal (Fig. 6). Wisconsin's capital set a new record for snowfall, with more than 196.3 cm (77.3 inches) of snow by the end of February. The previous record was set during the winter of 1978/1979, when 194.6 cm (76.6 inches) of snow fell. See also an interesting analysis of why it was snowing so much in southern Wisconsin this winter.
One notable area where the forecast failed was the southwestern United States. Contrary to the forecast, mean winter temperatures there were below average. The forecast was based primarily on the canonical correlation analysis between winter SAT over North America and preceding summer-fall SST anomalies in the North Atlantic by van den Dool et al. (2007) (see rule 109). In the summer-fall 2007, the SST pattern closely resembled the one associated with a positive SAT anomalies in the southwestern US.
These expectations were reinforced by the winter forecasts issued by the Climate Prediction Center (CPC) and the International Research Institute for Climate Prediction;(IRI). However, the anomalously high pressure center, which dominated;in the Southwest from March through November 2007, suddenly disappeared in December, and after a very warm fall season, temperatures there dropped to below normal values.




