Summer 2008 forecast for North America

Summary: The mean summer (JJA) SAT averaged over the contiguous United States is forecast to decrease compared to the summer of 2007, but still remain above the 1971-2000 average. Warmer and drier than average conditions are expected in the western and central U.S. The North American monsoon will likely be late and dry. Near to slightly below normal temperatures are forecast for the eastern U.S. The Midwest and the Southeast have an increased chance of above normal precipitation.

Fig. 1. A forecast of surface air temperature anomalies in North America for the summer (JJA) of 2008.  The blue (red) contours and the letter C (W) indicate colder (warmer) than normal temperatures. CL means near normal. The base period for calculating anomalies is 1971-2000. The numbers at the contours are confidence factors of the forecast.

Fig. 2. Same as Fig. 1, except for precipitation. The orange (green) contours and the letter D (W) indicate drier (wetter) than normal conditions.

Discussion

Currently the forecast rules in the knowledge base are available primarily for the contiguous United States. Less then 50% of the rules were used in this forecast. As the information about spring climate variables becomes available, more rules will be triggered.

Temperature

The mean summer (JJA) SAT averaged over the contiguous U.S. is forecast to decrease compared to the summer of 2007, but still remain above the 1971-2000 average (rule 363). The confidence in this statement is relatively high (CF = 50). Less certain, however, is the forecast of a spatial distribution of SAT anomalies.

The general pattern of SAT anomalies will probably be characterized by above normal temperatures in the western and central U.S. and near to slightly below normal temperatures in the eastern U.S. A high pressure ridge is expected to be positioned over the Rocky Mountains, which will result in anomalously warm and dry summer in the West. This forecast is also supported by an increased likelihood of above normal temperatures in Tucson (rule 423), Sacramento (rule 430 and rule 432) and Kansas City (rule 278).

Even near normal temperatures in the eastern U.S. would mean a significant cooling from the summer of 2007, when the temperatures there were well above normal (1971-2000 average). Summer temperatures in the Great Lakes region will experience a general short-term cooling trend in the next 5 years or so, as described in rule 415. However, if the current Warm-Cold SST pattern in the North Pacific (doc) persists through the summer, the trend may be delayed (rule 417). The forecast of cooling in the Southeast is supported by rule 402, linking SAT in Atlanta to the QBO and solar activity.

The forecast of near to below normal temperatures in the East is in disagreement with the majority of publicly available forecasting models. All six models used for the PSD seasonal forecast guidance (and, hence, their average forecast) show above normal temperatures in the eastern U.S. The guidance is similar to the forecast issued by the U.K. Met Office, as well as to the persistence forecast. Both the CPC outlook and IRI multi-model probability forecast show an increased probability of above normal temperatures along the Atlantic seaboard and near normal temperatures in the area stretching from the Great Lakes south to Texas. Colder than normal temperatures in Texas are predicted by the ECPC model. Only the CFS seasonal forecast is quite similar to ours.

Precipitation

Drier than normal conditions are forecast in the West, where anticyclonic circulation will likely to dominate. In the Southwest, the summer (JAS) monsoon is expected to be late and dry. This is suggested by rule 283, which describes the relationship between the monsoonal rain and preceding winter precipitation pattern along the West Coast. Also, given the negative phase of the PDO and positive phase of the AMO, there is a higher probability of drought conditions in the Southwest (rule 268).

When the Southwest is dry, there is a tendency for heavy rains in the Great Plains and the Midwest (rule 281). However, the probability of above normal precipitation in the Great Plains is reduced due to a lingering La Niña event (rule 295 and rule 103). 

Our forecast is consistent with other publicly available forecasts regarding the anomalously dry conditions in the West (see the CPC seasonal outlook, IRI multi-model probability forecast, U.K. Met Office forecast, and PSD seasonal guidance). It disagrees, however, with most of those forecasts regarding precipitation in the Great Lakes region, except, probably, the NCEP model in the PSD seasonal guidance.  The CCA forecast suggests above normal precipitation in the Southeast, and the ECPC model suggests that for the southern tier of the United States. Again, as with the temperature, only the CFS model output resembles our precipitation forecast.