STATISTICAL SIGNIFICANCE OF LONG-RANGE OPTIMAL CLIMATE NORMAL TEMPERATURE AND PRECIPITATION FORECASTS

A simple approach to long-range forecasting of monthly or seasonal qua ntities is as the average of observations over some number of the most recent years. Finding this ''optimal climate normal'' (OCN) involves examining the relationships between the observed Variable and averages of its values over the previous one to 30 years and selecting the ave raging period yielding the best results. This procedure involves a mul tiplicity of comparisons, which will lead to misleadingly positive res ults for developmental data. The statistical significance of these OCN s are assessed here using a resampling procedure, in which time series of U.S. Climate Division data are repeatedly shuffled to produce stat istical distributions of forecast performance measures, under the null hypothesis that the OCNs exhibit no predictive skill. Substantial are as in the United States are found for which forecast performance appea rs to be significantly better than would occur by chance. Another comp lication in the assessment of the statistical significance of the OCNs derives from the spatial correlation exhibited by the data. Because o f this correlation, instances of Type I errors (false rejections of lo cal null hypotheses) will tend to occur with spatial coherency and acc ordingly have the potential to be confused with regions for which ther e may be real predictability. The ''field significance'' of the collec tions of local rests is also assessed here by simultaneously and coher ently shuffling the time series for the Climate Divisions. Areas exhib iting significant local tests are large enough to conclude that season al OCN temperature forecasts exhibit significant skill over parts of t he United States for all seasons except SON, OND, and NDJ, and that se asonal OCN precipitation forecasts are significantly skillful only in the fall. Statistical significance is weaker for monthly than for seas onal OCN temperature forecasts, and the monthly OCN precipitation fore casts do not exhibit significant predictive skill.