Procrustes Target Analysis: a multivariate tool for identification of climate fluctuations

Agriculturally important climate fluctuation types are identified by using an expert systems approach to synthesize information concerning the sensitivity of various aspects of Midwestern corn production to climatic variability. This information forms target criteria which are the basis for subsequent multivariate analysis using a technique new to meteorology, Procrustes Target Analysis (PTA), to fit the target to climatological data. Mathematical derivation of PTA is presented, along with an example of its application. The results of the analysis indicate that significant climate anomalies exist in the Midwest which agree with the target coefficients. Their spatial evolution is presented. Further investigation reveals that coherent areas of these fluctuations persist for large periods of the 15-year windows examined and that they appear to impact corn production. -Authors