Statistical Techniques and Key Issues for the Evaluation of Operational Weather Modification

A major factor calling for evaluation of operational cloud seeding projects concerns the growth of its usage. A number of statistical techniques were selected and tested on simulated weather changes in western Kansas, and it was found that principal component regression is consistently one of the more powerful evaluation techniques. Key issues affecting statistical evaluations were identified, they were better comparison, uniform definition of sampling unit, uniform measurementof response, choice of statistical technique, bias- consciousness, starting and stopping effects, and validity of historical comparison.