A Possible Experimental Design for Testing the Effect of Artificial Cloud Seeding

摘要

Because of the clash of opinion on the effectiveness of artificial cloud seeding as a means of increasing precipitation, an objective method of appraisal is imperative. The problem is herein shown to be statistical, and the general philosophy of valid statistical inference is elucidated. In the light of this background, the peculiarities of meteorological data are pointed out, and the pitfalls of ordinary procedures are discussed. An experimental design capable of resolving the present controversy is presented in the Appendix, and evidence of its soundness and power is cited in the text. (Author)