1.4 A new look at statistical evaluations of cloud seeding experiments

Tuesday, 8 January 2013: 9:15 AM
Room 9A (Austin Convention Center)
Roelof Bruintjes, NCAR, Boulder, CO; and M. Dixon, C. Weeks, and S. Tessendorf

Historically randomized statistical experiments to evaluate rainfall enhancement experiments have focused on target/control areas with surface precipitation gauges being the primary evaluation data set, During the past twenty years radar based evaluation methods have become more generally used. From radar each storm is tracked objectively, and various characteristics are measured at 5-minute intervals for as long as the evaluation is intended out mostly to 60 minutes after seeding. Primary measures of the effects of seeding include radar derived rain flux and mass. Exploratory Data Analysis (EDA) techniques are used to display and compare the two groups (seeded and unseeded) of storms. The results of one-tailed hypothesis tests are presented. Permutation tests are used to compare the differences in the medians, quartiles, and means of both samples.

Variations in meteorological conditions can dominate the effects of seeding and are often times much larger than the effect of seeding (10-100 times). These variations can occur in space and in time and can significantly affect the results from any randomized seeding experiments depending on a single statistical test assuming that the samples are randomly drawn from the same distribution of potential values (treatment application for these measurements was at random). More statistically efficient means of analysis are required if we hope to gain significant results in realistic time frames such as multivariate statistical models by including covariates that influence the precipitation processes in a region to control for natural variations in rainfall. In contrast to pure randomization analysis, this type of analysis estimates the conditional contribution to rainfall by meteorological and for example aerosol effects.

The presentation will provide an overview of methods and analyses from recent experiments to highlight some of the issues related with statistical analyses and evaluation of rainfall enhancement experiments. It will also provide a future framework for conducting randomized cloud seeding experiments.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner