An evaluation of eleven operational cloud seeding programs in the watersheds of the Sierra Nevada Mountains

A target–control statistical evaluation of 11 operational cloud seeding programs carried out in watersheds of the Sierra Nevada Mountains was conducted using Monte Carlo permutation (re-randomization) analysis. The water year (October–September) streamflow served as the response variable in the evaluations. The evaluation estimated the effect of seeding on unimpaired streamflow at each of the Sierra targets using the controls that give the most precise evaluation results possible with the available data. It was found that operational cloud seeding succeeded in increasing the streamflow in 6 of the 11 major watersheds in the Sierra Nevada Mountains. All 6 major watersheds indicating a positive seeding effect are on the western (upwind) side of the Sierra Nevada Mountain range. There was insufficient statistical evidence to reject the null hypothesis of no seeding effect in the other 5 major watersheds that were evaluated. It is noteworthy that the 5 watersheds whose evaluation was inconclusive include the 3 watersheds on the eastern (downwind) side of the Sierra Nevada Mountain range. The results of these evaluations and, in particular, those for the San Joaquin, Upper American and Carson–Walker operational cloud seeding programs illustrate the complexities involved in the transport and dispersion of silver iodide plumes from ground-based generators in mountainous terrain. The results suggest that aircraft seeding, either by itself or as a supplement to ground seeding, was able to affect targets that could not be affected by ground seeding alone. There was a statistically significant, positive seeding effect at the West Walker River Near Coleville target that was most likely due to contamination from an upwind seeding program, most likely the Mokelumne operational seeding program. Although contamination may have been present at the other seeding targets, it was not strong enough to affect the statistical results. Follow-on physical-statistical studies are needed to identify and understand the physical reasons for the statistical results of this study. In the opinion of this author, a comprehensive set of silver iodide tracer studies would contribute most to our understanding of the results, especially the dichotomy of seeding results for the operational seeding programs on the western and eastern watersheds of the Sierra Nevada Mountains.