Cloud seeding trials in the rainy belt of western Colombia

Evaluation of a rain stimulation operation for the Río Anchicayá, near Buenaventura, Colombia, during ‘dry’ seasons since 1963 was hampered by lack of rainfall data. Daily streamflow data were manipulated to remove serial correlation and to derive a series of daily equivalent rainfalls by comparing each day's runoff with the maximum corresponding runoff decrement observed on rainless days. The Bayes criterion, applied to frequency distributions of equivalent daily rainfall before and during the seeding, consistently recommended continuation of the seeding. The total equivalent rainfall per month was found to correlate well with the number of days per month with equivalent rainfall above 2.5 mm, and comparison of the seeded period with the regression of equivalent rainfall per month on number of rainy days per month for unseeded history indicated an increase of 13%, with one chance in 8 of random occurrence, corresponding to 2100 kw added power generation and about a sixfold economic return.