Analysis of Radar Observations of a Randomized Cloud Seeding Experiment

Abstract

Magnetic tape records for radar observations of 80 moving one-hour test cases in a three-way randomized (no-seed, salt, silver iodide) cloud seeding experiment have been analyzed in terms of echoing areas and radar-estimated rainfall amounts. Individual test cases ranged from non-precipitating cumulus up to moderate thunderstorms with echoing areas exceeding 100 km² and rainfall estimated at 3000 kT in 1 h.Out of numerous predictor variables, cloud depth is found to be the best single predictor for both echoing area and radar-estimated rainfall. The echoing area and radar-estimated rainfall are very closely correlated. A cube-root transformation of the radar-estimated rainfall improves the correlation between cloud depth and the radar-estimated rainfall for the no-seed (control) sample to 0.91. For clouds of a given depth, both the echoing area and radar-estimated rainfall are larger in seeded than in unseeded cases. The differences between no-seed and salt cases are of marginal statistical significance, but the differences in echoing area and rainfall between no-seed and silver iodide cases are significant at the 1% level. The indicated effects, expressed as a percentage of the echoing area or radar-estimated rainfall in the no-seed cases, decrease with cloud depth.A comparison of no-seed and AgI cases with the aid of a one-dimensional steady-state cloud model shows that AgI seeding may have led to increases in maximum cloud height averaging 600 m.It is concluded that seeding affected the precipitation in the Cloud Catcher test cases through both the microphysical processes and the cloud dynamics.