A numerical technique for simulation of cloud seeding experiments

Two numerical cloud seeding experiments, using historic rainfall for the Deccan plateau region in Maharashtra state, were performed adopting different simulation techniques. The data used consisted of 1-day total rainfall for the 5-year period 1951–55. A double-area cross-over design with area randomisation was adopted. The first experiment, EXP-TR, was based on the simulation technique of Twomey and Robertson which involves about 100 hr of Robotron EC-1040 computer time. The second experiment, EXP-MMM was based on a different simulation technique proposed in the present study. The results of EXP-TR and EXP-MMM have shown close agreement. The numerical simulation technique of EXP-MMM is more promising for the following two reasons: (i) the computational time is reduced by about an order of magnitude without compromising the scientific value of the results, and (ii) a direct estimate of the lower limit of the double ratio value which can be detected at 5% level of significance is defined. The results of the two numerical experiments suggested that, for the Deccan plateau region, 15 and 20% increases in rainfall due to seeding could be detected with 80% or more probability in 5 years.