Results of On-Top Glaciogenic Cloud Seeding in Thailand. Part I: The Demonstration Experiment

Abstract

Randomized, cold-cloud, rain-enhancement experiments were carried out during 1991?98 in the Bhumibol catchment area in northwestern Thailand. Exploratory experimentation in 1991 and 1993 suggested increases in rainfall from seeding. A demonstration experiment followed in which the design was specified in advance (i.e., a priori) to determine the potential of on-top silver iodide (AgI) seeding for the enhancement of area (1964 km2) rainfall. It was conducted in accordance with a moving-experimental-unit design as discussed in the design document and summarized herein. The evaluation of the demonstration experiment until its scheduled termination in 1998 involved both cell and unit analyses. The S-band project radar collected 5-min volume-scan data to be used to evaluate cell and unit properties and to determine the mean cell motion vector for the advection of the experimental unit with time. The cell dataset comprises 353 AgI-seeded cells and 289 control cells that received simulated AgI (glaciogenic) treatment. All cells were tracked using the Rosenfeld long-track procedures. The proportional effect of seeding on cell rain volume as estimated by radar is 35% with a one-sided P value of 0.139, which falls short of the P-value threshold of 0.025 that is required for statistical significance. The lower and upper bound of the corresponding 90% confidence interval is ?14% and +111%, respectively. Analysis of the unit sample was limited to those cloud systems that postanalysis retracking revealed had been treated and had remained within the boundaries of the moving unit. The proportional effect of seeding on unit rainfalls at 300 min after unit qualification for the sample of 62 experimental units (31 seeded and 31 nonseeded) is 46% with a one-sided P value of 0.107. Thus, the effect of seeding on unit rainfalls also fell short of statistical significance at the threshold P value of 0.025. The 90% confidence interval for the strength of the seeding effect ranges from ?11% to +142%, and the one-sided probability that the seeding effect is ≥0%, ≥5%, and ≥10% is 90%, 86%, and 83%, respectively. Regression analysis to account for the potential impact of prequalification unit rainfall biases favoring the seeded sample had no effect on the results of the evaluation. Although the demonstration experiment did not reach statistical significance, much is to be learned about the potential effects of cold-cloud seeding in Thailand from exploration of the full cell and unit demonstration samples, which is done in a companion paper (Part II).