Cloud Seeding Research
As early as the 1940s, scientists and engineers have been conducting experimental trials, analyzing statistics, and searching to find specific methods that will result in increasing and augmenting natural rainfall and snowfall within different geographical areas of our planet. Despite extensive research over the years, up to only a few years ago statistical analysis of these weather and precipitation modification efforts commonly referred to as cloud seeding had not produced the scientifically convincing proof that is required to reach pertinent and reasonably accurate conclusions about the true effectiveness of cloud seeding efforts. This has changed with recent advances in instrumentation technology, additional …show more content…

The double ratio attempts to show the ratio of seeded rainfall to control rainfall and the sums of seeded and unseeded months (Morrison, Siems, Manton, & Nazarov, 2009). A further statistical procedure used to analyze the cloud seeding data was the use of bootstrap analysis. Bootstrap analysis is computer intensive approach that attempts to estimate the particular properties or variance of an estimator as a way of replacing the original observed set of data with a random resample as a way to produce a more accurate statistical inference. For example, bootstrapping is a straightforward procedure to produce confidence levels or standard errors for distribution parameters such as odds, ratios, proportions, and correlation coefficients (Morrison, Siems, Manton, & Nazarov, 2009).
All of the statistical procedures used to analyze the study dataset produced statistically significant results. While the simplistic arithmetic mean approach to seeded areas produced close to a 20% increase in precipitation over the control areas, the filtered results of the arithmetic mean produced more believable increase results in the 5.1%-14.5% range with 80% of the data showing exceeding the 0.05 significance level (Morrison, Siems, Manton, & Nazarov, 2009). Analysis of the double ratio approach produced similar results of between 5%-14% increases in precipitation levels with a confidence level approaching 95%. This confidence level