Results of a Randomized Hail Suppression Experiment in Northeast Colorado. Part VIII: The Representative Draw Analysis

View More View Less
  • 1 National Center for Atmospheric Research, Boulder, CO 80307
© Get Permissions Rent on DeepDyve
Restricted access

Abstract

A detailed analysis is made of the environmental conditions existing on each of the declared hail days during the randomized seeding experiment. From the many soundings available each day, the one most representative of the near-storm environment is carefully selected. This sounding is then used to compute several parameters known to influence hailfall. It is found that two parameters, both indicative of the thermodynamic instability, have a more instable mean value on the seed days than on the control days that in one case is statistically significant at the 10% level. Correcting for this draw would result in reducing the actual ratios of seed to control hail mass found in the primary statistical evaluation of the experiment. However, the reduction would not be sufficient in relation to the very wide 90% confidence limits to affect the statistical conclusions that the ratios were not significantly different from 1.0.

An analysis of the sequences of declared hail days showed that, in spite of the careful experimental design, the random selection process produced an actual partitioning of sequence starts into seed or control such that a sequence this extreme, or more extreme, had a chance of only 3 in 100 of occurring. However, it is not likely that this unexpected draw affected the evaluation of the experiment in any significant way, since it is taken care of indirectly in the analyses of the environmental parameters.

Abstract

A detailed analysis is made of the environmental conditions existing on each of the declared hail days during the randomized seeding experiment. From the many soundings available each day, the one most representative of the near-storm environment is carefully selected. This sounding is then used to compute several parameters known to influence hailfall. It is found that two parameters, both indicative of the thermodynamic instability, have a more instable mean value on the seed days than on the control days that in one case is statistically significant at the 10% level. Correcting for this draw would result in reducing the actual ratios of seed to control hail mass found in the primary statistical evaluation of the experiment. However, the reduction would not be sufficient in relation to the very wide 90% confidence limits to affect the statistical conclusions that the ratios were not significantly different from 1.0.

An analysis of the sequences of declared hail days showed that, in spite of the careful experimental design, the random selection process produced an actual partitioning of sequence starts into seed or control such that a sequence this extreme, or more extreme, had a chance of only 3 in 100 of occurring. However, it is not likely that this unexpected draw affected the evaluation of the experiment in any significant way, since it is taken care of indirectly in the analyses of the environmental parameters.

Save