WMOD00285 - Laserfiche WebLink

Home Browse Search

. procedure, both indices proved to be variable and difficult to forecast, resulting in counterproductive seeding and negative effects~ In view of these lessons, development and confirmation of improved recognition of seedability by direct measurement techniques is a primary objective. The advanced instrumentation and data depiction systems discussed in this report provide the means to this goal. The Operational Phase Operational Objective. The operational objective will be to integrate the modification hypotheses and design technology into a distinct, usable augmentation tool. The operations plan will focus on a strong physical measurement component with a reliable operational technique. Information developed during earlier pro- ject phases will result in a suspension criteria system for periods of excessively high snowpacks, avalanche risk, and other areas of public concern to insure the safety and well-being of project area residents and visitors. The statistical design will permit all positive opportunities to be treated to achieve maximum augmentation of the San Juan water supply and also to insure data collection for ongoing and post-hoc evaluation of project operations. The designs under considerations are: (1) a target-control design in which all potential experi- mental events are seeded in one area with a nearby area serving as a fixed control; or (2) a crossover design that requires random seeding of one of two candidate areas as a target with the other nonseeded area serving as the control. The target-control design requires a high degree of correlation between the target and the control; the higher the correlation, the fewer the samples needed to quantify and evaluate the precipitation increases. Since every suitable event in the target area is seeded, maximum additional precipitation and streamflow will result. Crossover designs are expected to provide verification of seeding effects sooner than other designs. Contamination of the control has been a serious problem in this mode; however, the combination of a dry ice seeding agent and aircraft seeding can eliminate this restraint. An additional consideration of the crossover design is that both the target and the control must be approximately the same size and must meet project criteria simultaneously. The final selection of the project design will be based on consideration of the characteristics of the San Juan Basin, the nature of the storms affecting the area, and consideration of evaluation requirements. Statistical tests will be based primarily upon permutation and nonparametric statistical techniques. Preference will be given to those tests that avoid assumptions about normality and other characteristics of the distributions. V-6