WMOD00461 - Laserfiche WebLink

Home Browse Search

The general objectives of the scientific experiment are to answer the following questions: o When (e.g., under what conditions) can precipitation and streamflow be increased? o How much (eog., total amount) can precipitation and streamflow be increased? ! o Where (e.g., in what areas) can precipitation and streamflow be increased? In a scientific experiment it is also important to check the presumed physical process at a few major points in its sequence, perhaps three to six, to verify that the conceptual model of seeding effects is a useful approximation of reality or, if not, where it must be modified. This knowledge is essential to optimizing the seeding effects and to trans- ferring the resulting technology to other areas of Thailand and, as appropriate, other areas of the world. - The core component in the design of the demonstr,ation project is the postulation of the physical conceptual model (s), that is a statement(s) of the alterations in cloud properties and resultant precipitation and streamflow that occur when the clouds (warm and cold) are seeded in a prescribed manner. The numerical cloud modeling stUdy described in Part IV, section B.3, guided by established theory and experimental experience, is the primary vehicle for formulating appropriate concep- tual model (s) for the warm and cold cloud types that occur in Thailand. The modeling studies should provide information Ion the relative effec- tiveness of various seeding materials and chemicals; the results of introducing them in the clouds at various times iand locations, and in various dosages; and provide the treatment effect models for the sta- tistical analyses (e.g., precipitation as a function of treatment level temperature, treatment amount, cloud liquid water content, etc.). In addi ti on, they shoul d provide insight into what ,events in the preci pi ta- tion process will be affected and what measurements will be needed to document them. The most promising seeding strategies that lead to the largest changes in precipitation and streamflow through a recognizable sequence of physical events, as determined throUigh the numerical cloud modeling study, would be the best candidates for conceptual models to be tested 1n the demonstration project. The assessment team urges the use of a randomized treatment plan in order to obtain a concurrent, unbiased sample of non-treated clouds in the target area with which the treated sample can be comparedo A separate randomized treatment plan will be required for each conceptual model being tested and each should acconrnodate statistical evaluations based on both precipitation and streamflow. Different randomized treat- ment plans that are tailored to the particular meteorological and geographical situation should be tried, eog., randomized pairs, ran- domized blocks, randomized crossover, etc., to identify the most power- fu 1 treatment plan. Th is wi 11 requ ire sens it i vity stud i es of the ~ ~ f" 52