WMOD00519 - Laserfiche WebLink

Home Browse Search

I IN'rRODUCTION 1. Under the sponsorchip of the World Heteor010gical Organization, an experi- ment is being planned with the aim of determining the feasibility of artificial stimulation ancl enhancement of precipitation. Planning for this Precipitation Enhancement Project (PEP) is going forward with the dual purpose of providing guidance for implementation of PEP and for use by cOillltries interested in undertaking projects of their ovrn in precipitation enhancement. In keeping ",ith this dual purpose, the present review of the hydrological aspects of evaluation of precipitation enhancement will go beyond the specific requirements of PEP. Although the measuI'ement of rainfall, an area of interest in hydrology', is being addressed else\>rhere j.n the planning for PEP, some of the problems encowltered in measuring precipitation aI'e a logical part of this discussion. 2. The history of the use of streamflo"l in evaluating the effecti vene8S of precipitation enhancement efforts is nearly as long as the modern day history of oloud seeding. The approach has been used in some long-established operational projects, as well as in some randomized experiments. Over the years, a nwnber of investigators (Thom, 1957; Crawford and Linsley, 1963; Yevdejevich, 1965; l1arkovic, 1966; Hastay and Gladwell, 1968, Watson and Denny" 1971; and Horel-Seytoux, 1972) have addressed the advantages and disadvantages of hydrological. approaches to evalua- tion, and examined the applioabilitY of available statistical tests in providing guidanoe as to the confidence to be placed in the evaluation results. Huch of what follows is drawn from their work. II THE RATIONALE FOR CONSIDERING HyJ)ROLOGICAL EVALUATION 3. Preoipitation enhancement is illldertaken with the aim of increasing the water supply of a region by useful amounts. Economic justification of such effor.ts requires convincing evidence that the water supply has been augmented over what would have occurred in the absence of the enhancement effort. The task of obtaining this evidence is rendered difficult by the great variability of precipitation events, coupled ,ofith a present inability to oompute, with acceptable accuracy, ,,,hat amoilllts of precipitation can be expected to result from a given set of meteorological conditions. Although rudimentary mathematical moclels of cloud processes can be used to estimate the amount of precipitation that simple clouds should produce, the errors of estimate involved have been too great to instil widespread confidence in the results of such computations. 4. In the absence of a generally accepted method of judging the effectiveness of precipitation enhancement efforts on a storm-by-storm basis, evaluation has usually taken the form of comparj,sons of grouped experiences of trea tecl and untreated precipitation events. If the group of treated events is truly comparable to the group of untreated events, from the standpoint of meteorological. conditions, a..YJd if the errors of estimating the precipitation for both groups are acceptably small, a credible comparison can result. 5. However, the investment involved in making the estimates of ])recipitation sufficiently accurate may exoeed the resources available, In addition to the illlcer- tainties about how effectively a particular type of precipitation gauge measures the precipitation falling on the area covered by the gauge, particularly when the preci- pitation occurs in windy conditions or as snow, a major source of uncertaintJr arises ,,,henever a network of point observations is used to infer the precilli tation experience Over large areas interveninG between the points at which the precipitation i~: measured. When convection is involved in the precipitation process, large amounts can fall undetected, lmless the gauging network is quite dense. In addition to the