Laserfiche WebLink
<br />II. EXPERIMENTAL PLAN <br /> <br />The purpose of the experimental plan is to tie <br />together theory and practice to produce a workable <br />engineering program. In the past, many programs <br />have employed annual precipitation as the princi- <br />pal predictor, and comparison of a target area and <br />control area as the means to assure that any detected <br />changes were real. This approach leads, in the first <br />case, to the requirement for a long test period-to <br />average out natural annual variations-and in the <br />latter case, to a nagging uncertainty as to the con- <br />sistency of the target/control area relationship. <br />It appeared to us that both of these objections <br />could be overcome if: <br /> <br />there were no predictors in the conventional sense-that <br /> <br />is, no "forecasting", <br /> <br />one verifier were actual artificial nuclei concentrations in <br />target snowmelt and snowcrystal replicas, <br /> <br />one verifier were single-storm characteristics as a function <br />of measurable, physical parametric quantities, and <br /> <br />the target and control areas were one and the same. <br /> <br />This is very much an applied effort-rather than <br /> <br />a theoretical approach to the problem. It is based <br /> <br />on rather fine-scale measurements of physical quan- <br />tities at brief time intervals, and measurements that <br /> <br />have a demonstrably high signal-to-noise ratio <br /> <br />(SIN). The questions requiring answers are: <br /> <br />did artificial nuclei reach the target area, and if so, in <br />what quantity and where? <br /> <br />did the artificial nuclei participate in snowcrystal pro- <br />duction because of a deficiency in natural nuclei, and if <br />so, in what proportion to natural nuclei? <br /> <br />in participating, did the artificial nuclei modify the <br />single-storm characteristics of precipitation-rate to an <br />appreciable extent-that is, did any real "weather modifi- <br />cation" occur? <br /> <br />The answers to the first two questions could <br />determine the precipitation augmentation achieved <br />-if we prove that there is a deficiency in effective <br /> <br />freezing nuclei to begin with. The answer to the <br />last question determines whether the storm struc- <br />ture itself was modified and by how much. All <br />three answers are needed if we are to gage local <br />effects and estimate broader consequences. <br />The experimental plan, then, included the re- <br />quirement to measure three principal quantities: <br /> <br />tracer diffusion over the target area-because we did not <br />wish to contaminate the area. while gathering natural <br />precipitation data, <br /> <br />artificial nuclei diffusion, and participation in snowcrystal <br />formation, and <br /> <br />the precipitation-rate and precipitation-total character- <br />istics of storms as they traverse the target and surround- <br />ing area. <br /> <br />As background for these measurements we desire <br />others, most of which have been gathered routinely <br />in the past for conventional meteorological studies. <br />They include surface winds and temperature over <br />the area of interest, yertical wind and temperature <br />profiles, radar pictures of storms, and snowcourse <br />and stream gage measurements for long term aver- <br />ages and loss~alculations. <br />It is important to remember that precipitation <br />augmentation, by definition, can only be measured <br />on a plane above the snowpack and quasi-instanta- <br />neously. It cannot be measured on a conventional <br />snowcourse-because one must constantly estimate <br />variable evaporation and melting effects, nor on a <br />stream-because one must constantly estimate varia- <br />ble evapo-transpiration, diversion and ground trans- <br />port effects. Snowcourse and stream gage measure- <br />ments, on the other hand, are the only way to gage <br />runoff water production by the snowpack regardless <br />of whether precipitation augmentation occurred. <br />Thus, when we include within the desired measure- <br />ments snowcourse and stream gage data, it is with <br />the intent of detennining in what proportion any <br /> <br />5 <br />