WMOD00063 - Laserfiche WebLink

Home Browse Search

necded to undcrstand the reasons for thc dccrcase in seeding efTectiveness and, thereby, to determine if adjustments to the secding procedures can be made that restore the sceding et1cctiveness of these programs to the morc succcssfullevels they achicved earlicr. The preliminary findings of this study does not. by itselC constitute "scientilic proof' that sceding of orographic clouds increases precipitation, snowpack and subscquent streamflow; howevcr, they certainly constitute "amplc evidcncc" to justify operational application of this seeding technology I. Introduction There are opcrational (non-randomized) cloud seeding programs being conducted in almost all of the major watersheds of the Sicrra Ncvada Mountains of California aimed at increasing precipitation in ordcr to enhance streamflow for increased hydroelectric power generation with additional bencfits to downstrcam agriculture and rescrvoir recreation. All of these programs have, for the most part, been conducted continuously since their inception, the earliest one starting in the San Joaquin River Basin in water year 1951. All the operational programs expect to increase precipitation according to the same seeding conceptual model, that is by seeding for microphysical efTects to improve the precipitation efliciency of the clouds. Some operational programs try to accomplish this by conducting secding operations on both slimmer and winter storms to increase rainfall and to augment snO\vpack, respectively, whcreas somc conduct seeding operations on only wintcr stonns to augment snowpack. Both ground- based and aircralt sccding is bcing applied on some of the operational programs using sceding systems such as silvcr iodide ground generators. airborne silver iodide gencrators, airbornc silver iodide flares. and/or airborne hygroscopic flares whereas some of the operational programs only use ground-based silver iodide generators. Relatively few of these programs have been subjected to a fornMl statistical evaluation the results of which are availablc in thc open literature. Most of the evaluations appear in seeding contractor reports to their sponsors, utilities or water conscrvation agencies, or in reports by the sponsors themselves, The most consistently used streamflow cvaluation method is the so-called historical regression method, whereby thc sceding etTect is calculated on an annual basis as the diffcrencc between the observed streamtlow at a station reprcsenting the unimpaired streamflow of the basin and that predicted for that station by a control st3tion(s) through a regression equation derived from a period of record before cloud seeding was initiated. All of the published statistical evaluations reported increases in streamflow with strong statistical support. For example, Hendcrson (1966) cvaluated the first 10 years of the Kings River program using thc historical regression method and reported annual incrcases from -6.1% to +15.0% with an average increase in streamflow of 6.1 % that was significant at the 0.005 level. Subsequently, Henderson (2003b) evaluated the Iirst 47 ycars of the Kings River program using the historical regression method and reported annual incrcases ranging from -9.5% to +25,6% with an average increase of 5.5% having a statistical probability of 99.9%. The Panel on Weather and Climatc Modification to the Committee on Atmosphcric Sciences, National Academy of Scicnces conducted. an evaluation of the first 14 years of the San Joaquin River Basin Operational Cloud Seeding Projcct (National Acadcmy of Sciencc, 1966) using ntnolT as the tcst variable and reported a 7% increase at a signilicancc Icvel of 0,04. Thcre are no reports ofcvaluations of any of these operational seeding programs in the published litcraturc that cover their cntire pcriod of operations, 33