WMOD00263 - Laserfiche WebLink

Home Browse Search

64 JOURNAL OF WEATHER MODIFICATION Volume 38 4.0 DISCUSSION The very small differences between seed and control precipitation predicted by the model were very disappointing and not expected at the onset of this project. Possible causes of such low seedability: . The model predicted seed ability could be real; however, because of the model over precipitation prediction bias and low amounts of supercooled liquid water con- tent, this possibility is doubtful. . The background CCN and IN concentra- tions are unknown but instead are deter- mined by our selected background concen- trations. Too low a background CCN con- centration would make clouds more effi- cient in natural precipitation formation thereby lowering seedability. Too high background IN concentrations would likely lead to lower seedability. . There is circumstantial evidence that the model-predicted supercooled liquid water content is too low, thereby lowering seed- ability. . The evaluated over-prediction bias in pre- cipitation may lead to reduced opportunities for precipitation enhancement in the model. . Banded patterns of seed - no seed differ- ences on daily totals suggest a possible very weak dynamic response to seeding. This pattern of differences results in much of the target area being in regions of reduced pre- cipitation. . The low-level warm temperature bias in the model results in delayed AgI nuclei activa- tion and reduced effectiveness of the seed- ing agent. However, this effect has overall a small impact on seedability. . The simulated transport and diffusion of seeding material from the generator sites is getting into the clouds too far downwind of the generator sites. However, the particle modeling suggests that seeding material is delivered to the target area at levels suitable for seeding, which argues against the notion that seeding material is not getting into the intended seeding zones. - Reviewed - 5.0 RECOMMENDATIONS It is recommended that additional modeling stud- ies are warranted because this was only a one-year contract and research funding was limited. One of the first things that needs to be done is to determine the cause of the model over-prediction bias in precipita- tion. Another is to explore the various hypotheses that have been put forward to explain the very small differences between seed and no-seed precipitation amounts. Still another area to explore is the low amounts of SL W in the 2-hr vertically integrated maps over the target area; additional sensitivity tests would be useful. Also, it would be desirable to rerun all or at least the 30 selected days with higher resolu- tion to determine if increased resolution reduced the precipitation bias and/or the seed, no-seed differ- ences. In support of future operational cloud seeding projects in which a model is used as part of the evaluation technique, it is urged that background CCN and IN concentrations be measured. Preferably this would be airborne but in lieu of that longer term ground-based measurements, particularly from higher-terrain sites, would be desirable. Other items that would be very useful in such a project would be a vertically-pointing radiometer near the summit on the target mountain barrier for SL W detection, and the use of scanning cloud radar for identifying re- gions of liquid water in the clouds and to follow pre- cipitation morphology. In addition the combination of model predictions and new observations such as cloud radar and radiometers could be used in a very sophisticated method of evaluation of an operational seeding project. Acknowledgements. The following individuals are acknowledged for their help in making this re- search possible: Joe Busto, Project Manager, Colo- rado Water Conservation Board, Flood Protection and Weather Modification Permitting Section; Steve Schmitzer, Becky Dechant, and Greg Bryant, Denver Water Operational Cloud Seeding Program; Ross Williams consultant to CWCB for GIS support; Larry Hjermstad of WWC who graciously provided us logs on seeding operations and enthusiastically critiqued the model output, taking a great deal of time at his own expense; and Brenda Thompson for assistance with manuscript preparation. This research was supported by the U.S. Bureau of Reclamation Weather Damage Modification Pro- gram, Denver Federal Center, Denver, CO 80225 under Financial Assistance Agreement No. 03-FC- 81-0925