WMOD00433 - Laserfiche WebLink

Home Browse Search

~ ~. r [ ~. " t I DECEMBEIl- 1978 LARRY VARDIMAN AND JAMES A. MOORE 1773 cated an increase in precipitation due to seeding, while Bridger and Jemez indicated decreases when LCTTI was between - 20 and -400C. Pyramid Lake and CENSARE showed a completely different seeding effect over the same temperature range. Since a number of physical factors, acting simultaneously, probably determines seedability, it is not surprising that projects which have different factors dominating should disagree under single-variable stratifications. Correlations be- tween the variables and precipitation were studied in an attempt to identify the basic physical factors which are dominant in orographic clouds. No single variable produced a correlation with precipitation greater than 0.50. Findings from the correlation study led to addi- tional studies based on multiple-variable stratifications. Some of the stratifications which have been investigated using this multiple-variable stratifications approach are discussed in Section 5. 5. Multiple-variable stratifications Multiple-variable stratifications were conducted on a combined set of data from six of the seven projects. . At this point in the study, Santa Barbara was dropped because of concern that computed cloud-top tempera- tures were unusually warm. This may be seen in Table 2, where the average LCTTI was -9.80e. Two reasons were evident for these very warm cloud tops: 1) soundings in Santa Barbara were launched into convective bands in an attempt to measure band variables and many - soundings could have exited the band through the side rather than the top or between bands inlower cloud decks giving erroneous estimates of cloud top and 2) the frequent presence of low-level moisture was not properly accounted for in the com- putation of the analysis variables. The computer program occasionally calculafed a cloud top for the low stratus layer rather than for the higher cloud layer. A total of 1248 cases for the crest group was available for study from the remaining six projects, Data from the six projects were combined in the study, The justifica- tion for combining the projects rests on the assumption that the variables selected for stratification adequately describe the major physical processes which determine seedability in all six projects, The results of this study have shown this assumption to be essentially correct. Initially, all projects were combined and the Wilcoxon test was run without stratification of variables. Table 4 summarizes the crest group statistics for the total sample of six projects with no stratifications. Assuming that the seed/no-seed ratio is correct and is typical of the seeding effect expected over mountains of the western United States, seeding without regard to meteorological variations from storm to storm or topographical differences from location to location would produce a 34% increase in precipitation, This increase is very high and suggests either bias or a very strong positive seeding effect. One source of the large TABLE 4. Statistics for crest without stratification. M and N are the number of seed and no seed events, respectively, "Zeros" are the number of zero precipitation events and "mean" is the average 6 h precipitation [mm(6 h)-I]. "Ratio" is the ratio of mean seed precipitation to mean no seed precipitation, The "test statistic" is computed using the Wilcoxon nonparametric two-sample test, "P value" is the probability by chance alone of having a test statistic value more extreme in either direction than the actually observed test statistic value. Projects All Precipitation group Crest Stratifications None M + N = Total number of cases 1248 Seeded Nonseeded M=499 N=749 Zeros = 51 Zeros = 145 Mean= 2,84 Mean= 2,12 Ratio (seededfnonseeded) = 1.34 Test statistic =4.956 P value<O,OOI ratio will be shown to be CENSARE. The determina- tion of whether this result is due to bias in the data or a real seeding effect has not been made. In Table 4 the number of zero precipitation events is also tabulated for the seed and no-seed cases. Nineteen percent of the no-seed cases are zeros versus only 10% for the seed cases, suggesting that seeding causes nonprecipitating events to begin precipitating, The primary mechanism for initiating precipitation in winter orographic clouds is the presence of ice crystals, and the main factor which determines the presence or absence of natural ice crystals is probably the cloud top temperature. Silver iodide seeding generators produce ice nuclei very inefficiently at temperatures warmer than about -Woe. Therefore, a greater difference would be expected in zero events between seed and no-seed cases when seeding is conducted in the range of cloud-top temperatures colder than -Woe. Table 5 shows stratifications over three ranges of cloud-top temperature, The percentage of zero events for the no-seed cases is 33, 21 and 10% for the three temperature ranges, respectively, This shows that warmer cloud tops for no-seed cases have a greater occurrence of zero-percipitation events. The percentage of zero events for the seed cases is 28, 10 and 5% for the three temperature ranges, respectively, At tempera- tures warmer than -lOoC, seeding only slightly reduces the number of zero events, probably because of the relatively poor seeding effectiveness at warm tempera- tures. At temperatures between -10 and -30oC and at temperatures colder than -300C, seeding reduces the number of zero events by half. It was noted earlier that the seed/no-seed ratio for all six projects at the crest appears to be too large as a possible result of bias. Therefore, the statistics were recomputed eliminating each project, one at a time.