WMOD00473 - Laserfiche WebLink

Home Browse Search

'\ 1994 JOURNAL OF CLIMAT~ AND APPLIED METEOROLOGY VOLUME 22 marked temperature dependence is believed to partially explain the results of the statistical analysis to be dis- cussed. . ' The generator at each seeding site was operated at approximately 30 g AgI h-1. A 3% (by weight) solution of AgI in acetone, complexed with NH4I, was used. Calibrated flowmeters, pressure settings and flame condition were monitored and recorded at 3-4.5 h intervals, day and night. The reliability of the seeding operations was high as the total downtime was only 9 and 24 h for the northern and southern generators respectively, out of a total of over 700 seeded h during the winter of 1970-71. Downtimes for the 1971-72 winter were estimated as less than 6 h at either sit~. 4. Transport, dispersion and deposition of the AgI The BRE gave particular attention to the generation, dispersion and transport of silver iodide (AgI) seeding agent. The question of proper targeting of desired con- centration of AgI has been,. and. continues to be, a serious uncertainty in most winter orographic seeding programs. For examples see Rottner et al. (1975), Reid (1979) and Miller and Hill (1981). It is believed that the BRE has some of the most convincing evidence of successful targeting obtained in a winter orographic program. This evidence will now be reviewed because of its importance in consideration of the suggestions of the statistical analysis. . a. Silver. iodide plume tracing Silver iodide plume characteristics over the Bridger Range have been reported by Super (1974). Some ad- ditional observations were reported in Part II. Con- sequently, only a brief summary of the plume tracing program and results will be given here. An NCAR ice nucleus counter was flown in a light twin aircraft on over a dozen different days, about half during February-March, 1971 and the rest during February-March, 1973. Flight passes were usually above the Main Ridge, but sometimes over the Bangtail Ridge as well. The counter was very similar to that described by Langer (1973) who reported quite good agreement between it and the CSU isothermal cloud chamber. All sampling missions were, of practical ne- cessity, flown under visual flight rule (VFR) conditions, usually while the atmospheric layer of interest was slightly stable. Diffusion would be expected to be greater than sam- pled under VFR conditions during storms with con- vection present. In order to assess the frequency of convection during Bridger storms, the convective in- stability iJOe!iJz of the 750-700 mb layer has been cal- culated from the BRE rawinsonde data. This layer should have been lifted with westerly flow over the Bridger Range. Therefore, cases with iJOe!iJz < 0 and cloud present at the bottom of the layer.should result in embedded convection which would enhance plume dispersion. Of the total of 363 available soundings examined, 80% were stable between 750-700 mb as indicated by iJOe!iJz> O. Further, when the averages of all available.- soundings for each experimental day were considered, only 10% were convectively unstable. It appears that convection was only present infrequently to increase the plume dispersion. Thus, the observations made under VFR conditions and in a slightly stable atmo- spheric layer should be a reasonable first approximation of actual plume conditions in most winter storms. It is believed that mechanical mixing was primarily re- sponsible for the plume dispersion. . Pairs of passes in opposite directions were normally made directly above the Main Ridge at pressure altitude. increments of -150 m (500 ft). These were flown from just above the surface to above the maximum altitude reached by the AgI plume(s). The known time delay in the counter's first response encountering AgI Was used to estimate the plume edge position for each entrance into it. All passes in a given vertical plane across the plume were sometimes used, together with wind data from pibais, to estimate the flux of AgI. Major findings of the AgI plume tracing program were as follows: 1) AgI plumes from both generators were usually identified moving over the Main Ridge and toward the intended target-the Bangtail Ridge. In fact, the plume. from the southern site, located fur- ther upwind, is believed to have been detected on prac- tically every pass ever made in the lowest 400 m above the Main Ridge, i.e., during 42 passes over 13 different days. 2) the AgI was largely confined to within -450 rri. (1500 ft) of the Main Ridge crestline, which placed the plume top at about the 700 mb level. 3) Plume widths from the southern generator were usually in the 10-300 range over the Main Ridge, corresponding to a crosswind distance of 1.7-2.7 km. Limited ob- servations over the Bangtail Ridge (see Part II) indicated further crosswind broadening of the AgI with widths between 4 and 5 km, but with little additional vertical ascent evident. Little data are available to calculate plume widths from the northern generator. 4) Mea7 sured ice nucleus concentrations above the Main Ridge typically ranged from 100 to 1000 per liter, effective at -200C. The values represent the NCAR counter raw data increased by a factor of 10 to compensate for known instrument losses (Langer and Weickmann, 1971). This was estimated to correspond to 10 to 100 per liter at the warmer temperatures prevalent within 450 m of the crestline. (5) Estimates of ice nuclei flux over the Main Ridge were in very good agreement with generator output as calibrated at the CSU Cloud Sim- ulation Laboratory. It is also noteworthy that the AgI produced by the Montana State University generators appeared to retain its ice nucleating ability for periods from one to a few