OCR Text |
Show the Buffalo Pass circle, significant at the 5% level. Assuming that the values of ~which were found for the areas outside the test circl~,. bu~ still i_n the proper direction sector, represent the natural unseeded prec1p1tat10n ratio, and computing the average precipitation rates for each sample, the average seeding - produced precipitation rates were computed as (a) Buffalo Pass circle, O. 018 in hr- 1 (82% increase over natural), and (b) Rabbit Ears circle, O. 003 in hr-1 (20% increase over natural). Since weighing bucket data generally under- e stimate the amount of precipitation, computed percentage values ar e more apt to be realistic than absolute rate values. The larger computed percentage increases at Buffalo Pass may be evidence of (1) a more inefficient natural precipitation process under conditions of wa.rm southwesterly flow aloft, (2) artificial snowfall by precipitation initiation in the warmer lower layers, or ( 3) insufficient data to obtain a true estimate of the quantitative seeding effect. The large difference between and ! /NL values suggest that significant snowfall (of the order of 0. 015 in hr-1) may be produced by seeding when only light sno"wfall (0. 003 to 0. 006 in hr-1) is occurring naturally in the unseeded area. Thus, the . period of significant snowfall may be prolonged by seeding. 5. 6 Reliability of Delivery Systems Ground generator systems are very reliable in operation. If the flow rate is metered and monitored, the output is known to a high confidence level. From this standpoint, and because they are economical, ground generators are desirable. The main problem with ground generators is in the reliability of producing a desired concentration of material in a given volume at the desired altitude. Unfortunately, with ground generators in mountainous terrain the dispersion and transport characteristics often change rapidly, so that the exact treatment given a cloud volume is largely beyond controL The most serious complicating factor is the presence of a complicated inversion structure, which is typical of any high _mountain-valley system. An inversion above the level of release will often completely inhibit vertical transport of nucleating material, effectively putting a lid on any release. These trapping inversions typically occur up to a height of at least half of the mean terrain height of the main barrier, but there are many variations. Often, sites for release of nucleating agents at the proper upwind distance and direction are not available above the typical inversion height. This is particularly true of high yield precipitation areas, since the heavy precipitation is usually the result of an absence of high terrain along the upwind vector. Even if suitable high release sites are available, there are so many variations in the dispersion characteristics in mountainous terrain that it is difficult to adequately describe the dispersion, much less to control the dosage and treatment of a given cloud system. Any realistic model would require 312 |