WMOD00297 - Laserfiche WebLink

Home Browse Search

longer lasting cloud. Good hydrometeor development generally requires a 20 minute lifetime in the cloud - and much longer if the cloud is diluted. For the stronger upcurrent cases, the larger starting sizes are required if the particle is to stay below the -40C slow-growth regime. Figure 10 is based in part on Kluzura's calculations. It summarizes some of these concepts, and for pre- cipitation initiation compares hygroscopic seeding to AgI seeding. It shows, for various upcurrents, how high a cloud must be for the growing hydrometeor to fall back down through the upcurrent rather than blow Ol)t the top. It shows the same for AgI. From the sm.all diagram on the ri'ght of Figure 10 it is evident that in Oklahoma many of the clouds fall into the category where hydro- meteors evolving from hygroscopic particles will grow well while tho~e from 'AgI (and from natural liquid embroyos from stochastic processes) start too late and move up too fast to grow large enough to fall through the upcurrent. The AgI may ma~e the cloud bigger and alter the situation, but that is not considered here. Hygroscopic seeding offers the chance of extracting a bigger percentage of water from a large cloud -- less is lost to the upper atmosphere in the form of cirrus. It also can extract water from smaller natural clouds, including ones which ar~ not supercooled. When considering the actual seeding of real clouds using real particle generating equipment, one is confronted with a broad s.pectrum of particle sizes. There is no way to know at this time the distribution of particles released by our spray system as used in Oklahoma. We may make some reasonable assumptions by assuming that of 1. 0 unit of liquid dispensed, 0.03 was near the 10 pm dia. size range, 0.04 was near the 20 pm dia. range, O. 10 was in the 30 pm dia. range, and O. 15 was in the 40 pm range. Note this totals only 0.34 of the material - - we assume 2/3 of the material is wasted, primarily in the form of inefficient, large droplets. The resulting multiplier considering the droplets distributed as given above is 4. 2. 106, if all become single 5 mm raindrops. The biggest contribution comes from the 10 pm size category. Using this derivation we should get about 4 acre-feet of water per kg of hygroscopic material, assuming perfect clouds. Hail growth and the Langmuir chain reaction increase this figure, while non-optimum matching of hydrometeor size to the upcurrent will decrease it, i. e., not every embryo will build a drop as large as 5 mm dia. Considering everything, getting 1 acre-foot of rain from 1 kg of spray seems to be a reasonable goal. It is instructive to hypothesize by several approaches how many embryos. are needed for practical seeding. The previous discussion was only con- cerned with individual, non-interfering particles. 24