WMOD00393 - Laserfiche WebLink

Home Browse Search

III. RESULTS OF CLOUD SEEDING TO MODIFY PRECIPITATION 607 source, while aircraft dry ice drops produce cur- tains of ice crystals. The modeling work, coupled with actual trials with AgI crystals or tracers [e.g., sulfur hexa- fluoride (SF6)] has given some insight into how to seed clouds. When seeding agents are in- jected into clouds, seeding lines are normally separated by only 400 to 500 m, this being the maximum distance at which "filling in" of the seeding agent by turbulence can take place in a reasonably short time, for example, 5 min. The distribution of seeding agents from aircraft oper- ating below cloud base requires consideration of the time required for the seeding agent to dis- perse widely and affect most of the updraft. In some cases it is necessary for the seeding air- craft to operate several hundred meters below base and at several kilometers distance from the cloud being treated to allow for the spreading of the seeding agent. Seeding from the ground normally allows suffi- cient time for wide distribution. Typical plumes have angular widths of 15 to 300 and may mean- der as the result of fluctuations in wind direc- tion. The principal concern with ground-based seeding is that stable layers may prevent the seeding agent from rising to the supercooled re- gions of the cloud. While orographic lifting fa- vors ascent, orographic clouds present their own problems. The mountains themselves de- flect the wind; sometimes a barrier jet is ob- served blowing parallel to the mountain range. Trapping of seeding agents below stable layers just ahead of mountain barriers has been re- ported in some cases and may be a contributing factor to persistent (carryover) effects in cloud seeding. The upward transport of seeding agents from the ground is favored by an unstable atmosphere with well-developed convective currents. Field tests using SF6 as a tracer have shown that seed- ing agents from ground-based generators do reach convective clouds under unstable condi- tions. The distribution of the seeding agents within convective clouds is sometimes less than optimum regardless of seeding method. Portions of a cumulus cloud often show no seeding agent as much as 10 min after seeding from below cloud base or by dropping of pyrotechnics. Wind shear coupled with differential fall speeds of par- ticles in a seeding plume can be very effective in increasing the horizontal extent. Seeding cur- tains can be tilted to acquire a horizontal extent of 10 km or more in less than 30 min in a typical winter storm. III. Results of Cloud Seeding to Modify Precipitation A. OPERATIONAL SEEDING OF WINTER OROGRAPHIC CLOUDS Supercooled orographic clouds offer a unique opportunity for potential increases in precipita- tion. They persist for hours over mountain ranges, sometimes producing little or no precipi- tation, apparently because they do not contain enough natural ice crystals to snow efficiently. In addition, the lifting of the air on the upwind side of the mountain barrier suggests that oro- graphic clouds can be seeded cheaply and effi- ciently by AgI generators on the ground. The economics appear especially attractive in areas equipped to collect, store, and distribute the ad- ditional water produced. On the basis of these considerations, opera- tional programs to increase precipitation over mountains were established in the 1950s in many countries, including France, Morocco, and the United States. A typical project involved the de- ployment of 10 to 20 manually operated AgI gen- erators on the ground upwind to seed a target area of several thousand square kilometers dur- ing winter storms. Aircraft seeding was con- ducted on some projects to supplement the ef- fects of the ground-based generators or to reach otherwise inaccessible areas. Eventually remote control generators were installed on some proj- ects to improve coverage. B. STATISTICAL EVIDENCE FOR INCREASES IN OROGRAPHIC PRECIPITATION AND RUNOFF It was inevitable that, after several years of operational cloud seeding, an evaluation of the effects produced would be wanted. One of the most comprehensive evaluations was conducted by the Advisory Committee on Weather Con- trol, which was established by the U.S. govern- ment in 1953. The Advisory Committee's statis- tical analysis of orographic cloud seeding projects was based principally on linear regres- sion techniques applied to historical data for tar- get and control areas. For a given target area station (precipitation gauge), appropriate con- trol stations were selected based on exposure to prevailing winds, similarity of elevation, etc. Care was exercised to avoid selecting control stations that appeared likely to be themselves affectl~d by the seeding. Because precipitation data tend to be highly skewed, the Advisory