WMOD00412

• Title: Final Report on Utah Cloud Seeding Experimentation Using Propane During the 2003/04 Winter
• Date: 3/1/2005
• State: UT
• Weather Modification - Doc Type: Report

I

I

cloud temperatures were definitely too wann for AgI to be effective. Consequently, prior experimfmtal
propane seeding in Utah was limited to a narrow and wann range of supercooled cloud temperatures, with
high altitude seeding site temperatures between -004 and -3ADC. Ice crystal growth rates are slow in this
range so impacts on snowfall would be expected to be limited. Some clear evidence was provided of
propane seeding effectiveness in producing light snowfall rates. But the high variability of natural
snowfall in time and space frustrated most attempts to directly detect a seeding signal (Holroyd and Super
1998). They concluded that obvious seeding effects were limited to periods with none to trace natural
snowfall while periods with even light natural snowfall usually masked any seeding signal. It is not
reasonable to extrapolate the results of a few obvious experiments, obtained in the virtual absence of
natural snow, to seasonal snowfall changes. Yet the water user community is interested in seasonal
increases in the snowpack and more directly in resulting increases in runoff.

I

I

I

The primary goal of the 2003/04 experimentation was to provide better documentation of propane
seeding effectiveness over an entire winter by conducting a randomized statistical seeding program.
Another goal was to expand that documentation across the entire range of SL W cloud temperatures found
near the surface of Utah's mountains during winter, not just the very mildly SL W clouds previously
experimentally seeded.

I

I

Liquid propane expansion can produce meaningful ice crystal concentrations at temperatures as wann
as -OSC, but production rates are limited until the cloud temperature is _2DC or less (Hicks and Vali
1973). In effect, seeding by propane expansion is seeding with ice crystals because they fonn in vast
numbers immediately downwind of the dispenser nozzles. Water vapor in the air is cooled by the
expanding propane to force condensation of water vapor into very high concentrations of tiny cloud
droplets which immediately freeze. Freezing of existing cloud droplets also occurs, but at much lower
concentrations. Propane dispensers must be located well up windward mountain slopes to be in SL W
cloud, or just below SL W cloud base where ice saturation exists. Else, the tiny seeding crystals will
quickly sublimate back into water vapor. High elevation releases have a major advantage in largely
eliminating the targeting problem common with winter orographic cloud seeding projects as discussed
below.

I

I

I

It has been recognized for many years that achieving adequate transport and dispersion of the
commonly-used AgI seeding agent is the major problem in seeding winter orographic clouds (Rangno
1986; Reynolds 1988; Super 1990; Warburton et al. 1995). Perhaps Smith and Heffernan (1967) stated it
best when they said that, " --- persons who release silver iodide from the ground would be wise to tind out
where it goes." This seemingly obvious advice has seldom been followed. Failure to document that
clouds are actually being seeded continues to seriously hamper the development of this promising
technology.

I

I

I

Numerous plume tracing studies over different mountain barriers have shown targeting to be routine
with high elevation releases (Super 1974, Holroyd et al. 1988, Super and Heimbach 1988, Holroyd et al.
1995, Sec. 4 of this report). Moreover, high elevation propane seeding can initiate growth of ice crystals
near cloud base, even when SL W temperatures are only slightly supercooled. This approach not only
provides a much larger "temperature window" of seeding effectiveness, but maximizes crystal growth
times. Ground-released AgI plumes would need to be transported to higher, colder cloud levels to create
ice crystals, providing a much smaller "temperature window" and usually briefer growth times if such
plumes are transported along windward mountain slopes. There are conditions where gravity waves may
loft valley-released AgI plumes to sufficiently high and cold altitudes upwind of mountain barriers
(Heimbach et aI., 1997; 1998). But gravity waves are transient in space and time, and are difficult to
predict. It has yet to be demonstrated that they provide a means of routine vertical transport for AgI
releases in intennountain valleys or foothills. Gravity waves can provide an important source of SL W
cloud over valleys between parallel barriers (Bruintjes et aI., 1994; Reinking et aI., 2000), but the best
seeding strategy for that SL W is likely the use of generators on the windward slope of the upwind barrier.

I

I

I

I

3

I
I