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

FINAL REPORT ON UTAH CLOUD SEEDING EXPERIMENT A TION USING PROPANE
DURING THE 2003/04 WINTER

I

March 2005

EXECUTIVE SUMMARY

I

A randomized winter orographic (mountain-induced) cloud seeding experiment was conducted on the
Wasatch Plateau of central Utah from mid-December 2003 through March 2004. The purpose was to
statistically test overall propane seeding effectiveness applied to stonn passages over a time period
approaching a winter season. Establishing whether propane seeding could produce meaningful net
snowfall increases from a population of several stonns was a significant advance in the development of
this potentially important seeding technology. Propane seeding has the important advantage in being able
create ice crystals within mildly supercooled liquid water (SL W) cloud, at temperatures just below OOC.
It is impractical to seed with silver iodide (AgI) at cloud temperatures wanner than about -6 to -80C
because of its large temperature dependence as effective ice nuclei. Therefore, propane seeding has the
potential to be an important adjunct to AgI seeding within this "warm temperature window" where SL W
cloud is often abundant in the Intennountain West and elsewhere during winter, but untreatable with Agl.

I

I

I

Previous laboratory and field investigations have demonstrated that expansion of liquefied propane in or
near SL W cloud or fog produces vast quantities of embryonic ice crystals by homogeneous nucleation. A
reviewer of Super and Heimbach (2005a), which summarized this final report, kindly pointed out that the
propane method was likely initially tested in supercooled mountain clouds as reported by Serpolay (1957)
in a French publication. English references to later propane work include Vardiman et al. (1971); Hicks
and Vali (1973); Kumai (1982); Super and Holroyd (1997) and Holroyd and Super (1998). Somewhat
similar work was reported by Fukuta (1996), using expansion of liquid carbon dioxide rather than
propane, released from moving surface vehicles. He noted that latent heat release enhanced mixing of the
seeding plumes, resulting in widespread SL W fog clearing. It is well known that supercooled cloud
droplets cannot continue to exist at temperatures less than about -40oC, but freeze into ice crystals.
Expansion of liquid propane is a convenient means of locally chilling the air well below that temperature
and, thereby, producing vast numbers of tiny ice crystals. Hicks and Vali (1973) reported that about 1012
ice crystals were produced per gram of propane at temperatures below -20C, based on both laboratory and
field work. Laboratory results by Kumai (1982) showed about an order of magnitude lower output. The
seeding-produced ice crystals will grow within SL W cloud, or within a limited vertical distance below
liquid cloud base where ice saturation exists. Snowfall can result when growth times are sufficient, as the
seeded crystals are transported by forced airflow up the windward slope and top of a mountain barrier.
The same forced airflow produces the orographic SLW cloud which is the "raw material" needed for
winter mountain-induced cloud seeding to work.

I

I

I

I

I

I

A limited number of prior physical case study experiments, and one in this report, have demonstrated
obvious seeding effects producing light snowfall rates. Such experiments have demonstrated that the
physical hypothesis for propane seeding is reasonably well understood. However, such case studies can
only be conducted during the infrequent periods with negligible natural snowfall in the presence of SL W
cloud. Even light natural snowfall, which varies rapidly over time, masks obvious seeding effects.
Moreover, it is not possible to accurately extrapolate from the results of a few case study experiments to
expected seasonal precipitation increases. Only a randomized experiment can reliably provide this
important infonnation. Conducting such an experiment was the logical next step in testing the
effectiveness of propane seeding.

I

I

I

The 2003/04 randomized experiment was limited in duration and target area size by available funding.
It was not designed to answer all remaining questions, in particular concerning the magnitude of seeding
effects more than about 6.5 Ian downwind of the seeding site. The gauge furthest downwind was only 2
Ian east of the plateau top's west edge because of targeting uncertainties with a single seeding site and

I

,I
I