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

by local terrain. The vertical separation between the HAS and TAR is about 300 m. The seeding plume
would be expected to disperse 300 to 600 m higher than the TAR according to considerable past airborne
plume tracking and modeling studies (e.g., Heimbach et al. 1998). Plume dispersion may be caused by a
combination of routine mechanical turbulence related to airflow interactions with rugged mountainous
terrain, sometimes present weak embedded convection, and the buoyancy caused by the heat of phase
change during seeded ice crystal growth as discussed by Fukuta (1996). Direct fallout of individual
seeded ice particles from the plume top to surface gauges is considered unlikely with the possible
exception of the most downwind target gauge, GDN. However, seeded ice particles could be carried a
significant vertical distance above the terrain downwind of the HAS and still fall into target gauges if a
significant growth mechanism was present. The most likely means of transporting higher level seeded
crystals onto surface gauges are: (1) by aggregation of many seeded crystals into snowflakes with much
larger fall velocities than individual crystals, (2) by natural snowflakes falling through the seeded crystal
plume, colliding with and sweeping out (scavenging) many of much smaller seeded crystals and carrying
them to the ground (also aggregation), and (3) by accretional growth (riming) of snowflakes containing
seeded crystals in the SL W rich environment which often exists in the cloudy air forced up the windward
slopes of the plateau. These aggregation and accretion growth processes could explain the apparent
seeding effects found in snowcourses only 4 kIn downwind of a high altitude Agl generator used during
the Bridger Range Experiment in Montana, reported by Super and Heimbach (1983), and from a later
operational/research program using the same AgI seeding site discussed by Heimbach and Super (1988).

I

I

I

I

I

Operations on a 24/7 basis are quite expensive if field technicians are required around the clock. This
was avoided by using an icing rate sensor at the HAS to determine the existence of SL W. A Goodrich
icing rate sensor, Model 0872F 1, was used to determine when SL W was present at the HAS seeding site.
Whenever the rate of icing exceeded a threshold, the next two hour block was "declared" an EU pair by a
programmed data logger no matter what subsequently happened to cloud and weather conditions. Of
course, SL W presence just prior to the beginning of the two hour experimental block provides no
guarantee that it will continue to be present throughout the period. However, the probability of continued
SLW availability, especially during the first EU of the pair, is much enhanced over any other available
approach of declaring an experimental period. It will be recalled that SL W must be available at the HAS,
or not far above it in the case of ice saturation at the HAS altitude, in order for seeding potential to exist.
Otherwise the embryonic seeding crystals will soon sublimate back into water vapor.

I

I

I

I

Once seedable conditions were measured by the icing rate sensor, and the start of a two hour
experimental period was declared, the same programmed data logger made a random decision to seed
either the first or second EU of the pair. The data logger controlled the three propane dispensers near the
HAS, releasing propane for exactly 40 min. All dispensers had two nozzles separated across the wind by
3 m. As with previous Utah and Sierra Nevada testing, the propane output per nozzle was approximately
3.5 gal h-I, equivalent to 1.9 gm S-I. According to previously cited laboratory and field investigations, the
expected ice crystal production per nozzle would be about 1012 seeded crystals per second. The center
dispenser was about 6 m northeast (downwind) of the HAS tower and the other two dispensers were about
30 m northwest and southeast (generally crosswind) of the center disperser. This configuration should
have reduced aggregation of embryonic seeded ice crystals near the dispensers, resulting in higher
concentrations downwind as the seeding plume approached target gauges. Based on earlier work
discussed by Holroyd and Super (1998), it was expected that seeded ice crystal concentrations at the TAR
would be approximately 50 to 60 L'(.

I

I

I

I

4. ANALYSES OF PRIOR PLUME TRACING AND SEEDING EXPERIMENTS

I

Important observations collected on the Wasatch Plateau during the 1990s were known to be of
significant interest for closer examination of HAS seeding plumes trajectories and their relationships with
wind observations. However, these data were believed to have been lost over the years until rediscovered
during the summer of 2004. The observations had never been analyzed in a manner suitable to address

I

7

I
I