WMOD00508

• Title: Physical Response of Winter Orographic Clouds over the Sierra Nevada to Airborne Seeding Using Dry Ice or Silver Iodide
• Date: 4/4/1990
• Weather Modification - Doc Type: Report

APRIL 1990

DESHLER, REYNOLDS AND HUGGINS

327

1986 was the first conducted after the OTM had been
updated to use soundings at both the base and crest of
the barrier for its windfield calculations, instead of only
one sounding at the base of the barrier. In these two
cases both OTM calculations and intermediate obser-
vations by aircraft and radar allowed precise timing of
seeding effects to KGV.

I'

5. Summary and conclusions

Experiments have been conducted over several win-
ter seasons in the central Sierra Nevada of California
to directly m~asure the chain of physical events that
result from seeding clouds by aircraft. A trajectory
model, that continually evolved during the course of
these experiments, was used to position the seeder air-
craft such that seeded snow crystals were likely to fall
at a fixed site where intensive surface observations were
collected. Clouds were selected for seeding based on
the amount and temporal duration of supercooled liq-
uid water. Several years of survey measurements in-
dicated that post-frontal stratus and stratocumulus that
form due to orography were the optimum clouds. Of
a total of 36 experiments, the complete chain of events
expected from seeding was documented by all mea-
surement platforms on only two occasions, and even
these results are not unequivocal. To illustrate the re-
sults from these experiments these two case studies,
and an overview of all experiments, have been pre-
sented.
The goal of measuring the chain of physical events
resulting from seeding was only marginally accom-
plished. A general result from these experiments is that
it is difficult to consistently measure effects from seed-
ing clouds. The research aircraft detected seeding effects
in only 35 percent of the seedlines sampled. Facilities
used for measurements below the aircraft altitude were
even less successful. The success rate of the radar was
4 percent; however, this was anticipated since seeding
effects would only be apparent in non echoing clouds
because of the small particles produced by seeding.
Seeding effects at the surface were only detected in light
precipitation, and were difficult to separate from the
background because of the small precipitation increases
and short time period for effects. Even with a highly
sensitive continuous recording device, the aspirated
2D-C, and an optimistic view towards the data, seeding
effects were measured in only 17 percent of the seed-
lines. Precipitation network data was not adequate to
resolve seeding effects even with gauges sensitive to 0.1
mm in 5 min. For 12 of the fixed-target experiments
data were available from all measurement platforms,
and seeded precipitation was predicted to arrive at the
fixed target. In only the two cases presented here was
the complete chain of events considered documented;
however, three out of the last six seeded cases of the
3-year experiment provided the best results. This could
be just a matter of circumstance, or it could be a re-

flection of increased expertise by the field personnel as
the various measurement platforms were more skill-
fully applied.
A variety of reasons contributed to the poor success
at measuring seeding effects. The liquid water content
in the target clouds was generally less than 0.1 g m -3
and variable along the seedline, thus limiting the mag-
nitude of seeding effects. High natural particle concen-
trations in the clouds masked seeding effects from the
aircraft. Seeding effects were most easily observed in
clouds with low concentrations of natural ice and rel-
atively high concentrations ofliquid water. The logistics
of coordinating the various measurement platforms in
shallow clouds over a mountain barrier was difficult
and sometimes led to long delays in initiating seeding.
Thus, some optimum periods for conducting seeding
were missed.
Even though the chain of events resulting from
seeding was completely measured on only two days,
the effects of seeding were similar on these two occa-
sions and provide some insight into the effects that
seeding may produce. Key results from the two case
studies were: 1) Seeding produced small (< 1 mm)
graupel that reached the surface after about 40 min.
2) Seedlines dispersed at 1 m s -I, leading to increases
in snowfall at the fixed surface target which lasted for
< 10 min. 3) Precipitation rate increases attributable
to seeding were at most 0.1 to 1.0 mm h -1.4) On radar
seeding produced increases in areal coverage of 1-10
dBZ echoes. 5) The operational targeting model, using
upwind and barrier crest soundings, was adequate in
positioning the seeder aircraft so that seeded precipi-
tation fell at the surface target. 6) The targeting model
estimates of seedline advection were consistent with
seeding effects observed in the radar and surface mi-
crophysical measurements. 7) Seedline advection cal-
culations, based on aircraft measured winds, were con-
sistent with seeding effects identified in the aircraft data.
8) Seeding at -60C with a 3 percent solution of AgI
NH4I NH4Cl04 produced ICC > 100 L -1.9) Seeding
with a line source of AgI and a vertical curtain of CO2
on different days produced nearly identical results at
the surface.
From these experiments several points pertinent to
cloud seeding operations are worth mentioning. One
problem limiting the amount of additional water that
can be obtained from seeding is the difficulty of treating
a large volume of cloud (Hill 1980b). Measurements
here suggest that seedlines disperse at 1-2 m s -I, and
that seeded precipitation falls out after 40-50 min. For
a 37 km long seedline that amounts to treating 100
km 2. Achieving fairly continuous coverage along the
direction of seedline advection requires seedlines to be
no longer than 37 km, yet treatable cloud may extend
for hundreds of kilometers along the barrier. Additional
coverage can only be achieved by using more than one
seeder aircraft per seeding mission.
Another point worth highlighting is the similarity of