WMOD00500

• Title: Observations of Silver Iodide Plumes over the Grand Mesa of Colorado
• Date: 10/10/1988
• State: CO
• Weather Modification - Doc Type: Report

1142

JOURNAL OF APPLIED METEOROLOGY

VOLUME 27

by a near zero correlation coefficient. However, if the
two low spreading rate cases of 10 March (in altostratus,
accompanied by the highest shear) are ignored, then
the remaining seven cases (in stratocumulus) yield a
correlation coefficient of 0.73. Thus, the spreading rate
may be partially dependent upon wind shear, but ad-
ditional observations should be obtained to test this
supposition.
It was thought that convective processes interacting
with wind shear should enhance the plume spreading
rates. Though the supporting measurements were crude
and the relationships were weak, there was a tendency
for the data to confirm this expectation. Such experi-
ments need to be repeated with more direct and ac-
curate measurements of convective processes, such as
precise aircraft-measured horizontal and vertical winds
and turbulence, that were not available on the aircraft
used. Until then, these observations suggest an along-
the-wind plume spreading rate of over 2 m s -I in clouds
over the Grand Mesa.

b. Vertical extents and spreading rates

Attempts were made during a few experiments to
measure the vertical expansion rate of the plumes, but
in general they were not successful. On those days when
it was intentionally tried, the air (and plume) rose in
altitude to cross the mesa and descended on the other
side. It became too difficult to identify whether the
vertical edges of the plume resulted from diffusion or
from orographic uplift and subsidence.

c. Concentrations

As in the case of the ground seeding experiments,
important aspects of the aircraft seeding experiments
are the concentrations of AgI and of ice particles in the
plumes. The total IN counts per pass were taken to be
an indication of AgI plume concentration. The clear
air experiments were analyzed separately from those
with ice particle plumes. Though the scatters were large,
in both cases the total counts tended to decrease linearly
with time as the concentrations became diluted by
plume expansion. But unlike the ground seeding cases,
there was a definite reduction in AgI counts in the data
with ice particle plumes by a factor of over 7. This
indicates that the IN were scavenged by the water
droplets and ice particles, as discussed by Super et al.
(1988). It is not yet known why the data from the
aircraft-release experiments should indicate scavenging
and the data from the ground-release experiments
should not.
Combining all 46 passes with ice particle plumes,
the average concentration of ice particles larger than
0.1 mm was 15.9 L -(, with arange of 1.4 to 40.1. Over
the crest of the mesa they were 3.3 to 37.2 L -( , with
an average of 15.4 L -lover the 28 acceptable passes.
Such values are similar to those achieved with the

ground seeding experiments over both the Grand Mesa
and the Bridger Range. The generators of the AgI ice
nuclei were different, with different effectivenesses, and
the weather conditions also varied.
The production of similar concentrations of ice par-
ticles under differing weather conditions and differing
seeding techniques may indicate that the clouds tended
to limit production of ice particles (Fukuta 1985;
Hobbs and Rangno 1985), regardless of how many IN
were available, as long as there were enough IN to reach
some preferred concentration, which seems to be near
10 L -1.

8. Discussion and conclusions

Ice nuclei and the resulting ice particle plumes (if
any) were detected nearly every time by aircraft sam-
pling during numerous ground- and aircraft-releases
of AgI over the Grand Mesa of western Colorado.
Ground-released plumes were tracked for up to 40
km and up to about 80 min of advection time from
the seeding site. The average plume spreading angle,
approximating an instantaneous plume, was almost
always within a factor of two of the median of 150.
The instantaneous plume meandered through a wider
angle, ranging from 240 to 960 with a median of380.
Farther downwind the behavior of the ground-released
plumes yaried, and ranged from little additional
spreading to an expanding coverage to a bending pat-
tern, until they were lost in the region oflee subsidence.
There are suggestions that ground generators should
not be placed much lower than about 700 m below the
crest or the seeding materials may not be transported
over the Grand Mesa, but the best elevation limit has
not yet been determined. In the Grand Mesa context,
it would be prudent to place generators as far back
from the crest as possible (for maximum diffusion and
ice crystal growth and settling time), yet not below
about 2500 m. The optimum cross-wind spacing
should allow the merger of about 150 plumes by the
time they reach the crest. The local topography with
respect to the winds will obviously force compromises,
as will possible budget constraints. If possible, the winds
at the seeding sites as well as above the crest should be
monitored. Then only those generators whose plumes
would be likely to reach the intended target area could
be turned on.
The increasing IN effectiveness at higher altitudes
(colder temperatures) compensates for the decreasing
concentrations of AgI with altitude. Even so, since a
typical plume rises more than 500 m above the crest,
the warmer one-third to one-half ofthe winter storms
over the Grand Mesa are not cold enough for adequate
nucleation rates at the plume tops with AgI-NH41.
Similar results were suggested over the lower altitude
but more northern (colder) Bridger Range of Montana
(Super and Heimbach 1983). This problem might be
remedied by using greater release rates of AgI and/or