WMOD00554

• Title: Seeding Summertime Convective Clouds to Increase Black Hills Rainfall
• Weather Modification - Doc Type: Report

r

which were often limited to the Black
Hills.

The static seeding approach was
followed. Ice crystal formation was
induced in supercooled cumulus clouds
for microphysical effects, with the
principal effect sought being an
increase in precipitation efficiency.
The experimenters were aware that
dynamic effects might be produced too,
but dynamic effects were not
deliberately sought. The seeding rate
was kept low to reduce the possibility
of overseeding. Several seeding agents
were used. By far the most common
treatment was the release of silver
iodide from an acetone generator on an
aircraft flying in updrafts below cloud
base. Sodium iodide was used as the
solubilizing agent. Only one generator
was used at a time, with the release
rate being approximately 300 g.h'l.

Seeding directions were radioed to
the pilot from the radar station east of
Rapid City. The objective was to treat
all promising cumulus congestus and
towering cumulus clouds that were over
or approaching the seed target area.
seeding was sometimes suspended when the
only target clouds available were large,
heavily glaciated cumulonimbus clouds.

3.2 Results
The Rapid project included 98
operational days, of which 27 were
shower days and 50 were storm days. A
more extensive breakdown is given in
table 2.

The Rapid Project was evaluated on
the basis of observations at a network
of almost 100 rain gauges. Under the
randomized crossover design, there was a
seed target area and a no-seed target
area each operational day. The rain
gauge data (table 2) show that the seed
target received more rain than the no-
seed target on shower days, but less
rain on storm days. statistical
analyses by Chang (1976) showed that the
differences on shower days were likely
real, with the p-values being 0.01 and
0.09 for southwesterly flow and
northwesterly flow days respectively.
However, the p-values for the storm
days, 0.35 and 0.61 for southwesterly
flow and northwesterly flow days
respectively, were too large to support
any conclusions about effects of seeding
on those days.

If one assumes that the differences
in rainfall patterns on north-seed and
south-seed days represent rainfall
increases in the seed target rather than

suppression effects in the unseeded
target, the Rapid Project results
foreshadowed t.hose obtained later on
Cloud catcher. That is, the most
promising clouds to be seeded for
pot.ential rainfall increases are cumulus
congestus and towering cumulus, and the
rainfall increases obtainable! on days
when such clouds predominate are very
large, when expressed as a percentage of
the natural rainfall on those, days.
Adding up the numbers in table 2 for
shower days pertaining to seed and no-
seed targets yields totals of 4.87 and
1.73, respectively. The ratio of those
two numbers is 2.8, which agrees
surprisingly well with the estimate of
2.3 obtained from the Cloud Catcher
data. One can hypothesize that the
Cloud Catcher cases correspond to the
shower days of the Rapid Project, and
that the storms on the Rapid Project
storm days are samples of a different
type of phenomenon, which is completely
outside of the shower size distributions
expressed by (3) and (4). While further
study might enable one to improve on
this assumption, it is not possible to
be more precise at this time.

Table 2*. - Average rainfall (in mm) per
gauge per operational day
and seed/no-seed ratios in
target areas for all days of
given type on Rapid Project

(Number of days by category f,ollows
north-area entries)

SW-flow

NW-flow

Shower days

North area, seed N 0.99 / 6 0.91 / 7
seed S 0.69 / 8 0.64 / 6
Seed/no-seed ratio 1.4 1.4
South area, seed S 1. 83 1.14
seed N 0.30 0.10
Seed/no-seed ratio 6.0 11

Storm days
Nor1:h area, seed N 2.18 /15 0.89 / 7
seed S 2.82 / 14 4.32 / 14
Seed/no-seed ratio 0.77 0.21
South area, seed S 2.36 1. 37
seed N 2.97 3.18
Seed/no-seed ratio 0.79 0.43

* after Dennis and Koscielski, 1969

\ve turn next to the possible effects
of seeding isolated convective! clouds or
cloud clusters upon total seasonal
rainfall. The indicated effect is
modest, in large part because of the
scarcity of shower days. The :Rapid
Project, which was in the field 6 days a
week from roughly June 1 to August 15

7