WMOD00442

• Title: A Model of Hygroscopic Seeding in Cumulus Clouds
• Date: 12/12/1978
• Weather Modification - Doc Type: Report

DECEMBER 1978

GERARD E. KLAZURA AND CLEMENT J. TODD

1765

14

12

10

em =23.20C
BASE = I k m
--'5'C M.R.~16.2g kg-I

CLOUD BASE 200e

'0

15

20

25

UPDRAFT (m 5-1)

FIG. 7. Variation of the minimum cloud depth required for
a particle to grow large enough to begin to fall through the
cloud as a function of updraft speed for initial hygroscopic
drop diameters of 40 to 200 I-'m, naturally formed hydrometeors,
and ice particles formed through the nucleation of AgI nuclei
in a warm, moist cloud (model computations).

discussion to consider the onset of natural precipita-
. tion to evolve from particles that can be approximated
by 10-20 ~m hygroscopic seeds released beneath cloud
base (they grow to about 50-70 ~m by the time they
reach cloud-base region).
With this concept in mind we can investigate the
height at which precipitation particles become large
enough so they are just balanced by the updraft for
various hygroscopic seeds, for large drops evolved
from continental clouds, and for ice particles grown
following the nucleation of AgI nuclei.
Calculations used for growth of ice particles from
nucleation of AgI are described by Smith et at. (1974).
Time and height at which precipitation particles form
may be critical factors to the efficiency with which
a cloud precipitates. The smaller the cloud depth
required for the precipitation particle to begin to fall
through the updraft, the less time it takes for rain
to begin and the better the chances for the Langmuir
chain reaction mechanism to occur.
We consider a warm, humid air mass with clouds
containing (naturally) 50-100 ~m size particles in the
base. One might expect hygroscopic seeding to be
more effective in producing colloidal instability earlier
and lower in the cloud than either natural cloud
processes or through ice-phase (AgI) seeding. Fig. 7
verifies this conceptual" feeling."
The height at which a precipitation particle be-
comes large enough so it is just balanced by the
updraft is given for hygroscopic seeds (40, 100 and
200 ~m initially), for large drops evolved naturally

30

from a continental cloud, and for ice particles grown
from AgI nuclei. Notice how much lower in the cloud
the large drop is formed in the hygroscopically seeded
case. A similar graph is illustrated in Fig. 8, except
the cloud conditions here are much drier and colder.
Nucleated AgI particles require less cloud depth than
the natural process for updraft speeds greater than
2 m s-r, and only slightly greater depth than 40 ~m
hygroscopic seeds for updrafts between 3 and 12 m S-I.
Thus, as would be expected, as the cloud base gets
colder, the ice-phase mechanism for initiating colloidal
instability becomes more efficient.
The variations in growth rate for the three cate-
gories just discussed have been observed to occur in
real clouds. Dennis and Koscielski (1972) have analyzed
first radar echoes in salt seeded, AgI seeded and un-
seeded clouds in South Dakota, and found quite a
variance in the height above cloud base of the first
echo between the three categories. The median height
above cloud base was 3353 m for unseeded clouds,
2103 m for AgI seeded clouds and 1615 m for salt
seeded clouds.

b. A cloud seeding case study

A case of rain apparently being produced as a result
of hygroscopic seeding occurred on 23 July 1970,
near Rapid City, South Dakota (Biswas and Dennis,
1971). The cloud system that produced the spec-
tacular results was observed visually, photographed
and scanned by radar (with signals digitally recorded)
before, during and after the rainshower. A best esti-
mate of the internal cloud profiles was predicted by

,;

BASE O. C

,.

'2

'0

em = 16. C
BASE = 3km
M.R.~S.9g kg-I

>-
:I:
~
W
:I:

--IS.C

.

_-50 C

10 15
UPDRAFT (m ,-')

25

20

FIG. 8. As in Fig. 7 except for colder, drier cloud.

</1