* 'I '
UTAH WEATHER MODIFICATION PROGRAM
For the ninth consecutive
winter season, North American Weather
Consultants (NAWC) is conducting
a cloud seeding program in the
state of Utah. The primary purpose
of these operations is to increase
snowpack by seeding suitable winter
storms in the central and southern
portions of the state.
This program was initiated
in 1973 by the Utah Water Resources
Development Corporation (UWRDC),
a consortium of county commissions
and water districts. For two conse-
cutive years, the UWRDC totally
funded the project without outside
contributions; then in 1975, the
state of Utah, through its Division
of Water Resources joined the program
both financially and administra-
tively. Since that time the Division
of Water Resources has cost-shared
She project expenses and has provided
technical review of the operations.
During the history of the
Utah Cloud Seeding Program, the
operational design has varied somewhat
due to both financial restrictions
and technological advancements.
During the early years of the project,
limited funding allowed only for
minimal equipment deployment.
Specifically, ground-based generators
were strategically placed upwind
of high elevation terrain in southern
Concurrent with the state's
financial participation, the program
was upgraded to provide project
specific data. During the 1975-76
winter season an aircraft was added
to the program with a dual purpose
in mind. First the aircraft
provided upper air data such as
vertical temperature profiles and
estimates of cloud water content.
The aircraft was also used as an
aerial seeding platform to supplement
the ground generator network. This
program advancement was followed
closely by another operation:
strategy modification adoptior
of seeding criteria, as formulate
from results of research projects
(Thompson et al., 1978). Seedinc
criteria are basically check lists
of storm characteristics that indicate'
whether or not a particular storr
is suitable for seeding. Storr
characteristics that are investigated
for these seeding determinations
include: wind directions and speeds,
cloud water contents, cloud toc
temperatures, low level inversions,
With the adoption of thes<
seeding crite-ria, it was necessary
to upgrade the data availability
for decision making purposes.
Therefore, during the 1977-78 season,
upper air data were made available
to the decision makers through
the addition of rawinsonde sites.
These weather balloon operations
were totally dedicated to the project
and were operated on a three- tc
six-hour launch schedule during
storms. Information made available
to the project personnel through
these operations included temperatures,
relative, humidity, and winds fror
the surface up to 50,000 feet above,
Another major improvement,
to the program was the additit..
of high elevation, remote-controlle.
seeding generators. The principal
reason aerial seeding was utilized
during the 1975-78 periods wa
due to concerns that the seeding
material released from ground locations
under stable atmospheric condition-
would not rise high enough to reach
effective seeding levels. However
experience gained with aircraft
seeding on the Utah project suggested
a number of reasons for considering
an alternate approach to the problem
of effectively seeding an area
the size of the Utah target. A
reassessment of the utility of
aerial seeding for an area this
large (approximately 10,000 sq mi,
indicated that several aircraft
would be required to effectively
seed the entire target area. Moreover.
due to minimum flight altitudes
allowable over mountainous terrain,
aircraft frequently could not fly
at the desired temperature levels.
Additionally, recent research (Hill,
JA I ,1 1j 1 .,
- i M .
1980) suggested that frequently
tkh amount of vertical diffusion
downwind from an aerial seeding
source is limited.
An alternate approach to
effectively seed during these stable
periods was to locate manually
operated ground generators at high
elevations (above inversions).
This had been done wherever possible,
but much of the target area is
not populated during winter. Therefore,
NAWC developed a remote controlled
'seeding generator. The first remote
unit was installed in the Mineral
Mountains (north of Beaver) at
an elevation of 7000 feet during
the 1979-80 season* The unit was
controlled from NAWC's Salt Lake
City office via phone line and
microwave signals through a repeater
site at Frisco Peak (northwest.
of Beaver, Utah). The remote site
-was --als- -equipped wi t-k-srome bas io-
(wind and temperature sensors),
which could be interrogated from
the Salt Lake City office. Three
additional units were installed
prior to the 1980-81 season. Plans
or future years call for the addition
of four to six more units to the
Summary of the 1980-81 Season
The 1980-81 season's operations
were conducted from mid- November
through mid-April. During this
period there were a total of fifty
days on which seeding took place.
More than two thirds of the total
seeding was done during the last
half of January and the month of
March. These fifty seeding days
consisted of 30 distinct seeding
periods, lasting from 6 to 36 hours
each, in which storm characteristics
met the established seedability
criteria (Thompsqn and Shaffer,
1981). During these seeding periods
-a tot al ... f- nearly 2., 0I yL ou r s
of seeding generator operations
were accumulated, with a total
of 70 kg (154 pounds)of seeding
material being dispensed.
In April 1979, NAWC completed
a statistical evaluation of the
project by examining precipitation
data for an indication of seeding
effectiveness through a technique
known as a target and control evaluation
(Thompson, 1979). This technique
utilizes mathematical relationships
established during non-seeded periods
,between target and control areas.
Shese relationships are then used
co predict the precipitation in
the target area during the seeded
period and to compare those values
with the observed precipitation. Fig. 1
is a simplified schematic of the
-.- CONTROL --- --- TARWT---
Figure 1. Simplifeld form of trgt/oontrol ewituoion.
In this theoretical example, the historical average
for the control area Is 6.0". The precipitation
for the control area during the period In question
Is 6.4". Adjusting the target area historical awra
(6.7") by the above were observed control preclp
tatlon (.3") would predlt the targe ara preclpifon
to be 7.1". The actual target srea precipitation
was 7.5", therefore Indicating a difference of 0.4"
ovr what was predicted.
technique used in the target/control
evaluation. The results indicated
that January-March precipitation
had been increased approximately
17% within the target area during
the five-year period of 1974-78.
Evaluation of the 1974-81 Operations
S- --- ____-T-_1
--__ In NiALC'.s...f for..t.. mo n.tor
the success of the operations,
the evaluation was updated to include
data from the 1979, 1980, and 1981
project seasons. Using again the
target/control method, the eight-year
seeded period revealed an apparent
increase in precipitation.
The control area consisted
of six precipitation measuring
stations in west-central Utah and
five precipitation stations located
in Nevada and Arizona.
equations were developed for each
of four separate target areas (Central,
South Central, Dixie and East Tooele
County) and one supplemental area
(Eastern). These areas are shown
in Figure 2. The Eastern area
was not in the intended target
area, but was included to investigate
potential extra area effects.
I .. Ij, I A k JR L -L, I I 1 I
t* h UTAH
T East Tooele
The Eastern supplemental
area ratios of observed to calculated
precipitation were high, with generally
igh statistical significance,
suggesting a rather strong possibility
of a positive downwind effect.
The reader should be reminded
at this point that the preceding
discussion has been a gross simpli-
fication of the evaluation process.
Numerous independent calculations
were performed, using different
statistical tests on various combin-
ations of target and control compar-
isons. If more detailed discussions
are desired, the reader is encouraged
to contact a NAWC staff member.
Again, in simlified terminology,
the results of these evaluations
indicate the following:
1. Area-wide January-March
precipitation was increased
In the target area during
the eight-year seeding period
by more than 10%.
2. The data suggesting such
increases is highly statis-
CN-pre2. Schemaeticf tergt(s) enM coitroll wee.
I usei for the~ 1974-81 vratlotme
Separate linear regressions
were developed relating average
January-March precipitation (1956-
1973) in each of the target areas
to that in the control group. These
equations were then applied to
1974-1981 control area precipi-
tation to estimate the target area
precipitation without seeding opera-
While the indicated results
of seeding for each of the sub-sec-
tions are important, perhaps the
most important aspect is the apparent
effect of seeding in the total
Primary Target. The results from
the combined eleven station control
indicate an 11% increase, significant
at about the 3 percent level.
That is, there is, only a three
Pa recent likelihood that the results
Ibuld be attributed to chance alone.
in eastern Tooele County results
of six years of seeding indicated
increases from 7% to nearly 12%,
3. There is evidence of positive
downwind effects from seeding.
Thompson, J. R., 1979: Evaluation
of five years of operations
of Ia central and southern
Utah weather modification
program, 1974-78. MAWC
Report No. SBWM-79-1, NAWC
to Utah Water Resources
Development Corp. and Utah
Div. of Water Resources.
Hill, G., 1980: Dispersion of
iodide in winter orographic
clouds. JAM, Vol. 19, No. 8,
Thompson, J. R., R. W. Shaffer,
C. Wisner, and D. A.
Griffith, 1978: A design
study for a cloud seeding
program for the State of
tah. NAWC Report No. 77-15
to State of Utah, Div. of
Thompson, J. R. and R. W. Shaffer,
1981: Summary of operations
and evaluation of a cloud
seeding program in Utah.
NAWC report No. SLWM-81-1
to State of Utah, Division
of Water Resources.