WMOD00412

• Title: Final Report on Utah Cloud Seeding Experimentation Using Propane During the 2003/04 Winter
• Date: 3/1/2005
• State: UT
• Weather Modification - Doc Type: Report

I

I

I

Figure 7 is a similar distribution using TAR wind directions. Case numbers per rating category vary
slightly between the two plots because of some missing data at both locations. Again, one outlier with a
poor rating is not plotted on the TAR figure. It had a wind direction of 191 deg. The remaining 59 cases
have rather similar distributions for all three targeting ratings. There is a tendency for a somewhat wider
range for the poor targeting, but differences are too small to use TAR wind directions for partitioning.
The north-south oriented Wasatch Plateau and the major canyon to the west resulted in all good and fair
cases being in the limited direction sector of 250 to 270 deg, based on the raw data. Virtually all available
cases were in that limited wind direction range.

I

I

5. PRECIPIT A nON MEASUREMENT CONSIDERA nONS

I

Any randomized winter orographic seeding program which hopes to demonstrate seeding effec:tiveness
in one to a few winters must rely on accurate high resolution precipitation measurements. In theory, very
long term programs could use measurements of the seasonal snowpack from snow courses, or spring and
early summer stream flow observations. But those approaches were impractical for the 2003/04 Utah
project.

I

I

5a. Precipitation Gauge Considerations

I

Obtaining accurate SWE measurements of snowfall is far more difficult than realized by most water
users, water resource managers or meteorologists for that matter. Snowboards left level with the
snowpack surface can provide accurate observations in sheltered locations. But such measurements are
very labor intensive and impractical for the short term snowfall data required by the Utah 2003/014
randomized program. The only reasonable method of obtaining short term SWE observations is with
recording precipitation gauges. Some recording gauges previously used and currently being marketed are
poor choices for snowfall observations of the time and amount accuracies needed by the Utah program.
One requirement is a gauge orifice diameter larger than the 8-in standard. Past use of Universal gauges
with low density snows common to high altitudes resulted in frequent "capping" episodes. Snow builds
up on the sloping gauge housing walls below the orifice and eventually covers the entire orifice without
falling into it. Snow may build a foot or two above the snow-covered orifice before falling in. The ability
of low density snowflakes to chain together and bridge over an 8 inch orifice is a surprising phenomenon
to view, but is not uncommon where winds are very light. The gauges used in the 2003/04 Utah project
preventing capping by having 12 inch orifices and no sloping side walls. That is, the entire gauge
housing was an upward pointed cylinder with thin orifice rim as is appropriate for SWE observations.
Capping was never observed with these gauges.

I

I

I

I

5b. Minimizing Wind-Induced Gauge Undercatch

I

The basic problem with using a gauge to measure precipitation in general, and especially snowfall, has
long been recognized as being wind-induced undercatch. When airflow encounters a gauge, it is partially
diverted around each side and partially up and over the gauge orifice, creating a local turbulent updraft.
The upward component of the moving air stream reduces the numbers of precipitation particles e:ntering
the gauge orifice, causing an undercatch which is related to the wind speed and also to the gauge size and
shape. The undercatch is much greater with snowfall than rainfall because snow particles are less dense
than liquid water particles and are therefore more influenced by moving air. Attempts to use a wind
shield to reduce gauge undercatch go back to around 1850 according to Warnick (1956) who further notes
that the first version of the well known Nipher wind shield was reported in 1878. Numerous
investigations into gauge catch of snowfall as functions of wind speed, various types of wind shi,elds and
gauge housing shape were reported throughout the 20th Century. The evidence is overwhelming that
gauges measuring snowfall should be sited in locations protected from the wind and should be equipped
with wind shields.

I

I

I

I

16

I