Weather Mod Critical Issues Report

• Project Type: General OWC
• Project Name: Weather Modification
• Title: Critical Issues in Weather Modification Research
• Date: 1/1/2004
• Water Conservation - Doc Type: Final Report

-

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

expense of a full-bin-resolving model and can readily be implemented in three-
dimensional storm models.

To better model the precipitating ice, Farley developed a hybrid method that
utilizes twenty categories (now twenty-one), or bins, for these particles, The sizes range
in diameter from 100 ~m to 5,0 cm (recently increased to 7.0 cm by adding the extra bin).
Bulk-water microphysical methods are used for the cloud liquid, cloud ice and rain fields,
hence the hybrid terminology, The dynamic framework for the microphysics has been a
two-dimensional, time-dependent cloud model and a three-dimensional, time-dependent,
cloud-resolving mesoscale model developed by Clark (1977,1979), Clark and Farley
(1984) and Clark and Hall (1991). The lAS two-dimensional framework has been used
to simulate hail formation and fallout in an Alberta hailstorm (Farley, 1987) in both
seeded and unseeded conditions, and in a North Dakota hailstorm (Farley et aI., 1996,
2004a,b). Good agreement with radar observations was obtained in the Alberta and
North Dakota hailstorms, This model framework allows the type of hailstone embryos,
either frozen raindrop or graupel, to be identified (Kubesh et aI., 1988), A critical
component of the Kubesh study was the data provided by the armored T-28 aircraft
involving particles types and sizes inside the strong updrafts of a supercell storm. Both
model and observations indicated the importance of shedding from graupel and hail
particles to produce rain for fallout and for hailstone embryos in the rich supercooled
liquid water environment.

In addition, the three-dimensional Clark model has been used to simulate snow
and rainfall over the Black Hills of South Dakota and Wyoming (Farley et aI., 2000)
during a four-day storm period, Simulation of the cold precipitation period produced
reasonably accurate precipitation patterns, but not as accurate for the warmer, weakly
forced situation, This last result indicated that a simulation of a drizzle process should
improve the rainfall comparisons, as was also called for in the NRC report. A simulation
of ground-generator cloud seeding of the storm system was reported in Farley et al.
(1997), which showed that the cloud seeding was effective on only one of the four days.
Other orographic simulations of precipitation formation and the effects of cloud seeding
are found in Meyers et al. (1992, 1995).

Some success has also been obtained in the three-dimensional modeling of
convective clouds and storms. The 3D cloud-resolving model of Clark (with the bin ice
microphysics of Farley) has been used to simulate hailfall in northern Italy (Wobruck et
aI., 2000) and in southern France (Wobruck et aI., 2003). This last study showed good
agreement with observations of the hailstone spectrum at the ground, Johnson et aI.,
(1993) simulated the 2 August 1981 CCOPE supercell with both liquid water
microphysics (L WM) only and with a hail category version (HCM) of the model (similar
to the hail formulation of Farley). The ice microphysics was shown to be important for
the better comparison with the actual storm, but the L WM simulation reproduced some of
the important dynamics of the supercell storm, The T-28 armored aircraft provided
critical information from inside the storm that was used for the comparisons. Farley et al.
(1992) used the 3D Clark model with bulk-water microphysics to simulate a moderate

26