WMOD00307

• Title: A Proposal for Cooperative Weather Research Between the Santa Barbara County Water Agency and the US Department of Interior Bureau of Reclaimation in the Context of the Weather Damage Modification Program through the State of CA 2002-2003 Winter Season
• Date: 10/17/2002
• State: CA
• Weather Modification - Doc Type: Report

.
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t

is likely to be. The crosswind sampling line end points will be adjusted in accordance with
updated wind observations.

Airborne sampling may sometimes be unsafe when strong winds and turbulence are present, and
missions may be terminated at the pilot's discretion. The pilot-in-command always has ultimate
responsibility for maintaining aircraft and crew safety.

2.3.1 Field Preparation

Preparation for data collection effort will focus on preparation of the WMI transport and
dispersion model for the Santa Barbara terrain, selection of preferred site for the release of
rawinsondes, pre-project preparation and calibration of the SF6 detector, preparation of the
research aircraft itself, and the development of a detailed field operations plan.

Transport And Dispersion Modeling
The WMI transport and dispersion model follows the approach of the "GUIDE" model initially
developed under the auspices of the Bureau of Reclamation during Project Skywater. The first
use was for the Sierra Nevada Cooperative Pilot Project (SCPP) in the 1980's (Elliott 1981).
Subsequent development of the model occurred in 1984-1985 (Rauber et ai. 1988), and the first
applications were for aerially-released lines of AgI seeding agent. The model was intended to be
a simple operational application that could be run in near real-time. Since SCPP, other versions
of the model have been applied; e.g., testing of propane seeding in 1993-1994 (Reynolds 1996).
A GUIDE-type model was selected for development and application by WMI because it can be
run rapidly on a PC in the field, and configuration for specific target areas is straightforward.

GU \ 9 ~
~ 1l1.

oJ-
w",1?

51) ~

The flow over mountainous terrain is modeled by arbitrarily dividing the lower atmosphere into
several flow channels, or layers. The channels' vertical thicknesses are a function of elevation,
with higher terrain producing vertical constriction. At some level and above, there is no chimge
of channel thickness in response to terrain. In the examples presented herein, this level was
assumed to be 650 hPa. The mass flux perpendicular the barrier is assumed to be constant for
each channel. This produces an increased cross-barrier wind component over higher terrain.
The flow component parallel to the barrier is assumed to be unchanged.

Each grid point has a representative elevation. If Ui,j is the wind component parallel to the up- or
downslope direction at grid point i,j, then,

UiJ = Uc (APc)(LlPi,jyl ,

(1)

where Uc is the mean wind for channel c, derived from a sounding, LlP c is the depth of the
initialized channels of the sounding, and LlPi,j is the depth of the channel at grid point i,j.
Equation (1) forces the flows within channels to meet the conservation of mass flux crite::1um
throughout the model grid. For all but the surface channel, LlP c has been set to 50 hPa. The
distance between adjacent grid points can be varied, but there are tradeoffs that effect model run
time, terrain smoothing, computer memory requirements, and resolution. Larger spa.cing
decreases run time, memory requirements, and resolution, and results in less detail in the terrain.
Smaller spacing increases run-time, memory requirements, and resolution, and allows more
Weather Damage Modification Program 31