WMOD00268

• Applicant: CWCB
• Sponsor Name: USBR
• Project Name: Final Report Weather Damage Mitigation Program
• Title: Numerical Simulations of Snowpack Augmentation for Drought Mitigation Studies in the Colorado Rocky Mountains
• Prepared For: USBR - WDMP
• Prepared By: Curt Hartzell, Dr. William Cotton, Joe Busto
• Date: 9/1/2005
• State: CO
• Weather Modification - Doc Type: Scientific Study

30

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

The products generated by CSU for the project's Web site are created
using NCAR Graphics code lhat is distributed as part of the overall RAMS
package. This package analyzes data directly on the RAMS grid, without any
interpolation necessary (unless the analysis is on a constant pressure surface, in
which case there is vertical interpolation). On the precipitation maps the inward
poinling lick marks indicate the center of each 3 x 3-km grid cell. This method
uses the actual grid-poinl data, which allows an analyst to zoom in and look at
exactly what the model values are at a specific point. making troubleshooting
easier. The RAMS Real-time forecast graphics are generated using the GrADS
graphics package, which is in widespread use by lots of modelers (RAMS and
otherwise) because of its open sourcing and ease of using. With this package,
the RAMS data are horizontally interpolated to a latitude-longitude grid with about
the same resolution as the 3-km data. The area plotted on a GrADS Grid 3 map
is a little smaller than the actual RAMS Grid 3 domain (because of differences
between the RAMS Polar-Stereographic mapping and the GrADS latitude-
longitude mapping). But for all practical purposes, lhe resultanl fields are the
same regardless of which package is used.

The difference between the 24-hr simulated precipitation accumulations
from the NCAR Graphics and GrADS graphics packages is visually noticeable.
Figure 2.11 was generated using Grads. The precipitation accumulation is for
the 8-hr to 32-hr forecast period ending November 11, 2003 at 0800 UTC (0100
MST). Figure 2.14 is the simulated precipitation accumulation for the same 24-hr
period generaled after-the-fact by using NCAR graphics.

At the time the WDMP Colorado Financial Assistance Agreement was
awarded (October 2003), it was expected that the existing CSU RAMS PC
Cluster would be available for daily model non-seeded forecast runs until the new
PC processors were procured and the cluster expanded. However, due 10 delay
in finalizing the CWCB-CSU Interagency Agreement for the research project (not
signed until December 9, 2003), CSU was not able to purchase the new PC
processors on the schedule they had anticipated. Consequently, starting the
evening of December 1, 2003, CSU switched the real-time 0000 UTC RAMS
forecast cycle from their faster cluster to a slower one. This change was due to
impending deadlines of two other research projects where the use of the faster
cluster was critical; i.e., they required the greater memory and processing speed
of the faster cluster. The model-forecast products still got to the CSU Web site,
bul at a 3-limes slower rate than usual. (The slower forecast run took 12 to 14
hours of compuler time to finish.) The model-forecast output was still archived
on the CSU project Web site as usual.