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

36

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

This task was comprised of two parts, viz. selling up the MRBP code for the
evaluation, and then using the adapted code for the MRBP analysis. Results of
the MRBP evaluation of model performance are summarized in Subsection 3.7.

Dr. Paul Mielke provided a copy of the MRBP Analysis Code. His original
code was translated from FORTRAN?? 10 FORTRAN90 (a more widely used
version) and modified for the project by Dr. Gustava Carri6. However, the modified
program strictly follows Mielke's methodology. The updated version of MRBP
consisted of two codes (note - this brief summary does not intend to describe the
methodology associated with the MRBP analysis that is explained in detail in
Mielke et aI., 2001):

INPUT.f90: This code reads observational and simulated precipitation fields for a
series of stations. These input files (obse.dat, reglr.dat, and seed.dat) are written
in a free format matrix with rows and columns corresponding to days and
stations, respectively. The number of stations (N_STAT) and the number of days
to be considered in the test (N_DAYS) are input parameters. The code generates
input matrices in the format needed by the MRBP analysis for RAMS standard
simulations and for seeded runs (rglr_mrbp.ln and seed_mrbp.in, respectively).
These files consist of a series of blocks corresponding to different lest conditions.
An example of the format of each block of the file that1NPUT.f90 would generate
for three stations and four days is given as a comment at the beginning of
MRBP.f90. Each block is preceded by a line that tests conditions.

MRBP.f90: This code computes the test statistic and associated P-value of a
randomized block experiment. The program performs repeated MRBP analysis in
one operation considering different input parameters such as alignment within
blocks, distance function commensuration and rank tests (see Mielke, 2001).
While running the program, it gives the user the option to compare the
observation vs. the regular simulations or observation VS. the seeded runs and
automatically selects the corresponding input matrix. The outputs for the
comparison of the observation vs. the non-seeded and seeded runs are written in
two text files: seed_mrbp.out and rglr_mrbp.oul. respectively, The agreement
measure, P-value, as well as the value of delta, its expected value, the variance
and skewness are given for all above-mentioned options. This code performs
repeated MRBP analysis in one operation considering different test conditions
and therefore the corresponding results are also written in each output file.

2,10 Task 6 . Research study supervision and reports

The grant agreement rrom Reclamation for the Colorado WDMP required
three quarterly technical progress reports, a mid-project meeting between
Reclamation research staff and the Colorado WDMP Project Team, a presentation
on the overall project after project research studies and evaluations had been
essentially completed, and a final report.