WMOD00281

• Title: Application of a Simple Local-Scale Numerical Model in the Study of Altered Climate Impacts on Watershed Precipitation
• Prepared By: J. G. Medina
• Date: 1/1/1989
• State: AZ
• Weather Modification - Doc Type: Report

.

In the case of the Delaware River watershed,
the orographic effect is also evident (Fig. 6), but low-
elevation precipitation is more prominent than in
Arizona. This is principally due to the particular
configuration of the model for Delaware conditions.
Study of the monthly values for individual grid points
and gauge locations indicated four of the grid points at
the highest elevations had model precipitation
exceeding 500 mm (20 in), which might appear
excessive. .However, six gauges recorded more than
250 mm (10 in) of precipitation and one recorded
399 mm (15.7 in), despite the fact that gauges were
generally not located at the higher elevations.

Given in table 1 are the results from linear
correlation analysis (Pearson correlation) of daily
mean precipitation values, where the mean of
11 gauge site values is developed for 29 days of
January for Arizona, and, similarly, the mean of
24 gauge site values is obtained for 31 days for the
Delaware. Comparisons of the monthly gauge totals
with the respective model results indicate the model
appears to have good skill as, with local soundings, it
underestimated gauge monthly values by 18 percent in
Arizona and 3 percent in the Delaware. With MM4-
produced soundings, the model overestimated gauge
values for Arizona by 2 percent.

Table 1. Comparison of model results with gauge
group mean values (in nun)

Monthly
pcp

Pearson
correlation

Gauges (AZ)
Gauges (DEL)
Model (AZ local
(soundings)
Model (AZ-MM4
(soundings)
Model (DEL local
(soundings

125.5
226.8

102.6

0.82 (N = 29)
0.75 (N =29)
0.66 (N =31)

128.3

220.0

Cor,e1ation " :dues in table 1 suggest the
model has useful capability to estimate daily areal
precipitation, as values range from 0,66 (44 percent of
the variance explained) for the Delaware River
watershed to 0.82 (67 percent of the variance
explained) for the Mogollon Rim. Assessment of the
model's capability to estimate point values is currently
under study. In an application of the model to the
Atlas Mountains of Morocco by EI Majdoub (1989),
correlation values ranged from 0.7 to 0.9 with monthly
totals at single sites, with 20-month samples.
Correlation values near 0.75 were obtained for several
sites with daily precipitation, suggesting the model has
useful skill at estimating point precipitation.

4. SUMMARY

A precipitation model has been separately
adapted for estimation of winter precipitation at
10-km grid points in the Delaware River watershed
and the Mogollon Rim of Arizona. Correlation
analysis suggests that the model has useful skill at
'~stimating daily precipitation averaged over a number
of points on the order of a watershed. Work is
continuing to assess the capability of the model to
I~stimate point precipitation values in the two
watersheds under study,

Individual model runs take little computer
time, particularly on 33-MHz microcomputers or a
Icomparable machine. The model's capability to
estimate precipitation over a watershed, when
initialized from regional-scale model output, suggests
that it may be a useful tool in the study of global
climate change effects at the local level. Eventually,
initialization of the precipitation model will be
attempted with soundings interpolated from GCM
runs that simulate doubled-C02 conditions.

5. ACKNOWLEDGMENTS

This study was supported in part through an
Intergovernmental Agreement with the State of
Arizona, Department of Water Resources, and in
cooperation with the Geological Survey, Delaware
River Basin Climate Change Project. Data from the
MM4 regional model runs were obtained from
Filippo Giorgi and Gary Bates of the National Center
for Atmospheric Research. Arnett Dennis and
David Matthews of the Bureau of Reclamation
provided assistance in the manipulation of the MM4
sounding data.

6. REFERENCES

Anthes, R. A., 1977: A cumulus parameterization
scheme utilizing a one-dimensional cloud model.
Mon. Wea. Rev., 105, pp. 270-286.

, and T. T. Warner, 1978: Development
of hydrodynamical models suitable for air pollution
and other mesometeorological studies. Mon. Wea.
Rev., 106, pp. 1045-1078.

, E. Y. Hsie, and y, H. Kuo, 1987:
Description of the Penn State/NCAR Mesoscale
Model Version 4 (MM4). NCAR Technical Note,
NCAR/TN-282 + STR. National Center for
Atmospheric Research, Boulder, Colorado, 66 pp.

Deardorff, J. W., 1972: Parameterization of the
planetary boundary layer for use in general circulation
models. Mon. Wea. Rev., 100, pp. 93-106.