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

;.

,^

t

APPLICATION OF A SIMPLE LOCAL-SCALE NUMERICAL MODEL IN TIlE STUDY
OF ALTERED CLIMATE IMP Acrs ON WATERSHED PRECIPITATION

J. G. Medina
U.S. Department of the Interior
Bureau of Reclamation
Denver, Colorado U.S.A

1. INTRODUCTION

Prospects of global climate changes over the
next few decades have led to enhanced interest in
climate modeling. The Bureau of Reclamation
operates numerous water storage facilities and
hydroelectric dams and thus is concerned about
climate change effects that may occur on the scale of a
watershed.

Hydrological processes can amplify the effects
of altered precipitation that may occur from climate
changes. Consequently, there is substantial interest in
the study of effects on the local scale of 5 to 50 kIn.
Currently available general circulation models
(GCM's) utilize grid scales on the order of 200 to
400 kIn (e.g., Washington and Meehl, 1989).
Regional-scale models developed for climate studies
employ scales on the order of 50 to 100 kIn (e.g.,
Giorgi and Bates, 1989). While the regional-scale
models substantially improve the description of effects
of features such as mountain chains over the GCM's,
there remains the need to study precipitation at the
watershed scale in order to study climate change
scenarios on, for example, reselVoir management.

This paper discusses the application of a
simple local-scale model to two geographically and
meteorologically very different areas to diagnose
winter precipitation on the scale of a watershed.
Specifically, the model is separately adapted to the
Delaware River watershed and the Mogollon Rim of
Arizona. Additionally, in the case of the Mogollon
Rim, input sounding data required by the model are
obtained from sites that are part of the national
rawinsonde network and, in a separate study, from
output from the Pennsylvania State University /
National Center for Atmospheric Research regional
meteorological model (MM4) (Anthes and Warner,
1978; Anthes et al., 1987; Deardorff, 1972; Zhang and
Anthes, 1982; Anthes, 1977).

2. MODEL DESCRIPTION

The model of interest was originally
developed by J. Owen Rhea in the 1970's (Rhea,
1977). Unlike complex orographic cloud models that
attempt to simulate complex cloud processes and
consequently require considerable computer resources,
Rhea's model attempts to simulate only the most
important physical processes that account for much of
mountain winter precipitation.

Advantages of such a simple model include ease of
performing many iterative runs on relatively
inexpensive microcomputers or work stations and
operational applicability for hydrometeorological
purposes. Additionally, simple orographic
precipitation models generally require only routinely
collected atmospheric sounding data as input.

The Rhea model is two-dimensional, steady-
state, and multi-layer. It accounts for moisture flow
from any direction and terrain effects such as rate of
rise and "shadowing" by upstream barriers.

The model requires a topographic grid unique
to each wind direction. Thus, 36 separate grid arrays
of topography are employed, one for each 10" intelVal
of the compass, at a selected grid spacing that in the
current study was set at 10 kIn. The wind direction at
the 850- or 700-mb level at the center of the
geographical area of interest is employed to select the
particular grid array to be used for model
computations.

The model keeps track of condensate
formation or evaporation as air layers experience
vertical displacements in interaction with the
underlying topography. As the model is multi-layer,
precipitation falling into a subsaturated layer will
partially or totally evaporate. Precipitation developed
at higher layers reaches the ground if it is not fully
consumed by evaporation.

Simple formulations are included in the model
to account for blocking at bw layers and streamline
vertical displacement. Enhanced streamline
displacement over the highest terrain is used to
simulate effects of convection, bUI the model cannot
simulate convective cloud processes in any rigorous
way. When atmospheric sounding data are available
for three or more locations, the model can estimate
divergence and include some compensation for large-
scale vertical motion.

The model employs a spatially constant
precipitation efficiency. The relationship, E = - kTo ,
where E is the efficiency, k is a positive constant, and
To is the temperature in degrees Celsius of the highest
layer with relative humidity equal to or greater than
65 percent, was found to give satisfactory results
provided E was not allowed to exceed 25 percent.