WMOD00472

• Title: Estimation of Cloud Liquid Water in Winter Storms on the Mogollon Rim
• Date: 5/1/1993
• Weather Modification - Doc Type: Report

was that in a future effort, estimated CLW values could then be used with corresponding
precipitation data to estimate PE in winter storms on the Mogollon Rim. Meaningful cloud
Eifficiency estimates would be quite useful in the determination of cloud seeding suitability.

1rhe availability of radiometer measurements of CLW at HJ for a major portion of January
through March 1987, along with companion soundings from CVR, and an available 10-winter
sounding record for DRA, INW, and TUS that included the 1987 period, set the stage for this
study. Two approaches were used, one more statistical in nature and the other involving the
application of a one-dimensional model that estimates CLW content. Scatterplots were
developed between sounding variables for the various sites and between sounding variables and
HJ CLW measurements. The latter's estimate found best related to the sounding information
was a 7-h mean centered on the sounding hour.

.With the exception of temperature, no apparent linear (or other) relationship was indicated in
the scatterplots of DRA sounding variables and like information from the other sounding sites.
A strong linear relationship was evident for temperature between the various other sounding
sites. Weaker, but noticeable, linear relationships between sites were evident at 750 and 700
mb in relative humidity, wind direction, and wind speed. For these two levels, the HJ CLW was
linearly related to particularly the INW 750-mb wind speed.

Indications of linear relationships with HJ CLW suggested MLR as an approach to relate the
sounding information to CLW and develop forecasting equations. From a number of
independent variables (HJ CLW the dependent variable), stepwise regression yielded a
multiple correlation of 0.78 with the mean of CVR 700- and 750-mb relative humidity and a
second independent variable, INW 750-mb wind speed, suggesting that local sounding
information can yield useful estimates of CLW on the Mogollon Rim. This finding should be
useful in the conduct offuture experiments and perhaps operational seeding. With TUS relative
humidity and INW wind speed as independent variables, the multiple correlation decreased to
0.67, but still a significant and useful result. In an attempt to improve on this figure, a
one-dimensional model was acquired that could use soundings as input to develop estimates of
CLW content.

The Smith-Feddes model was selected, obtained, reconfigured, and adapted for application to
the Mogollon Rim. Essentially, the model calculates the adiabatic CLW. Adjustments are made
to the adiabatic values, depending upon estimated cloud type, with the use of empirical curves
developed from aircraft data. Once configured, the model is easily applied on micro- or
mini-computers.

F'or comparison with measurements, the model-estimated maximum LWC was employed for
each sounding. Comparison with HJ CLW measurements (7-h means) produced a correlation of
0.49 when TUS soundings were employed. Running with INW soundings resulted in a
somewhat lower value apparently caused by dynamic effects produced by the Mogollon Rim.
When the INW 750-mb wind speed was introduced as a second independent variable in MLR,
the multiple correlation was 0.59, short of the 0.67 obtained with the more statistical approach.
The less than desirable performance by the model may be caused, in part, by improper cloud
typing and entrainment adjustments for the Mogollon Rim. Resources did not allow the
illlvestigation of the sensitivity ofthese model features.

16