WMOD00472

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

MODEL ESTIMATION OF CLW

Model features

A one-dimensional model, known as the SFM (Smith-Feddes model), currently employed in the
Winter Icing and Storms Project (conducted in the Colorado front range) for CLW and icing
studies, was obtained for application in Arizona. The model could be configured to employ
soundings as input. For each 100-m layer of each cloud deck, the model developed estimates of
CLW concentration, temperature, and probability of the cloud layer in question being all liquid.
Also generated for each cloudy layer was an estimate of the cloud type, volumetric mean drop
diameter, and the probability of encountering cloud.

The model was originally developed by Feddes (1974) and Smith (1974) and improved by
Rogers et al. (1985). The early Smith and Feddes procedures estimated CLW and some other
cloud parameters as a function of cloud type and temperatures. Improvements have led to more
accurate estimates of CLW. Others have altered the input/output of the computer code for
interactive operation and driving (of the SFM) from the LAPS (Local Analysis and Prediction
System) (McGinley, 1989) data base. The LAPS analyses produce estimates of temperature and
cloud descriptions on a grid that is updated hourly by objective analyses on data from various
sensors including surface mesonet, doppler radar, wind profiler, satellite and rawinsonde.
Technical details of the model are given in Haines et al. (1989) and Luers et al (1988). Only a
brief summary of the more important model details appears here.

In the configuration. for Arizona, model input/output was altered to enable running files
containing a number of soundings. Those files from CVR, INW, and TUS were employed to
provide estimates of CLW and other outputs of secondary interest. The latter included
estimates of the severity of icing for helicopters and fixed-wing aircraft, as developed by using
an ice accretion algorithm.

Given on figure 15 is a listing of the model inputs and outputs. With a sounding as input, the
program PROC-SND develops, for each cloud determined, estimates of the type of cloud, base
and top elevations, percentage cover, temperature, and pressure, every 100 m, and percent total
coverage. The SFM uses the outputs ofPROC-SND to develop various estimates for each cloudy
layer. A sample of model output is given on figure 16 for the TUS OOOOZ UTC sounding of 10
February 1987. The CLW content is given in the column entitled "LWC" (liquid water content)
in units of glm3.

Of primary interest to this study was the estimated CLW content produced by the model. The
SFM generates CLW according to the Rogers et al. (1985) approach in which the adiabatic CLW
profile is calculated and then adjustments are made based on estimates of entrainment that
depend on cloud type. The adiabatic CLW is calculated for each 10 m of a cloud layer by
computing the water condensed in an air parcel that rises 10 m and cools at the saturated
adiabatic lapse rate at the bottom of the layer. Entrainment adjustments for cumulus and
stratocumulus clouds use a curve developed by Skatskii (Rogers et al., 1985) that depicts the
ratio of observed mean condensed Ill,oisture content (CMC) to the adiabatic CMC. A similar
curve developed by Warner (1979) is employed in adjustments for stratoform clouds. In the case
of cirroform clouds the liquid water is assumed zero and any condensed moisture is ice water.

13