WMOD00527

• Title: Quantitative Precipitation Forecasting for Improving Reservoir Operations
• Date: 4/1/1995
• Weather Modification - Doc Type: Report

3. Depth and duration of moist air presence were estimated from the mean 1000- to 500-mb
predicted relative humidity charts.

4. The future 700-mb temperatures were estimated from the predicted 1000- to 500-mb
thickness charts and the initial 700-mb temperature.

5. The estimates from steps 2 through 4 were compared to the known relevant quantities
of the "reference sounding" of step 1 to derive multiplicative "correction factors" which
were applied to the model reference precipitation amounts in step 1. A QPF resulted.
The correction factors were for wind speed, moisture depth, temperature, and duration.
The 700-mb wind direction determined which part of the "reference table" to use.

To expand this past method to other areas of California, and to use NMC model prognostic
gridded field data, "reference tables" of orographic precipitation model output were
established for many more watersheds in the manner described in step 1 above. However,
the determination of the "correction factor" to apply for each forecast period was made in a
different manner than described in steps 2 through 5 above, although the factor still amounts
to a comparison of the gridded field data "predicted sounding" with the "reference sounding."
The program makes no attempt to correct for a duration of less than the time interval
between gridded field prediction intervals. The present method is described below.

2.7.2 Present Method of Using the Orographic Precipitation Model

The hypothetical reference sounding is used with a simple inclined plane having the
dimensions of 70.0 km in the horizontal by 1.22 km in the vertical up which the air flows, to
compute a "reference CSR" (condensate supply rate). Then, using the predicted sounding
from the NMC gridded field data and the same simple inclined plane, the program computes
a "predicted CSR." Dividing the "predicted CSR" by the "reference CSR" gives a
dimensionless "relative CSR," which serves as a total correction (multiplying) factor. This
correction factor is used with the 700-mb wind direction and the reference average watershed
precipitation from the "reference table" to determine the 6-h QPF.

The HED71 headwater runoff simulation model requires input QPFs for every 6-h interval.
The NGM and Eta model data are available in 6-h intervals and can be used directly.
However, because the A VN model data are available in 12-h intervals, interpolated gridded
field data are used for the intermediate 6-h intervals for all areas of interest out to 72 h into
the future. These data sets are then used to predict the CSRs and QPFs for the watersheds.

The QPFs from the four NMC models provide 'Jumping off' points (objective aids) for
assessing the weather and preparing QPFs for distribution and as input to the HED7l
headwater runoff model. Forecasters at the NWS's CNRFC will need to adjust these QPFs
based upon all of the latest available data, especially observed river stage and precipitation,
in preparing the hydrologic forecasts.

2.8 Status of QPFs and Numerical Models

2.8.1 Rhea Orographic Precipitation Model

The 2-D orographic precipitation model computed reasonable QPFs for historical major
storms in the ARB using OAK rawinsonde data. However, model accuracy, using NWS model

19