WMOD00527

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

The NWS MAR includes the formation of the HPC (Hydrometeorological Prediction Center)
in Washington, D.C. Quantitative precipitation forecasting remains a difficult problem. In
the future, the preparation of QPF guidance will be the primary responsibility of the HPC,
where forecasters will base their forecasts on the components of the numerical model suite,
model-output statistics, and radar- and satellite-based techniques that are designed to focus
on precipitation forecasting. The QPF guidance prepared at the HPC will be transmitted to
the RFCs and WFOs, where local forecasters use it as the starting point for their preparation
of local forecast products (McPherson, 1994).

The HPC will include an applications development group that will design and construct
techniques to assist HPC, RFC, and WFO forecasters in adding value to the model output;
i.e., this group will serve as a bridge between the model builders at the NWS's Environmental
Modeling Center and the model users (McPherson, 1994).

· Recommendation - Reclamation should support development of value-added products
(e.g., objective forecast aids) for the NWS model output which focus on specific water
resources management issues. This development would likely include the testing and
verification of QPFs and headwater runoff forecasts received from the RFC for selected
watersheds where accurate and timely forecasts are critical for Reclamation's water
operations.

2.8.3 Future Mesoscale Numerical Models

Strong indications exist of a coming revolution in computing based on massively parallel
architecture. The NWS's long-range plans assume that supercomputers capable of 1014
operations per second will be available early in the next century. Such computing power
would enable model resolutions in the HPC regional suite of 4 to 5 km. The NWS planning
establishes a goal of such high-resolution mesoscale models covering the contiguous United
States at that resolution in about 10 yr (McPherson, 1994). These 3-D mesoscale models are
anticipated to be full physics models, with sigma-z terrain-following coordinates in the
vertical axis. The future operational use of such mesoscale models will significantly improve
the accuracy of QPFs.

NWS's internal research will not be able to do all that is required to develop, test, and add
value to these future mesoscale models; consequently, the NWS encourages r:esearch partners
in other institutions and Federal agencies.

· Recommendation - Reclamation should cultivate a partnership with the NWS to study
watersheds of special interest for Reclamation water operations. Studies should include
the comparison of the input prognostic grid point data for the mesoscale models with
upper-air observations, assessment of model weaknesses and biases, and verification of
model precipitation estimates with recording gauge observations. Case studies for heavy
precipitation events would be of particular interest to Reclamation.

22