Mesoscale Modeling

Mesoscale modeling at the Environmental Technology Laboratory (ETL) supports the traditional instrument development and measurement programs of the laboratory. The modeling work is being used 1) as an aid in planning field programs, 2) to better understand the physical context of observations, 3) as a test bed for new parameterizations derived from the observations, 4) as a tool for testing the utility of new data sources such as the anticipated offshore buoy-based profilers, and 5) to validate the models using the measurements. Both the Penn State/NCAR MM5 and CSU RAMS models have been used for these purposes.

Most of ETL's recent modeling applications have been over oceanic and coastal regions. Some examples of these are:

• During the San Clemente Ocean Probing Experiment (SCOPE), synthetic aperture radar (SAR) data obtained from satellite indicated very minimal amounts of scattering on the southeast side of the island in one case. A RAMS model simulation of the event indicated that this region was in the wake of the island and the surface winds were very weak, thus suggesting that the lack of back scatter was due to much smaller wind waves.
• An array of CODAR radars for measuring ocean currents near Monterey Bay observed a diurnally varying gyre that is frequently present in the Bay. Simulations using RAMS showed that an atmospheric vortex, shed off of the nearby Santa Cruz mountains, would propagate across the Bay. The surface stress from the diurnally varying atmospheric flow over the bay was consistent with the observed varying ocean currents. Subsequent wind profiler measurements of the atmospheric flow around the Bay were found to be consistent with the model predictions.
• "Southerly Surge" events on the U.S. West Coast are well-known for their rapid change of wind direction and speed, corresponding changes in sea-state, and for the generation of stratus clouds. MM5 simulations of a southerly surge event along the central California coast predicts the existence of a potential vorticity plume offshore and suggests that it induces the southerly flow along the coast. The model indicates that this plume is frictionally generated along the northern California coastal mountains, where the model low-level flow is in excellent agreement with measured profiler winds.
• MM5 simulations of an explosive cyclogenesis case from the ERICA experiment show that the use of different planetary boundary layer (PBL) parameterization schemes alters the mesoscale characteristics of the cold front and its precipitation pattern. Combinations of different PBL schemes with different convective parameterization schemes yield significantly different mesoscale structures associated with the cyclone. These results suggest that 1) measurements of the PBL structure and surface fluxes of heat and momentum, similar to those ETL has performed in more tranquil areas, are needed to improve parameteriztions under extreme but important conditions, and 2) such measurements and/or PBL parameterization improvements could have significant impacts on the forecasting of explosive cyclones.

The current ETL focus of observational studies over oceans or in coastal areas has motivated the development a coupled atmosphere-ocean-wave model to be used in the laboratory. This development is in collaboration with researches at the University of Colorado, and is currently in progress. The MM5 is being used for the atmospheric component and the University of Colorado version of the Princeton model is being used for the oceanic component. The oceanic component will provide 3-D forecasts of u, v, w, T, and salinity, which will provide simulations of ocean circulations, sea-surface temperature, and storm surges, as well as provide feedback to the atmosphere . The WAM wave model will soon be incorporated. The development of this coupled model is being done using the excellent COPE data set taken by ETL along the coast of Oregon in the fall of 1995. As part of this coupled system, a data assimilation system will be developed to incorporate atmospheric, oceanic, and wave observations into the model initialization.