Extended Abstract (132K)8.1
Modeling of coastal meteorology in weakly forced situations
Ananthakrishna Sarma, SAIC, McLean, VA
The accurate modeling of dispersion is inherently tied to the accurate modeling of winds and temperature structure in the region of the atmosphere containing the plume. The mesoscale models currently in use show good skill in such forecasts when the synoptic forcing is relatively strong. However, in weakly forced situations the models exhibit considerably lower skill, especially in situations where the topography might play a significant role. One such scenario is the complex flow patterns that are observed over coastal regions. We use the Operational Multiscale Environment model with Grid Adaptivity (OMEGA) to explore the impact of high-resolution surface characteristics datasets on the quality of the wind and temperature forecasts, during a weakly forced summer-time scenario. OMEGA is based upon an unstructured adaptive grid, which is unique and new in the field of numerical weather prediction. The strength of the unstructured grid is that it can provide high-resolution where needed, based on any surface characteristics or meteorological fields. Also the grid can be constructed so that the edges of the computational elements follow the coastline as closely as possible, thus providing a better representation of the underlying surface. This is important in the weakly forced scenarios as the winds and temperature structure are strongly influenced by the surface characteristics. When the dynamic grid adaptivity feature of OMEGA is used, the grid can adapt to the time-evolving fields such as plume location, pressure minima, momentum deformation, or the gradients in the potential temperature field.
In this paper we present the results from modeling experiments performed to support a field experiment organized and conducted by the Defense Threat Reduction Agency (DTRA), during 15-31 July 2001. The experiment region covered the Chesapeake Bay region. Special surface observations were taken at several sites in the experiment region. This paper concentrates on a few specific days during the experiment period during which weak synoptic forcing was observed. The model results were compared against all available observations to obtain mean, mean absolute and root mean squared errors of temperature, dew point, wind speed and wind direction. The results show the significance of including accurate surface data such as the sea surface temperature in the model initial conditions.
Supplementary URL: http://vortex.atgteam.com
Session 8, advanced modeling techniques for dispersion on all scales (e.g., Lagrangian particle models, large-eddy simulations, etc.)(Parallel with Session J5)
Thursday, 23 May 2002, 1:30 PM-4:30 PM
Next paper