Demonstration of linked UAV observations and atmospheric model predictions in chem/bio attack response

Both the faculty and staff from the Departments of Meteorology and Aeronautics evaluated the integration of components for a near-real time decision aid designed to enable small units to respond in a focused way to a ChemBio attack. This effort included the field-testing of an atmospheric dispersion prediction model, an instrumented UAV for collecting meteorological data, and the means for linking the UAV data to real-time dispersion prediction. The primary modeling effort focused on an adaptation of the 'Wind On Constant Streamline Surfaces' (WOCSS) model developed to run on a small computer with input from an external mesoscale model (MM5). The combined models were run for approximately one month for the region surrounding Camp Roberts, CA. In situ meteorological data were collected at the Camp Roberts airfield from 2 October to 5 November 2002 to validate the model predictions. The model results showed promise in capturing the diurnal evolution of near-surface temperatures that drive the local circulations in the warm season. Linking WOCSS with the atmospheric mesoscale model forecasts showed no significant improvement in wind forecasts when compared to the mesoscale model wind forecasts alone. Linking WOCSS to the trajectory visualization code revealed that vertical wind component estimates needed to be improved. The linked model/UAV demonstration of 7-9 October 2002 tested the synthesis of UAV measurements and dispersion model predictions. Although a UAV mishap occurred soon after the demonstration began, the instrumented UAV performance during this early period and in preliminary flight tests indicate that the hardware/software architecture for UAV data collection and its linkage with real-time dispersion prediction will be successful. Overall, the demonstration proved the feasibility of linking a coarse grid mesoscale model to a fine-scale diagnostic wind model for producing fine resolution forward and backward trajectories.