J1.6
A forensic approach to source location applied to FFT-07 data
Shawn Rottmann, ENSCO, Melbourne, FL; and A. Siegel
The FUSION Field Trial 2007 (FFT-07) field test was conducted for the purpose of evaluating different source term algorithms. This short range (~500 meters) field study, performed in September 2007 at Dugway Proving Grounds, collected data using tracer releases in a highly instrumented network of digiPID samplers and meteorological measurement sites. Algorithm developers were provided isolated meteorological observations and select sampler measurements. The developers were then asked to determine source term characteristics such as location and release rates.
ENSCO developed an approach which does not require brute force tactics, but instead combines sampler signal, forward and backward trajectories, plume/puff dispersion characteristics, and semi-automated, manual intervention to converge on source-location solutions. The approach begins with the inclusion of only the most statistically significant points from the sampler data time series and triangulates back to the source regions. Higher concentrations, when associated with forward and/or backward trajectories and correlated with upwind and/or downwind sensors, represent centerline, or near-centerline hits that provide a first guess estimate of the azimuth of a source. These analyses reveal not only single sources, but the existence and location of multiple sources.
Source location and release rate results were tabulated and submitted to the FFT-07 independent analysis team. This team compared the results with actual data and thus far suggests the multi-step trajectory and decision-making methodology predicts average source locations within 100 meters. A fully-automated technique is being considered for future analyses. This presentation will present a discussion of the method along with the most pertinent preliminary results.
Joint Session 1, Applications of Artificial Intelligence Techniques to Air Pollution Problems
Tuesday, 19 January 2010, 3:30 PM-5:30 PM, B308
Previous paper Next paper