An operational event reconstruction tool used with biological agent collectors for source inversion applications

In this paper we describe a new methodology for performing source inversion relatively quickly and within the bounds of the uncertainties introduced by the measurements and modeling tools. The goal of event reconstruction is to start with measurements of airborne contaminant dosage and winds at sensors and then “back” calculate where the source originated from and how much material was released (i.e., source inversion) and then project “forward” to indicate where the plume traveled. In our application, the source release occurs only once (e.g., a terrorist release of a biological agent, an accidental release of chemicals from an industrial facility), so that repeated dosage measurements over many different days are not available as in receptor modeling. Our approach utilizes what we call the collector footprint method. It is essentially an approach in which we compute plume transport and dispersion using the airborne contaminant sampler locations as the source locations, reversing the direction of the wind field, and solving for the source strength Q instead of the dosage. A “plume” of Q contours is produced upwind of the airborne samplers and through geometrical overlap and intersection operations, the location of the point source can be obtained. The collector footprint approach is very fast compared to methods that try to determine the source location using different optimization methods as the collector footprint approach only requires a handful of plume model calculations (equal to the number of collectors), whereas the forward approaches often must perform tens to hundreds of thousands of calculations at discrete source locations. In our presentation, we will describe the methodology, and then show examples from validation studies and real-world problems. The tool is being used as part of a national program supported by the Department of Homeland Security.