Critical Data Processing for Chemical Warfare Simulant Field Testing - Advanced Data Fusion

To properly measure (and understand) the performance of a standoff detector, properties of the challenge cloud must be known. Meteorological models, such as HPAC and VLSTRACK, have proven to be reasonably good tools for predicting cloud properties. However, a problem with these tools is that they deliver ensemble averages and may not describe a specific cloud to the detail required. The Dugway approach is to use several suites of analytical instruments (point detectors, infrared imagers, spatially scanning FTIRs, and line-of-sight FTIR) to best measure cloud parameters and test conditions. A problem arises because each referee system measures differing parameters at various sensitivities, which makes determining the "best" cloud parameters complicated. Data obtained can include CL (concentration path length), average concentration, cloud width, length, height, and cloud centroid. What is needed is to use the measured data as feedback to a cloud model, so that a truly representative description of individual clouds can be developed and allow more accurate correlation of detector measurements to the current and future properties of the cloud. This would not only allow a better assessment of detector performance, but would also allow correlation of the detectors performance to real world operational needs (example, correlating Dugway Proving Ground [DPG] performance for a Battery attack at 5 km to expected performance in Baghdad). The contractor should produce a software package that will run a met model, predict referee responses, calculate a figure of merit, and rerun the met model with varying parameters to minimize this figure of merit. The software should be configurable and useable by both the Army's and DPG modeling and simulation program and be consistent with their protocols. In addition, the software output should be formatted in a manner compatible with the Army's Future Combat Systems protocol (TENA). There are inherent advantages for using existing meteorological models due to their application across a wide group of users and virtual platforms - however; original pure approaches are also encouraged. PHASE I: Provide a feasibility of technical approach and proof of process using previously collected field data. PHASE II: Produce a software package that will run a model, predict referee responses, calculate a figure of merit, and iteratively run the model with varying parameters to minimize this figure of merit. PHASE III: This software package will produce data that can be used to be used in future distributed tests, where real chemical and biological releases can be inserted.