Nuclear, Biological, and Chemical Hazard Prediction Modeling: A New Approach

There are a number of community-accepted NBC transport and diffusion models. Each has inherent science strengths and weaknesses. Each uses similar input (weather, terrain, agent information), yet can produce different output. This causes confusion as to which model is “correct”. Litton-TASC is taking an “ensemble” approach to NBC hazard modeling by grouping a number of these accepted models into an ensemble, and developing algorithms to weight the strengths and weaknesses according to their unique characteristics. Additionally, ensembling algorithms are able to assimilate real-time observational data from newer generation NBC detectors, as well as meteorological sensors to adjust the predictive model output. Outputs are then presented as a single predictive downwind hazard, or individual model output can be selected. The predicted hazard corridor plots are integrated into command and control systems, and used as input for various decision aids. These automated decision aids assist operators in making rapid assessments of various courses of action. This methodology is designed to allow the model developers to continue to improve their particular scientific approach, and then have the improved model inserted as a “plug and play” module. The real-time data assimilation into the NBC predictive downwind hazard will ensure that observational data has a higher weighting than predictive.