The often prohibitive costs of comprehensive field trials coupled with relatively cheap and abundant computational power leads to a strong desire to use modelling tools to supplement field testing of system components. These modelling tools must be capable of reproducing key environmental variables present during field testing and require rigorous validation. The Virtual THreat Response Emulation and Analysis Testbed (VTHREAT) modelling system is composed of a suite of models designed to provide a virtual Chemical, Biological, Radiological and Nuclear (CBRN) release environment. Two key variables that VTHREAT is designed to realistically simulate are agent concentration and wind velocity. Typical validation studies compare mean predicted and observed quantities of interest such as mean concentration and mean wind speed and direction. This paper attempts to develop techniques to evaluate fluctuations - in particular, two–dimensional wind vector fluctuations.
Keywords: virtual environments, transport, pollutant dispersion, concentration fluctuations, wind fluctuations, validation methodology, p–values, air pollution, dispersion modelling, hazardous release, virtual chemical sensors, biological sensors, meteorological sensors, wind vector fluctuations, agent concentration, wind velocity, dispersion modelling, NBC events