Extended Abstract (308K)452
Determining Turbulence Scaling Variables and Source Characteristics from Contaminant Concentration Data
Andrew J. Annunzio, Penn State Univ., University Park, PA; and S. E. Haupt and G. Young
Characterizing the source of a contaminant is a contemporary issue in homeland and defense security the solution to which is required to mitigate the effects of the contaminant and to predict subsequent atmospheric transport and dispersion (AT&D). We determine these source characteristics by matching contaminant concentration observations from sensors with contaminant concentration predictions from a dispersion model via an optimization technique. The dispersion model produces a concentration field that is a function of the source location as well as other variables such as the boundary layer depth, zi, and the convective velocity scale, w*. These length and velocity scales of the boundary layer spanning eddies uniquely determine the lateral and vertical spread of the contaminant. By matching the contaminant spread between simulation and observations, these variables can be ascertained at short ranges, well before contaminant entrapment provides information on boundary layer depth. This method is viable for a continuous contaminant release where it is possible to average concentration values in time. For an instantaneous release, it is not be possible to take an average, because there is only one realization. In this case, we must rely on information from entrapment, and thus sample concentration values at longer downwind ranges where the contaminant is well mixed in the vertical. We test this method for instantaneous and continuous contaminant releases in the atmospheric surface layer and in the mixed layer. For each scenario, boundary layer physics provides implications for the proper sensor domain size, and the appropriate dispersion model and spread parameters. By determining these scaling variables it is possible to accurately locate the source as well as to predict AT&D further downwind.
Joint Poster Session , Applications of Artificial Intelligence Techniques to Air Pollution Problems
Wednesday, 20 January 2010, 2:30 PM-4:00 PM, Exhibit Hall B2