Computational modeling of indoor dynamics of chemical warfare agents: Case study with indoor releases of sarin
P.G. Georgopoulos, S.K. Stamatelos, P. George, and S. S. Isukapalli
Environmental and Occupational Health Sciences Institute (EOHSI)
Estimation of short term human inhalation exposures to airborne contaminants indoors poses special challenges depending on the physical, chemical and toxicological properties of the contaminant considered. The majority of currently used models for either residential or occupational applications typically consider idealized configurations of source and receptor settings and employ various simplifying assumptions with respect to mixing and transport processes. However, indoor airflows can be very complex due to factors such as heat sources, obstacles such as furniture, electrical equipment, and the presence and motion of humans. These factors may result in concentration profiles that cannot be adequately captured by "conventional" methods but may be important, especially in the case of highly toxic agents, where human movement and posture will affect contaminant transport and exposure patterns. Alternatively, Computational Fluid Dynamics (CFD) techniques allow estimation of detailed transport patterns through the numerical solution of 3D mass, energy and momentum balance equations, accounting for local mechanical and thermal convection effects on the transport, mixing and deposition of contaminants. Case studies are presented that compare the application of a series of both screening and CFD approaches in calculating exposures for scenarios involving the release of the chemical warfare agent, sarin, in idealized as well as realistic indoor environments. The exposure metrics include temporal profiles in terms of Acute Exposure Guideline Limits (AEGL) values, corresponding to different lengths of exposure. Sensitivities of the estimates developed through different parameterizations and approaches are presented to complement their evaluation. Limitations and strengths of the different methods are identified and their implications in modeling situations, such as emergency events in specific occupational settings (e.g. healthcare facilities), are discussed.