A simulation study of short-range atmospheric dispersion for hypothetical air-borne effluent releases using different turbulent diffusion methods

Abstract

Radiological impact due to hypothetical air-borne effluent releases for an upcoming fast reactor (PFBR), at a tropical coastal site in Kalpakkam is studied using the HYSPLIT dispersion model. Short range air dispersion simulations are conducted with HYSPLIT for a weakly forced synoptic condition on 3rd March, 2008 with three turbulent diffusion methods, i.e., standard velocity deformation (A), short range isotropic similarity (B) and turbulent kinetic energy (TKE) (C) schemes to assess their performance in predicting the dose distribution at a local scale. The time varying 3-d meteorological parameters for the case study are predicted using a nested grid meso-scale dynamical atmospheric model MM5. Results indicate that the velocity deformation method gives spatially more complex dose pattern while the short-range similarity and TKE methods provide smoother horizontal dose profiles. The downwind dose in different pathways due to elevated releases is found to be the highest for the method A followed by B and C. For the ground releases, the TKE method produced higher doses than the other two. The difference in the dose values from the three methods is due to the variation in the horizontal and vertical mixing calculation and the assumptions involved in the respective methods. The total dose at the site boundary from different pathways is highest from the TKE method. Values from all the methods are about one to two orders less than the values given by a Gaussian Plume model for similar atmospheric conditions. The total dose estimated from the TKE method is 20.62 mSv at the release location and 1.25 mSv at the site boundary.