Although widely used to calculate the concentration of chemical pollutants or radionuclide released from power plants in flat terrain, the Gaussian Plume model (GP) still remains several limitations due to assumptions of homogenous turbulent, non-zero wind condition and/or steady state release that make the model inaccurate in predicting pollutant concentration in flat terrain. Estimation of pollutant concentration in complex terrain is more difficult for GP because of terrain-induced effects.
In this study, a modified Lagrangian Puff model is proposed for more accurate estimation of ground level pollutant concentration both in flat terrain and complex terrain than is found with GP. The new model also is capable of dealing with inhomogeneous turbulent, calm wind condition and unsteady state release, and is designed for releases within an area of a hundred of kilometers in a time scale of several hours. A compact program was built in Matlab environment to calculate ground-level pollutant concentration. The results are compared with experiments and with GP. Examples of accidental unsteady releases are illustrated. The gathered results show that the model has a potential application for regulatory purpose.