Back-forward model analysis for spatial localization of pollutant sources

The present paper deals with the implementation, calibration and validation of a back-forward mathematical model for pollutant source spatial localization and characterization. In particular, a Gaussian model, used as a sort of reverse engineering tool, was implemented to analyse plume dispersion. In this paper particulate matter was the pollutant agent. In order to calibrate model parameters and verify its accuracy, an extensive experimental campaign of plume dispersion was carried out in a controlled environment. In details, a wind tunnel was used to generate wind effect on plume and aerosol particles spatial concentrations were measured by means of a sensor network. On the basis of the obtained results it is possible to state that the implemented model is able to identify position of the sources and emission rate with low errors.