Atmospheric dispersion simulation of an accidental smoke plume using a heat diffusion algorithm into a LES-STO coupled model

Abstract

A high resolution earth satellite image with the visible smoke plume of a huge fire in the city of La Plata, Argentina, is used to verify the simulated plume with the fluid particle dispersion model LES-STO (large-eddy simulation - lagrangian stochastic one particle model). The initial and boundary conditions are provided by a boundary layer model forecast, and local meteorological observations validate the wind and temperature forecast with good results. The smoke plume is modeled by the continuous emission of fluid particles, initiated 4 h before the time of the satellite image in order to allow the model spin-up. Due to the lack of technical information about the fire, the emission parameters required by the model are estimated considering the dimensions of the burnt area and the products stored there. The spatial orientation and horizontal lateral extension of the modeled plume coincide with the smoke plume clearly visible in the satellite image. The conclusion of the study is that it is possible to use LES-STO with initial and boundary conditions provided by operational meteorological forecasts to make a diagnosis of those areas that could be affected by accidental emissions. • The new smoke plume simulation algorithm is performed into LES-STO for obtained the geographical plume dispersion of smoke. • High resolution satellite image is used to verify the result of the smoke plume simulation. • The initial and boundary conditions determine the dispersion of the smoke plume and its geographical distribution. • LES-STO can be used with boundary conditions provided by operational weather forecasts for diagnosis of fire affected areas.