Filling station short-range impact on the surrounding area: A novel methodology for environmental monitoring based on the shadows study

Abstract We studied the polluting impact of a filling station on the surrounding area by means of a novel approach. Given the fact that a precise relation can be found between the influence of a physic obstacle on the dispersion of gaseous substances and the “shadow” effect of the same obstacle (i.e. the shielding of a light source), we collected data about the shadowing effect of any kind of obstacle present in an urban area (e.g. buildings, trees, etc.). A machine learning algorithm was trained with the collected data in combination with historic anemometric conditions to predict the wind field of the area under investigation without prior acquiring detailed information about the presence of buildings, obstacles, trees, overpasses, etc. We used the numerical resolution of the Advection–Dispersion equation to calculate in real time the concentration of pollutants. The analytical detection of the pollutant concentration in the air of the area under investigation confirmed the accuracy of our strategy. These unique results demonstrate that our original approach can be a very promising technique for short-range environmental studies inside complex area.