MODELS OF TURBULENCE. APPLICATIONS TO PARTICULATE MIXING INDUCED BY TRAFFIC FLOW IN URBAN AREAS.

In this work we address our attention to the estimation of the contribution of non-exhaust sources, like brake abrasion, tire and road wear and resuspension of particles, to the final PM air concentration; particularly we focus our investigation on the resuspension of PM deposited on road pavement surfaces and raised by the air turbulence produced by the vehicles flux, under urban and extra-urban traffic conditions. Our approach to the problem is based on modeling techniques. We refer to measurement data from literature to determine the selected empirical parameters contained in our models. Analytical models based on algebraic eddy diffusivity hypothesis are applied to describe the mean statistical component of flow generated by air recirculation inside a canyon and by the far-wake structure besides moving vehicles of simplified geometrical shapes. The analysis of the far wake solutions is suitable to the description of vehicle wakes interaction, which permits to apply our analysis to different driving cycles conditions. Numerical simulations based on finite element discretization of suitable two-equation turbulence models are employed to describe near-wake structures, which cause the strongest mixing of atmospheric pollutants and resuspension of road dust. These different components of turbulence fields at different scales of the street geometry are composed to define a set of operational and numerical models for the dispersion dynamics at the canyon scale of two classes of PM10 pollutants, corresponding to a Soot and a road dust components. The deposition and the resuspension of pollutants are described by resistance and filtration models on porous asphalts, inserting the corresponding terms in the dispersion equations as suitable boundary conditions on the ground. We estimate the resuspension fraction of traffic-related PM10 emissions at the tailpipe, through a simplified linear-emission model, considering representative data describing traffic statistics coming from empirical data. Profile laws of resuspension factors are drawn, for different vehicles geometries and velocities, and how resuspension changes with different asphalt characteristics. The results are applied to typical traffic situations in the city of Milan, studying the effect of implementations of different reduction scenarios to the total amount of traffic-related PM10 emissions. The results point at a new approach to the local PM10 reduction policies, based on more effective asphalt design and maintenance. Finally, we apply one of the dispersion operational models to the case of a congested urban traffic configuration in a canyon street, in order to obtain the pollutant spatial distribution.