Abstract: The erodibility of particles on stockpiles strongly depends on particles size distribution and friction velocity imposed by the wind over the pile surface. Larger particles are often categorized as non-erodible due to their inertia. These particles create areas (low shear stress levels) acting as a shelter to smaller particles which would be emitted if not protected by the larger ones. This is an important effect to be explicitly included in emission models because it strongly influences the quantities of emitted particles. The USEPA model for fugitive dust emissions estimation due to wind erosion, which is the most widely used model, accounts globally the influence of non-erodible particles on emissions but does not incorporate the proportion of erodible and non-erodible particles in the modeling parameters. Therefore, a modification on the emission factor formulation is proposed based on emitted mass data obtained by wind tunnel experiments. The experimental data involves a set of 21 different experiments, concerning one stockpile and two stockpiles varying the proportion of the type of sand that forms the pile, erodible or non-erodible, the free-stream velocity, for the case of two piles the distance between piles and the presence of an obstacle before the pile. A good agreement was found between experimental data and the emitted mass modeled by the modified emission factor. Besides that, the errors found when the standard USEPA and the modified model are compared shows that the standard model can underestimate the emitted mass from 38 to 2,000 times. The present results strongly indicate that the formulation proposed in this work is a relevant evolution for a more accurate quantification of emitted particles and it is also an important step for the improvement of the original USEPA methodology.

Keywords: Aeolian erosion, USEPA emission model, Erosion potential, Wind tunnel, CFD.