Statistical characterization of sedimentation velocity of natural particles

Aeolian sediment transport has major repercussions for global climatic variations, air quality, human health, agricultural areas, air and ground transportation, civil structures and infrastructures. This led to the development of models to understand the involved physical process, and to predict the dynamics of aeolian sediment transport. The sedimentation velocity is one of the key model parameters characterizing sediment behavior. It allows the prediction of the mode of transport, the distribution of particles above the ground, and the sediment transport rate. Despite its importance, there is substantial discrepancy among the sedimentation velocity laws for natural particles, also due to the large dispersion of experimental data. This study proposes the statistical characterization of the experimental measurements of sedimentation velocity available from the literature. 1812 experimental measurements were recovered from 11 studies. Their variability was discussed, providing an aerodynamic reading. Nonlinear regression was carried out on the consolidated dataset in order to assess the average value. Two well-known laws initially conceived for spheres were refitted to natural particles providing high coefficients of determination equal to 0.90 and 0.93. Then, copula-based regression was performed in order to seize the variability of sedimentation velocity. Finally, the results from the two approaches were compared and critically discussed.