5. Conclusions
Experimental observation of particles within a fully saturated saltation cloud has historically been a challenging task because of inadequate lighting, difficulties with surface detection, and a high degree of error associated with particle detection and trajectory identification. In this paper the EPAS-PTV method, based upon particle radius comparison, is first optimized to reduce trajectory identification errors and then validated through comparison with measurements obtained using laser Doppler anemometry (LDA). Application of this technology in a novel wind tunnel investigation of the spanwise component of trajectories within a saturated saltation cloud reveals that less than 1/8th of particles travel directly along the path of the mean air flow. However, 95% of the particles sampled are contained within 0 6 h 6 45. This study provides the first direct measurements of the x, y and z components of particle velocity and their respective probability distributions. The alignment of the flight path is found to systematically alter the total velocity of a given particle, as well as its launch/ impact angle. The observed decline in the proportionate particleborne kinetic energy with increasing spanwise angle, however, is found to be driven primarily by the waning particle counts and not speed. At high angles, the primary mode of transport appears to shift from saltation to reptation. Such observations may have important implications for the parameterization of emerging 3D saltation models, as well as for understanding the inception and growth of small-scale aeolian bedforms in the context of particle diffusion, both of which are beyond the scope of the present paper. An extension of this 2.5D study is presently underway to quantify particle acceleration for the full parabolic trajectories captured and to determine the variation in U and d with elevation above the bed surface.
