Estimating the impact threshold for wind-blown sand

In many aeolian studies, it is commonplace to use Bagnold's (1936) equation to calculate threshold shear velocity (u∗t), which includes an empirical constant, A, typically set at about 0.082 for maintaining saltation (the dynamic, or impact, threshold). Here, we present data from a pilot study to assess the variability of A to improve estimations of u∗t, which in turn, should improve transport rate predictions. Using field data from three coastal environments, we measured or calculated all parameters within the Lettau and Lettau (1978) model and u∗t equation. In Jericoacoara, Brazil (BRA), Inch, Ireland (IRE), and Esposende, Portugal (POR) wind velocities were measured with cup anemometer towers and transport rates were measured using traps for 31 data runs lasting 120 to 1020 seconds each. Mean grain sizes were 0.17 mm (IRE), 0.31 mm (POR), and 0.30 mm (BRA), and mean shear velocities were 0.38 m s^-1 (IRE), 0.40 m s^-1 (POR), and 0.49 m s^-1 (BRA). Empirically determined, adjusted A values ranged from 0.02 to 0.21 with a mean and standard deviation of 0.11 and 0.04. No relationship exists between estimates of A and grain Reynolds number. A statistically significant (p < 0.001), negative relationship was found between A and mass transport rate, leading to substantial over-prediction of transport rates near the threshold and under-prediction during fast winds if a constant of 0.082 is used.