Test of sediment initial‐motion theories using irregular‐wave field data

Sediment initial‐motion theories predict the flow conditions that coincide with onset of sediment motion, but usually the most that can be determined from burst‐sampled field data is whether or not sediment was in motion. Furthermore, initial‐motion theories are usually based on regular‐wave laboratory data, with little or no guidance given on how such theories are to be transferred to the real world of irregular waves. A field dataset obtained from the zone of wave shoaling beyond the surfzone is used to compare the performance of three initial‐motion theories in a way that takes explicit account of wave irregularity and the limitations of burst sampling. The dataset comprises video images of the seabed and measurements of waves, currents and suspended sediment. The initial‐motion theories tested were Komar & Miller’s (1975, J. Sedim. Petrol ., 45, 362–367) ‘wave‐orbital‐speed’ theory, a ‘wave‐stress’ theory, and a ‘wave‐plus‐current‐stress’ theory. Fourteen transitions from no sediment motion to sediment motion were observed, but not all of those represented a challenging test of theory. Using the criterion that the best theory is the one that minimizes errors in classifying bursts as ‘no sediment motion’ or ‘sediment in motion’, Komar & Miller’s (1975) theory was found to perform the best when waves were characterized by significant wave height and mean spectral period, both of which can be estimated directly from pressure data alone. Komar and Miller’s theory was better at predicting the onset of extended transport events, which occurred during confused seas, rather than the transition back to moribund seabed under clean swell at the end of such events. The conclusion regarding best theory is dependent on the choice of scales used to characterize wave motion. The way forward therefore is to standardize and define terms precisely.