Temporal variation in bed configuration and one-dimensional bottom roughness at the mid-shelf STRESS site

Abstract

Time-lapse stereophotographs were taken over a 90-day period from mid-November 1990 to late-February 1991 at a 90-m silt-bottom site on the central California shelf as part of the STRESS (Sediment Transport Events on Shelves and Slopes) project. Five distinct bed configurations were observed, in order of decreasing abundance, these are: (1) bioturbated bed; (2) smoothed bed; (3) current-rippled bed; (4) scour-pitted bed; and (5) wave-rippled bed. Concurrent measurements of the flow field implicate along-shelf currents, rather than waves, as the primary agent forming the physical bed configurations. The presence of a wave-induced cross-shelf gradient in near-bottom suspended sediment during storm events and the redistribution of this sediment by upwelling or downwelling currents is postulated to control the appearance of depositional current-ripples (northwest poleward flow, downwelling) and erosional scour-pits (southeast equatorward flow, upwelling). All physical bed forms are destroyed by bioturbation processes in periods of hours to days. Analytical photogrammetric techniques were used to extract high-resolution sea floor height data from the stereophotographs. Results indicate maximal relief over a 0.25-m 2 area at this site never exceeded 5 cm. Root-mean-square (rms) height varied by a factor of 3 (3.2–9.2 mm) and is a weak function of bed configuration. Current ripples have the largest rms-height, smoothed and scour-pitted beds the smallest. Rms-heights of bioturbated beds are variable and appear to depend on the previously produced physical bed configuration. Changes in rms-height can be abrupt with factor of 2 changes observed over a 12-h period. Horizontal descriptors of roughness such as peak spacing or peak width cannot separate bed configurations. Results from surface slope distributions are broadly coherent with the rms-height data, in that surfaces with large rms-heights have broad slope distributions and vice versa. Slope distribution data also indicate that all bed configurations except the current-rippled bed are isotropic. These preliminary data suggest that time series information is needed to adequately resolve both the micro-scale roughness of the sea floor on continental shelves and the presence of short lived, but potentially flow-diagnostic bed configurations.