Measurement of size and settling velocity of suspended aggregates on the northern California continental shelf

Abstract

A high-resolution video camera attached to a miniature sediment trap was used to make time-lapse images of suspended aggregates settling through the water column. The video-trap system was placed 2 m above the seabed in the STRATAFORM study area off Eureka, California, and was programmed to take a 7-second time sequence of settling particles every 6 h over a 4-month period. The camera field of view within the trap is 15 mm and the minimum particle resolution is approximately 130 μm. Data analysis was carried out with a video frame grabber and digital imaging software to determine the particle size, shape, and settling velocity. Results of the first deployment of this system (September 1995) show that particle sizes (elliptical nominal diameter) determined from the video images ranged from 130 to 740 μm with a modal size of 300 μm. The average form factor (ratio of short to long semi-axis) of the aggregates was 0.71. Measured settling velocities varied from 0.09 to 8.13 mm/s. In terms of weight-percent, or mass, the median size of the aggregate distribution is 600 μm. Analysis of settled particles collected from the sediment trap shows disaggregated or component grains ranging from 1.0 to 63 μm with a modal size of 26 μm and a median size of 11 μm. Additionally, an analysis was carried out to separate the relative importance of particle shape and excess density on settling velocity, and these results were compared to observations. Using the equations for settling velocity of spheres and prolate spheroids, and substituting published relationships for size versus excess density into these equations, the settling velocity for combinations of shape and excess density has been calculated. Results show that differences between sphere and prolate spheroid (with major axis both parallel and perpendicular to flow) approximations caused less than a 13% difference in computed settling velocity. In contrast, the various size/excess density relationships used to compute settling velocity caused order of magnitude deviations from measured values. In all cases, however, the approximations of settling velocity for aggregates provide a significantly better fit to data than approximations based on disaggregated sediment size distribution.