Settling speeds at each depth are significantly different from settling speeds at the other depths (p < 0.05).

Settling speeds at each depth are significantly different from settling speeds at the other depths (p < 0.05).

Diameter at surface is significantly different from diameters at middle and bottom, which are not significantly different from each other (p < 0.05).

velocity explains much of their co-variability. It is notable that combining the data from all three depths resulted in an artificially high estimate of the fractal dimension and an artificially low estimate of the reference particle size, reinforcing the idea that these (especially the bottom) represented distinct particle pools.

Other data are included in Figure 10.9 for comparison. The LISST estimates are from the same locations and times as the VISTA data, calculated in the same way as the data presented in Figure 10.8b. While they are consistent with the fractal relationships, the LISST estimates were at the low end of the scale for both particle size and settling velocity. This is most likely due to the fact that the VISTA estimates were for large, visible settling flocs, while the LISST estimates incorporated the entire spectrum of particles and thus are strongly affected by the fine, slow settling fraction. The Valeport tube estimate of ws50 indicated in Figure 10.9c (1.9 mm sec-1) is from a sample mid-way between the second and third VISTA bottom samples, as indicated in Figure 10.8. It is quite encouraging that it is in the center of the cluster of VISTA estimates. The second and third Valeport tube estimates were very similar, but the last sample of the anchor station contained significantly more material settling at very high speeds (>7 mm sec-1). The long-dash line in the "all depths" panel corresponds to the reference particle size and fractal dimension quoted for the Ems estuary flocs by Winterwerp61; it is reproduced here because of its striking similarity to our surface layer data. Finally, the short-dash line in the "all depths" panel corresponds to the Chesapeake ETM data collected by Gibbs.17 Clearly, though the basic trends are similar to our data (with a slightly lower fractal dimension), the ranges of settling speeds and particle sizes are considerably lower in Gibbs' data. This is likely because the sampling techniques available at the time disrupted the largest, fastest settling flocs.

A summary of all of the Valeport tube data we collected in upper Chesapeake Bay during 1996, 2001, and 2002 is presented in Figure 10.10. Some arguments against the use of static bottom withdrawal settling tubes51 have focused on the possibility of biases induced by flocculation in the tube itself due to differential settling, which would result in the previously reported strong correlation between settling velocity

0 0

Post a comment