comparisons for the following reasons: First, the sediment core at Km 41.5 (RM 25.8) was collected in Piermont Marsh - an area that has limited exchange with suspended solids in the main portion of the river. Second, the tri-CB concentrations at Km3.9 (RM2.4) are exceptionally high compared to other homologues in the sample, strongly suggesting that there was an analytical error in quantification (for example, matrix interference) or a very localized source of tri-CB near this sampling location. The resulting longitudinal distributions for di-CB (not shown) and tri-CB (Fig. 25.4) exhibit a more pronounced decline in concentrations with distance downstream. This is due to the lower partitioning of di- and tri-CB to suspended solids and DOC which results in more of these homologues remaining in the freely-dissolved phase and available for loss by volatilization.

The above model evaluations clearly demonstrate that both the magnitude and longitudinal distributions of PCBs in sediments are strongly dependent on homologue-specific partitioning behavior. Enhanced partitioning of the lower chlorinated homologues through the adjustment of "aphyto" was critical in increasing surface sediment concentrations of di- and tri-CB, and reduced partitioning of the high chlorinated homologues through the specification of "kg/ku" had an important (albeit smaller) effect on sediment concentrations of penta- and hexa-CB. These results are consistent with PCB partitioning to phytoplankton (as observed by Skoglund and Swackhamer, 1994), andstrongly suggest that phytoplanktonare important in controlling PCB partitioning to suspended solids. In addition, phytoplankton production in the water column and subsequent decomposition of phytoplankton-derived material in sediments play a key role in scavenging PCBs from the water column and accumulating them in sediment. Without this trapping mechanism, PCB accumulation in sediments would be greatly reduced.

Time history/projections of dissolved PCBs in the water column and particulate PCBs in sediments are givenin Figure 25.5 for the model simulation period of January 1987 to December 2001. Dissolved PCB concentrations at Km 207 (RM 128.5), which is 40 km (25 mi) downstream from Troy, clearly show the effect of changing loads from the Upper Hudson (Fig. 25.2). As shown, dissolved PCB concentrations at Km 207 (RM 128.5) decreased during the period of exponentially decreasing loads in the late 1980s. A large peak in dissolved PCB

0 0

Post a comment