Bed-sediment analyses are an effective means of integrating contamination episodes in stream segments over extended time periods since many metals and organic chemicals sorb onto fine sediment particles. Sediment traps were used collect sediment for the assessment of transport rates of trace metals and PCBs at all stream sites.

Physical Characteristics of Sediment-Trap Contents

Sediment traps were deployed at each of the 15 stream-sampling locations. The mass of sediment recovered from traps deployed at the 15 stream sites is depicted in figure 54 (left panel). Particle sizes can be described as a percentage of sand, silt, and clay (right panel). Samples from traps that yielded the greatest mass of recovered sediments were composed primarily of sand.

Figure 54. Mass and particle-size distribution of sediment captured in Phase II sediment traps in the Milwaukee Metropolitan Sewerage District planning area, Wis. Site abbreviations listed in table 1.

Traps deployed in 2005 collected a higher fraction of silt- and clay-size particles and a correspondingly lower fraction of sand-size particles compared to traps deployed in 2004. The differences seen in the particle-size distributions between the 2004 and 2005 field seasons were likely related to differences in rainfall, stream discharge, and the time period chosen for trap deployment. Area streams in 2004 had higher flows than in 2005, and may be the reason for the higher fraction of coarse-grained sediments in traps.

Sediment samplers deployed as part of Phase II were not designed to provide quantitative estimates of sediment transport. Nevertheless, the results shown in figure 55 provide information on the approximate particle-size classes that were transported as suspended sediment. Because of the design of the samplers, the smallest particle-size classes (clay size and smaller) probably were underestimated because the sediment-trap efficiency for those size classes is thought to be quite low.

Figure 55. Particle-size distribution in sediment-trap samples, by Phase II sampling location, in the Milwaukee Metropolitan Sewerage District planning area, Wis. Site abbreviations listed in table 1.

return to top

Comparison of Sites by Use of Average Effect Concentrations

The material collected in the sediment traps was analyzed for arsenic, cadmium, chromium, copper, lead, mercury, nickel, zinc, total phosphorus, and total PCBs. In some cases, not enough material was collected to allow all analyses to be run. The sediment-trap results were summarized by looking at specific chemicals alone and by combining the potential effects of all chemicals detected.

Results from two different methods of describing the significance of all chemicals detected in the sediment-trap samples are shown in figure 56. The left panel shows the total number of constituents that exceed the consensus-based threshold effect concentration (TEC) as listed in MacDonald and others (2000). The TEC is a conservative indicator of possible problems related to sediment chemistry; exceedence of even several TECs does not necessarily indicate toxicity. The right panel shows each sampling result divided by a consensus-based probable effects concentration (PEC), which yields a “PEC quotient.” The PEC quotient was developed to relate bulk sediment-chemistry results to toxicity-test results involving experiments with benthic organisms (MacDonald and others, 2000). A mean PEC quotient can be computed from all individual PEC quotients for each sample.

Figure 56. Number of consensus-based threshold effect concentration (TEC) exceedences and the mean consensus-based probable effects concentration (PEC) quotients for contaminants in sediments from the Milwaukee Metropolitan Sewerage District planning area, Wis. (MacDonald and others, 2000). Site abbreviations listed in table 1.

The developers of the PEC quotient noted that a PEC quotient that exceeds 0.5 is associated with increased toxicity in test organisms, and a PEC quotient between 1 and 5 is associated with a test-organism mortality rate exceeding 50 percent (MacDonald and others, 2000). Milwaukee River at Mouth was the only site where the mean PEC quotient exceeded 0.5, suggesting the possibility of increased sediment toxicity.

The PEC quotients presented here did not account for effects of PAHs or chlorinated pesticide compounds, since these constituents were not analyzed for in conjunction with Phase II sampling efforts. Both of these classes of compounds could potentially change the mean PEC quotients relative to those presented above.

return to top

Comparison of Sites by Use of Individual Contaminant Effects Concentrations

The PEC quotients for individual contaminants at each sample site are summarized in figure 57. For certain sites, the mean PEC quotient differed greatly from PEC quotients calculated for individual contaminants. For example, Kinnickinnic River had the highest individual PEC quotient for copper, zinc, or mercury. Milwaukee River at Milwaukee and Milwaukee River near Cedarburg had high individual PEC quotients for total PCBs. Underwood Creek and Menomonee River at Wauwatosa had high individual PEC quotients for lead. By contrast, when the overall mean PEC quotient was considered, the Milwaukee River at Mouth site had the highest result, reflecting the additive effects of contaminants transported throughout the watershed.

Figure 57. Consensus-based probable effects concentration (PEC) quotients for nine contaminants in the Milwaukee Metropolitan Sewerage District planning area, Wis. (MacDonald and others, 2000). Site abbreviations listed in table 1.

Total PCBs were detected at almost all sites, and most of the detections were at concentrations near the level of detection (LOD). It was useful to compare the results for maximum total PCB concentrations to the TEC of 0.060 mg/kg, as defined by MacDonald and others (2000). Sediments with contaminant concentrations below the TEC generally do not cause toxicity. Samples from five of the sites met or exceeded the TEC for PCBs: Milwaukee River at Mouth (1.2 mg/kg), Milwaukee River at Milwaukee (1.1 mg/kg), Milwaukee River near Cedarburg (0.3 mg/kg), Menomonee River at Wauwatosa (0.22 mg/kg), and Little Menomonee River (0.063 mg/kg) (table 23). Results for all other sites fell below the TEC. The main stems of the Milwaukee River and the Kinnickinnic River both have several known areas of PCB contamination, as documented in numerous reports, including Southeastern Wisconsin Regional Planning Commission (1987) and Steuer and others (1999). PCB contamination along the main stem of the Milwaukee River may be attributed to the Cedar Creek Superfund alternative site located upstream. PCB concentrations exceeding the TEC on the Menomonee River may be a result of former industrial land-use practices within the Menomonee River watershed (Southeastern Wisconsin Regional Planning Commission, 1987).\

Table 23. Maximum total polychlorinated biphenyl (PCB) concentrations in sediment-trap samples for 15 Phase II stream sites collected during two surveys (June 2004 and April 2005) of the Milwaukee Metropolitan Sewerage District Corridor Study.

[mg/kg, milligram per kilogram; <, less than; --, not available; bold indicates concentrations that met or exceeded the threshold effect concentration (TEC) of 0.060 mg/kg (MacDonald and others, 2000]

   ^a No test performed due to insufficient amount of sediment collected

Concentrations of the nine contaminants shown in figure 57 (Phase II data) were compared separately to Phase I results (appendix 5). Comparisons were made with a limited number of sites. For almost all sediment trace elements (cadmium, chromium, copper, lead, mercury, nickel, and zinc), most sites indicated decreases in concentrations from Phase I to Phase II. The one exception was arsenic, which increased at most sites. Ever decreasing LOD’s for PCB’s may have allowed more detections of low-level PCB’s. This greater analytical sensitivity over time is important to keep in mind when drawing comparisons between current and historical data. Many of the Phase I samples may have contained PCBs at concentrations undetectable by older laboratory methods (LOD 0.05 mg/kg), but which might have been detectable using current laboratory methods (LOD of 0.024 µg/g). Data for total PCB concentrations in sediment were available for Phase I and Phase II comparisons at four sites (appendix 5). During Phase I, PCBs were detected in samples from three of the sites; no PCBs were detected in the single sample from Little Menomonee River. During Phase II, two of the four sites had concentrations detectable at the previous WSLH LOD: Little Menomonee River (0.06 mg/kg) and Menomonee River at Wauwatosa (0.12 mg/kg).

return to top

Comparison of Sites by Use of Total Phosphorus Enrichment

Phosphorus concentrations also were measured in conjunction with sediment sampling. In order to show the level of enrichment above background levels (that is, upland soil content), observed concentrations were divided by the average soil-sample total-phosphorus result (38.75 mg/kg) for Waukesha, Milwaukee, Ozaukee, and Washington Counties reported by Combs and Peters at the University of Wisconsin-Madison Soil & Plant Analysis Lab between 1995 and 1999 (n.d.). The resulting enrichment factors (expressed as a ratio) for samples varied widely between sites, ranging from 4.8 at the Menomonee River at Wauwatosa site to 55 at the Milwaukee River at Mouth site (fig. 58). The total phosphorus concentration in sediment-trap samples reflected the overlapping influences of urban and rural nonpoint-source pollution, and point-source inputs from treatment plants, combined and sanitary sewer overflows, cooling-water discharges, and phosphorus associated with soil erosion.

Figure 58. Total phosphorus enrichment factor relative to background upland soil concentrations in the Milwaukee Metropolitan Sewerage District planning area, Wis. Site abbreviations listed in table 1.

return to top
continue to Water-Column Toxicity