APPROACH AND METHODS:

Home Background Purpose and Scope Approach and Methods Results and Discussion Conclusions

Bottom-sediment samples were collected at 11 sites (Fig. 2; Table 1) in October 1998 by using a stainless steel Ekman dredge. Barron River Canal is cut through limestone with little soil at the surface, and bottom sediments in some places are composed mainly of limestone debris, sand, and organic detritus. We attempted to locate and sample depositional areas containing mainly fine organic sediment. At each sampling site, bottom sediments were composited from 5 or more surface (top 2-3 cm) sediment samples to obtain a representative sample. Bottom sediment was taken from the top of the dredge, avoiding any contact with sides of the dredge.

Figure 2. Location of sampling sites along the Barron River Canal and on the Turner River. [larger image]

Samples were passed through a 2-mm sieve and shipped to the USGS National Water Quality Laboratory (NWQL) for analyses. All bottom-sediment concentrations are reported on a dry-weight basis. The grab samples were composited in pre-cleaned glass or Teflon bowls. The dredge, stainless steel sieve, bowls, and processing tools for organic analyses were cleaned with Liquinox, distilled water, and methanol after sampling each site. The bowls, sieves, and tools for processing trace elements were cleaned with liquinox, deionized water, 5 % HCl solution, and deionized water again. A new, clean rope was used for the dredge at each site. Samples were processed immediately after collection, preserved appropriately, and shipped to the NWQL. Quality control procedures were followed. One concurrent replicate sample was collected from the Jerome site on Barron River Canal and analyzed for the same constituents as the environmental samples.

Filtered water samples were also collected at four sites on the Barron River Canal during October 26-29, 1998, and analyzed for a suite of 83 pesticides or pesticide degradation products. The water samples were collected by using a depth integrated procedure at several points across the width of the canal, processed according to USGS parts-per-billion protocols (Shelton, 1994), and shipped to the NWQL for analysis.

For the purpose of interpretation, concentrations of trace elements in bottom sediments were plotted against or divided by aluminum concentrations to normalize their concentrations. The normalization reduces the variation in concentration due to the variation of the clay concentration from one site to the next and sometimes allows natural background concentrations to be distinguished from concentrations enriched by human activities (Schropp and Windom, 1988). The trace element-to-aluminum plots show the 95 % confidence intervals derived from "clean" Florida coastal bed sediment data of Schropp and Windom (1988). Data that fall within the 95 % confidence intervals are considered as "background." Data above the 95 % prediction limits may indicate trace element enrichment above natural background concentrations.

Concentrations of trace organic compounds were normalized against organic carbon concentrations to reduce the influence of bulk sediment composition on spatial patterns. To normalize the data, the concentrations of each organic compound were divided by the concentration of organic carbon. The original reporting units are µg/kg for SVOCs and g/kg for organic carbon; consequently, the ratios from the reported concentrations are the same as if the concentrations were converted to the same units and multiplied by a unit conversion factor of 1 million. If the ratios for one compound were much larger than for the other compounds, a smaller multiplier was used in graphs to show more detail for the other compounds. Hydrophobic organic compounds have an affinity for organic matter and, consequently, bottom sediments that are rich in organic carbon compounds adsorb more SVOCs than do bottom sediments that are mainly composed of inorganic matter. The normalization procedure enhances the ability to distinguish local inputs of SVOCs from variations that are simply due to variations in the bulk concentration of organic matter in the sediment samples.

Analytical data below the usual method reporting limits were used in some cases in the interpretation and presentation of data. If the presence of an analyte is verified by spectra and is below the method reporting limits, the concentration is reported as an estimated value. The use of analytical data below the usual method reporting limits was justified by Pritt (1994) in a U.S. Geological Survey Technical Memorandum.

Available aquatic-life criteria for bottom sediment and water were used to evaluate potential adverse biological effects of analytes detected during this study. Data from the Barron River Canal and Turner River sites were compared with aquatic-life criteria developed by Environment Canada (1999), U.S. Environmental Protection Agency (USEPA, 1996), National Oceanic and Atmospheric Administration (Long and Morgan, 1990; Long et al, 1995), and Florida Department of Environmental Protection (FDEP, 1994). Gilliom and coworkers (1998) summarized a number of the aquatic-life criteria from the above sources. Where available, we used the Canadian probable effects level (Environment Canada, 1999) for evaluations of water and sediment concentrations. For a few substances, other criteria were used in the absence of Canadian criteria. Aquatic-life criteria are usually based on research done on individual chemicals or classes of compounds and do not account for the synergistic effect of mixtures of chemicals that occur in the environment (USGS, 1999).

The criteria for trace elements in bottom sediments were developed by using analyses of bottom sediments that were partially digested with acid in an attempt to measure bioavailability. Trace element analyses were determined by a relatively complete digestion commonly used for geochemical investigations. The essentially complete digestion tends to yield higher concentrations than partial digestion and might, in some cases, make contamination appear more severe.