Polychlorinated Biphenyls and other Synthetic Organic Contaminants Associated with Sediments and Fish in the Mississippi River

Colleen E. Rostad, LaDonna M. Bishop, Geoffrey S. Ellis, Thomas J. Leiker, Stephanie G. Monsterleet, and Wilfred E. Pereira

Hexachlorobenzene and PCBs in Silts

(Click on image for a larger version, 66K)

Figure 46. -- Many of the sparingly soluble organic contaminants found in the Mississippi River, such as hexachlorobenzene and PCBs (polychlorinated biphenyls), are associated with the suspended sediments. Their distributions on the sediments in the river reflect the sources of the contaminants. Once they are in the river, many of these contaminants are most concentrated in the fractions of the sediment that contain the most organic carbon. Data shown here are for suspended silts collected from the river during downriver sampling cruises in March 1989 and June 1989. Total PCBs are defined here as the sum of the penta-, hexa-, hepta-, and octachlorinated biphenyl homologs. Complete data are tabulated in the report by Rostad, Bishop, and others (1995).

A

Concentrations of hexachlorobenzene in the suspended silt of the Mississippi River main stem clearly show the two principal sources of this contaminant. The marked increases near river kilometer 1530 are due to the input from the Ohio River, which supplies many times more hexachlorobenzene than does the Upper Mississippi River. A further increase in hexachlorobenzene occurs as the Lower Mississippi River flows through Louisiana between St. Francisville (river kilometer 428) and Belle Chasse (river kilometer 117); this increase is probably related to the large number of halogenated hydrocarbon manufacturing industries along this part of the river between Baton Rouge and New Orleans.

B

Organic carbon in the suspended sediment is a strong determinant in the adsorption of sparingly soluble organic contaminants like hexachlorobenzene. When the contaminant concentrations are normalized to organic carbon, as they are in this graph, some of the irregularities in the contaminant-silt relation are smoothed out. The suspended silt collected at river kilometer 1920 (Winfield, Missouri) during March 1989 contained a very high proportion of organic carbon and a high concentration of hexachlorobenzene for the silt of the Upper Mississippi River (fig. 46A). However, when the hexachlorobenzene concentration is expressed in relation to organic carbon, as it is here in figure 46B, the concentration is consistent with concentrations measured at other times and locations in the upper river.

C

The concentrations of total PCBs in the suspended silt in the middle and lower reaches of the Mississippi River are less visibly related to specific sources than those of a contaminant like hexachlorobenzene. One reason for this is that the sources of PCBs have been more diverse and disseminated than those of hexachlorobenzene. Perhaps even more important is that in the nearly two decades since the disposal of PCBs into the river has been curtailed, the repeated deposition and resuspension of contaminated silts has resulted in a partial homogenization of PCB concentrations along the length of the river and a blurring of some of the more abrupt changes in distribution that indicated specific sources.

D

PCBs also are strongly associated with the organic carbon in the suspended sediments of the Mississippi River. By expressing the PCB concentrations in relation to organic carbon, this graph eliminates the seemingly anomalous concentration of PCBs measured at kilometer 1920 during March 1989 (fig. 46C) and emphasizes the marked increase in PCB concentration associated with the influx of suspended sediment from the Ohio River.

Introduction

Many toxic chemicals, including most heavy metals and the majority of the U.S. Environmental Protection Agency (USEPA) Priority Pollutants, are primarily associated with sediments (Ongley and others, 1992). These sparingly soluble compounds may be transported in the environment by direct surface runoff or by resuspension of contaminated bed sediments in rivers. Because they are not water soluble, they associate with sediment or move into organisms and thus enter into food chains. They can bioaccumulate; for example, after a fish eats a contaminated insect, the contaminants are not degraded or metabolized but instead are retained by the fat in the fish tissue. The fish eats many insects and continues to accumulate the contaminants. As additional contaminated insects are consumed, higher concentrations build up in the fish tissue. Biomagnification takes place, when each level of the food chain contains higher concentrations than the level below it. If a bird eats this highly contaminated fish, the bird may also retain the contaminant in its tissue. If a human eats the fish, the contaminant can also accumulate in human tissue.

PCBs, chlordane, DCPA, and hexachlorobenzene usually are detected in association with the suspended material in the river but not in the water itself. This behavior is expressed numerically as "partition coefficients," which for these compounds are large. The partition coefficient is a laboratory-determined ratio of the amount of a compound associated with the sediment to the amount associated with the aqueous phase. It is characteristic of each chemical and ranges from 1 to 10,000,000. This ratio allows prediction of how the compound behaves in the environment, because it indicates whether a compound will be found in the aqueous phase or the sediment phase. A small partition-coefficient value means the compound will have an affinity for the aqueous phase, not the sediment phase, and will most likely remain dissolved under most conditions. A large value means the compound will instead associate with an organic phase, like the organic carbon of sediments, or in the fat of fish tissue. Compounds with large partition coefficients are more likely to be found associated with available organic matter, whether it is bed sediment, suspended sediment, fish tissue, or aquatic insects.

Synthetic compounds enter the river from point and nonpoint sources. Point sources are discrete, such as manufacturing plants or accidental spills. Although there can be many point sources, each has an individual effect on the environment at that point as well as downstream. Nonpoint sources are more general, reflecting overall widespread usage of chemicals in the environment, such as atrazine from farmlands. PCBs, which probably came from industrial point sources in the past, are now widely distributed in the environment, and as a result, are now considered to be from nonpoint sources. Chlordane was used extensively on corn and, more recently, on lawns and in residences for termite control before it was banned by USEPA. DCPA use is currently residential and agricultural. Both chlordane and DCPA have been so widely used that their sources are now considered nonpoint. Hexachlorobenzene, however, has specific industrial uses as well as agricultural uses, and therefore can come from point sources as well as nonpoint sources.

The toxicity of these compounds is quite high to some organisms. Most of the time, however, only trace concentrations of these compounds are present in environmental waters. The precise health effects of such trace concentrations are often subtle and, therefore, difficult to pinpoint in the environment or the general human population. For example, chlordane is classified by USEPA as a probable human carcinogen (U.S. Environmental Protection Agency, 1986). Because chlordane is present in trace amounts in food and water (Cochrane and others, 1975), and because humans can barely metabolize it, the amount present in human fatty tissue has been found to increase with age. The main target organ for carcinogenic effect from chlordane is the liver, but other organs of the body are also susceptible to its effects. In a 1973--74 study conducted in Mississippi and Arkansas, 54 percent of human breast milk samples contained chlordane residues at trace levels or higher (Strassman and Kutz, 1977). At large concentrations, chlordane is known to cause blood anemia, leukemia, and problems of the central nervous system.

PCBs from Ohio and Missouri Rivers

(Click on image for a larger version, 50K)

Figure 47. -- Concentrations of organic contaminants on suspended sediment in the Mississippi River are variable, depending on which major tributary is contributing the most sediment at the time of sampling. Shown here are concentrations of total PCBs on suspended silts in the Lower Mississippi River during two periods of 1990---one when most of the silt in the main stem was being supplied by the Ohio River, and another when most was being supplied by the Missouri River. Complete data are tabulated in the report by Rostad, Bishop, and others (1995).

A

During March 1990, most of the silt being transported in the Mississippi River system was in or from the Ohio River, and PCB concentrations were accordingly high. Data in this graph extend from Uniontown, Kentucky, on the Ohio River, down the Ohio to its confluence with the Mississippi River, and down the Mississippi to the Gulf of Mexico. That is, the uppermost 400 kilometers on the graph represent the lowermost 400 km of the Ohio River, and the Upper Mississippi is represented as a tributary to the Ohio and Lower Mississippi Rivers. Concentrations of total PCBs on the suspended silts were greatest in the Ohio River. All the major tributaries that joined the Ohio (Wabash, Cumberland, Tennessee, and Upper Mississippi Rivers) transported silts whose PCB concentrations were less than those in the Ohio River, which drains a more industrialized area.

B

During June 1990, most of the silt being transported in the Mississippi River was coming from the Missouri River, and PCB concentrations were accordingly lower than they had been during March. Data in the graph show PCB concentrations in the main-stem Mississippi River from Grafton, Illinois, to below New Orleans, Louisiana. PCB concentrations on silts were greater in the tributary Illinois and Ohio Rivers than in the main stem. Although most of the silt was coming from the Missouri River, no silt sample was taken at the time from the Missouri River itself. The probable concentration range of 10--40 nanograms per gram (ng/g) shown in the graph for PCBs in the Missouri River was inferred from the five other measurements that were made between 1988 and 1991 on suspended silts from the Missouri. Of the major tributaries to the Mississippi River that were sampled from 1988 through 1992, the Missouri River consistently supplied suspended silts having the smallest concentration of PCBs.

Contaminants Associated with Suspended Silt

Chlordane also was detected in almost every silt sample we analyzed from the Mississippi River and its tributaries. In general, its distribution parallels that of PCBs throughout the river system: highest concentrations in the Upper Mississippi are found near Lake Pepin; largest inputs to the Lower Mississippi come from the Ohio River. Local exceptions do occur, such as the elevated concentrations of chlordane detected near Keokuk, Iowa, during July and October 1991. The higher concentrations of chlordane associated with the silt at this site also are reflected in the fish, for which the State of Missouri has issued a fish-consumption advisory, (Missouri Department of Health, 1990).

Total chlordane in this study is the sum of cis-chlordane, trans-chlordane, and trans-nonachlor. Of the three components, trans-nonachlor is the most resistant to environmental degradation. The ratio of cis- and trans-chlordane to trans-nonachlor is an indicator of the proportion of recently released chlordane to "weathered" chlordane. It does not, however, provide a specific time estimate of the weathering period involved. Weathered chlordane was associated with the silt during sample periods of spring runoff (March 1989 and 1990), which indicated the source of the chlordane to be resuspension of bed sediment or runoff of previously contaminated sediment rather than more recent environmental inputs of chlordane.

DCPA also is ubiquitous on the suspended silts of the Mississippi River and its principal tributaries. Concentrations of DCPA on silts usually do not vary widely from one part of the river to another, which indicates that DCPA comes from widely dispersed nonpoint sources rather than from specific regions or tributaries.

Hexachlorobenzene, in contrast, has more specific sources in the Mississippi River Basin. Suspended silts of the Ohio River carry concentrations of hexachlorobenzene that are consistently greater than those of the Upper Mississippi and its tributaries (Illinois and Missouri Rivers) by factors of 5 or more. The concentration of hexachlorobenzene increases significantly between St. Francisville (above Baton Rouge) and Belle Chasse (below New Orleans), Louisiana, as the river flows through an industrial corridor.

PCBs in Silts and Colloids

(Click on image for a larger version, 66K)

Figure 48. -- Large proportions of the organic contaminants are associated with the colloidal fraction of the sediments suspended in the Mississippi River. "Colloid" was defined operationally in this study as the suspended matter that passed through the ultracentrifuge but was removed by the ultrafilter (see the chapter on "Sampling the Big Rivers"). In effect, the boundary between "colloid" and "silt" was about 0.1 micrometer, or one ten-thousandth of a millimeter. Data shown in the figure are from samples collected in the Mississippi River and four major tributaries during the downriver sampling cruise of July-August 1991. Complete data are tabulated in the report by Rostad, Monsterleet, and others (1995).

A

In suspended sediments of the Mississippi River, concentrations of total PCBs usually are greater for colloids than for silts---not only because the colloidal material has more exposed surface area onto which PCBs can be adsorbed, but because colloids usually contain larger proportions of organic carbon. Data in the graph show that PCB concentrations are greatest in the uppermost Mississippi River for both the silt and colloid. Throughout most of the river system and the major tributaries, however, concentrations for the colloids are consistently greater than those for the silts.

B

In terms of the total loads of PCBs transported by the Mississippi River, silt transport is more important than colloidal transport, simply because there is much more silt than colloid in suspension in the river. Transports shown in the graph are computed by multiplying the concentrations shown in figure 48A by the concentrations of silt and colloid in suspension and the discharge of the water. Even though concentrations of PCBs in colloids may exceed those in silts by factors of 2 or 3, the quantities of silt in suspension usually exceed those of colloids in suspension by factors near 10. Therefore, the total quantities of PCBs transported down the Mississippi River by silts are usually 3 to 5 times greater than those transported by colloids.

Contaminants Associated with Suspended Colloid

Most toxic compounds travel in association with organic carbon, and the attribute of colloids that makes them especially important in the transport of pollutants is their high content of organic carbon. The colloid sampled in the Mississippi River averaged 15 percent organic carbon, which was substantially higher than that of the silt (3 percent). Colloidal material from the Upper Mississippi River was higher in organic carbon content than that from the Lower Mississippi River.

Colloid particles do not settle. Once a contaminant is adsorbed to a colloidal particle, its transport is probably consistent along the length of the river system. Silt transport, however, varies greatly in response to the variations in velocity, turbulence, and discharge of the river. The amount of contaminants transported by the colloids should be consistent, whereas the amount transported by the silt may vary widely.

Contaminants Associated with Fish

A study by the Tennessee Department of Health and Environment that included more than 500 km of the Mississippi River upstream and downstream from Memphis showed how, in addition to the consistent background of these compounds, concentrations in fish can sharply increase near more heavily contaminated areas (Collins and others, 1989). Local studies have found that most severe cases of fish contamination occur near large urban areas (Iowa Department of Natural Resources, 1992).

PCBs in Bed Sediments

(Click on image for a larger version, 66K)

Figure 49. -- Concentrations of PCBs stored in the bed sediments in the navigation pools of the Upper Mississippi River vary from place to place and according to the times at which the sediments were deposited.

A

In the surficial bed sediments (that is, the sediments no more than 10 cm below the bottoms of the pools), the concentrations of total PCBs are large below Minneapolis-St. Paul, reach their greatest values in Lake Pepin, and are significantly smaller in the pools downriver of Lake Pepin. Data in the graph are from composite samples that were collected during 1991--92 at 15--20 locations in the lower reaches of each navigation pool and combined into a single sample (one per pool) for the analysis of total PCBs. Locations of the navigation pools are shown in the map on the left side of the figure. "UP" and "LP" in the graph denote Upper and Lower Lake Pepin. Data are tabulated in the report by Iverson and others (1995). Lake Pepin is the only large natural lake on the Mississippi River main stem below Minneapolis-St. Paul, and it apparently functions as the principal repository of many of the sediment-associated contaminants discharged from the metropolitan area. Furthermore, the high concentrations of organic carbon in this region of the river (see fig. 29D) probably increase the attraction between PCBs and the local sediments.

B

The history of PCB contamination in the Upper Mississippi River is recorded in the sediments that have been deposited and stored in the navigation pools. Shown in the graph are the concentrations of total PCBs in a core taken in March 1989 from the bed sediments in a shallow backwater in the lower end of Pool 2, a few miles downriver from Minneapolis-St. Paul. Data were provided by Dan Helwig of the Minnesota Pollution Control Agency, Water Quality Division. Very high PCB concentrations of 2,000--3,000 nanograms per gram (ng/g) are found in the sediments deposited when large quantities of PCBs were allowed to enter the river, probably during the 1950s and 1960s. Following the cessation of the manufacture of PCBs, and the passage of State and Federal regulations restricting their use and disposal during the 1970s, concentrations of PCBs on sediments decreased markedly. The uppermost 35 cm of the core consisted of sediments in which concentrations consistently averaged about 150 ng/g. Because of the proximity to the Twin Cities, even these more recent PCB concentrations are greater than those in sediments stored farther downriver.

Chlordane and PCBs in Catfish

(Click on image for a larger version, 50K)

Figure 50. -- Although banned for years, many halogenated organic contaminants persist in rivers and are ingested or absorbed by fish. Shown in the graphs are the concentrations of chlordane and PCBs in catfish taken from the Mississippi River and four of its tributaries during July-August 1987. One to four fish caught by hook and line at each location were homogenized in a blender and analyzed for PCBs and chlordane, as well as for other compounds such as hexachlorobenzene, DCPA, and DDT and its degradation products. Complete data are tabulated in the report by Leiker and others (1991).

A

Chlordane is especially concentrated in catfish. Concentrations of chlordane shown here are substantially greater than those on the suspended sediment in the Mississippi River, which are almost always less than 10 (often less than 5) nanograms of chlordane per gram of suspended silt. Chlordane is especially lipophilic-that is, especially soluble in the fatty tissues of the fish-and difficult to metabolize. The action level for chlordane of 300 nanograms per gram, specified by the U.S. Food and Drug Administration (USFDA, 1986) corresponds to the top of the graph.

B

PCB concentrations in catfish in the Mississippi River are comparable to those on suspended sediment. Although the "total" PCB shown here refers to the sum of the tetra-, penta-, hexa-, and heptachlorobiphenyls, this total is comparable to that of the penta-, hexa-, hepta-, and octachlorobiphenyls portrayed in the other figures of this chapter. The action level for PCBs of 2,000 nanograms per gram, specified by USFDA (1986), is well above the range of values shown here.

REFERENCES

Baker, J.E., Capel, P.D., and Eisenreich, S.J., 1986,

Influence of colloids on sediment-water partition coefficients of polychlorobiphenyl congeners in natural waters: Environmental Science and Technology, v. 20, p. 1136-1143.

Cochrane, W.P., Lawrence, J.F., Lee, Y.W., Maybury, R.B., and Wilson, B.P., 1975,

Gas-liquid chromatographic determination of technical chlordane residues in food crops--Interpretation of analytical data: Journal of the Association of Official Analytical Chemists, v. 58, no. 5, p. 1051-1061.

Collins, R., Robinson, M., and Sinclair, R., 1989, Mississippi River fish contamination study, 1982:

Tennessee Department of Health and Environment, Division of Water Pollution Control, November 1989, 18 p.

Iowa Department of Natural Resources, 1992,

Report of the analysis of fishes collected during 1990 from the Regional Ambient Fish Tissue Monitoring sites in Iowa: Iowa Department of Natural Resources, Water Quality Section, Surface and Groundwater Protection Bureau, Environmental Protection Division, 47 p.

Iverson, J.L., Rostad, C.E., Bishop, L.M., Sullivan, J.L., and Moody, J.A., 1995,

Selected organic compounds in Moody, J.A., ed., Hydrologic, sedimentologic, and chemical data describing surficial bed sediments and water in navigation pools of the Upper Mississippi River, July 1991-April 1992: U.S. Geological Survey Open-File Report 95-708.

Leiker, T.J., Rostad, C.E., Barnes, C.R., and Pereira, W.E., 1991,

A reconnaissance study of halogenated organic compounds in catfish from the lower Mississippi River and its major tributaries: Chemosphere, v. 23, no. 7, p. 817-829.

Missouri Department of Health, 1990,

1990 Health advisories on chlordane in fish: Jefferson City, Missouri Department of Health, 16 p.

Ongley, E.D., Krishnappan, B.G., Droppo, I.G., Rao, S.S., and Maguire, R.J., 1992,

Cohesive sediment transport--Emerging issues for toxic chemical management: Hydrobiologia, v. 235/236, p. 177-187.

Rees, T.F., and Ranville, J.F., 1989,

Characterization of colloids in the Mississippi River and its major tributaries, in Mallard, G.E., and Ragone, S.E., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the Technical Meeting, Phoenix, Arizona, September 26-30, 1988: U.S. Geological Survey Water-Resources Investigations Report 88-4220, p. 513-522.

Rostad, C.E., Bishop, L.M., Ellis, G.S., Leiker, T.J., Monsterleet, S.G., and Pereira, W.P., 1995,

Organic contaminant data for suspended sediment from the Mississippi River and some of its tributaries, May 1988-June 1990: U.S. Geological Survey Open-File Report 93-360.

Rostad, C.E., Monsterleet, S.G., Bishop, L.M., and Ellis, G.S., 1995,

Polychlorinated biphenyls associated with suspended silt and clay and colloid from the Mississippi River and some of its tributaries, July-November 1991: U.S. Geological Survey Open-File Report 94-484.

Royal Society of Chemistry, 1991,

The agrochemicals handbook, 3d ed.: Cambridge, England, Royal Society of Chemistry, variously paged.

Schmitt, C.J., Ribick, M.A., Ludke, J.L., and May, T.W., 1983,

National pesticide monitoring program: Residues of organochlorine chemicals in freshwater fish, 1976-79: U.S. Fish and Wildlife Service Resource Publication 152, 62 p.

Schmitt, C.J., Zajicek, J.L., and Ribick, M.A., 1985,

National pesticide monitoring program--Residues of organochlorine chemicals in freshwater fish, 1980-1981: Archives of Environmental Contamination and Toxicology, v. 14, p. 225-260.

Strassman, S.C., and Kutz, R.W., 1977,

Insecticide residues in human milk from Arkansas and Mississippi, 1973-1974: Pesticide Monitoring Journal, v. 10, p. 130-133.

U.S. Environmental Protection Agency, 1986,

Guidelines for carcinogen risk assessment: Federal Register, v. 51, no. 185, September 24, 1986, p. 33992-34003.

U.S. Food and Drug Administration, 1986,

Compliance policy guide 7141.01: Pesticide residues in food and feed--enforcement criteria, chap. 41--Pesticides.