Volume II Chapter 4.0 Pages 1 of 4 page next page 2

4. LEVELS OF CDD, CDF, AND PCB CONGENERS IN ENVIRONMENTAL

4.1. INTRODUCTION 4-1

4.2. CONCENTRATIONS IN SOIL 4-2

4.2.1. North American Data 4-2

4.2.2. European Data 4-4

4.2.3. Soil Summary 4-5

4.3. CONCENTRATIONS IN WATER 4-6

4.3.1. North American Data 4-6

4.3.2. European Data 4-7

4.3.3. Water Summary 4-7

4.4. CONCENTRATIONS IN SEDIMENT 4-8

4.4.1. North American Data 4-8

4.4.2. European Data 4-11

4. LEVELS OF CDD, CDF, AND PCB CONGENERS IN ENVIRONMENTAL MEDIA AND FOOD

4.1. INTRODUCTION

Polychlorinated dibenzo-p-dioxins (CDDs), polychlorinated dibenzofurans (CDFs), and polychlorinated biphenyls (PCBs) have been found throughout the world in practically all media including air, soil, water, sediment, fish and shellfish, and other food products such as meat and dairy products. Also, not unexpectedly, considering the recalcitrant nature of these compounds and their physical/chemical properties (i.e., low water solubilities, low vapor pressures, and high Kows and Kocs), the highest levels of these compounds are found in soils, sediments, and biota (ppt and higher); very low levels are found in water and air (ppq and lower).

The widespread occurrence observed is not unexpected considering the numerous sources that emit these compounds into the atmosphere (discussed in Chapter 3), and the overall resistance of these compounds to biotic and abiotic transformation. (See Chapter 2.) Part-per-trillion levels of CDDs/CDFs have been found in everyday materials that are contaminated with dust--clothes dryer lint (2.4 to 6.0 ng TEQ/kg; vacuum cleaner dust (8.3 to 12 ng TEQ/kg); room air filters (27 to 29 ng TEQ/kg); and house furnace filter dust (170 ng TEQ/kg) (Berry et al., 1993). Although Berry et al. (1993) only analyzed one or two samples of these materials, the findings suggest that these compounds may be ubiquitous.

The purpose of this chapter is to provide an overview of the concentrations at which these compounds have been found in the environment and food based on data presented in the published literature. This literature summary is not all inclusive, but is meant to present the reader with a general overview of values reported in the recent literature. Only data from Government-sponsored monitoring studies and studies reported in the peer-reviewed literature are discussed in this chapter.

Data are presented as they were presented in the original studies/reports. No attempt has been made to verify or assess the adequacy of the quality assurance/quality control measures employed in these studies beyond those described in the published reports. The final section of this chapter discusses the mechanisms by which these compounds could enter the food chain.

4.2. CONCENTRATIONS IN SOIL

Tables B-1 and B-2 (Appendix B) contain summaries of data from several of the numerous studies in the published peer-reviewed literature regarding concentrations of CDDs and CDFs in soil. Data on coplanar PCB congener soil concentrations were not found in the literature; the PCB soil concentration data found in the literature were reported as either total PCB concentrations or concentrations of Aroclor PCB mixtures. Descriptions of several of the studies summarized in Appendix B are presented below.

4.2.1. North American Data

Soil sampled in 1987 from the vicinity of a sewage sludge incinerator was compared with soil from rural and urban sites in Ontario, Canada, by Pearson et al. (1990). Soil in the vicinity of the incinerator showed a general increase in CDD concentration with increasing degree of chlorination. Of the CDFs, only OCDF was detected (mean concentration 43 ppt).

Rural woodlot soil samples contained only OCDD (mean concentration of 30 ppt). Soil samples from undisturbed urban parkland settings revealed only HpCDDs and OCDD, but all CDF congener groups (Cl4 to Cl8) were present.

Those samples showed an increase in concentration from the HpCDDs to OCDD and PeCDFs to OCDD. The TCDFs had the highest mean value (29 ppt) of all the CDF congener groups. Resampling of one urban site in 1988, however, showed high variability in the concentrations of CDDs and CDFs.

Data were collected on CDD and CDF levels in soil samples from industrial, urban, and rural sites in Ontario and some U.S. Midwestern States (Birmingham, 1990). The levels of CDD/CDF in rural soils were primarily nondetected (ND), although the HpCDDs and OCDD were found in a few samples.

In urban soils, the tetra- through octa-congener groups were measured for both CDDs and CDFs. The HpCDDs and OCDD dominated the homologue profile and were two orders of magnitude greater than in the rural soils. These soils also contained measurable quantities of the TCDDs and PeCDDs. Industrial soils did not contain any TCDDs or PeCDDs, but they contained the highest levels of the TCDFs, HpCDFs, and OCDF.

In another study, soils from industrialized areas of a group of cities from Midwestern and Mid-Atlantic States (Michigan, Illinois, Ohio, Tennessee, Pennsylvania, New York, West Virginia, Virginia) were analyzed for levels of 2,3,7,8-TCDD (Nestrick et al., 1986). Many of the samples were taken within 1 mile of major steel, automotive or chemical manufacturing facilities, or municipal solid waste incinerators. Concentrations of 2,3,7,8-TCDD measured in this study ranged from ND to 9.4 ppt.

EPA conducted a 2-year nationwide study to investigate the national extent of 2,3,7,8-TCDD contamination (U.S. EPA, 1987). The results of this large study were summarized broadly in the primary reference (i.e., the number and types of samples per site and range of detection).

The method used to analyze samples for five of the seven "tiers" of the study had a detection limit in soil, sediment, and water of 1 part per billion (ppb). [Each "tier" of sites is a grouping of sites with a common past or present use (e.g., industrialized, pristine, etc.)]. Only Tier 5 (sites where pesticides derived from 2,4,5-trichlorophenol (TCP) have been or are being used for commercial purposes), and Tier 7 (ambient sampling for fish and soil) had detection limits of 1 ppt. Consequently, the data from this study are not included in the tables, but some observations from this study with regard to soil contamination are discussed below.

Soil concentrations found in most of the 100 Tier 1 and 2 sites (i.e., sites already on or expected to be on the NPL list) were in the ppb range; although in a few sites where concentrated 2,4,5-TCP production wastes were stored or disposed, concentrations were as high as 2,000 parts per million (ppm).

Offsite soil contamination of concern was confirmed in 7 of the 100 Tier 1 and 2 sites, with soil concentrations in the ppb range. Eleven of 64 Tier 3 sites (facilities and associated disposal sites where 2,4,5-TCP and its derivatives were formulated into pesticide products) were found to have soil concentrations exceeding 1 ppb, and in 7 of 11 sites where contamination was found, only one or two soil samples were above 1 ppb.

Fifteen of 26 Tier 5 sites (areas where 2,4,5-TCP and pesticide derivatives had been or were being used) had concentrations above 1 ppt, and one of those had a single detection of 6 ppb. Two-thirds of all detections at the Tier 5 sites were below 5 ppt.

Three of 18 Tier 6 sites (organic chemical and pesticide manufacturing facilities where improper quality control on production processes could have resulted in 2,3,7,8-TCDD being introduced into the wastestreams) had soil concentrations that exceeded the detection limit of 1 ppb, although these levels were limited to one or two samples per site.

Seventeen of the 221 urban soil sites and 1 of the 138 rural sites from Tier 7 (background sites not expected to have contamination) had soil concentrations exceeding 1 ppt. The highest concentration detected (11.2 ppt) was found in an urban sample. The results from Tier 7 are consistent with the other studies discussed above regarding soil concentrations of 2,3,7,8-TCDD in nonindustrial settings.

4.2.2. European Data

Soil samples from rural and semi-urban sites in England, Wales, and lowland Scotland showed a general increase in concentration from the TCDDs to OCDD, whereas the CDF levels showed very little variation between the congener groups (Creaser et al., 1989). Concentrations of 2,3,7,8-TCDD at those sites ranged from <0.5 to 2.1 parts per trillion (ppt).

The median values for the TCDDs to OCDD were 6.0, 4.6, 31, 55, and 143 ppt, respectively. The median values for the TCDFs to OCDF were 16, 17, 32, 15, and 15 ppt. Evaluation of soil data from urban sites in the same geographical area showed that the mean levels for the CDD and CDF congeners were significantly greater (p<0.01) than those for rural and semi-urban background soils (Creaser et al., 1990).

Concentrations of 2,3,7,8-TCDD at the urban sites ranged from <0.5 to 4.2 ppt. The median values for the TCDDs to OCDD were 40, 63, 141, 256, and 469 ppt, respectively. The median values for the TCDFs to OCDF were 140, 103, 103, 81, and 40 ppt.

The significantly elevated levels of the lower congeners, together with higher overall CDD/CDF concentrations, are indicative that local sources and short-range transport mechanisms are major contributors of CDDs and CDFs to urban soils.

Analysis of four sites in Hamburg, Germany, contaminated by an organochlorine pesticide manufacturing company showed patterns of CDD and CDF distribution that are similar to the urban and industrial sites examined in England, Wales, and Scotland (Sievers and Friesel, 1989).

The study indicated that CDDs and CDFs showed a regular increase in concentration with increasing degree of chlorination (although individual data points were not presented). The maximum concentrations of 2,3,7,8-TCDD ranged from 900 ppt to 874,000 ppt. The very high concentrations of 2,3,7,8-TCDD at the sites were attributed to an admixture of wastes from 2,4,5-T production.

A soil sampling survey in Salzburg, Austria, also showed that the concentrations of CDD/CDFs were higher in urban and industrial sites than in rural sites (Boos et al., 1992). The total CDD content of the soils ranged from 33.7 to 1236.7 ppt for urban sites; 92.2 to 455 ppt for industrial sites; and 7.1 to 183.6 ppt for rural sites. The total CDF content of the soils ranged from 45.6 to 260.8 ppt for urban sites; 53.0 to 355.3 ppt for industrial sites; and 12.0 to 77.7 ppt for rural sites.

Rotard et al. (1993) measured CDD/Fs in soil samples collected from forest, grassland. and plowland sites in western Germany. The highest mean concentration of CDD/Fs were found in the sursoil and topsoil layers of deciduous (38.0 ng TEQ/kg dry matter) and coniferous forests (36.9 ng TEQ/kg dry matter). Grassland and plowland sites had mean concentrations of 2.3 ng TEQ/kg dry matter and 1.7 ng TEQ/kg dry matter, respectively.

4.2.3. Soil Summary

Some general observations for CDD and CDF levels in soils are possible from the data presented in the various soil studies discussed above:

Generalizations about the prevalence of specific congeners within a congener group are not possible.

As the degree of chlorination increases, the concentrations increase. Concentrations of the hepta- and octa-chlorinated congeners are generally higher than the tetra-, penta-, and hexa-chlorinated congeners.

Concentrations associated with industrial sites clearly are the highest, with concentrations in the hundreds to thousands of parts per trillion.

Concentrations in settings identified as urban are higher than those in areas identified as rural.

Based on the above studies, 95 samples were selected as representing background conditions in the United States. The mean TEQ level was estimated to be 8 ppt assuming that nondetects equal half the detection limit. Similarly, 133 background samples were selected from the European studies and were estimated to have a mean of 9 ppt of TEQ.

4.3. CONCENTRATIONS IN WATER

Tables B-3 and B-4 (Appendix B) contain summaries of data from the limited number of published studies regarding concentrations of CDDs and CDFs in water. Data on coplanar PCB congener water concentrations were not found in the literature. Several of these studies are discussed below.

4.3.1. North American Data

A survey of 49 drinking water supplies in Ontario, including supplies in the vicinity of chemical industries and pulp and paper mills, was initiated in 1983 (Jobb et al., 1990). As of February 1989, 4,347 congener analyses had been performed on 399 raw and treated water samples. OCDD was detected in 36 of 37 positive results and ranged from 9 to 175 ppq in raw samples (33 positive samples) and 19 to 46 ppq in treated samples (4 positive samples).

These low concentrations were found primarily in water obtained downstream of industrialized areas in the St. Clair/Detroit River system. Concentrations of 2,3,7,8-TCDD were not detected in any sample. Because CDDs and CDFs are hydrophobic concentrations of compounds and consequently have a tendency to sorb onto particulate matter in water, conventional water treatment processes are expected to be effective in removing the contaminants along with the particulates.

This is substantiated by the fact that 33 of the 37 positive results were raw water samples. Because of the relatively low levels of CDDs detected in the samples, it is difficult to ascertain whether the CDDs were particulate-associated or dissolved.

A survey of 20 community water systems throughout New York State was conducted in 1986 (Meyer et al., 1989). The sampling sites were representative of the major surface source waters in New York. They included sources receiving industrial discharges or known to contain dioxin-contaminated fish, as well as waters in more remote areas. TCDFs were detected in the finished water at the Lockport facility (duplicate samples had concentrations of 2.1 and 2.6 ppq).

Except for a trace of OCDF detected at one location, no other CDDs/CDFs were detected in finished water at any of the other 19 community water systems surveyed. Raw water sampled at the Lockport facility contained concentrations of TCDDs (1.7 ppq) as well as TCDFs to OCDF (18, 27, 85, 210, and 230 ppq, respectively). As can be seen from the data, the CDF congener group concentrations increased with increasing chlorine number.

4.3.2. European Data

CDDs in surface water samples collected from the Eman River in southern Sweden generally increased in concentration from the TCDDs to OCDD; whereas the CDF levels showed very little variation between the congener groups (Rappe et al., 1989b). In general, however, the levels of CDFs were higher than the levels of CDDs.

Concentrations of 2,3,7,8-TCDF were 0.022 parts per quadrillion (ppq) in Jarnsjon and 0.026 ppq in Fliseryd. The filtered water, before chlorination and distribution as drinking water, had no detectable tetra-, penta-, or hexa-chlorinated congeners of CDDs or CDFs, but the HpCDDs and OCDD were detected at 120 and 170 ppq, respectively.

4.3.3. Water Summary

Some general observations for CDD and CDF levels are possible from the limited data available from the various water studies above:

CDDs/CDFs are seldom detected in drinking water at ppq levels or higher.

Raw water samples generally have higher concentrations of CDDs/CDFs than finished water samples.

The concentration of CDDs and CDFs in surface water generally increases from the tetra-chlorinated to the octa-chlorinated congener groups.

Based on the above studies, a total of 214 samples were selected as representing background conditions in North America. The mean TEQ level was computed as 0.0056 ppg, assuming that nondetects equal half the detection limit.

It should be noted, however, that OCDD and OCDF were the only congeners for which background data were available. Of the 214 samples analyzed for OCDD, 4 were positive, and 2 out of 22 samples analyzed for OCDF were positive. No appropriate data could be found for Europe.

4.4. CONCENTRATIONS IN SEDIMENT

Tables B-5, B-6 & B-7 (Appendix B) contain summaries of data from several of the numerous studies in the published literature regarding concentrations of CDDs, CDFs, and coplanar PCB congeners in sediment. Several of these studies are discussed in the following paragraphs.

4.4.1. North American Data

In sediment samples collected from estuaries adjacent to an industrial site in Newark, New Jersey, where chlorinated phenols had been produced, the level of OCDD was many times higher than that of 2,3,7,8-TCDD (Bopp et al., 1991).

The study indicated that there probably is a significant regional source (i.e., combustion and/or use of a common wood preservative, pentachlorophenol) for OCDD depleted in 2,3,7,8-TCDD relative to the local industrial source. A high correlation was found between 2,3,7,8-TCDD and 2,3,7,8-TCDF (R2=0.87), which suggests that the industrial site was a major source of 2,3,7,8-TCDF to the natural waters of the study area.

An interesting note is that the bottom section (108-111 cm) of one sediment core contained 2,3,7,8-TCDD at a concentration of 21,000 ppt, the highest concentration measured in the study. This value was consistent with deposition of that sample during the mid to later stages of active 2,4,5-T production at the site from the late 1950s to early 1960s. CDD and CDF in Hudson River sediment samples contained primarily the higher chlorinated (Cl6 to Cl8) CDD and CDF congeners (Petty et al., 1982).

Concentrations of the HpCDDs and OCDD homologues ranged from 5 to 15 ppb, and the OCDD homologue in most instances accounted for more than half of the total CDD residue. Likewise, the HpCDFs and OCDF occurred at the highest levels (ca. 1 ppb).

Surface sediment samples were collected from several estuaries in the United States (Norwood et al., 1989). The sampling sites included Black Rock Harbor in Bridgeport, Connecticut, (an industrialized urban estuary); central Long Island Sound (a relatively clean reference site); Narragansett Bay, Rhode Island, (where chemical industries may have contributed to the input); New Bedford Harbor, Massachusetts, (a section of which is a National Superfund Site because of PCB contamination); and Eagle Harbor, Washington, (the site of a creosote wood treatment facility).

The sediments in New Bedford Harbor were reported to be more heavily contaminated with CDFs, especially with regard to the HxCDF congeners that were greater by a factor of 40 (although individual data points were not presented). In contrast, sediments from Eagle Harbor were practically devoid of CDFs and showed a large increase in the HpCDD and OCDD congeners closer to the treatment facility.

Narragansett Bay and Black Rock Harbor were similar in both concentration and distribution of CDDs and CDFs, and Black Rock Harbor contained slightly higher levels of the tetra- to hepta-CDD and CDF congeners. Sediment from Long Island Sound was cleaner and had a distribution of CDFs between that of Narragansett Bay and Black Rock Harbor. Sediment with the least contamination was collected in New Bedford Harbor, up-river from the PCB facilities; the highest OCDD concentration (1400 ppt) was detected in Eagle Harbor.

Sediment samples from Siskiwit Lake, on Isle Royale, Lake Superior, were examined to evaluate the atmospheric input of CDDs and CDFs to the lake (Czuczwa et al., 1984). The water level in Siskiwit Lake is 17 meters higher than that in Lake Superior, and in addition, there are no anthropogenic inputs in the drainage basin of Siskiwit Lake.

Consequently, the atmosphere is the only source of anthropogenic chemicals in that lake. OCDD was most predominant, and the HpCDD and HpCDF congeners also were abundant. The study indicated that the considerable decrease in concentration of all CDD and CDF between 6 and 8 cm of the sediment core depth (i.e., sediment believed to have been deposited about 1940).

Surficial sediments collected from Jackfish Bay on the north shore of Lake Superior contained moderate concentrations of the TCDF and OCDD congeners, with trace concentrations of other congeners (Sherman et al., 1990). The concentration of OCDD was similar to that found in the Siskiwit Lake sediment samples. The OCDF and OCDD profile for a sediment core collected from Moberly Bay was similar to the surficial sediment pattern.

These congener groups predominated at all depths where detectable concentrations occurred. In addition, low concentrations of the HpCDD and PeCDF and HpCDF congeners were detected. The concentration profile of the HpCDF congener group showed a relatively high value that dropped abruptly to nondetectable (<60 ppt) below a depth of 10 cm. This abrupt change corresponded to a section date 1973 that reflects an operational change at the pulp mill.

A survey of surficial harbor sediments collected near a wood preserving plant in Thunder Bay, Ontario, Canada, on the north shore of Lake Superior, found CDDs and CDFs; the highest concentrations of which occurred at stations closest to the plant dock, and lower concentrations at locations further from the source (McKee et al., 1990). No CDDs or CDFs were detected below the surficial layer. TCDD and PeCDD congeners were below analytical detection limits in all samples.

However, the concentrations of the HxCDDs to OCDD congeners increased with the degree of chlorination. The maximum concentrations of the HxCDDs to OCDD ranged from 5,700 ppt for the HxCDDs to 980,000 ppt for the OCDD. As with the CDD distribution profile, the HxCDFs to OCDF increased with the degree of chlorination.

Bottom surficial sediments (0-3 cm) were collected from the sedimentation basins of Lake Ontario to assess the levels of the various PCB congeners (Oliver and Niimi, 1988). Concentrations of 2,3',4,4',5-PeCB; 2,3,3',4,4'-PeCB; and 2,3,3',4,4',5-HxCB in the sediment were 15, 10, and 2.1 ppb, respectively. A baseline assessment of CDDs and CDFs was performed on the Elk River, a semi-rural area located about 25 miles northwest of Minneapolis-St. Paul, Minnesota (Reed et al., 1990).

Sediment samples were collected from Lake Orono, a reservoir on the Elk River, and from an abandoned gravel pit. Although none of the sediment samples contained 2,3,7,8-TCDD, the gravel pit sediments contained measurable concentrations of TCDFs. Only one Lake Orono sample contained measurable concentrations of 2,3,7,8-TCDF (0.31 ppt) and total TCDF (0.54 ppt).

The gravel pit samples also contained HpCDDs to OCDD and PeCDFs to OCDF. Lake Orono samples contained HpCDDs, OCDD, and HpCDF congeners. The HpCDDs ranged from 7.3 ppt in the lake inlet to 110 ppt in the gravel pit and the lake, near the dam. OCDD concentration ranged from 450 ppt in the gravel pit to 600 ppt in the middle of Lake Orono. The sediment profiles reflected combustion source influences.

The Sheboygan River, a Wisconsin tributary to Lake Michigan, is polluted with PCBs from the mouth to about 14 miles upstream (Sonzogni et al., 1991). That portion of the river is a Superfund site as well as one of the Great Lakes "Areas of Concern." Sediment cores were collected at Rochester Park, near the original source of the PCBs, about 14 miles upstream from the mouth.

The PCB congeners 2,3',4,4',5-PeCB; 2,3,3',4,4'-PeCB; and 3,3',4,4'-TCB were detected in all samples and ranged from about 5 to 1500 ppb. The remaining coplanar PCB congeners were detected less frequently and ranged from nondetectable to slightly over 100 ppb. The PCB congener 2,3',4,4',5-PeCB appears to be the most common coplanar PCB in environmental samples and was found in the Sheboygan River sediments in the highest weight percent. The eight toxic PCBs detected in this study were present in relatively low concentrations compared to total PCBs or other more abundant congeners.

Sediments collected from Waukegan Harbor in Lake Michigan contained the coplanar PCB congeners 3,3',4,4'-TCB and 2,3,3',4,4'-PeCB (Huckins et al., 1988). The percentage of 3,3',4,4'-TCB in the samples averaged (0.16 percent 0.15) varied by 1.4 orders of magnitude, with concentrations ranging from 13 to 27,500 ppb. The percentage of 2,3,3',4,4'-PeCB averaged 0.66 percent 0.37, with concentrations ranging from 102 to 131,000 ppb.

In another Lake Michigan study, sediment samples collected from Green Bay contained concentrations of all 11 coplanar PCB congeners (Smith et al., 1990). The dominant congeners were 2,3,4,4',5-PeCB and 2,3,3',4,4'-PeCB with concentrations of 11 and 5.8 ppb, respectively.

4.4.2. European Data

Sediment samples from the vicinity of a magnesium production plant in Norway were analyzed for CDDs and CDFs (Oehme et al., 1989). The concentration distribution of CDD and CDF congeners was rather homogeneous except for a slight decrease at a sampling station further downstream of the plant. However, the deeper sediments (4-6 and 11-13 cm depth) at that site had somewhat higher levels.

Another sampling station even further downstream had concentrations that were a factor of 4 to 10 lower, thereby indicating substantial transport of CDDs and CDFs. The TCDF congener profiles were the same as those for magnesium production. In addition, the PeCDF congener profiles were very similar to those found in the wastewater.

Trapped sediments from the archipelago of Stockholm, Sweden, displayed CDD and CDF congener distribution patterns that were very similar to those exhibited in total air and air particulates (Rappe and Kjeller, 1987). The HpCDDs, OCDD, and HpCDF were the dominant congener groups in the sediment.

Bottom surface sediment samples collected from the Baltic Sea showed interesting CDD and CDF distribution patterns (Rappe et al.,1989a). The background samples, one between the Swedish and Soviet coasts and the other between the Swedish and Finnish coasts, contained similar levels and distribution profiles.

The study indicated that the pattern of the TCDF congeners at these sites was typical of the "incineration pattern" (i.e., patterns resulting from MSW incineration, car exhausts, steel mills, etc.) which also had been found in samples of air and air particulates.

However, sediment samples collected at a distance of 4 to 30 km from a pulp mill revealed a congener distribution pattern typical of bleaching mills. The TCDFs found in the sediment 4 km from the pulp mill contained only two major congeners. The sediment collected 30 km from the mill displayed the same pattern.

Surface sediments collected from 18 lake areas in central Finland were analyzed for CDDs, CDFs, and PCBs. Although 2,3,7,8-TCDD was not detected in any samples, two other TCDD congeners that are common and abundant in pulp mill effluents were detected.

In addition, the HxCDD to OCDD congener groups as well as the HpCDF and OCDF congener groups not normally linked to pulp mills, also were detected. The study suggested that these may have resulted from combustion operations in the more densely populated and industrialized areas in south Finland. Coplanar PCBs were detected at low background levels.

The most toxic PCB congener, 3,3',4,4',5-PeCB (IUPAC No. 126), was found only in one area located nearest to a PCB leakage point. The concentration of PCB 126 at that site was 110 ppt.

Evaluation of sediments in Hamburg Harbor in Germany revealed high concentrations of the TCDDs through OCDD (mean concentrations of 564, 1112, 2744, 4040, and 7560 ppt, respectively) and the TCDFs through OCDF (mean concentrations of 526, 2980, 4106, 2358, and 2712 ppt) (Gotz et al., 1990).

The average concentration of 2,3,7,8-TCDD was 375.3 ppt. The high concentrations of 2,3,7,8-TCDD, especially in the Moorfleeter Canal and the Auserer Vering Canal, were attributed to discharges from an organochlorine pesticide manufacturing plant.

The patterns of 2,3,7,8-TCDD and the other HpCDD congeners are characteristic of the patterns resulting from the production of 2,4,5-T and 2,4,6-trichlorophenol. In addition, the pattern of the HpCDF congeners can be linked to emissions from thermal processes employed by chemical industries in the production of chlorinated organic chemicals. The high concentrations of hepta- and octa-CDDs/CDFs may also be the result of other industrial combustion processes in the Hamburg area.