No abstract

Ball clay was the source of dioxin contamination discovered in selected chickens analyzed as part of a joint U.S. Department of Agriculture/U.S. Environmental Protection Agency national survey of the U.S. poultry supply conducted in 1997. The affected animals, which had been raised in the southern United States, represented approximately 5% of the national poultry production. All of these chickens and other animal food sources (i.e., farmraised catfish), similarly contaminated, were fed a diet of animal feed containing ball clay as an anti-caking additive. The clay was mined in northwestern Mississippi within a geological formation referred to as the Mississippi Embayment. Individual raw and processed ball clay samples were analyzed for the presence of the 2,3,7,8-PCDDs/ -PCDFs. The average toxic equivalents (TEQs) for the raw and processed samples were 1513 and 966 ppt dw, respectively. Other mined clay-based products used in animal feeds revealed lower TEQs. All of the products exhibited either an absence of detectable concentrations of 2,3,7,8-PCDFs or concentrations 2-3 orders of magnitude lower than the PCDDs. The isomer distribution, specific isomer identification, and congener profile of the PCDDs in the clay were established and compared to known sources of dioxin contamination. Several unique features of this isomer distribution are characteristic of the clays and are distinguishable from those of other known sources. These characteristics found in prehistoric clay deposits in the United States and Germany have led to speculation attributing their presence to natural geologic processes.

Long-term measurements of the atmospheric concentrations of PCDDs, PCDFs, and coplanar PCBs were taken in rural and remote areas of the United States by the National Dioxin Air Monitoring Network (NDAMN). A total of 21 quarterly sampling moments occurred from June, 1998 to December, 2002 at 34 locations geographically distributed throughout the United States. Sampling sites were located in rural and remote areas to obtain background air concentrations of dioxin-like compounds. Results were reported as the toxic equivalent (TEQ) of the mix of PCDDs/ PCDFs (TEQDF) and the mix of coplanar PCBs (TEQPCB). At the studied rural sites the mean annual TEQDF for each of the NDAMN sampling years was 10.43, 11.39, 10.40, and 10.47 fg m(-3) for 1999, 2000, 2001, and 2002, respectively. There was no statistically significant difference in the rural mean TEQDF air concentrations across the sampling years (at 0.05 level of significance), although the mean concentration in sampling year 2000 increased 10% relative to the other sampling years. The 95th percent confidence interval of TEQDF air concentrations in rural areas of the United States is from 6.4 to 15.4 fg m(-3), indicating there is a 95% probability that the true mean falls within this range. Mean annual atmospheric concentrations (TEQDF) at the remote sites were 1.41, 0.99, 0.7, and 1.07 fg m(-3) in 1999, 2000, 2001, and 2002, respectively, The 95th percent confidence interval of TEQDF air concentrations suggest that the true mean annual atmospheric TEQDF concentration in remote areas of the United States is between 0.1 and 3 fg m(-3). The remote sites have average air TEQDF concentrations that are approximately 10 times lower than those of the rural sites. For the rural sites, there was close agreement in the mean annual air concentrations of coplanar PCBs (TEQPCB) among the years 1999, 2000, 2001, and 2002 (i.e., 0.62, 0.69, 0.59, and 0.7 fg m(-3), respectively). However, as is the case with PCDDs/PCDFs, there was a marked increase (i.e., approximately equal to 13%) in the annual rural mean air concentration in 2000 as compared to the other sampling years. The confidence intervals across the NDAMN sampling years suggests a 95% probability that mean TEQPcB atmospheric concentrations in rural and remote areas of the United States are within the range of 0.5-0.9 fg m(-3) and 0.1-0.5 fg m(-3), respectively. The congener distributional patterns of PCDDs/PCDFs in air were relatively constant at all locations, and match the profile of urban air. We propose the hypothesis that urban areas are regional sources of PCDDs/PCDFs and are affecting atmospheric levels in rural and remote areas of the United States.