The Hanford Reach is an unimpounded 81 km (50 mi) stretch of the Columbia River upstream from McNary Dam in southeastern Washington that ranges from approximately river km 628 to 547 (fig. 1). The Hanford Site, a 1,517 km² (586 mi²) Federal facility situated alongside the Hanford Reach, was originally established as part of the Manhattan Project (Gosling, 2010). The Hanford Site is of interest because of the legacy from its past operations, which included liquid waste disposal that contributed to the contamination of local groundwater that ultimately reached the Columbia River. A plutonium production facility operated at the site from 1944 to 1988. As a result of these activities, the Hanford Site had several known and unknown waste disposal locations that included subsurface burial sites, liquid waste disposal sites, and contaminated facilities. Radioactive, nonradioactive, mixed, and hazardous waste was produced at the Hanford Site during its time of operation (PNNL, 2000), and several of the hazardous waste disposal locations at the facility contained bioaccumulating chemicals such as lead and mercury, pesticides such as DDT, and other chemicals (including PCBs and dioxin). These waste disposal practices contaminated upland soils and groundwater beneath the site (Hulstrom, 2011). Although operations at the Hanford Site ended in the late 1980s, contamination of the surrounding land and water continues to be a concern, likely due to groundwater movement from the contaminated areas to the Columbia River. Additionally, various other sources are located upstream from the site, such as agricultural runoff, mining operations, paper mill production, atmospheric deposition, runoff from urban land, and permitted wastewater discharges that are potential sources of bioaccumulating chemicals (EPA, 2009; Hulstrom, 2011).

The Hanford Site is divided into operational areas to facilitate security, safety, and functionality (PNNL, 2000); the location of the sampling sites used for this study are within the predetermined sub-areas currently used at the Hanford Site (PNNL, 2000; Hulstrom and Tiller, 2010). The sub-areas at the Hanford Site relevant to this study are shown in figure 1. The sturgeon sampled for this study were collected at five sites within sub-areas 100 and 300. Sub-areas 100 and 300 were chosen for their proximity to the Hanford Site. Approximately 32 km (20 mi) of the Columbia River flows alongside sub-area 100, and this area contained all nine plutonium production reactors, several waste disposal sites, barrel landfills, contaminated soil containing aldrin, dieldrin, DDT, paint waste, petroleum waste, and sodium dichromate (EPA, 1997; Capron, 2008). Sub-area 300 previously housed the fuel manufacturing operations and experimental laboratory facilities and is located alongside the Columbia River upstream from the City of Richland. Sub-area 300 had more than 70 waste burial grounds and unlined surface waste ponds and more than 130 contaminated facilities, all of which are contaminated with uranium, transuranic waste (plutonium), lead, toxic metals, benzene, PCBs, polycyclic aromatic hydrocarbons (PAHs), and various solvents (Peterson and others, 2008; Agency for Toxic Substances and Disease Registry, 2011).

The focus of this study was the bioaccumulation of four classes of persistent chemicals in white sturgeon collected from the Hanford Reach of the Columbia River near the Hanford Site. The four classes consisted of OC pesticides and pesticide degradation products, industrial or personal care products, PBDE congeners, and PCB congeners. Some of the waste from local contaminated groundwater produced by historical operations at the Hanford Site that reached the Columbia River was eventually deposited on the riverbed sediment (PNNL, 2005; Hulstrom, 2011). Although not included in this study, radioactive materials and chemicals were among the contaminants released from the Hanford Site over time, and their occurrence is described elsewhere (PNNL, 2005; Capron, 2008; Hulstrom, 2011). The study described in this report had three objectives: (1) to determine whether persistent chemicals of concern are bioaccumulating in white sturgeon residing in the Hanford Reach, (2) to identify where in the fish the chemicals are accumulating, and (3) to determine whether there are implications for human health related to the bioaccumulation of persistent chemicals of concern in white sturgeon.

The chemicals analyzed in this study consisted of 25 OC pesticides and degradation products (table 1), 6 industrial or personal care products, 10 PBDE congeners, and 15 PCB congeners (table 2). These 56 chemicals (lipid content was not included) were analyzed in brain, muscle, gonad, and liver tissues and, therefore, provided insight into the bioaccumulation that occurs in resident white sturgeon near the Hanford Site. The sampling results from this study were used to determine whether persistent chemicals of concern are bioaccumulating in white sturgeon and to assess the relative amount of bioaccumulation occurring in different tissues within those fish. The results also were used to assess implications for human health by comparing the measured concentrations to available health benchmarks. Because the samples were obtained from five sites and four types of fish tissue and organs, and included 56 chemicals per sample, the raw data contained more than 1,000 discrete measurements. All resulting data were previously published, including metadata for each fish—its sex and weight, as well as its location and date collected (Nilsen and others, 2016).

White sturgeon were collected from five locations in the Hanford Reach of the Columbia River during July 16–September 15, 2009 (Hulstrom and Tiller, 2010). Sturgeon tissues from that sampling effort were provided to the U.S. Geological Survey (USGS), and the USGS did not capture the fish. The fish were captured by use of long-line techniques, whereby circle hooks were attached at 4.6 m (15 ft) intervals along a 61–91 m (200–300 ft) rope let out along the river. One hundred sixty-six sturgeon were captured during the collection period, but only 30 were initially retained to comply with State regulations enacted to ensure that mature fish are not removed from the fishery. Fish that did not meet the legal requirements of fork lengths from 110 to 137 centimeters (cm) were released back to the river after recording their physical measurements. After capture and weight measurements were taken, the fish that complied with State regulations were retained and euthanized by severing the spinal cord, and the gills and organs were preserved for analysis. Twelve of these 30 fish had anatomical abnormalities, including duplicate fins, lesions, and discolored gonads. Of the 30 fish retained, tissues to be analyzed from 16 sturgeon were provided to the USGS, and the resulting data (Nilsen and others, 2016) were examined for this study. Sturgeon analyzed in this study had a mean length of 133 cm and a mean weight of 10.4 kilograms (kg). Samples of brain, gonad, and liver tissue and a subsample of the final fillet sample were individually collected from the sturgeon. The samples then were composited, by tissue type and among all the fish at each site location, resulting in one sample of each of the four tissues, from each of the five locations, with masses of 10–20 grams or more, as available. These composited tissue samples were wrapped in aluminum foil, frozen, and shipped to the USGS National Water Quality Laboratory (NWQL) in Lakewood, Colorado, for analysis.

Sturgeon samples were analyzed in 2011 at the NWQL. Laboratory procedures were based on techniques described in Zaugg and others (2007) and were modified for this study. These modified laboratory procedures were subsequently used and published in Nilsen and others (2014). After receipt at the NWQL, the frozen samples were thawed and homogenized using a blender with specialized glass, stainless steel, and Teflon^® internal parts. Extraction of the chemicals of interest from the homogenized samples was completed by using pressurized liquid extraction. The chemicals of interest were isolated from the extract after clean-up by capillary column gas chromatography. The results then were detected by negative ion mass spectrometry using selected ion monitoring. Concentrations of the chemicals of interest were reliably quantified at levels above the method quantitation limit. The qualitative identification of compounds at concentrations less than the method quantitation limit were reported as an estimate. The results for each of the samples were reported in micrograms per kilogram wet weight. All results from this study are available in the associated data release (Nilsen and others, 2016).

The results from this study showed that persistent chemicals of concern are bioaccumulating in white sturgeon residing in the Hanford Reach of the Columbia River and that the most prevalent of the chemicals measured were endocrine disruptors that can impact the reproductive health of the fish (Beckvar and Lotufo, 2011). The highest concentrations of these chemicals were measured within the gonad tissues of the white sturgeon sampled in this study. Highly lipophilic endocrine disrupting chemicals, such as OC pesticides, PBDEs, and PCBs, can transfer from contaminated tissues in adults to developing eggs and concentrate in lipid-rich gonads (Beckvar and Lotufo, 2011). Although the pathways that these chemicals take within white sturgeon are not fully understood, maternal transfer of pollutants might explain the high concentrations of persistent chemicals measured in the gonads of the fish sampled in this study and others. Additionally, white sturgeon is a long-lived and a late maturing species, which allows for a longer period of time for chemicals to accumulate and eventually transfer into developing eggs. Growth and reproductive challenges were observed in Columbia River white sturgeon studied by other researchers (Feist and others, 2005) and in white sturgeon sampled in the San Francisco Estuary (Gundersen and others, 2017) that had similar chemical exposures to those measured in this study. In both studies, elevated concentrations of DDE and PCBs were detected at high frequencies in the liver and gonads, with similar detections observed in this study. The high concentrations of DDE and PCBs were attributed in the previous studies as a likely cause for growth, reproductive, and feminization effects on the white sturgeon. Further assessments likely would help in understanding the toxic effects on resident fish like white sturgeon and would help differentiate between the effects of endocrine disruption and carcinogenesis from the bioaccumulation of persistent chemicals.

Although total concentrations are useful for assessing patterns in the results from this study, comparing the results to those from previous studies helps inform the overall understanding of the bioaccumulation of chemical compounds in white sturgeon. The maximum concentrations of DDE and total PCB congeners for the fillet tissues analyzed in this study were compared to the fillet results from two other studies that sampled white sturgeon from the Hanford Reach (EPA, 2002; Hulstrom, 2011). The results from this study also were compared to the results from studies on white sturgeon sampled in other reaches of the Columbia River (Foster and others, 2001; Feist and others, 2005; and Gundersen and others, 2008) and in the San Francisco Estuary (Gundersen and others, 2017) to get a more complete understanding of the relative contaminant burdens of these fish.

Although the total concentrations of DDE and PCB congeners in fillets measured in this study were lower than comparable results from the two previous studies of white sturgeon in the Hanford Reach (EPA, 2002; Hulstrom, 2011), the overall results generally were consistent with those from previous studies of white sturgeon collected from other reaches of the Columbia River and in the San Francisco Estuary (Foster and others, 2001; Feist and others, 2005; Gundersen and others, 2008, Gundersen and others, 2017). The maximum total DDE concentrations from the EPA (2002) and Hulstrom (2011) studies were 3.7 and 2.2 times greater, respectively, than the maximum total DDE concentration from this study; the mean total PCB concentration from the EPA (2002) study and the maximum total PCB concentration from the Hulstrom (2011) study were 2.4 and 8.2 times greater, respectively, than the maximum total concentration of PCB congeners from this study. Like this study, the data from white sturgeon in other reaches of the Columbia River and in the San Francisco Estuary showed that DDT degradation products were the most commonly detected chemicals in white sturgeon sampled and were present at higher concentrations than other types of chemicals detected within their respective studies. Persistent chemicals such as DDE, PCBs, and PBDEs also had high detection frequencies in the liver and gonad tissues of white sturgeon in the San Francisco Estuary, reinforcing previous findings that bioaccumulation of persistent chemicals in white sturgeon is an issue beyond the Columbia River Basin. However, distribution of some compounds within the tissues can vary among studies. For example, total DDT and PCB concentrations measured in the Columbia River study by Feist and others (2005) and the study in the San Francisco Estuary by Gundersen and others (2017) were higher in liver tissue than in gonad tissue, whereas, in this study, total concentrations of DDT degradation products and PCB congeners were much higher in gonad tissues. Although the lack of liver samples at sites 1 and 3 potentially biased the results, detections of DDT degradation products and PCB congeners were approximately 10 times and 3.5 times higher in gonad tissues than in liver tissues, respectively.

A select set of available results from this study were compared to the EPA human health screening values (SVs) (EPA, 2000) and Washington State Department of Health (WSDOH) screening levels (SLs) (WSDOH, 2012, 2017) to evaluate the potential human health risks from consuming white sturgeon from the Hanford Reach (table 7). The SVs and SLs are the acceptable concentrations of bioaccumulative chemicals in fish tissue relative to the human general/recreational population and subsistence/Tribal populations. The SVs and SLs for the latter group are lower (and therefore more protective) because people in that group are expected to consume fish at higher rates than the general/recreational population and generally use whole fish in their food preparations. The SVs and SLs are calculated by the carcinogenic effects, defined as health effects that manifests over an entire lifetime, and noncarcinogenic effects, defined as acute health effects over a relatively short period of time (EPA, 2000; WSDOH, 2012, 2017). The EPA SVs and WSDOH SLs were available for eight of the chemicals or chemical combinations that were analyzed in this study (tables 8 and 9). Of the industrial or personal care products analyzed for this study, only one product in that chemical class (triclosan) was detected, with no available SV or SL for that chemical; therefore, no industrial or personal care product was used for any comparison. The risk to the general/recreational population was determined by comparing the fillet tissue results from this study to the applicable SVs and SLs. In contrast, the risk to subsistence/Tribal populations was determined by comparing the mean results for all tissues to the applicable SVs and SLs, based on the assumption that a whole-fish value likely would be within the range between the minimum and maximum tissue concentrations, and close to the mean value. The minimum and maximum concentrations for each chemical and chemical combination also are presented in tables 8 and 9.

A comparison of the results from this study to the EPA’s SVs and the WSDOH SLs indicated that the general/recreation population and subsistence/Tribal populations have increased health risks associated with the consumption of white sturgeon from the Hanford Reach. In general, EPA carcinogen SVs (fig. 2) had more chemical detections above their respective chemical threshold values for the general/recreation and subsistence/Tribal consumers than for non-carcinogen values. No non-carcinogen values exceeded the WSDOH SL for three of the six chemicals (dieldrin, hexachlorobenzene, and total chlordane) for both the general/recreational and subsistence/Tribal consumers (fig. 3). However, SLs for total DDT, total PBDE, and total PCB had detections above the threshold value for both general/recreational and subsistence/Tribal consumers, with all total PBDE detections for subsistence/Tribal consumers above the SL value. For the general/recreation population, the total concentration in fish fillet exceeded the carcinogenic SV at four of five sites for DDT degradation products and at one site for PCB congeners. None of the non-carcinogen fillet SVs for the general/recreation population were exceeded, however, site 4 exceeded one SL for PCB congeners. The results from this study also indicated that the health risks are greater for subsistence/Tribal populations than for the general/recreation population. The mean concentration in all tissue types exceeded the subsistence/Tribal, noncarcinogenic SV and SL at all five sites for DDT degradation products and exceeded the subsistence/Tribal, noncarcinogenic SV and SL at all five sites for PCB congeners. The mean concentration in all tissue types also exceeded SLs at all five sites for PBDE congeners. Additionally, the mean concentration in all tissue types exceeded the subsistence/Tribal, carcinogenic SVs at all five sites for DDT degradation products, dieldrin, and PCB congeners. The mean concentration in all tissue types also exceeded the subsistence/Tribal carcinogenic SVs at three and four sites for total chlordane and hexachlorobenzene, respectively.