PUBLIC HEALTH ASSESSMENT
LOWER DUWAMISH WATERWAY
SEATTLE, KING COUNTY, WASHINGTON
Frequency of detection for contaminants with screening values in LDW sediment
Frequency detected : Frequency analyzed
|count||95th % (ppm)||Comparison Value (ppm)||Source||Contaminant of Concern?|
|4-Bromophenyl phenyl ether||1:734||0.120||NA||NA||No|
|4-Chlorophenyl phenyl ether||1:734||0.120||NA||NA||No|
|Aroclor-1242||136:984||0.319||0.220||Region 9||Total PCBs|
|Aroclor-1248||208:983||0.869||0.220||Region 9||Total PCBs|
|Aroclor-1260||745:986||2.400||0.220||Region 9||Total PCBs|
|Benzo(a)anthracene + Chrysene||735:813||1.100||0.137||MTCA||cPAHs|
|Butyl benzyl phthalate||437:818||0.360||10000||RMEG||No|
|Di–octyl phthalate||50:819||0.130||20000||int EMEG||No|
|Dioxin/furan TCDD toxicity equivalent||29:29||0.0001||0.00005||EMEG||Explain|
|Methyl ethyl ketone||18:61||0.024||48000||MTCA||No|
|Polychlorinated Biphenyl Dioxin-like Congeners||0.00009||0.000007||MTCA||Explain|
|Total Petroleum Hydrocarbons||65:73||4300|
|Total Polynuclear Aromatic Hydrocarbons||599:613||16||NA||NA||No|
|Total Polychlorinated Biphenyls||1194:1325||4.446||0.4||CREG||Yes|
|Total of 6 isomers: pp,op-DDT,-DDD,-DDE||83:213||0.080||NA||NA||No|
|TPH - Diesel||6:31||1415||200||MTCA||Explain|
|TPH - Gasoline Range||0:8||No|
|TPH - Heavy Fuel Oil Range||2:8||328||200||MTCA||Too Few|
Generally speaking, contaminants that exceeded screening criterion were considered to be of concern and were evaluated further. In some cases, contaminants exceeded screening criterion, but were not considered as contaminants of concern for other reasons. These explanations are listed below.
Iron was found in sediment at levels that exceeded the Region 9 PRG. This level, however, is wellwithin the background range of iron that occurs naturally in the Puget Sound region soils. The 95thpercentile of iron found in LDW sediments (39520 ppm) much lower than the 90th percentile ofbackground in the Puget Sound region (58700 ppm).
Thallium was found in sediment at levels that exceeded the Region IX PRG. The reference dose usedto calculate the PRG is based on thallium sulfate, and the critical effect for that chemical is alopecia(hair loss) in female rats. This endpoint is weak, and therefore, thallium in sediment is notconsidered to be of great concern.
Bis(2-ethylhexyl)pthalate (DEHP) was found in coho and Chinook at levels that exceeded acalculated comparison value. DEHP is a chemical for which there appears to be a threshold forcarcinogenicity. In other words, there is a dose of DEHP below which there is no cancer risk, butabove which results in some cancer risk. The evidence for this threshold comes from studies of ratsand mice dosed with DEHP. Liver cancer in these animals is thought to result from the process of peroxisome proliferation after exposure to DEHP. Without peroxisome proliferation, there were nosigns of carcinogenicity. Studies determined a NOEL for peroxisome proliferation at 20 mg/kg/dayin mice. Furthermore, rats and mice are considered to be especially sensitive to peroxisomeproliferation compared to humans and other primates. Based on this information, a margin ofexposure (MOE) of 10 was determined to be protective for potential risks to humans from DEHPexposure.(f) For comparison purposes, a dose calculation for a subsistence consumer of chinook is shown below because the highest level of DEHP in LDW tissue was found in a chinook sample.
DEHP dose =C x IR
Concentration [C] - 5.4 mg/kg
Fish Ingestion Rate (IR) - 0.00058 kg fish / kg body weight /day
DEHP dose = 5.4 mg/kg * 0.00058 kg fish / kg body weight /day
= 0.003 mg/kg/day
This dose can be used in conjunction with the observed NOAEL from the mice study to determine a marginof exposure (MOE) for this consumption scenario.
Margin of Exposure = NOAEL / Dose
= 20 mg/kg/day / 0.003 mg/kg/day
An MOE of 10 was determined to be protective of human health with regard to DEHP exposure, and anMOE of 6,700 was obtained using a reasonable conservative exposure scenario. In other words, the exposurescenario resulted in a MOE that was more than three orders of magnitude more protective than a MOE thatis considered to be protective. For this reason, DEHP was not considered a contaminant of concern.
Too few samples
Though there were data for more than 1000 sediment samples from the LDW, many chemicals wereanalyzed infrequently. Among them were some chemicals that may have been detected in a few samples, butin too few samples with which to conduct a worthwhile assessment over a wide area. The lack of completedata is a great source of uncertainty. The contaminants below met initial screening requirements, but werenot evaluated due to the low number of samples. In general, fewer than 50 samples over the entire waterwaywas considered to be a paucity.
Benzidine - detected in 2 of 16 sediment samples.
Total Petroleum Hydrocarbons (diesel range) - detected in 6 of 31 samples.
Total Petroleum Hydrocarbons (gasoline range) - detected in 2 of 8 samples.
Heavy Oil - detected in 5 of 13 samples.
Total TCDD - detected in 29 of 29 samples.
The draft public health assessment was released for comment on July 11, 2002. The public was given anopportunity to provide comments to DOH, and attempts were made to address all of them. Some commentswere addressed by simply amending the text within the document, while others comments are responded tobelow.
1. The contaminant screening process in the draft public health assessment is difficult to follow.
Efforts were made to make the screening process more transparent. Appendix F shows the contaminantscreening process.
2. There seems to be discrepancies with the conclusions of the Public Health Assessment and theRemedial Investigation conducted by the LDWG. For example, the RI concluded that arsenic, cPAHs,and PCBs were the largest contributors to risk in that order, and the PHA states that the maincontaminants of concern are PCBs and mercury. Please explain these discrepancies.
The health assessment and the Remedial Investigation each use similar methods in assessing risk or hazardsassociated with the LDW site. However, there are some different assumptions and approaches made in eachdocument because the purposes of the Remedial Investigation and public health assessment are different.The RI is designed to support site-specific decisions on the need for cleanup and remediation. The healthassessment is more qualitative and is designed to determine the relative hazard associated with the site andneed for any recommendations to reduce exposure.
Using cPAHs as an example, the remedial investigation presented relatively high cancer risks associatedwith consumption of LDW fish contaminated with cPAHs; however, there were not any finfish or crabs thathad detected levels of cPAHs in their tissues. The cancer risks presented in the RI were based onassumptions that the fish contained ½ of the limit of detection. While this is a sound approach fordetermining potential data gaps for the baseline risk assessment (i.e., necessary to get detection lowerdetection limits for cPAH levels in finfish), no reliable conclusion could be made regarding health hazardsfrom such a data set.
PCBs, on the other hand, were detected in all fish species. Hazard quotients associated with PCB exposureranked highest in all the fish exposure scenarios. This resulted in a fish advisory for the general populationbased on immune effects of PCBs and for pregnant women or those considering pregnancy based thecombined developmental effects of PCBs, mercury, and DDE.
3. Crab consumption should be included as a completed pathway of exposure based on reports fromWDFW that observed people harvesting crabs from the LDW. Furthermore, LDW crab consumptionwould clearly result in elevated health risks which is further compounded by the consumption habitsof LDW API consumers. The elevated risk needs to be recognized and clearly stated.
DOH recognizes that crabs are being caught from the LDW (though we do not know how many or howoften), and have considered the crab consumption pathway as a completed pathway of exposure in the finalversion of the PHA. DOH also recognizes that, based on sparse data, consumption of Dungeness crab mightresult in an elevated health risk, but only three individual Dungeness crab samples are used to calculate riskand hazard associated with the consumption of this species. Red rock crab samples are more numerous andalso contained PCB and mercury levels, therefore, crab consumption limits are included in therecommendation section of this document. We also recognize that API consumers might eat the entire crab.Data from Elliot bay and other studies indicate that the hepatopancreas in crabs tends to accumulatecontaminants. Therefore, the PHA recommends that this organ not be eaten.
4. The PHA reported risks for exposure from direct contact to sediment in the 10-6 range whichaccording to DOH's classification system would indicate a slight increase in cancer risk. Yet DOHconcludes that "exposure to sediments in the LDW represents no apparent public health hazard."
All sediment exposure scenarios resulted in doses that were well below noncancer reference doses whichindicates that exposure to LDW sediments is not expected to result in adverse noncancer health effects.Cancer risk attributable to direct contact with LDW sediments is lower than 1 x 10-5. Exposure tocarcinogenic chemicals at any level will result in some theoretical risk if it is assumed that there is nothreshold. Regulatory decisions are usually made when a risk of cancer exceeds a probability of 1 x 10-6 to1x10-4. The purpose of this health assessment is not to establish cleanup levels in sediment, but to informpeople of their potential risks.
5. The text states that DOH advises against harvesting shellfish from King County, "except forVashon/Maury Island." As DOH is aware, serious health concerns related to arsenic contaminationexist on Vashon/Maury Island. In light of this, DOH should update its advisory to include theVashon/Maury Island shoreline.
Vashon/Maury Island is outside the scope of this document. However, the DOH Office of Food Protectionand Shellfish Programs has not found higher levels of arsenic in shellfish from King County or VashonIsland compared to other parts of Puget Sound. In fact, the arsenic levels found have been very consistentthroughout Puget Sound regardless whether the shellfish tested came from pristine areas or urbanized areas(or areas downwind of Asarco). Advice from DOH against harvesting shellfish from King County'surbanized east shore beaches is based primarily on microbial contamination concerns.
6. WA DOH lists the EMEG Comparison Value of 20 ppm for arsenic in soil. Washington State is wellaware that this is not a protective value for arsenic. WA DOH should consult with the WashingtonState Department of Ecology (Ecology) and others regarding appropriate human health protectivelevels for arsenic. Ecology records indicate that 0.67 ppm has been determined to be protective.
Ecology has established 20 ppm as the cleanup level for arsenic in residential soil. The cleanup levels isbased on an upper-bound of background concentrations in Washington State. Ecology does acknowledgethat 0.67 ppm would be protective based on a 10-6 increased cancer risk, but that level is well belownaturally occurring levels in soil.
7. The PHA states that, "factors such as background exposure are considered when formulatingconclusions." Yet nowhere in the document are existing body burdens of chemicals presented,discussed or apparently considered in determining health effects. Please present information onexisting body burdens of chemicals such as lead, mercury, PCBs, arsenic and others, and explain howthese pre-existing body burdens are taken into account when determining the impacts of sedimentexposure and consumption of Duwamish River fish.
Generally, the PHA attempts to determine the risk for adverse health effects that would occur as a result ofexposure at a site. These risks often need to be put into perspective by comparing them to risks that wereceive as part of our daily lives. An example of this can be seen in fish contaminant levels. PCB levels arehigher in LDW English sole compared to nonurban areas of Puget Sound, but mercury levels are similar.This indicates that risk associated with PCBs is more of a site related problem (even though PCBs are foundin all fish), whereas risk associated with mercury reflects regional conditions. Cleanup of the DuwamishRiver will have a future impact on PCB levels in resident fish, but may not have a huge impact with regardto mercury levels.
8. Are there no existing data on toxicological mixtures for chemicals found in the Duwamish River, atrecorded levels? At a minimum, known synergistic effects for chemicals present in the river should bepresented, with a discussion of any uncertainties associated with reaching conclusions in specific fieldcircumstances.
ATSDR's Division of Toxicology recently prepared a draft interaction profile for persistent chemicals foundin fish. The weight-of-evidence analyses of available data on the joint toxic action of mixtures of thesecomponents indicate that scientific evidence for greater-than-additive or less-than-additive interactionsamong these components is limited and inadequate for characterizing the possible modes of joint action onmost of the pertinent toxicity targets. Therefore, it is recommended that additivity be assumed as a publichealth protective when assessing exposure to mixtures of these contaminants.
9. The PHA states that, "little difference exists between contaminant levels in salmon caught from theLDW versus other areas of Puget Sound." Yet Table 7 appears to contradict this statement, especiallyfor Aroclor-1254 levels in coho. Please present the results of a statistical analysis that would help toexplain this discrepancy.
A recent statistical analysis performed by the WDFW at the request of DOH compared PCB levels inDuwamish River coho with those from nonurban basins in Puget Sound. The analysis took intoconsideration several factors such as lipid content, whether the fish were hatchery reared or wild, gender,and size. While PCB levels in Duwamish coho were significantly higher than those from the Nooksack andSkagit rivers, they were not significantly different than those from the Nisqually or Deschutes rivers. Thisresult supports the notion that difference in the level of PCBs in the south Puget Sound is at least in partrelated to the amount of time coho spend feeding in Puget Sound.
10. The PHA states that in the case of average consumers, salmon consumption from the LDW is notexpected to pose a risk for any noncancer health effects and for high-end consumers states that thedoses are still well below actual toxic effects (due to safety factors applied). Given the potential risksindicated using approved methodologies, advising the public to disregard the potential risks indicatedappears irresponsible.
One of the difficulties in communicating health risks attributable to fish consumption is balancing the veryreal health benefits of eating fish versus theoretical or uncertain risks. Salmon are often regarded as being arelatively "clean" fish that have a high level of omega-3 fatty acids which reduce the risk of heart disease,and have other health benefits. For this reason, people are often advised to eat salmon and other lowcontaminant fish as opposed to fish that typically have high levels of contaminants. DOH agrees that there isa risk of eating an unlimited amount of salmon, or any other fish in the world, for that matter. DOHconsiders salmon the preferred fish to eat from the LDW due to the relatively low level of contaminants, andthe high level of omega-3 fatty acids. Consumption limits have not been set for LDW salmon due to thehealth benefits.
11. A discussion of cancer risk from consuming salmon should be presented, especially in light ofrecent studies on the Columbia River determining that salmonid consumption there poses anunacceptable cancer risk to tribal fishers. A comparison of fish tissue and sediment concentrationsbetween the Columbia and Duwamish Rivers should be provided.
EPA's Columbia River Basin Fish Contaminant Survey reported individual cancer risks of 1 x 10-3associated with a high-end consumers of coho, Chinook, Steelhead, Eulachon, and Pacific Lamprey. Themajority of these cancer risks were attributable to arsenic, and dioxin TEQ (includes dioxin-like PCBcongeners).
The cancer risk associated with a high-end consumer of all anadromous fish from the LDW is 9 x 10-4(Table C5). The majority of this risk is attributable to cPAHs and arsenic. It is important to reemphasize thatcPAHs were not detected in any finfish, and the amount of inorganic arsenic assumed to be in finfish isuncertain. A key difference between the data that were available for the Columbia River Basin FishContaminant Survey and the data used in the PHA was that there was no dioxin TEQ data available in theLDW fish.
Aroclor levels in the few Columbia River Basin salmon do appear to be lower than in LDW salmon,however, there may be explanations for these differences based on the geographic differences of the twowaterways. For instance, many of the fish in the Columbia were sampled from fresh water locations a greatdistance from salt water whereas LDW salmon were sampled in a marine environment or in brackish water.Returning salmon stop feeding as they swim up fresh water streams relying solely on their fat reserves. Inthe process of mobilizing their fat, they release PCBs into their blood stream where it is either repartitionedto remaining fat, other organs, or excreted. A comparison between these two different populations of coho,therefore, is not appropriate.
12. The PHA concludes that there is no apparent public health hazard associated with children'scontact with contaminated sediment at public access locations. The RI identified potential sedimentdata gap near public access points.
The PHA acknowledges a paucity of intertidal sediment samples near public access areas. However, it wasassumed that the 1,200 samples taken from the LDW were taken in areas thought to be contaminated, thusbeing biased toward finding areas with the most contamination. While this may not be true in all cases, itwas assumed that the sediment was adequately characterized with respect to direct contact pathway. DOHunderstands that more sediment sampling may occur at public access areas and will reevaluate the data oncethey are available.
13. Why does the PHA consider that all the seafood people consumes comes from the LDW?
In terms of assessing hazards associated with a site, DOH chose to evaluate a worst-case scenario. Theresults of the evaluation revealed that a subsistence level consumption of resident fish from the LDW couldresult in adverse health effects. Based on this, DOH has issued a fish advisory for resident fish in theDuwamish River (See Recommendations on page 58). Furthermore, consumers that eat a lot of fish as aroutine part of their diet should avoid eating resident fish from the LDW due to the fact that all fish havesome level of contamination.
14.What consideration was given to lead exposure at beaches?
Lead was not considered as a contaminant of concern in sediment because the 95th percentile leadconcentration of all sediment samples was below the comparison value (see table F5). However, 32sediment samples contained lead above comparison values, and nearly half of those samples were takenfrom a single area. None of the samples with elevated lead levels were located at or near public access areas.EPA and LDWG plan to further characterize sediment contaminant levels near public access points at whichtime DOH will be available to reevaluate lead and other contaminant exposure at public beaches.
15. EPA's revised guidelines for assessing cancer risks to children should be used to reassess cancerrisks for children in the Public Health Assessment for the Lower Duwamish River.
See the child health considerations section on page 48.
The Washington State Department of Health prepared this public health assessment under a cooperative agreement with the Agency for Toxic Substances and Disease Registry (ATSDR). The document is in accordance with aproved methodology and procedures existing at the time the health assessment was begun.
Technical Project Officer,
SPS, SSAB, DHAC
The Division of Health Assessment and Consultation, ATSDR has reviewed this public health assessment and concurs with the findings.
Lisa C. Hayes
for Roberta Erlwein
SPS, SSAB, DHAC
f Doull J, Cattley R, Elcombe C, Lake BG, Swenberg J, Wilkinson C Williams G, and van Gemert M. A cancer risk assessment of di(2-ethylhexyl)phthalate: Application of the new U.S. EPA Risk Assessment Guidelines. Regulatory Toxicology and Pharmacology 29: 327-357 (1999).