Tittabawassee River Floodplain Dioxin Contamination
EPA FACILITY ID: MID980994354
A Midland resident and two Michigan-based environmental organizations petitioned the Agency forToxic Substances and Disease Registry (ATSDR) to conduct a public health assessment of dioxinand dioxin-like compound (DLC) contamination in Midland, Michigan, and adjacent communities. The present public health consultation addresses floodplain soil contamination in the TittabawasseeRiver watershed downstream from the city of Midland. A draft of this health assessment wasreleased for public comment. The data necessary to determine if DLC-contaminated floodplain soilin the Tittabawassee River watershed poses a public health risk are not available; therefore, the siteposes an indeterminate public health hazard.
DLCs were detected at concentrations exceeding the ATSDR action level for residential soils (1 partper billion dioxin toxic equivalents) at several locations in the floodplain near the confluence of theTittabawassee and Saginaw Rivers. Additional soil sampling is necessary to determine if nearbyresidential properties are similarly contaminated with DLCs.
A soil sampling study conducted by the Michigan Department of Environmental Quality (MDEQ)indicates that floodplain soil in Tittabawassee River watershed near and upstream of the confluenceof the Tittabawassee and Saginaw Rivers is contaminated with DLCs. These data, along with DLClevels detected in fish, indicate that DLC contamination may be widespread throughout theTittabawassee River watershed below Midland. Additional soil sampling is necessary to determinethe extent and severity of DLC contamination in these areas.
The Michigan Department of Community Health (MDCH) recommends that the MDEQ implementsampling plans to determine if DLC contamination is present in the Tittabawassee River watershedbetween the city of Midland and the confluence of the Tittabawassee and Saginaw Rivers. MDCHspecifically recommends soil sampling at residential properties in closest proximity to previouslyidentified areas of concern. These plans should include contingencies for potential public healthactions if DLCs are detected at concentrations greater than the ATSDR action level, and acomprehensive evaluation of site-specific exposure factors if DLCs are found at concentrationsgreater than the ATSDR screening level. The MDCH will request ATSDR collaboration and supportfor health education and an Exposure Assessment for affected communities if indicated by the resultsof soil studies.
On May 1, 2001, a Midland resident and two Michigan-based environmental organizations petitionedthe federal Agency for Toxic Substances and Disease Registry (ATSDR) to conduct a public healthassessment of DLC contamination in communities adjacent to Midland, Michigan, wherecontamination originating from the Midland area has been "transported...via air and water"(Attachment A). The petitioners stated that the "likely source of this contamination is the DowChemical Company" (Petitioners for the Dow Midland site 2001). The petitioners specifically notedthat DLC contamination found in fish in the Tittabawassee River below Midland has been andcontinues to be chronically high.
ATSDR and the Michigan Department of Community Health (MDCH) have a cooperative agreementfor conducting assessments and consultations for potential health hazards at sites of environmentalcontamination within the state of Michigan. On July 30, 2001, MDCH staff and a representativefrom the ATSDR Region 5 office visited the Midland area and toured both the Tittabawassee Riverfloodplain and the Dow Chemical Company (Dow) plant site. The MDCH completed a "PetitionScoping Report" and provided the information obtained from these activities to ATSDR on August31, 2001. A copy of the report is provided in Attachment B. MDCH met with the petitioners onOctober 3, 2001, to discuss the health assessment process and to provide an opportunity for exchangeof additional information.
ATSDR responded in writing (Attachment C) to the petitioners on November 2, 2001, stating that,"After reviewing the public health issues and community concerns about potential DLCcontamination and the Dow Midland facility, ATSDR has found a reasonable basis to prepare publichealth consultations to address the concerns associated with the Dow facility"(ATSDR 2001). MDCH has agreed to prepare these public health consultations. ATSDR will review MDCH's workand provide technical support as needed.
The present consultation addresses concerns related to DLC contamination found in soil samplestaken from the floodplain of the Tittabawassee River below Midland. Additional consultations forMidland and affected adjacent communities that address other contaminated media may bedeveloped in the future with the ultimate goal of providing a full multimedia, multipathway publichealth assessment.
The Dow Chemical Company, founded in 1897, operates a chemical manufacturing plant in thecity of Midland, Michigan. The Dow plant encompasses approximately 1,900 acres on thesouthern perimeter of the city (Figure 1). The Tittabawassee River forms the southern boundaryof the plant site and flows east to the Saginaw Bay of Lake Huron.
Chemicals that have been produced at the Dow plant include: styrene, butadiene, picric acid,mustard gas, Saran Wrap, Styrofoam, Agent Orange, and various other chemicals includingchlorpyrifos (i.e., Dursban) and 2,4,5-trichlorophenoxyacetic acid (2,4,5-T). Chlorophenolproduction began in 1915. Wastes generated from this process were initially disposed of in 600acres of on-site waste ponds. During high flow periods in the early 1900s, wastes from these pondswould be intentionally released to the Tittabawassee River (Brandt 1997). Dow currently operatesits own on-site wastewater treatment plant .
In 1986, the State of Michigan experienced a significant flood event that incapacitated the Dowwastewater treatment plant and flooded areas of the plant where soils were contaminated with DLCs. Run-off from 10 feet of floodwaters overwhelmed containment systems, overflowed dikes, andentered the river (Wilkerson 1986). Untreated or partially treated chemical wastes entered theTittabawassee River before plant operations were shut down (Schmidt 1986). Chemical odorsassociated with the Dow plant were noticed 7 miles downstream from the Dow plant site.
In April 2000, soil samples were collected under the direction of the Michigan Department ofEnvironmental Quality (MDEQ) in an area near the confluence of the Tittabawassee and SaginawRivers (Figure 2) as part of a wetland mitigation project (MDEQ 2001a). This project was requiredby the United States Army Corps of Engineers to replace the shallow water habitat that would be lostthrough implementation of proposed fill and containment response activities at a site ofenvironmental contamination located downstream on the Saginaw River in Bay City.
Soil samples were collected by the responsible party to ensure that the selected mitigation site on the north side of the Tittabawassee River did not contain contaminant concentrations that would pose an unacceptable risk to terrestrial and aquatic organisms that would be expected to inhabit the new wetland resource. A sampling grid was established over the entire sample area, and surface soil was composited from nine locations located from within each grid area. At the request of the MDEQ, two of the composite samples were randomly selected for DLC analysis and concentrations of 1,474 and 2,199 parts per trillion (ppt) total dioxin toxicity equivalents (see box) were detected (Table 1).
|Soil Samples Location||Date Sampled||Number of Samples||Range of TEQ Detected (ppt)||DEQ Residential Cleanup Criterion (ppt)||ATSDR Screening Level (ppt)||ATSDR Action Level (ppt)|
|Mitigation Site Composite Samples||April 2000||9||1,474 - 2,199||90||50||1,000|
|Mitigation Site Discrete Samples||Dec. 2000||5||338 - 7,261||90||50||1,000|
De Rosa et al. 1997a ; MDEQ 2000, 2001a.
No immediate local source could be identified to explain the presence of the high concentrations ofDLCs at the mitigation site. The only known upstream source of significant concentrations of DLCsis the Dow plant in the city of Midland, approximately 20 miles upstream of the mitigation site.
To further define the extent and level of DLC contamination, the MDEQ developed a Phase I soilsampling program to achieve the following:
- Confirm the mitigation site sampling results.
- Identify whether other areas of the Tittabawassee River floodplain in the vicinity of the mitigation site contain DLCs above the MDEQ residential soil criterion of 90 ppt for DLCs.
- Generate sufficient information to make determinations regarding the necessity ofimplementing a Phase II soil investigation within the Tittabawassee River floodplain.
In December 2000, MDEQ staff collected and analyzed five discrete surface soil samplesfrom the mitigation site. Concentrations of DLCs ranged from 338 to 7,261 ppt total dioxinTEQs (Table 1). These data confirmed the previous composite sample results and alsoindicated that DLC contamination was present throughout mitigation site soils atconcentrations exceeding the MDEQ Residential Cleanup Criterion.
Based on the sampling results for the mitigation site, three upstream locations were selectedto assess whether other areas of the Tittabawassee River floodplain in the near vicinitycontained DLCs (Figure 2):
- A site of environmental contamination, located 1 mile upstream along the north bank of the river;
- A forested area within the Shiawassee National Wildlife Refuge (SNWR), located 1½ miles upstream along the south bank of the river; and,
- An open, upland area within the SNWR, located slightly less than 1½ miles upstream along the south bank of the river.
Four discrete surface soil samples were collected near the site of environmentalcontamination upstream of the mitigation site, two were collected upstream of the site froman adjacent farm field and two downstream from an adjacent golf course. All sample resultsidentified total dioxin TEQ concentrations above the MDEQ Residential Cleanup Criterionof 90 ppt (Table 2).
|Soil Samples Location||Date Sampled||Range of TEQ Detected (ppt)||Number of Samples||DEQ Residential Cleanup Criterion (ppt)||ATSDR Screening Level (ppt)||ATSDR Action Level (ppt)|
|Farm field||May 2001||180 - 424||2||90||50||1,000|
|Golf course||May 2001||2,529 - 2,588||2||90||50||1,000|
|SNWR Forested Area||May 2001||10||90||50||1,000|
|< 6 inches||35 - 134||90||50||1,000|
|< 12 inches||57 - 1,055||90||50||1,000|
|SNWR Upland Area||June 2001||15||90||50||1,000|
|0-3 inches||386 - 765||90||50||1,000|
|3-6 inches||424 - 663||90||50||1,000|
|12-15 inches||58 - 275||90||50||1,000|
De Rosa et al. 1997a ; MDEQ 2000, 2001a.
Five locations were sampled at two depths from within the forested area of the SNWR. Sample depths varied somewhat from one location to another due to root mass, leaf litter,and other location-specific factors. Five of the 10 samples contained total dioxin TEQconcentrations above the MDEQ Residential Cleanup Criterion (Table 2).
Five locations were sampled at three depths from within the SNWR Upland Area. Sampleswere collected from the 0-3 inch, 3-6 inch, and 12-15 inch soil strata. The five samplinglocations were selected from the higher ground elevations found within the open upland area. Thirteen of the 15 samples contained total dioxin TEQ concentrations above the MDEQResidential Cleanup Criterion (Table 2).
Based upon the soil sampling results described above, the MDEQ proposed a Phase IIsampling plan in October 2001 (MDEQ 2001b). Phase II soil samples were collected fromthree locations within the 12-mile stretch of the Tittabawassee River located between the cityof Freeland and Center Road in Saginaw Township (Figure 3). The selected samplinglocations were Freeland Festival Park, Imerman Park, and West Michigan Park. TheFreeland Festival Park is located approximately 7 miles downstream of the city of Midlandand is operated by the community of Tittabawassee Township. Imerman Park is located 4½miles downstream of the Freeland Festival Park. Imerman Park is a large, heavily utilizedpark that is operated by Saginaw County. West Michigan Park is operated by the communityof Saginaw Township and is located just over 6 miles downstream of Imerman Park andapproximately 4 miles upstream of the confluence of the Tittabawassee and Saginaw Rivers. The objectives of the Phase II sampling program were to:
- Identify whether DLCs have come to be located throughout the Tittabawassee River floodplain in Saginaw County.
- Identify whether DLC contamination in the Tittabawassee River floodplain soil located within Saginaw County is consistently above the MDEQ Residential Cleanup Criterion.
- Identify if DLC contamination found in floodplain soil varies between upstream and downstream locations.
- Observe the distribution of DLCs with respect to soil depth.
- Begin to make determinations regarding the source or sources for the observed DLCconcentrations.
- Make determinations regarding the need to implement a Phase III investigation and assessment program for residential properties within the Tittabawassee River floodplain.
The Phase II sampling plan has been implemented (Phase II sampling data will be discussed in afuture health consultation). More data are necessary to determine if floodplain soils throughout theTittabawassee River floodplain pose a public health risk. No additional information concerningDLC concentrations in the Tittabawassee River floodplain is available at this time.
The Tittabawassee River flows southeast from Midland to the confluence of the Saginaw andShiawassee Rivers near the mitigation site. From there, the Saginaw River flows northeast to theSaginaw Bay of Lake Huron. The MDCH has issued fish advisories for the Tittabawassee Riverbelow Midland based on detected levels of DLCs and polychlorinated biphenyls (PCBs). Theadvisories recommend no consumption of carp, channel catfish, or white bass, and limitedconsumption of smallmouth bass and all other species (one meal per week for the general populationand one meal per month for children and women of childbearing age). Similar advisories have beenissued for the Saginaw River. Advisories for the Saginaw Bay recommend no consumption of carpand channel catfish based on detected DLC levels. The advisory for Lake Huron, based on detectedDLC levels, recommends that women of childbearing age and children limit consumption of rainbowand steelhead trout to one meal per month, smaller whitefish (less than 18 inches in length) to onemeal per week, and recommends no consumption of larger whitefish (MDCH 2001).
To determine whether people are or could be exposed to contaminants associated with a property,ATSDR and MDCH evaluate the environmental and human components that lead to humanexposure. An exposure pathway contains five major elements: 1) a source of contamination, 2)contaminant transport through an environmental medium, 3) a point of exposure, 4) a route of humanexposure, and 5) a receptor population. An exposure pathway is considered a complete pathway ifthere is evidence that all five of these elements are, have been, or will be present at the property.
|Source||Environmental Transport and Media||Chemicals of Concern||Exposure Point||Exposure Route||Exposed Population||Time Frame||Status|
|The Dow Chemical Company, Midland||Deposition of river sediments in the Tittabawassee River Floodplain||Chlorinated Dioxins and Furans||Soil in residential yards, farm fields, and public access sites||Incidental ingestion, |
Dermal contact, Inhalation
|Residents in the Tittabawassee Floodplain, farmers, recreational visitors||Past||Complete|
The MDEQ Phase I sampling program has conclusively shown that elevated levels of total dioxinTEQ concentrations are present in the Tittabawassee River floodplain near the confluence with theSaginaw River. Human use of the floodplain increases upstream of the Phase I sampling area. Numerous residential properties are located within the floodplain. The majority located in the cityof Shields and in Saginaw Township, but are also scattered along the stretch of the river. Someagricultural operations are also located within the floodplain and as are several public parks.
The likely source of these contaminants is chemical manufacturing activities at the Dow plant sitein Midland. The likely mechanism of transport is deposition of contaminated river sediments in theTittabawassee River floodplain. People living within the floodplain, farmers, and recreation usersof the Tittabawassee River could be exposed to dioxins and furans in floodplain soil throughincidental ingestion, direct dermal contact, and inhalation of soil and dust. Chlorinated dioxins andfurans are very persistent in soil, therefore, exposure is likely to have occurred in the past and willcontinue in the absence of any remedial action.
Human use of the Tittabawassee River floodplain increases upstream of the Phase I sample area. Numerous residential properties are located within the floodplain, the majority being within thecommunities of Shields and Saginaw Township. Of particular concern are approximately 12 homeslocated adjacent to the river less than half a mile upstream from the mitigation site where total dioxinTEQs were detected at concentrations up to 7,261 ppt. Agricultural operations also occur within thefloodplain as well as a variety of public recreational facilities including parks, boat launches, andpublic access sites.
Dioxins and dioxin-like compounds are a group of over 210 chlorinated chemicals with similarstructures and chemical properties. This group of chemicals, which includes chlorinated dioxins,furans, and some polychlorinated biphenyls, is often referred to collectively as simply "dioxins" or"dioxin-like compounds (DLCs)." When found in the environment, DLCs are usually a mixture ofseveral of these chemicals. DLCs are not intentionally produced and have no known use. Not allDLCs have the same toxicity or ability to cause illness and adverse health effects. However, it isassumed that dioxins and dioxin-like compounds cause adverse health effects through a similarbiological mechanism of action. Further, the available science indicates that the health effectsresulting from exposure to multiple dioxin-like compounds are additive.
The most toxic chemical in the group is 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD). Toxicequivalency factors (TEF) have been developed to compare the relative toxicity of other dioxins anddioxin-like compounds to that of 2,3,7,8-TCDD. The levels of other dioxin-like compoundsmeasured in the environment are multiplied by a TEF to produce a 2,3,7,8-TCDD toxic equivalent or TEQ concentration. The resulting TEQs for all dioxin-like compounds measured in a sample arethen added together to determine the total dioxin TEQ concentration for that sample.
People who have been exposed to high levels of DLCs (such as those found in an industrialsetting or due to a significant industrial explosion) have developed chloracne, a skin disease withsevere acne-like pimples. Chloracne can persist for years, sometimes clearing only to recurseveral years later. Changes in blood and urine that may indicate liver damage have also beenseen in some people. Exposure to these high concentrations of DLCs may cause long-termalterations in glucose (blood sugar) metabolism and slight changes in hormone levels (ATSDR1998).
Exposure to lower levels of DLCs in laboratory animals has resulted in a wide variety of adversehealth effects, such as cancer, liver damage, and disruption of the endocrine system. In many speciesof animals, DLCs weaken the immune system and cause a decrease in the system's ability to fightinfection. In other animal studies, exposure to DLCs has caused reproductive damage and birthdefects. Some animal species, including monkeys, exposed to DLCs during pregnancy hadmiscarriages. The offspring of animals exposed to DLCs during pregnancy often had birth defectsincluding skeletal deformities, kidney defects, weakened immune responses, andneurodevelopmental effects (ATSDR 1998).
It is not known whether people exposed to low levels of DLCs will experience the same healtheffects seen in animal studies. However, based on the available information, DLCs are believed tohave the potential to cause a wide range of adverse effects in humans, including cancer. The U.S.Environmental Protection Agency (EPA) (EPA 2000) has characterized the mixture of DLCs towhich people are commonly exposed as "likely human carcinogens." The EPA has alsocharacterized 2,3,7,8-TCDD as a "human carcinogen" (EPA 2000). The U.S. Department of Healthand Human Services, National Toxicology Program 9th Report on Carcinogens (NTP 2001) lists2,3,7,8-TCDD as a substance "known to be a human carcinogen."
ATSDR Interim Guidance
Because of the potential for adverse health effects in human populations exposed to environmentallevels of DLCs, the ATSDR has developed interim policy guidelines to assist health assessors inidentifying soil concentrations of potential concern (Attachments D and E). The guidelinesrecommend the tiered approach shown in the table below to evaluate DLC concentrations in soil.
|Screening Level |
< 50 ppt TEQs
|Evaluation Level |
> 50 ppt but < 1,000 ppt TEQs
|Action Level |
> 1,000 ppt TEQs
|Health effects are unlikely and further evaluation is not necessary, unless there are community health concerns.||Evaluation of site-specific factors, such as ||Potential public health actions considered, such as |
(De Rosa et al. 1997a)
The screening level of 50 ppt total dioxin TEQ is the environmental media evaluation guide (EMEG)for 2,3,7,8-TCDD. The EMEG was developed from the ATSDR minimum risk level (MRL) basedon neurodevelopmental effects observed in the offspring of female rhesus monkeys exposed duringpregnancy (ATSDR 1998). EMEGs are very conservative and protective values. Generally, if soilconcentrations do not exceed the EMEG, ATSDR assumes that exposure is not likely to result inadverse health effects. However, if soil concentrations exceed the EMEG, this does not mean thatadverse human health effects will always occur. Instead, soil concentrations greater than 50 ppt totaldioxin TEQ indicate further site-specific evaluation is necessary (De Rosa et al. 1997a).
The action level of 1,000 ppt TEQ is a concentration of DLCs in residential soil at which variousactions may be considered to prevent or limit exposure. The action level is based on the analysis byKimbrough et al. (1984) of the carcinogenic potential of 2,3,7,8-TCDD. ATSDR recommends thatthe action level for residential soil be used in full consideration of site-specific factors that may affecttotal exposure to DLCs through all media and exposure pathways (De Rosa et al. 1997b).
ATSDR recommends that the action level be compared with the "maximum concentrations identifiedat the site" (De Rosa et al. 1997b). Residential soil levels were not available for this healthconsultation. However, total dioxin TEQ concentrations greater than 1,000 ppt have been detectedat several locations in the Tittabawassee River floodplain including: the mitigation site (up to 7,261ppt), an upstream golf course (up to 2,588 ppt), and the forested area of the SNWR (up to 1,055 ppt). MDCH has taken a conservative approach of using ATSDR's residential action level as acomparison for these non-residential areas.
Soil concentrations of dioxin TEQs that fall between the screening level and the action level warrantfurther site-specific study (De Rosa et al. 1997b). Many factors, such as those shown in the tableabove, can affect how much DLC people are exposed to, how much is absorbed into the body, andwhether or not adverse health effects will result. All of the soil samples collected from themitigation site, the upstream golf course and farm field, the upland area of the SNWR, and some ofthe samples from the SNWR forested area fall within 50 and 1,000 ppt.
Background Exposure to DLCs
An important consideration when evaluating DLC levels in soil is the level of exposure from allsources of DLCs, or the "background exposure." People can be exposed to DLCs from many sourcesother than contaminated soil. The general population is mainly exposed to DLCs through their dietby eating plants and animals that contain DLCs. People who live near or work at hazardous wastesites containing DLCs, waste incinerators, or manufacturing facilities that produce DLCs as a by-product may have additional DLC exposures beyond their diet. When people are exposed to andabsorb DLCs, they are stored in fatty body tissues where they may persist for months or years. Thehalf-life (the time needed for the body to rid itself of half the contaminants absorbed) for DLCs inhumans is 5 to 14 years. Because they remain for a long time, DLCs accumulate in the body and cancause health effects long after exposures have ended. The amount of DLCs accumulated over timeis referred to as the "body burden." The best available science suggests that body burden levels ofDLCs are closely associated with the likelihood of health effects. Therefore, many scientistsrecommend comparing DLC body burden levels in at-risk populations with those associated withhealth effects observed in animal and human studies (EPA 2000, De Rosa et al. 1997b).
Because people may be exposed to DLCs from a variety of sources, and because all theseexposures contribute to the body burden of DLCs accumulated over time, ATSDR recommendsevaluation of the contribution of soil exposures to total exposures from all sources (De Rosa etal. 1997b). Evaluation of soil exposures would require sufficient knowledge of DLCconcentrations in soil, bioavailability of DLCs in site soils, and human behaviors that can affectexposures. Additionally, evaluation of all other sources of DLC exposure would be required toevaluate the incremental contribution of soil exposures to the total body burden of DLCs in theat-risk population. This information is not currently available for residents, farmers, andrecreational users of the Tittabawassee River floodplain area.
This document was released for public comment in March 2002. The comment period lasted for 90days. A public meeting was held to solicit comments. The comments that were received during thecomment period are addressed in Attachment F.
Children may be at greater risk than adults from certain kinds of exposure to hazardous substancesat sites of environmental contamination. They engage in activities such as playing outdoors andhand-to-mouth behaviors that increase their exposure to hazardous substances. They are shorter thanadults, which means they breathe dust, soil, and vapors close to the ground. Their lower body weightand higher intake rate result in a greater dose of hazardous substance per unit of body weight. Thedeveloping body systems of children can sustain permanent damage if toxic exposures are highenough during critical growth stages. Prenatal exposures and those that occur in the first few yearsof life are more likely to cause permanent damage.
Fetuses, infants, and children may be especially sensitive to DLC exposure because of their rapidgrowth and development. In animal studies, exposure to DLCs has caused reproductive damage andbirth defects. Some animal species exposed to DLCs during pregnancy had miscarriages. Theoffspring of animals exposed to DLCs during pregnancy often had birth defects including skeletaldeformities, kidney defects, weakened immune responses, and neurodevelopmental effects (ATSDR1998).
Evaluation of all other sources of DLC exposure would be required to evaluate the incrementalcontribution of soil exposures to the total body burden of DLCs in the at-risk population. The datanecessary to determine if floodplain soil throughout the Tittabawassee River floodplain poses apublic health risk are not available; therefore, the site poses an indeterminate public health hazard. ATSDR classifies sites of environmental contamination into the indeterminate category when thedata to make a final decision are lacking.
The levels of total dioxin TEQs detected at concentrations exceeding the ATSDR residential actionlevel at the mitigation site (7,261 ppt), the golf course (2,588 ppt), and some areas of the SNWR(1,055 ppt) might pose a public health hazard if these levels are widespread and people are routinelyexposed to soil at these levels at residential locations. The level of exposure on these properties iscurrently not known. In addition, soil sampling is necessary to determine if nearby residentialproperties are contaminated with DLCs.
The results of the MDEQ Phase I sampling study indicate that floodplain soil samples collected fromthe confluence of the Tittabawassee and Saginaw Rivers at the mitigation site, the upstream golfcourse and farm field, and the forested and upland areas of the SNWR are contaminated with DLCs. These data, along with DLC levels detected in fish, indicate that DLC contamination may bewidespread throughout the Tittabawassee River watershed below Midland. Additional soil samplingis necessary to determine the extent and severity of DLC contamination.
- Implement the MDEQ Phase II soil-sampling plan to determine if DLC contamination is presentin floodplain soil in the Tittabawassee River watershed between the Phase I sampling locationsand the city of Midland.
- Design and implement a soil-sampling plan for the residential properties adjacent to the Tittabawassee River and in closest proximity to the mitigation site and other identified areas of concern. The plan should include contingencies for potential public health actions if total dioxin TEQs are detected on these properties at concentrations greater than the ATSDR action level, consistent with evaluation of site-specific exposure factors.
- Allow the sampling plans for the residential areas to be reviewed and commented upon by MDCH, ATSDR, and U.S. EPA prior to finalization and implementation. Feedback fromMDCH and ATSDR will be solicited to assess whether the sampling plan will be adequate tocollect the information necessary to better characterize the public health implications.
- Design a comprehensive evaluation of site-specific exposure factors for residents of theproperties in closest proximity to the mitigation site and any other areas of concern identified inthe Phase I and II studies, including biota sample analysis if feasible. Implement the exposureevaluation if total dioxin TEQs are detected at these properties at concentrations greater than theATSDR screening level.
- Design and implement a sampling plan to address residential properties throughout the Tittabawassee River watershed, contingent upon the results of the Phase II soil sampling study.
- The MDEQ has implemented the Phase II sampling plan for the Tittabawassee River watershed. Phase II sampling results will be discussed in a future health consultation. MDEQ is designing and implementing a sampling plan for at-risk residential properties, including contingencies for potential public health actions if necessary, consistent with evaluation of site-specific exposure factors. The MDEQ has begun implementation of this sampling.
- The MDEQ should implement additional soil studies as indicated by the results of Phase II sampling. The MDEQ is committed to additional samples if necessary.
- The MDCH will request ATSDR collaboration and support for health education and an exposure assessment for the affected communities if soil studies indicate an evaluation of site-specific exposure factors is necessary to determine the level of public health hazard posed by DLC contamination in the watersheds.
- The MDCH in cooperation with the appropriate local health departments will undertake health education activities to define and respond to the information needs of the affected communities.
- The MDCH in cooperation with the appropriate local health departments will be available to consult on the appropriateness and efficacy of future remedial actions.
If any citizen has additional information or health concerns regarding the Tittabawassee River Floodplain consultation or the Dow Chemical Company Midland petitioned health assessment, please contact the Michigan Department of Community Health, Division of Environmental and Occupational Epidemiology at 1-800-648-6942.
ATSDR (Agency for Toxic Substances and Disease Registry). 1998. Toxicological Profile for Chlorinated Dibenzo-p-Dioxins. December 1998.
ATSDR (Agency for Toxic Substances and Disease Registry). 2001. Letter to Petitioners for Dow Midland site dated November 2, 2001.
Brandt, E. N., 1997. Growth Company: Dow Chemical's First Century, Michigan State University Press, East Lansing, MI.
De Rosa, Christopher T. et al. 1997a. Dioxin and Dioxin-Like Compounds in Soil, Part 1: ATSDR Interim Policy Guideline. Toxicology and Industrial Health, Vol. 13, No. 6, 1997. pages 759-768.
De Rosa, Christopher T. et al. 1997b. Dioxin and Dioxin-Like Compounds in Soil, Part 2: Technical Support Document for ATSDR Interim Policy Guideline. Toxicology and Industrial Health, Vol. 13, No. 6, 1997. pages 769-804.
EPA (United States Environmental Protection Agency). 2000. Draft Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin (TCDD) and Related Compounds. September 2000.
MDCH (Michigan Department of Community Health). 2001. Michigan 2001 Fish Advisory.
MDEQ (Michigan Department of Environmental Quality). 2000. Part 201 Generic Cleanup Criteria Tables. June 7, 2000.
MDEQ (Michigan Department of Environmental Quality), Environmental Response Division. 2001a. Unpublished: Greenpoint - Tittabawassee River Dioxin Study Area, Phase I Sampling Study Report. October 2001.
MDEQ (Michigan Department of Environmental Quality), Environmental Response Division. 2001b. Unpublished: Greenpoint - Tittabawassee River Dioxin Study Area Phase II Sampling Program. October 2001.
NTP (National Toxicology Program). 2001. 9th Report on Carcinogens. U.S. Department of Health and Human Services. Public Health Service. January 2001.
Petitioners for Dow Midland site. 2001. Petition letter to ATSDR. May 1, 2001.
Schmidt, Wayne. Pollution rising to dangerous levels, experts say. Bay City Times. September 13, 1986.
U. S. Census Bureau (United States Census Bureau). 1990. Census of Population and Housing.
Wilkerson, Isabel. 1986. Tainting of Fish Feared After Michigan Floods. The New York Times. September 23, 1986.
Michigan Department of Community Health
Linda D. Dykema, Ph.D.
Toxicologist, Principal Investigator
Robin Freer, M.S.
Geographic Information System Specialist
ATSDR Regional Representative
Office of Regional Operations, Region V
ATSDR Technical Project Officer
Alan W. Yarbrough
Division of Health Assessment and Consultation
Superfund Site Assessment Branch
This Tittabawassee River Floodplain Health Consultation was prepared by the Michigan Department of Community Health under a cooperative agreement with the Agency for Toxic Substances and Disease Registry (ATSDR). It is in accordance with approved methodology and procedures existing at the time the health consultation was begun.
Alan W. Yarbrough
Technical Project Officer, CAT, SSAB, DHAC, ATSDR
The Division of Health Assessment and Consultation, ATSDR, has reviewed this public health consultation and concurs with the findings.
Sven E. Rodenbeck
for Roberta Erlwein
Chief, Cooperative Agreement Team, SSAB, DHAC, ATSDR
Click here to view Attachment A in PDF format (PDF, 2.7MB)
Click here to view Attachment B in PDF format (PDF, 2.7MB)
Click here to view Attachment C in PDF format (PDF, 2.7MB)
Click here to view Attachment D in PDF format (PDF, 2.7MB)
Click here to view Attachment E in PDF format (PDF, 2.7MB)
MDCH provided an initial 60-day comment period that was extended an additional 30 days at therequest of several commenters. The comment period closed on June 14, 2002. Comments receivedafter this date will not be addressed here.
Some commenters expressed concerns related to comments made by other individuals, companies,or organizations. These comments are not germane to the public health consultation and will not beaddressed further here.
Several commenters provided journal articles, research papers, and other reference materials, andthe MDCH thanks the contributors for this information.
1. Comment: We are in support of and encourage the MDCH and the MDEQ to move forward withthe recommendations contained in the petition, assess the health of the community with a properlydesigned survey, and collect soil samples in accordance with the Phase II sampling plan.
Response: The MDCH with the assistance of the ATSDR is developing a protocol for anExposure Investigation (EI) in the Tittabawassee River flood plain. When this protocol issufficiently developed, the draft will be shared with citizens, local government agencies, andother State of Michigan agencies. Recommendations for a "health study" - anepidemiological study of disease rates among people living in the flood plain or other similarassessment - will be considered after the results of the EI are known.
The MDEQ has completed Phase II soil sampling and has issued a report of the findings. These results will be discussed in a future health consultation.
2. Comment: The survey design for the health assessment should be forwarded to the SaginawCounty Department of Public Health (SCDPH) for review and comment before implementing thesurvey.
Response: See response #1 above.
3. Comment: The sampling strategy should be forwarded to the SCDPH for review and comment.
Response: The MDEQ is responsible for developing sampling strategies. SCDPH shouldmake this request to the MDEQ.
4. Comment: The identification of the study population should be forwarded to the SCDPH forreview and comment.
Response: See response #1 above.
5. Comment: Provide a timeline on the health assessment survey.
Response: See response #1 above.
6. Comment: Notify the SCDPH before any contact is made with the study population, media,public officials, or others and before an information relative to the survey results, whether apreliminary or final report is released to the petitioners, study population, media, public officials orothers.
Response: The SCDPH will be kept informed of the progress of any studies conducted inthe Tittabawassee River flood plain. However, the SCDPH may not always receive thisinformation before other stakeholders. For example, the property owners whose property hasbeen sampled are entitled to receive information without undue delay.
7. Comment: Collect "control" samples to identify potential sources of dioxins in the Flint,Saginaw, and Shiawassee Rivers. These samples may identify (or eliminate) pesticide application,manufacturing processes, or other activities that could have elevated the levels of dioxin along thefloodplain of the Tittabawassee River. Provide a timeline on the Phase II sampling.
Response: The MDEQ has conducted extensive sampling in their Phase II FloodplainSampling Program and these results have been disseminated to the public. A comprehensivereport was available in 2003. The MDCH will produce a revised public health consultationbased on these new data.
8. Comment: Notify the SCDPH before the collection of soil samples occur, before contact is madewith the media, public officials, or others regarding the collection of the soil samples, and before anyinformation relative to the sample results, whether preliminary or final report, is released to thepetitioners, study population, media, public officials or others.
Response: Requests regarding notification about soil sampling should be directed to theMDEQ as the agency responsible for sampling.
9. Comment: Why were the public meetings scheduled in Saginaw for May 1st and Midland forMay 2nd cancelled for unknown reasons?
Response: MDCH had some difficulty identifying an available and appropriate location tohold a public meeting in Saginaw. When a location was identified, MDCH postponed themeeting for a few days to allow the public adequate notification time.
10. Comment: How can we determine the number of Multiple Sclerosis (MS) cases reported in thearea of dioxin contamination?
Response: Diseases like MS are very difficult to study. MS is insidious in onset, difficultto diagnose, and chronic but not usually fatal. There are no registries of MS sufferers asthere are for cancer or birth defects: therefore, cases are not reported. While there are somecrude MS prevalence estimates at the regional and national levels, there are no existingregistries that would enable us to easily estimate the prevalence of cases among people livingin the Tittabawassee River flood plain.
11. Comment: Could the Thomas Township wells that supplied water to the township beforeSaginaw water became available have become contaminated?
Response: It is unlikely that these wells were contaminated. Dioxin binds tightly to soil andis almost insoluble in water unless there is an organic solvent present. The MDEQ hasconducted well water sampling at several locations with the flood plain and has not detecteddioxin concentrations above federal drinking water standards in any well. In most of the wellwater samples, no dioxins were detected.
12. Comment: Could dioxin contamination have had an effect on public health before the LaborDay flood in 1980s?
Response: It cannot be known with certainty what levels of total dioxin TEQs were presentin the flood plain soils in the 1980s or what health effects could have resulted from exposureto historical levels. However, it now appears that repetitive annual flooding, rather than asingle flood event, is responsible for depositing dioxin-contaminated sediments on theupland areas. The MDEQ has found elevated levels of dioxin 3 and 4 feet below the currentground surface and, in many cases, levels increase in deeper samples.
13. Comment: Can dioxin contribute to or cause multiple sclerosis?
Response: The MDCH is not aware of any literature or study suggesting a link betweendioxin exposure and multiple sclerosis.
14. Comment: In 1985 the EPA said the Tittabawassee fish had the highest national levels ofdioxins in fish. What are the current levels? What is the risk?
Response: In general, the levels of dioxin in fish taken from the Tittabawassee aredeclining. Further studies are ongoing and will be made public when the results areavailable. In the interim, follow the recommendations in the Michigan Fish Advisory forconsumption of sport-caught fish.
15. Comment: Given the known pollution of the Pine River (upstream from the Tittabawassee andDow), would it not be essential to sample upstream to fairly establish the effect of Dow's Midlandplant?
Response: The MDEQ has conducted this sampling, and a comprehensive report of thefindings was available in 2003. MDCH will discuss this data in a future health consultation.
16. Comment: Your own stated health action criteria were met at the outset with the results of thePhase I tests and have been ignored (4 of 5 samples exceeded the action level criteria of 1,000 ppt). How can this be explained?
Response: The State of Michigan criterion has not been ignored; however, furtherinvestigation is needed before the extent of contamination and the public hazard can bedetermined. MDCH remains committed to work with the MDEQ and other stakeholders toaddress the situation through appropriate public health actions.
17. Comment: Did DEQ issue directives or guidelines on the Phase 1 or Phase 2 testing in thismatter? If so, what were they?
Response: The MDEQ is the lead agency on Phase I and II sampling and directs all theseinvestigation efforts. MDCH provides technical support upon request.
18. Comment: Who was informed of the Phase 1 tests results and when?
Response: This question should be directed to the MDEQ. The MDCH became aware ofthe Phase I test results in December 2000.
19. Comment: When was Michigan's 90 ppt residential clean-up standard established and bywhom?
Response: The 90 ppt standard was calculated by the MDEQ in 1995 when all the cleanupcriteria were revised consistent with amendments to Part 201 of the Natural Resources andEnvironmental Protection Act, Act 451 of 1994, as amended. MDEQ standard exposureassumptions used for all chemicals, and the cancer slope factor developed by the U.S. EPAfor dioxin were used to derive the 90 ppt value.
20. Comment: Has that 90 ppt standard been changed, is it currently under review, and if so whatis the review process? On what basis can the 90 ppt be changed?
Response: The MDEQ has decided to maintain the 90 ppt criterion pending finalization ofthe U.S. EPA Dioxin Reassessment. When this document is final, the MDEQ willincorporate new toxicity information into the criterion.
21. Comment: Given the 20 or so soil samples taken in 2001 as part of the Taylor and McCabeBaseline Characterization of Saginaw Bay Watershed Sediments, shouldn't these be used tosupplement the Phase I test results?
Response: The results of the MDEQ Saginaw Bay Watershed Sediment Study will beincorporated into future health consultations
22. Comment: Do all mammals have essentially the same dioxin uptake receptors?
Response: All vertebrates (animals with backbones) including mammals have arylhydrocarbon (Ah) cellular receptors that bind dioxins. When vertebrate animals absorbdioxin into their bodies, the dioxin is believed to bind to the Ah receptor and set off a seriesof cellular events that could lead to adverse health effects. The effects that can result fromactivation of the Ah receptor vary significantly from one animal species to another. It isimportant to remember that human beings are animals too; and while humans may not be assensitive to dioxin as some other species, there is no evidence to suggest that humans areinsensitive to these effects.
23. Comment: What is the highest known level of dioxin found in a human (where, when)?
Response: One individual, a 30-year old woman believed to be intentionally poisoned withdioxin, showed concentrations of 2,3,7,8-TCDD of 144,000 picogram per gram of bloodlipids upon admission to a hospital at the University of Vienna. This person experiencedsevere chloracne that covered her entire body. Besides chloracne, the patient hadexperienced gastrointestinal symptoms including nausea, vomiting, epigastric pain, loss ofappetite, and had not menstruated since she had started a new job at a textile research facilityin 1997.
24. Comment: Does a baby "inherit" its mother's dioxin level? If so, would it be through the bloodsupply as well as through breast milk?
Response: Babies are exposed to dioxin before birth if the mother has elevated levels ofdioxin in her blood stream. Dioxin will accumulate in the baby's body just as it does in themother, although the rate of elimination could be higher due to the baby's more rapidmetabolism. Research in laboratory animals and in human populations suggests that theprenatal period (i.e., before birth) is the most sensitive time for dioxin exposure. Babies arealso exposed to dioxin if they are breast-fed and dioxin intake at this time can make asignificant concentration to a child's body burden of dioxin.
25. Comment: I want to register my support of the Michigan Department of Community Health'sefforts to further assess the health affects of dioxin contamination in …affected areas along theTittabawassee River.
Response: The comment is gratefully noted.
26. Comment: The data in the draft consultation from the confluence must be supplemented bydata from the full Saginaw River watershed and evaluated by considering all that data.
Response: MDCH agrees that data from the Saginaw River watershed must be acquired andevaluated to determine the full impact of dioxin contamination on both the TittabawasseeRiver flood plain and that of the Saginaw River. MDCH is aware of the other potentialsources of contamination in the watershed and supports MDEQ's efforts to fully characterizethe source of the dioxin contamination. These data will be evaluated in subsequentdocuments.
27. Comment: Data on health effects likely to be experienced by residents of Midland anddownriver areas are generally not reliably collected or not collected at all (with the exception ofgenital malformations). Most would be impossible to measure at the clinical level and are onlymeasurable at the population level. Yet they represent the most important risk to area residents.
Response: MDCH agrees with the comment and will consider ways to acquire the necessarydata to assess these health effects.
28. Comment: Levels of dioxin in indoor dust must be considered as part of the healthconsultation.
Response: MDCH agrees and is compiling information to conduct dust sampling.
29. Comment: Both the ATSDR and the MDCH have a role to play in accurate characterizationsof the public health issues raised by contamination of the Midland and downstream areas, as withany public health concern. We strongly urge ATSDR and MDCH to provide accurate publiccommunications to the community about dioxin and its hazards.
Response: MDCH agrees and, along with the MDEQ, is taking a proactive approach tocommunity involvement.
30. Comment: ATSDR 1 ppb standard is flawed. The level, derived more than 15 years ago, hasbeen referred to as a policy-based level, rather than a risk or health-based level. The number doesnot incorporate the newest synthesis of dioxin science in the dioxin reassessment.
Response: ATSDR does not set standards but does set media specific comparison values to assist in evaluation of hazardous waste sites and releases. These comparison values identify those contaminants for which further evaluation is performed. This consists of a more detailed examination of exposure pathways and available toxicological information. Based on the results of these evaluations, recommendations are made for publichealth actions and appropriate remedial actions, but again, these recommendations depend upon exposure-specific considerations.
ATSDR's Policy Guidance on Dioxin in Soil concludes that "the action level of 1 ppb (TEQ) for dioxin and dioxin-like compounds, when coupled to a site-specific context of evaluation for the range >50 ppt to < 1 ppb TEQs in residential soils, is protective of public health and continues to represent a level at which consideration of health action to interdict exposure, including cleanup shouldoccur." While ATSDR is aware of the scientific debate over the health-related impacts of lowlevel non-occupational exposures to dioxin and dioxin-related compounds, based on ourexperience at other sites, we find ATSDR's action level for residential soils to be protectiveof public health. These guidelines are applicable only to scenarios involving humanexposure via direct ingestion of dioxin-contaminated soils in residential areas and are notappropriate for other exposure scenarios.
31. Comment: Michigan residential and industrial soil cleanup levels compared with some EPARegions are not protective.
Response: The authority to develop Michigan clean up criteria rests with the MDEQ. MDCH will forward this comment to the appropriate MDEQ staff.
32. Comment: Midland and downriver soils are significantly elevated compared to state andnational averages.
Response: MDCH concurs with this comment.
33. Comment: No reviews of the epidemiological data have been conducted for the potentiallymost exposed population along the Tittabawassee and Saginaw Rivers.
Response: MDCH concurs with this comment and is considering methods to conduct sucha review. The public will be asked to participate in planning any such data review.
34. Comment: It has been suggested that dioxin congeners found downriver indicate dioxin-contaminated PCBs as the source.
Response: During Phase II sampling, the MDEQ initially analyzed soil samples for PCBs. It soon became apparent that PCBs were present in only very low concentrations andcontributed very little to the total dioxin-like toxicity of the samples. PCB analysis wasdiscontinued on later samples. However, if biomedical sampling of the human populationis conducted in the future, these samples will be analyzed for the co-planar PCBs todetermine the contribution of these compounds to the total dioxin-like toxicity.
35. Comment: Wildlife data must be considered in the health consultation, as often this dataprovides important information about extent of the contamination, historical trends, and potentialhealth effects in humans.
Response: The MDCH concurs with the comment. The MDEQ is conducting an ecologicalstudy of the flood plain, and this information will be incorporated into future consultations.
36. Comment: Area residents may have elevated rates of PCB's in their bodies because of majorPCB contamination in the Saginaw area.
Response: The MDCH concurs with the comment. Please see the response to Comment 34 above.
37. Comment: In order to be complete and accurate, all dioxin-like compounds must be consideredwhen assessing potential health impacts, not only those likely to be present in the general population,but also those specific to this region.
Response: The MDCH concurs with the comment. Please see the response to Comment 34 above.
38. Comment: Numerous reports have surfaced from surgeons, pediatricians, and other healthprofessionals about the anomalous health effects from Midland area residents. One report from asurgeon in Saginaw suggested more surgery for cleft palate (associated with dioxin exposure) in thearea than he had ever encountered in a practice that spanned a number of regions.
Response: The MDCH will be following up on these reports as part of the on going publichealth assessment process.
39. Comment: We urge biological sampling of wildlife and humans to determine if area residentshave elevated levels of dioxin.
Response: The MDCH in collaboration with the MDEQ is developing plans for biologicalsampling in both wildlife and human populations. This information will be presented infuture health consultations.
40. Comment: If biological sampling reveals elevated levels of dioxin, further public healthinterventions should be considered.
Response: Comment noted. The MDCH welcomes suggestions for specific health actionsin the flood plain.
41. Comment: Conduct soil, water, sediment, and indoor dust sampling (and water where relevant.
Response: The MDEQ is developing plans to further characterize the contamination in theTittabawassee and Saginaw River watersheds. The MDCH continues to participate in theseplanning efforts to ensure that the data acquired is useful for public health assessment.
42. Comment: Dramatically improve availability and monitoring of health data for humans andwildlife. A comprehensive, community approved new monitoring system should be set up to collectdata on relevant health effects for the Midland and downriver communities. A special form shouldbe designed for health practitioners specific to this region to collect data.
Response: The comment is noted. Such a monitoring system may be considered as MDCHand ATSDR progress through the public health assessment process.
43. Comment: A comprehensive, community-approved wildlife and farm animal monitoring planshould also be put in place.
Response: As stated above in Comment 35, the MDEQ is conducting an ecologicalassessment of the Tittabawassee River flood plain. The Michigan Department of Agricultureis responsible for monitoring farm animals and has conducted some limited testing of farmproducts in the area.
44. Comment: Develop proactive public education plan and materials.
45. Comment: I do not think enough is being done to let the public know about health risksassociated with dioxin.
Response: See the response to Comment 44 above
46. Comment: Is there not a cumulative effect of dioxin?
Response: Yes, dioxin accumulates in the body if exposure is on going. Dioxin is storedin fatty tissues and is only slowly eliminated from the body. After even a single highexposure, dioxin will persist in the body for many years.
47. Comment: The meeting held on May 13, 2002, at Swan Valley High School was a sham! Whata cover up!
Response: Comment noted.
48. Comment: I ran a day care in my home. How safe are these children and my children? I havealmost 2 acres and we are flooded many times a year. Also, in the flood of '86, my home wasflooded. How safe are we in our home?
Response: It is not possible to answer this question with the limited data currently available. This is why the MDCH and the ATSDR have recommended further study in theTittabawassee River flood plain and a comprehensive exposure investigation that includesall potentially contaminated media and routes of exposure. In order to help prevent possibleexposures, prudent public health practices should be followed such as washing one's handsbefore eating, removing shoes when coming indoors, and maintaining a grass coverthroughout the yard.