Untreated drinking water provided by Delavan Well No. 4 is contaminated with VOCs. This contamination apparently originates from the nearby Sta-Rite Industries. Sub-surface soils and groundwater under the Sta-Rite property contained higher levels of contamination than has been found in Well No. 4. Clean-up actions on the Sta-Rite property removed the sources of this contamination, and ongoing groundwater and soil treatment systems further minimize the movement of contaminated groundwater away from the property.
Available information indicates that VOC-contaminated groundwater may be wide spread in the Delavan area, but the full extent of this contamination has not been determined. Several investigations have not discovered any Delavan homes with private wells that draw VOC-contaminated groundwater, but sampling of private wells at three Delavan businesses found VOC contamination. Additionally, unrelated DNR investigations have found VOC contamination at five other sites located in eastern Delavan. The sources of this contamination have not been identified. One natural spring located near Comus Lake was found to be contaminated with TCE and the City of Delavan closed access to this spring. The source of this contamination is not known.
There are three bodies of surface water in the vicinity of Delavan Well No. 4 that were possibly affected by contamination coming from the Sta-Rite property. Swan Creek sediments and surface water were found to contain low levels of VOCs that are not a health concern. No contamination was found in surface water samples from Comus Lake. There has been no sampling of sediments or surface water from a five-acre pond southeast of the Sta-Rite property.
This following section describes how "Chemicals of Concern" are distributed in soil, sediment, water, and biota in and near a Superfund site. "Chemicals of Concern" are those that occur above a level where the maximum plausible exposure to the contaminated material might affect human health. This public health assessment addresses only those contaminants that the authors judge to be present at levels of concern. Levels of concern are listed as "comparison values" in relevant tables incorporated into this public health assessment. These values frequently differ from regulatory standards or health advisory levels. Subsequent sections of this public health assessment discuss whether chemicals of concern pose a potentially significant threat to public health.
A. Source of Contamination in Well No. 4
The Sta-Rite property on Wright street is identified as the apparent source of VOC contamination found in Well No. 4. Investigations at the Sta-Rite property indicate that contamination originates at Plant 2 [55]. A liquid sample taken from a sump at Plant 2 in December 1982 showed levels of TCE at 77 µg/L and PCE at 28 µg/L. Samples collected from solvent catch basins at Plant 1 in 1982 showed very high levels of VOCs. The sludge from one catch basin had TCE at 1,200,000 µg/L and PCE at 60,000 µg/L [61, TABLE 2-2]. A January 1984 report concluded that VOC concentrations in soils near the Plant 2 sump were high enough to impact nearby groundwater [42].
Sub-surface Soil at the Sta-Rite Property
Soils underneath a sump at Plant 2 were shown in December 1983 to be contaminated with TCE at 980,000 µg/kg and PCE at 280,000 µg/kg. The sump and contaminated soils at this location were excavated and removed in 1983 [61, P2.6]. Soil boring samples were collected in 1991 and 1992 from 16 locations at Plant 2, with 12 locations showing contamination. Soil samples from one location near the former sump (SB-2008) had the highest levels of PCE (greater than 50,000 µg/kg) and TCE (greater than 8,200 µg/kg).
A total of 15 different VOCs were detected in these 12 soil samples [61, TABLE 4-4]. At Plant 1 the highest TCE concentrations, found in soil borings in 1991 and 1992, were at the drainage field for former sump #15 (281 µg/kg) and active sump #8 (240 µg/kg). A number of Plant 1 sumps have been identified as known or suspected points of VOC disposal [61, TABLE 4-11]. The contamination found in soil boring samples is summarized in Table 3.
Groundwater at the Sta-Rite Property
B. Groundwater Contamination in Delavan Municipal Wells
Recent analysis of untreated water samples from Wells No. 3 and No. 4 show TCE continues to be present at low levels. The possible sources of contamination found in Well No. 3 has not been determined. See Table 1 for selected well testing of Delavan municipal wells from 1982 to 1985. The background section and Table 2 summarizes the history of VOC contamination found in Delavan municipal Wells No. 3 and No. 4.
C. Other Delavan Groundwater Contamination
Information suggests groundwater contamination in Delavan could be more widely spread than previously suspected. Groundwater sampling from wells at Delavan locations, other than the Sta-Rite property, has found contamination by some of the same chemicals found in Wells No. 4 and No. 3. The source of this contamination is not clear. While no homes in the Delavan area are known to use contaminated water from private wells, three Delavan businesses have wells with low levels of VOCs. Further sampling activities are needed before an accurate picture can be drawn that depicts the full extent of groundwater contamination in Delavan.
Contaminated Groundwater Migrating from the Sta-Rite Property
Groundwater monitoring data collected during the Sta-Rite Remedial Investigation shows contamination up to the furthest monitoring points located west-northwest of the Sta-Rite property [63, TABLE 4-5]. One monitoring well cluster (D-5 & D-6) [44, TABLE 6], located just south of the railroad right-of-way and across Wright street from at the Sta-Rite property, has shown VOC contamination (Table 6). In 1983 and 1984, samples from both wells in the cluster revealed elevated levels of VOCs (Wells D-5 and D-6 are cased at 30-50 and 100-110 feet respectively). Groundwater samples collected in 1991 found some of these VOCs, but only in the shallower D-5 and at lower levels than in 1983 and 1984. From 1991 to 1994, TCE was consistently found in D-5 at levels between 6 and 9 µµg/L [45 TABLE 2].
Contamination was also found at another monitoring well and soil boring location (MW-1030 & SB-1030), which is approximately 1,100 feet northwest of the Sta-Rite property. TCE was detected twice (4 µg/L) in groundwater samples collected from this monitoring well. Sources other than Sta-Rite were suggested for some of this contamination [64, P5.16], yet these VOCs were also in groundwater at the Sta-Rite property, and plume modeling indicated the sampling point is probably in the path of contaminants migrating from the property [41]. Between 1991 and 1994, TCE was consistently detected in MW-1030, ranging between 3 and 57 µg/L, with the higher levels detected in during the last two years. In 1994 PCE was also measured in MW-1030 at 1.4 µg/L [45, TABLE 2]. In 1995, a water sample collected from a naturally flowing spring, located 3,500 feet west-northwest of the Sta-Rite property, was found to contain TCE at 6.5 µg/L [42].
While modeling of Sta-Rite plume migration concluded that contaminated groundwater flows into Comus Lake, no investigations have examined groundwater movement in the vicinity of Comus Lake to substantiate this model. Comus Lake is a reservoir on Turtle Creek, and Swan Creek enters Turtle Creek immediately below the Delavan dam. Groundwater movement south of Comus Lake could be to the west or southwest, which could direct contaminant plumes much further to the west than suggested by previous investigations. There may also be seasonal variations in groundwater flow near and around Comus Lake, which also could affect the location and movement of contaminant plumes.
Other Delavan-Area Groundwater Investigations
Delavan Environmental Repair Project (ERP) Sites
The Jacobus Oil ERP site is located on Sunshine street, approximately 800 feet north of the Sta-Rite property. TCE was found in three shallow groundwater monitoring wells, with the highest reported level at 8.1 µg/L [36].
Delavan Leaking Underground Storage Tank (LUST) Sites
The Holiday/Super America service station, 803 East Geneva Street, is located approximately 2,500 feet west-southwest of Well No. 4 and approximately 1,250 southwest of Well No. 3. Groundwater samples were collected twice from seven locations and analyzed for VOCs. VOCs were detected in samples from three upgradient locations. PCE was detected both times in samples from each location, with the lowest and highest levels at 3 and 18 µg/L, respectively. TCE was found only once and at only one location, at 1 µg/L. The 1991 remedial investigation report states that VOC contamination of groundwater under this LUST site may be from another source [68]. Another assessment reported that PCE and TCE contaminated groundwater at this site is "not derived from on-site sources" [62]
Campbell's Mobil, at 746 East Geneva Street, is across the street from the Holiday/Super America service station. Three monitoring wells were installed at this LUST site to evaluate groundwater contamination from gasoline. In addition to finding an assortment of gasoline-related products in groundwater, PCE was detected at 3 and 5 µg/L in the two upgradient monitoring wells. The Tank Removal Assessment report did not evaluate the possible sources of this VOC contamination [50].
The Delavan Mobil/Del-Mart Phillips 66 service station is located at 338 East Walworth Avenue, approximately 5,200 feet west-northwest of Well No. 4 and 3,500 feet west-northwest of Well No. 3. Investigations at this LUST site found petroleum products in groundwater and soils. Two rounds of groundwater samples were collected from each of three groundwater monitoring wells. PCE was detected at least once in each groundwater monitoring well (but was not found in a sample from the soil vapor extraction well). PCE was found in six of eight groundwater samples, with the highest level measured at 2.2 µg/L. A report written by the consulting hydrogeologist said "the uniformity of the levels [of PCE] suggest that it has been caused by an upgradient source" [8].
Private Drinking Water Investigations
Outside Delavan City Limits
The DNR reported to DOH, in 1994, that three private wells located in the Mound Road Industrial Park (northeast of the City of Delavan and within the Town of Delavan) were contaminated with VOCs below levels of health concern. This industrial area is approximately 1,000 feet north of the Sta-Rite property. Two of these wells were contaminated with TCE at 2.6 and 4.5 µg/L. The levels of VOC contamination measured in these wells are a not health concern [43].
In June 1995, the DNR placed a public notice in two local newspapers in order to identify any previously unknown private wells, which then could be tested. Only one homeowner in the immediate Delavan area requested to have a well sampled. In July 1995, DOH collected a water sample from this private well, which is located on State Highway 11 and approximately 2,500 feet northeast of the Sta-Rite property. No detectable VOC contamination was found in this well [23].
Within Delavan City Limits
In a separate effort, Sta-Rite had a notice stuffed into a free newspaper that was distributed to each Delavan household [60, P3.3]. The Sta-Rite survey targeted an area covering a corridor from the Sta-Rite property northwest to Comus Lake [60, FIGURE 3-3], and no new wells were identified in the targeted survey area [60, FIGURE 4-2]. Water samples were collected at least three times from one well (PW-1) and no contamination was detected [58, APPENDIX I] [43, P3.3].
Water from this well was reported to be exclusively used for a swimming pool. It was also reported that water from a second well located in the area targeted by Sta-Rite was not used for potable purposes and the owner did not allow a water sample to be collected [60, P4.3]. An additional private well was previously identified in 1984 in an area west of the targeted area [43, P3.3], but there are no reports that a water sample was collected from this well, nor was the location of the well identified.
In April 1995, DOH collected a water sample from a naturally flowing spring located in Spring Grove Cemetery, within the City of Delavan and approximately 3,500 feet west-northwest of the Sta-Rite property. TCE was detected at 6.3 µg/L. The City of Delavan and the cemetery caretaker closed the spring during June 1995 [22].
D. Surface Water Contamination
There are three bodies of surface water in the vicinity of Delavan Well No. 4 that might previously or currently be affected by contamination coming from the Sta-Rite property. These are: the Swan Creek tributary below the storm sewer effluent discharge point; a five-acre pond directly southeast of Sta-Rite (Figure 4); and Comus Lake. Swan Creek sediments and surface water were found to contain low levels of VOCs. No contamination was found in surface water samples from Comus Lake. There has been no sampling of sediments or surface water from the five-acre pond located near the Sta-Rite property.
It is feasible that contaminants could have been carried in waste water and deposited onto this field. However, any VOCs present in this surface water likely would have evaporated quickly. The concentrations of some VOCs in extracted water were also found to be elevated, with the highest levels found in 1982 (TCE at 5,000 µg/L). Table 5 summarizes the levels of VOC contamination in water pumped from each of the seven extraction wells.
It is not known if other contaminants were present in the waste water received by these areas, but sampling in December 1982 suggests that this waste water contained other, non-VOC constituents. The earliest available analytical results of Sta-Rite storm sewer water discharge were reported in December 1982. While results showed the levels of heavy metals at "typical" concentrations, the laboratory reported the "strength of...sanitary wastewater is considerable with elevated concentrations of...total phosphorous and Oil and Grease," which were entering the sewer line [45].
Surface Water from Storm Sewer Effluent
A permit for the temporary discharge of contaminated groundwater from Sta-Rite and into the Delavan municipal storm sewer was approved by the Wisconsin DNR in 1989 for PCE and TCE at 865 µg/L, and 3,840 µg/L, respectively [31]. There are no reports that levels of PCE and TCE in grab samples exceeded these levels. This permit was renewed by the DNR effective October 1990, until July 1995. This new permit did not set discharge limits for these VOCs [31].
Sediments from Storm Sewer Effluent
Ambient Air Releases from Storm Sewer Effluent
Calculations in the Sta-Rite Remedial Investigation Report estimated that 75 percent of extracted water VOC contaminants would volatilize between the point of entry into the storm sewer system and the outlet to the Swan Creek tributary, at SS-2 [63, P4.22], as supported by the sampling data presented in Table 7. However, the Sta-Rite Remedial Investigation Report did not fully evaluate the fate of the compounds volatilized within the storm sewer. The Remedial Investigation implies that volatilized VOCs are mixed with air in the sewer to such a degree that by the time it reaches SS-2 the concentration is too low to detect. Unfortunately no air samples were collected at or within access covers along the path of the sewer to verify these calculations and subsequent conclusions.
Surface Water Run-Off from Sta-Rite Property
It is difficult to evaluate how storm water moved around and off the Sta-Rite property. Portions of the property were regraded in the past when Sta-Rite connected to the Delavan storm sewer in 1982. Contaminants present on the ground would have been carried by runoff into drainage ditches on the Sta-Rite property, including the drainage ways servicing the older storm sewer system. An earlier document described "storm runoff from a portion of the facility near Plant 2, not currently served by a storm sewer drain, discharges through a drainage ditch southeast of Plant 2 into a five-acre pond southeast of the Sta-Rite property [59, P4.24]." Given the volatile nature of VOCs, this contamination probably no longer exists on the surface of the existing drainage ways.
In 1991 and 1992 a number of soil borings were taken east of Plants 1 and 2, and in a former drainage way. Tests showed relatively low VOC concentrations. It was concluded in one report that this contamination was not expected to impact groundwater quality [61, P4.20]. However at one location (SB-1025) over 21 different contaminants were detected, more than any other soil boring sample. Runoff from the adjacent parking lot and driveway could account for some contamination found in certain samples.
An artificial pond, located less than 500 feet southeast of the Sta-Rite property, may have received contaminated run-off in the past. A review of aerial photographs indicate this catchment basin was created between in 1972 and 1975, during the construction of the nearby Highway 43 [46]. A drainage ditch, which empties into this pond, is located immediately east of Plants 1 and 2. Contaminated storm water runoff may have flowed east, into this ditch, and ultimately entered the pond. Contaminated soils are present in some of the areas served by the drainage ditch. A soil sample collected from the 13-15 foot depth of a boring on the northern edge of the ditch (SB-1031) showed the presence of a number of VOCs, including TCE. There are no reports of water or sediment samples collected from this pond. If VOCs reached this pond, detectable levels would probably not have remained in the surface water for more than 30 days. The lack of surface water sampling from this pond represents a data gap.
E. Toxic Chemical Release Inventory
A Toxic Chemical Release Inventory (TRI) search was conducted by the Division of Health for the City of Delavan zip code (53115). The TRI is used to investigate any other sources of the same type of environmental contamination as found at the Superfund site. Certain manufacturers are required to report to the U.S. EPA of releases to the environment of over 300 hazardous chemicals. This reported information is entered into the automated TRI system. The TRI was searched for reports of releases for the years 1989, 1990, and 1991. The search reported releases of one VOC by Sta-Rite Industries, which was not identified as a contaminant of concern. Concentrations of this VOC in nearby air was estimated using a plume dispersion model with the highest reported annual release level. The model indicated that this VOC released from Sta-Rite does not pose an apparent health concern to nearby residents.
F. Quality Assurance and Quality Control
The Wisconsin Division of Health assumes the DNR, the City of Delavan, Sta-Rite Industries, and all contractors fully met standard sampling protocol, unless stated otherwise, including those cited as appendices in referenced reports. These quality assurance and quality control measures were to be followed during the field sampling and measurements, the chain of custody activities, laboratory analytical procedures, and data reporting. The ability of the Division of Health to make valid conclusions depends on the amount and quality of data provided.
There are no apparent physical hazards associated with Delavan Well No. 4.
City of Delavan residents were evidently exposed in the past to contaminated water originating from Delavan Well No. 4. The Delavan municipal water supply system is an indeterminate past public health hazard because a lack of information does not allow confident estimates of past human exposure. The presence of VOC-contaminated groundwater at several Delavan locations suggests that contamination may be widespread and area residents who obtain their drinking water from private wells may be exposed. The private wells at three Delavan businesses are contaminated with low levels of VOCs that are not a health concern. Several investigations have not discovered any homes in the Delavan area that use private wells and draw VOC-contaminated groundwater.
People may be exposed to VOC-contaminated surface water and sediments where storm sewer effluent enters Swan Creek, but detected levels were not a health concern. Contaminated air released from storm sewer effluent is an indeterminate public health hazard. The Preliminary Health Assessment for Delavan Well No. 4 recommended testing of air for contamination released from extracted water disposed into the storm sewer, which comes out at Swan Creek. Testing was only conducted outside of one disposal point. Surface water and sediment samples were not collected from a five-acre pond located just south of the Sta-Rite property. This is a data gap.
The following discussion describes how people may become exposed to site-related contaminants. There are several ways people are exposed to contamination from a site. "Completed Exposure Pathways" are those pathways where there are indications people were exposed to contaminants from the site and sufficient information exists to evaluate such an exposure. All five of the pathway elements must exist for there to be a "Completed Exposure Pathway" (a description of these five elements is found in Appendix C). This considers exposures that likely occurred in the past and exposures that are currently occurring. A "Potential Completed Pathway" is when there is insufficient information to link a contaminant or chemical to a known level of exposure among an identified population. This is an exposure that may have occurred in the past, is possibly occurring, or which may occur in the future. An exposure pathway can be eliminated from consideration if at least one of the five elements is missing and will never be present.
A. Completed Human Exposure Pathways
Starting in June 1993, VOC contamination was removed from water entering in the Delavan water supply. This was a result of the new water treatment facility that the City of Delavan brought on-line at that time.
It is not known when Delavan Well No. 4 first began drawing VOC-contaminated water originating from the Sta-Rite property, but available information suggests the well may have become contaminated between 1968 and 1970. Well No. 4 first began operating in 1968 [33]. Sta-Rite Industries reported TCE was used at Plant 1 as a paint thinner and degreaser from 1960 to 1977 [66], and the use of TCE for cleaning and degreasing occurred at Plant 2 from 1968 till 1977 [56, P4.1]. Solvents were released to the ground at Sta-Rite via floor drains, catch basins, and sumps, starting in the early to mid 1960's. Regional groundwater flow in the vicinity of Well No. 4 is to the northwest and estimated to be between 135 and 500 feet per year [59, P4.18] [55, P22]. The velocity of groundwater flow increases the closer it comes to the well because of the gradient caused by the drawing of water into the well [58, P4.18]. Well No. 4 is approximately 500 feet from the contaminated area at Sta-Rite Plant 2. Given this distance, the estimated flow rates, and assuming that the contamination in Well No. 4 came exclusively from Plant 2 (as suggested by the Sta-Rite Remedial Investigation), it is reasonable that VOC contamination from Sta-Rite first reached Well No. 4 between 1968 and 1970. However, Plant 1 could have contributed to the contamination detected in Well No. 4. Investigations on Sta-Rite property conclude that groundwater contamination originating at Plant 1 does not reach Well No. 4. If it is assumed that only contaminated groundwater from Plant 2 affected Well No. 4, then the well may have drawn contaminated groundwater during 1968.
It is not possible to confidently estimate contaminant concentrations in tap water arriving at Delavan homes and businesses between 1982 and 1993. There are no reports of tap water samples collected from any residences or business in Delavan. Water samples were collected from taps at two schools and the Delavan City Hall [52], but no contamination was reported. Further complicating this issue, the Delavan water supply system does not uniformly mix water from all municipal wells before distribution.
In 1991 and 1992 water samples were collected from two distribution points located near Well No. 4, and the Delavan central treatment plant (located near Well No. 3). These results are presented in Table 8 and the sampling locations are shown in Figure 4. These data show a relationship between the levels of contamination in water at on-line municipal wells and that found in nearby distribution points. This probably reflects contaminant concentrations found in nearby taps. The concentration of contaminants in water arriving at each Delavan tap is probably affected by a number of factors, including: the concentration of contamination in each operating well; the proximity to each operating well; proximity of each tap to reservoirs; relative pumping rates of each well on a given time and date; the relative degree of water use in the distribution network; and sizes of distribution pipes serving the tap.
The worse-case average contaminant concentrations in tap water can be estimated by extrapolating concentrations observed in 1982 and 1983 to levels in the past. The levels of contaminants in Well No. 4 remained stable for the first two years following detection (Table 1). The decrease in contaminant levels after 1983 could be attributed to any of a number of actions, including: increased pumping of Well No. 3; decreased pumping of Well No. 4; removal of contaminant sources; or movement downgradient of higher pockets of one or more groundwater contaminant plumes. Consequently, prior to 1982 the levels of contamination may have been lower, higher, or the same as that found during 1982 and 1983. Therefore, using the middle range, the highest monthly level for each VOC contaminant detected in Well No. 4 in 1982 and 1983 were averaged to derive the following estimated pre-1982 concentrations. The estimated worse-case average concentrations for tetrachloroethylene (PCE) and trichloroethylene (TCE) are 33 µg/L and 244 µg/L, respectively.
Even if these worse-case concentrations accurately reflect past levels of contaminants in Well No. 4, the discussion above describes the pitfalls of estimating contaminant concentrations in Delavan residential tap water. Despite this, it might be useful to create a worse-case scenario of the highest potential human exposure to contaminants in residential tap water. This scenario could use the middle range, worst-case concentrations and assume that water from Well No. 4 received very little mixing, and the number of Delavan residents receiving undiluted water from Well No. 4 was equal to the proportion of water provided by that well.
As reported above, in 1981 Well No. 4 pumped an average of 12 million gallons per month, approximately 58 percent of the City of Delavan's municipal water needs. The number of Delavan municipal water users in 1981 was estimated to be 4,200. Using this approach, an estimated 2,440 (58% of 4,200) Delavan residents were possibly exposed between 1968 and 1982 to VOC contaminated water from Well No. 4. These users of contaminated, municipal water probably ingested VOCs when drinking contaminated water, inhaled VOC vapors released from domestic use of contaminated water, and absorbed TCE through their skin while bathing and washing in contaminated water.
It is apparent that Delavan residents were exposed to contaminated municipal water prior to 1982. However, the lack of information about contaminant concentrations in tap water does not permit confident estimates of past levels of human exposure. Consequently, the Delavan municipal water supply system is an indeterminate public health hazard before 1982.
B. Potential Human Exposure Pathways
Since 1982, three private wells within the City of Delavan have been identified in the area presumed to be in the path of contaminant plume(s) flowing from the Sta-Rite property. Two of the wells were not reported to be used for human consumption purposes; a water sample collected from one of these wells did not reveal contamination. Information is not available about the use or testing of the third well.
Samples were also collected from several private wells located within Delavan city limits and in the immediate vicinity. All private wells used by homes were found to be uncontaminated. Three wells used by businesses in an industrial park just beyond the northeastern corner of Delavan, located less than 1,000 feet north of Sta-Rite property, were contaminated with low levels of VOCs that are not a health concern, including TCE at 2.6 and 4.5 µg/L.
Surface Water Run-off from Sta-Rite
Public use of this pond is not discouraged as the City of Delavan has made public access improvements northwest of the pond. While there are no reports of people swimming at the pond, people are fishing here.
Comus Lake Surface Water
Sediment from Sta-Rite Run-off
Contaminated Air Released from Storm Sewers
VOC-contaminated air released from storm sewers is an indeterminate public health hazard. The lack of data does not permit an adequate evaluation to determine people are exposed to VOC-contaminated air from the Delavan storm sewer.
Low levels of VOC contamination measured in water provided by Delavan Well No. 4 between 1983 and 1993 does not represent a public health hazard. A review of reported health data did not show an unexpected number of cancers or birth defects among Delavan residents.
Tetrachloroethylene was consistently found in Well No. 4 water samples collected from March 1982 to December 1983. The highest level detected was 97 µg/L, and, during this period, the average of monthly peak levels was 33 µg/L. After 1983 PCE has been detected less frequently in Well No. 4, and at much lower levels (see Table 1 and Table 2).
Delavan residents were probably not exposed to PCE during the time Well No. 4 was disconnected, from 1983 to 1990. When Well No. 4 was brought back on-line in 1990, PCE was occasionally detected, but never above 1 µg/L. PCE has never been found in water samples collected from Well No. 3. Therefore, since 1982 the exposure of Delavan residents to PCE is not an apparent public health hazard.
Before 1982 the levels of PCE in Well No. 4 are unknown because no water samples were collected and analyzed for VOCs. A worse-case would be to assume that 33 µg/L is representative of PCE concentrations in Well No. 4 prior to 1982. A worse-case scenario estimates that PCE may have been present at this level in Well No. 4 for 12 years, from 1970 till 1982 (see page 22 for a further discussion). Using this scenario, past exposure from PCE possibly posed "no apparent increased cancer risk"" [4, P47] to people who used municipal water provided by Delavan Well No. 4. The highest level of PCE detected in Well No. 4 (97 µg/L) is not expected to cause any other adverse, non-cancer health effects from a long-term exposure.
EPA formerly categorized PCE as a probable human carcinogen, but this classification is currently being re-evaluated. The Wisconsin Department of Health and Family Services is aware that PCE may reasonably be anticipated to be determined a carcinogen because it causes cancer in laboratory animals. Some studies have suggested a potential relationship between exposure to PCE and some forms of cancer, but these human and laboratory animal studies are inconclusive [3, P60]. There is no empirical evidence proving PCE causes cancer in humans. Studies of laboratory mice have shown increases in liver cancer when exposed to much higher levels of PCE than what was detected in Delavan Well No. 4 (386 mg/kg/day or the drinking water equivalent of 13,510 mg/L) [3, P28]. Assuming PCE is a carcinogen, a person would have "no apparent increased risk" [4, P47] of cancer if they were exposed for 12 years to drinking water contaminated with PCE at a level of 33 µg/L. Refer to Appendix D for cancer risk estimation methods.
Trichloroethylene was found in 88 of 91 water samples collected from Well No. 4 over the period of March 1982 to December 1985. The City of Delavan halted the use of Well No. 4 in July 1982. The highest levels of TCE observed in Well No. 4 were between March 1982 and December 1983, with the highest level of 1,300 µg/L. The average of monthly peak level of TCE over this period was 244 µg/L. Low levels of TCE continue to be regularly found in untreated water from Well No. 4. In 1991 TCE was found in 64 of 65 water samples collected from Well No. 4, with the range of concentrations between 1 and 7 µg/L (Table 2).
Initially, TCE-contaminated water in the Delavan water supply came exclusively from Well No. 4. But in June 1983, one year after halting the use Well No. 4, low levels of TCE were detected in water from Well No. 3. Between June 1983 and December 1985 TCE was found in 32 of 53 water samples collected from Well No. 3. The highest TCE concentration detected was 230 µg/L, though other peak monthly levels during this period were between 16 and 2 µg/L. The levels of TCE in Well No. 4 before March 1982 are not known because no water samples were collected. See Table 1 and Table 2 for a summary of TCE levels in Delavan Wells No. 3 and No. 4.
Delavan residents were exposed to low levels of TCE from 1983 to 1990, when Well No. 4 was disconnected from the municipal water system. After this time, all of their TCE exposure came from Well No. 3. During this time, the highest level of TCE detected in Well No. 3 was 230 µg/L, though tests of many samples collected during this time did not detect TCE. If TCE was detected it was typically between 2 and 20 µg/L. When Well No. 4 was brought back on-line in 1990, TCE was usually detected, yet never above 7 µg/L. There are no known adverse health effects from exposure to TCE at such levels. Consequently, between 1982 and the present, there are no adverse health effects expected to occur from exposure to water from the Delavan municipal water supply system.
The levels of TCE in Well No. 4 before March 1982 are unknown because no water samples were previously collected and analyzed for VOCs. The worse-case would be to assume that 244 µg/L is representative of TCE concentrations in Well No. 4 prior to 1982. Such a scenario estimates 244 µg/L as the average concentration present in Well No. 4 for 12 years, from 1970 till 1982 (refer to page 22 for a further discussion). Using this scenario, such a past exposure to TCE probably posed a "no apparent increased cancer risk" [4, P47] to people who used municipal water provided by Delavan Well No. 4. The highest level of TCE detected in Well No. 4 (1,300 µg/L) is not expected to result in any adverse, non-cancer health effects in people.
The EPA formerly categorized TCE as probable human carcinogen, but this classification is currently being re-evaluated. There is no definitive evidence that shows TCE causes cancer in humans, but some studies suggest higher concentrations of the chemical may cause cancer in laboratory animals. Two recent studies indicate that exposure to water contaminated with TCE might be associated with congenital cardiac malformations. In one study, infants of mothers exposed during the first trimester of pregnancy to TCE-contaminated water were significantly more likely to have a congenital heart disease than did infants whose mothers did not have a similar exposure [49]. Another study examined rats for a dose-dependent relation between fetal exposure to TCE and various congenital cardiac defects and a possible relationship was noted [16]. An increase in liver cancer was found in laboratory mice when exposed to a high level of TCE (the drinking water equivalent of 35,000 mg/L), but the results are not conclusive [2, P27]. If it is assumed TCE is a carcinogen, as previously defined, a person would have "no increased" risk of cancer if they were exposed for 12 years to drinking water contaminated with TCE at a level of 244 µg/L. Refer to Appendix D for cancer risk estimation methods.
B. Health Outcome Data Evaluation
A review of health outcome data is appropriate when there is evidence that people living near a Superfund site have been exposed to contaminants at levels that might plausibly lead to an increase in rates of death or disease. "Health Outcome Data" refers to records of death and/or illness. A review also may be appropriate if there are reports of unusual clusters of disease near the site or due to specific community health concerns. As discussed previously in the Pathways Analyses section, Delavan residents have apparently been exposed to contaminated water pumped by Well No. 4., though it is unclear if residents were exposed to contaminant levels which were at levels of health concern. The Wisconsin Department of Health and Family Services, Division of Health, is not aware of any reports of clusters of chronic disease or cancers in the vicinity of this site.
The Wisconsin Division of Health requested data from the Wisconsin Cancer Reporting System for reported cases of specific cancers diagnosed in people who described Delavan as their zip code of residence. These data were obtained for the period of 1980 to 1991 reporting the occurrence of liver cancers, kidney cancers, bladder cancers, cancers of the urinary tract, and leukemias. The incidence rate was calculated for each of these reported cancers and none were above expected levels. Additionally, birth certificate data were searched for Delavan births for the period 1968 to 1989. A record of each birth certificate was flagged and pulled based on the occurrence of any reported birth defect. There were no apparent birth defects clusters, and the number of reported birth defects were not above expected levels.
C. Community Health Concerns Evaluation
No community health concerns pertaining to Delavan Well No. 4 have been received by the Division of Health or the Walworth County Nursing Service agency.
Next Section Table of Contents
