PUBLIC HEALTH ASSESSMENT
SILRESIM CHEMICAL CORPORATION
LOWELL, MIDDLESEX COUNTY, MASSACHUSETTS
To identify possible facilities that could contribute to the air, surface water, and surface soil contamination near the Silresim site, MDPH searched the Toxic Release Inventory (TRI) database for 1987 and 1988. TRI is a database developed by EPA from chemical release information provided by certain industries. TRI contained information reported by seven industries within a one-mile radius of the Silresim site. All reported releases within that radius were to the air rather than to soil or surface water. Chemicals released to the air include bromomethane, xylene, 2-methoxyethanol, toluene, methyl ethyl ketone, sulfuric acid, sodium hydroxide, hydrochloric acid, methyl isobutyl ketone, and di(2-ethylhexyl)phthalate.
The following sections briefly describe the known compounds of concern detected in the study
area. Compounds of concern are compounds with concentrations in at least one environmental
medium exceeding the health assessment comparison values established by ATSDR. That
section is used for preliminary screening of the contaminants; the contaminants listed will be
further assessed in the following sections. Information on the health implications of identified
compounds is discussed in the Exposure Pathways and Toxicological Implications sections. The
presence of a contaminant on the compound of concern list does not imply that a human health
threat exists. Some contaminants may be eliminated because of an incomplete past, present, or
future exposure pathway.
A. On-Site Contamination
Each of the five groundwater investigations will be described before the qualitative presentation of their results. Table C-1 provides quantitative results of all groundwater sampling and analysis. Sampling locations can be found on figure A-2.
During the hydrogeological investigation (HI), groundwater samples were collected from nine newly installed monitoring wells (5 deep wells, 4 shallow wells) . Samples were also collected from two previously installed wells on the southeastern boundary of the site and on the corner of Canada and Main streets. Samples were collected and analyzed for EPA priority pollutants, including VOCs, herbicides, pesticides, and PCBs. The Addendum to the HI, which involved the installation of off-site monitoring wells, will be discussed in the section dealing with off-site contamination .
Phase I of the RI/FS entailed the monitoring of groundwater from 39 newly installed monitoring wells in 19 locations and from 18 previously installed monitoring wells in 12 locations. Six monitoring wells were installed on-site, and 57 samples were collected and screened for VOCs.
During the Phase II groundwater investigation, 16 additional monitoring wells and nine piezometers were installed at 21 locations. A total of 83 (on-site and off-site) groundwater samples were collected and screened for VOCs, and seven samples were collected from on-site monitoring wells for laboratory analysis. During the RI Addendum, 72 samples were analyzed for Hazardous Substance List (HSL) compounds or VOCs (13 on-site, 59 off-site).
In June 1991, the EPA completed a Supplemental Remedial Investigation which included the installation of six monitoring wells in three locations. In addition, six previously installed wells were also sampled. Three on-site and ten off-site samples were collected and analyzed for VOCs, semi-VOCs, and metals .
Compounds of concern detected during the analysis of samples from both the overburden and bedrock aquifers included these: trichloroethene (TCE), tetrachloroethene (PCE), benzene, carbon tetrachloride, styrene, chloroform, 1,1,1-trichloro-ethane (TCA), 1,1-dichloroethane (1,1-DCA), 1,2-dichloro-ethane (1,2-DCA), styrene, 1,1-dichloroethene (1,1-DCE), lead, arsenic, aroclor 1242, carbon disulfide, lindane, and 1,2,4-trichlorobenzene (1,2,4-TCB).
With the exception of benzene, the highest concentrations were detected in MW-405B. The boring log noted that a solvent odor and a petroleum odor were detected in the soil from 1.5 to 3.0 feet and 3.0 to 4.5 feet below ground surface, respectively, at MW 404B. With the exception of TCA and 1,1-DCA, the contaminant concentration showed large variation between the wells and the various studies. Generally, the compound concentrations were higher during the RIs than during the HIs and higher in the overburden when compared to the bedrock wells.
The on-site soil data reviewed for this health assessment were obtained from the same studies as
the groundwater data [21,22,44,45]. Quantitative results are presented in Tables C-2 and C-3,
and sample locations are shown on Figure A-2.
a. Subsurface Soil
The HI soil sampling consisted of test pit samples collected at a depth of four to eleven feet in high-use areas of the site, primarily where past chemical storage had occurred. The Phase I RI involved the collection of samples at five-foot intervals in the boring locations. The Phase I RI soil sampling also consisted of a test pit program designed to further characterize potential buried objects that had been reported on site by NUS Corporation in 1984, and surveyed by S.A. Alsup and Associates in February 1986 . Nine test pit soil samples, from either the test pit walls or the backhoe bucket, were collected.
Additional soil sampling was conducted in November 1988, as part of the Addendum to the RI in order to supplement existing data on contamination in unsaturated zone soils. All samples from the 21 on-site sampling locations were screened for total VOCs, and selected soil samples were screened for PAHs, PCBs, and pesticides. Generally, the sample with the highest VOC level for each boring was submitted for HSL organic and inorganic analysis.
In December 1990 as part of the Supplemental RI to determine the nature and extent of contamination, six soil samples were collected from three locations. Four on-site and two off-site samples were collected and analyzed for VOCs and semi-VOCs.
The most prevalent subsurface soil contaminants of concern are: TCE, PCE, benzene, chloroform, 1,1,1-TCA, styrene, 1,1-DCA, 1,2-DCA, 1,1-DCE, lindane, aroclor 1242, aroclor 1221, aroclor 1016, aroclor 1248, 1,2,4-TCB, 2,4-dichloropropane, 2,4,5-trichlorophe-noxyacetic acid (2,4,5-T), arsenic, lead, aldrin, 4,4'-DDD, 4,4'-DDT, PAHs, and dioxins/furans.
Predominant areas of subsurface dioxin/furan contamination are the northeastern and southeastern corners of the site at total concentrations of 668 ppb and 62.15 ppb, respectively. Various forms of dioxins/furans, called isomers or congeners, can be found in the environment. Those congeners vary by the number of chlorine atoms present in the molecule, as well as by the location of the chlorine atoms within the molecule. Generally, the 2,3,7,8-TCDD congeners are considered the most dangerous to human health. The OCDD/OCDF congeners were the most frequently detected; the 2,3,7,8- TCDD/TCDF congeners were not detected. Other congeners present in the subsurface soil were TCDD/TCDF, PeCDD/PeCDF, HxCDD/HxCDF, HpCDD/HpCDF. The 2,3,7,8-TCDD total toxic equivalent concentrations ranged from 0.61 ppb to 10.42 ppb. Concentrations were generally highest at a depth of 4-7.5 feet.
3. Surface Water
Surface water monitoring was conducted on February 26, 1986, and April 10, 1986, during the Phase I RI. In the Phase I RI workplan, three on-site sampling points were established to monitor surface water entering the on-site storm drains and flowing to River Meadow Brook. However, it was observed that, even during heavy precipitation, runoff was insufficient to generate flow in the storm drain flowing from the site to River Meadow Brook. Therefore, those sampling points were eliminated, and no on-site surface-water samples were collected during any of the three surface-water monitoring programs.
4. Ambient Air
The results of on-site ambient air monitoring , presented in Table C-5, were obtained from five investigations conducted between 1982 and 1983 prior to the installation of the clay cap [33,35,41,42,58]. During the air monitoring studies, the sampling stations were relocated based on changing wind directions. The following compounds were detected during the four studies that monitored for VOCs: benzene, TCE, PCE, toluene, xylene, carbon disulfide, 1,1,1-TCA, styrene, chloroform, and carbon tetrachloride. With the exception of carbon disulfide, all VOCs were detected at low concen-trations (<20 ppb). Carbon disulfide was detected at concentrations ranging from 40.3 to 144.7 ppb. One study analyzed the air for oxygen- and nitrogen-based compounds. Low levels of phenols and amines were detected.
A particulate sampling station was established at the site perimeter downwind from the working area in order to measure particulates during the Interim Remedial Measures (IRM). The samples were analyzed for semi-volatile and non-volatile constituents, including PCBs and pesticides. The following compounds were detected in the downwind particulate samples at low levels (<1.5 µg/m3): 4,4'-DDT, 4,4'-DDE, endrin aldehyde, dieldrin, aldrin, endrin, BHC (1,2,3,4,5,6- hexachlorocyclohexane), and endosulfan sulfate. Throughout the particulate sampling, styrene was detected at concentrations ranging from 6.8 to 79.6 µg/m3.
5. DNAPL Monitoring
The Phase II RI sampling program included an investigation to determine the presence of dense non-aqueous phase liquid (DNAPL) in selected monitoring wells on site and immediately north of the site. Monitoring was performed by lowering a transparent PVC bailer into the well and checking for any measurable amount of DNAPL upon retrieval of the bailer. Ten on-site wells and nine off-site wells were monitored for DNAPL. One on-site monitoring well (MW-405B, set at 4 to 14 feet), contained 0.8 feet of DNAPL. Results of the off-site well sampling will be discussed in the section on off-site contaminants.
6. Air Vent Monitoring
Air vent monitoring for VOCs was conducted on November 5, 1986, as part of the Phase I RI for
the purpose of quantitatively and qualitatively analyzing emissions from the vents. Five
ventilation networks, each equipped with two vents and charcoal filters, are on site. Each
network was screened, and the vent with the highest total VOC concentration was selected for
sampling. Five vents were sampled; the maximum concentrations of volatile compounds found
are as follows: 377 ppm TCE, 69.7 ppm PCE, 0.96 ppm benzene, 83.5 ppm 1,1-DCA, 7.23 ppm
1,2-DCA, 18.9 ppm 1,1-DCE, 564 ppm 1,1,1-TCA, and 14.1 ppm chloroform. Charcoal filters
were removed from the vents before sample screening and collection.
B. Off-Site Contamination
For ease of presentation, the off-site sampling areas will be described as follows (see Figure A-2):
- Area 1: encompasses the area north of the site, including Lowell Iron and Steel/Scannell Boiler Works, bordered to the east by the B&M railroad tracks and to the west by Tanner Street.
- Area 2: encompasses the area to the east of the railroad tracks, including East Pond and the Maple Street Condominiums. The area is bordered to the south by the Maple Street residential area, and to the north by the Olive Street residential area.
- Area 3: encompasses the area extending south of the Silresim fence line, bordered to the east by the railroad tracks, and to the west by Tanner Street. The area includes the Lowell Used Auto Parts property, formerly, the Arrow Carrier Property.
- Area 4: encompasses all property west of Tanner Street, including River Meadow Brook and extending to the Lowell Connector.
Groundwater monitoring studies are described in the section on on-site groundwater contamination. A total of 116 samples were collected from the off-site wells: 50 from Area 1, 26 from Area 2, 27 from Area 3, and 13 from Area 4.
Area 1 (Lowell Iron and Steel Property): Groundwater monitoring in this area consisted of the collection of 45 samples between December 1981 and February 1989. Laboratory analysis detected the following contaminants of concern: TCE, PCE, benzene, carbon tetrachloride, chloroform, 1,1,1-TCA, 1,1-DCA, 1,2-DCA, 1,1-DCE, styrene, lead, arsenic, and trace levels (<1.0 ppm) 1,2,4-TCB, aroclor 1260, 1,1-DCE, and carbon disulfide. Generally, the highest concentrations were detected in MW-102B and the MW-309 series wells (overburden and bedrock).
Area 2 (east of the railroad tracks): Groundwater monitoring involved the collection of a total of
26 samples from 17 locations between December 1981 and February 1989; the majority (17) of
the samples were collected in 1989. Some samples were only qualitatively screened and
therefore are not included in Table C-1. Contaminants of concern detected in the eastern area
wells were these: TCE, PCE, benzene, 1,1,1-TCA, 1,2-DCA, aroclor 1260, 1,1-DCA, 1,1-DCE,
and chloroform. With the exception of TCE, all compounds were detected at less than 1.0 ppm.
The highest concentrations were detected in MW-502 which is located approximately 25 feet
south of the eastern edge of the pond (see figure A-2). Concentrations detected during screening
were generally higher than those detected during laboratory analysis.
Area 3 (south of the site): 30 samples were collected from 1981-1989 from a total of 24 locations. Only four sampling locations showed levels of contaminants of concern above 1.0 ppm: B-2, GW-1, MW-306A, and MW-306B (see figure A-2). Detected contaminants of concern were these: TCE, PCE, benzene, carbon tetrachloride, chloroform, 1,1,1-TCA, 1,1-DCA, 1,1-DCE, 1,2-DCA, lindane, aroclor 1260, arsenic, and lead. GW-1 and MW-306B contained the highest levels of the previously mentioned contaminants.
Area 4 (west of Tanner Street): Samples were collected from nine locations. Five of the nine locations contained the following contaminants of concern at low concentrations (<0.5 ppm): TCE, PCE, benzene, carbon tetrachloride, chloroform, 1,1,1-TCA, 1,1-DCA, 1,1-DCE, 1,2-DCA, aroclor 1260, arsenic, and lead.
Off-site soil monitoring was conducted during the HIs and the RIs (Phase I and II, and
addendum). Details on the soil monitoring program are presented in the section on on-site soil
contamination. A brief description of the off-site soil contamination follows and is summarized
in Table C-2.
a. Subsurface Soil
Area 1 (north of the site): A total of 94 soil samples were collected from 26 locations situated primarily on the land owned by Lowell Iron and Steel. Twenty subsurface samples from seven locations were analyzed; the following contaminants of concern were detected at concentrations greater than 1.0 ppm: 1,1,1-TCA, TCE, styrene, arsenic, lead, and PAHs.
Area 2 (east of site): Sampling near the vicinity of the B&M railroad tracks involved collection of samples from three subsurface points. Subsurface contaminants of concern detected were these: TCE, PCE, 1,1,1-TCA, styrene, PAHs, aroclor 1254, arsenic, and lead. Generally, the higher concentrations of volatile organic compounds were detected in the subsurface soil samples, and the non-volatile compounds were found in surficial soils.
Area 3 (former Arrow Carrier property): Subsurface sampling to the south of the site involved collection of four samples. The following VOCs were detected at concentrations greater than 1.0 ppm: TCE, 1,1,1-TCA, carbon tetrachloride, benzene, and PCE. Other contaminants of concern detected in soils south of the site were 1,2,4-TCB, PAHs, aroclor 1221, 2,4,5-T, arsenic, and lead.
b. Soil (0 - 12 inches)
ATSDR requires that surface-soil samples be collected within the top three inches of soil. During the RI, surface-soil samples were generally composited from the top six inches of soil. For the purpose of this health assessment, these samples will be classified as soil (0 - 12 inch) samples rather than surface-soil samples.
Additional Phase I soil monitoring consisted of the collection of two off-site soil samples for HSL organic compounds, priority pollutant metals, iron, manganese, cyanide, and dioxin analysis. Using the results of this analysis, an expanded sampling program consisting of three soil samples was conducted.
Soil sampling was also conducted during the Phase II RI program. Sampling during the Phase II RI took place in five phases, from October, 1986, to June, 1987. Five samples were collected for HSL analysis, and three samples were collected for analysis for mercury, arsenic, and chromium from the eastern site perimeter and the former Arrow Carrier property.
Seven additional samples were collected near the expanded fence line at the southern corner of the site because of the possible presence of dioxin in that area. Samples were analyzed for dioxin and mercury, and screened for PCBs. In June, 1987, additional sampling consisted of eight surficial soil samples for dioxin/furan analysis and 11 samples for PCB and metals analysis. Three dioxin samples were grab samples; five samples were composites of 25- by 50-foot areas.
Phase III of the soil sampling program consisted of 25 samples analyzed for VOCs, semi-volatile compounds, PCBs, and pesticides. This sampling focused on an area of the Lowell Iron and Steel property where runoff from Silresim had been released. Ten of the locations were resampled for Hazardous Substances List (HSL) organic compounds, arsenic, chromium, mercury, and lead, and five composite samples were collected for herbicide and dioxin analysis.
Area 1 (north of the site): Soil samples were analyzed for VOCs, Acid/Base/Neutral Compounds (ABN's), pesticides, PCBs, dioxins, and inorganic compounds (metals). Five composite samples were also analyzed for selected pesticides and PCBs. Generally, volatile organic compounds were found at levels less than 1.0 ppm in the soil samples, except for 1,1,1-TCA, TCE, and PCE. On the other hand, aroclor 1254, 4,4'-DDE, 4,4'-DDT, lindane, arsenic, lead, and OCDD/OCDF were detected at elevated levels.
Area 2 (east of the site): Soil samples were analyzed for VOCs, ABNs, pesticides, PCBs, and inorganics. VOCs, with the exception of TCE in one sample, were less than 1.0 ppm. Aroclor 1254 was the only pesticide/PCB detected at levels greater than 1 ppm. Elevated concentrations of PAHs, arsenic, and lead were also detected in this area.
Area 3 (former Arrow Carrier property): Samples were collected and analyzed for VOCs, semi-volatile organic compounds, dioxins, and metals (inorganic). Only one surface sample contained VOCs at concentrations greater than 1.0 ppm. Dioxins/furans were detected at the southeastern corner of the fenceline. Highest concentrations were generally within 10 to 20 feet of the fence, in an area currently covered with crushed stone.
Area 4 (west of Tanner Street): Three samples were analyzed for VOCs, pesticides, and PCBs. No contaminants of concern were detected at levels greater than 1.0 ppm.
3. Surface Water
Off-site surface-water monitoring was conducted in February 1978, as part of the "Analysis of Hazardous Waste Mismanagement Incident" report; in October, 1981, during the HI; and in October 1989, in accordance with the National Oceanic and Atmospheric Administration (NOAA) request dated May 15, 1989 [23,45]. Surface-water data are summarized in Table C-3. During the investigations, surface-water samples were collected from River Meadow Brook, Merrimack River, Concord River, East Pond, and the basement of the former Arrow Carrier building. TCE, PCE, and lead were the only compounds of concern detected in the surface-water samples. The concentrations of those compounds, however, did not exceed 19 ppb.
Sediment monitoring took place during the Phase I RI and in accordance with an NOAA request
dated May 15, 1989 . Sampling was conducted on February 26, 1986, and October 26, 1989.
Sediment samples were collected from the same locations as the corresponding surface-water
samples. Several of the samples were composites from different locations. Chloroform,
1,1-DCA, TCE, 1,2,4-TCB, lead, and PAHs were detected in off-site sediment. Sediment data
are summarized in Table C-3.
5. Ambient Air
Five investigations involving ambient air monitoring were described in the on-site sampling section. Sampling stations were established in the neighborhoods around Silresim (both upwind and downwind) as well as throughout the city (for background samples). Sampling stations (upwind and downwind) varied, depending on the wind direction on the day of sampling. Ambient air data are presented in Table C-4.
Four of the investigations analyzed samples for VOCs, and the following compounds were detected around the site and in the background samples: benzene, PCE, TCE, toluene, dimethyl formamide (DMF), chloroform, carbon disulfide, 1,1,1-TCA, styrene, 1,2-DCA, carbon tetrachloride. Generally, the concentrations of those contaminants did not exceed 20 ppb, and concentrations at the background stations were comparable to levels found around the site. Carbon disulfide was detected at concentrations up to 100 ppb downwind from the site. Amines, phenol, and cresol were detected in one investigation, but total concentrations did not exceed 1.0 ppm. Those compounds were generally detected at higher concentrations on site than off site.
Because of the presence of low levels (4.0 ppb) of DMF both upwind and downwind of the site,
an air-monitoring investigation was conducted throughout the City of Lowell. DMF was not
known to have been stored or used at Silresim, and its presence in the air could be from an
off-site source. The investigation focused on the Majilite and Compo Corporations, which are
within one mile of the site (see figure A-1). Elevated levels of DMF ( up to 2800 ppb), toluene
(up to 340 ppb), and xylene (up to 840 ppb) were detected downwind of both facilities; low
concentrations of the compounds were detected upwind of the facilities. The presence of DMF
more than likely was a result of emissions from those facilities rather than from operations at the
Silresim site. Refer to the section on Health Outcome Data Evaluations for a discussion of the
health evaluation performed for residents living near the two facilities.
6. DNAPL Monitoring
During the Phase II RI sampling program nine off-site wells were monitored for DNAPL hydrocarbons. A 0.5-inch floating layer was observed in the one well south of the site (Area 3), and between 0.5 and 2.5 feet of DNAPL were observed in five of the eight wells north of the site (Area 1).
7. Private/Municipal Water Supplies
a. Private Water Supplies
Two private water supplies exist in the general vicinity of the site. A residential water supply well on Main Street, approximately 500 feet south of Silresim, was sampled during the HI and was found to contain low levels of PCE, 1,1,2,2-tetra-chloroethane, toluene, ethylbenzene, styrene, and DMF. The well was supposed to be used for gardening purposes only. According to a resident, however, the well was also being used as a source of drinking water. The Department of Environmental Quality Engineering (DEQE) and the Lowell health department informed the residents of the sampling results and advised them to discontinue using the well as a source of potable water [29,34]. The well was removed in September 1986.
The Lowell Car Wash (approximately 2,000 feet south of the site) uses a groundwater well during the car wash processes. No information is available about the depth of the well; where the well screen is set; the volume of water obtained from the well by the car wash; or any chemical analyses performed on the well .
The latest sampling results (September 1991) indicate that the groundwater contamination plume has not reached the monitoring wells located between the site and car wash. It is unlikely that (1) the contamination in the private well was site related and (2) individuals at the car wash are being exposed to site related contaminants in the well water.
b. Municipal Water Supplies
Lowell has received its water supply from the Merrimack River since 1960. Monitoring of the finished waters for VOCs is performed on a quarterly basis and reported to DEP. Between March 1988 and March 1989, three of the four finished water samples contained chloroform (7.9 to 52 ppb); two of the four finished water samples contained bromodichloromethane (2.0 to 9.5 ppb); and one of the four water samples contained chlorodibromo-methane (1.2 ppb) . The presence of three trihalomethanes is due to the chlorination of the water supply . No other volatile organic compounds were detected in the four finished water samples.
8. Sewer Monitoring
Three monitoring programs were established in order to assess the impact of site contaminants on the city's combined sewer system, which transports both sanitary wastewater and stormwater . Twelve waste-water samples were collected from 10 sampling locations during the three studies. Nine samples were collected west of the site (Area 4), including three upstream samples, four downstream samples, and two samples west of Tanner Street. Three samples were collected from Area 2 east of Silresim. No VOCs were detected in the five upstream samples or the two samples west of Tanner Street. Chloroform (3 ppb), 1,1,1-TCA (45 ppb), TCE (31 ppb), and benzene (34 ppb) were the compounds of concern detected in the samples to the east of the site. The downstream samples contained the highest concentrations of VOCs, including 890 ppb TCE, 10 ppb chloroform, 29 ppb benzene, 18 ppb 1,1-DCA, 43 ppb 1,2-DCA, 4.1 ppb 1,1-DCE, 3,100 ppb 1,1,1-TCA, 63 ppb styrene, 43 ppb carbon disulfide, and 38 ppb carbon tetrachloride.
9. Indoor Air Sampling
An initial study was conducted on October 5, 1988, to determine if VOCs within the groundwater plume were volatilizing into the basement air space in the Administration Building on the Lowell Iron and Steel property, which is north of the site (Area 4). The results of the initial study were not of acceptable quality from which to draw eight-hour concentration averages. A second eight-hour sampling study was conducted on December 6, 1988. Six samples were analyzed for selected VOCs. Compounds of concern detected were TCE, benzene, carbon tetrachloride, 1,1-DCE, 1,1-DCA, chloroform, and 1,1,1-TCA. Four of the six samples had total VOC concentrations of less than 25 ppb. The two samples with higher VOC results contained higher levels (69 and 71 ppb) of methylene chloride and were collected in the basement sump connected to the adjacent 84-inch sewer. The EPA believes that the high hits of methylene chloride are attributable to the sewer because high concentrations of methylene chloride were not detected in the groundwater samples collected adjacent to the building. All other VOCs in the two samples were found at concentrations less than 10 ppb.
As previously discussed, the groundwater contamination plume is also migrating to the south of the site. Based on recent monitoring data (August 1991), it was determined that the groundwater plume has not reached the residences to the south of the site. Therefore, it has not been necessary to conduct indoor air sampling at locations other than the Lowell Iron and Steel facility.
Each of the major investigations summarized in the previous sections on environmental contaminants has its own quality assurance and quality control (QA/QC) procedures. Overall, the sampling, analytical techniques, and data were adequate for the purpose of the health assessment. However, several problems were encountered in the various studies which could have compromised the quality of the data.
Several surface-water, surface-soil, subsurface soil, and sediment samples were collected as composites from different locations. In some studies, the surface soil was also composited from the top 12 inches of soil. Compositing samples in either of those ways could grossly over- or underestimate the actual contaminant concentration.
A number of samples for VOC analysis were analyzed beyond the maximum holding time. Those results may underestimate the VOC concentrations due to volatilization and biodegradation of the contaminants during the holding time. Also, the presence of acetone and dichloromethane in samples during those investigations were determined to be the result of laboratory contamination. Styrene and DMF concentrations were estimated from ethylbenzene and tetrahydrofuran concentrations, respectively. Therefore, those concentrations could be over- or underestimated.
There was no established QA/QC program for the TRC Environmental Consultants, Inc., report.
(Field Investigation of the Silresim Chemical Waste Site). That report contained several inherent
data limitations: 1) concentrations were determined based on previous instrument calibration; 2)
the analytical instrument is unable to distinguish between isomers of the same compound; and 3)
the instrument can identify a specific class of compound but cannot identify the specific isomer
within the class (i.e., C2-amine).
D. Physical and Other Hazards
The presence of physical and other hazards was investigated during the November 20, 1990, site visit. Based on that visit, it is believed that the site is generally clean and well-maintained. All on-site buildings have been demolished, and no permanent structures are currently on site.
A storage trailer, the work-staging area, and an empty purge water tank remained on site following the remedial investigation. Several empty, 50-gallon drums were near the storage trailer. A pile of crushed stone was left on site after recent work performed by EPA contractors.
The items mentioned here would be accessible to unauthorized individuals who access the site.
However, it is unlikely that people will gain entry; the site is surrounded by an eight-foot-high
fence topped with barbed wire. The two access gates are well-maintained, and warning signs are
prominently placed on the gates. A small child might be able to fit through the fence, but there
was no evidence of human activity on site. Substantial fence repairs were performed in 1991 due
to vandalism. DEP personnel occasionally walk the perimeter of the fence for signs of fence
damage and unauthorized entry.
As discussed in the Site Description and History subsection, many past industrial operations at
the site have resulted in soil and groundwater contamination via chemical spills and leaks. The
following subsections discuss the environmental and human exposure pathways at the site.
A. Environmental Pathways (Fate and Transport)
1. Hydrogeology of the Site and Study Area
Overall, the site gradually slopes upward to the south and downward toward River Meadow Brook. The overburden materials at the site and study area are characterized as glacial deposits with overlying organic deposits and fill. Glacial till and outwash (sand and silt) layers of varying thickness were found in borings dispersed throughout the site and study area. Sandy clay layers were occasionally observed in the outwash. Organic soils were found in areas such as River Meadow Brook and East Pond, and were sporadically dispersed underneath the site and throughout the study area.
The on-site fill is composed of the following three distinct units: 1) clay or crushed stone surface cover approximately one foot thick; 2) gravel ranging from 0 to 4 feet thick; and 3) miscellaneous granular fill from one to nine feet thick. The miscellaneous granular fill consists of fine to coarse sands interspersed with cinders, ash, brick, wood, and various other debris. Pockets of clay material were noticed in the fill near the center of the site.
The on-site bedrock topography varies from the low bedrock elevations found north of the site, to rock found at depths greater than 90 feet. From the low point, the bedrock slopes sharply to the north and more gradually to the south, forming a bedrock valley that runs east-west. That valley is believed to be part of the ancient Merrimack River Channel, which runs north-south along approximately the same course as River Meadow Brook and turns east near the site. North of the bedrock valley, rock surface elevations continue to rise. The extent of weathering and fracturing of the bedrock varies across the study area.
Surface and groundwater movement around the site is variable because of the presence of various structures, surface water bodies, and the extensive municipal sewer system. Surface water on the site is routed to a drainage swale (installed in 1984) and ultimately discharges into River Meadow Brook west of the site. Runoff from the western perimeter of the site and from the clay capped portion of the site enters a drainage swale and moves westward towards a catch basin located southwest of MW105 at the northwest corner of the site. During major rainfall events flow moves from the catch basin towards the Tanner Street storm drain/combined sewer system. Water from the northern portion of the site collects in a catch basin northeast of monitoring well MW-105 (on Lowell Iron and Steel property and near the northeastern corner of the site. South of the site, runoff appears to pool on the former Arrow Carrier property. Runoff east of the site is constrained by the railroad grade and routed north or south along the railroad tracks.
Groundwater in the overburden aquifer shows a predominantly north-northwest and downward flow gradient toward River Meadow Brook, which ultimately discharges into the Concord River. However, secondary flows exist east and west of the site, with some restricted flow to the south. A groundwater mound in the northeastern section of the site supports radial flow to the north, east, and west. A groundwater high southwest of the site probably limits groundwater flow in that direction. Data about the southeastern section of the site are unclear, but a rise in the topography would probably limit groundwater flow somewhere between Maple and Blossom streets. Limited data on deep overburden and bedrock strata suggest that deeper flow gradients are similar to the flow in the shallow aquifer. The natural groundwater flow has been altered historically via the installation of utilities and groundwater wells; construction of buildings and the Lowell Connector; rerouting of River Meadow Brook and the active filling of lowland areas; and, recently, because of the installation of the clay cap on site.
The River Meadow Brook originates approximately six miles southwest of the site in a wetlands area. North of the site, it bends east and converges with the Concord River approximately one mile south of its convergence with the Merrimack River. The River Meadow Brook is a low-flow, shallow brook in the sub-basin of the Concord River Basin. The East Pond is a 30,000-square-foot pond with variable water levels based on local groundwater elevations and seasonal fluctuations.
To determine whether nearby residents are exposed to contaminants migrating from the site, factors influencing human exposure were evaluated. The pathways analysis consists of five elements: a source of contamination, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population.
Pathways are identified as completed, potential, or eliminated. For a completed pathway to exist, five elements must be identified, and there must be evidence that exposure to a contaminant has occurred, is occurring, or will occur. Potential pathways exist when at least one of the five elements is missing. Potential pathways indicate that exposure to a contaminant could have occurred in the past or could occur in the future. Eliminated pathways require that at least one of the five elements is missing and will never be present. The discussion that follows incorporates only those pathways that are important and relevant to the site.
Two pathways, namely, groundwater and subsurface soil, have been eliminated from further discussion in this report because there has been no exposure to the contaminants in those media, and there will not be any exposure in the future that would result in adverse health effects. Elevated levels of VOCs, PAHs, and metals have been detected in at least one of the media. However, the groundwater is not used as a drinking water source, because the city is supplied with municipal water from a surface-water source. There are no drinking-water wells in the site vicinity which could be impacted by the groundwater plume, and all buildings in the city are required to use the municipal drinking water source . Concern has been expressed that the groundwater plume could migrate to the River Meadow Brook and eventually reach the Merrimack River, which is the drinking water source for the city. However, high levels of contamination have not yet reached River Meadow Brook. If high levels of contaminants eventually reach the brook, they will not adversely impact the Merrimack River because of the diluting effect as the contaminants migrate from River Meadow Brook, to the Concord River, and then to the Merrimack River. Subsurface contaminants are not currently an exposure route because the City of Lowell requires environmental testing before allowing potentially disruptive activities, such as building or digging, which could expose residents and workers in the area to subsurface soil contaminants .
1. COMPLETED EXPOSURE PATHWAYS (see Table E-1)
Ambient Air Pathway
Past exposures to site-related contaminants are possible from the ambient air around the site. Potentially impacted populations include workers on site and in the site vicinity and residents in the surrounding neighborhoods.
On-site and off-site air data collected after the removal actions show migration of benzene, TCE, PCE, toluene, xylene, carbon disulfide, 1,1,1-TCA, styrene, chloroform, and carbon tetrachloride. Those data are from sampling conducted over a short period and do not represent the range of conditions that might have existed over the years. The air-monitoring data show that VOC concentrations in the neighborhoods around the site are generally less than 20 ppb, with the exception of carbon disulfide. Those concentrations were comparable to concentrations detected at background stations throughout the City of Lowell. DMF, toluene, xylene, and 1,2-DCA, however, were detected at much greater concentrations around the site because of other industries in the area, such as Compo and Majilite. Amines, phenol, and cresol were detected in one investigation, but concentrations generally did not exceed 1 ppm.
Using available information, MDPH has determined that levels of VOCs detected in the past are not known to represent a risk to public health. Although certain amines are known to represent a threat to human health at elevated concentrations, it is not possible to assess the historical threat of the Silresim site because the sampling methodology used could not distinguish between the various amine isomers. Therefore, it is not known if the isomers detected were toxic or non-toxic forms of amines.
While no data exist on VOCs in ambient air during site operations or during EPA removal actions, it is suspected that air concentrations of VOCs during those events may have been similar to, or greater than, levels detected during the various investigations. Therefore, past health concerns would be similar to those expected during site-removal activities.
Because of the lack of groundcover, contamination in the surface soils could have been spread via fugitive dusts. Limited air data exist for particulate matter. During remedial measures, particulate samples were collected from one downwind station at the perimeter fence. The compounds detected were 4,4'-DDT, 4,4'-DDE, endrin aldehyde, dieldrin, aldrin, endrin, BHC, endosulfan sulfate, and styrene. All compounds, except for styrene, were detected at <1.5 g/m3. Those samples were collected when work was being performed on site during removal activities, and probably overestimate daily concentrations in particulate matter. Also, the presence of the clay cap and gravel over areas of surface-soil contamination serves to decrease exposure via that route. It is difficult to assess actual exposure because no air particulate data exist for daily conditions at the site.
Soil (0 - 12 inches) Pathway
Past exposure pathways are possible from soil contamination at the following locations: Silresim site, areas near the railroad tracks, and the southeastern corner fence line. On-site contamination originated from chemical spills, leaks, and runoff. Past transport of soil contaminants has probably occurred via several pathways: (1) in solution with surface runoff, (2) bound to particulates and transported by storm runoff, (3) by fugitive dust generation and vapor emissions; and (4) by company workers and children playing nearby carrying the contamination home on their clothes and shoes. Organic solvents (TCE, PCE, benzene, 1,1,1-TCA, 1,2-DCA, and 1,1-DCE) tend to be transported in water filtering through soil into groundwater. Other compounds (PAHs, PCBs, pesticides, metals, and dioxins/furans) tend to adhere to soil particles and are transported via runoff with soil particles or through the air attached to particulate matter.
Residents, including children, and workers in the site area could have been exposed to lead, mercury, PCBs, and dioxins in the past by ingestion of soil, inhalation of fugitive dusts, or dermal contact with contaminants. Those routes of exposure are currently diminished because of the installation of the clay cap and fence, and the interim remedial measures. Currently, in areas outside the fence line, soil ingestion is an important route of exposure, particularly for children less than six years old. Concern about soil ingestion is more pronounced for young children because of their greater hand-to-mouth activity. Young children typically ingest 200 milligrams (mg) of soil per day; adults and older children ingest fewer than 100 mg per day.
East of the site, in the area of the Boston and Maine Railroad tracks, elevated levels of aroclor 1254, PAHs, arsenic, and lead were detected. It has been reported that neighborhood children have used and currently use this area as a shortcut walkway. Principal exposure would be to children, via ingestion. However, adult residents, workers, and children in the area might also be exposed via inhalation of particulate matter and via absorption following dermal contact. Both of those exposure pathways, however, would be minimal compared to the ingestion route. It is important to note that contaminants in the area could be related to the railroad rather than to the site.
Indoor Air Pathway
Past exposure pathways are possible from VOCs in the groundwater plume volatilizing into the basements of residences and industries around the site. Low levels of contaminants (generally <10 ppb) were detected in the basement indoor air samples of the Lowell Iron and Steel building north of the site. Sampling consisted of 8-hour, time-weighted averages and are representative, therefore, only of the time sampling occurred. Actual concentrations could be higher or lower, depending on varying environmental and atmospheric conditions. Assuming that the levels detected are representative of the average concentrations found in the basement, indoor air contaminant levels are unlikely to represent a past health concern for workers at the Lowell Iron and Steel facility.
Past, current, and future exposure pathways are possible from contamination of the sediment in East Pond and River Meadow Brook. The population most impacted by those contaminants would be children who have played or currently play in the pond or brook.
Low levels of VOCs, PAHs, and arsenic and elevated levels of lead have been detected in the sediment of East Pond and River Meadow Brook. Exposure to the VOCs, PAHs, and arsenic do not represent a community health concern because of the low levels detected. However, the lead concentrations pose a health risk to certain members of the population. It has been reported that children have access to River Meadow Brook and East Pond; they would be at the greatest risk of exposure to the lead, through ingestion of contaminated sediment. The levels cited were detected during the RI/FS, and current levels are not known. Also, because of the presence of other industries along River Meadow Brook and previous reported dumping in East Pond, it is unlikely that the lead contamination is related to the site, and it is not expected that sediment lead concentrations would increase drastically in the future. Current and future exposures, therefore, are expected to be similar to past exposures.
2. POTENTIAL EXPOSURE PATHWAYS
Ambient Air Pathway
Historically, migration of contaminants has been detected in the ambient air and particulate matter around the Silresim site. Current and future inhalation exposures to site-related contaminants have been decreased because of the installation of the clay cap and placement of gravel over the areas of high surface- soil contamination. It is not expected that volatilization of contaminants from the surface soil, or migration of contaminated particulate matter, currently or in the future will result in adverse health affects to workers and residents in the site vicinity.
VOCs, however, could still be emitted to the air via the air vent system on the site. Elevated levels of VOCs were detected during the sampling of the air vent system on site. Those samples were collected after charcoal filters were removed from the vents. The charcoal filters are currently on each of the vents in order to prevent the release of VOCs into the ambient air. The condition of the charcoal filters is not known. If the filters are old and saturated, contaminants detected during vent sampling could be passing through the filters and into the ambient air. However, based on the vent sampling results and the air flow rates inside the vents, EPA determined that it was not necessary to filter the vent emissions. Without updated data, it is not possible to fully assess exposures and concomitant health effects.
Soil (0-12 inches) Pathway
Transport of and exposure to contaminants in the soil is currently diminished due to the installation of the clay cap and interim clean-up measures. The site is fenced, and most areas of high contamination outside of the fence are covered with gravel.
Soil at two off-site areas (Lowell Iron and Steel, and the former Arrow Carrier property) are contaminated with VOCs, PAHs, and lead. Soil lead levels ranged up to 130 ppm north of the site on the Lowell Iron and Steel property. PAHs were detected north of the site and around the southeastern corner fenceline at concentrations up to 3.6 ppm. The Lowell Iron and Steel and former Arrow Carrier property are fenced, and areas of highest contamination are covered with gravel; therefore, neighborhood children are unlikely to access areas of contamination on those properties. Individuals working at those businesses are most likely to contact contaminated areas. Ingestion of contaminants on the properties is not a concern, because of the population accessing the areas, and dermal absorption will not be a primary exposure route because of the chemical nature of the contaminants in the surface soils.
Surface Water Pathway
Past, current, and future exposure pathways are possible from VOC contamination of the East Pond and River Meadow Brook. Children playing in or near the pond and brook are the individuals most likely to be affected by the contaminants. The groundwater plume from the Silresim site migrates primarily northwest and secondarily east-southeast. Low levels of contaminants (< 9.1 ppb) have been detected in East Pond and River Meadow Brook. Although the contaminant plume has not yet reached River Meadow Brook, surface- water discharge from the site does enter the brook via a pipeline. Low levels of contaminants from the site have reached East Pond. However, previous dumping in and around the pond has been reported and could contribute to the levels detected.
As stated earlier, it has been reported that children living near the site play in the East Pond and River Meadow Brook. Children could be exposed to VOCs through inhalation of vapors, ingestion of surface water, and dermal absorption of contaminants. However, past and current surface-water contaminant levels do not represent a threat to human health.
If the groundwater plume continues to migrate to those surface water bodies, individuals accessing the pond and brook could be exposed via inhalation, ingestion, and absorption. It is difficult to predict any future exposure to surface-water contaminants without knowledge of contaminant concentrations. Groundwater treatment during remediation is expected to decrease the concentration of contaminants migrating to the surface water bodies and consequently to decrease any future health threat. However, future exposures cannot be determined without analytical data for the surface water.
Indoor Air Pathway
Indoor air sampling has not been conducted since 1988. Therefore, it is difficult to assess current
or future exposure pathways. However, if contaminants in the groundwater plume continue to
migrate, it is possible that elevated contaminant levels could be found in structures north and
south of the site. Groundwater treatment during remediation is expected to decrease the
migration of contaminants from the site and to prevent the groundwater plume from reaching
structures south of the site. Without sampling data, it is not possible to assess current or future
This section will discuss health effects in persons exposed to specific contaminants; evaluate state and local health databases; and address specific community health concerns. To evaluate health effects, ATSDR has developed a minimal risk level (MRL) for contaminants commonly found at hazardous waste sites. The MRL is an estimate of daily human exposure to a contaminant below which non-cancer, adverse health effects are unlikely to occur. MRLs are developed for all routes of exposure, and for lengths of exposure, such as acute (fewer than 14 days), intermediate (15 to 364 days), and chronic (greater than 365). ATSDR presents those MRLs in the Toxicological Profiles, which are chemical-specific profiles that provide information on health effects, environmental transport, human exposure, and regulatory status. In the following discussion, ATSDR's Toxicological Profiles and other sources were used for arsenic, lead, aroclors, and 1,2,4-trichlorobenzene.
Arsenic exposure via ingestion has occurred and is occurring in children who live near the site and who access sediments in East Pond or River Meadow Brook. Exposure has also occurred in the past to children or adults who may have contacted surface soil at the southeastern corner of the site and east of the site.
The primary health outcome of arsenic ingestion is irritation of the gastrointestinal tract. Long-term exposure to increased arsenic levels via ingestion can result in anemia, neuropathy, skin lesions (hyperpigmentation, hyperkeratosis), and liver and kidney damage. Possible outcomes of chronic exposure include increased risk of skin and internal cancers [1,18,26].
It is estimated that young children ingest 200 mg of soil per day. ATSDR estimated that the average arsenic intake for soil containing an average of 5 ppm of arsenic would be 1 g; therefore, the MRL is 1 g/kg/day . Sediment and soil at the site contained a maximum of 600 ppm of arsenic. Although intake of sediment/soils with those concentrations could result in an ingestion of greater than 1 g of arsenic, the levels ingested are probably well below concentrations related to adverse health effects. Studies indicate that characteristic signs of arsenic toxicity develop in the most sensitive individuals at between 20 and 60 g/kg/day. Assuming the average child weighs 10 kg and the average adult weighs 70 kg, arsenic toxicity for the most sensitive individuals would develop between 200 and 600 g arsenic/day for children and 1400 to 4200 g arsenic/day for adults. Exposure to even the maximum concentration of arsenic in sediments/soils would not result in ingestion of greater than 200 g/day of arsenic. It is not expected that ingestion of sediments/soils around the site would result in arsenic toxicity. Dermal exposure to arsenic is believed to lead to dermatitis. However, ATSDR has no information on dose-response relationships pertaining to exposure to arsenic and development of dermatitis . Also, only small amounts of arsenic may enter the body via the skin, but ATSDR has no methodology to determine the level of absorption of chemicals through the skin . For that reason, it is difficult to determine other health effects of dermal exposure to arsenic.
Exposure to lead has occurred and could be occurring in workers at the Lowell Iron and Steel facility who have contacted surface-soil contamination. Primary exposure, however, is occurring in children and other residents who might be ingesting or contacting sediments in East Pond and River Meadow Brook and surface soil east of the site. ATSDR has no MRLs, and EPA has no RfDs, for lead, but exposure could cause adverse health effects. In infants and young children, lead exposure via ingestion can cause a decrease in intelligence (IQ) scores, slow growth, and hearing problems. The effects can persist as children get older and can interfere with successful performance in school . Such children could be classified as learning disabled. A blood lead level greater than 25 micrograms per deciliter (g/dL) in children is considered lead poisoning .
Lead exposure is especially dangerous for the fetus because of potential harm during development. Pregnant women exposed to lead can pass lead to unborn children, resulting in premature birth, low birth weight, and miscarriages . Soil ingestion is the primary exposure pathway related to lead-contaminated soils around the site. Therefore, it is not expected that pregnant women residing near the site would be at an increased risk, because of their minimal ingestion of soil.
In addition, exposure to high levels of lead can damage the brain and kidneys of adults and children . Lead exposure can increase blood pressure in men, but it is unknown whether that effect occurs in women. Lead exposure may affect sperm or damage other parts of the male reproductive system . However, adult residents are not expected to be at an increased risk of such effects because soil ingestion is the primary exposure route. It is not expected that adult men and women will ingest large quantities of contaminated soil (0 - 12 inches).
Young children who live or play near or at the site are most sensitive to lead exposure. It is not expected that lead levels in surface soil are adversely affecting adults or children via dermal absorption, because only small quantities of lead can pass through the skin. Biologic tests, such as measuring lead in blood, can determine whether excessive lead exposure is occurring as a result of ingesting contaminated sediment. Blood lead levels for children in the neighborhood do not exist. However, the Massachusetts Lead Poisoning Prevention Program was able to provide information on blood lead levels in children on a city-wide basis . Data for July 1, 1989, through June 30, 1990, indicate that there are 9,533 children in Lowell aged 6 months through 5 years. Of those children, 5,665 (59%) were screened for blood lead levels; 35 children were confirmed as having lead poisoning, indicating that their blood lead levels were greater than 25 g/dL. The incidence ratio for lead poisoning for children in Lowell is 0.62%; the statewide average is 0.36% . There was no information available on the source of lead exposure for, or the geographic distribution of, the children with elevated blood lead levels.
Exposure to PCBs has occurred and could be occurring due to contamination of soil (0 - 12 inches) on the Lowell Iron and Steel property, the Boston and Maine Railroad area, the northeastern corner of the site, and the southeastern corner of the fenceline. Some areas of soil contamination have been covered with gravel in order to mitigate current and future exposures. Children are most at risk of adverse health effects resulting from past exposures, primarily by ingestion of and contact with contaminated surface soils. Adults (workers at Lowell Iron and Steel and residents around the site) might have been exposed to low levels of PCBs caused by dermal absorption resulting from contact with contaminated surface soils.
Oral exposure to PCBs generally results from ingestion of contaminated fish. Few data are available about health effects related to ingestion of contaminated soils. ATSDR has set the chronic MRL at 1 g/kg/day. PCBs are known to primarily affect the liver and cutaneous tissues. Dermal exposure to PCBs could result in chloracne, hyperpigmen-tation, erythema, dryness, and skin rashes [1,25]. Chloracne is a severe, deformative skin lesion that can take years to clear and is usually associated with occupational exposures. Ingestion of contaminated soils could result in liver damage to children [1,25]. Generally, with the exception of two highly contaminated samples, soil concentrations of PCBs were below the EMEGs established by the ATSDR. No pertinent data exist to correlate dose response levels to actual observed health effects. Therefore, it is difficult to assess if children could have been adversely affected by levels found in soil within the study area.
ATSDR does not have a Toxicological Profile or MRLs for 1,2,4-TCB. EPA has established
RfDs of 0.2 and 0.02 mg/kg/day for subchronic and chronic ingestion exposure, respectively.
1,2,4-TCB was detected in one sediment sample from River Meadow Brook at a concentration of
0.17 mg/kg. Children could be exposed via ingestion and dermal contact; adults would be
exposed via dermal absorption only. Exposure is expected to be low because of the infrequency
with which the compound was detected. Dermal exposure could result in skin irritation and
rashes . However, until more data are available, it is not possible to determine whether
adverse health effects are likely to occur in children and other residents in the site area.
B. Health Outcome Data Evaluation
1. Health Studies
Three health studies were conducted before the current descriptive health study was initiated. The first study was conducted by Lowell Fair Share to try to determine if there were a significant number of health effects associated with Silresim . Investigators from Boston University, School of Public Health, conducted two epidemiologic investigations. The first descriptive study was to confirm or refute the results of the Lowell Fair Share study . The second was a cross-sectional approach and incorporated air-sampling data that detected DMF emitting from the Compo and Majilite facilities . The study was conducted to determine whether residents near the Silresim site and near the Compo and Majilite facilities were being affected by the DMF.
Health Survey, Lowell Fair Share
Lowell Fair Share and Ayer City Fair Share conducted a door-to-door survey in the neighborhoods around the Silresim site. They documented a variety of health problems, including skin rashes, headaches, respiratory infections, urinary/bladder infections, allergies, cancer, heart disease, miscarriages, stillbirths, and developmental effects. Of primary concern to Fair Share were reports of cancer, respiratory infection, birth defects, and miscarriages/stillbirths (12 out of 22 pregnancies). It is not known, however, how the interviews were administered. Improper data collection techniques could lead to biased results. Also, factors such as nutritional status, smoking, alcohol consumption, and occupational exposure were not evaluated in the survey. Those non-environmental factors could increase the risk of certain health outcomes.
Silresim Area Health Study, Boston University, School of Public Health
Between February and April, 1983, Boston University conducted telephone and in-person interviews with residents near the Silresim site (target group) and Lowell residents at least three fourths of a mile from the site (control group). Information was collected on approximately 2,000 adults and 1,100 children in the two areas. Death certificates from 1970 to 1981 were analyzed to see if there were unusual mortality patterns. The purpose of the study was to confirm or refute the data gathered by Lowell Fair Share. Health effects such as cancer, adverse reproductive outcomes, and self- reported symptoms such as headache, respiratory complaints, fatigue, colds, and bowel dysfunction were evaluated. Factors such as smoking, occupational exposure, age, and socioeconomic status were accounted for in the analysis.
Respiratory problems, fatigue, headaches, and bowel dysfunction were reported more often by the target population than by the control group. In children, there was a higher frequency of sore throats and bronchitis in the target group. The investigation did not reveal unusual patterns of mortality, increased cancer risk, birth defects, or adverse reproductive outcomes.
Boston University concluded that the observed effects were more numerous than what would be expected based on the levels of contaminants around the site. Therefore, they could not conclude whether the adverse health effects were the result of the site or other factors.
Health Investigation of Residents Living Near the Compo and Majilite Facilities, Boston University, School of Public Health:
This study, conducted by the Boston University School of Public Health, was an extension of the Silresim Area Health Study following the discovery of DMF emissions from the Compo and Majilite facilities. As with the Silresim study, the survey method involved interviewing 301 residents and using data from 948 residents from the Silresim control group. The objective was to assess whether the adverse health effects could be related to the two industrial facilities. A cross-sectional survey design was implemented to examine the relationship between emission exposure and self reporting of adverse health impacts. Acute and chronic conditions reflecting current and past exposures were examined.
This investigation found a statistically significant association between emissions (based on computer modeling) and gastrointestinal tract complaints, eye/nose irritation, wheezing/tightness in the chest, kidney problems, earaches (pre-school children), and never being pregnant or having a miscarriage. Respiratory complaints, fatigue, and dizziness were too confounded by seasonal differences to determine an association. Causality was determined based on a number of criteria; however, a conclusive relationship between emissions and adverse health effects could not be established.
2. Current Descriptive Health Study
As part of the Silresim health assessment prepared on behalf of ATSDR by the Massachusetts Demonstration Program, the MDPH Community Assessment Unit (CAU) investigated cancer incidence in the city of Lowell. Standardized Incidence Ratios (SIRs) were calculated for several cancers for the five-year period 1982-1986, both by census tract and for the city as a whole. SIRs were calculated for cancers of the bladder, liver, lung, and prostate, as well as for leukemia, Hodgkin's disease, non-Hodgkin's lymphoma (NHL), and soft tissue sarcoma (STS). Cancers were selected for investigation based upon documented community concerns and/or because of a potential relationship to the types of contaminants present at the site.
For this assessment, particular emphasis was placed on census tracts within a one-mile radius of
the Silresim Superfund site, including 3101, 3111-13, 3115-22, and 3124 (see accompanying
tables and figure in Appendix D).
Cancer incidence data were obtained from the Massachusetts Cancer Registry of the MDPH Bureau of Health Statistics, Research and Evaluation. In order to determine whether an elevated rate of cancer exists in Lowell, data were analyzed to compare the actual (or observed) number of cancer cases of each primary site to the number that would have been expected based on the state-wide cancer incidence experience [38,57].
An SIR is an estimate of the occurrence of disease in a population in relation to what might be expected if the population had the same cancer experience as some larger population designated as "normal" or average. That population is usually the state as a whole. Specifically, an SIR is the ratio of the observed number of cancer incident cases to the expected number of cases. An SIR equal to 100 indicates that the number of cancer cases in the population is equal to the number of cases expected in a "normal" population. An SIR greater than 100 indicates that more cancer cases occurred than expected; an SIR less than 100 means that fewer cases occurred than expected. Accordingly, an SIR of 150 is interpreted as an excess of 50% of cases over the expected number; an SIR of 90 indicates 10% fewer cases than expected.
Caution should be exercised, however, when interpreting an SIR. The interpretation of an SIR depends on both the size of the SIR and on how stable the SIR is. Two SIRs can be the same size, but not have the same stability. An SIR of 150 based on two expected cases and three observed cases indicates a 50% excess of cancer. The excess is attributed to only one excess case, and that case may have occurred by chance alone. Conversely, an SIR of 150 based on 200 expected cases and 300 observed cases shows the same 50% excess in cancer, but, because the SIR is based on a greater number of cases, the estimate is more stable. It is unlikely that 100 excess cases would occur by chance alone.
In order to determine if the observed number of cases is significantly different from the expected number, or if the difference may be due to chance, a 95% confidence interval is calculated. A 95% confidence interval is the range of estimated SIR values so that the range has a 95% probability of including the true SIR for the population. If the range does not include the value 100, that means the study population is significantly different from the "normal" population. If the range excludes 100, and the observed SIR is greater than 100, there is a significant excess in the number of cancers. Similarly, if an observed SIR is less than 100, and the range excludes 100, the number of cancer cases is significantly lower than expected. If the range includes 100, that means the true SIR may be 100, and it cannot be concluded, with sufficient confidence, that the observed SIR reflects a true cancer excess (or deficit). Statistical significance is not assessed if fewer than five cancer cases are observed.
Place of residence at the time of diagnosis was examined for each case (by primary site) to assess any possible geographic clustering of cases. The available occupational information and smoking status of individuals diagnosed with specific cancer types were also evaluated. That evaluation was accomplished by using existing cancer registry data and census data.
Accurate population information is critical when calculating incidence rates. The population figures for the state of Massachusetts and the city of Lowell were obtained by calculating the percentage increase or decrease according to the 1980 Federal Census data and the 1990 MARS file, which is an adjusted listing of the 1990 Federal Census data. The percentage of change is than applied to the intercensal years through straight line interpolation. For the purpose of consistency, and because of the gap between census years, the assumption is made that the change in population occurs at a constant rate throughout the ten year period. From these calculations the 1989 midyear estimates are obtained.
Overall, for men and women combined, cancer incidence in the city of Lowell was about what would be expected based on Massachusetts' state-wide cancer incidence rates. Non-Hodgkin's lymphoma incidence occurred significantly less often than expected among women (26 cases observed/51.7 cases expected, SIR = 50; confidence interval [CI] = 33-74). During a review of rare cancer types, one case of ependymoma was observed in Lowell during the period 1982-1989.
Census Tract Analysis
Analysis of census tracts 3101, 3111-13, 3115-22, and 3124 combined revealed that, for the most part, cancer cases have occurred essentially as expected among men and women. Men in the combined census tracts, however, experienced a significant elevation in lung cancer (152 cases observed/124.2 cases expected, SIR = 122; CI = 104-143). Total combined male and female lung cancer incidence was also significantly elevated during the same eight year period (231 cases observed/192.1 cases expected, SIR = 120: CI = 105-137).
In most of the census tracts, lung cancers occurred at a rate that is about what would be expected. However, statistically significant elevations were observed among men and women combined in census tracts 3101 (33 observed/22.2 expected, SIR = 149; CI = 102-209), 3111 (14 observed/7.6 expected, SIR = 183; CI = 100 - 308), and 3124 (22 observed/11.6 expected, SIR = 190; CI = 119-288). Statistically significant elevations were also observed among men in census tracts 3111 (11 observed/4.9 expected, SIR = 227; CI = 113-406), and 3124 (14 observed/6.3 expected, SIR=222; CI=121-373). Those data are summarized in Table D-9.
Non-Hodgkin's lymphomas occurred significantly less often than expected among women in the combined census tracts (9 observed/20.5 expected, SIR = 44; CI = 20-83). Those data are summarized in Table D-2. Prostate cancer incidence also occurred significantly less often than expected in all census tracts combined (100 cases observed/123.4 expected, SIR = 81; CI = 66-99).
Analysis of individual census tracts revealed that most cancers -- Hodgkin's disease, non-Hodgkin's lymphoma, leukemia, soft tissue sarcoma, and bladder cancer -- are occurring at approximately the rate that would be expected. The results for census tracts by cancer type are presented in Tables D-3 to D-4, D-7, and D-8.
In general, liver cancer occurred as would be expected, in census tracts 3111 - 3116, 3118, 3119, 3121, and 3124. Non-significant elevations were noted in census tracts 3101, 3117, 3120, and 3122. Although these SIR's are large, they are all based on a fewer than two cases each. Those data are summarized in Table D-6.
A statistically significant elevation in leukemia was observed among women in census tract 3122 based on data analyzed for the period 1982-1989. Updated information on cancer incidence (1982-1990) which became available before the release of this report was checked for additional leukemia incidence in this census tract. The data show no change in the incidence of leukemia among women (6 cases observed/0.8 cases expected, SIR = 691; CI = 252-1504 for 1982-1990) in this census tract. The data is summarized in Table D-5.
Place of residence at the time of diagnosis was examined for individual cases of Hodgkin's disease, NHL, leukemia, STS, liver, bladder, and lung cancers in the census tracts within a one-mile radius of the Silresim site. Analysis of the geographic distribution of cancer incidence revealed that cancer cases were fairly evenly distributed within each census tract, with the exception of lung cancer and leukemia. The areas where lung cancer cases appear to be clustered include census tracts 3111, 3119, and 3120. Although it may appear that lung cancer cases are also unusually clustered in census tracts 3101 and 3124, those areas represent elderly housing complexes in which the population is dense and residents are unlikely to have resided there during the latency (developmental period) for this type of cancer.
No clusters were noted in census tract 3122, which contains the Silresim Chemical Corporation. Of the seventeen female cases of leukemia observed in the combined census tracts, six of them (35%) occurred in census tract 3122. There was no geographic clustering of cancer cases in the immediate vicinity of the site (see Figure D-1).
Because of statistically significant elevations observed in lung cancer incidence, the smoking status of cases was evaluated for individual cases in census tracts 3101, 3124, and in the 13 census tracts combined. The majority of persons with cancer were as current or former smokers.
Of a total of 33 persons with lung cancer in census tract 3101, smoking status was available for 30 cases (91%); 25 cases (76%) reported being a current or former smoker at the time of their diagnosis, and 5 cases (15%) reported that they had never smoked.
In census tract 3124, of 22 persons with lung cancer, smoking status was available for 21 cases (96%); 19 cases (86%) reported being a current or former smoker, and 1 case (5%) reported that she had never smoked.
For the 13 census tracts combined, smoking status was available for 215 (93%) of 231 persons with lung cancer; 192 cases (83%) reported being a current or former smoker, and 23 cases (10%) reported that they had never smoked.
Because of statistically significant elevations observed in lung cancer incidence, available occupational information for individual cases in the combined census tracts was reviewed. Specific occupational titles as reported to the cancer registry are noted in order to identify individuals whose occupation may have placed them at an increased risk for development of lung cancer.
Among persons with lung cancer in the combined census tracts who reported an occupational status, eight individuals reported occupational exposure to asbestos, a known causative agent in the development of the disease. Sixteen cases (7%) reported other occupations associated with an increased risk of lung cancer. Seventy-four cases (32%) were either retired or their occupations were unknown at the time of diagnosis. Of the remaining reported occupations, information was either incomplete and could not be evaluated, or there was no known association with lung cancer.
Occupational exposure to ionizing radiation, benzene, and other chemicals are known or suspected risk factors for the development of leukemia. Genetic factors, exposure to anti-cancer drugs and (more recently) smoking have also been cited as risk factors for the development of this disease. Of the thirty-three individuals diagnosed with leukemia, 3 cases (10%) reported an occupation where exposure to chemicals is likely to have occurred. None listed an occupation in which exposure to ionizing radiation is known to occur.
Cancers, in general, have a variety of associated risk factors which are assumed to be related to the etiology (development) of the disease. Many cancers are believed to be related largely to life-style factors such as cigarette smoking, diet, and alcohol consumption. Epidemiologic studies of humans and laboratory animals have related several cancers to adverse chemical exposures in the workplace or in an individual's environment. Other factors associated with cancer are socioeconomic status, race, heredity/genetics, and geography. Review of scientific and medical literature provides known or assumed risk factors for the cancers of interest.
With the exception of lung cancer, the city of Lowell does not appear to have excessive rates of the cancer evaluated. SIR's reveal that the incidence of cancer in census tracts 3101, 3111-13, 3115-22, and 3124 was about what would be expected when compared with the statewide cancer incidence experience. However, several statistically significant elevations were observed in lung cancer incidence: 1) among men and the total population in the combined census tracts; 2) among the total population in census tract 3101; and 3) among men and the total population in census tract 3111 and 3124 (see Tables D-1 and D-9). None of these elevations occurred in areas near the Silresim site.
Evaluation of the smoking status of each lung cancer case indicates that approximately 83% of cases in the combined census tracts reported being a current or former smoker. In the individual census tracts that showed significant elevations in lung cancer, the majority of persons with cancer also reported being current or former smokers. It is also worthy to note that cases were not concentrated in any one area of the site with the exception of those in the elderly housing complex. As previously mentioned this information suggests that the environment was not likely to be related to the development of these cancers.
Lung cancer represents approximately 15% of all cancers nationwide (22% among men and 8% among women). Cigarette smoking is known to be the principal causative factor in the development of lung cancer . The federal Office of Smoking and Health estimates that, in the United States, cigarette smoking accounts for 95% of lung cancer deaths in men and approximately 80% of lung cancer deaths in women. The risk of developing lung cancer depends upon the intensity of one's smoking habits (i.e., duration of habit, "tar" yield of cigarette, and amount smoked). Fortunately, the risk of lung cancer declines after smoking cessation. However, investigators estimate that it can take more than 10 years of not smoking for long-term, heavy smokers to reduce their risk to a level similar to that of someone who has never smoked. There is increasing evidence that passive smoking (exposure to secondhand smoke) may increase an individual's risk of developing lung cancer. The degree of that risk has not yet been fully established [3,43].
Lung cancer trends reveal that lung cancer has become a disease increasingly associated with populations of lower socioeconomic status, because, according to current research, those individuals may smoke more. Epidemiologic studies indicate that several occupations are associated with an increased risk of developing lung cancer, particularly those involving exposure to asbestos (i.e., mill workers, miners, textile and shipyard workers, etc.). Other known occupational exposures include radon and radon daughters (uranium mining), polycyclic hydrocarbons (coke ovens, roofing, smelting), chromium (plating, spray painting), nickel (smelting, roasting, electrolysis), inorganic arsenic (sheep dip manufacturing, copper smelting, production and use of pesticides/herbicides), chloromethyl ethers (chemical industry), and ionizing radiation. It is important to note that a combination of just one of those occupational exposures and regular cigarette smoking dramatically increases the risk of developing lung cancer [4,43,57].
Review of available occupational information for lung cancer cases in the combined census tracts revealed that 32% of individuals were retired, or their occupations were unknown at the time of diagnosis. Approximately 7% (n=16) of individuals with lung cancer who reported an occupational status in the combined census tracts reported an occupation that has been associated with the disease.
Several industrial exposures have occurred, including, but not limited to, vinyl chloride and beryllium, which are suspected to play a role in the development of lung cancer, although epidemiologic evidence is preliminary. An increased risk is also suspected among rubber workers. Urban air pollutants have long been suspected in the etiology of lung cancer; however, that association has not been substantiated .
Liver cancer is a general term representing a group of malignancies classified by separate histologic types (cell types). Epidemiologic studies have revealed that each histologic type of liver cancer is an individual disease with specific characteristics, patterns of survival, and etiologic factors [4,57].
Of the six liver cancer cases observed in the census tracts examined in Lowell, four different histologic types were identified. The majority of individuals were diagnosed with hepatocellular carcinoma, the most common type of liver cancer. The major etiologic factors associated with liver cancer include chronic hepatitis B virus, aflatoxin, and cirrhosis. Other suspected risk factors include a number of hereditary metabolic diseases and congenital disorders [4,20,56]. Occupational exposure to vinyl chloride has been identified as a causative agent in angiosarcoma, a rare type of liver cancer, but no cases of angiosarcoma were reported in Lowell during the period 1982-1989.
This investigation is descriptive and can only provide an evaluation of certain types of cancer incidence in the city of Lowell compared to cancer incidence in the state. Descriptive assessments have inherent limitations. Only routinely collected data are analyzed, and information regarding personal risk factors (i.e., occupation, diet, smoking, etc.) that may influence disease incidence is often limited and not of an historical nature.
The major finding in this investigation is the elevation in lung cancer incidence noted in the neighboring census tracts 3101, 3111 and 3124, and among men in all census tracts combined.
Since cigarette smoking is the major risk factor for lung cancer, and a large majority of cases observed in Lowell were persons who were current or former smokers at the time of diagnosis, it is likely that cigarette smoking played a large role in the development of these cancers. Lung cancer cases appear more pronounced in census tracts 3111, 3124, and along the border of census tracts 3120 and 3121. Although there may also appear to be a cluster of lung cancer cases in census tract 3101, those individuals reside in densely populated, multi-unit residences, many of which cater almost exclusively to an elderly population that have not likely lived there long enough to have been exposed to a common environmental agent. The latency period for lung cancer is approximately 20-40 years; therefore, it is likely that individuals may have developed cancers from past exposures, not from exposures at their current addresses. It is important to note that there was no geographic clustering of cases in the vicinity of the site.
The lack of definitive occupational information did not allow for determining the role occupational exposures might have played in the development of lung cancer in Lowell. Data have been presented, however, which suggest that occupation probably played a role in some incidents. In addition, preliminary 1990 census data suggest a 76% increase in population since 1980 within census tract 3101 . Because of this apparent increase in population, those results may not be indicative of a true elevation in lung cancer.
Leukemia incidence among females in census tract 3122 was significantly elevated for the period 1982 - 1989. A geographic analysis of the census tract, however, revealed no evidence of spatial clustering in relation to the Silresim site. In addition, a closer evaluation indicated that the 6 cases of leukemia in this census tract represented five different histological types of leukemia. These factors make it less likely that the cases are the result of a single risk factor, such as a common environmental exposure.
The one case of ependymoma developed in an individual who lived more than two miles from the site, and therefore, is probably not attributable to any site related contaminants.
It is beyond the scope of these analyses to determine the causal and/or synergistic roles that occupation, smoking status, and/or environmental exposures may have played in the development of cancers in Lowell. The findings in this health assessment can only be reported and discussed in the context of the available information. The Community Assessment Unit will continue to monitor cancer incidence in the city of Lowell through the Massachusetts Cancer Registry.
- We have addressed each of the community concerns about health as follows:
- Are children who play in East Pond and River Meadow Brook being exposed to
Children who play in East Pond are not being exposed to levels of surface-water contamination that pose a risk to their health. Therefore, exposure via inhalation, ingestion, and absorption of surface water will not pose a risk to human health.
However, the sediment in East Pond contains elevated levels of lead that pose a risk to children if they ingest the contaminated sediment. In children, lead exposure via ingestion can cause decreased IQ scores, slow growth, and hearing problems. Exposure to high lead levels can also result in brain and kidney damage in both adults and children. Children who may be exposed to lead from other sources (i.e., lead paint in houses) are at an increased risk of adverse health effects if they are exposed to lead-contaminated sediment.
As stated earlier, ingestion of lead and subsequent adverse health outcomes are possible due to the contaminated sediment in East Pond. However, young children (<6 years old) are the individuals who are most likely to ingest large quantities of soil or sediment. Based on the location of the pond and the debris present in and around the pond, it is unlikely that young children would have frequent access to the pond. They probably would no ingest the quantity of sediment necessary to cause adverse health outcomes.
- Are children who walk through contaminated soils off-site being exposed?
The off-site contamination, except around the Boston and Maine Railroad, is currently covered with gravel or is inaccessible because of fencing. However, area residents could have been exposed in the past. Because of the nature of the contaminants in surface soil, the primary exposure route would be ingestion rather than inhalation or absorption. However, using available data, it is not expected that ingestion of the contaminated soils would result in adverse health effects.
- Could contamination from the site have been spread by wind dispersion of contaminated
soils? Could particulates from the site have accumulated in the soil over time and
currently pose a threat to human health?
Currently, the major areas of surface soil contaminants are covered by the clay cap or gravel. Although some contamination could have been spread via wind dispersal, it is expected to be a minor source, with little accumulation of contaminants because of particulates. Therefore, such contamination is not expected to represent a threat to human health.
- Are employees at Lowell Iron and Steel at risk from contamination associated with the
Surface soil at the Lowell Iron and Steel property is contaminated with compounds such as PCBs, dioxins, and lead. Those compounds are not expected to be absorbed or ingested in large quantities by workers at the site. Inhalation of contaminated particulate matter could occur, but that pathway is not expected to pose a health risk because gravel covers the most contaminated areas of surface soil.
Indoor air sampling in the basement of the administrative building on the Lowell Iron and Steel property detected low levels of VOCs in the air. The levels detected do not pose a health risk to workers. However, it is important to note that the sampling was done on a single day, and concentrations could vary daily because of weather and atmospheric conditions. Also, concentrations are expected to be higher during times of flooding because more potentially contaminated groundwater could be entering the sump pit in the basement.
Future levels of contaminants in the basement could increase if highly contaminated areas of the groundwater plume reach the Lowell Iron and Steel buildings. It is impossible to predict actual levels that might be detected in the indoor air in the future. However, Lowell Iron and Steel has removed workers from the basement so exposure would be on an intermittent basis. ATSDR recommends future testing of indoor air if EPA determines that increased levels of contaminants are reaching the administrative building.
- There appears to be a lot of cancer in the area. Is it related to the site?
The Department of Public Health reviewed cancer incidence data from the Massachusetts Cancer Registry. Incidence rates were calculated for cancers of the bladder, liver, lung, and prostate, as well as for soft tissue sarcoma, leukemia, Hodgkin's disease, and non-Hodgkin's lymphoma. The incidence for most of the cancers was about what would be expected using the state-wide averages. Significant elevations in lung cancer were observed in certain areas. However, a review of occupational and smoking status information indicates that lifestyle factors may have played a role in the lung cancer cases. It is also important to note that the elevations were observed at a distance from the site; therefore, it is unlikely that exposures resulting from residential proximity could be largely responsible. There was an apparent elevation in liver cancer, but it cannot be assigned statistical significance because of the small number of cases.
It is difficult to correlate exposure to site contaminants with cancer incidence because of the many confounding factors: (1) exposure to chemicals from other industries around the site; (2) possible occupational exposure to chemicals; (3) life-style factors such as diet, alcohol consumption, and smoking status; (4) genetics; and (5) residential history. Also, most cancers have a long latency period, which makes a correlation between exposure and development of cancer difficult at best. MDPH will continue to monitor cancer incidence in Lowell through the Massachusetts Cancer Registry.
- Is it possible for contaminants to migrate into the basements of residences around the
The groundwater plume has not yet reached the residences to the south of the site. Although sewer lines around the site will slow down migration of the groundwater plume, the lines will not completely eliminate the movement. It is possible that contaminants could eventually reach the residences via the groundwater plume and volatilize into the basements. In the future, that exposure pathway will be eliminated by remediation of the groundwater by EPA.