Printer Friendly

The New Jersey residential well-testing program--a case study: Randolph Township.


More than 50 percent of the U.S. population relies on groundwater as a source of drinking water, and a greater proportion of residents rely on potable or treatable groundwater in rural areas. In total, 81 percent of community water systems depend on a groundwater supply.

Many localities experience significant contamination to potable water sources. The most common sources of groundwater pollution are leaking underground storage tanks, septic tanks, municipal landfills, and farming practices such as pesticide use in agricultural production (U.S. Environmental Protection Agency [U.S. EPA], 1992). The most common contaminants associated with these sources include coliform bacteria, nitrates, metals, volatile organic compounds (VOCs), and pesticides. While municipal water supplies are regulated and routinely monitored for these contaminants by state and federal agencies, no similar protective regulations exist for private well users. It is the responsibility of the individual well owner to ensure the safety of his or her water supply.

The focus of the Randolph Township Private Well Survey was to provide a sample of residential well water to serve as the basis for implementation of a local policy protecting private well water supplies from contamination. The Randolph Township Health Department sampled 50 private and nonpublic wells. The data generated by the study will be used to develop a model ordinance on the testing of private and nonpublic wells for volatile organic compounds and other inorganic contaminants. The well study could also prove valuable in the development of the township's master plan, and may have considerable impact on the future of Randolph Township's water and sewer utilities.



The Randolph Township Well Survey was conducted by the Randolph Township Health Department, with involvement from the Planning Department, the Engineering Department, the Mayor and Council, and the Environmental Commission. A public-private partnership was established with Garden State Laboratories, a private New Jersey state-certified laboratory, to provide a cost-reduced scan of 50 water samples. All sample collection and analyses were conducted by the health department. The Environmental and Occupational Health Sciences Institute, at the University of Medicine and Dentistry of New Jersey, provided quality assurance and quality control support.

Project Procedure

Fifty private and nonpublic wells were randomly selected from a listing of 1,488 wells. Questionnaires were mailed to the homeowners at these 50 sites, and samples were collected and evaluated for 64 VOCs, nitrate, coliform bacteria, and other inorganic water quality criteria.

Sample Plan

Before a plan can be written, site-specific information must be gathered to ensure that the plan is logical, that it will meet the required objectives, and that the course of action is feasible (New Jersey Department of Environmental Protection [NJDEP], 1992). The first step in a site investigation should be the gathering of background information. Sources of information utilized in the development of a sampling plan for the Randolph Township Well Survey included records from the departments of health and planning and the Tax Assessor's Office. The Township Well Database and a map depicting the location of all private, nonpublic, public, and public noncommunity wells provided an inventory of 1,544 wells throughout Randolph Township, including the town's 1,488 private wells. The database inventoried street address; block and lot designation; type of well; method of sealing; and well depth, diameter, length, and yield.

Six factors were used to determine the number of samples required for site characterization:

1. exposure pathways,

2. statistical performance objectives,

3. quality assurance objectives,

4. background samples,

5. sampling objectives, and

6. site specific conditions (NJDEP, 1992).

A stratified random sample was also used. Because the intent of the study was to generate data on water quality throughout Randolph Township without regard to prior contamination or areas potentially affected by topographic features and so forth, biased sampling methods were not used. Stratified random sampling provides a more precise means of estimating contamination than use of simple random sampling without stratification of the area. The Randolph Township Well Database stratifies its 1,488 private wells with respect to geographic location by block and lot designation. A total of 50 samples were collected for cost consideration, statistical significance, and presumed ease of utilization. This number reflected the most cost-efficient method of gaining an adequate sampling of potable-water quality.

To achieve randomization, the authors chose every 30th private well site from the Well Database (i.e., 30, 60, 90, etc.). Once an initial sample of 50 locations was selected, these sites were plotted on the Township Well Map to ensure an even geographic representation of all Randolph Township's private well areas, as intended with the block and lot stratification. Adequate geographic randomization was confirmed through mapping of the chosen locations.

Because all sites to be sampled were private properties, accessibility was dependent on resident cooperation and access to sampling sources. Fifty samples were obtained from 179 locations contacted by mail, representing a return rate of 28 percent.

Sample Collection and Quality Assurance

The Randolph Township Private Well Survey used the New Jersey Department of Environmental Protection's Field Sampling Procedure Manual for standard operating procedures in collecting and handling water samples.

Most water samples were collected from the kitchen faucet, and all samples were drawn from the cold-water faucet. Systems were flushed for 15 minutes. The aerator and screen were removed, and the faucet was wiped down with an isopropyl alcohol swab for sterilization. Water samples were collected over a period of five months.

Upon completion of laboratory analysis, results were mailed to the health department. All sample results were immediately entered into a results spreadsheet, including all well-related information for each address sampled as provided by the laboratory, residents, and health department staff. When sample results exceeded New Jersey maximum contaminant limits, residents were notified immediately and follow-up samples were obtained for analysis. A need for both notification of residents and follow-up sampling occurred only with respect to total coliform bacteria. Of the seven coliform-positive samples, only two retested positive. One resident chlorinated his well, which was sampled a third time for total coliform and tested negative. To date, residents at the other twice-tested coliform-positive locations have yet to contact the health department for a third sampling, despite being advised to treat their well for presence of the bacteria.

Questionnaire and Accompanying Materials

In an effort both to solicit resident participation in the Randolph Township Well Survey and to supplement forthcoming laboratory drinking-water data with other geographical, structural, and aesthetic-quality characteristics, a questionnaire was mailed to all prospective sample locations.

Homeowners were asked to provide information about their well (i.e., depth, casing, location, driller, usage of water treatment, pump capacity date of drilling, and sketch), as well as data about existing or abandoned septic systems, and fuel tanks. In addition, residents were asked various questions about lawn/agricultural treatment, the aesthetic quality of drinking water, and the condition of plumbing, well, delivery system, and kitchen/bathroom fixtures (Table 1).

An introductory letter and educational water quality materials were sent to residents along with the questionnaire. Among the materials received by prospective well survey participants was a brochure produced by Randolph Township, Protecting Your Drinking Water. This informative brochure provides basic well water information to homeowners and addresses some common problems and solutions regarding water quality. Residents were also sent brochures about proper fertilizer and pesticide maintenance, and a health department publication entitled You and Your Septic Tank! A Homeowners Manual for Septic Systems.


The Randolph Township Well Survey revealed the presence of several volatile organic compounds (VOCs) in residential water supplies, although none of the detectable VOCs occurred at above the established maximum contaminant levels. The VOCs found in this survey were chloroform, methyl-tertiary butyl ether, trichloroethylene, isopropyl ether, dibromochloromethane, 1,1,1-trichloroethane, and cis-1,2 dichloroethylene. The presence of chloroform and cis-1,2 dichloroethylene may be due to the degradation of the other above-mentioned compounds. It is possible, however, that the source of these organics may be other solvents or compounds. Methyl-tertiary butyl ether (MTBE) was detected at five sample locations, making it-along with chloroform-the most prevalent of the VOCs found in the survey. MTBE is used as a gasoline additive and can be found at service stations and other facilities that store gasoline for company/vehicle use. With its three major traffic corridors of Route 10, Route 513, and Sussex Turnpike, Randolph Township has many gasoline service stations. Because of its central location in northern New Jersey, the township is also home to several facilities that use individual transportation fleets (i.e., trucks, buses). Therefore, many sources may be contributing MTBE to the groundwater in Randolph. In addition to service stations and private motor vehicle fleets, sources of MTBE in the groundwater include gasoline spills, pipe ruptures, and miscellaneous discharges.

Thirteen sample sites contained detectable VOCs in their well water. A comparison of the locations revealed that six of the 13 sample sites (46 percent) were located less than one quarter of a mile from stationary sources of VOCs. When these sites were mapped and compared with other geographic features, it was clear that some of the volatile-organic findings are clustered around the Route 10 corridor, the hub of Randolph's commercial and industrial development (e.g., service stations, dry cleaners). These findings are consistent with those of previous water quality studies.

Total coliform bacteria were present in seven of the 50 samples. The coliform-positive samples were not accompanied by unusually high standard plate counts, which serve as an additional indicator for the presence of bacteria and water contaminants. A look at the raw data does reveal one trend regarding the presence of total coliform bacteria in Randolph's private wells. Of the seven coliform-positive samples, four were drawn from wells shallower than 75 feet, and two were found in wells deeper than 75 feel. For one sample site with total coliform bacteria, well depth data were missing. Although there are few samples to base this data item on, it is of interest that 67 percent of the coliform-positive wells were of shallow depth.

Of the nine private wells that revealed nitrate levels above 4 mg/L, four were shallower than 75 feet in depth; for the five remaining wells, well depth data were missing. Of the six additional wells with depths less than 75 feet, three had nitrate levels above 3 mg/L, the level recognized in literature as being affected by anthropogenic activity. In total, of the 10 wells known to have a depth of 75 feet or less, seven (70 percent) contained moderately high levels of nitrate. Forty percent of all measured wells contained nitrate levels of 3.0 mg/L or greater, regardless of well depth, so it appears that moderately high levels of nitrate are consistent throughout Randolph Township (Figure 1).

A similar relationship was seen among the total coliform-positive sample locations and the locations known to be served by individual subsurface sewage disposal (septic) systems. Six of the seven locations positive for total coliform bacteria are served by septic systems rather than public sewer, although the number of coliform-negative locations has not been further evaluated in the raw data for the presence of septic systems.

One additional finding in the raw data was the presence of more than one contaminant at many sample locations. Multiple contaminants were present at 19 of the 50 sites (38 percent), with a variety of combinations observed. Most notable may be a high level of water hardness along with many of the other measured water criteria. Hard water was found combined with high or moderately high levels of other measured contaminants (as defined in the study) in seven different wells. Only one well was found to exceed criteria levels of all four measured contaminants.

Because the questionnaire data on the presence of water treatment devices at individual sample sites were incomplete, only 37 samples were used for independent t-test results. In summary, nitrate was the only water contaminant found to be inversely associated with water treatment. Homes without water treatment devices were found to have mean nitrate levels of 3.0 (the number cited by sources as being due to anthropogenic activity). Homes with water treatment devices were found to have a mean nitrate level of 1.875.

The mean standard plate count (bacterial indicator) was higher in homes without water treatment, at 409; homes with water treatment had a much lower mean standard plate count of 68 (p = .36). A few outlier samples, however, may have contributed to this difference in mean. Locations without water treatment also had higher mean total-coliform levels and higher water hardness, although these differences in mean were negligible, especially when the lack of significance is considered. Mean VOC levels were actually higher at locations with water treatment devices, but again, the result was without statistical significance. It is important to remember that the water treatment variable suffers several inadequacies: presence/nonpresence of water treatment was not reported by 13 of 50 respondents; it is an inexact variable that does not specify the type of treatment; and some respondents may not have known what, by definition, a water treatment device is. Regardless of the shortcomings of this variable, it is worth-while to examine the potential relationships between the use of water treatment--albeit vague in definition and response--and the presence of measured contaminants.


The small number of cases (samples) used for many of the statistical analyses presented a recurring problem. Despite a sample of 50 private wells, some of the cited associations between variables and contaminant presence were based on as few as 26 samples. The sample number was limited by financial and logistical considerations, and more samples are clearly needed to achieve greater confidence in statistical calculations.

One other problem of the study was the poor response rate of residents. Although homeowners were offered free testing and were contacted with attractive, informative packets via mail, a return rate of only 28 percent was achieved. This rate is consistent with the 26 percent of homeowners who responded to the Middle Township well study, but such a low rate is not encouraging (Cape May County Department of Health [CMCDOH], 1996). Several residents were concerned about personal liability in the event of contamination, and this concern was a deterrent to their participation in the study. Residents were assured that the results were confidential, and that they would not be penalized in any way if well samples were found to have high levels of measured contaminants. They were further advised that the health department would help with any remediation or treatment options they wished to explore.

Potential biases were inherent in the study design. The questionnaire was subject to recall bias, as residents may not clearly remember aesthetic water quality criteria. Selection bias is also a possibility, since residents with a pre-existing concern over the quality of their water may have been more likely to participate in such a study, In addition, residents who live near possible pollutant sources such as farms, industrial areas, and so forth may have been more apt to participate in the study than residents without similar concerns. Another possible limitation of the study is that all aesthetic water quality criteria were based on self-reporting via questionnaire, lending to potential information bias. There was no way to validate the respondents' reporting of water quality problems, particularly because such observations were not time-specific and may have occurred years ago.


Randolph Township is comprised of 7,240 individual residences, approximately 20 percent of which draw their water from private wells. Because such a substantial number of homes receive their water from private, unregulated sources, this study was necessary to verify the presence of common water contaminants and other water quality criteria. The New Jersey Department of Environmental Protection has placed an even greater emphasis on drinking-water quality by proceeding with a well cluster study program, presumably in partial recognition of a lack in the monitoring of well water sources. In further recognition of this lack of data, the New Jersey Department of Health and Senior Services recently implemented the Private Well Testing Act, making the testing of private wells requisite for all real estate transfers. As groundwater resources are placed under the microscope, it is more important than ever to assess the quality of private drinking-water supplies.


The results of the study reported here are consistent with those of earlier studies in which the presence of drinking-water contaminants in private water supplies was found to be associated with other factors such as well construction and distance to potential sources (Briggins, 1995; Bush et al., 1995; Mayberry, 1996; Richardson, 1988; Seigley, Hallberg, Walther, & Miller, 1989; Tucker & Burke, 1978). Because of the small sample size, the results must be interpreted with caution. Despite its potential drawbacks, the study does give reason to test private municipal water sources further for the contaminants previously assessed, as well as for other pollutants with implications for public health. The findings--particularly the number of homes found to contain unsuitable or unacceptable levels of nitrate, VOCs, and total coliform--are sufficient not only to warrant further testing of private water supplies, but also to warrant taking regulatory and public-utilities planning measures for their protection.

Acknowledgements: This document was prepared with the aid of a grant from the New Jersey Department of Environmental Protection, Office of Intergovernmental Affairs, and the Environmental Services Program. It was supported by NIEHS Grants ES07148 and ES05022 and the Environmental and Occupational Health Sciences Institute.

Corresponding Author: Mark G. Robson, Associate Professor and Chairman, Assistant Dean of Student Affairs, UMDNJ--School of Public Health, Department of Environmental and Occupational Health, 683 Hoes Lane West, Piscataway, NJ 08854. E-mail:


Briggins, D.R. (1995). Pesticides, nitrate-N and bacteria in farm wells of Kings County, Nova Scotia. Water Quality Research Journal of Canada, 30(3), 429-442.

Bush, P.B., Berisford, Y.C., Hitchcock, R.N., Perkins, R.G., Segars, W.I. & Tyson, A.W (1995). Water quality in Georgia: 1994 summary of well water testing in Georgia. Athens, GA: University of Georgia, College of Agricultural and Environmental Sciences.

Cape May County Department of Health. (1996). Cape May County wellhead protection program: Middle township domestic well cluster report. Cape May, NJ: Author.

Mayberry, L. (1996). Nitrate survey report: Breakwater Estates, Erma, New Jersey. Cape May, NJ: Cape May County Department of Health.

New Jersey Department of Environmental Protection. (1992). Field sampling procedure manual. Trenton, NJ: Author.

Richardson, W.D. (1988). Groundwater contamination in Roxbury Township--An analysis of the risks and potential sources. Succasunna, NJ: Roxbury Township Department of Health.

Seigley, L.S., Hallberg, G.R., Walther, P.R., & Miller, G.A. (1989). Well water quality data from a volunteer sampling program: Audubon County, Iowa. Journal of the Iowa Academy of Science, 100(1), 15-20.

Tucker, R.K., & Burke, T.A. (1978). A preliminary report on the findings of the state groundwater monitoring project. Trenton, NJ: New Jersey Department of Environmental Protection.

U.S. Environmental Protection Agency. (1992). The quality of our nation's water: 1992 (EPA Publication No. 264/9-92/3003). Washington, DC: U.S. Environmental Protection Agency, Office of Water.

Peter N. Tabbot, M.P.H.

Mark G. Robson, Ph.D., M.P.H.
TABLE 1 Summary of Raw Data for Selected Variables

Variable (Outcome Criteria) Percentage of Wells

Presence of water treatment 43
Use of pesticides at residence 21
Use of herbicides at residence 30
Use of insecticides at residence 23
Use of fertilizer at residence 65
Failure of, or repair to, septic system 26
Presence of calcium deposits in sinks/tubs 19
Difficulty in producing soap suds/lather 12
Presence of egg odor in water 12
Presence of rust stains in plumbing fixtures 47
Metallic taste in water 19
Black stains to silverware 0
Black particles in water 16
Corrosion of metal parts on well pump, piping, 16
Presence of turbidity in water 19
Presence of gas/oil odor in water 0
Presence of chemical odor in water 5
Stomach problems in family or regular guests 2
Total coliform (presence of detectable level) 14
Nitrate (level >4.0 mg/L) 18
Nitrate (level >3.0 mg/L) 40
Volatile organics (presence of level >0.5 ppb) 26
Hardness ([greater than or equal to] 120 mg/L) 28
Depth of well (75' or shallower) 32
Distance to by-way (<1/4 mile) 31
Distance to fixed source of VOCs (<1/4 mile) 23
COPYRIGHT 2006 National Environmental Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Tabbot, Peter N.; Robson, Mark G.
Publication:Journal of Environmental Health
Geographic Code:1U2NJ
Date:Sep 1, 2006
Previous Article:The public health response to disasters in the 21st century: reflections on Hurricane Katrina.
Next Article:Changes in immunological and hematological parameters of female residents exposed to volatile organic compounds in the city of Kaohsiung, Taiwan.

Related Articles
Sammis Co. to build 500 homes.
Seeking new opportunities in tri-state area and beyond.
Industrial facility coming to Somerset.
The Business Employment Incentive Program (BEIP) grant offers powerful appeal for businesses in New Jersey.
TEDC names chair, officers.
The New Jersey residential well-testing program--a case study: Randolph Township.
Smart growth in the Pinelands: a vision becomes reality.
Russian Peasants Go to Court: Legal Culture in the Countryside, 1905-1917.
320-acre mixed-use development inches closer for Walters Group.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |