Data sources and methodology

November 2021

EWG’s Tap Water Database includes information about drinking water quality for nearly 50,000 community water systems nationwide. The database displays information about the chemical and radioactive contaminants detected in drinking water, and how these concentrations compare to federal legal limits and health guidelines. EWG has been collecting and analyzing tap water quality data since 2003.

The 2021 edition of the database features the results of water quality tests conducted by water systems between 2014 and 2019. EWG’s drinking water quality database includes a total of nearly 31 million test results for 534 chemicals, 324 of which were detected in drinking water, making it the largest resource of free and openly accessible data on U.S. drinking water.

How EWG collected the data

State drinking water authorities, typically health or environmental departments, maintain the records of water test results by drinking water systems or associated water testing laboratories. To build the database, EWG requested test results from authorities responsible for drinking water quality oversight in all 50 states and Washington, D.C.

Additionally, EWG obtained the results of drinking water tests for unregulated contaminants – substances that are known to contaminate drinking water but that have no national legal limits – from the Environmental Protection Agency. The EPA’s third and fourth Unregulated Contaminant Monitoring Rules ( UCMR3 and UCMR4) required that community water systems serving more than 10,000 people test water for 30 unregulated contaminants. Data collected from 2013 to 2016 for UCMR3 and from 2018 to 2021 for UCMR4 is included. The EWG database also includes drinking water violation information from the EPA Enforcement and Compliance History Online, or ECHO, website.

The EPA classifies the water systems that purchase, or otherwise receive, some or all of their finished water from one or more community water systems as “ consecutive systems,” which have different water quality monitoring requirements. Such systems are required to test for disinfection byproducts, lead and copper only.

For consecutive systems, EWG used publicly accessible records to list where the system likely purchases its water and to provide links to the supplier systems. If the consecutive system purchases or receives water from one supplier company only, EWG lists the water quality results for the supplier company on the page for the consecutive system.

For example, D.C. Water, which supplies tap water to the nation’s capital, purchases its water from the Washington Aqueduct, a federally owned and operated public water supply agency under the authority of the U.S. Army Corps of Engineers. Therefore, test results conducted by the Washington Aqueduct are also displayed on the page for Washington, D.C., water in the EWG Tap Water Database.

Population statistics for the community water systems were obtained from state drinking water programs and supplemented by information from the U.S. EPA Envirofacts database. All population numbers represent an estimate, and the specific number of customers served by individual community water system may differ.

EWG’s database does not include water quality information for private wells, because there are no federal requirements for private wells to be tested, and data on private wells are not readily available.

Data quality control

After assembling our database, EWG reviewed the data for inconsistencies, systematically searching for, identifying and removing data points that could contain errors, such as the use of measurement units different from what is typically used for a specific contaminant. During this process, EWG removed many thousands of test results flagged as potentially erroneous.

Several months ahead of the release, EWG briefed the Association of State Drinking Water Administrators, the American Water Works Association and the Association of Metropolitan Water Agencies about the upcoming edition of the database. EWG provided the American Water Works Association and the Association of Metropolitan Water Agencies member utilities with advance access to the EWG database so they could review the assembled data and make corrections, if necessary. For the 2021 Tap Water Database release, water utilities took the opportunity to review the data for their systems, and then contacted EWG during the review period to provide corrections, verify the accuracy of the data or suggest feedback about data display.

Calculation of average contaminant levels

For the 2021 edition of the EWG’s Tap Water Database, the following periods were used to calculate average contaminant concentrations for individual community water systems:

• All samples available for contaminants monitored under the EPA’s UCMR3 and UCMR4 programs.
• All samples for the 2014-2019 data period for radioactive elements.
• All available samples for 2017-2019 for federally regulated water contaminants that are monitored annually and all other contaminants.

These periods were chosen to allow EWG to capture the information about the occurrence of contaminants that are not monitored every year, such as radiological contaminants, or chemicals that were monitored during a specific time frame, such as the UCMR3 contaminants.

For most contaminants, we selected three years of data from 2017 to 2019 to reflect the most recent contaminant concentrations.

EWG calculated arithmetic mean averages of the contaminant concentrations for each community water system for the analyzed data period. Test results reported as “non-detects” were assigned a value of zero and included in the overall data array for the calculation of averages.

This approach is conservative and exerts a downward effect on the overall exposure estimates because, at least in some states, the detection limit for purposes of reporting is higher than what is achievable with the analytical capabilities of the most sensitive test methods.

Further, all individual test results are listed for each community water system in the database for the 2014-2019 time period and from the UCMR3 and UCMR4 programs. For every water sample, EWG database listed the date, numerical result, and laboratory identification number, where available.

Some water systems blend water from different water sources in differing amounts, and this mixing may reduce the average contaminant amount in the finished tap water served to consumers. The water blending may be adjusted seasonally or more frequently, depending on overall water demand, supply changes and contaminant levels. Water systems may use different methods for averaging, and the values published within a water quality report may not necessarily match what is shown in the database.

EWG compared the calculated average contaminant concentrations for individual community water systems to state and national population-weighted contaminant concentration averages for the same period. The state and national averages were calculated according to the formula:

where average contaminant concentration corresponds to values for individual community water systems of different sizes; total population corresponds to the population served by community water systems state- or nation-wide; and the resulting values are summed for all community water systems in a state or nationwide. This averaging includes systems where the contaminants were listed as non-detects. These systems were treated as having a contaminant level of zero for the calculation of population-weighted averages.

Health guidelines

EWG compared average contaminant concentrations to the federal legal limits and health guidelines published by government agencies and those from EWG’s own research:

• EPA Maximum Contaminant Level, or MCL – the enforceable federal standard that defines the highest level of a contaminant allowed in drinking water. The EPA has not set a new tap water standard in almost 20 years. Among the existing MCLs for chemical contaminants, two are over 40 years old (MCLs for nitrate and radium); eight were published in 1987, and the majority were developed in the early 1990s. Available here.
• California Public Health Goal, or PHG – defined by the California Office of Environmental Health Hazard Assessment, or OEHHA, as the level of contaminant in drinking water that poses no significant health risk if consumed for a lifetime. Public health goals are based solely on health effects. For carcinogens, the level represents one additional case of cancer per million people drinking the water over a lifetime. Available here.
• EPA one-in-a-million cancer risk level – the concentration of a chemical in drinking water corresponding to an excess estimated lifetime cancer risk of one in a million, published by the EPA Integrated Risk Information System. Available here. 
• Reference concentration under EPA Unregulated Contaminants Monitoring Program – a non-enforceable health guideline for drinking water contaminants that represents a benchmark for unregulated contaminants included in the EPA monitoring program. Available here.
• Minnesota Health Risk Limit – defined by the Minnesota Department of Health as the concentration of a groundwater contaminant, or a mixture of contaminants, that can be consumed with little or no risk to health. Available here. 
• EWG standards – health guidelines EWG developed based on recent scientific research. Available here.

Non-enforceable health guidelines represent the maximum concentration of a contaminant in water that scientists consider safe. This value is based only on protecting health and does not consider treatment costs, treatment feasibility or other factors.

For chemicals that can cause cancer, health guidelines typically represent one excess case of cancer per million people drinking the water over a lifetime. For chemicals that are not considered carcinogens, health guidelines are set using toxicity information from animal or human studies after applying additional safety factors.

These safety factors are used to provide a margin of safety based on differences between people and animals, differences among individuals in their susceptibility to chemical exposure, and general uncertainty – the fact that for many chemicals, scientists do not fully understand the potential health effects. In addition, the safety factors typically incorporate the increased likelihood of harm from exposure to toxic chemicals during pregnancy or childhood.

Most health guidelines in the database reflect the public health goal values published by the OEHHA. Where multiple health-based guidelines are available, EWG used the most recent health assessments completed by federal or state health agencies.

Exposure to contaminants at concentrations above the health guidelines increases the possibility of additional adverse health effects. For chemicals with health guidelines, we have identified a critical health effect, as well as a list of the other health effects that were observed in research studies. The primary sources of information for the other health effects were drinking water toxicity assessments completed by the state of California or the EPA.

Additionally, EPA published Maximum Contaminant Level Goals , or MCLG, along with the MCLs for individual chemicals. MCLGs are non-enforceable federal health guidelines. For cancer-causing chemicals, the EPA typically set these health goals to zero.

The EPA published MCLGs of zero for acrylamide, alachlor, arsenic, benzene, benzo[a]pyrene (PAHs), bromate, bromodichloromethane, bromoform, carbon tetrachloride, chlordane, 1,2-dibromo-3-chloropropane, 1,2-dichloroethane, dichloromethane, 1,2-dichloropropane, di(2-ethylhexyl) phthalate, dioxin (2,3,7,8-TCDD), epichlorohydrin, ethylene dibromide, heptachlor, hexachlorobenzene, lead, PCBs, pentachlorophenol, tetrachloroethylene, toxaphene, trichloroethylene, vinyl chloride, alpha particles, beta particles, radium and uranium.

Displaying lead test results

Water utilities test for lead and copper levels inside their customers’ homes, unlike their tests for other contaminants, to account for the fact that lead corrosion and leaching occur within water lines and household plumbing. Lead test results indicate the potential magnitude of lead hazards within a water system but do not reliably predict the specific levels of lead for tap water within individual homes. For this reason, EWG displays both the individual lead test results and the 90th percentile lead test values, where available.

The EPA’s current Lead and Copper Rule allows 10 percent of homes sampled to exceed the federal lead action level of 15 parts per billion, or ppb; 90 percent of the sampled homes must have lead levels below 15 ppb for the community water system to comply with the federal regulations. EWG obtained information about the 90th percentile lead tests from the EPA’s Envirofacts database and displayed the data for the most recent years available.

Water utilities also report individual test results to the government agencies responsible for overseeing drinking water safety in each state. EWG collected individual household lead test results, where available, and displayed this information in our database. Whenever a water system tested at least 10 household water samples for lead in the most recent year analyzed by EWG, the test results are displayed in a pie chart to help viewers distinguish the proportion of tests above health guidelines.

EWG analysis of drinking water compliance data

To examine whether tap water provided by different community water systems complied with federal drinking water standards, EWG analyzed information from the EPA’s ECHO database for the latest three-month period for which the EPA posted assessment information online, as well as information on the system’s compliance status over the past three years.

Drinking water quality is regulated on a state level, with state agencies in all but one state responsible for compiling and reviewing drinking water data, confirming that community water systems comply with federal laws, and addressing noncompliance issues through enforcement actions.

Wyoming is an exception, because the regional EPA office is directly responsible for the oversight of the state’s drinking water systems. Drinking water systems collect and report the water quality data to their designated authorities, and the information about Safe Drinking Water Act violations is compiled and reported quarterly to EPA authorities.

The EPA describes the Safe Drinking Water Act compliance process as follows:

After a violation has been reported, there is typically a three-to-six month review period before the official quarterly status of the [public water system] is updated. The most recent “official” compliance status is displayed in the 12th quarter on the Detailed Facility Report. If states have provided new violation information for the next quarter, these violations are displayed in the 13th quarter. Until the next quarter becomes official, these data are considered draft, have not been fully quality assured, and are subject to change.

As defined by the EPA , there are several types of drinking water violations, the most serious of which is a health-based violation, which indicates a lack of compliance with the federal MCLs or other legal requirements limiting the levels of contaminants in the water, such as the application of appropriate treatment techniques for lead. The EPA assigns a greater number of violations points – five or 10 – for such significant compliance issues.

Other types of violations include monitoring and reporting violations, public notice violations, and failure to keep required records of water treatment. Such compliance issues typically receive one violation point.

Violation points accrue for that system based on its compliance history, and remain associated with it, until specific violations are corrected. The EPA’s ECHO database Help page explains the types of health-based violations, other types of violations and the weighted point system the EPA uses to reflect a water system’s degree of noncompliance.

In addition to identifying specific violations, the EPA identifies community water systems with unresolved serious, multiple, and/or continuing violations. Such systems are designated “serious violators,” so that the drinking water system and the responsible agency act quickly to address the issue. Where provided by the law, formal enforcement action might be applied.

The EPA’s ECHO site also contains detailed facility information that can be reached through the link at the bottom of each water system page in the EWG Tap Water Database. This detailed facility report includes a summary graphic for the most recent 12-quarter period, where each quarter is identified as “no violation,” “noncompliance,” “significant violation” or “unknown.” These reports also provide detailed violation histories for each quarter and each drinking water quality regulation. In the EPA’s summary graphics for the detailed facility reports, quarters in violation are classified as “noncompliance,” and quarters with serious violations are classified as “significant violation.”

EWG’s Tap Water Database pages for individual water systems indicate with a “Yes” or “No” whether the system was listed by the EPA as being “in violation” or “a serious violator” in the past three years of reviewed data. Where information was available, EWG also listed the number of quarters over the past three years a specific water system was designated as having been “in violation” or “a serious violator” of drinking water standards. On the state pages in the database, community water systems with violations are listed by the number of current uncorrected violation points, according to information accessible on the ECHO website.

Disclaimer

EWG has worked to ensure the accuracy of the information provided in the Tap Water Database. The database is dynamic. The utility information, contaminant levels, and other information in the database may change based on evolving science, new information, or other factors. Please be advised the database frequently relies on data obtained from many sources, and accordingly, EWG cannot guarantee the accuracy of the information provided or any analysis based thereon.

EWG’s database is provided solely for your personal, non-commercial use. It is meant to help consumers make informed choices about their health and environment. Please be advised that any and all information EWG makes available through the database is for general informational purposes only and is not intended as, nor should they ever be considered a substitute for, professional medical advice. You can find EWG’s full terms and conditions here.