PFAS Map: 1,127 Water Systems Averaged Above EPA's PFOA or PFOS Limit

On the PFAS map, 44% of Connecticut and New Jersey water systems averaged above 4.0 ng/L

Across the 50 states and DC, 43 have at least one water system whose UCMR 5 sampling-point average for PFOA or PFOS was above EPA's 4.0 ng/L limit. The rate ranges from 44.4% of systems in Connecticut to zero in 8 jurisdictions.

Connecticut (28 of 63 systems, 44.4%) and New Jersey (114 of 258, 44.2%) have the highest shares of systems averaging above the PFOA or PFOS limit in EPA's 2023 to 2025 UCMR 5 sampling. The national share is 11.1%, so New Jersey's rate is 4.0 times the national figure. Delaware (33.3%), Florida (32.2%) and South Carolina (31.8%) follow.

Nine jurisdictions have shares of 20% or more, all of them eastern states. The median state share is 5.6%. Several large states sit far lower: Texas 2.4% (27 of 1,124 systems), Illinois 2.7% and Michigan 1.6%. No UCMR 5 system averaged above the limit in Arkansas, the District of Columbia, Hawaii, Iowa, Montana, North Dakota, Vermont and Wyoming.

54.3 million people get water from a system that averaged above the PFOA or PFOS limit

Matching each UCMR 5 system to its population in EPA's SDWIS file (2026 Q3) gives 54,316,137 people served by systems with an above-limit average, 18.1% of the 300.2 million served by all fully sampled systems.

Florida (9.74 million), California (8.9 million) and New Jersey (5.12 million) account for 43.7% of the 54.3 million. 111 of the 1,127 systems serve 100,000 people or more.

The largest system on the list is MDWASA – MAIN SYSTEM (FL), serving 2,377,460 people, with a highest sampling-point average of 31.4 ng/L PFOS. A system is counted when any one of its sampling points averaged above the limit, so the population figure is an upper bound on the people actually drinking water above 4.0 ng/L.

For a homeowner, the system name and PWSID on the water bill or in the annual water quality report (Consumer Confidence Report) are the link to these results. EPA's UCMR occurrence data page lists every result by system.

One sample at 4 ng/L is not a violation: 1,739 systems hit it once, 1,127 on average

Because UCMR 5 could only report PFOA and PFOS at 4 ng/L or higher, every reported value is at or above the limit. 1,739 systems (16.9% of 10,307 tested) had at least one such result; 612 of them did not average above the limit once all samples were counted, or lacked a full set.

Compliance under the 2024 rule is judged on a running annual average of quarterly results at each sampling point, not on a single sample. UCMR 5 collected two samples a year at ground water points and four at surface water points, so EPA averaged those and counted an average as above the PFOA or PFOS limit at 0.00405 ug/L or more.

Our reproduction of EPA's method matches the agency's published counts exactly: 662 of 4,277 large systems and 474 of 5,852 smaller systems averaged above at least one 2024 standard. EPA's weighted national estimate, which corrects for the partial small-system sample, is 7.8% of all US public water systems.

If EPA drops four of the six standards, 9 systems leave the above-limit list

Of the 1,136 systems that averaged above any 2024 standard, only 9 (0.8%) were above solely the PFHxS, PFNA, HFPO-DA or Hazard Index standards that EPA proposed to rescind on May 20, 2026.

PFOS (875 systems) and PFOA (779) drive almost every case. The Hazard Index mixture (80), PFHxS (66), PFNA (9) and HFPO-DA (8) average above their limits far less often, and 77 of the 86 systems above one of them were also above for PFOA or PFOS. The 9 systems above only a proposed-rescission standard serve 0.14 million people.

These counts are UCMR 5 screening results. They show how much of the measured problem the rescission would remove from federal requirements; they do not show the health effect of the change, which EPA did not reopen for comment in the rescission proposal.

When do water utilities have to meet the PFAS limits? April 2029 in law, 2031 if EPA finalizes its proposal

The 2024 rule took effect June 25, 2024. Initial monitoring is due April 26, 2027 and the limits apply from April 26, 2029. A May 2026 proposal would let systems request until April 26, 2031 for PFOA and PFOS; as of 2026-11-02 we found no final rule in the Federal Register.

The legal challenge, American Water Works Association v. EPA, No. 24-1188, is pending in the D.C. Circuit. The court denied EPA's request to summarily vacate the four non-PFOA/PFOS standards in January 2026 and denied its request to sever and pause those claims on March 19, 2026 (Harvard EELP tracker). With that ruling, all six limits stay in place until a merits decision or a final rescission (MG+M). We found no merits decision as of 2026-11-02.

The April 2024 rule set the PFOA and PFOS health goals at zero and the enforceable limits at 4.0 ng/L; it set 10 ng/L for PFHxS, PFNA and HFPO-DA and a Hazard Index of 1 for mixtures (89 FR 32532). EPA estimated compliance costs and quantified health benefits at about $1.5 billion a year each (EPA fact sheet).

Filters certified to NSF/ANSI 53 or 58 reduce PFAS, but no label yet promises 4 ng/L

EPA tells consumers to look for certification to NSF/ANSI 53 or NSF/ANSI 58 for PFAS reduction, and notes that certifications as of April 2024 do not show removal down to the new limits.

EPA names three point-of-use technologies that can reduce PFAS: activated carbon, ion exchange resin and reverse osmosis. Its April 2024 fact sheet prices home filters from $20 to more than $1,000 before maintenance, and lists 5 ANSI-accredited bodies that certify filters for PFAS reduction (EPA home filter fact sheet; EPA filter page).

Filters only work while cartridges and membranes are replaced on the maker's schedule. An under-sink reverse osmosis unit also needs a drain connection and cabinet space, which is easier to plan during a kitchen remodel; see our kitchen remodel timeline for where plumbing rough-in falls. Mineral scale is a separate question covered in our hard water map.

What EPA and ATSDR link to PFAS exposure

EPA lists 6 groups of health outcomes linked to exposure to certain PFAS, from some cancers to immune and cholesterol effects; ATSDR says research on mixtures and low-level exposure is ongoing.

• Reproductive effects, including decreased fertility and higher blood pressure in pregnancy (EPA).
• Developmental effects in children, including low birth weight and accelerated puberty (EPA).
• Increased risk of some cancers, including prostate, kidney and testicular cancers (EPA); kidney and testicular cancer associated with PFOA (ATSDR).
• Reduced immune response, including lower antibody response to some vaccines (EPA, ATSDR).
• Interference with the body's hormones (EPA).
• Higher cholesterol and/or obesity risk (EPA); cholesterol increases associated with PFOA, PFOS, PFNA and PFDA (ATSDR).

Sources: EPA, updated October 30, 2026; ATSDR. Both agencies say risk depends on dose, duration and individual factors. Our lead service line counts are in the separate lead pipes map.

State-by-state findings

1. 1,127 of 10,128 UCMR 5 systems (11.1%) averaged above 4.0 ng/L for PFOA or PFOS at one or more sampling points, 2023 to 2025, United States.^[1][2]
2. Systems averaging above the PFOA or PFOS limit serve 54,316,137 people (SDWIS, 2026 Q3).^[1][4]
3. Connecticut has the highest share of systems above the limit, 44.4% (28 of 63), UCMR 5 2023 to 2025.^[1]
4. New Jersey has 114 systems above the limit, 44.2% of its 258 fully sampled systems, serving 5.12 million people.^[1][4]
5. Florida has the most people served by above-limit systems, 9.74 million, UCMR 5 2023 to 2025.^[1][4]
6. 43 of 51 jurisdictions (50 states and DC) have at least one system averaging above the PFOA or PFOS limit.^[1]
7. PFOS averaged above its limit in 875 systems and PFOA in 779, UCMR 5 2023 to 2025.^[1][2]
8. EPA reports 15.5% of large, 8.3% of medium and 7.1% of small UCMR 5 systems averaged above at least one 2024 PFAS limit.^[2]
9. EPA's weighted estimate is that 7.8% of all US public water systems would have an average above a 2024 PFAS limit.^[2]
10. 66% of UCMR 5 sampling locations with any PFAS at or above the reporting level had more than one PFAS (EPA, July 2026).^[2]
11. 9 systems (0.8% of 1,136) were above only standards EPA proposed to rescind on May 20, 2026.^[1][8]
12. The 2024 rule's PFOA and PFOS compliance date is April 26, 2029; EPA proposed April 26, 2031 for systems that request it.^[5][7]
13. EPA says certifications as of April 2024 do not show that a home filter removes PFAS down to the new limits.^[11][12]

Rankings and ratios we derived

Download the calculations (CSV)

Executive summary

• 1,127 of 10,128 UCMR 5 water systems (11.1%) had a sampling-point average above EPA's 4.0 ng/L limit for PFOA or PFOS in 2023 to 2025 sampling.
• Those systems serve 54.3 million people, 18.1% of the population served by all fully sampled systems.
• Connecticut (44.4%) and New Jersey (44.2%) have the highest state shares; 8 jurisdictions had none.
• Florida, California and New Jersey hold 43.7% of the people served by above-limit systems.
• A single-sample screen flags 1,739 systems; averaging, as EPA does, cuts that to 1,127.
• Only 9 systems exceed solely the PFHxS, PFNA, HFPO-DA or Hazard Index standards EPA proposed to rescind in May 2026.
• The 2024 rule requires compliance by April 26, 2029; EPA's May 2026 proposal would let systems request April 26, 2031. Neither proposal was final on 2026-11-02.

Frequently asked questions

Related guides, services and research

• How long a kitchen remodel takes
• Remodelers by city
• How Long Do Ovens Last? Gas, Electric and the Federal Recall Record
• How Long Do Refrigerators Last? DOE Lifespan Data and the Age of US Fridges
• How Long Do Toilets Last? Flush Rules, Breakdowns and Parts
• How Long Do Water Softeners Last? Resin, Valves and the Hardness Factor
• All kitchen and bath research

Reproduce this analysis

Every figure comes from the files below. The scripts that pulled and computed them are kept with the report and can be re-run.

Data quality and exclusions

• Kept only PFOA, PFOS, PFHxS, PFNA, HFPO-DA and PFBS rows from the 30-contaminant file; the other 23 PFAS and lithium were dropped from the state table.
• 179 tested systems had no sampling point with a full set of PFOA or PFOS results and are excluded from the average-based shares, as EPA does.
• 13 systems have a population of 0 in SDWIS and add nothing to population totals.
• Duplicate results for the same sampling point, event and contaminant were collapsed by keeping the higher value.
• 115 Tribal systems (EPA region codes and Navajo Nation) and 110 territory systems are reported as separate rows and excluded from the tile map.

Methodology

Selection: the primary dataset is EPA's UCMR 5 occurrence file (UCMR5_All.txt in ucmr5-occurrence-data.zip, file dated 2026-10-15), the final release covering samples from January 2023 to December 2025. Population served comes from EPA's ECHO SDWA download (SDWA_PUB_WATER_SYSTEMS.csv, submission quarter 2026Q3). Regulatory dates come from the Federal Register and EPA pages. Litigation events come from the Harvard EELP tracker, with a law firm note as support.

Inclusion: all rows for PFOA, PFOS, PFHxS, PFNA, HFPO-DA and PFBS. A system counts as tested if it has any PFOA or PFOS result. A sampling point has a full set when it has 2 results (ground water) or 4 results (surface water, ground water under surface influence, or mixed). Results reported as below the reporting level are set to 0.

Averages and cutoffs follow EPA's July 2026 Data Summary: an average counts as above the PFOA or PFOS limit at 0.00405 ug/L or more, above PFHxS, PFNA or HFPO-DA at 0.015 ug/L or more, and above the Hazard Index at 1.5 or more with at least two of the four PFAS at or above the reporting level. Hazard Index values are computed per sample event, then averaged. These rules reproduce EPA's Table 4 counts exactly (table t2).

Conflicts: EPA says UCMR 5 results do not indicate compliance; we report them as a screening comparison. Where two sources date the same court order differently (January 12 versus January 20, 2026), we use the tracker date and note the other in the fact sheet. Tribal systems without primacy and territories are reported as separate rows and are not drawn on the state map.

Derived metrics are in derived_metrics.csv with formulas and inputs; scripts/build_data.py recomputes everything from raw files. Last updated 2026-11-02.

This report presents published data with its limitations. It is not legal, financial or investment advice.

Sources, ranked by quality

Tier 1: primary and official sources (19)

• [1] U.S. EPA. UCMR 5 Occurrence Data Text Files (UCMR5_All.txt, file dated 2026-10-15). Samples 2023 to 2025; final release August 2026. Accessed 2026-11-02.
• [2] U.S. EPA. The Fifth Unregulated Contaminant Monitoring Rule (UCMR 5) Data Summary: July 2026 (EPA 815-S-26-007). 2023 to 2025. Accessed 2026-11-02.
• [3] U.S. EPA. Occurrence Data from the Unregulated Contaminant Monitoring Rule (web page, updated October 15, 2026). 2023 to 2025. Accessed 2026-11-02.
• [4] U.S. EPA ECHO. ECHO SDWA data download: SDWA_PUB_WATER_SYSTEMS.csv (submission quarter 2026Q3). 2026 Q3. Accessed 2026-11-02.
• [5] U.S. EPA, Federal Register. PFAS National Primary Drinking Water Regulation, 89 FR 32532 (April 26, 2024). 2024. Accessed 2026-11-02.
• [6] U.S. EPA, Federal Register. PFAS National Primary Drinking Water Regulation; Correction, 89 FR 49101 (June 11, 2024). 2024. Accessed 2026-11-02.
• [7] U.S. EPA, Federal Register. Extending the Compliance Deadline for the PFOA and PFOS Maximum Contaminant Levels, 91 FR 29425 (May 20, 2026). 2026. Accessed 2026-11-02.
• [8] U.S. EPA, Federal Register. Rescission of Regulatory Determinations and Removal of Related Provisions for Four PFAS Substances, 91 FR 29413 (May 20, 2026). 2026. Accessed 2026-11-02.
• [9] U.S. EPA. EPA Announces It Will Keep Maximum Contaminant Levels for PFOA, PFOS (news release, May 14, 2025). 2025. Accessed 2026-11-02.
• [10] U.S. EPA. Final PFAS National Primary Drinking Water Regulation (web page, updated May 18, 2026). 2024 to 2026. Accessed 2026-11-02.
• [11] U.S. EPA. Identifying Drinking Water Filters Certified to Reduce PFAS (web page, updated November 18, 2025). 2024 to 2025. Accessed 2026-11-02.
• [12] U.S. EPA. Fact Sheet: Reducing PFAS in Your Drinking Water with a Home Filter (April 2024). 2024. Accessed 2026-11-02.
• [13] U.S. EPA. Fact Sheet: PFAS National Primary Drinking Water Regulation (April 2024). 2024. Accessed 2026-11-02.
• [14] U.S. EPA. Our Current Understanding of the Human Health and Environmental Risks of PFAS (web page, updated October 30, 2026). 2026. Accessed 2026-11-02.
• [15] CDC Agency for Toxic Substances and Disease Registry. PFAS: Potential Health Effects of PFAS Chemicals. current page. Accessed 2026-11-02.
• [16] U.S. EPA, Federal Register. Revisions to the Unregulated Contaminant Monitoring Rule (UCMR 5), 86 FR 73131 (December 27, 2021). 2021. Accessed 2026-11-02.
• [17] U.S. EPA, Federal Register. Revisions To Establish the Sixth Unregulated Contaminant Monitoring Rule (UCMR 6), 91 FR 39952 (July 1, 2026). 2026. Accessed 2026-11-02.
• [18] U.S. EPA. Fifth Unregulated Contaminant Monitoring Rule (web page, updated August 28, 2026). 2021 to 2026. Accessed 2026-11-02.
• [21] Office of the Federal Register. Federal Register API search: EPA documents matching PFOA PFOS drinking water, published 2026-05-21 to 2026-11-02. 2026-05-21 to 2026-11-02. Accessed 2026-11-02.

Tier 2: trade bodies, market research, company data (1)

• [19] Harvard Law School Environmental and Energy Law Program. Regulatory Tracker: PFAS in Drinking Water. 2024 to 2026. Accessed 2026-11-02.

Tier 3: journalism and expert commentary (1)

• [20] MG+M The Law Firm. DC Panel Puts the Brakes on EPA's Attempt to Undo PFAS MCLs. January 2026. Accessed 2026-11-02.