Is NYC Tap Water Safe to Drink? What the City's Tests Show, by Borough and Building

NYC water quality in 2025: results against each maximum contaminant level

The Department of Environmental Protection (DEP) ran 332,800 analyses on 24,050 samples from the city's pipes in 2025. Its Drinking Water Supply and Quality Report states that the water met all drinking water standards. The chart shows the measures that have a maximum contaminant level; closest were Haloacetic acids and Trihalomethanes. Lead, copper and coliform follow other kinds of rule and are in the next section.

Where the water comes from. Catskill and Delaware reservoirs supplied 91.3% in 2025, unfiltered under a state waiver through at least 2027; the Croton system supplied 8.7%, filtered in the Bronx. DEP adds phosphoric acid and sodium hydroxide to limit metal dissolving from building pipes. Hillview Reservoir, the last stop before city mains, is uncovered; a 2019 consent decree sets a schedule to cover it.

Legal limits and health goals: where they differ

Reports that NYC water exceeds "health guidelines" use a different yardstick from the law. DEP's report defines a maximum contaminant level goal (MCLG) as "the level of a contaminant in drinking water below which there is no known or expected risk to health"; legal limits are set "as close to the MCLGs as feasible." Lead has an action level, "not an MCL": if more than 10% of sampled homes exceed it, the utility must act. Coliform has a treatment technique that "triggers a Level 1 assessment" above 5% in a month. For the contaminants DEP detected in 2025 with an EPA goal of zero, the two differ:

Source: NYC DEP, Drinking Water Supply and Quality Report 2025 (results, New York limits, MCLG column, footnotes 4 and 21); EPA, lead and PFAS rules. EPA announced on May 18, 2026 a proposal that would let systems request two more years, to 2031, to meet the PFOA and PFOS limits; it is a proposal, not a final rule.

"Under the legal limit" and "no known risk" are different statements. The lead action level of 15 µg/L is the first kind.

Bacteria: 106,394 street-station tests since 2015

We recomputed the bacteria record from DEP's raw rows on NYC Open Data. Coliform bacteria, which DEP lists as naturally present in the environment, are tested as a warning sign; the citywide monthly share peaked at 4.3% (38 of 887 samples).

E. coli turned up in 3 of 106,394 samples from 2015 to 2025 (two in 2017 and one in 2018); DEP's reports for those years record no violation, which requires a positive repeat sample. DEP's 2026 report, due in 2027, will say whether the repeat sample after the 2026 E. coli result was clean.

By borough. Over 2021–2025, the coliform-positive share was 0.97% in Queens, 0.65% in Staten Island, 0.39% in the Bronx, 0.38% in Brooklyn and 0.34% in Manhattan. Split by borough, seven of the 705 borough-months since January 2015 were above 5%, all in Queens and Staten Island; the highest was Staten Island in July 2022, 13 of 103 samples (12.6%). None was a violation: the 5% trigger applies to the whole system's samples each month.

How much chlorine is in NYC tap water?

About half a milligram per liter. The 9,205 street-side samples DEP took in 2025 averaged 0.51 mg/L of free chlorine, 79% of them between 0.3 and 0.8 mg/L, and the highest was 1.18 mg/L. The limit is 4 mg/L as a running annual average. DEP must keep a residual in the pipes so that bacteria cannot regrow, and its FAQ says chlorine "is not considered hazardous or harmful in the amounts used to treat the water supply."

Yearly averages stayed between 0.39 and 0.53 mg/L since 2015; 2025 borough averages ran from 0.43 mg/L in Manhattan to 0.67 mg/L in the Bronx. DEP's tip for the taste: keep a pitcher of tap water in the refrigerator overnight, or pour it between two containers about 10 times. (DEP's report prints 0.6 mg/L, its highest running annual average over 13,027 samples including entry points.)

Lead in NYC water: no lead mains, some lead at the tap

DEP states that its roughly 7,000 miles of aqueducts, tunnels and water mains contain no lead. Its 313 samples at street-curb stations in 2025 ranged from not detected to 2.5 µg/L. Lead gets in from a lead service line (the pipe from main to building, banned since 1961), lead solder (banned since 1987) and brass fixtures. EPA's health goal for lead in drinking water is zero; EPA says the best available science shows "there is no safe level of exposure to lead." The 15 µg/L action level is a trigger for the utility, not a safe level.

In the official Lead and Copper Rule sampling, which targets homes likely to have lead plumbing, 90% of the 326 homes tested in 2025 measured 10 µg/L or less; 16 were above the 15 µg/L action level and the highest was 155 µg/L.

EPA's Lead and Copper Rule Improvements lower the action level to 10 µg/L. Systems must begin complying three years after publication on October 30, 2024, which EPA expects to be late 2027 (EPA Q&A). NYC's 90th percentile was above 10 µg/L in 9 of 12 years. The rule also changes sampling and requires most systems to replace lead service lines within 10 years, so later results will not compare directly.

What 54,355 free lead test kits found

DEP publishes every free kit result. Each kit holds a "first draw" after hours unused and a sample after running the tap 1 to 2 minutes. Of 54,308 kits returned from 2014 through 2025, 4.5% had lead above 15 µg/L in at least one sample and 34% had detectable lead in the first draw. The yearly share above 15 µg/L ranged from 3.2% to 8.4%; in 2025 it was 6.0%.

What running the tap does to lead levels

Of the 1,411 kits since 2014 whose first-draw sample was above 15 µg/L, 1,117 (79%) had a flushed sample at or below it, and the median fell from 32 to 4 µg/L. Another 1,019 kits went the other way: a first draw at or below 15 µg/L and a flushed sample above it. The dataset has no plumbing field, so it cannot say why. EPA says a lead service line is typically the largest source where one exists, and water from the service line reaches the tap only after the indoor pipes empty; that fits a higher flushed sample without proving it. A clean first draw does not clear a building.

NYC lead test results by ZIP code

The table ranks the 79 ZIP codes with at least 150 kits from 2021 to 2025 by the share above 15 µg/L in either sample. The 15 highest split as 8 in Brooklyn, 6 in Queens and 1 in Staten Island; the 9 lowest are all in Manhattan. Harlem's ZIP codes are in the table, and its neighborhood page covers the rest of its record.

Kits collected 2021–2025 with both samples. Linked names are DwellCheck neighborhoods with the most homes in the ZIP; unlinked names are the 2020 neighborhood tabulation areas holding at least a quarter of its homes. A ZIP's share describes the households that requested a kit, not every home, and a low share does not clear any building. Source: NYC DEP via NYC Open Data • pulled 2026-11-12.

Lead service lines by borough

DEP's records list a lead service line, or a galvanized one that must be replaced, at 120,611 of the 817,567 tax lots with a water service line (14.8%), and the material is unknown at another 114,473 (14.0%). Together that is 28.8% of lots where the line is, or may be, lead. Most entries come from plumbing filings, not inspection, and 39,818 "non-lead" entries are a statistical prediction.

Look up an address on DEP's lead service line map. The owner is responsible for replacing it; DEP's free replacement program covers eligible properties in the Bronx and Queens.

Rooftop water tanks: who has one and what the inspections say

In buildings with more than six floors, pumps push water to a rooftop tank. The owner must have each tank inspected yearly, with a bacteria test, and file by January 15. Inspectors filed 6,077 tank reports for 4,627 buildings for 2025, 84% of them in Manhattan; the number of buildings filing rose from 3,629 in 2015.

Of 61,490 tank reports for 2015–2025, 15 recorded coliform, E. coli or both. Five of those record E. coli present with total coliform absent, an inconsistent coding we could not resolve from the dataset. 1,112 reports (1.8%) recorded sediment, growth, debris or animal signs that were corrected. The Health Department says the results "have not been verified by a City Agency," and buildings that filed nothing do not appear. Owners must keep results for five years and let residents see them: ask, or search the inspection dataset by address (NYC Health rules).

Taste, smell and color: what New Yorkers report to 311

New Yorkers filed 1,324 water quality complaints with 311 in 2025,15 per 100,000 residents, from 11 in Queens to 24 in Manhattan. A complaint is a report, not a lab finding.

Taste and smell codes included musty (196), chlorine (125), bitter or metallic (121) and sewer (119). DEP's FAQ:

• Chlorine. Required in the pipes and more noticeable in warm weather, DEP says. Chill the water or pour it between containers.
• Cloudy or milky. Usually trapped air that clears once the water stands; DEP says it is "not a public health concern."
• Brown or discolored. DEP says it is not harmful but residents "should not drink obviously discolored water"; run the cold tap for 2 to 3 minutes.
• Metallic, sewer or anything else new. No specific cause given; DEP asks residents to report them to 311.

How to check the water at your own address

1. Look up the service line on DEP's map; indoor solder and fixtures are not on it.
2. Order the free lead test kit through the 311 online form or by calling 311. If you rent and the result is high, tell the landlord.
3. Above six floors, ask about the tank: the latest annual inspection and bacteria test.
4. Until you have results, follow DEP's precautions. Its lead page says to run the tap in the morning until the water is colder, which "usually takes about 1 to 2 minutes"; its annual report says 30 seconds to 2 minutes when water has sat for several hours. Use cold water for drinking and cooking, clean the faucet aerator, and if you use a filter, choose one certified by NSF or UL for lead removal.

What this data cannot tell you

• Nothing here measures your tap. Lead depends on one building's pipes and can vary between apartments and from one morning to the next.
• Kits are self-selected and the official panel is worst-case, so neither is a citywide rate.
• Service-line and tank records are paperwork: 14.0% of lots are of unknown material, and tank results are filed by the owner's inspector without city verification.
• Unregulated substances without a federal or state limit are outside these tables.
• This is not medical advice. For health questions about lead, especially for a child or during pregnancy, ask a doctor or call 311.

Method and sources

1. City supply. DEP's 2025 report; share of limit = result ÷ limit.
2. Bacteria and chlorine. DEP Distribution Water Quality (bkwf-xfky), sample type "Compliance", January 2015 through September 2026. A coliform or E. coli result of "<1" MPN/100 mL is negative. Monthly share = positives ÷ samples, citywide and by borough.
3. Official lead samples. DEP Compliance at-the-tap Lead and Copper Data (3wxk-qa8q); 90th percentile interpolated at rank 0.9 × n.
4. Free kits. DEP Free Residential at-the-tap Lead and Copper Data (k5us-nav4), 54,355 rows; 47 without both samples dropped. Over 15 = above 0.015 mg/L. The borough column disagrees with the ZIP code in about 5% of rows, so borough comes from the ZIP (PLUTO).
5. Service lines, tanks, 311. jqfp-uff7 (856,912 lots), gjm4-k24g, erm2-nwe9 (rates on 2020 Census populations).
6. Rules. EPA lead, LCRI and PFAS pages; DEP lead and FAQ pages; NYC Health tank rules.

Cross-checks. The raw rows reproduce DEP's printed 2025 coliform and lead figures (9,203 samples, 47 positive, worst month 3.1%; lead 90th percentile 10 µg/L); all 24 checks passed. Open Data rounds lead to whole µg/L, so it shows 12 homes above 15 µg/L where DEP reports 16.

Frequently asked questions