Glyphosate in Drinking Water: Health Concerns & Removal

Glyphosate is the single most widely used herbicide in the United States and the world. It reaches drinking water almost exclusively through agricultural runoff, it is regulated at one of the highest pesticide limits EPA maintains, and it sits at the center of a decade-long disagreement between the International Agency for Research on Cancer and national regulators over whether it causes cancer in humans. This profile lays out the chemistry, the exposure pathways, the science, and the removal options for households on affected water supplies.

What Is Glyphosate?

Glyphosate — chemically, N-(phosphonomethyl)glycine — is a synthetic, broad-spectrum, systemic herbicide introduced by Monsanto in 1974 under the trade name Roundup. It kills nearly all annual and perennial plants on contact, which is what made it an ideal companion technology to the “Roundup Ready” genetically modified corn, soybean, and cotton varieties Monsanto began commercializing in the mid-1990s.

US agricultural use of glyphosate has grown roughly fifteen-fold since Roundup Ready crops arrived on the market. The most widely cited USGS estimate puts annual US use at about 280 million pounds of active ingredient per year (2017 figures), making it the highest-volume pesticide in American agriculture by a large margin. Globally, total use exceeds 1.6 billion pounds per year.

Glyphosate works by inhibiting EPSP synthase (5-enolpyruvylshikimate-3-phosphate synthase), an enzyme in the shikimate pathway that plants and many microorganisms use to produce aromatic amino acids. Because animals do not have the shikimate pathway — they obtain aromatic amino acids from diet — the mechanism was long considered mammalian-safe. That argument has been partially complicated by the observation that gut bacteria do have the shikimate pathway, raising questions about potential microbiome effects from ingested glyphosate.

The molecule itself is highly water-soluble, non-volatile, colorless, odorless, and tasteless at concentrations found in water. In soil, microbial degradation typically breaks glyphosate down within days to a few weeks, although half-lives can extend to months in cold or anaerobic conditions. In water, it persists somewhat longer, particularly in sediments. Glyphosate also binds tightly to soil particles through its phosphonate group, which limits leaching into deep groundwater — a property that makes it primarily a surface water concern rather than a well water one. This is a different behavior than atrazine, which is more mobile and more widely detected in shallow wells.

How Glyphosate Gets Into Drinking Water

Glyphosate reaches drinking water through a small number of well-characterized pathways, dominated by agriculture. Because the molecule adsorbs to soil and breaks down relatively quickly, the concentrations that reach public water systems are usually low — but detections are extremely common.

Agricultural Runoff

Row-crop agriculture is the primary pathway. Glyphosate is sprayed on corn, soybeans, and cotton throughout the growing season and is increasingly used as a pre-harvest desiccant on wheat, oats, barley, and lentils to dry down the crop and synchronize harvest. When storm events follow application, a fraction of the dose washes off fields into ditches, streams, and rivers. Peak detections in surface water tend to follow spring planting applications in the Corn Belt and late-season desiccation pulses in wheat country.

Residential and Municipal Use

Glyphosate is the most widely sold consumer herbicide in the United States, used on lawns, gardens, sidewalk cracks, and ornamental beds. Municipalities and transportation departments apply it for right-of-way vegetation control along roads, rail corridors, and utility easements. Some water utilities and park districts also use glyphosate formulations labeled for aquatic weed control. These uses are smaller than agricultural application in absolute volume but contribute meaningfully to urban and suburban watershed loads.

AMPA Degradate

Glyphosate breaks down into aminomethylphosphonic acid (AMPA), a degradation product that is more persistent and more mobile than the parent compound. USGS monitoring consistently detects AMPA more frequently than glyphosate itself, and often at similar or higher concentrations. AMPA can also form from phosphonate-based detergents and industrial chemicals, so its presence in wastewater-influenced streams is not solely attributable to herbicide use. For drinking water purposes, AMPA is generally considered to share glyphosate’s low acute toxicity profile, although it is less thoroughly studied.

Groundwater vs Surface Water

Glyphosate is primarily a surface water contaminant. Its phosphonate group binds strongly to iron, aluminum, and clay in soil, which slows vertical transport through the vadose zone. USGS sampling finds glyphosate and AMPA at substantially lower detection frequencies in groundwater than in streams. Shallow wells in agricultural areas with sandy, permeable soils or karst geology can still show periodic low-level detections, and private wells close to heavily treated fields are the most likely to be affected.

Geographic Hotspots

Region	Context
Corn Belt (IL, IA, IN, NE, MN)	Highest application volumes in the country; surface water detections common during planting and post-emergence applications
Mississippi River basin	Watershed-scale exports into the Gulf of Mexico; cumulative applied volumes measured in billions of pounds
Pacific Northwest	Wheat desiccation, orchards, and forestry uses contribute regional loads
Cotton belt (TX, MS, AR, LA)	Extensive in-season and defoliation use; AMPA frequently elevated
California Central Valley	Roundup Ready crops, orchards, and vineyards drive sustained detections

Glyphosate commonly co-occurs with atrazine and nitrate in Corn Belt surface water, and with PFAS in watersheds with mixed industrial and agricultural inputs.

Health Effects

The central health debate over glyphosate is not about acute toxicity — on that metric it ranks below table salt — but about whether it causes cancer, particularly non-Hodgkin lymphoma, at real-world exposure levels. On that question, the world’s leading cancer research body and the major national regulators have reached different conclusions using different evidence frameworks. Both positions are defensible within their respective methodologies, and a reader trying to understand the science should hold both in view rather than picking a side.

IARC Classification (Group 2A — “Probably Carcinogenic”)

In March 2015, the International Agency for Research on Cancer (IARC), part of the World Health Organization, classified glyphosate as Group 2A — probably carcinogenic to humans. The classification rested on “limited” evidence of non-Hodgkin lymphoma in humans from agricultural worker studies, “sufficient” evidence of cancer in experimental animals, and “strong” evidence of genotoxicity and oxidative stress mechanisms. IARC’s methodology is hazard-based: it asks whether a substance is capable of causing cancer under any conditions, not whether typical exposures produce measurable risk.

EPA and Regulatory Agency Positions

EPA’s 2020 interim registration review decision concluded that glyphosate is “not likely to be carcinogenic to humans” at dietary exposures, and the agency maintained the existing tolerances and drinking water limits. The European Food Safety Authority (EFSA) reached a similar conclusion in its 2023 review, finding that available data did not support classifying glyphosate as a carcinogen at regulatory exposure levels, and the EU renewed glyphosate’s authorization for another ten years. Health Canada, the German BfR, and Australia’s APVMA have reached broadly comparable conclusions. These agencies use risk-based frameworks that weigh the full dataset — including large pooled epidemiology studies and guideline-compliant toxicology — and give greater weight to exposure-adjusted outcomes than IARC’s hazard-based approach.

Non-Hodgkin Lymphoma Litigation

Non-Hodgkin lymphoma is the most studied individual health outcome in the glyphosate literature and the centerpiece of US civil litigation. Bayer, which acquired Monsanto in 2018, has faced tens of thousands of Roundup cancer lawsuits. According to recent company disclosures, total provisions and payments related to the Roundup litigation now reportedly exceed ten billion dollars, with additional billions set aside for future claims and a proposed multi-billion-dollar class settlement under court review in 2026. The outcome of ongoing cases and the class settlement will likely continue to evolve — precise totals quoted in this profile should be treated as point-in-time figures.

Endocrine and Other Effects

Beyond cancer, some studies report endocrine-disrupting activity, effects on the gut microbiome (via the shikimate pathway in gut bacteria), kidney and liver effects at high doses, and developmental effects in animal models. The weight of evidence here is mixed and more preliminary than the cancer data. Regulators have generally concluded that these endpoints do not currently justify tightening the drinking water standard, while independent researchers continue to study low-dose and mixture effects.

Children and Pregnant Women

Drinking water epidemiology specifically targeting glyphosate in children and pregnant women is limited. Most human glyphosate exposure is dietary rather than water-borne — residues on grain products, legumes, and some produce dominate the intake of non-occupationally exposed populations. Biomonitoring studies find measurable glyphosate in the urine of most sampled US adults and children, with the dominant source appearing to be food.

Drinking Water Context

At the EPA MCL of 700 ppb, drinking water is not expected to be a meaningful contributor to total glyphosate exposure for most people. Typical finished water concentrations are roughly three to four orders of magnitude below the MCL. The more substantive concern with the US drinking water standard is not that water routinely exceeds it but that the standard itself was derived from non-cancer endpoints and from toxicity studies that predate the IARC review — a point health advocates and some scientists have raised in calls for a lower limit.

EPA Regulation and Limits

EPA regulates glyphosate under the National Primary Drinking Water Regulations at a Maximum Contaminant Level (MCL) of 700 ppb (0.7 mg/L). This is one of the highest MCLs for any regulated pesticide in the US, reflecting glyphosate’s relatively low acute mammalian toxicity and the non-cancer (kidney effects in chronic rodent studies) endpoint EPA selected as the basis for the standard. The MCL was promulgated in the early 1990s as part of EPA’s Phase V round of chemical contaminant regulation and has not been revised. EPA reaffirmed the standard in its 2020 interim registration review.

The Maximum Contaminant Level Goal (MCLG) for glyphosate is also 700 ppb, equal to the MCL. MCLGs are health-based, non-enforceable targets; setting the MCLG equal to the MCL reflects EPA’s determination that glyphosate is not a carcinogen at regulated doses, under which a non-zero MCLG is permitted.

Standard	Value	Notes
EPA MCL	700 ppb	Among the highest pesticide MCLs in US drinking water rules
EPA MCLG	700 ppb	Matches MCL; based on non-cancer endpoint
EU Drinking Water Directive	0.1 ppb per pesticide	Blanket per-pesticide limit, not toxicity-derived
California Public Health Goal (OEHHA)	Reported in the sub-ppm range	Non-enforceable health goal; consult OEHHA for current value
EWG health guideline	Well below 700 ppb	Advocacy guideline, not regulatory

The most striking number in the table is the 7,000-fold gap between the US MCL and the EU’s 0.1 ppb per-pesticide limit. That gap is not primarily a scientific disagreement about glyphosate toxicity — the EU limit applies uniformly to every pesticide in drinking water regardless of individual risk, as a precautionary backstop against chronic low-level exposure to mixtures. The US standard, by contrast, is derived from compound-specific toxicology. Both approaches are internally consistent; they answer different questions.

How Widespread Is Glyphosate?

Glyphosate is among the most frequently detected pesticides in US surface water. USGS National Water Quality Assessment (NAWQA) sampling finds glyphosate in roughly 40 percent of assessed US streams and AMPA in roughly 60 percent — AMPA is detected more often because it is more persistent and because it also arises from non-herbicide sources. Detection frequencies vary by land use, with agricultural watersheds showing the highest rates and urban watersheds also contributing meaningfully.

Concentrations, however, rarely approach the MCL. Typical USGS stream medians fall in the low fractions of a part per billion, with ninety-fifth-percentile values still well below 10 ppb in most watersheds. No US community water system has a publicly documented MCL violation for glyphosate in finished drinking water. The policy-relevant concern is therefore not regulatory exceedance but the combination of near-universal low-level detection, a decades-old standard, and the unresolved IARC-versus-regulator disagreement about low-dose cancer risk.

Private wells in agricultural areas show lower detection frequencies than surface water but are not always tested for glyphosate at all, since the compound is not part of most routine well panels. Where sampling is done, low-level detections are occasionally reported in shallow wells adjacent to heavily treated fields.

Glyphosate sits in an unusual category among drinking water contaminants: nearly ubiquitous at trace levels, almost never at regulated levels, and persistently controversial in the space between those two facts. Emerging contaminants like 1,4-dioxane and PFAS occupy a related but distinct position — detected widely, with standards still evolving.

How WaterVerge Tracks Glyphosate

WaterVerge pulls glyphosate monitoring data from EPA SDWIS (Safe Drinking Water Information System), which collects compliance sampling results reported by state drinking water programs. Because glyphosate is one of the unregulated-in-practice contaminants from a violations standpoint — detection is common but MCL exceedances are essentially absent — many community water systems operate on reduced monitoring frequency, which means a given city page may show a glyphosate result from several years ago rather than a recent sample.

Where a system has reported monitoring data, the most recent glyphosate result appears on that city’s profile along with its date and the ratio to the EPA MCL. Private well users are not covered by SDWIS and should arrange independent testing through a state-certified lab. Glyphosate-specific testing runs roughly $50 to $120 per sample through consumer-facing lab panels; it is not included in most standard well screening packages. See well water testing and how to test your tap water for lab selection and sampling guidance.

How to Remove Glyphosate

The first thing to know about glyphosate removal is what does not work. Boiling does not remove glyphosate — it concentrates it as water evaporates. Standard sediment filters do nothing for dissolved herbicide. Most basic refrigerator and faucet-mount filters are not certified for pesticide reduction and should not be relied on for glyphosate.

What does work is a narrower list than for many other pesticides, because glyphosate’s chemistry works against conventional treatment. The molecule is highly polar, hydrophilic, and zwitterionic, which means it does not adsorb strongly to activated carbon the way hydrophobic pesticides like atrazine do. That makes reverse osmosis — which rejects solutes by size and charge rather than by sorption — the most reliable home option.

Method	Removal Rate	Certification	Best For
Reverse osmosis	90-99%	NSF/ANSI 58	Under-sink, comprehensive contaminant reduction
Activated carbon block (high-quality)	70-90%	NSF/ANSI 53 pesticide reduction	Under-sink and countertop; verify specific claim
GAC whole-house	70-90%	NSF/ANSI 42 / 53	Point-of-entry for well water
Nanofiltration	90-95%	Membrane process	Utility-scale and high-end residential
Standard pitcher carbon	Limited / variable	Rarely certified for glyphosate	Not reliable for glyphosate

A few practical notes. NSF/ANSI 53 certification specifically listing glyphosate reduction is relatively uncommon compared to more routinely claimed pesticides; when a carbon filter claims glyphosate removal, look for the specific reduction claim in the NSF data sheet rather than relying on general “pesticide reduction” language. Reverse osmosis systems certified to NSF/ANSI 58 generally reduce glyphosate as part of overall dissolved-solid rejection, and are the most reliable option for households wanting to minimize glyphosate in drinking water — see best reverse osmosis systems for certified units. Under-sink carbon systems with verified glyphosate claims are covered in best under-sink water filters. For well water households who want whole-house protection, see best whole-house water filters — a GAC tank sized for the household flow rate is the typical approach.

Membrane fouling and pre-filtration matter. Iron, manganese, and sediment in raw water can foul RO membranes and reduce rejection performance; a properly sized sediment and carbon pre-filter extends membrane life and maintains removal rates.

Check Your City

Drinking water is, for most Americans, a minor route of glyphosate exposure relative to food — residues on grain-based products typically dominate total intake. That is not an argument for ignoring water glyphosate, particularly for private well households in agricultural areas, where concentrations are more variable and where routine compliance monitoring does not apply. Testing is warranted if you are on a well near treated fields, if your system shows repeated surface water detections in its consumer confidence report, or if you simply want to quantify one input to your household exposure. Search your city to see WaterVerge’s latest glyphosate data for your public water system, or review well water testing for private well guidance.