EPA's Biosolids PFAS Guidance Binds No One. It Will Still Shrink Land Application Faster Than a Rule.

Nobody at a wastewater plant decides where the biosolids go. The farmer decides, or the farmer's landlord, or the county attorney who drafts the spreading agreement, or the insurer who reads that agreement two years later and asks a question nobody prepared an answer for. That's the audience EPA's biosolids PFAS guidance is actually written for, which is why the words "non-binding" on the cover page tell you almost nothing about what the memo will do.

The notice hit the Federal Register on July 6 (91 FR 41020), opening a sixty-day comment window that closes September 4 under docket EPA-HQ-OW-2026-2509. Assistant Administrator for Water Jessica Kramer signed the underlying memo on June 29. It recommends that utilities identify PFAS sources in their collection systems, work pollution prevention with industrial dischargers, monitor what comes out of the digester, and steer land application away from water features, ground where children play, and crops with higher exposure pathways. It changes no permit condition, amends no rule, and in EPA's own words "does not have the force and effect of law."

So the question I keep fielding from utility boards is whether they can wait for something firmer. They're already late, and not because a rulemaking is coming. Guidance that binds nobody moves faster than a rule precisely because it skips everything that slows a rule down: no regulatory impact analysis to defend, no cost record to litigate, no final agency action anyone can challenge in the D.C. Circuit. It goes straight into the documents that decide where sludge lands, which are the spreading contract, the county ordinance, and the lender's environmental questionnaire.

What the memo says, and the number it refuses to give

Set the recommendations next to EPA's January 2025 draft sewage sludge risk assessment and the difference is a missing figure. That assessment modeled land application at one part per billion of PFOA or PFOS in biosolids and found exceedances of the agency's own risk threshold. EPA now says the framing was flawed, that it leaned on atypical industrially contaminated sites, and that it left everyone with the impression that all use and disposal practices are unsafe. Reasonable criticism. But the 2026 memo answers a contested number with no number at all.

For anyone building a compliance program that's the harder outcome. A threshold you disagree with is still a threshold you can design monitoring around, budget against, and defend at renewal. Qualitative advice to keep biosolids off "areas where children may be present" is a standard somebody else gets to specify, and the somebody else will be a state agency, a county board, or an expert witness. I've watched this run in air permitting for a decade. Vague federal language doesn't stay vague; it gets pinned down locally, at a stringency the generator never had a chance to comment on.

Meanwhile 40 CFR Part 503 sits untouched. It still governs land application and surface disposal, still sets ceilings for nine metals, and still says nothing whatsoever about PFAS. That gap is the entire story of the next three years.

Criterion one: which counterparty ends up holding the bag

Operators tend to compare biosolids endpoints on cost per ton. Compare them instead on who carries the residual liability and how long that party's memory is, because that's the variable that actually moved in July.

Land application spreads liability outward. The generator ships a product, the landowner accepts it, and the exposure sits in soil that will be sampled by somebody eventually. Maine's farm cases made the shape of that clear enough. Once a state opens the question, the sampling doesn't stop at the property line and the generator's indemnity language gets read very carefully.

Landfilling concentrates the liability in one place, which is easier to manage right up until the landfill's own discharge permit tightens. The PFAS then leaves in the leachate rather than the truck, and the cost of that transfer is not small; we walked through the numbers in our analysis of what Superfund designation does to landfill leachate treatment economics. Disposal contracts written before that repricing are the ones being renegotiated now.

Thermal treatment moves the whole question to a different agency. Your counterparty stops being a farmer and starts being an air permit writer, which changes the renewal cycle, the monitoring obligations, and the people who show up to the hearing. It doesn't make the problem smaller. It makes it somebody else's file.

Endpoint	Who can close it on you	Warning you get
Land application	A state agency, a county board, or the landowner alone	Weeks. A handshake isn't a permit, and it can be withdrawn like one.
MSW landfill	Whoever writes the landfill's leachate discharge limits	One permit cycle, then passed to you as price
Surface disposal, monofill	The groundwater monitoring wells at your own unit	None. A trip is a trip.
Thermal, dedicated or co-fired	The air permit writer, at renewal rather than issuance	Longest of the four, and the most expensive to lose

Criterion two: what it costs the week your endpoint closes

Maine ran this experiment for everyone. LD 1911 banned land application of biosolids in April 2022, and the sludge had to go somewhere. Utilities that had been paying about seventy dollars a wet ton for handling in 2018 were quoted roughly $190 to $240 a wet ton to haul and landfill by 2023, per reporting in Treatment Plant Operator, with one district's annual biosolids line moving from about $1.5 million to $3.2 million. Then the state-owned Juniper Ridge landfill, short of the bulking agent it needed to take that much wet material, started turning away municipal waste. Exports to New Brunswick ran at a few thousand tons a month until emergency legislation shut that valve too.

None of those costs are the interesting number. The interesting number is one: how many endpoints a given plant can legally use tomorrow. A utility with a single contracted outlet has no negotiating position at all, and the vendor knows the renewal date better than the board does. Optionality is the asset here, and it's the asset the guidance quietly taxes, because every recommendation in the memo narrows the acreage that qualifies without narrowing the tonnage that arrives. So how many of your outlets survive a county ordinance drafted by someone who read this memo and nothing else?

Roughly four million dry metric tons of sewage sludge were generated in 2024, of which about 2.39 million went to land, 982,000 to landfill, and 558,000 to incineration, per EPA's own biosolids reporting. Actually, that needs a qualifier: those figures cover large publicly owned treatment works in states where EPA is the permitting authority, not the whole country, so treat the split as a strong indicator rather than a census. Even as an indicator it says the same thing. The endpoint under pressure is the one carrying three fifths of the load.

Criterion three: the permit path, and how long it really takes

Dedicated sewage sludge incinerators sit under 40 CFR Part 60 Subpart MMMM if they existed before October 2010, Subpart LLLL if they didn't, with limits on nine pollutants and no PFAS-specific requirement anywhere in either. That absence reads like freedom on the first pass. It isn't, because permit writers fill absences.

In 2024 I sat through a PFAS-related air permit review where the state inserted a continuous monitoring condition into the draft roughly halfway through the cycle. The installed analyzer's range didn't reach the new limit. Not a calibration problem, a hardware problem, and the applicant had already ordered long-lead items against the original draft. That's the failure mode with thermal: the technology works, the schedule doesn't, and the condition that hurts is almost never in the version you started commenting on. I had a similar 2023 Title V renewal where a modification everyone had agreed was minor tripped a PSD applicability review and cost eight months.

Co-firing dewatered or dried cake at an existing energy-from-waste unit is usually the faster route than permitting a new dedicated incinerator, because the combustion unit and its monitoring are already permitted and you're arguing about a feedstock change rather than a new source. It's not free. Dryers are capital and they're parasitic on the plant's heat balance (an Andritz drum dryer pulling cake to ninety percent solids is a real thermal load, not a bolt-on), and the receiving facility will want the biosolids specified tightly enough to protect its own permit. Operators evaluating that route should look hard at how the receiving unit's existing waste-to-energy technology handles a high-moisture, high-ash side stream before signing anything multi-year.

Whatever the endpoint, the destruction question is separate from the disposal question. Sending PFAS-bearing solids to a combustion unit is a transfer until you can show what came out of the stack, and the instruments that would show it are not standard equipment on most sludge burners.

Criterion four: what you can actually prove at renewal

Only one intervention in the memo changes the concentration rather than relocating it, and that's source control. Michigan has been running it since 2021, when EGLE began restricting land application of biosolids above 150 ppb PFOS and pushing industrial pretreatment programs at the plants that failed. Seven wastewater plants working their industrial users reported PFOS reductions in sludge of up to ninety-nine percent. The state has since tightened to a tiered structure: at or above 100 ppb the material is treated as industrially impacted and stays off fields, and between 20 and 100 ppb applications are cut to 1.5 dry tons per acre or paired with other mitigation.

Michigan's numbers are the strongest argument in this entire debate, and they cut against the reflex to go shopping for a thermal vendor. If a plant's elevated PFOS traces to three platers and a paper mill, the cheapest fix is a pretreatment permit condition, not a dryer. But it only works where the industrial user exists and the utility has pretreatment authority to reach it.

Which is the caveat that matters most. A one-MGD plant in a farming county with no significant industrial users has nothing to source-control; its PFAS arrives from consumer products and firefighting foam legacy, diffuse and unattributable, and no monitoring program will change that. Neither will the memo's recommendations, which assume an industrial pretreatment program that thousands of small systems simply don't have. Those utilities get all of the siting restrictions and none of the tools, and they are the ones for whom this becomes a rate case. Small systems in states with weak biosolids programs may also see no change at all for several years, which is its own planning hazard, because the change arrives later and all at once.

Which endpoint fits which utility

There's no winner here, only a fit. A mid-size plant with identifiable industrial dischargers and pretreatment authority should spend the next eighteen months on source identification and keep land application alive, because that's the cheapest ton it will ever move and the memo's recommendations are survivable when the numbers are low. A plant whose sludge is already elevated and whose land base is dominated by row crops near surface water should assume its acreage shrinks and start the thermal permitting conversation now, since the schedule risk dwarfs the capital risk. A coastal or urban utility with no land base at all is really choosing between landfill contract terms and a share of somebody's combustion capacity, and it should be negotiating both simultaneously rather than sequentially.

Everyone should be sampling. Not because the memo says to, but because when your state adopts a number, and it will, your negotiating position depends entirely on whether you have three years of your own data or none. I've sat on the wrong side of that table with clients who had no baseline, and there's no argument you can make from an empty file.

Utilities looking at this as a systems problem rather than a disposal problem tend to land in better shape, which is the framing behind most credible zero-waste-to-landfill solutions for municipal residuals: fewer single points of failure, more measurement, contracts that assume the rules will move. The ones that get hurt are the ones whose entire biosolids strategy is a single handshake with a single hauler.

The memo doesn't have the force of law. What it has is a plain-English list of things a county attorney can paste into next year's spreading agreement, and county attorneys read faster than courts do.

Disclosure: Renewable Waste Energy advises developers and municipalities on permitting and thermal treatment pathways for wastewater residuals.

Sources & Notes

• Dates, docket number, and the non-binding language come straight from the Federal Register notice published July 6, 2026. The memo itself is signed June 29 and comments close September 4.
• The recommendation set, and the criticism of the January 2025 draft risk assessment, are summarized in EPA's announcement of the draft guidance.
• Generation and disposal split for 2024: EPA's basic information page on sewage sludge and biosolids. Note the coverage caveat discussed above.
• For the tiered thresholds and the pretreatment results, see Michigan EGLE's interim strategy for land application of biosolids containing PFAS.
• Maine's post-ban handling costs and the Juniper Ridge capacity squeeze were reported by Treatment Plant Operator in March 2023. The 2024 air permit review and the 2023 Title V matter are from my own consulting files and are described without identifying the clients.