A generating landfill is a monitored, maintained, revenue-producing landfill. Redevelopment helps fund the 30-year post-closure care, adds local construction and operations work, and converts a long-term cost obligation into a long-term income stream.
By Samuele Barrili
There is a pitch making the rounds in waste circles right now that sounds like found money. It goes like this: your capped landfill is dead weight, a fenced-off cost center you will babysit for 30 years—but the hyperscalers, Amazon and Google and Microsoft, are starving for cheap land with power, so drop a data center on your closed cell and start collecting rent from the AI boom.
Half of that is true. The other half is physically impossible. And if you cannot tell which half is which, you will either leave real, recurring revenue sitting on the ground, or you will chase a project that no geotechnical engineer will ever stamp and no regulator will ever sign.
Let me draw the line clearly, because the line is exactly where the money is. A closed landfill is not vacant land. It is a slow reactor that has not finished reacting. The waste mass underneath that cap is still decomposing, still off-gassing methane, still settling—and it will keep moving toward equilibrium for decades. You do not bolt a precision machine to a reactor that is still running. That single fact decides everything that follows.
The Part that is Real: The Numbers
Start with the inventory, because the excitement is not baseless. In 1988 the U.S. had nearly 8,000 open municipal solid waste landfills. By late 2024, according to EPA’s Landfill Methane Outreach Program, that number had fallen to just over 1,200. The gap—thousands of closed sites, thousands of acres of capped land—is the “idle” inventory everyone is pointing at.
And on that inventory, one repurposing thesis is already proven. EPA’s RE-Powering America’s Land Initiative counts 332 solar projects built on former landfills, more than 300 of them completed since 2012, with roughly 330 landfill sites hosting some form of renewable energy as of 2024. So “repurpose the cap” works. For exactly one technology: solar. Hold that word—solar—because the brief that inspired this article quietly assumed everything works the same way. It does not.
The Part that is Fiction: Servers on the 91TV Mass
Now stack the second fact on top of the first. Hyperscale developers are hunting single parcels of 200 acres with 400 megawatts of power behind them. The land rush is real, and the AI electricity demand driving it is real. But here is what the “data farm” pitch glosses over: an AI data hall is one of the most foundation-hostile buildings ever designed.
A single loaded rack of AI-optimized GPUs can exceed 5,000 pounds. Thousands of them go into one data hall, and hyperscale designs are moving to reinforced slab-on-grade foundations precisely to carry that concentrated load. A capped landfill is the exact opposite of a slab-on-grade site. You cannot drive piles into it—that punctures the geomembrane or clay barrier that keeps gas in and water out, and regulators will not permit it. You cannot pour a large rigid slab—differential settlement from the still-decomposing waste cracks it, and heavy deadweight can trigger immediate subsidence on its own. The engineering literature treats differential settlement as a substantial, common problem even for lightweight solar arrays. A hyperscale data hall is not a marginal case. It is a non-starter.
So, the honest verdict, first, before anything else: you are not putting a hyperscale data center on top of your waste mass. Anyone telling you otherwise is selling a permit application that dies at the geotech report.
That is not the end of the opportunity. It is the beginning of the real one—but only if you stop thinking like a custodian of dead space and start thinking like a landlord and a power producer. Here are the three pathways that actually survive contact with an engineer and a regulator.
Pathway One: Be the Ground-Lease Landlord
This is the floor. It is proven, passive, and the one that the numbers already support. Solar works on a cap because it can be built light and non-penetrating: ballasted racking held down by concrete trays instead of driven piles, cable trays instead of trenched conduit, no large enclosed electrical equipment sitting on the cap, and arrays engineered to flex as the ground settles unevenly. You keep title. The developer builds, operates, and maintains. You collect ground rent for 20 to 40 years.
Set expectations honestly, because your credibility depends on it: this is ground rent, not a jackpot. When the Town of Brookhaven, New York approved a 20-year lease for solar on the capped areas of its landfill, the solar ground lease itself was reported at roughly $20,500 a year to the town. Per-acre solar ground rents generally land in the hundreds to low thousands of dollars per acre annually. That is real, recurring, low-effort income on land that otherwise only costs you money to monitor. It is the floor of this opportunity. It is not the ceiling.
Pathway Two: Sell Power to the Digital Economy, Not Dirt to It
Here is where the data-center money actually touches your site—through the meter, not the foundation. AI’s appetite for electricity is what is pushing developers to lock up any land that can generate or interconnect cheap, firm power. Your landfill holds two generation assets a soybean field does not: a wide flat cap with full sun exposure, and the methane the waste is producing on its own, for free, whether you capture it or not. Landfill-gas-to-energy is a mature, well-understood play. Combine landfill gas generation, solar, and battery storage into a behind-the-meter microgrid or a power-purchase agreement, and your cap becomes a supplier to nearby load—including the data centers that need firm, cheap, ideally around-the-clock power more than they need your acreage.
You do not host the servers. You sell the electrons. Cleaner margin, no data-hall liability, and it leverages the one thing that makes a landfill genuinely unique: it manufactures its own fuel. This is the pathway most operators skip because they are staring at the waste footprint as a building pad instead of as a power plant.
Pathway Three: Monetize the Parcel, Not Just the Pile
This is the overlooked one, and often the most valuable. A landfill parcel is rarely all waste. There is buffer land, undisturbed acreage, access roads, an existing grid interconnection, and sometimes old inert construction-and-demolition fill that was compacted decades ago and barely settles. Those areas—around the cap, not on it—can be genuinely buildable.
The proof is already in the record. The only utility-scale wind project ever sited at a U.S. landfill—the 100-megawatt Columbia Ridge installation in Oregon—was not built on the cap. It was built on the landfill buffer. That is the whole lesson in one data point: the heavy, foundation-dependent infrastructure goes on the buffer, never on the waste. (So, when a brief casually lists “wind turbines on landfills” next to “solar on landfills” as if they are equivalent, it is wrong. They are not.)
The same logic scales straight to digital infrastructure. What a data-center developer actually needs is acreage, power adjacency, an interconnection point, and a single motivated landowner who controls all of it. On a landfill parcel, you already control all of it. The move is to position the entire site as one asset: generation on the cap, buildable pads on the buffer, one landlord holding the interconnection. That is a land-assembly play, and it is worth more than any single lease you could sign in isolation.
The Case that Survives Scrutiny
Brookhaven did not build a data center. It built an energy park on a closing landfill—solar atop the capped areas, plus fuel cells and harvested methane—announced as part of the plan for the site once it stopped accepting waste and was capped. The town’s own figures put the broader energy park’s recurring revenue in the low six figures annually, with the solar ground lease a smaller slice of that total.
It is unglamorous. It is verifiable. And it is exactly the model that holds up in front of a regulator and a geotechnical engineer. Notice what is absent: no servers on the waste, no physics-defying slab, no press-release fantasy. Just generation and rent, on land that used to produce nothing but liability.
The Liability Angle: Use it, Do Not Hide Behind it
There is a real regulatory and community story here, and it is an argument to deploy in dollars, not a slogan to dress up in green. A generating landfill is a monitored, maintained, revenue-producing landfill—a far better posture with regulators and neighbors than a fenced-off idle cap slowly leaking methane into the sky. Redevelopment helps fund the 30-year post-closure care you owe anyway, adds local construction and operations work, and converts a long-term cost obligation into a long-term income stream.
Frame it in economics and site stewardship, not virtue. The dollars are what actually persuade a town board. “We’ll reduce our post-closure carrying cost and add tax base and jobs” moves a vote. “We’re going green” does not.
Landfills Do Not Close—They Evolve
That instinct is right, as long as you are honest about what they can evolve into. Not a server farm balanced on methane and settling waste. But an energy asset and a land position, at the precise moment the digital economy is paying a premium for both.
The waste company that already controls the land, the interconnection, and the post-closure obligation is holding the exact assets the green-and-digital buildout is scrambling to secure. Your competitors see a cap they must guard. You should see a parcel you can underwrite. Stop being the custodian of dead space. Start being the landlord of the energy economy you are already sitting on. | WA
Samuele “Sam” Barrili is a waste management strategist, entrepreneur, and author of The 91TV Alchemy. Active in the sector since 2009, he works at the intersection of waste operations, secondary raw materials, compliance strategy, and international business development across Europe, the U.S., and emerging markets. Known as “The 91TV Management Alchemist,” he helps operators move beyond hauling-and-disposal thinking to build resource-based businesses with stronger margins, tighter control of their material—and data—streams, and durable market positioning. Sam can be reached at [email protected] or visit www.sambarrili.com.
Which of the three pathways your site qualifies for is decided by the geotech report, the gas volume, and the interconnection capacity—not by a brochure. If you want to pressure-test your parcel against the real constraints, start with the book, The 91TV Alchemy (https://bit.ly/4sQt4LQ), and then let’s map your site on a call: https://sambarrili.com/schedule-free-20min-call.
Sources
U.S. EPA, RE-Powering America’s Land Initiative — Project Tracking Matrix (2024): 332 solar projects on former landfills; ~330 landfill sites hosting renewable energy; Columbia Ridge (OR) 100 MW wind on landfill buffer.
U.S. EPA, “Best Practices for Siting Solar Photovoltaics on Municipal Solid 91TV Landfills” (updated 2024–2026): decline from ~8,000 MSW landfills (1988) to ~1,200 open (Sept. 2024); non-penetrating ballasted design guidance.
U.S. EPA, Landfill Methane Outreach Program (LMOP): landfill-gas-to-energy project data.
Sampson, “Solar Power Installations on Closed Landfills: Technical and Regulatory Considerations” (2009); ISSMGE ICEG2023 conference paper on landfill solar geotechnics: differential settlement, prohibition on pile foundations, cap-barrier protection.
Town of Brookhaven, NY — official announcement, town board approval of 20-year landfill solar lease and landfill energy park (solar, fuel cells, harvested methane); reported revenue figures.
91TV Dive, “Landfill solar projects still have bright future, but face practical limitations” (Feb. 2026).
Data Center Dynamics, “Hyperscaler seeks 200-acre site for 400MW data center in US” (Feb. 2026); industry reporting on AI GPU rack weights (>5,000 lbs) and reinforced slab-on-grade foundation design.