Federal Appeals Court Upholds EPA Approval of Florida Road Built With Radioactive Fertilizer Waste

A federal appeals court has upheld the Environmental Protection Agency's approval of a Florida pilot project that used phosphogypsum, the radioactive byproduct of phosphate fertilizer production, as a road construction material, rejecting a challenge brought by the Center for Biological Diversity.

A three-judge panel of the U.S. Court of Appeals for the Eleventh Circuit ruled unanimously on August 12 that the EPA did not violate its own regulations when it granted Mosaic Fertilizer a case-by-case approval to build experimental roads on the company's private land in Florida.

The decision matters well beyond the specific stretch of road at issue. Florida holds the largest concentration of phosphogypsum in the United States, stacked in enormous piles known as gypstacks across the phosphate mining region of Central Florida, and the question of whether that material can be put to beneficial use rather than stored indefinitely has been contested for decades.

What the court decided

The panel held that the EPA's regulations unambiguously permit the agency to grant case-by-case approval for other uses of phosphogypsum, including road construction, provided that health protections are satisfied. That reading of the regulatory text was the core of the ruling.

On the health question, the court pointed to the EPA's risk estimates. The agency calculated cancer risk at less than 1 in 1 million for nearby residents and facility workers, and at less than 2 in 1 million for construction workers handling the material. The court noted that even the highest of those estimates was at least 45 times lower than the threshold the environmental group itself had proposed as acceptable.

That comparison did substantial work in the opinion. Courts reviewing agency risk determinations generally defer to technical judgments within the agency's expertise, and a petitioner arguing that a risk is unacceptable has a difficult path when the agency's estimate falls well below the petitioner's own stated standard.

The EPA approved Mosaic's application in December 2024. The roads were laid in November 2025, beginning an 18-month experiment, and the agency's approval requires Mosaic to monitor groundwater for at least 18 months after construction.

What phosphogypsum is and why it is stacked

Phosphogypsum is produced when phosphate rock is processed into fertilizer using sulfuric acid. For every ton of phosphoric acid produced, the process generates several tons of phosphogypsum, and the material concentrates naturally occurring radium from the original ore.

Because of that radium content and its decay product radon, the EPA has regulated phosphogypsum under the Clean Air Act's radionuclide provisions since 1989, generally requiring that it be stored in stacks rather than used in commerce. Limited exceptions have existed for agricultural and research uses.

Florida's phosphate industry, concentrated in Polk, Hillsborough, Manatee, Hardee and surrounding counties, has produced an enormous inventory. The gypstacks that hold it are among the largest engineered structures in the state, and they carry their own management burden: the process water ponds atop them must be maintained, and stack failures have caused significant environmental incidents in Florida's past.

The Piney Point facility in Manatee County is the case that shapes Florida's public understanding of the issue. Its stack system produced repeated discharges of nutrient-laden process water into Tampa Bay, most consequentially in 2021, and the state has since been working through a costly closure.

The argument for beneficial use

The industry case for road use is that the material has engineering properties suited to base construction, that using it reduces the volume that must be stored perpetually, and that other countries have permitted similar applications. Reducing stack inventory addresses a long-term liability that currently has no endpoint.

The pilot structure reflects the regulatory caution around that argument. Mosaic's approval covers roads on the company's own private land rather than public highways, the experiment runs for a defined period, and groundwater monitoring is required for at least 18 months after construction.

What the pilot is designed to test is whether radon emanation and leaching behave as models predict once the material is placed in a road base and exposed to Florida's rainfall and water table conditions. Modeling and field performance do not always agree, which is the point of a field experiment.

The argument against

The Center for Biological Diversity's challenge rested on the position that a radioactive waste stream regulated precisely because of its radium content should not be distributed into the environment in dispersed form, and that once phosphogypsum enters road construction, it becomes practically impossible to retrieve.

Environmental groups have also argued that a pilot on private land functions as a precedent for broader authorization. If the 18-month experiment produces acceptable monitoring results, the pathway to approval for public road use becomes shorter, and that eventual decision is the one with real scale consequences.

Florida's hydrogeology sharpens the concern. Much of the state sits on karst limestone with a shallow water table and rapid connection between surface and aquifer, and the Floridan aquifer supplies drinking water to a large share of the state's population. Contaminant transport behaves differently in that setting than in the geology where much U.S. groundwater modeling was developed.

What it means for Floridians

Nothing changes immediately for anyone driving Florida roads. The approved project covers private roads on Mosaic property, and the experiment is running under monitoring requirements. Public road construction with phosphogypsum would require separate approvals.

The longer-term stake is the future of the gypstacks. Florida currently carries a permanent storage obligation for a material inventory measured in the hundreds of millions of tons, with the costs of stack maintenance and eventual closure falling on operators and, when operators fail, on the state.

Communities in the phosphate district have the most direct interest. Polk, Hillsborough, Manatee and Hardee counties host both the mining operations and the stacks, and residents there have been the constituency most engaged in phosphogypsum policy over the past several decades.

Florida's phosphate industry

Phosphate mining has operated in Florida since the late 19th century, and the state has been the dominant U.S. source of the mineral for most of that period. The Bone Valley formation, spanning Polk, Hillsborough, Hardee, Manatee and DeSoto counties, contains deposits that supplied a large share of global fertilizer production during the 20th century.

The industry remains a significant employer in Central Florida, supporting mining operations, processing plants, rail transportation and port export facilities. Mosaic is the largest operator, formed through consolidation of earlier phosphate companies, and its Florida operations anchor its global fertilizer business.

Land reclamation is a regulated obligation. Florida requires mined land to be restored, and reclaimed phosphate land across Central Florida has been converted to pasture, wetlands, residential development and recreational use. The quality and ecological function of that reclamation has been the subject of long-running dispute.

Phosphogypsum is the byproduct that reclamation cannot address, because it is generated by chemical processing rather than by mining and because regulation requires it to be stacked rather than dispersed. That constraint is the origin of the beneficial use question the court addressed.

What the gypstacks require

A phosphogypsum stack is an engineered structure holding both solid material and a pond of acidic process water on top. Managing that water is the operational challenge, because rainfall adds to the pond volume and the water cannot simply be released.

Stack failures have consequences that Florida has experienced directly. The Piney Point facility in Manatee County produced repeated discharges into Tampa Bay, most consequentially in 2021 when an imminent breach forced emergency releases of hundreds of millions of gallons of nutrient-rich water and prompted evacuations in surrounding areas.

The state has been funding a closure of Piney Point involving deep well injection of process water and permanent capping of the stack, a project running into the hundreds of millions of dollars. That cost fell to the state because the facility's prior operators were no longer able to bear it.

Active stacks operated by solvent companies carry financial assurance requirements intended to prevent a repeat of that outcome. Whether those assurances are adequate for structures that require management in perpetuity is a question that regulators revisit periodically.

The regulatory history of beneficial use

The EPA's 1989 rule establishing the general prohibition on phosphogypsum distribution was issued under Clean Air Act authority addressing radionuclide emissions. The rule allowed limited exceptions, including for agricultural use on soils meeting specified conditions and for research and development.

The agency approved a road use application in 2020 and subsequently withdrew that approval following litigation and procedural challenges, an episode that shaped how the current application was structured. The 2024 approval that the Eleventh Circuit reviewed followed a more developed administrative record.

Other countries have permitted phosphogypsum in construction applications, and the industry cites that international practice as evidence that the material can be used safely. Critics note that radium content varies by ore source and that experience elsewhere does not automatically transfer to Florida's specific material or hydrogeology.

The pilot structure represents a compromise between those positions: limited scope, private land, defined duration and mandatory monitoring, with the resulting data intended to inform whatever comes next rather than to settle the question by assertion.

What's next

The Center for Biological Diversity may seek further review, either by requesting rehearing before the full Eleventh Circuit or by petitioning the U.S. Supreme Court. Neither route is likely to succeed as a statistical matter, and the panel's unanimity reduces the prospects further.

The pilot's monitoring data will accumulate over the 18-month period following the November 2025 construction. Those groundwater results are the substantive information that will inform any future regulatory decision, and they will become available through the EPA's oversight of the approval.

The broader policy question remains open in Congress and at the agency. Legislative proposals to authorize phosphogypsum in road construction have been introduced in past sessions, and the outcome of this pilot is likely to feature in future rounds of that debate.