Last Friday, in American Farm Bureau Federation v. EPA, Judge Sylvia Rambo upheld EPA’s Chesapeake Bay TMDL. As Judge Rambo noted in her conclusion, while the environmental problems associated with the Chesapeake Bay are massive and the issues complicated, her review was not that difficult.
Notwithstanding the expansive administrative record, and the complexity of the numerous issues implicated herein, the court’s scope of review in this case is relatively narrow. In accordance with the deferential standards applicable to a court’s review of an agency’s actions, this court must give EPA’s interpretation of the CWA and its use of scientific models and data due deference in light of EPA’s scientific and technical expertise.
Even so, the decision does contain some specific holdings that are worth noting. These include:
EPA had authority to separate the TMDL into Waste Load Allocations for point sources and Load Allocations for non-point sources. (This perhaps should not be surprising, since EPA has done so countless times, but it was apparently an issue of judicial first impression.)
EPA’s allocations for specific sectors and individual sources was appropriate and did not constitute improper intrusion into the states’ role in implementing the TMDL.
EPA had authority to “backstop” the individual states’ decisions about parts of the TMDL within their jurisdiction, where EPA found that the state decisions were not sufficiently protective.
EPA could require states to provide “reasonable assurances” that water quality standards will be met and load allocations will be achieved. Here too, the court rejected a challenge which argued that requiring such assurances intruded on the states’ role in implementing the TMDL.
The TMDL does not impose binding allocations on individual sources. The Court found that states can propose modifications to EPA. “WLAs are not permit limits per se; rather they still require translation into permit limits….” The Court also noted that the TMDL provides for effluent trading systems, thus providing yet more flexibility.
EPA has authority to issue allocations to states upstream of the Bay.
This watershed-wide approach also appears to be consistent, if not specifically authorized by CWA Section 303(d), which requires TMDLs to be established for impaired waters “at a level necessary to implement the applicable water quality standards.”
Finally, EPA had discretion to use the models that it did and rely on the data that it did.
The court must defer to EPA’s use of data, even if that data is imperfect, unless the data bears no rational relationship to the reality it purports to represent.
At bottom, as the court noted:
cooperative federalism can be, at times, messy and cumbersome. It is unavoidable that states and the federal government will occasionally disagree. EPA worked with the states to ensure that the proposed allocations were sufficient to achieve water quality standards. Complete unanimity between the states and EPA in resolving all the complex issues involved here is likely impossible. Disagreements between the states and the federal government regarding some of the allocations necessary to achieve water quality standards was to be expected, and the debate and discussions that ensued were of nature that is required in a cooperative federalism scheme. Moreover, although Plaintiffs believe that this process was coercive, it is noteworthy that no state has filed suit challenging the TMDL, let alone alleged that their participation in the TMDL drafting process was a result of coercion.
Hard to disagree with that. The lesson for practitioners? The messier it is, the more discretion EPA probably will receive from the courts.