Listing Impaired Waters under the TMDL Program

Hopping Green Sams & Smith, P.A.
Tallahassee, Florida

The Total Maximum Daily Load ("TMDL") program has been part of the regulatory landscape since the inception of the modern Clean Water Act ("CWA") in 1972. It lay dormant, however, as regulatory priorities were directed elsewhere. In the mid-1990s, spurred by citizen suits, courts began directing the U.S. Environmental Protection Agency ("EPA") to implement the TMDL program where states have failed to do so. States, in turn, have sprung into action to stay one step ahead of EPA in listing impaired waters and developing TMDLs within their jurisdictions. This article examines the current efforts by the states – especially Florida – to establish methodologies for developing lists of impaired waters for which TMDLs will be developed.

The TMDL Program

The TMDL program was a relatively inconsequential feature of surface water regulation for over two decades. Compared to other pressing responsibilities under the 1972 CWA – such as promulgating effluent limitation guidelines and implementing the National Pollutant Discharge Elimination System ("NPDES") permit program – EPA initially perceived the TMDL program as "an evolving and time consuming process which does not lend itself to imposition of a deadline date." 43 Fed. Reg. 60662, 60664 (Dec. 28, 1978). EPA reasoned that missing TMDL deadlines would be no big deal because "[d]evelopment of TMDLs . . . is not a necessary prerequisite to adoption or enforcement of water quality standards. . . ." Id. The states generally followed EPA’s lead. For example, in 1989, Florida’s Department of Environmental Protection ("FDEP") informed EPA that TMDLs are a "lesser priority"; the state agency observed that "[o]bviously, EPA also felt that TMDLs were not a priority. . . ." March 21, 1989, Rhodes (FDEP) letter to McGhee (EPA). The states did list impaired waters from time to time, but seldom promulgated – much less allocated and implemented – TMDLs. Accordingly, the regulated community typically displayed minimal interest in the states’ listing decisions.

As is now well known, in the 1990s, several courts were harshly critical of the agencies’ laid back regulatory perspective. See, e.g., Alaska Center for the Environment v. Reilly, 796 F. Supp. 1374, 1379 (W.D. Wash. 1992) ("The only ‘consistently held interpretation’ that the EPA has demonstrated with respect to the CWA’s TMDL requirements has been to ignore them.") The negative reinforcement associated with 40 lawsuits in 38 states, coupled with an emerging perception that the TMDL program represents a promising launching pad for new initiatives (pertaining to non-point source regulation, air deposition, and regulation of mercury emissions) appears to have resulted in an attitude adjustment: the federal agency now enthusiastically embraces the TMDL program, as reflected in EPA’s expansive, newly promulgated TMDL regulations (65 Fed. Reg. 43586, July 13, 2000).

Now that it is on the front burner, the state agencies are struggling to meet the demands of the TMDL program. The difficulty of their task is exacerbated by short deadlines imposed by consent decrees and the complexity generated by the additional overlay of new federal regulations (to go into effect later in 2001) as well as emerging EPA guidance documents. And now that allocation and implementation of TMDLs are no longer merely abstract possibilities, the regulated community has been looking more closely at the states’ impaired waters listing decisions.

How It Works

In general, and without the many qualifications required for complete accuracy, TMDLs establish the maximum levels of pollution that a water body can assimilate, after application of technology-based effluent limitations, without exceeding water quality standards. All states are expected to:

• list and prioritize the impaired waters,
• establish TMDLs for those waters,
• allocate pollution loading responsibilities, and
• implement the allocations.

CWA § 303(d). EPA looks over each state’s shoulder along the way and intervenes if it decides that a state has not fulfilled these tasks.

Special Challenges in Listing Waters

The first step in the TMDL process – listing impaired waters – can be very problematic. In many circumstances it is difficult to reliably identify and quantify water quality impairment. Even with regard to objective, numerical water quality criteria – such as the dissolved oxygen criterion of 5.0 mg/l that is found in many states – it is often hard to determine what constitutes impairment. Depending on numerous factors, including the season of the year or even the time of day, many waterbodies routinely and naturally feature dissolved oxygen levels below 5.0 mg/l. To further complicate listing decisions, the validity of some of the data that states typically rely on is being called into question. Agency officials have openly admitted that some historical water quality data for certain metals are particularly unreliable due to problems in field collection or laboratory analysis techniques. Inherent difficulties in determining what constitutes impairment become especially problematic when state agencies consider narrative water quality standards, such as the typical state standard for nutrients. For example, what exactly does it mean to "cause an imbalance in natural populations of aquatic flora and fauna?"

Neither EPA nor the states appear to have the resources required for the level of monitoring and surveillance required to get the impairment determinations right in many cases. Accordingly, EPA’s apparent game plan is for state agencies to simply do the best they can with the resources available, erring as necessary in favor of the environment. See Houck, "TMDLs IV: The Final Frontier," 29 ELR 10469, 10474, 10476 ("Pollution control systems based on ambient standards have always relied more on science than science can deliver;" EPA’s guidance on listing impaired waters tends to be "all inclusive" and "to err on the side of listing.") EPA’s benchmarks for the quality of data to be considered in impaired water listing decisions are that the data "exist" and be "readily available." 40 CFR § 130.7(a)(5).

Requiring Credible Data

This over-inclusiveness in listing waters as impaired under the TMDL program has been questioned on both policy and legal grounds. In response, a growing number of state laws require the state environmental agency to develop a scientifically-based methodology for determining whether waters truly are impaired and mandate that the agency provide an opportunity to contest listing decisions under the state’s Administrative Procedures Act ("APA"). Florida’s 1999 legislation, reportedly the "first" and "most expansive" of the state laws (Inside EPA Water Policy Report, May 24, 2000, p. 8), contains the following features:

• A pre-existing list of impaired waters that FDEP submitted to EPA in 1998 is characterized as a list of waters to be assessed for impairment; "the [1998] list, priority ranking, and schedule cannot be used in the administration or implementation of any regulatory program."
• FDEP must "adopt by rule a methodology for determining those waters which are impaired." Impairment designations must be based on "objective and credible data."
• Using the new methodology, FDEP must issue a "subsequent, updated list of those water bodies or segments for which total maximum daily loads will be calculated." FDEP orders listing impaired waters will be subject to challenge under Florida’s APA.

§ 403.067, Fla. Stat. Montana, Idaho, Indiana, Arizona, and Wyoming also have enacted bills or rules that require "credible" data to be considered in making listing decisions.

One policy rationale underlying such laws is that the limited resources available to implement the TMDL program should be focused where the environmental benefit is the greatest. Many states have hundreds of impaired water bodies or segments on 303(d) lists that predate the promulgation of a listing methodology or meaningful attention to data quality. Unless states make sure that listed waters truly are impaired there is a substantial possibility that states will spend significant sums on waters that are not in fact impaired, thereby diverting resources from waters that do require further regulatory attention. Stated another way, erring in favor of listing waters is not necessarily erring in favor of the environment.

Criticism of the Requirement for Credible Data

Not everybody has been receptive to this rationale. A prominent environmental activist in Florida was quoted as characterizing FDEP’s draft listing methodology rule as follows: "It will simply assure that most of Florida’s polluted waters will never be protected and restored to an unpolluted condition." St. Petersburg Times, November 6, 2000.

EPA weighed in on this issue via a May 18, 2000 letter sent to its "State Water Quality Colleagues" in various states. Characterizing "credible data legislation" as a "troubling trend," EPA’s letter emphasized that state listing decisions should be based on consideration of "all existing and readily available water-quality related data and information." EPA warned the "proponents of credible data legislation" that they could be facing "EPA’s disapproval of the list."

Later, EPA Region IV submitted approximately twenty pages of detailed comments in response to a preliminary draft of FDEP’s listing methodology rule. EPA expressed concern that the Florida program would result in a "highly conservative" list of impaired waters because "available data" that is not of "high enough quality" (under Florida’s methodology rule) would be excluded from consideration. Banister (EPA) letter to Brooks (FDEP), Sept. 5, 2000. EPA criticized FDEP draft rule language that would exclude consideration of data more than five years old and draft rule language that would require a minimum of ten samples before assessing impairment. Id. EPA asserted, "It is reasonable to expect that anecdotal and other types of information would be appropriate and significant sources of information in determining whether narrative nutrient criteria are being met." Id. EPA also criticized the provision in Florida’s 1999 legislation that characterizes Florida’s 1998 list as merely waters to "be assessed":

We continue to find the most troubling aspect of the draft rule to be the numerous provisions that would allow the State to remove from its current approved 1998 303(d) list many waters that fail to meet data sufficiency provisions as outlined in the proposed rule . . . . In short, by setting a high bar for listing waters, many waters that do not reach that bar would in effect be delisted from the approved list. Since this approach does not meet the "good cause" requirements set out in the Clean Water Act (CWA) regulations for delisting waters, any list created following this process would be inconsistent with the CWA. This would lead to EPA’s disapproval of a future State 303(d) list, and a subsequent requirement on EPA to establish a 303(d) list for the State.

Banister (EPA) letter to Brooks (FDEP), Sept. 22, 2000.

EPA perceived the 1998 list as Florida’s "approved 303(d) list" that can be altered only in accordance with formal delisting procedures. One could argue, perhaps, that EPA committed itself to this position in a June, 1999 Consent Decree in Florida Wildlife Federation v. Browner, 4:98CV356-WS (N.D. Fla. 1999), wherein EPA agreed to promulgate TMDLs for most waters on the 1998 list if FDEP fails to do so by prescribed deadlines that extend through 2011. Id. On the other hand, one could argue that it was inappropriate for the June, 1999 Consent Decree in essence to implement FDEP’s 1998 impaired waters list, given that the Florida Legislature – on account of the procedural and substantive shortcomings of the 1998 list – had declared just a few months earlier that the 1998 list "cannot be used in the administration or implementation of any regulatory program." § 403.067(2), Fla. Stat. Also, because Florida’s 1998 list was not signed by the Secretary of FDEP and had not been subjected to Florida’s relatively expansive APA, it is arguable that Florida’s 1998 submittal was never really a valid submittal under Clean Water Act § 303(d).

Resolving the State/Federal Tension

Florida’s Legislature has directed FDEP to base its listing methodology on "objective and credible" data, while EPA prefers a methodology that utilizes all data that is "readily available." Clearly, "objective and credible" data will comprise a smaller subset of all "readily available" data, so the two standards are in tension. Perhaps EPA’s position will evolve to some extent in the context of its ongoing development of Consolidated Assessment and Listing Methodology ("CALM") guidance, which "is intended to help states . . . develop and document a consolidated and scientifically defensible approach for assessing water quality, developing lists of impaired waters under Section 303(d). . . and preparing water quality inventories under Section 305(b). . . ."

FDEP currently intends to promulgate a rule finalizing its listing methodology in late winter of 2000 or early spring of 2001. The most recent FDEP draft methodology rule (dated November 30, 2000) can be found at www.dep.state.fl.us/water/division/tmdl/

ruleworkshop.htm/. Recent indications are that EPA and FDEP have resolved many of the issues raised in EPA’s comment letters. Whether FDEP, EPA, environmental groups, and the regulated community will find common ground on all methodology issues remains to be seen.