Mr. Chairman and Members of the Committee:
My name is Robert W. Adler. I am a Professor of Law and Interim Director of the Wallace Stegner Center for Land, Resources and the Environment at the University of Utah College of Law. I submit this testimony an individual(1) who has had a longstanding interest and involvement in the proper implementation of the Clean Water Act. Over the past twenty years I have represented a state government, environmental organizations and private individuals and businesses in matters involving implementation of the NPDES and water quality standards programs, as well as other aspects of the CWA. I have written extensively about Clean Water Act (CWA) law and policy, including a book on the history and effectiveness of the Act,(2) a recent law review article comparing the TMDL program with the similar State Implementation Plan (SIP) program under the Clean Air Act,(3) and a large number of other articles and book chapters regarding water quality and watershed protection issues. I was a member of EPA's Federal Advisory Committee Act (FACA) Committee on TMDLs. Finally, over the past year I have also participated as one of the principle researchers in a study of innovations in environmental programs being conducted for Congress by the National Academy of Public Administration (NAPA), some of the preliminary results(4) of which are relevant to the issue being studied by this Committee. A copy of my full resume is attached.
General Comments
The Committee requested testimony on the impact on agriculture and forestry of the Administration's proposed regulations. An equally important and integrally related question, however, is the impact of U.S. agriculture and forestry on human health and on the health of our aquatic ecosystems, and the long history of prior efforts to address those impacts. I intend to provide some perspective on that issue as well, and to explain why the answer to the second question is critical to a proper analysis of the first.
Representatives of the agriculture and forestry industries fear that EPA's proposed regulations, if adopted, will result in economic impacts to their members. I have absolutely no doubt they are correct. The proposed regulations will clearly change the manner in which the industry must address the environmental impacts of their operations. In some cases, operators will face increased costs of doing business, just as all other major U.S. industries have been asked to incur reasonable increased costs in order to do their fair share to protect the waterways we all rely on for drinking water, recreation, fish and wildlife and other environmental resources. It does not necessarily follow, however, that the net result of the regulations ultimately be to the detriment of those critically important sectors of our economy. First, in some cases the types of operational changes necessary to reduce or eliminate the environmental impacts of agriculture can benefit operators economically by decreasing annual input costs and increasing annual yields.(5) Second, and more importantly for the issue before this Committee, by increasing the efficiency with which both public and private resources are dedicated to agricultural and silvicultural pollution, EPA's proposed regulatory changes have strong potential to benefit both the environment and our agriculture and forestry industries. They also could help to ensure that the agricultural pollution control assistance programs conducted under the auspices of this Committee are conducted in a smarter, more cost-effective way.
To explain that position, however, I want to begin by answering my second question, the impact of U.S. agriculture on water quality and aquatic ecosystem health, from both an historical and a current perspective. Of course, this legislative body has been well aware of the extensive environmental impacts caused by agricultural runoff and other forms of nonpoint source pollution for at least three decades. The Senate Committee Report accompanying the 1972 amendments to the Clean Water Act said:
One of the most significant aspects of this year's hearings on the pending legislation was the information presented on the degree to which nonpoint sources contribute to water pollution. Agricultural runoff, animal wastes, soil erosion, fertilizers, pesticides and other farm chemicals that are part of runoff ... are major contributors to the Nation's water pollution problem. Little has been done to control this major source of pollution . It has been clearly established that the waters of the Nation cannot be restored and their quality maintained until this very complex and difficult problem of nonpoint sources is addressed .... However, there are many programs that can be applied to each of the categories of nonpoint sources, and the Committee expects that these controls will be applied as soon as possible.(6)
Subsequent research confirmed the findings of the 1972 Senate Report, even many years after the Act was passed. In 1991, EPA released a report entitled Managing Nonpoint Source Pollution, based on individual reports submitted by the states, summarizing ongoing environmental threats caused by runoff pollution. EPA's findings included the following:
Similar results were reported by the states in their lists of impaired waters submitted under section 304(l) of the Act, added in 1987. Of the more than 17,000 water bodies identified as seriously degraded, only about 600 were identified as polluted "wholly or substantially" by point sources alone. Runoff pollution contributed to the impairment of the remaining waters, over 16,000 altogether. Major causes included soil erosion from row crops, logging and grazing; pesticide pollution; eutrophication due to runoff of excess phosphorus and nitrogen fertilizers; pathogens from confined livestock facilities; discharges of salts and metals from irrigation return flows; and degradation of riparian habitat from a range of farming and logging operations.
Even more disturbing, however, is how little the data have changed in yet another decade. Despite efforts to control agricultural and silvicultural water pollution under sections 208 and 319 of the CWA, as well as numerous programs undertaken under various Farm Bill and other agricultural assistance programs, agriculture remains the leading single source of water pollution in the country. In its most recent national assessment, EPA identified agriculture as the number one cause of impairment of rivers and lakes, and the fifth leading source of estuarine pollution.(7) For those waters adequately surveyed by the states, agriculture was identified as a leading source of impairment of 25 percent of assessed river miles, nearly 20 percent of assessed lake acres, and over 10 percent of assessed square miles of estuaries. In some cases, agricultural pollution continues to cause serious threats to human health, as well as environmental quality. In my home state of Utah, for example, the state Health Department conducted a study which indicated that communities adjacent to a large agricultural feeding operation faced higher risks of certain types of illness than people in other communities.
Members of the agricultural community, along with federal and state agriculture agencies, counter that significant efforts have been taken over the past thirty years to reduce the impacts of agricultural water pollution. Millions of public and private sector dollars have been spent to implement various best management practices on farms, ranches and logging operations around the country. Serious efforts have been taken to educate farmers about the environmental impacts of their activities, and to identify and disseminate information about more environmentally-friendly practices. As show in a recent survey conducted by the Environmental Law Institute, in some states the traditional voluntary approach to agricultural pollution control has even given way to regulatory approaches.(8)
Absent adequate data on the efficacy of agricultural BMPs and the degree to which such practices are maintained over time after their initial implementation, however, it is difficult to assess the degree to which discrete control efforts or programs have succeeded in addressing the problem of agricultural pollution. For example, one study in Utah found that some farmers had failed to properly maintain irrigation system improvements installed under cost-sharing programs designed to reduce salinity inputs into the Colorado River system.(9) Unfortunately, similar follow-up studies have not even been conducted for many cost-sharing and education-based programs. Therefore, for many programs it is impossible to know the degree to which BMPs are properly installed and maintained over time, much less to assess their actual effectiveness in solving runoff pollution problems.
Even assuming that most BMPs are properly installed and maintained, however, the facts continually produced by state water quality monitoring and assessment programs remain undeniable. Despite all of these laudable efforts, agriculture remains the leading cause of water pollution in the United States. The question is, why? How can it be true that severe agricultural pollution remains despite over thirty years of dedicated control efforts under the CWA as well as the large range of USDA assistance programs?
I believe that one explanation for this paradox is that the dollars spent and the efforts made under past and ongoing agricultural pollution programs have not been targeted properly and used as effectively as possible. Dollars have been spent in ways and in places that are not calculated to produce the maximum improvements in water quality. Absent a reasonable method to correlate pollution control efforts with real-world environmental conditions, those efforts are not likely to produce the best possible results. And that is precisely why the TMDL process can be used to help improve the efficiency and the effectiveness of existing as well as proposed future programs to address agricultural water pollution. It is why this Committee and the agricultural community should welcome TMDLs as ways to properly target pollution controls, to gauge their effectiveness over time, and to revise control efforts over time until actual water quality improvements can be documented.
The TMDL Process Will Help, Not Hurt, Existing Agricultural Pollution Control Programs
While the TMDL process can be complex in some respects, its basic premise is simple and eminently logical. It is designed first to identify those water bodies that remain impaired despite the implementation of the initial round of the CWA's technology-based controls on point sources. As discussed in more detail below, these include waters impaired by nonpoint as well as point sources.
The TMDL process next requires states to identify the pollutants causing the impairment for each specific water body, and the major sources of pollution in the watershed. By allocating needed pollution reductions fairly among the responsible sources, pollution control efforts can then be targeted in ways that are far more likely to produce real water quality gains than are programs conducted without this basic information. This targeting process is improved further if it is iterative, that is, if water quality trends are monitored over time, and if programs are modified as appropriate to ensure continuing environmental improvements. The net result is that whatever dollars are spent on pollution control, by governmental or private sources, are spent in ways that are more likely to improve the health of the water bodies we all share for drinking water, fish and wildlife and other uses. For example, through the proper use of TMDLs, cost-sharing dollars available for use within a particular region can be targeted at those watersheds identified as needing reductions in particular types of pollutants. Within those watersheds, programs can properly be targeted at those sources of pollution that most likely contribute to the actual problem to be addressed.
The Colorado River Basin Salinity Control Program, one of the major large watershed programs that a colleague and I studied as part of the NAPA study, uses the equivalent of a basinwide TMDL to accomplish this type of targeting. Using an available model developed by the Tennessee Valley Authority (TVA) and modified to apply to the Colorado River system, the basin states (with assistance from USGS, the Bureau of Reclamation and other agencies), calculated total salt reductions necessary to meet interstate water quality standards for the river, as well as international treaty requirements regarding the quality of water delivered to Mexico. Next, the program has continuously identified particular sources of salinity throughout the basin, and used the modeling process to identify the most cost-effective targets for salinity control. More recently, the program has instituted a competitive bidding process by which anyone within the basin can propose salinity control projects, with the most cost-effective solutions chosen during each bidding cycle. Thus, the TMDL-type process is used to target and select the best possible control projects, but does not inherently mandate particular solutions for particular sources. Available data shows that this program has succeeded in producing significant reductions in salinity inputs into the river system, and that water quality standards have been maintained in the basin as a result. Thus, because it has been properly targeted using a TMDL-type process, the salinity program stands as one of our most successful nonpoint source pollution programs in the country in terms of meeting actual water quality goals. Moreover, because the targeting approach facilitates rather than forecloses innovations such as the new competitive bidding process, the cost-effectiveness of salinity controls has improved dramatically.
It is important to understand in this regard that like the Colorado River Basin Salinity Control Program, the TMDL process will not, if implemented properly, result in rigid federal intrusion into either state or agricultural agency water quality programs. Rather, it is designed to provide the necessary information to ensure that state and other programs operate effectively. In particular, the proposed rules do not dictate particular regulatory or other pollution control strategies. Instead, they merely impose reasonable accountability to the goal of improving water quality by requiring a TMDL implementation plan to include "reasonable assurances" that the load allocations assigned to nonpoint sources will be met.(10) For nonpoint sources, "reasonable assurances" are defined as "specific procedures and mechanisms that ensure load allocations will be implemented for that waterbody;" and examples explicitly identified in the regulation include, at the option of the state water quality agency, "regulations, local ordinances, performance bonds, contracts, cost-share agreements, memorandums of understanding, site-specific or watershed-specific voluntary actions, and compliance audits of best management practices."(11)
For these reasons, I believe that agricultural agencies and the agriculture community at large should welcome rather than oppose EPA's proposed new TMDL program. They are designed to help guide, rather than to replace or to dictate the content of, the agricultural pollution control programs chosen by other responsible agencies. If implemented successfully, they will not transfer authority to EPA to decide what nonpoint source controls must be adopted and by what sources. Rather, they will help to ensure that the choices made by the other responsible agencies are wise, cost-effective, and guided by sound science. This Committee will benefit if the cost-sharing and other assistance programs conducted under its auspices are more effective, and if the limited available federal dollars are spent in ways that produce better water quality improvements. The agriculture and forestry industries will benefit when pollution control programs are conducted in locations and in ways that are more likely to improve water quality in their communities. And the public at large will benefit when the thousands of water bodies around the country that continue to be impaired due to agricultural and silvicultural pollution despite three decades of concerted control efforts are restored to healthy conditions.
Waters Impaired by Nonpoint Source Pollution Should Be Included in the TMDL process
As shown above, after almost thirty years of progress in reducing impacts from sewage treatment plants, industries and other point sources, the bulk of the remaining impairment of the nation's waters comes from nonpoint sources. Excluding those sources from the TMDL program would render the program of extremely limited value in solving the most prevalent and intractable problems facing America's waterways. Moreover, excluding nonpoint sources from a program designed to address aggregate pollution rather than discrete sources would make virtually no sense. Indeed, such a focus would unfairly continue to force point sources to bear the lion's share of the water pollution control price tag despite clear evidence that unregulated nonpoint sources contribute substantially, and in many watersheds exclusively, to water body impairment. Finally, the protracted schedule for listing and completion of TMDLs in the proposed regulations would be entirely unnecessary, and in fact would constitute unreasonable and unlawful delay of an already-overdue program, if it were limited to the comparatively easy task of identifying and quantifying point source wasteload allocations.
From a legal perspective, EPA is well within its authority, if not subject to a legal duty, to include nonpoint sources in this program. Section 303(d)(1)(A) requires states to identify all waters for which the first round of technology-based pollution controls are not sufficient to implement any applicable water quality standard. Section 303(d)(1)(C), in turn, requires the development of TMDLs for all of the waters identified under (d)(1)(A). It is entirely illogical to contend that the fact that Congress excluded from the listing process those waters that could attain water quality standards based on the earliest round of point source controls means that Congress similarly intended to exclude, by implication rather than expressed intent, waters that could only be cleaned up through the control of nonpoint sources or more complex combinations of point and nonpoint sources. Indeed, the opposite interpretation, that Congress intended to exclude only that limited category of waters identified expressly in the opening sentence of section 303(d)(1)(A), is a far more logical and straightforward way to read the statutory language. Given this realization, Congress could have indicated an intent to exclude nonpoint source waters with far more precision than it is alleged to have done here. While the legislative history of this provision is sparse and far from a model of clarity, it clearly reflects that Congress understood that nonpoint sources contribute substantially to the pollution of many watersheds and should be taken into account in the TMDL process.(12) As my colleague Oliver Houck, a Professor of Law at Tulane Law School, wrote:
... the only logical interpretation of [the] legislative history behind section 303(d) is that nonpoint sources were a big fact of life in achieving water quality standards, and they would have to be included in the assessments of polluted waters and their TMDL allocations. Were they not included, a process to ensure that municipal and industrial limits were "consistent with water quality standards would make no sense; it, literally, could not be done.(13)
TMDLs Should Include Implementatioon Plans
The new regulations would depart from existing practice by requiring states to include implementation plans as part of their TMDLs, rather than leaving implementation to other, disparate parts of the Act. Ironically, at appears that some representatives of both the regulated community and the environmental community object to the inclusion of implementation plans as a mandatory component of TMDLs, although for different reasons. Regulated groups object because implementation plans, as defined in the proposed rules, will at long last require the identification of more precise and firm obligations to address the many sources of pollution that have escaped control obligations in the past. Representatives of some environmental groups will object because the inclusion of implementation plans will further delay the TMDL program relative to TMDLs that seek to produce numeric allocations alone and leave implementation to other provisions of the law.
Equally ironic, however, is the fact that the desirability of implementation plans in the TMDL program was one of the recommendations of the FACA Committee that came with unanimous support, despite the diverse representation on the Committee.(14) In my view, especially with respect to nonpoint sources, the inclusion of implementation plans represents one of the most significant recommendations that the Committee had to offer, and one that will result in the most important contributions to long-term water pollution control. Without implementation plans and the more precise identification of on-the-ground controls that will be included in such plans, many pollution sources will remain uncontrolled, as has occurred under countless past "planning" efforts under this provision, section 208, section 319, and section 303(e) of the Act. As a matter of fairness, it is also essential to note that the agreement to allow up to 15 years for states to complete their TMDLs was linked expressly in the FACA deliberations to the fact that the TMDLs would include implementation plans, which will be more difficult and time-consuming to develop. If EPA deleted the implementation plan component from the regulations, there would be no plausible justification for the extended schedule allowed by other aspects of the proposed rules.
The only real legal dispute about EPA's authority to require implementation plans as part of TMDLs is whether such authority resides in section 303(d) or section 303(e) of the Act. The implications of this distinction will be purely theoretical if the states implement the TMDL program properly. In that case, all state TMDLs will be approved by EPA and there will be no need for EPA to step in and prepare TMDLs and their accompanying implementation plans instead. It is reasonable to expect that most or all states will meet their TMDL obligations once clarified in the final regulation. If some states fail to do so, however, there are important reasons why EPA should step in and implement the program fully in their stead. First, from the perspective of interstate fairness, one of the major values Congress sought to protect in passing the law in the first place, it will be unfair if most states and their citizens fulfill their TMDL obligations properly, but some do not. In essence, equally-situated farmers (and other operations) in a few states will be given an unfair competitive advantage with respect to those in other states. Equally important, the citizens of most states will be protected from the adverse effects of agricultural pollution, while the unfortunate citizens of other states will not. Moreover, for these reasons Congress clearly provided in section 303(d) of the Act that EPA must step in and implement the TMDL program if a state fails to do so. This authority will be rendered meaningless if EPA is allowed only to produce raw, numeric TMDL calculations, but not to develop the implementing steps necessary to ensure that the TMDL wasteload and load allocations are actually achieved. The purpose of the CWA is to restore and maintain the integrity of the nation's waters, not simply to produce meaningless calculations.
EPA Has Ample Authority to Require NPDES Permits for Silvicultural Point Sources
Finally, the proposed EPA regulations seek to facilitate better implementation of the TMDL program by amending its existing regulatory exemptions for certain silvicultural point sources to require, on a case-by-case basis, the issuance of NPDES permits by EPA or delegated states where necessary to implement wasteload allocations identified in the TMDL process. This authority can only be applied where EPA has written a TMDL, presumably due a state failure to do so, and only based on a specific finding that the source contributes to a violation of a water quality standard or is a significant contributor of pollutants to waters of the United States. Where such conditions exist, it is entirely logical and appropriate to use the NPDES system as the mechanism to ensure that appropriate pollution controls are adopted by those sources.
Some representatives of the forestry industry unfortunately have mischaracterized this proposed regulation. In particular, they allege that EPA is attempting, by regulation, to convert to point sources activities that are designated as nonpoint sources in the CWA. This is not correct as a matter of law. With only limited exceptions that apply to agricultural stormwater and irrigation return flows, but not to silvicultural activities, point sources are defined in the statute by reference to the nature of the discharge, and not by category of economic activity. In particular, the statute defines "point source" to mean:
... any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged. This term does not include agricultural stormwater discharges and return flows from irrigated agriculture.(15)
Federal courts have indicated repeatedly that, in order to fulfill the remedial purposes of the statute, this definition should be interpreted broadly.(16) Thus, based on the plain language of the statute alone, discharges of pollutants to waters of the United States from any silvicultural operations through the types of discrete conveyance listed in the statutory definition should require NPDES permits. By contrast, silvicultural pollution that reaches waters of the United States through means other than point sources, that is, through runoff rather than discharges through discrete conveyances, constitutes nonpoint source pollution and is statutorily exempt from the NPDES program. Again, except for the express agricultural stormwater and irrigation return flow exemption, it is the manner in which pollution reaches water bodies and not the type of economic activity in question that dictates whether a point source exists. When Congress includes such express exemptions in a statute, it is presumed that other, similar exemptions were not intended.
In the past, however, EPA has by regulation decided to exempt certain silvicultural point sources from NPDES permitting requirements. In reality, it is these regulatory exemptions rather than the current proposal to remove some of those exemptions, that are arguably illegal under the CWA. When EPA originally attempted to exclude entire categories of point sources from the NPDES program, environmental groups successfully challenged that practice.(17) The D.C. Circuit Court of Appeals did indicate that EPA had limited authority to determine the proper scope of the NPDES program through its regulatory interpretation of the terms "point source" and "nonpoint source." And EPA has used that technique to exempt, among other sources, a large number of silvicultural activities that otherwise would constitute point sources under the plain language of the CWA. Thus, what EPA proposes to do now is simply to modify those regulatory exemptions under the limited circumstance of silvicultural point sources that contribute to water quality violations, as identified through the TMDL process. It is not converting statutory nonpoint sources into point sources.
This does not mean, as industry representatives suggest, that all silvicultural activities will now be treated as point sources. First, nothing in the proposed regulation purports to define as a point source any silvicultural pollution other than discharges through confined and discrete conveyances. Unconfined silvicultural runoff will continue to be treated, as the statute currently requires, as a nonpoint source. Second, as explained above, EPA will only treat discharges through such discrete conveyances as point sources based on specific findings of water quality impairment through the TMDL process.
Neither section 208 nor section 404 of the CWA exempt silvicultural activities categorically from the definition of point source. First, if Congress intended this result, it could have done so much more clearly by simply adding "silvicultural stormwater" to the exemptions expressly included in the definition of "point source," quoted above. Second, the alleged exemption is simply not supported by the language of either section 208 or section 404 of the Act. Section 208, the provision of law originally designed as one of the primary means of planning for nonpoint source pollution control, required states to identify and to design best management practices to address "silviculturally related nonpoint source pollution," along with other forms of nonpoint source pollution(18) This does not, on its face, in any way indicate that all silvicultural pollution is nonpoint source pollution. Rather, it simply instructs that some types of silvicultural pollution, to be identified by the states, was in the form of nonpoint source runoff and should be addressed in the 208 process.
Section 404(f) of the CWA,(19) by contrast, does expressly exempt "normal" silvicultural operations and the construction and maintenance of forest roads from permitting requirements under both section 402 and section 404 of the Act, but only for limited purposes. In particular, the provision indicates that "the discharge of dredged or fill material" from such activities requires a permit under either section 402 or section 404. Again, however, where Congress includes an express exemption for certain limited purposes, it is presumed that other, broader exemptions were not intended. Here, the exemption applies only to discharges of dredged or fill material, and not to other types of pollution normally addressed in the NPDES program.
Conclusion
I do not agree with all aspects of the proposed regulations, and given the difficult competing issues the agency faces, it is not likely that anyone does. In general, however, EPA has done an excellent job of integrating the recommendations of the FACA Committee and other relevant considerations into a coherent regulatory package that has the potential to make significant progress in implementing an important aspect the CWA that has remained largely unfulfilled for almost three decades. In light of these long delays, and given the need to strike a reasonable balance between the competing interests who have objected to these proposed changes, I hope that EPA proceeds expeditiously with the promulgation and full implementation of final regulations. While far from perfect, the TMDL program is the best tool available under the CWA as currently written to redress the wide range of impairments that affect watersheds across the country in a comprehensive rather than a fragmented way. Moreover, the TMDL process provides an excellent structure within which to implement EPA and state watershed-based policies and programs.
I appreciate the opportunity to testify today on these important regulatory changes. I would be happy to answer any questions the Committee might have.
1. These comments do not reflect the views of my current employer, the University of Utah College of Law. My affiliation is included above for identification purposes only.
2. Robert W. Adler, Jessica C. Landman and Diane M. Cameron, The Clean Water Act 20 Years Later (Island Press 1993).
3. Robert W. Adler, "Integrated Approaches to the Water Quality Problem: Lessons from the Clean Air Act," 23 Harvard Environmental Law Review 203-95 (1999).
4. NAPA is in the process of reviewing the research conducted to support this study. The final report is still in process.
5. See generally, National Research Council, Alternative Agriculture (National Academy Press 1989).
6. Sen. Rept. 414, 92d Cong. 2d Sess., reprinted in 1972 U.S. Code Cong. & Ad. News 3668, 3705-06.
7. EPA. The Quality of Our Nation's Waters: 1996, at 13.
8. Environmental Law Institute, Enforceable State Mechanisms for the Control of Nonpoint Source Water Pollution (1997).
9. Hill, Robert W. and Terry A. Messmer, et al., Colorado River Salinity Control Program Uintah Basin Monitoring and Evaluation Final Report (Utah State University, April 1999).
10. Proposed 40 C.F.R. 130.33(b)(10(3).
11. Proposed 40 C.F.R. 130.2(p).
12. H.R. Rep. No. 92-911, at 105 (1972).
13. Oliver A. Houck, The Clean Water Act TMDL Program: Law, Policy and Implementation, at 41 n.93 (Environmental Law Institute 1999).
14. Committee members disagreed, of course, about whether the inclusion of implementation plans was appropriate under section 303(d) or 303(e) of the CWA.
15. CWA §502(14), 33 U.S.C. §1362(14).
16. E.g., U.S. v. Earth Sciences, 599 F.2d 368, 373 (10th Cir. 1979); Quivira Mining Co. v. EPA, 765 F.2d 126, 129-30 (10th Cir 1985); Community Ass'n. For Restoration of the Environment v. Sid Koopman Dairy, 54 F.Supp.2d 976, 981 (E.D. Wash. 1999).
17. NRDC v. Costle, 568 F.2d 1369 (D.C. Cir. 1977).
18. CWA §208(b)(2)(F), 33 U.S.C. §1288(b)(2)(F).
19. 33 U.S.C. §1344(f).