Natural Resources Defense Council, Inc. v. Environmental Protection Agency et alCross MOTION for Summary Judgment and Opposition to Plaintiff's Motion for Summary JudgmentD.D.C.June 2, 2017IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA ____________________________________ ) NATURAL RESOURCES DEFENSE ) COUNCIL, INC. ) ) Plaintiff, ) ) Civil Action No. 1:16-cv-01861-JDB v. ) ) UNITED STATES ENVIRONMENTAL ) PROTECTION AGENCY, et al., ) ) Defendants. ) ____________________________________) EPA’S CROSS-MOTION FOR SUMMARY JUDGMENT AND OPPOSITION TO NRDC’S MOTION FOR SUMMARY JUDGMENT Of Counsel: James Curtin Office of General Counsel U.S. Environmental Protection Agency 1200 Pennsylvania Avenue, N.W. Washington, D.C. 20460 Stefania Shamet Kelly A. Gable Office of Regional Counsel U.S. Environmental Protection Agency, Region 3 1650 Arch Street Philadelphia, PA 19103 Jeffrey H. Wood Acting Assistant Attorney General Environment and Natural Resources Division United States Department of Justice Justin D. Heminger Senior Attorney Environmental Defense Section P.O. Box 7611 Washington, D.C. 20044 (202) 514-2689 justin.heminger@usdoj.gov Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 1 of 50 i TABLE OF CONTENTS Table of Contents ............................................................................................................................................... i Table of Authorities .......................................................................................................................................... iii Introduction ........................................................................................................................................................ 1 Background ......................................................................................................................................................... 2 I. Statutory and Regulatory Background ................................................................................. 2 A. The States’ Development of Water Quality Standards ........................................ 3 B. The TMDL Process .................................................................................................. 4 C. The NPDES Permitting Process ............................................................................ 7 II. Factual Background ................................................................................................................ 8 A. The Anacostia River’s Trash Problem ................................................................... 8 B. Development of the Anacostia River Trash TMDL ........................................... 9 1. TMDL Data Development ...................................................................... 11 2. Calculation of the Baseline Load of Trash ............................................ 13 3. TMDL Establishment ............................................................................... 14 C. Consideration of Public Comments on the Trash TMDL ............................... 16 D. EPA’s Approval of the Trash TMDL .................................................................. 17 Standard of Review .......................................................................................................................................... 19 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 2 of 50 ii Argument .......................................................................................................................................................... 20 I. EPA’s approval of the Trash TMDL is consistent with the Clean Water Act’s language and EPA’s implementing regulations. ............................................................... 20 A. The phrase “total maximum daily load” in Section 1313(d)(1)(C) of the Act is ambiguous. ............................................................................................. 20 B. EPA’s approval of the Trash TMDL is based on a reasonable interpretation of the ambiguous language in Section 1313(d)(1)(C) and a reasonable application of its own regulations. ......................................... 23 1. The Trash TMDL sets a maximum load for trash. .............................. 24 2. EPA reasonably interpreted the Act and its regulations to allow the Trash TMDL to express the maximum load as the amount of trash to be prevented from entering the River. ................. 25 C. EPA played an appropriate role in reviewing and approving the Trash TMDL developed by Maryland and the District. ............................................... 29 II. EPA rationally approved the District’s and Maryland’s determination that the Trash TMDL is set at a level necessary to implement their water quality standards. ............................................................................................................................... 31 A. EPA properly approved the District’s and Maryland’s conclusion that removal of the baseline load of trash, with a margin of safety, would ensure compliance with the water quality standards. ......................................... 31 B. The baseline load is a conservative and reliable estimate of the average amount of trash loaded annually into the Anacostia River. ............... 35 III. The Court should decline to grant the relief that NRDC requests. .............................. 39 Conclusion ........................................................................................................................................................ 41 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 3 of 50 _________________ * Authorities chiefly relied upon are marked with an asterisk. TABLE OF AUTHORITIES CASES Allied-Signal, Inc. v. United States Nuclear Regulatory Comm’n, 988 F.2d 146 (D.C. Cir. 1993) ................................................................................................................... 40 *Am. Farm Bureau Fed’n v. EPA, 984 F. Supp. 2d 289 (M.D. Pa. 2013), aff'd, 792 F.3d 281 (3d Cir. 2015), cert. denied, 136 S. Ct. 1246 (2016) ................................................................................................... 4, 36, 38 *Am. Farm Bureau Fed'n v. EPA, 792 F.3d 281 (3d Cir. 2015), cert. denied, 136 S. Ct. 1246 (2016) ................................................................................................. 21, 22, 23 Am. Petroleum Inst. v. EPA, 684 F.3d 1342 (D.C. Cir. 2012) ................................................................................................................. 19 Anacostia Riverkeepers, Inc. v. Jackson, 713 F. Supp. 2d 50 (D.D.C. 2010) ............................................................................................................ 40 Anacostia Riverkeeper, Inc. v. Jackson, 798 F. Supp. 2d 210 (D.D.C. 2011) ................................................................................................. 3, 4, 34 Auer v. Robbins, 519 U.S. 452 (1997) ..................................................................................................................................... 19 Banner Health v. Sebelius, 715 F. Supp. 2d 142 (D.D.C. 2010) ................................................................................................... 19, 31 Bennett v. Donovan, 703 F.3d 582 (D.C. Cir. 2013) ................................................................................................................... 39 Chevron U.S.A., Inc. v. Hammond, 726 F.2d 483 (9th Cir. 1984) ........................................................................................................................ 3 Chevron, U.S.A., Inc. v. NRDC, Inc., 467 U.S. 837 (1984) ................................................................................................................. 19, 20, 21, 23 City of Albuquerque v. Browner, 97 F.3d 415 (10th Cir. 1996) ...................................................................................................................... 29 City of Arcadia v. EPA, 265 F. Supp. 2d 1142 (N.D. Cal. 2003) ...................................................................................................... 4 Ctr. for Biological Diversity v. EPA, 749 F.3d 1079 (D.C. Cir. 2014) ................................................................................................................. 38 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 4 of 50 iv Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct. 1326 (2013) ................................................................................................................................. 23 Defenders of Wildlife v. EPA, 415 F.3d 1121 (10th Cir. 2005) .................................................................................................................... 4 District of Columbia v. Schramm, 631 F.2d 854 (D.C. Cir. 1980) ................................................................................................................... 30 EPA v. Nat’l Crushed Stone Ass’n, 449 U.S. 64 (1980) ....................................................................................................................................... 37 Ethyl Corp. v. EPA, 541 F.2d 1 (D.C. Cir. 1976) ........................................................................................................................ 37 Friends of the Earth v. EPA, 446 F.3d 140 (D.C. Cir. 2006) ............................................................................................................ 22, 28 Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983) ....................................................................................................................................... 19 Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644 (2007) ..................................................................................................................................... 22 Nat’l Wildlife Fed’n v. Browner, 127 F.3d 1126 (D.C. Cir. 1997) .......................................................................................................... 19, 27 Natural Resources Defense Council v. EPA, 489 F.3d 1364 (D.C. Cir. 2007) ................................................................................................................. 40 North Carolina v. EPA, 550 F.3d 1176 (D.C. Cir. 2008) ................................................................................................................. 40 N. Air Cargo v. U.S. Postal Serv., 674 F.3d 852 (D.C. Cir. 2012) ................................................................................................................... 39 NRDC v. EPA, 16 F.3d 1395 (4th Cir. 1993) ...................................................................................................................... 30 *NRDC v. Muszynski, 268 F.3d 91 (2d Cir. 2001) ........................................................................................................ 7, 21, 22, 33 Pa. Dep’t of Envtl. Res. v. EPA, 932 F.2d 269 (3d Cir. 1991) ....................................................................................................................... 37 Pronsolino v. Nastri, 291 F.3d 1123 (9th Cir. 2002) ........................................................................................................... 4, 7, 30 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 5 of 50 v PUD No. 1 of Jefferson Co. v. Wash. Dep’t of Ecology, 511 U.S. 700 (1994) ....................................................................................................................................... 2 Sierra Club v. EPA, 167 F.3d 658 (D.C. Cir. 1999) ................................................................................................................... 36 Sierra Club v. Meiburg, 296 F.3d 1021 (11th Cir. 2002) ............................................................................................................. 7, 29 TOMAC v. Norton, 193 F. Supp. 2d 182 (D.D.C. 2002) .......................................................................................................... 19 United States v. Homestake Mining Co., 595 F.2d 421 (8th Cir. 1979) ...................................................................................................................... 29 Upper Blackstone Water Pollution Abatement Dist. v. EPA, 690 F.3d 9 (1st Cir. 2012) ........................................................................................................................... 37 West Virginia v. EPA, 362 F.3d 861 (D.C. Cir. 2004) ................................................................................................................... 19 Zuni Pub. Sch. Dist. v. Dep’t of Educ., 550 U.S. 81 (2007) ....................................................................................................................................... 38 STATUTES 5 U.S.C. § 706(2) .............................................................................................................................................. 39 5 U.S.C. § 706(2)(A) ......................................................................................................................................... 19 33 U.S.C. § 1251(a) ............................................................................................................................................ 2 33 U.S.C. § 1251(b) ................................................................................................................................. 2, 3, 29 33 U.S.C. § 1311(b)(1)(C) .................................................................................................................................. 7 33 U.S.C. § 1313(a) ............................................................................................................................................ 3 33 U.S.C. § 1313(b) ............................................................................................................................................ 3 33 U.S.C. § 1313(c)(1) ........................................................................................................................................ 3 33 U.S.C. § 1313(c)(2) ........................................................................................................................................ 3 33 U.S.C. § 1313(c)(3) ........................................................................................................................................ 3 33 U.S.C. § 1313(d) ....................................................................................................................................... 4, 5 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 6 of 50 vi 33 U.S.C. § 1313(d)(1)(A) ................................................................................................................................. 3 33 U.S.C. § 1313(d)(1)(B) .................................................................................................................................. 3 33 U.S.C. § 1313(d)(1)(C) ............................................................................... 4, 5, 6, 7, 20, 21, 22, 25, 26, 31 33 U.S.C. § 1313(d)(2) ............................................................................................................................ 4, 7, 29 33 U.S.C. § 1313(e) ............................................................................................................................................ 4 33 U.S.C. § 1362(6) ............................................................................................................................................ 9 33 U.S.C. § 1362(12)(A) .................................................................................................................................... 4 33 U.S.C. § 1362(14) .......................................................................................................................................... 4 CODE OF FEDERAL REGULATIONS 40 C.F.R. § 122.44(d)(1)(vii)(A) ....................................................................................................................... 7 40 C.F.R. § 122.44(d)(1)(vii)(B) ........................................................................................................................ 7 40 C.F.R. § 130.2(e) ................................................................................................................................ 5, 6, 28 40 C.F.R. § 130.2(f) ..................................................................................................................................... 5, 28 40 C.F.R. § 130.2(g) .................................................................................................................................... 5, 28 40 C.F.R. § 130.2(g)-(i) ...................................................................................................................................... 6 40 C.F.R. § 130.2(h) .................................................................................................................................... 5, 28 40 C.F.R. § 130.2(i) ..................................................................................................... 5, 18, 23, 25, 26, 27, 28 40 C.F.R. § 130.2(j) ....................................................................................................................................... 3, 4 40 C.F.R. § 130.7 ................................................................................................................................................ 6 40 C.F.R. § 130.7(b)(1) ...................................................................................................................................... 4 40 C.F.R. § 130.7(c)(1) .................................................................................................................................. 4, 6 40 C.F.R. § 130.7(c)(1)(ii) .................................................................................................................................. 6 40 C.F.R. § 130.7(d) ........................................................................................................................................... 6 40 C.F.R. § 130.7(d)(1) ...................................................................................................................................... 4 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 7 of 50 vii 40 C.F.R. § 130.7(d)(2) ...................................................................................................................................... 7 40 C.F.R. § 131.3(i) ............................................................................................................................................ 3 LOCAL RULES OF CIVIL PROCEDURE Local Civil Rule 7(h)(2) ..................................................................................................................................... 8 Local Civil Rule 7(n) .......................................................................................................................................... 8 Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 8 of 50 INTRODUCTION As with most urban waterbodies, the Anacostia River faces enormous pollution challenges. But more than six years ago, two local governments, the District of Columbia and the State of Maryland, took a bold step to improve the River—a step that only one other jurisdiction, California, has taken. In August 2010, as the result of a years-long partnership, the District and Maryland established Total Maximum Daily Loads (TMDLs) under the Clean Water Act for trash in their respective portions of the Anacostia River (together, the Trash TMDL). The District and Maryland were not alone in their work—they collaborated with many stakeholders, including environmental groups. In developing the Trash TMDL, the District and Maryland considered the unique nature of trash pollution, and they confronted the challenges posed by translating their narrative water quality standards for trash—which speak in subjective terms such as “objectionable” and “unsightly” rather than setting numeric limitations—into a defined daily limit that will ensure achievement of those criteria. To solve these challenges, the District and Maryland expressed the “total maximum daily load” in the Trash TMDL as prevention, capture, or removal of 105 percent of the average daily load of trash in the River (100 percent, plus an additional five percent for a margin of safety). In September 2010, the U.S. Environmental Protection Agency (EPA) fulfilled its statutory role under Section 1313(d) of the Clean Water Act when it reviewed and approved the Trash TMDL. Over the six years since then, the District and Maryland have made progress toward implementing and achieving the Trash TMDL’s goal of capturing or removing 105 percent of the average daily trash load that would otherwise be discharged into the River. Even today, the Trash TMDL remains one of only a few TMDLs in the entire country to address trash pollution. Enter the Plaintiff, Natural Resources Defense Council (NRDC), which claims that EPA lacked the statutory authority to approve the Trash TMDL. NRDC’s Mot. for Summ. J., ECF No. 10 (NRDC Br.) at 15–22. But when EPA reviewed the Trash TMDL, it concluded that the “total Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 9 of 50 2 maximum daily load” selected by the District and Maryland—prevention or removal of 105 percent of the average daily trash load—was consistent with a reasonable interpretation of the Clean Water Act, EPA’s implementing regulations, and the District’s and Maryland’s narrative water quality standards as applied to trash pollution. NRDC cannot show that the phrase “total maximum daily load” has a single, plain meaning that invalidates EPA’s approval, or that EPA’s interpretation of its own regulations is plainly erroneous. Thus, EPA acted within its statutory and regulatory authority when it approved the Trash TMDL. NRDC also claims (Br. at 22–28) that EPA’s approval of the Trash TMDL is arbitrary and capricious, but EPA rationally explained why removal of 100 percent of the baseline load of trash, and another five percent to account for a margin of safety, would satisfy the District’s and Maryland’s narrative water quality standards. EPA’s explanation is supported by substantial evidence in the record. And as a matter of technical and scientific judgment within its expertise, EPA’s conclusion warrants a high degree of deference. Because EPA’s decision to approve the Trash TMDL is within the Agency’s authority and follows from a reasoned explanation with solid support in the administrative record, the Court should grant summary judgment to EPA. The Trash TMDL should remain approved and in place, as it has for the past six years. BACKGROUND I. Statutory and Regulatory Background Congress enacted the Clean Water Act to “restore and maintain the chemical, physical, and biological integrity of the Nation’s waters” by eliminating the discharge of pollutants to those waters. 33 U.S.C. § 1251(a). The Act is a “complex statutory and regulatory scheme . . . [that] establishes distinct roles for the Federal and State Governments.” PUD No. 1 of Jefferson Co. v. Wash. Dep’t of Ecology, 511 U.S. 700, 704 (1994); see also 33 U.S.C. § 1251(b) (emphasizing the States’ role in Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 10 of 50 3 preventing pollution). “A core element of the [Clean Water Act] is a two-step approach to improving water quality, which delegates certain responsibilities to EPA and others to the states in furtherance of the Act’s stated purpose of promoting cooperation between federal and state governments.” Anacostia Riverkeeper, Inc. v. Jackson, 798 F. Supp. 2d 210, 214 (D.D.C. 2011) (citing 33 U.S.C. § 1251(b)). Under this cooperative federalism structure, States have “primary responsibility for abating pollution in their jurisdictions.” Chevron U.S.A., Inc. v. Hammond, 726 F.2d 483, 489 (9th Cir. 1984). This is particularly true with regard to Section 1313 of the Act, 33 U.S.C. § 1313, the section that governs the process for establishing water quality standards and TMDLs. A. The States’ Development of Water Quality Standards Section 1313 of the Act “requires each State, subject to federal approval, to institute comprehensive water quality standards establishing water quality goals for all intrastate waters.” PUD No. 1, 511 U.S. at 704. In step one of this comprehensive process, States, with federal approval and oversight, adopt water quality standards for each waterbody or waterbody segment within their boundaries. 33 U.S.C. §§ 1313(a), (b) & (c)(1). A State sets a water quality standard by identifying (1) the “designated uses” for a particular waterbody (for example, public water supply, support of aquatic life, and/or recreational uses) and (2) a “water quality criterion” generally expressed as a level (for example, a pollutant-specific concentration or a narrative condition) that must not be exceeded so that the waterbody can support those uses (for instance, dissolved oxygen levels for fish). Id. § 1313(c)(2); 40 C.F.R. § 131.3(i). Several water quality standards may apply to the same waterbody or waterbody segment. EPA either approves a State’s proposed water quality standards or, if it disapproves, promulgates standards for the State. Id. § 1313(c)(3). After adoption and approval of water quality standards, Section 1313(d) directs the States to identify and prioritize water-quality-limited segments, which are the individual water segments that do not, or are not expected to, meet applicable water quality standards even after implementation of Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 11 of 50 4 technology-based effluent limitations. 33 U.S.C. § 1313(d)(1)(A) & (B); 40 C.F.R. §§ 130.2(j) & 130.7(b)(1). EPA’s regulations specify that the States submit their water-quality-limited segments in lists, known as “Section 303(d)” lists, to EPA for approval or disapproval on a biennial basis. 40 C.F.R. § 130.7(d)(1). If EPA disapproves, it must identify the water-quality-limited segments to be added within 30 days from the date of disapproval. 33 U.S.C. § 1313(d)(2). In the final step, discussed further in the next section, the State addresses impairments to its water-quality-limited segments through various planning mechanisms, including the establishment of TMDLs under Section 1313(d) of the Clean Water Act and a continuing planning process under Section 1313(e) of the Act. 33 U.S.C. §§ 1313(d), (e); Am. Farm Bureau Fed’n v. EPA, 984 F. Supp. 2d 289, 295–98 (M.D. Pa. 2013), aff’d, 792 F.3d 281 (3d Cir. 2015), cert. denied, 136 S. Ct. 1246 (2016). B. The TMDL Process A TMDL is a planning document, developed to establish the total maximum daily load of a pollutant “necessary to implement the applicable water quality standards” for water-quality-limited segments. 33 U.S.C. § 1313(d)(1)(C); 40 C.F.R. § 130.7(c)(1); Anacostia Riverkeeper, 798 F. Supp. 2d at 216; City of Arcadia v. EPA, 265 F. Supp. 2d 1142, 1144 (N.D. Cal. 2003). Each TMDL addresses all sources of a particular pollutant, that is, point sources subject to a National Pollutant Discharge Elimination System (NPDES) permit and nonpoint sources beyond the purview of direct federal regulation. Pronsolino v. Nastri, 291 F.3d 1123, 1126–29 (9th Cir. 2002).1 TMDLs establish a “pollution budget” for a waterbody to meet water quality standards and assign pollutant allocations to contributing point and nonpoint sources. 33 U.S.C. § 1313(d)(1)(C); 40 C.F.R. § 130.2(i); Anacostia Riverkeeper, 798 F. Supp. 2d at 216. 1 “A point source is “any discernible, confined and discrete conveyance . . . from which pollutants are or may be discharged.” 33 U.S.C. §§ 1362(12)(A), (14). A nonpoint source is any source that does not meet the definition of “point source” and generally includes agricultural areas and non-regulated stormwater runoff of pollutants. Defenders of Wildlife v. EPA, 415 F.3d 1121, 1124 (10th Cir. 2005). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 12 of 50 5 The phrase “total maximum daily load” is not defined anywhere in the Clean Water Act. The statute provides only that “[s]uch load shall be established at a level necessary to implement the applicable water quality standards with seasonal variations and a margin of safety which takes into account any lack of knowledge concerning the relationship between effluent limitations and water quality.” 33 U.S.C. § 1313(d)(1)(C). The Definitions section of EPA’s regulations on Water Quality Planning contains a nested series of definitions for the phrase “total maximum daily load.” There, EPA defines a TMDL to be the “sum of the individual [wasteload allocations] for point sources and [load allocations] for nonpoint sources and natural background.” 40 C.F.R. § 130.2(i). The definition of a TMDL also states that “TMDLs can be expressed in terms of either mass per time, toxicity, or other appropriate measure.” Id. The term wasteload allocation is in turn defined as “[t]he portion of a receiving water’s loading capacity that is allocated to one of its existing or future point sources . . . [and] constitute[s] a type of water quality-based effluent limitation.” 40 C.F.R. § 130.2(h). The term load allocation is defined as “the portion of a receiving water’s loading capacity that is attributed either to one of its existing or future nonpoint sources of pollution or to natural background sources.” Id. § 130.2(g). The term “loading capacity” is defined as “[t]he greatest amount of loading that water can receive without violating water quality standards.” Id. § 130.2(f). Finally, the terms “load or loading” are defined as “[a]n amount of matter or thermal energy that is introduced into a receiving water; to introduce matter or thermal energy into a receiving water. Loading may be either man-caused (pollutant loading) or natural (natural background loading).” Id. § 130.2(e). States are to develop a TMDL for each impaired waterbody and the particular pollutant for which that waterbody is impaired. 33 U.S.C. § 1313(d). TMDL development requires States to identify the maximum amount of pollutant “loading,” that is, the quantity of a particular pollutant, that the impaired waterbody can receive from all sources and still meet the relevant water quality Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 13 of 50 6 standard. Id. § 1313(d)(1)(C); 40 C.F.R. § 130.2(e). Each TMDL must, among other things: (1) be set at a level necessary to meet the applicable water quality standard for which it is established; (2) include, as appropriate, both wasteload allocations from point sources and load allocations from nonpoint sources; (3) consider the impacts of background pollutant contributions; (4) consider seasonal variations; (5) include a margin of safety; and (6) be subject to public participation. 40 C.F.R. §§ 130.7, 130.7(c)(1), 130.2(g)-(i). Calculating a pollutant’s total maximum daily load that is necessary to ensure attainment of water quality standards is a highly technical undertaking. The total maximum daily load accounts for numerous factors, including the waterbody’s flow and physical characteristics, the manner in which the applicable water quality standard is expressed, the amounts and sources of pollutants, and the pollutant’s impact on designated beneficial uses. States may use a variety of technical and analytical tools to derive the pollution load targets. The lack of scientific exactitude will not render a TMDL invalid. Instead, the daily loads established in a TMDL must include a margin of safety that “takes into account any lack of knowledge concerning the relationship between effluent limitations and water quality.” 33 U.S.C. § 1313(d)(1)(C). Developing a TMDL typically requires significant data gathering and complex analyses, and once initiated, may take years to complete depending, among other things, on the data and studies required. Once the wasteload and load calculations (with a margin of safety) necessary to implement State-defined water quality standards are generated, the proposed TMDL is subject to public review and comment, which occurs through the State’s (and in this case, the District’s) administrative process. Id. § 130.7(c)(1)(ii). After considering public comments, the State finalizes the TMDL and submits it to EPA for approval. Id. § 130.7(d). Once a State submits a TMDL to EPA, EPA has only thirty days to consider the TMDL’s conclusions, scientific analysis, modeling, and underlying record, and to either approve or Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 14 of 50 7 disapprove the TMDL. 33 U.S.C. § 1313(d)(2); 40 C.F.R. § 130.7(d)(2). EPA’s central task in conducting its 30-day review is to assess whether the TMDL is “established at a level necessary to implement the applicable water quality standards.” 33 U.S.C. § 1313(d)(1)(C); NRDC v. Muszynski, 268 F.3d 91, 100 (2d Cir. 2001) (“EPA’s primary concern in determining whether to approve the TMDL is whether or not the TMDL will ‘implement the applicable water quality standards.’”) (citation omitted); Anacostia Riverkeeper, 798 F. Supp. 2d at 238. If EPA disapproves a State- submitted TMDL, EPA must establish a federal TMDL for the water-quality-limited segments covered by that TMDL within 30 days of the Agency’s disapproval. 33 U.S.C. § 1313(d)(2). Once approved by EPA, the implementation of a TMDL rests with the State. Sierra Club v. Meiburg, 296 F.3d 1021, 1031 (11th Cir. 2002) (the Clean Water Act “puts the responsibility for implementation of TMDLs on the states.”); Pronsolino, 291 F.3d at 1140. C. The NPDES Permitting Process TMDLs function primarily as planning devices and are not self-executing. Pronsolino, 291 F.3d at 1129. Each TMDL represents a pollutant reduction goal that the State or EPA may implement by adjusting pollutant discharge requirements in individual NPDES permits for point sources, by encouraging nonpoint source controls, or both. Meiburg, 296 F.3d at 1025; 40 C.F.R. § 122.44(d)(1)(vii)(B) (effluent limits in NPDES permits must be consistent with the assumptions and requirements of any TMDL).2 Thus, TMDLs form a basis for further State actions for particular pollutant discharges. Even absent a TMDL, NPDES permits must include effluent limits as stringent as necessary to meet water quality standards. 33 U.S.C. § 1311(b)(1)(C); 40 C.F.R. § 122.44(d)(1)(vii)(A). 2 Most States administer their own NPDES permit program, but EPA manages the program for a few States and for the District of Columbia. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 15 of 50 8 II. Factual Background Pursuant to Local Civil Rule 7(h)(2), EPA provides the following statement of facts with references to the administrative record in support of its motion for summary judgment.3 A. The Anacostia River’s Trash Problem The Anacostia River is an important shared water resource in the region—about 80 percent of the watershed is in Maryland and the rest is in the District.4 From its headwaters in Montgomery and Prince George’s Counties, Maryland, to its confluence with the Potomac River in the District of Columbia, the Anacostia’s watershed drains more than 170 square miles.5 The watershed is highly urbanized. Forty-five percent of the land is residential, and fifteen percent is commercial and institutional, while only thirty percent of the land is undeveloped.6 This high degree of urbanization has led to a trash problem. Maryland and the District have each established designated uses for their respective portions of the Anacostia River and developed water quality criteria to protect those uses.7 The water quality criteria that are applicable to trash are expressed as narrative descriptions, rather than numeric levels, as detailed in Section 1.4 of the Trash TMDL.8 As Maryland and the District explained, their respective narrative water quality criteria for the Anacostia River “describe unacceptable levels of trash in subjective terms such as objectionable, nuisance, and unsightly.”9 Maryland and the District do 3 Citations to “AR” numbers are references to pages in the administrative record. Pursuant to Local Civil Rule 7(n), the parties will jointly prepare an appendix containing those portions of the administrative record that the parties cite or rely upon, which NRDC will file at the conclusion of summary judgment briefing. 4 Trash TMDL at 1 (AR0003006). 5 Trash TMDL at 1 (AR0003006). For a map depicting the location of the Anacostia River watershed, see Figure 1 in the Trash TMDL. Id. at 3 (AR0003008). 6 Trash TMDL at 2 (AR0003007). 7 Trash TMDL at 8–11 (AR0003013–16). 8 Trash TMDL at 8–11 (AR0003013–16). 9 Comment Response Document at 3 (AR0003075). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 16 of 50 9 not interpret their narrative criteria to require zero trash, reasoning that “[e]ven extremely toxic substances have acceptable discharge limits.”10 Based upon their best professional judgment in interpreting their narrative criteria as applied to trash, both the District and Maryland identified the Anacostia River as water-quality-limited segments that are impaired by trash.11 That designation led them to develop the Trash TMDL.12 B. Development of the Anacostia River Trash TMDL Trash is one of the pollutants regulated by the Clean Water Act. 33 U.S.C. § 1362(6) (defining pollutant to mean, among other things, “solid waste,” “garbage,” “discarded equipment,” and “industrial, municipal and agricultural waste”). But among pollutants, trash has unique characteristics. As anyone who has emptied a trash can or driven by a garbage dump knows, trash comes in all shapes and sizes—from a bubble gum wrapper, to an old table lamp, to a wrecked car.13 And trash is made from every sort of material imaginable—metal, plastic, rubber, paper, Styrofoam, wood, and glass, to name just a few.14 In many ways, trash is defined not by its physical properties but by its status as something that someone has discarded.15 Trash’s unique characteristics as a pollutant make it a challenge to regulate, measure, and assess. Trash is not associated with a single category of dischargers, nor is trash naturally-occurring 10 Comment Response Document at 3 (AR0003075). 11 Decision Rationale at 5 (AR0003117); Draft TMDL at 3 (AR0002454). 12 Decision Rationale at 5 (AR0003117). 13 Trash TMDL at 20 (AR0003025) (defining 43 categories of trash and debris used for in-stream monitoring in the District). 14 Trash TMDL at 6–8 (AR0003011–13) (describing different types of trash found in the District’s portion of the watershed, from bottles and cans, to Styrofoam, to plastic bags, to “decades-old tires and construction lumber”). 15 For purposes of this TMDL, Maryland and the District defined trash as “all improperly discarded waste material, including, but not limited to, convenience food, beverage, and other product packages or containers constructed of steel, aluminum, glass, paper, plastic, and other natural and synthetic materials thrown or deposited on the land or water.” Trash TMDL at 1 (AR0003006). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 17 of 50 10 within waterways. Unlike many other pollutants, flow does not dilute trash; it merely transports it.16 Because trash is primarily the result of activity by individuals on land within the watershed, there are countless ways to reduce trash pollution—the District’s bag law is one example of a program designed to alter consumer behavior in a way that reduces trash pollution.17 Because trash is a challenging pollutant to regulate, there are only a small number of TMDLs in the entire country that address trash.18 To date, the District and Maryland are the only jurisdictions that have collaborated to tackle this difficult environmental issue by developing a joint TMDL for a shared waterbody. The Trash TMDL was developed through a collaborative effort by the District, Maryland, EPA, and a number of stakeholders, including environmental groups, as part of a broader strategy to reduce trash in the region’s waters.19 More specifically, the District and Maryland developed the Trash TMDL to complement efforts by the Washington D.C. Metropolitan Conference of Governments and the Alice Ferguson Foundation to address the trash-related problems in the Potomac River and its tributaries (including the Anacostia) through the Potomac River Watershed Trash Treaty.20 The District and Maryland formed a workgroup to focus on TMDL development. The workgroup included EPA, the District, Maryland, the governments of Montgomery and Prince 16 E-mail (Dec. 4, 2007) (AR0002654) and Attachment (AR0002657) (“The Regional Board noted that for trash, flow does not dilute the pollutant, but merely serves to transport it.”). 17 Trash TMDL at 50 (AR0003055) (“The Anacostia River Clean-Up and Protection Act was signed by the District’s mayor on July 7, 2009. A plastic bag ban created under the law will take effect on January 1, 2010. Customers will be charged 5 cents for every disposable paper or plastic bag. The proceeds go to the Anacostia River Clean-Up and Protection Fund, which will pay for the distribution of reusable bags, public environmental education, and Anacostia River cleanup projects.”). 18 Although EPA reports that there are 70,833 TMDLs across the country, there are only 92 TMDLs for trash. See Chart listing National Cumulative TMDLs by Pollutant, https://ofmpub.epa.gov/waters10/attains_nation_cy.control (last visited June 2, 2017). 19 Trash TMDL at viii (AR0003001); id. at 49 (AR0003054). 20 Trash TMDL at 47–49 (AR0003052–54) Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 18 of 50 11 George’s counties, the Anacostia Watershed Society, the Alice Ferguson Foundation, NRDC, and others, ensuring that the TMDL would be the product of inter-jurisdictional cooperation and collaborative efforts involving government and stakeholders.21 For more than a year, workgroup members participated in regularly scheduled monthly conference calls and meetings to discuss and work through various issues that arose in connection with TMDL development.22 Workgroup members also participated in the Alice Ferguson Foundation’s Potomac Trash Summit,23 coordinated with the Anacostia Watershed Trash Reduction Group,24 and worked with the Washington D.C. Metropolitan Conference of Governments.25 1. TMDL Data Development Developing the Trash TMDL for the Anacostia River involved many steps. First, Maryland and the District collected, developed, and analyzed data for the Anacostia River watershed, including data used to characterize the watershed and its water quality conditions, to identify pollutant sources, and to support the calculation of the TMDL for trash.26 Maryland and the District developed data for the TMDL using two methods: (1) collecting and monitoring trash from stormwater outfalls; and (2) counting trash in streams.27 Both the outfall and in-stream methods involved extensive efforts that are described in detail in the TMDL.28 As one 21 Comment Response Document at 23 (AR0003095). 22 For example, see E-mail (August 20, 2007) (AR0002586) and Anacostia Trash TMDL Workgroup Conference Call Agenda for Wednesday, June 3, 2009 (AR0003242). 23 Action Recommendations and Minutes from Alice Ferguson Foundation’s Potomac Trash Summit (June 14, 2007) (AR0002499–2526). 24 E-mail (July 2, 2007) (AR0002972–73). 25 E-mail (May 8, 2007) (AR0002479–80) (discussing involvement of the Conference of Governments (COG)). 26 Trash TMDL at 12–26 (AR0003017–31). 27 Trash TMDL at 13 (AR0003018). 28 Trash TMDL at 13–26 (AR0003018–31). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 19 of 50 12 illustration of the first method, the District chose ten storm sewer drainage areas for monitoring.29 The District selected the storm sewers to be representative of different land uses, such as low- density residential or parks and open space.30 The ten storm sewer sampling locations are depicted in Figure 3 in the TMDL.31 Then the District installed trash traps with one-inch diameter netting to collect trash that flowed through the outfalls.32 The District conducted monitoring from March through August 2009 whenever there was a rainfall event of at least .25 inches.33 Rainfall played an important role in monitoring trash pollution because rain increases the transportation of trash to and through streams and sewer systems.34 Precipitation in 2009 (46.90 inches) was well above the longer term average rainfall of 39.35 inches; likewise, seasonal rainfall in the Spring of 2009, when 14.24 inches of rain fell, was well above the long-term Spring average of 9.00 inches.35 The District summarized the results of the monitoring in Table 5 in the TMDL.36 Maryland used a similar methodology37 to conduct stormwater outfall monitoring at eight locations from October 2008 through July 2009.38 Both the District and Maryland also conducted in-stream trash monitoring. This involved sending volunteers from local watershed groups, such as the Anacostia Watershed Society, to walk along the River and its tributaries to identify and categorize trash.39 The District conducted in-stream 29 Trash TMDL at 13 (AR0003018). 30 Trash TMDL at 13 (AR0003018). 31 Trash TMDL at 14 (AR0003019). 32 Trash TMDL at 13 (AR0003018). 33 Trash TMDL at 13 (AR0003018). 34 Trash TMDL at 46 (AR0003051). 35 Trash TMDL at 46 (AR0003051). 36 Trash TMDL at 15 (AR0003020). 37 Trash TMDL at 13 (AR0003018) (“Consistent monitoring methodologies were used in Maryland and the District.”). 38 Trash TMDL at 16–19 (AR0003021–24). 39 Trash TMDL at 20 (AR0003025). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 20 of 50 13 trash sampling at 46 locations once every season between August 2007 and June 2008.40 Between June 2008 and April 2009, Maryland conducted in-stream trash sampling at 30 monitoring sites, 15 in Montgomery County and 15 in Prince George’s County.41 The District and Maryland also considered what trash was likely to come from point sources (sources with a NPDES permit, like a sewer system) and what trash was likely to come from a nonpoint sources (diffuse, non-permitted sources).42 The District and Maryland decided that for purposes of establishing wasteload allocations and load allocations, all trash small enough to fit through the storm drain system would be treated as coming from point sources (that is, the municipal separate storm sewer systems known as MS4s and the combined sewer overflows known as CSOs).43 On the other hand, the District and Maryland determined that the nonpoint source load would be all trash that is too large to be transported through the storm drain systems—trash that came from “[a]ccidental or intentional dumping of materials, such as construction materials, vehicles, appliances and bricks.”44 2. Calculation of the Baseline Load of Trash Next, the District and Maryland calculated the baseline load of trash for point and nonpoint sources for purposes of establishing the Trash TMDL. This baseline load of trash, calculated during periods of rainfall ranging from average to higher than average, represented the typical annual trash load that would be added to the Anacostia River.45 40 Trash TMDL at 20 (AR0003025). 41 Trash TMDL at 24–25 (AR0003029–30). 42 Trash TMDL at 27–28 (AR0003032–33). 43 Trash TMDL at 28 (AR0003033). 44 Trash TMDL at 28 (AR0003033). 45 Trash TMDL at 39 (AR0003044) (defining the “baseline load” as “the annual trash load calculated from monitoring data obtained through storm drain and CSO monitoring and in-stream sampling”). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 21 of 50 14 The District and Maryland established the maximum baseline loading rates for point sources by taking all of the trash monitoring data that they collected and normalizing it into pounds per acre per year.46 This baseline load calculation involved a complex series of calculations with conservative assumptions. First, the jurisdictions established a unit loading rate, which is the maximum pounds of trash per acre per inch of rain for each sampling event at each monitoring location, representing the loading rate from various land uses.47 Next, the jurisdictions obtained an annual loading for each land use by multiplying the unit loading rate (pounds per acre per inch of rain) by the average annual rainfall of 39.13 inches per year.48 The average annual rainfall figure was conservatively based on the mean annual rainfall at Washington Reagan National Airport over the 50-year period from 1959 to 2008.49 Lastly, the jurisdictions conservatively applied the annual loading rate for each land use to all acres within that land use in the watershed to calculate an annual load for each land use.50 The District and Maryland also established baseline loading rates for trash from nonpoint sources following a methodology detailed in the TMDL.51 3. TMDL Establishment Finally, the District and Maryland established the total maximum daily load of trash for the Anacostia River. They recognized that the TMDL “is the total amount of pollutant that can be 46 Trash TMDL at 29 (AR0003034); id. at 32 (AR0003037) (although Maryland considered using an alternative method of developing point source loading rates through a regression model, the State ultimately chose to use the same methodology as the District). 47 Trash TMDL at 29 (AR0003034). 48 Trash TMDL at 29 (AR0003034). 49 Trash TMDL at 29 (AR0003034). 50 Trash TMDL at 44 (AR0003049) (“The [wasteload allocations] are conservative estimates of actual loads because they were calculated under the assumption that all land in the watershed (including non-point source lands not regulated under NPDES stormwater permits) contributes to the point source trash load.”). 51 Trash TMDL at 31–32, 36–38 (AR0003036–37, AR0003041–43). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 22 of 50 15 assimilated by the receiving waterbody while still achieving water quality standards or goals.”52 As they explained, a TMDL is “composed of the sum of individual [wasteload allocations] for point sources and [load allocations] for nonpoint sources and natural background levels,” and must include a margin of safety “to account for any uncertainty in the relationship between pollutant loads and the quality of the receiving waterbody.”53 The District and Maryland explained that their narrative water quality criteria were expressed in subjective terms, such as “objectionable, nuisance, and unsightly,” because “an objective, quantifiable threshold cannot be developed.”54 Accordingly, the District and Maryland expressed the TMDL, consistent with the applicable water quality criteria, as “the quantity of trash that must be captured or removed for the waterbody to achieve the narrative criteria, rather than as an amount of trash that can be added to the waterbody without being objectionable, unsightly or constituting a nuisance.”55 Thus, the jurisdictions set the total maximum daily load to be “equal to 100 percent removal of the baseline load [of trash], calculated as an average (because of high seasonal and annual variability) of the measured or estimated removal rate.”56 In setting the total maximum daily load equal to prevention, capture, or removal of 100 percent of the baseline load (plus a five percent margin of safety), the District and Maryland noted that this target was “not the same as zero (0) trash in the waterway, but it should result in compliance with the narrative standard, as determined by the agencies responsible for interpreting the standard.”57 Moreover, the jurisdictions found that “[t]his target provides an objective and 52 Trash TMDL at 39 (AR0003044). 53 Trash TMDL at 39 (AR0003044). 54 Trash TMDL at 39 (AR0003044) (italics omitted). 55 Trash TMDL at 39 (AR0003044). 56 Trash TMDL at 39 (AR0003044). 57 Trash TMDL at 39 (AR0003044). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 23 of 50 16 measurable basis for compliance, consistent with stormwater and other discharge permits.”58 The jurisdictions acknowledged that “there might be a quantity of trash that could be discharged to the Anacostia River before being deemed by the general public as objectionable.”59 But they concluded that it was “not necessary to calculate that quantity for purposes of this TMDL” because “[w]hatever that level might be,” they had “concluded that removal of 100 percent of the baseline load would achieve the applicable narrative water quality criteria” and “would be sufficient to avoid interference with designated uses.”60 C. Consideration of Public Comments on the Trash TMDL As discussed above, the Trash TMDL was developed through a collaborative, inter- jurisdictional effort that included substantial input from stakeholders. During the public notice and comment process, Maryland and the District received a number of comments, some contending that the Trash TMDL was not strict enough and some contending that it was too stringent.61 Some commenters, including NRDC, argued that the Trash TMDL should be set at a more stringent level of zero trash. In their view, the Clean Water Act and EPA’s regulations “require this TMDL to be expressed as a number that represents the highest amount of trash pollution that is allowed to enter the river in compliance with water quality standards.”62 According to NRDC, “[t]he solution is obvious – set a TMDL equivalent to zero.”63 Other commenters, including ones regulated under NPDES permits, objected that the Trash TMDL was too stringent.64 For example, the District Water and Sewer Authority contended that setting the TMDL target at 100 percent removal 58 Trash TMDL at 39 (AR0003044). 59 Trash TMDL at 39 (AR0003044). 60 Trash TMDL at 39 (AR0003044). 61 Comment Response Document at 1–32 (AR0003073–104). 62 Comment Response Document at 10 (AR0003082). 63 Comment Response Document at 11–12 (AR0003083–84). 64 Comment Response Document at 2–4 (AR0003074–76). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 24 of 50 17 of the baseline trash load could be “effectively imposing a no-trash discharge standard for the combined sewer system,” and this created practical and legal problems because the “technology and resources simply do not exist to eliminate the discharge of all trash from the [combined sewer system].”65 Similarly, the Maryland Association of Municipal Wastewater Agencies and the Stormwater Association of Maryland asserted that Maryland’s water quality standards “do not require that Maryland’s waters are entirely free of an annual average amount of trash. This is simply not achievable, and [Maryland’s] decision to go to such an extreme is unwarranted.”66 Thus, the Associations recommended that Maryland “impose a much lower trash removal (or prevention) target.”67 The District and Maryland disagreed with both the not-strict-enough and too-strict views. They explained that the TMDL was set at a level designed to achieve their narrative water quality standards and that those standards did not require that the TMDL be set at a “zero trash” level.68 D. EPA’s Approval of the Trash TMDL Maryland and the District submitted the Trash TMDL to EPA on September 8, 2010 and September 15, 2010, respectively.69 EPA reviewed the Trash TMDL for consistency with the applicable regulatory requirements, including: (1) that the TMDL is “designed to implement applicable water quality standards”; and (2) that the TMDL “include[s] a total allowable load as well as individual wasteload allocations (WLAs) and load allocations (LAs).”70 65 Comment Response Document at 2 (AR0003074). The U.S. Department of Defense Regional Environmental Coordinator noted that “[i]n effect, the TMDL becomes a goal of zero trash in the Anacostia” and, “while this is a worthy goal, from a science based standpoint zero trash in the Anacostia is not necessary.” Comment Response Document at 7 (AR0003079). 66 Comment Response Document at 2–4 (AR0003074–76). 67 Comment Response Document at 10 (AR0003082). 68 Comment Response Document at 2–7 (AR0003074–79). 69 Decision Rationale at 1 (AR0003113). 70 Decision Rationale at 1 (AR0003113). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 25 of 50 18 EPA recognized that the Trash TMDL, unlike most TMDLs, was expressed in preventative rather than allowable terms. In other words, the Trash TMDL was expressed “in terms of quantities of trash that must be captured, prevented from entering, or removed from the waterbody.”71 EPA concluded that, “given the nature of trash and how it is transported to the waterbody,” the preventative expression of the trash TMDL was an “appropriate measure” under EPA’s regulations.72 EPA also concluded that the Trash TMDL’s endpoint of removing 100 percent of the baseline load of trash (plus an additional five percent margin of safety) was a proper numeric interpretation of the District’s and Maryland’s narrative water quality standards.73 After completing its review and analysis of the TMDL submitted by the District and Maryland, EPA approved the TMDL on September 21, 2010, within the 30-day period allowed by the Clean Water Act. 71 Decision Rationale at 1 (AR0003113) (emphasis added). 72 Decision Rationale at 8–9 (AR0003120–21) (quoting and discussing the definition of “total maximum daily load” at 40 C.F.R. § 130.2(i)); id. at 2 (AR0003114) (concluding that the maximum load is “appropriate to the pollutant” and appropriate to the “water quality conditions to be address”). 73 Decision Rationale at 8 (AR0003120). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 26 of 50 19 STANDARD OF REVIEW Under the Administrative Procedure Act (APA), the Court may reverse EPA’s action only if it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). This standard “is narrow and a court is not to substitute its judgment for that of the agency.” Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983). EPA acts arbitrarily if it “entirely failed to consider an important aspect of the problem” or “offered an explanation for its decision that runs counter to the evidence before the agency.” Am. Petroleum Inst. v. EPA, 684 F.3d 1342, 1350 (D.C. Cir. 2012) (citations and quotations omitted). Where EPA has considered the relevant factors and articulated a rational connection between the facts found and the choices made, its decisions should be upheld. State Farm, 463 U.S. at 43. In making this determination, the Court’s review is limited to the administrative record. See TOMAC v. Norton, 193 F. Supp. 2d 182, 194 (D.D.C. 2002) (citation omitted). Agency determinations based on technical matters, in particular, are “entitled to great deference.” West Virginia v. EPA, 362 F.3d 861, 867 (D.C. Cir. 2004) (internal quotation marks omitted). The plaintiff bears the burden to demonstrate that an agency has acted arbitrarily. Banner Health v. Sebelius, 715 F. Supp. 2d 142, 153 (D.D.C. 2010). In reviewing EPA’s interpretation of the Clean Water Act, the Court must apply the statute’s plain language where it reflects “the unambiguously expressed intent of Congress.” Chevron, U.S.A., Inc. v. NRDC, Inc., 467 U.S. 837, 842–43 (1984). But where the Act is “silent or ambiguous with respect to the specific issue,” the Court defers to EPA’s interpretation if it is “based on a permissible construction” of the Act. Id. at 843. Courts also give substantial deference to EPA’s interpretation of its regulations; so long as EPA’s interpretation “does not violate the Constitution or a federal statute,” the interpretation is entitled to “controlling weight unless it is plainly erroneous or inconsistent with the regulation.” Nat’l Wildlife Fed’n v. Browner, 127 F.3d 1126, 1129 (D.C. Cir. 1997) (citations and quotations omitted); see also Auer v. Robbins, 519 U.S. 452, 461 (1997). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 27 of 50 20 ARGUMENT I. EPA’s approval of the Trash TMDL is consistent with the Clean Water Act’s language and EPA’s implementing regulations. NRDC’s leading argument is that EPA exceeded its statutory authority. NRDC Br. at 15– 19). Boiled down, that argument rests on three propositions, only the first of which is correct. NRDC’s first proposition (Br. at 15–19) is that the Clean Water Act and EPA’s regulations both require TMDLs to include maximum loads. This is undisputed. But NRDC’s second proposition (Br. at 17) is that the terms “maximum” and “load” each have a single, unambiguous meaning in Section 1313(d)(1)(C) of the Act. That reading of the Act is incorrect. NRDC’s third proposition (Br. at 19–22) is that the Trash TMDL does not set a maximum load for trash. That is also incorrect. The Trash TMDL does set a maximum load for trash, and EPA reasonably concluded that the maximum load is consistent with the Clean Water Act and its implementing regulations. A. The phrase “total maximum daily load” in Section 1313(d)(1)(C) of the Act is ambiguous. Under Chevron Step One, if Congress’ intent is “unambiguously expressed” in the Clean Water Act, then Court should enforce its plain meaning, but if the Act is ambiguous, then under Chevron Step Two, the Court should defer to EPA’s interpretation, if reasonable. Chevron, 467 U.S. at 842–43. Here, the requirement in Section 1313(d)(1)(C) that States establish the “total maximum daily load” for impaired waters is not an unambiguous command to be carried out mechanically. Rather, placed in context, the phrase is ambiguous. As a threshold matter, Congress did not define the phrase anywhere in the Act. The only further clarity is provided later in Section 1313(d)(1)(C), where Congress put minimal flesh on the statutory bones by stating that “[s]uch load shall be established at a level necessary to implement the applicable water quality standards with seasonal variations and a margin of safety which takes into account any lack of knowledge concerning the relationship between effluent limitations and water Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 28 of 50 21 quality.” 33 U.S.C. § 1313(d)(1)(C). Thus, the only express parameters that Congress placed on a “total maximum daily load” are that it must be (1) established at a level (2) that is necessary to implement the applicable water quality standards with (3) seasonal variations and (4) a margin of safety. Id. In all other respects, Congress left EPA with discretion to determine how to create a “total maximum daily load” for a waterbody. Courts agree that this phrase is ambiguous. Facing “challenges from both environmental and industry groups,” several courts have affirmed EPA’s “authority to fill the Clean Water Act’s considerable gaps on how to promulgate a ‘total maximum daily load.’” Am. Farm Bureau, 792 F.3d at 296 (collecting cases). As the Second Circuit noted when it held that the phrase was ambiguous, “Congress, in one sentence, directs EPA to approve TMDLs for hundreds of different pollutants in thousands of different waterbodies, and it is excessively formalistic to suggest that EPA may not express these standards in different ways, as appropriate to each unique circumstance.” Muszynski, 268 F.3d at 97. More recently, in American Farm Bureau, the Third Circuit rejected an industry argument that the words “total load” had an unambiguous meaning. Id. at 297. The Third Circuit reasoned that “Congress was ambiguous on the content of the words ‘total maximum daily load’: they are not defined in the statute, and ‘total’ is susceptible to multiple interpretations.” Id. at 298; see also id. (“[A]lthough Congress explicitly required the EPA to establish ‘total maximum daily loads,’ it nowhere prescribed how the EPA is to do so.”). NRDC contends (Br. at 15–22) that Section 1313(d)(1)(C) is unambiguous and so EPA’s approval of the Trash TMDL is foreclosed at Chevron Step One. According to NRDC, “[b]ecause Congress has ‘directly spoken’ to the question of whether TMDLs must include maximum loads, EPA cannot approve TMDLs lacking maximum loads without violating Clean Water Act § 303(d).” NRDC Br. at 16 (quoting Chevron, 467 U.S. at 842–43). Without question, Congress directly spoke to the threshold question of whether a TMDL must include a maximum load. It must. NRDC’s Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 29 of 50 22 statutory argument, however, rests on the flawed follow-up premise that Congress directly spoke to how a maximum load must be expressed in a TMDL. It did not. NRDC insists that the terms “maximum” and “load” in Section 1313(d)(1)(C) can only have a single plain meaning. To reach this outcome, NRDC reads the terms “maximum” and “load” in isolation, focusing on a dictionary definition of “maximum” and EPA’s regulatory definition of “load.” NRDC Br. at 17. NRDC asserts that the plain meaning of “maximum load” in the phrase “total maximum daily load” is as follows: “the highest amount of pollution that can be added to a waterbody while still ‘implemen[ting] the applicable water quality standards.’” Id. at 17 (quoting 33 U.S.C. § 1313(d)(1)(C)). NRDC’s piecemeal approach to statutory interpretation is incorrect. “In making the threshold determination under Chevron,” courts do not restrict their inquiry to “examining a particular statutory provision in isolation” because “[t]he meaning—or ambiguity—of certain words or phrases may only become evident when placed in context.” Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 666–67 (2007) (citations and quotations omitted). And as discussed above, in context, the phrase “total maximum daily load” is ambiguous. Just as in the cases cited above, NRDC’s “textual argument at [Chevron] Step One fails to persuade,” Am. Farm Bureau, 792 F.3d at 298. Indeed, NRDC concedes (Br. at 17) that it cannot point to any case that interprets “maximum load” in Section 1313(d) to have the single plain meaning that NRDC has assigned to the terms at Chevron Step One. To be sure, the D.C. Circuit split with the Second Circuit when it held that the term “daily” in Section 1313(d)(1)(C) is susceptible of only one ordinary meaning. Compare Friends of the Earth v. EPA, 446 F.3d 140, 144 (D.C. Cir. 2006) with Muszynski, 268 F.3d at 97. But this “split of authority on . . . the word ‘daily,’” American Farm Bureau, 792 F.3d at 297, does not control here, where the meaning of two other words in the phrase is at issue. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 30 of 50 23 Lacking a clear statutory definition, EPA has defined “total maximum daily load” in its implementing regulations, which are not subject to challenge in this case. There, EPA defines the phrase to mean “[t]he sum of the individual [wasteload allocations] for point sources and [load allocations] for nonpoint sources and natural background.” 40 C.F.R. § 130.2(i). EPA’s definition proceeds to stress the importance of flexibility in expressing a total maximum daily load: “TMDLs can be expressed in terms of either mass per time, toxicity, or other appropriate measure.” Id. As discussed next, EPA reasonably interpreted and applied this regulatory definition in approving the Trash TMDL. B. EPA’s approval of the Trash TMDL is based on a reasonable interpretation of the ambiguous language in Section 1313(d)(1)(C) and a reasonable application of its own regulations. Because the language of Section 1313(d)(1)(C) is ambiguous, the Court should defer to EPA’s interpretation as long as it is reasonable. Chevron, 467 U.S. at 843. Here, EPA’s interpretation is reasonably drawn from its Clean Water Act regulations, which expressly define the phrase “total maximum daily load.” See American Farm Bureau, 792 F.3d at 298 (the TMDL requirements in Section 1313(d)(1)(C) “taken together, tend to suggest that ‘total maximum daily load’ is a term of art meant to be fleshed out by regulation, and certainly something more than a number”). When EPA applies its own regulations, the Court should defer to the Agency’s interpretation “unless that interpretation is plainly erroneous or inconsistent with the regulation.” Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct. 1326, 1337 (2013); see id. (EPA’s interpretation of its Clean Water Act regulations “need not be the only possible reading . . . or even the best one.”). Under these governing principles, and given the unique nature of trash as a pollutant and the way that trash reaches the Anacostia River, EPA’s approval of the Trash TMDL should be upheld. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 31 of 50 24 1. The Trash TMDL sets a maximum load for trash. Consistent with the Clean Water Act—and contrary to NRDC’s claims (Br. at 19–22)—the Trash TMDL fulfills the requirement in Section 1313(d)(1)(C) to set a total maximum daily load for trash. Decision Rationale at 2–3, 8 (AR0003114–15, AR0003120). For Maryland, the total maximum daily load is defined as prevention, capture, or removal of 2,793.9 pounds of trash per day. Id. at 2 (AR0003114). For the District, the total maximum daily load is defined as prevention, capture, or removal of 665 pounds of trash per day. Id. at 3 (AR0003115). Thus, in numeric terms, the total maximum daily load in the Trash TMDL for both jurisdictions is 3,458.9 pounds of trash to be prevented from entering, captured, or removed from the Anacostia River on a daily basis. Id. at 2–3 (AR0003114–15). This amounts to an annual prevention, capture, or removal target of over 1.2 million pounds of trash—1,262,492 pounds, to be precise. Id. Maryland and the District developed this maximum load based on a sound methodology. Among other steps, they (1) monitored the trash running off the land into the River; (2) calculated the average amount of trash discharged for each land type per acre per rainfall event; (3) calculated the annual loading rate based on that data; and (4) applied that rate to develop a baseline load that represented a typical annual and daily load of trash from each source to the River. Trash TMDL at 29 (AR0003034). Then they used their best professional judgment to determine that 100 percent removal or capture of the baseline load would attain their narrative water quality standards. Decision Rationale at 8 (AR0003120). Next, the jurisdictions expressed the total maximum daily load as prevention, capture, or removal of 100 percent of the baseline load of trash. Id. Finally, the jurisdictions concluded that an additional 5 percent of the baseline load of trash should be prevented, captured, or removed to provide an adequate margin of safety. Id. at 11 (AR0003123). Thus, the Trash TMDL set the total maximum daily load as a numeric figure that requires prevention, capture, or removal of 105 percent of the baseline load of trash to the River. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 32 of 50 25 2. EPA reasonably interpreted the Act and its regulations to allow the Trash TMDL to express the maximum load as the amount of trash to be prevented from entering the River. NRDC’s true quarrel appears to be with the expression of the TMDL as an amount of trash that must be prevented from reaching the Anacostia River (in this case, 105 percent of the average annual baseline load of trash to the River) rather than an amount that can be added to the River. NRDC Br. at 19. EPA concluded, however, that this preventative expression of the maximum load was consistent both with the Clean Water Act and with the Agency’s implementing regulations, which provide that a TMDL may be expressed “in terms of either mass per time, toxicity, or other appropriate measure.” 40 C.F.R. § 130.2(i) (emphasis added); see Decision Rationale at 8-9 (AR0003120- 21) (explaining that the Trash TMDL did “include a total allowable load” and was properly “expressed as the quantity of trash that must be captured, prevented from entering, or removed” for the Anacostia River “to achieve the narrative criteria” in the water quality standards set by Maryland and the District.). EPA first concluded that the Trash TMDL satisfied the requirements in Section 1313(d)(1)(C) of the Act, which requires loads to be “established at a level necessary to implement the applicable water quality standards with seasonal variations and a margin of safety.” 33 U.S.C. § 1313(d)(1)(C). Decision Rationale at 6–11 (AR0003118–23). EPA found that the jurisdictions had correctly concluded that achieving the TMDL’s target of 105 percent prevention or removal of trash would “result in compliance with their narrative water quality standards.” Id. at 6–8 (AR0003118– 20). And EPA concluded that the Trash TMDL properly accounted for seasonal variations and included explicit and implicit margins of safety, as required by the Act. Id. at 11 (AR00023). EPA also interpreted and applied its regulations to conclude that the Trash TMDL properly identified a total allowable load that was expressed as a quantity of trash that had to be captured, prevented from entering, or removed from the River. Decision Rationale at 8 (AR0003120) (quoting Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 33 of 50 26 33 U.S.C. § 1313(d)(1)(C)). Those implementing regulations define the phrase “total maximum daily load” (which is undefined in the statute) as “the sum of individual [wasteload allocations] for point sources, [load allocations] for any nonpoint sources, and any natural background.” Id. at 9 (AR0003121) (citing 40 C.F.R. § 130.2(i)). Here, EPA correctly concluded that the District and Maryland had expressed the maximum load in the Trash TMDL as the sum of the daily and annual wasteload allocations and load allocations for the Anacostia River. Id.; see id. at 2 (AR0003114) (Tables 1 through 4). EPA recognized that the maximum load in the Trash TMDL is not expressed, as are most TMDLs, as an allowable load of the pollutant that can be added to the waterbody without violating the water quality criteria. Decision Rationale at 8 (AR0003120). Instead, after considering how to account for the unique characteristics of trash pollution and how to translate the narrative water quality standards into an objective, measurable target, the District and Maryland concluded that the maximum load was best expressed in preventative terms as the amount of trash that must be removed, captured, or otherwise prevented from reaching the Anacostia River. Id. In reviewing this preventative expression of the maximum load as removal of 105 percent of the baseline load of trash, EPA stressed that its regulatory definition of the phrase “total maximum daily load” provides “flexibility on how the TMDLs can be expressed in terms of ‘either mass per time, toxicity, or other appropriate measures.’” Decision Rationale at 8 (AR0003120) (quoting 40 C.F.R. § 130.2(i)). And EPA concluded that the District and Maryland had properly exercised that flexibility in the Trash TMDL because, “given the nature of trash and how it is transported to the waterbody, expression of the [wasteload allocations] and [load allocations] in terms of trash to be captured, prevented from entering, or removed for the waterbody is an ‘appropriate measure.’” Id. at 8–9 (AR0003120–21) (quoting 40 C.F.R. § 130.2(i)). EPA further concluded that the preventative expression of the total maximum load was consistent with the requirement in its regulation that total Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 34 of 50 27 allowable load be expressed as the sum of the individual wasteload allocations and load allocations. Id. at 9 (AR0003121) (citing 40 C.F.R. § 130.2(i)). EPA’s conclusion that the preventative expression of the maximum load in the Trash TMDL is an “appropriate measure” and that it is consistent with the Agency’s regulatory definition of the phrase “total maximum daily load” should be accorded controlling weight. See Nat’l Wildlife Fed’n, 127 F.3d at 1129. EPA’s recognition of flexible TMDL development approaches is not novel. In a 1999 technical guidance document for developing sediment TMDLs, EPA stated that “[a]s an alternative to maximum allowable loads, allocations can be expressed in terms of percentage reductions in sediment loading allocated among sources . . . .” Protocol for Developing Sediments TMDLs at 7-4 (AR0000650). According to EPA, “[p]ercent reduction targets enable the analyst to account for the uncertainties and variabilities in the analysis of dynamic watershed settings while providing a quantitative basis for allocations and subsequent monitoring.” Id. EPA also noted that “[f]or this approach to work, estimates of baseline sediment loading conditions by source are needed to determine the appropriate percentage reductions needed.” Id. Just as EPA’s 1999 guidance indicated that expressing a TMDL as a percentage reduction was an appropriate alternative approach for sediment TMDLs, EPA here concluded that expressing the total maximum daily load in preventative loading terms was an appropriate alternative for this Trash TMDL. NRDC argues (Br. at 20) that EPA’s approval of the Trash TMDL renders the word “maximum” in Section 1313(d) meaningless. But as EPA concluded, the Trash TMDL appropriately expressed the maximum load as a preventative figure—that is, the maximum load is a target to remove 100 percent of the baseline load (plus a five percent margin of safety). Decision Rationale at 2, 8 (AR0003114, AR0003120). Given the unique nature of trash and the way that trash is transported to the Anacostia River, this is a reasonable interpretation of “maximum.” Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 35 of 50 28 NRDC further argues that EPA’s regulations “foreclose a TMDL from being expressed as an amount of pollution that must be removed from a waterbody.” NRDC Br. at 18. EPA’s definitions of wasteload allocations and load allocations incorporate the term “loading capacity,” 40 C.F.R. § 130.2(g), (h). In turn, EPA defines “loading capacity” as “[t]he greatest amount of loading that a water can receive without violating water quality standards.” Id. § 130.2(f). And EPA defines “load” or “loading” as “[a]n amount of matter or thermal energy that is introduced into a receiving water.” Id. § 130.2(e). From these definitions, NRDC argues that a “TMDL without a loading capacity does not meet EPA’s definition of a TMDL,” and contends that the wasteload allocations and load allocations are contrary to EPA’s regulations because they “are not expressed as loads, but as pounds of trash to be removed from the river.” NRDC Br. at 18, 21. NRDC focuses on only one aspect of the TMDL, removal of trash from the River, without even acknowledging that the TMDL also reflects prevention and capture of trash that otherwise would be discharged into the River from point sources. Consistent with the regulatory definition of “load,” EPA concluded that preventing, capturing, or removing 105 percent of the trash would limit the amount of trash that would otherwise be introduced into the River. Decision Rationale at 8 (AR0003120) (quoting 40 C.F.R. § 130.2(i)). And consistent with the regulatory definition of “loading capacity,” EPA concluded that preventing, capturing, or removing 105 percent of the annual trash load would ensure that any residual amount of trash that the Anacostia River did receive would not violate the jurisdictions’ water quality standards. Id. at 8 (AR0003120). Finally, invoking the D.C. Circuit’s Friends of the Earth decision, NRDC seeks to dismiss EPA’s recognition of the nature of trash and how it is transported to the waterbody as a “policy justification” for approval of the Trash TMDL. NRDC Br. at 21 (quoting Friends of the Earth, 446 F.3d at 145–46). For all of the reasons discussed above however, EPA’s consideration of the nature of trash, and the means by which trash reaches the River, is not a policy justification for EPA’s Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 36 of 50 29 action—instead, that discussion accurately reflects the unique characteristics of trash as a pollutant. Having properly considered those characteristics, EPA reasonably interpreted the statutory phrase “total maximum daily load” and EPA’s own regulations to authorize the Trash TMDL’s expression of “maximum load” as a preventative, rather than additive, figure. C. EPA played an appropriate role in reviewing and approving the Trash TMDL developed by Maryland and the District. NRDC’s contention that EPA overstepped its authority in approving the Trash TMDL must be assessed in light of the limited role that the Clean Water Act places on EPA when reviewing State-established TMDLs. Consistent with the “vigorous federalism underlying the Clean Water Act,” United States v. Homestake Mining Co., 595 F.2d 421, 429 (8th Cir. 1979), the Act sets forth a highly cooperative system for the exercise of the States’ primary authority for setting water quality standards and, ultimately, TMDLs. See also 33 U.S.C. § 1251(b) (“It is the policy of the Congress to recognize, preserve, and protect the primary responsibilities and rights of States to prevent, reduce, and eliminate pollution, to plan the development and use (including restoration, preservation, and enhancement) of land and water resources, and to consult with the Administrator in the exercise of his authority under this chapter.”). Thus, the Act contemplates that Maryland and the District of Columbia will play the leading role in developing both water quality criteria and the TMDLs that aid in the implementation of those criteria for waterbodies within their boundaries. In contrast, EPA’s role with respect to State water quality management is typically one of oversight, in which it reviews State-established water quality criteria, 303(d) lists, and TMDLs to ensure conformance with federal statutory and regulatory requirements. 33 U.S.C. § 1313(d)(2). Courts have thus held that the Clean Water Act assigns EPA “a limited, non-rulemaking role in the establishment of water quality standards.” City of Albuquerque v. Browner, 97 F.3d 415, 425 (10th Cir. 1996); see also Meiburg, 296 F.3d at 1025 (States are primarily responsible for preparing lists of water- Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 37 of 50 30 quality-limited segments and corresponding TMDLs); NRDC v. EPA, 16 F.3d 1395, 1399 (4th Cir. 1993) (States have primary role in establishing appropriate water quality criteria, and EPA’s sole function is to review those standards for approval and determine whether the State’s decisions are scientifically defensible and protective of designated uses); District of Columbia v. Schramm, 631 F.2d 854, 860 (D.C. Cir. 1980) (primary duty of compiling water quality criteria rests with the State and the EPA merely retains a supervisory role). Although these cases involved EPA’s approval of water quality criteria, the same principle that EPA is acting primarily in an oversight capacity applies with equal force when EPA is approving a State-established TMDL. Water quality criteria define how clean a State’s waters must be; TMDLs are primarily informational tools that serve as a planning link between establishment and achievement of a State’s water quality standards. Pronsolino, 291 F.3d at 1129. As the regulatory bodies with the most relevant direct experience and knowledge about the local conditions in the Anacostia River, the District and Maryland were uniquely positioned to craft a blueprint for achieving their water quality standards through the Trash TMDL. To be sure, EPA worked extensively with the District and Maryland throughout the development of the Trash TMDL. But EPA’s focus during that process was on providing technical support and assistance to the District and Maryland, while allowing those jurisdictions to take the lead in crafting a TMDL that, in their expert judgment, would best help them meet their water quality standards. Once this process was complete, however, and the Trash TMDL was submitted to EPA, EPA’s role was limited to reviewing that TMDL for consistency with the Clean Water Act and EPA’s implementing regulations. As discussed in the preceding and following sections, EPA’s exercise of that role was based on reasonable statutory and regulatory interpretations and is fully supported by the administrative record. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 38 of 50 31 II. EPA rationally approved the District’s and Maryland’s determination that the Trash TMDL is set at a level necessary to implement their water quality standards. When EPA approved the District’s and Maryland’s conclusion that the Trash TMDL is set at a level necessary to implement the applicable narrative water quality standards, EPA provided a reasoned explanation for its approval, which is supported with substantial record evidence. Although NRDC challenges that decision, it has not met its burden to show that EPA’s approval is arbitrary and capricious. Banner Health, 715 F. Supp. 2d at 153. A. EPA properly approved the District’s and Maryland’s conclusion that removal of the baseline load of trash, with a margin of safety, would ensure compliance with the water quality standards. In the Decision Rationale, EPA concluded that the Trash TMDL was established at a “level necessary to implement the applicable water quality standards,” 33 U.S.C. § 1313(d)(1)(C). NRDC disputes that conclusion (Br. at 23–25), contending that EPA never determined the necessary “level.” But in the Trash TMDL, the District and Maryland did identify the level of trash loading that they concluded, in their best professional judgment, was necessary to satisfy their water quality standards—removal of 100 percent of the baseline load, with an additional 5 percent margin of safety to account for uncertainty. And EPA rationally explained why its approval of that level was supported by the record. In the Trash TMDL, the District and Maryland described how the level of trash loading they selected would achieve their water quality criteria. Because those criteria are in narrative form, the jurisdictions had to translate those narrative requirements into a concrete quantitative value. Trash TMDL at 11 (AR0003016). The narrative water quality criteria used subjective terms such as “objectionable, nuisance, and unsightly” to describe the level of trash that is aesthetically unacceptable in the Anacostia River. Id. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 39 of 50 32 Much of the public comment on the draft TMDL centered around whether the only target for the TMDL that would be consistent with the narrative water quality criteria would be zero trash in the Anacostia River. Comment Response Document at 1–31 (AR0003073–3104). Maryland and the District concluded, however, that “the more logical reading” of their narrative water quality criteria “is that they do not require an endpoint of zero.” Comment Response Document at 5 (AR0003077). Both jurisdictions noted that an interpretation requiring a zero trash target would not account for terms in their standards such as “amounts that . . . form objectionable deposits,” “amounts that . . . form nuisances,” and “amounts sufficient to . . . [create a nuisance]”—terms that necessarily assume that some amount of trash may be present in the River. Id. The jurisdictions also recognized the difficulty of quantifying these subjective terms in precise numeric figures. They noted EPA’s guidance (known as the Gold Book), which explained that aesthetic uses can “vary within the minds of individuals encountering the waterway.” Trash TMDL at 11 (AR0003016). To reconcile their narrative criteria with the need to set the TMDL at an objective and measurable “level necessary to implement” those criteria, the District and Maryland chose to express that level as “the quantity of trash that must be captured or removed for the waterbody to achieve the narrative criteria, rather than as the amount of trash that can be added to the waterbody without being objectionable, unsightly or constituting a nuisance.” Trash TMDL at 11 (AR0003016). The jurisdictions were aware that this approach would not result in “zero trash in the water.” Comment Response Document at 3 (AR0003075). But it was their best professional judgment that such amounts would not preclude achievement of the subjective narrative water quality criteria. Trash TMDL at 11 (AR 0003016); Comment Response Document at 5 (AR0003077). The record also reflects that the District and Maryland calculated the baseline load of trash conservatively as the average annual load of trash in the River, which accounted for high seasonal and annual variability, including critical conditions caused by high rainfall. Trash TMDL at ix–x Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 40 of 50 33 (AR0003002–03). In particular, the jurisdictions made sure that the baseline load accounted for “high flow events because these events represent conditions during which trash is most easily transported to and through streams and the sewer system.” Id. By collecting data over four seasons, at a large number of sites across the Anacostia watershed, during a range of rainfall events, the District and Maryland accounted for higher than average rainfall, as well as possible localized seasonal variations in trash loading. Id. After the jurisdictions set the TMDL endpoint at 100 percent removal of the baseline load, they also included explicit and implicit margins of safety. Id. at xi (AR0003004). The District and Maryland set an explicit margin of safety by requiring that another 5 percent of the baseline load be removed or prevent from entering the River. Id. That explicit margin accounted for uncertainties and lack of knowledge regarding the relationship between trash loadings to the River and the jurisdictions’ water quality standards. Id. The District and Maryland also established an implicit margin of safety by making conservative assumptions in calculating the wasteload allocations and load allocations. Id. (describing the conservative assumptions). The explicit and implicit margins of safety further ensure the Trash TMDL accounts for seasonal variability in the annual trash loading. NRDC argues (Br. at 23–24) that the District and Maryland failed to give a reasoned explanation for their decision to choose a level of 100 percent removal of the baseline load because, in setting that level, the jurisdictions invoked their best professional judgment. But the District and Maryland are entitled to exercise judgment when they are interpreting the water quality criteria that they are responsible for developing. Comment Response Document at 5 (AR0003077). And exercising that judgment is all the more appropriate where the criteria they are applying are couched in subjective terms that lack mathematical precision. See also Muszynski, 268 F.3d at 103 (“But simply to reject EPA’s efforts to implement the [Clean Water Act] because it must respond to real water quality problems without the guidance of a rigorously precise methodology would essentially nullify Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 41 of 50 34 the exercise of agency discretion in the form of ‘best professional judgment.’”). As the District and Maryland observed, the narrative terms in the water quality criteria “must be assigned meaning.” Comment Response Document at 5 (AR0003077). Next, NRDC points to Anacostia Riverkeeper to support its view that it is improper to exercise best professional judgment to apply narrative water quality standards. See NRDC Br. at 24 (citing and quoting Anacostia Riverkeeper, 798 F. Supp. 2d at 237–39). But in Anacostia Riverkeeper, the court did not fault EPA for exercising judgment. Instead, the court rejected EPA’s decision because the Agency had entirely failed to “consider whether the Final TMDL would protect designated uses or satisfy water quality criteria related to recreational or aesthetic uses in the Anacostia River.” 798 F. Supp. 2d at 238. Although the court found these “omissions” were “fatal,” id., the court affirmed that an agency’s judgment plays an important role in decision-making: “where the Agency determines—through exercise of its expert judgment and scientific review—that a proposed TMDL will meet all applicable water quality standards, the Court will generally defer to that judgment.” Id. NRDC also contends that EPA’s conclusion that removal of the baseline load of trash will implement the water quality standards is contradicted by the jurisdictions’ “factual finding that trash loading to the Anacostia is highly variable from year to year.” NRDC Br. at 24. NRDC has misinterpreted the Trash TMDL. What the District and Maryland actually explained is that the baseline load was “calculated as an average (because of high seasonal and annual variability) of the measured or estimated removal rate.” Trash TMDL at 11 (AR0003016). In other words, the jurisdictions selected an average removal rate precisely to account for variability in trash levels. Id. Without record support, NRDC also claims (Br. at 24) that because of seasonal variability, “in some years, trash loading will exceed the baseline load significantly.” But as discussed above, NRDC fails to acknowledge that the District and Maryland took several measures to ensure that the Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 42 of 50 35 baseline load of trash was a conservative calculation that accounted for critical conditions, including seasonal variations. Id. at 46 (AR0003051). Lacking record support, NRDC offers two hypothetical analogies to bolster its position. First, NRDC imagines a scenario in which 1,200 tons of trash are discharged into the Anacostia in a year but the baseline load is only 600 tons. NRDC Br. at 25. NRDC has not identified any evidence, either in its public comments or anywhere else in the record, to show that such a scenario is likely. Without record support, NRDC’s hypothetical lends no weight to its argument. NRDC’s crowded-room analogy is even more divorced from the record. In a footnote, NRDC posits that if the maximum capacity of a room were expressed as a number of people to be removed from the room or prevented from entering the room every hour, then there would be no way to ensure that the room does not become overcrowded. NRDC Br. at 25 n.7. But to make its analogy work, NRDC has to assume that no one knows how many people will enter the room every hour, or in terms that matter for this case, how much trash is entering the River every year. The record shows the opposite. The District and Maryland reasonably estimated how much trash is entering the River annually. Based on that information, they established the baseline load as an average figure that incorporated conservative assumptions. Trash TMDL at viii–xi (AR0003001–04). Then, they made sure that 105 percent of the baseline load would be prevented from reaching, or removed from, the River. In sum, NRDC’s analogies are not supported by record evidence. B. The baseline load is a conservative and reliable estimate of the average amount of trash loaded annually into the Anacostia River. Finally, NRDC challenges EPA’s conclusion that the baseline load of trash is a reliable estimate of the average amount of trash annually entering the Anacostia River, but the calculations are supported by the scientific record, and the Court should defer to EPA’s conclusion that the baseline load is accurate. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 43 of 50 36 NRDC contends that the “actual amount of trash that enters the river annually is unknown.” NRDC Br. at 25. That perfect precision is not the standard. To the contrary, it is wholly reasonable for EPA (and the District and Maryland) to develop monitoring data that is representative of trash discharges into the Anacostia River and then to use that data to estimate the baseline load. “EPA typically has wide latitude in determining the extent of data-gathering necessary to solve a problem,” and a court “generally defer[s] to an agency’s decision to proceed on the basis of imperfect scientific information.” Sierra Club v. EPA, 167 F.3d 658, 662 (D.C. Cir. 1999) (citation omitted). Thus, “[t]he court must defer to EPA’s use of data, even if the data is imperfect, unless the data bears no rational relationship to the reality it purports to represent.” Am. Farm Bureau, 984 F. Supp. 2d at 343. Here, the monitoring data that the District and Maryland developed bears a rational relationship to trash discharges from the watershed into the Anacostia River. In the Trash TMDL, the jurisdictions explained each step that they took to develop data from stormwater outfall trash collection and monitoring and from in-stream trash counts that are representative of various categories of land uses. Trash TMDL at 13–28 (AR0003018–33). And the jurisdictions also explained how they used this data to establish the wasteload allocations and load allocations in the Trash TMDL. Id. at 29–38 (AR0003034–43). NRDC argues that the jurisdictions did not collect enough data, pointing out that the District and Maryland monitored only 18 out of 3,225 stormwater outfalls that discharge into the Anacostia River. NRDC Br. at 26. But NRDC offers no explanation as to how many outfalls it believes the jurisdictions should have monitored to obtain sufficiently accurate data. As the District and Maryland noted, it would be unreasonable to require the jurisdictions to monitor every single outfall in the Anacostia watershed “or even a majority” of the outfalls. Comment Response Document at 18 (AR0003090). To increase the accuracy and relevancy of the data, the jurisdictions selected outfall locations that “represented specific land use drainage areas.” Id. Neither in its Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 44 of 50 37 comments nor in its brief does NRDC offer any suggestion as to what would be an appropriately representative set of outfalls. To meet its burden, NRDC must do more than assert “not good enough,” especially in the face of the extensive reasoning and evidence deployed in the TMDL and approved by EPA. Where NRDC has offered nothing more than a blanket criticism of the data, the Court should defer to the reasoned explanation offered by the jurisdictions and reviewed and approved by EPA. Decision Rationale at 9–10 (AR0003121–22); see Pa. Dep’t of Envtl. Res. v. EPA, 932 F.2d 269, 272 (3d Cir. 1991), citing EPA v. Nat’l Crushed Stone Ass’n, 449 U.S. 64, 83 (1980) (“On the merits of EPA’s refusal to consider the updated . . . data, we defer to its expertise.”); Ethyl Corp. v. EPA, 541 F.2d 1, 36 (D.C. Cir. 1976) (“We must look at the decision not as the chemist, biologist or statistician . . . but as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality.”); see also Upper Blackstone Water Pollution Abatement Dist. v. EPA, 690 F.3d 9, 23 (1st Cir. 2012) (“In almost every case, more data can be collected, models further calibrated to match real world conditions; the hope or anticipation that better science will materialize is always present, to some degree, in the context of science-based agency decisionmaking. Congress was aware of this when it nonetheless set a firm deadline for issuing new permits.”); id. (“As in many science- based policymaking contexts, under the [Clean Water Act] the EPA is required to exercise its judgment even in the face of some scientific uncertainty.”). NRDC also claims that the monitoring data is unreliable because the loading rates for the same land uses vary widely between the District and Maryland. NRDC Br. at 26–27. But the District and Maryland explained that these differences are likely accounted for by “inherent differences in the jurisdictions,” including differences in the ways that the two jurisdictions define land use categories. Comment Response Document at 19 (AR0003091). The jurisdictions noted that differences between the jurisdictions’ respective loading rates for land that is used for transportation Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 45 of 50 38 were attributable to differences in the way that the two jurisdictions defined transportation in the geospatial data. Id. at 20 (AR0003092). For certain land uses such as open land, forest and agriculture, where the land use categories were compatible, the TMDL uses the District’s data to calculate the loading rates in Maryland. Id. at 19 (AR 0003091). The District and Maryland thus explained the technical reasons why they selected the land use loading rates that they did. And when EPA approved the Trash TMDL, it rationally reviewed and approved the land-use-based trash loading rates set by the District and Maryland. Decision Rationale at 9 (AR0003121). On technical and scientific questions such as whether the land use loading rates in the Trash TMDL are accurate, EPA is entitled to substantial deference. See Ctr. for Biological Diversity v. EPA, 749 F.3d 1079, 1087– 88 (D.C. Cir. 2014) (sampling circuit law). “[T]here is no question that the calculation of a TMDL is a ‘highly technical, specialized interstitial matter that Congress . . . delegate[d] to specialized agencies to decide.’” Am. Farm Bureau, 984 F. Supp. 2d at 318 (quoting Zuni Pub. Sch. Dist. v. Dep’t of Educ., 550 U.S. 81, 90 (2007)). Finally, NRDC asserts that the Trash TMDL fails to account for the possibility that a population increase in the future may lead to increased trash being discharged to the Anacostia. NRDC Br. at 27. In responding to a public comment by NRDC, the District and Maryland agreed that if the amount of trash entering the Anacostia substantially changed in future years, either increasing or decreasing, then the Trash TMDL should be revised to reflect the removal rate needed to achieve the water quality standards. Comment Response Document at 21 (AR0003093). But the Clean Water Act does not require States to frame TMDLs based on speculation regarding materially different conditions that might manifest themselves in the future. See, e.g., Am. Farm Bureau, 984 F. Supp. 2d at 328 (“Thus, the TMDL framework anticipates future modifications which can originate from either EPA or the states.”). Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 46 of 50 39 III. The Court should decline to grant the relief that NRDC requests. For all the reasons discussed above, the Court should hold that EPA’s approval of the Trash TMDL is within the Agency’s authority and is a rational decision that is supported by the record. If, however, the Court were to hold that EPA’s action is in some respect arbitrary and capricious under the APA, then the Court should decline to grant NRDC’s requested relief. Instead, the most appropriate relief in this case would be for the Court to order a remand without vacatur. First, the Court should decline NRDC’s request for an order that EPA must take a specific action on remand (issuing a federal TMDL for trash that complies with the Clean Water Act) within a specific timeframe (a “reasonable amount of time”). NRDC Br. at 28. The APA authorizes the Court to remand an agency action for further proceedings consistent with its opinion, and it can do so with or without vacating the agency action at issue. See 5 U.S.C. § 706(2) (reviewing court should “hold unlawful and set aside agency action” that is arbitrary or unlawful). But ordering EPA to take a specific agency action is not the proper remedy under the APA. Bennett v. Donovan, 703 F.3d 582, 589 (D.C. Cir. 2013) (“Appellants brought a complaint under the Administrative Procedure Act to set aside an unlawful agency action, and in such circumstances, it is the prerogative of the agency to decide in the first instance how best to provide relief.”) (citation omitted); N. Air Cargo v. U.S. Postal Serv., 674 F.3d 852, 861 (D.C. Cir. 2012) (“When a district court reverses agency action and determines that the agency acted unlawfully, ordinarily the appropriate course is simply to identify a legal error and then remand to the agency, because the role of the district court in such situations is to act as an appellate tribunal.”). Ordering EPA to issue a federal TMDL would also be inconsistent with the status of the Trash TMDL because remanding or even vacating EPA’s approval does not automatically undo the underlying submission of the Trash TMDL by the District and Maryland. Under Section 1313(d)(2) of the Act, EPA must first make a decision to disapprove a State TMDL; only then is the Agency required to develop its own TMDL. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 47 of 50 40 Second, the Court should also decline NRDC’s request to specify a timetable for EPA to act. See Natural Res. Def. Council v. EPA, 489 F.3d 1364, 1375 (D.C. Cir. 2007) (“We decline to set a two year limit on EPA’s proceedings on remand as the NRDC requests; mandamus affords a remedy for undue delay.”); North Carolina v. EPA, 550 F.3d 1176, 1178 (D.C. Cir. 2008) (per curiam) (declining invitation to “impose a definitive deadline by which EPA must correct [a Clean Air Act rule’s] flaws”). The Court can and should presume that in the event of a remand, EPA will act diligently to reach a final decision consistent with the Court’s decision. Third and finally, to extent that NRDC seeks relief that is available under the APA (vacatur), it asks the Court to immediately stay that vacatur. In Anacostia Riverkeepers, Inc. v. Jackson, 713 F. Supp. 2d 50, 51–55 (D.D.C. 2010), this Court granted a judicial stay of vacatur. Although such unusual relief may be appropriate in some circumstances, it should not be granted here, where a straightforward remand to the Agency without vacatur will accomplish the same purpose. See Allied- Signal, Inc. v. United States Nuclear Regulatory Comm’n, 988 F.2d 146, 150–51 (D.C. Cir. 1993) (“The decision whether to vacate depends on the seriousness of the . . . deficiencies (and thus the extent of doubt whether the agency chose correctly) and the disruptive consequences of an interim change that may itself be changed.”). In this case, remand without vacatur would be warranted because that relief will ensure that (1) the environmental benefits provided by the Trash TMDL remain in place and (2) EPA, the District, Maryland, and all of the entities affected by the Trash TMDL can continue to rely on it, as they have for the past six-plus years. Remand without vacatur also will allow EPA to either further explain its reasoning or to work cooperatively with the District and Maryland to give those jurisdictions the opportunity to address any deficiencies in the Trash TMDL in the first instance. Finally, affording the District and Maryland the initial opportunity to address the Trash TMDL would respect the Clean Water Act’s emphasis on State leadership in the TMDL process. Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 48 of 50 41 CONCLUSION For the foregoing reasons, the Court should grant EPA’s cross-motion for summary judgment and deny NRDC’s motion for summary judgment. Respectfully submitted, Dated: June 2, 2017 Of Counsel: James Curtin Office of General Counsel U.S. Environmental Protection Agency 1200 Pennsylvania Avenue, N.W. Washington, D.C. 20460 Stefania Shamet Kelly A. Gable Office of Regional Counsel U.S. Environmental Protection Agency, Region 3 1650 Arch Street Philadelphia, PA 19103 JEFFREY H. WOOD Acting Assistant Attorney General Environment and Natural Resources Division /s/ Justin D. Heminger JUSTIN D. HEMINGER Senior Attorney Environmental Defense Section U.S. Department of Justice P.O. Box 7611 Washington, D.C. 20044 (202) 514-2689 justin.heminger@usdoj.gov Counsel for Defendant EPA Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 49 of 50 CERTIFICATE OF SERVICE I hereby certify that on June 2, 2017, I filed the foregoing using the Court’s CM/ECF system, which will electronically serve all counsel of record registered to use the CM/ECF system. /s/ Justin D. Heminger Case 1:16-cv-01861-JDB Document 15 Filed 06/02/17 Page 50 of 50 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA ____________________________________ ) NATURAL RESOURCES DEFENSE ) COUNCIL, INC. ) ) Plaintiff, ) ) Civil Action No. 1:16-cv-01861-JDB v. ) ) UNITED STATES ENVIRONMENTAL ) PROTECTION AGENCY, et al., ) ) Defendants. ) ____________________________________) [PROPOSED] ORDER Upon consideration of Plaintiff Natural Resources Defense Council’s Motion for Summary Judgment [ECF #10], the Cross-Motion for Summary Judgment and Opposition filed by Defendant U.S. Environmental Protection Agency (EPA), and all other related filings, the Court grants summary judgment to Defendant EPA. SO ORDERED. ____________________________ The Honorable John D. Bates United States District Judge DATED: ________________________ Case 1:16-cv-01861-JDB Document 15-1 Filed 06/02/17 Page 1 of 1