Puget Soundkeeper Alliance, V State Of Wa Dept. Of Ecology

Filed Washington State Court of Appeals Division Two February 22, 2017 IN THE COURT OF APPEALS OF THE STATE OF WASHINGTON DIVISION II PUGET SOUNDKEEPER ALLIANCE, No. 48267-3-II Appellant, v. UNPUBLISHED OPINION STATE OF WASHINGTON, DEPARTMENT OF ECOLOGY; and STATE OF WASHINGTON POLLUTION CONTROL HEARINGS BOARD, Respondents. MAXA, A.C.J. – Puget Soundkeeper Alliance (Soundkeeper) appeals the decision of the Pollution Control Hearings Board (Board) to uphold in part a National Pollutant Discharge Elimination System (NPDES) permit issued by the Washington Department of Ecology (Ecology) to Seattle Iron and Metals (SIM) for SIM’s wastewater and stormwater discharges into the Lower Duwamish Waterway. Soundkeeper challenges the permit provisions that (1) require discharges to be tested for polychlorinated biphenyls (PCBs)1 using Method 608 instead of the more sensitive Method 1668C, and (2) establish limitations on copper and zinc levels in 1 PCBs are a group of manmade chlorinated organic chemicals that contain multiple individual compounds (“congeners”) and are highly toxic to humans and animals. No. 48267-3-II untreated stormwater discharges based on the benchmarks in Ecology’s 2009 Industrial Stormwater General Permit (General Permit) instead of based on site-specific water quality standards for those substances. We hold that (1) SIM’s permit properly required the use of Method 608 for testing PCBs because we defer to Ecology’s determination that Method 608 is the testing method approved by the United States Environmental Protection Agency (EPA) and allowed under Washington law; and (2) substantial evidence does not support the Board’s conclusion that there was insufficient data to calculate site-specific water quality-based effluent limitations (WQBELs), and Washington law requires that SIM’s discharges be subject to WQBELs instead of the less restrictive limitations based on the General Permit. Accordingly, we affirm in part and reverse in part the Board’s decisions on the two challenged NPDES permit provisions. We remand to Ecology for revision of the effluent limitations for copper and zinc consistent with this opinion. FACTS SIM’s Discharges into Lower Duwamish Waterway SIM operates an auto shredding and metal recycling facility adjacent to the Lower Duwamish Waterway (LDW). The SIM facility is located in the LDW federal and state cleanup site, which includes the approximately 5.5 mile stretch of the Duwamish River that flows into Elliot Bay. The LDW is heavily contaminated because of major industrial activity in the area over the last 100 years. Ecology is the lead agency for source control at the LDW site. SIM’s operations produce two types of water that must be discharged from the facility. A mix of wastewater from SIM’s operations and some stormwater (referred to as “outfall 001”) is collected and treated before discharge. Stormwater runoff from rooftops and parking lots 2 No. 48267-3-II (referred to as “outfall 002”) is not treated before discharge. SIM discharges both the treated wastewater and the untreated stormwater into the LDW. SIM’s discharges into the LDW are recognized as a possible source of contaminants in the LDW sediments. NPDES Permit Ecology first issued an NPDES permit specific to the SIM site in 2007. The 2007 permit imposed WQBELs for SIM’s treated discharges from outfall 001, with numeric effluent limits for cooper, zinc, total PCBs, and other pollutants. That permit did not regulate SIM’s discharge of untreated stormwater from outfall 002. On September 16, 2013, Ecology issued an NPDES waste discharge permit to SIM relating to the discharges of both outfall 001 and outfall 002 into the LDW.2 The permit imposed daily limitations for PCBs, copper, zinc, and other contaminants at both outfalls. Regarding PCBs, the permit imposed daily limitations of 0.0089 micrograms per liter (µg/L) for outfall 001 discharges. That limitation was based on the PCB human health criteria of 0.00017 µg/L adjusted for a dilution factor for the “mixing zone,” the area surrounding the discharge point where wastewater mixes with receiving water.3 The permit stated that Method 8082A would be used to test PCB levels in outfall 001.4 2 The permit was first issued in 2007, but NPDES permits expire after five years and must be reissued. On August 26, 2014, before the Board’s review, Ecology modified certain portions of the permit. The Board reviewed the permit as modified, but still referred to it as the “2013 permit” in its ruling. 3 Pollutant concentrations within mixing zones may exceed the numeric standards without penalty on the theory that the pollutants will dilute quickly into the receiving water. 4 Before the Board hearing, Ecology modified the 2013 NPDES permit for outfall 001 and replaced the requirement to use Method 8082A with the requirement to use Method 608. 3 No. 48267-3-II For outfall 002, the permit imposed a daily PCB limitation of 0.25 µg/L, significantly higher than the PCB human health criteria used for outfall 001. This limitation was determined based on the detection limit of Method 608, the EPA-approved analytical test that Ecology required for outfall 002 PCB testing. The limitation level represented the minimum value that Method 608 could detect. Regarding copper and zinc, Ecology’s permit writer Ed Abassi calculated WQBELs for outfall 001 using historical data from the site. But for outfall 002, Ecology had only two data points because that discharge had not previously been regulated. Instead of calculating WQBELs, Abassi imported numeric benchmark values from the 2009 General Permit. The General Permit is an NPDES permit that Ecology issued to regulate more than 1,000 facilities statewide that discharge industrial stormwater. Using the General Permit benchmarks, Ecology imposed daily limitations of 14 µg/L for copper and 117 µg/L for zinc in outfall 002 discharges. Board Appeal On October 14, 2013, Soundkeeper filed a petition for Board review of certain portions of SIM’s permit. Soundkeeper challenged (1) the inclusion of a mixing zone for PCBs, (2) the imposition of different PCB limits for outfall 001 and outfall 002, (3) the use of Method 608 for PCB testing instead of more sensitive methods, and (4) the imposition of limits on copper and zinc levels for outfall 002 based on General Permit benchmark values instead of site-specific WQBELs. The Board reviewed the permit, as modified by Ecology, during a four-day hearing in March 2015. The Board entered extensive findings of fact and conclusions of law. The Board agreed with Soundkeeper that Ecology could not grant a mixing zone for PCBs because the LDW was 4 No. 48267-3-II known to be saturated by PCBs and PCBs do not dilute easily. The Board also agreed with Soundkeeper that there was no basis for Ecology to impose higher PCB limits for outfall 002 than for outfall 001. The Board remanded the permit to Ecology for correction of the discharge limitations for PCBs.5 However, the Board rejected Soundkeeper’s two other challenges. The Board ruled that the use of Method 608 for PCB testing was consistent with existing law because Method 608 was the only method approved by the EPA. The Board also ruled that Ecology’s use of the General Permit’s benchmark values to impose limitations on daily copper and zinc levels in outfall 002 discharges was reasonable and that those limitations were consistent with applicable law. The Board deferred to Ecology’s determination that it lacked sufficient data to develop site-specific limitations. APA Appeal Soundkeeper petitioned for judicial review in the superior court, and this court granted its petition for direct review of the Board’s order. Ruling Accepting Direct Review, Puget Soundkeeper All. v. Dep’t of Ecology, No. 45609-3-II, at 3 (Wash. Ct. App. Dec. 22, 2015). ANALYSIS A. STANDARD OF REVIEW The Administrative Procedures Act (APA) governs our review of agency decisions, which includes decisions by the Board. RCW 34.05.510; Cornelius v. Dep’t of Ecology, 182 5 The Board did not state what PCB limitation should be imposed on remand for outfall 002. Presumably, the limitation will be the same as for outfall 001: 0.00017 µg/L. 5 No. 48267-3-II Wn.2d 574, 584-85, 344 P.3d 199 (2015). We can provide direct review of an environmental board’s decision if that board files a certificate of appealability. RCW 34.05.518(1). Under the APA, we may grant relief from the Board’s order based on one of nine reasons listed in RCW 34.05.570(3), including that the order is (1) outside the agency’s statutory authority, (2) based on an erroneous interpretation or application of the law, (3) unsupported by substantial evidence, (4) inconsistent with an agency rule, or (5) arbitrary and capricious. RCW 34.05.570(3)(b), (d), (e), (h), (i). The party challenging the Board’s decision has the burden of demonstrating the invalidity of that decision. RCW 34.05.570(1)(a). We review questions of law and an agency’s application of the law to the facts de novo. Cornelius, 182 Wn.2d at 585. We give great weight to an agency’s interpretation of a statute when the statute is ambiguous and falls within the agency’s area of expertise, if the interpretation does not conflict with the statutory language or intent. Puget Soundkeeper All. v. Pollution Control Hr’gs Bd., 189 Wn. App 127, 136, 356 P.3d 753 (2015). We show the same deference to an agency’s interpretation of its own regulations. Id. More specifically, Ecology’s interpretation of environmental statutes is entitled to great weight “[g]iven that the legislature designated Ecology as the agency to regulate the State’s water resources.” Snohomish County v. Pollution Control Hr’gs Bd., ____ Wn.2d ____, 386 P.3d 1064, 1075 (2016). And the Board’s review of Ecology’s actions also is entitled to deference. Id. However, we are not bound by an agency’s interpretation of the law. Puget Soundkeeper All., 189 Wn.2d at 136; see also RCW 34.05.570(3)(d). “[D]eference to an agency is inappropriate where the agency’s interpretation conflicts with a statutory mandate.” Dep’t of Labor & Indus. v. Granger, 159 Wn.2d 752, 764, 153 P.3d 839 (2007). 6 No. 48267-3-II B. LEGAL PRINCIPLES 1. General Water Quality Policy The goal of the federal Clean Water Act (CWA)6 is to “restore and maintain the chemical, physical, and biological integrity of the Nation’s waters” and attain water quality which provides for the protection and propagation of fish, shellfish, and wildlife. 33 U.S.C. § 1251(a)(2). The CWA expresses “the national policy that the discharge of toxic pollutants in toxic amounts be prohibited,” 33 U.S.C. § 1251(a)(3), and states that “the discharge of any pollutant by any person shall be unlawful,” except as authorized by specified statutory provisions. 33 U.S.C. § 1311(a). The CWA prohibits any discharge of pollutants into the nation’s waters unless the discharge is made according to the terms of an NPDES permit. 33 U.S.C. §§ 1311(a), 13427. Congress authorized the EPA to delegate the NPDES permitting program to the states. 33 U.S.C. § 1342(b). The EPA delegated authority to Ecology to implement the NPDES permitting program in Washington. RCW 90.48.260(1). The legislature has recognized that Ecology has “[c]omplete authority to establish and administer” the program. RCW 90.48.260(1)(a); Snohomish County, 386 P.3d at 1067. The Washington legislature also has adopted a water quality policy, which seeks to “maintain the highest possible standards to insure the purity of all waters of the state.” RCW 90.48.010. And RCW 90.48.520 states, “In no event shall the discharge of toxicants be allowed 6 The CWA’s formal name is the Federal Water Pollution Control Act, 33 U.S.C. §§ 1251-1388. 7 33 U.S.C. § 1342 has been amended since the events of this case transpired. However, these amendments do not impact the statutory language relied on by this court. Accordingly, we do not include the word “former" before 33 U.S.C. § 1342. 7 No. 48267-3-II that would violate any water quality standard, including toxicant standards, sediment criteria, and dilution zone criteria.” 2. NPDES Permit Compliance with Water Quality Standards Under federal law, NPDES permits must impose limits on discharges as necessary to meet water quality standards set by both state and federal statutes and regulations. 33 U.S.C. § 1311(b)(1)(C); Snohomish County, 386 P.3d at 1067. Specifically, State agencies may not issue NPDES permits if “the conditions of the permit do not provide for compliance with the applicable requirements of CWA, or regulations promulgated under CWA” or if “the imposition of conditions cannot ensure compliance with the applicable water quality requirements of all affected States.” 40 C.F.R. § 122.4(a), (d). Similarly, WAC 173-220-130(1)(b)(i) provides that any NPDES permit shall apply and ensure compliance with limitations necessary to “[m]eet water quality standards . . . pursuant to any state law or regulation.” And WAC 173-201A-510(1) states that NPDES permits “must be conditioned so the discharges authorized will meet the water quality standards” and that no permit can be issued that “causes or contributes to a violation of water quality criteria.” These provisions demonstrate that the purpose of the NPDES permitting system is to ensure compliance with state water quality standards. Port of Seattle v. Pollution Control Hr’gs Bd., 151 Wn.2d 568, 603, 90 P.3d 659 (2004). The Washington legislature has “in no uncertain terms” prohibited Ecology from issuing NPDES permits that allow discharges of toxic substances in violation of applicable standards. Puget Soundkeeper All., 189 Wn. App at 138. As a result, “NPDES permits may be issued only where the discharge in question will comply with state water quality standards.” Port of Seattle, 151 Wn.2d at 603. 8 No. 48267-3-II Finally, WAC 173-220-150(1)(c) provides that each NPDES permit shall require that “[a]ny discharge of any pollutant . . . at a level in excess of that identified and authorized by the permit” constitutes a violation of permit terms and conditions. (Emphasis added.) Under this regulation, NPDES permits must require that each discharge comply with applicable water quality regulations. See Puget Soundkeeper All., 189 Wn. App at 138. 3. Washington Water Quality Standards Washington has developed its own water quality standards. Port of Seattle, 151 Wn.2d at 590. These standards include narrative water quality statements and numeric criteria for toxic substances. Id. WAC 173-201A-240(1) provides the narrative water quality standard governing discharges of toxic substances.8 Toxic substances shall not be introduced above natural background levels in waters of the state which have the potential either singularly or cumulatively to adversely affect characteristic water uses, cause acute or chronic toxicity to the most sensitive biota dependent upon those waters, or adversely affect public health, as determined by the department. See also Puget Soundkeeper All., 189 Wn. App at 138-39. WAC 173-201A-240(5) and the attached Table 240 provide specific numeric water quality standards for numerous toxic substances. The human health criteria for PCBs is 0.00017 µg/L. WAC 173-201A-240(5), tbl.240. The toxic substances criteria for marine water aquatic life for 8 WAC 173-201A-240 has been amended since the events of this case transpired. However, these amendments do not impact the statutory language relied on by this court. Accordingly, we do not include the word “former" before WAC 173-201A-240. 9 No. 48267-3-II copper is 4.8 µg/L (acute) and 3.1 µg/L (chronic) and for zinc is 90 µg/L (acute) and 81 µg/L (chronic).9 WAC 173-201A-240(5), tbl.240. C. USE OF METHOD 608 FOR TESTING PCB LEVELS SIM’s NPDES permit requires the use of Method 608, an EPA-approved PCB testing method, to measure PCBs in discharges from outfall 002. But the minimum detection limit of Method 608 is only 0.25 µg/L and Method 608 has a practical quantitation limit (PQL) of 0.5 µg/L.10 This PQL is significantly higher than the PCB human health criteria of 0.00017 µg/L.11 Soundkeeper argues that Ecology violated Washington law by issuing an NPDES permit that required the use of Method 608, because that method is not sensitive enough to determine whether SIM’s discharges violated the applicable water quality standard for PCBs. Soundkeeper claims that Ecology could not lawfully have issued the permit unless it specified the use of Method 1668C, a more sensitive test that can quantify PCB concentrations in the range of the water quality standard. Ecology argues that it was required to specify Method 608 in the permit under WAC 173-201A-260(3)(h) because it is the only testing method approved by the EPA. We agree with Ecology. 9 “Acute” refers to short-term exposure, and “chronic” refers to long-term exposure. WAC 173- 201A-020. The permit’s “daily” limits relate to acute limits. 10 The PQL represents the lowest level at which a pollutant concentration reliably can be quantified. 11 Ecology imposed an effluent limitation for PCBs of 0.25 µg/L on outfall 002 discharges based on the minimum detection limit of Method 608. However, the Board ruled that this high detection limit did not justify imposing a higher effluent limit than the 0.00017 µg/L limitation for outfall 001. The Board remanded to Ecology for the revision of effluent limits for PCBs. Presumably, on remand Ecology will impose the 0.00017 µg/L limitation for outfall 002. 10 No. 48267-3-II 1. Legal Principles Under federal law, monitoring must be done using “sufficiently sensitive” test methods. 40 C.F.R. § 122.44(i)(1)(iv). A method is sufficiently sensitive when either (1) the method minimum level is at or below the effluent limit established in the permit for the measured pollutant or (2) the method has the lowest minimum level of the analytical methods approved under 40 C.F.R. part 136 for the measured pollutant. 40 C.F.R. § 122.44(i)(1)(iv)(A)(1)-(2). Washington law provides additional regulations regarding testing methods. WAC 173- 201A-260(3) outlines how Ecology should set and measure water quality criteria. When setting numeric criteria for water quality, Ecology “will give consideration to the precision and accuracy of the sampling and analytical methods used, as well as the existing conditions at the time.” WAC 173-201A-260(3)(g). Further, WAC 173-201A-260(3)(h) provides: The analytical testing methods for these numeric criteria must be in accordance with the ‘“Guidelines Establishing Test Procedures for the Analysis of Pollutants’” (40 C.F.R. Part 136) or superseding methods published. [Ecology] may also approve other methods following consultation with adjacent states and with approval of the [EPA]. This regulation allows the use of a testing method that is (1) listed in 40 C.F.R. Part 136, (2) a superseding method that has been published, or (3) approved for use by Ecology following consultation with the EPA. Method 608 is listed in 40 C.F.R. Part 136 for monitoring PCBs, but Method 1668C is not. 40 C.F.R. 136, app. A. And Ecology has not approved Method 1668C for testing PCBs. The EPA developed Method 1668C with the intention of listing it as an approved PCB testing method in 40 C.F.R. Part 136. The EPA also “published” Method 1668C for use in CWA programs. In April 2010, the EPA stated: 11 No. 48267-3-II The Office of Science and Technology (OST) in EPA’s Office of Water developed Method 1668C . . . for use in Clean Water Act (CWA) programs. EPA is publishing this Method for users who wish to measure PCBs as congeners now, and in 2010, EPA expects to publish a proposal in the Federal Register for public comment to add this Method to other CWA Methods published at 40 CFR Part 136. Administrative Record (AR) at 2751 (emphasis added). Although the EPA proposed rulemaking to add Method 1668C to the list in 40 C.F.R. Part 136, it chose not to add the method. The EPA did not reject Method 1668C, but merely deferred approval. The EPA noted that it “is aware that this method is being used in some states in their regulatory programs and by other groups for some projects with good success.” AR at 3587. But the EPA stated that it was “still evaluating the large number of public comments and intends to make a determination on the approval of this method at a later date. . . . This decision does not negate the merits of this method for the determination of PCB congeners in regulatory programs.” AR at 3587. 2. Interpretation of WAC 173-201A-260(3)(h) The Board concluded that Ecology’s specification of Method 608 as the PCB testing method in SIM’s NPDES permit was consistent with WAC 173-201A-260(3)(h) because Method 608 is the only method the EPA has approved. Soundkeeper argues that Ecology could have required Method 1668C for PCB testing because that method qualifies as a “superseding method[] published” under WAC 173-201A-260(3)(h). To interpret agency regulations, we apply the same principles used to interpret statutes. Puget Soundkeeper All., 189 Wn. App. at 136. Statutory interpretation is a matter of law that we review de novo. Jametsky v. Olsen, 179 Wn.2d 756, 761, 317 P.3d 1003 (2014). The purpose of statutory interpretation is to determine and give effect to the legislature’s intent. Gray v. Suttell 12 No. 48267-3-II & Assocs., 181 Wn.2d 329, 339, 334 P.3d 14 (2014). To determine legislative intent, we first look to the plain language of the statute, considering the text of the provision, the context of the statute, related provisions, and the statutory scheme as a whole. Id. If a statutory term is undefined, we may use a dictionary to determine its plain meaning. Nissen v. Pierce County, 183 Wn.2d 863, 881, 357 P.3d 45 (2015). The parties apparently agree that Method 1668C is a “published” method. The question is whether Method 1668C is a “superseding” method. WAC 173-201A-260(3)(h) does not define the term “superseding.” Supersede has numerous dictionary definitions, including “[1] to make obsolete, inferior, or outmoded, [2] to make superfluous or unnecessary, [3] to take the place of and outmode by superiority: supplant and make inferior by better or more efficiently serving a function.” WEBSTER’S THIRD NEW INTERNATIONAL DICTIONARY 2295 (2002). Soundkeeper argues that Method 1668C falls within the definition of a superseding method. Method 1668C has a PQL as low as 0.000022 µg/L.12 Method 608’s PQL is only 0.5 µg/L. Because Method 1668C’s detection limit is much lower than Method 608’s detection limit, Method 1668C can be considered a superior testing method that can take the place of Method 608. But Ecology emphasizes that the EPA decided not to add Method 1668C to the list in 40 C.F.R. Part 136, and therefore Method 1668C cannot be said to have “superseded” the approved Method 608. Method 608 is not “superfluous or unnecessary” because it is still the only EPA- approved testing method. Ecology also argues that WAC 173-201A-260(3)(h)’s reference to a 12 Method 1668C tests each of the 209 congeners that comprise the total PCBs individually, so the PQL may vary among the congeners. 13 No. 48267-3-II superseding method refers only to new versions of methods already included in 40 C.F.R. Part 136, not entirely new methods. The term “superseding method” is ambiguous. But Ecology and the Board have interpreted WAC 173-201A-260(3)(h) as not applying to Method 1668C. Because the regulation is ambiguous and its interpretation falls within Ecology’s area of expertise, we will defer to Ecology’s interpretation of its own regulation.13 See Snohomish County, 386 P.3d at 1075. We hold that under Ecology’s interpretation of WAC 173-201A-260(3)(h), Method 1668C is not a published superseding method, and therefore Ecology could not consider that method for use in SIM’s NPDES permit. 3. Use of Method 608 Soundkeeper also argues that even if Method 608 is the only approved method for testing PCBs, Washington law precludes Ecology from using Method 608 because it is not sensitive enough to enforce compliance with water quality standards. Soundkeeper’s position is that Ecology’s only lawful option is to refuse to issue the NPDES permit. We disagree. The human health criteria for PCBs is 0.00017 µg/L. WAC 173-201A-240(5), tbl.240. Ecology adopted that standard as the effluent limitation for outfall 001, and the Board ruled that there was no justification for a higher effluent limitation at outfall 002. The problem is that Method 608 has a PQL of 0.5 µg/L. This means that Method 608 cannot detect when the PCB levels in SIM’s discharges are higher than the 0.00017 µg/L limitation but less than 0.5 µg/L. 13 Under WAC 173-201A-260(3)(h), Ecology also could use Method 1668C in NPDES permits if it approved that method after consulting with adjacent states and with the approval of the EPA. But the regulation states that Ecology “may” give such approval, WAC 173-201A-260(3)(h), and the Board noted that it had no authority to require Ecology to seek EPA approval of a different method. 14 No. 48267-3-II Therefore, Soundkeeper argues that the use of Method 608 is improper because it potentially would allow SIM to discharge PCBs in concentrations that would violate the water quality standards in its NPDES permit. But Soundkeeper’s argument is inconsistent with federal and state law regarding testing methods. Federal law requires that monitoring be done using “sufficiently sensitive” test methods. 40 C.F.R. § 122.44(i)(1)(iv). Under 40 C.F.R. § 122.44(i)(1)(iv)(A)(2), a method is sufficiently sensitive when it has the lowest minimum level of the analytical methods approved under 40 C.F.R. part 136 for the measured pollutant. Method 608 is the only approved method for PCBs, and therefore it necessarily is the method with the lowest minimum level. We hold that it is lawful for Ecology to issue an NPDES permit that calls for the use of Method 608 to test PCBs. D. EFFLUENT LIMITATIONS FOR COPPER AND ZINC IN OUTFALL 002 In developing effluent limitations for copper and zinc discharges from outfall 002, Ecology imported numeric benchmark values from the 2009 General Permit. Use of the General Permit benchmarks resulted in daily effluent limitations of 14 µg/L for copper and 117 µg/L for zinc. These limitations are significantly higher than what Soundkeeper asserts site-specific WQBELs would be – daily limits of 4.8 µg/L for copper and 90 µg/L for zinc. Soundkeeper argues that the Board erred in allowing Ecology to use copper and zinc limitations taken from the General Permit, which it characterizes as technology-based limitations, instead of calculating site-specific WQBELs. Ecology argues that the permit had to apply copper and zinc limitations taken from the General Permit because there was insufficient data for the permit writer to calculate site-specific WQBELs. Ecology also claims that the 15 No. 48267-3-II General Permit limitations were water quality-based, not technology-based. We agree with Soundkeeper.14 1. Imposition of Effluent Limitations When addressing the discharge of pollutants in an NPDES permit, Ecology must first determine whether an effluent limitation is required. An NPDES permit must contain effluent limits for a pollutant if there is a reasonable potential that a discharge will contain the pollutant in excess of water quality standards. 40 C.F.R. § 122.44(d)(1)(iii). A permit writer determines if an effluent limitation must be included in the permit by conducting a reasonable potential analysis: whether a facility’s discharge will cause, has the reasonable potential to cause, or will contribute to a violation of water quality standards. 40 C.F.R. 122.44(d)(1)(ii), (iv). Ecology’s Permit Writer’s Manual contains instructions for conducting a reasonable potential analysis. In order to perform a statistical reasonable potential analysis, a permit writer must develop an estimate of variability over time for each pollutant in a discharge. The most commonly used estimator is the coefficient of variation (CV), which is based on site discharge data. The CV is also used in the formula for calculating effluent limits for a permit. Here, permit writer Abassi stated that in order to accurately calculate a CV, he needed at least 10 to 12 data points. But only two data points from SIM’s outfall 002 discharge were available. Abassi testified that based on the lack of outfall 002 data, he could not calculate a CV and therefore could not perform a statistical reasonable potential analysis. 14 The Board stated that Ecology considered the copper and zinc limitations to be interim limitations. Soundkeeper argues, and Ecology concedes, that the technology-based copper and zinc limits cannot be justified as interim limits because they are not part of a compliance schedule. 16 No. 48267-3-II However, the Board concluded that Ecology actually did perform a reasonable potential analysis and determined that SIM’s outfall 002 discharges had the reasonable potential to exceed water quality standards. The Board stated that although Abassi did not perform a statistical calculation of reasonable potential, he nevertheless decided that effluent limitations were necessary. And the Board noted that Abassi’s supervisor testified that Abassi’s evaluation of the outfall 0002 discharge was the equivalent of a reasonable potential analysis. Ecology does not dispute the Board’s conclusion that Abassi essentially conducted a reasonable potential analysis and that effluent limitations were required for zinc and copper for outfall 002 in SIM’s NPDES permit. The question here is how to calculate those limitations. 2. Calculation of Effluent Limitations Once Ecology determines that an effluent limitation is required, it next must determine the level of that limitation. Ecology claims that Abassi had insufficient data to develop WQBELs for copper and zinc at outfall 002. Abassi testified that because he could not calculate a CV, he could not calculate site-specific effluent limits. The Board deferred to “Ecology’s technical determination that it lacked sufficient monitoring data for SIM’s untreated stormwater discharge to develop site-specific numeric effluent limits.” Clerk’s Papers (CP) at 50. And the Board concluded that Abassi’s decision to rely on the General Permit under these circumstances was reasonable. Under the APA, we may grant relief from an agency order if it is not supported by substantial evidence. RCW 34.05.570(3)(e). Substantial evidence does not support the Board’s conclusion for three reasons. First, Ecology did not make a “technical determination” that it had insufficient data to develop site-specific limitations. Abassi did testify about the absence of 17 No. 48267-3-II sufficient data, but primarily in the context of his inability to calculate a CV for a specific effluent limit and to conduct a statistical reasonable potential analysis. Ecology points to Abassi’s statement that he would not use two data points “for enforcement or for limit.” Report of Proceedings at 537. But this is Abassi’s only reference to insufficient data in the context of developing effluent limitations. Further, Abassi did not expressly state that he was forced to use the General Permit benchmarks because he had insufficient data. He simply stated that the effluent limits in the permit came from the General Permit and that they seemed accurate and protective. This testimony did not establish a “technical determination that it lacked sufficient monitoring data” to develop site-specific limitations. CP at 50. Second, the evidence shows that Abassi could have calculated site-specific WQBELs for outfall 002 despite the lack of data. Soundkeeper’s expert, Allan Chartrand, testified that effluent data was not necessary to calculate water quality-based limits for an NPDES permit. Ecology’s Permit Writer’s Manual states that when there are fewer than 20 data points available to calculate a CV, a default CV of 0.6 may be used instead of a calculated CV. Therefore, Abassi could have calculated site-specific WQBELs using the default CV. Ecology does not address why this default CV was not used. Third, Abassi testified that assuming a finding of reasonable potential, he could have determined the WQBELs for outfall 002. He stated that he would have used the human health calculations in Ecology’s fact sheet: water quality standards for copper of 4.8 µg/L (acute) and 3.1 µg/L (chronic) and water quality standards for zinc of 90 µg/L (acute) and 81 µg/L (chronic). Because the Board found that Ecology had determined that SIM’s discharges had the reasonable 18 No. 48267-3-II potential to exceed water quality standards, this testimony means that Abassi did have sufficient information to determine site-specific WQBELs for outfall 002. We hold that the Board’s conclusion that Ecology lacked sufficient data to develop site- specific effluent limits for outfall 002 is not supported by sufficient evidence. Because this conclusion depends on an evaluation of the applicable facts rather than an interpretation of statutes or regulations, we do not give special deference to Ecology or the Board on this issue. See Port of Seattle, 151 Wn.2d at 594 (stating the standard of review for factual findings inherently includes an element of deference to the Board). As a result, we hold that the Board erred in concluding that Abassi acted reasonably when he relied on the General Permit. 3. Inadequacy of NPDES Permit Limitations The Board concluded that the effluent limits in the NPDES permit for copper and zinc, which were based on the General Permit benchmarks, were consistent with applicable law. Soundkeeper argues that Washington law requires Ecology to use the lower site-specific WQBELs instead of the higher General Permit limitations. We agree with Soundkeeper. Initially, Ecology argues that the limitations based on the General Permit were consistent with applicable law because they were in fact water quality-based limitations. Ecology claims that these limitations are water quality-based because the General Permit benchmarks involved pollutant discharge levels that would not exceed water quality standards for the likely pollutants found in industrial stormwater and were designed to protect water quality in the majority of receiving water conditions. However, the Board referred to the limitations based on the General Permit benchmark as technology-based limits. Ecology does not challenge the Board’s reference to the permit 19 No. 48267-3-II limitations as technology-based. In addition, Ecology’s own fact sheet for SIM’s NPDES permit refers to the limitations as technology-based. More significantly, even if the General Permit limitations were based on water quality standards generally applicable to all industrial dischargers, Ecology does not explain why those limitations complied with Washington law. The evidence shows that the limitations Ecology imposed do not comply with the specific water quality standards applicable here. Both Abassi and Chartrand’ testified that properly calculated WQBELs for the 002 outfall would have been the same as the water quality criteria in WAC 173-201A-240(5), Table 240: 4.8 µg/L (acute) and 3.1 µg/L (chronic) for copper and is 90 µg/L (acute) and 81 µg/L (chronic) for zinc.15 But the permit limitations were significantly higher: daily limitations of 14 µg/L for copper and 117 µg/L for zinc. Therefore, SIM’s NPDES permit would allow the discharge of pollutants in concentrations that would far exceed established water quality standards. As stated above, Washington law is clear that Ecology cannot issue NPDES permits that would allow discharges of toxic substances that would violate applicable water quality standards. RCW 90.48.520; Port of Seattle, 151 Wn.2d at 603; Puget Soundkeeper All., 189 Wn. App at 138. Therefore, we hold that the Board erred in concluding that the effluent limitations in SIM’s NPDES permit – which were significantly higher than the water quality standards – were consistent with applicable law. 15 Normally the water quality criteria are adjusted to account for a mixing zone and dilution to develop WQBELs. But for the untreated wastewater at outfall 002, there was no mixing zone and no dilution factor. This means that the water quality criteria would have been the effluent limit. 20 No. 48267-3-II CONCLUSION We affirm in part and reverse in part the Board’s rulings on the proper PCB testing method and on the effluent limitations for copper and zinc. We remand to Ecology for revision of the effluent limitations for copper and zinc consistent with this opinion. A majority of the panel having determined that this opinion will not be printed in the Washington Appellate Reports, but will be filed for public record in accordance with RCW 2.06.040, it is so ordered. MAXA, A.C.J. We concur: WORSWICK, J. SUTTON, J. 21