Colorado Environmental Coalition v. Salazar

OPINION AND ORDER VACATING AND REMANDING AGENCY DETERMINATION MARCIA S. KRIEGER, District Judge. THIS MATTER comes before the Court for review of agency action pursuant to the Administrative Procedure Act (“APA”), 5 U.S.C. § 701 et seq. The Court has reviewed the Administrative Record (# 55, as supplemented # 59), the Plaintiffs’ opening brief (# 66), the Intervenors’ Response Brief (#81), the Defendants’ response brief (# 92), and the Plaintiffs’ reply brief (# 93),1 as well as the additional [1239] arguments and citations by the parties at an oral argument2 on May 22, 2012. FACTS A. Planning Area For purposes of the discussion here, the Roan Plateau Planning Area is a tract of land of more than 120,000 acres, located predominantly in Garfield County, Colorado. Roughly speaking, it is bounded on the south by the Colorado River (also encompassing a stretch of Interstate 70 between Rifle, Colorado and Parachute, Colorado), on the west by Parachute Creek, on the east by State Highway 13, and on the north by an artificial line that runs roughly parallel to and a few miles south of the boundary line between Garfield and Rio Blanco Counties (although in the northeast corner of the parcel, a small triangular finger extends north into the southeast corner of Rio Blanco County). The lands within the Planning Area include both privately-and publicly-owned land. Approximately 58% of the Planning Area is public land, managed by the Bureau of Land Management (“BLM”). The Planning Area consists of two zones, divided by steep cliffs. The lower portion, sometimes described as the “base of the cliffs,” is mostly located in the southern and easternmost portion of the Planning Area. It is comprised of approximately 38,000 acres of semi-desert habitat. The upper portion of the parcel is referred to as the “top” of the Plateau (and the terms “Roan Plateau” and “Plateau” are often used to refer only to this upper section). The top of the Plateau includes approximately 35,000 acres of higher-elevation moist montane and sub-alpine habitat. The top of the Plateau has been sculpted into several distinct drainage basins by the East Fork and East Middle Fork of Parachute Creek, Trapper Creek, Northwater Creek, and Good Creek. High ridges separate the basins. The Court need not extensively address the resources of the Planning Area. It is sufficient to note that, like many areas of Colorado, the area has been blessed with an abundance of two major resources, the uses of which are often in conflict. Its surface offers extensive and largely unspoiled (particularly in the areas atop the Plateau) scenic, ecological, and wildlife virtues. Below the surface, the Planning Area contains significant and valuable oil and gas reserves. B. History of the Planning Area Early in the 20th Century, certain public lands were set aside as Naval Petroleum and Oil Shale Reserves in order to provide a ready source of fuel in cases of war or national emergency. During the 1920s, two large portions of the Planning Area were designated as Naval Oil Shale Reserves: “NOSR 1,” located on the top of [1240] the Plateau, and “NOSR 3,” located at the area at the base of the cliffs. Initially, these lands were managed by the Department of Defense, but in the 1970s, management was transferred over to the Department of Energy. In the early 1980s, private natural gas development began in fields adjacent to NOSR 3. Eventually, the Department of Energy became concerned that natural gas reserves beneath NOSR 3 were being drained away by private entities, so the Department of Energy thus authorized the drilling of several dozen gas wells in NOSR 3. Those wells produced a significant amount of natural gas through 1996. In the mid-1990s, Congress decided that maintaining these reserves no longer served a meaningful strategic purpose, and that such reserves (and the lands under which they lay) presented an opportunity to generate public revenue. The Department of Energy decided that leasing, rather than selling the lands for mineral development, was preferable because lease revenue would be produced while retaining the potential for public access to the recreational, scenic, and biological aspects of the surface. Consistent with the Department of Energy’s recommendation, in 1997, Congress passed the Transfer Act, 10 U.S.C. § 7439. That Act directed the Department of Energy to transfer jurisdiction over NSOR 1 and NOSR 3 to the Department of the Interior. Two provisions of the Transfer Act are particularly germane to the issues presented here. First, Congress directed that: Beginning on November 18, 1997, or as soon thereafter as practicable, the Secretary of the Interior shall enter into leases with one or more private entities for the purpose of exploration for, and development and production of, petroleum ... located on or in public domain lands in [NOSR 1 and NOSR 3]. 10 U.S.C. § 7439(b)(1). In addition, Congress provided that: [t]he Secretary of the Interior, acting through the Director of the Bureau of Land Management, shall manage the lands transferred under [this Act] in accordance with the Federal Land Policy and Management Act of 1976 [43 U.S.C. § 1701 eb seq.1 and other laws applicable to public lands. 10 U.S.C. § 7439(c). C. Development of the RMP/EIS Until approximately 2000, the BLM managed the Planning Area in accordance with various Resource Management Plans (“RMPs”) issued between 1984 and 1997. RMPs provide a comprehensive statement of land management priorities established by the BLM for a given tract. On November 16, 2000, the BLM announced its intention to amend the existing RMPs for the Planning Area, to “[f]acilitat[e] management of the natural resources of the Planning Area for multiple-use and long-term value,” “[e]nsur[e] a consistent, coordinated approach to managing lands” within the area, and “[c]ompl[y] with the provisions of [the Transfer Act].” Because the amendment of the RMP constitutes a “major federal action,” the BLM directed the preparation of a comprehensive Environmental Impact Statement (“EIS”). The BLM began initial scoping on the project in 2001 and early 2002. By mid-October 2002, the BLM had formulated a preliminary slate of six alternatives for the new RMP, four of which are relevant: • Alternative A3: This is the statutorily-required “no action” alternative. It [1241] would simply continue existing land management policies. Under this alternative, the area of NOSR 3 that was under oil and gas development would remain so, and all other areas would remain as currently managed, leaving approximately 97% of the top of the Plateau undisturbed.4 The BLM estimated that this alternative would render about 60% of the total Planning Area closed to oil and gas leasing. • Alternative B: This alternative anticipated oil and gas leasing on “slightly more than half’ of those portions of NOSR 1 and NOSR 3 currently closed to leasing. However, much of the leased area would be subject to “no surface occupancy” or “no ground disturbance” stipulations that, according to the BLM, would result in a relatively small number of additional wells and a 36% increase in the total amount of land made available for development when compared to the “no action” alternative. Certain portions of the Planning Area that had previously been determined to have exceptional wilderness or wildlife characteristics would receive various forms of heightened environmental protection. • Alternative C: This was the BLM’s “preferred alternative.” It was designed “to emphasize multiple resource use.” All of the Planning Area would be made available for oil and gas leasing, but almost 50% of the Planning Area would be subject to “deferred leasing.” In other words, the lands on the top of the Plateau would be leased only after approximately 80% of the wells that would be drilled on the lower-elevation lands were completed and producing. (The BLM estimated that reaching this threshold would take 10-20 years.) As in Alternative B, some areas (although fewer than in Alternative B) would receive additional environmental protection., • Alternative F: This alternative prohibited most leasing of lands on top of the •Plateau for oil and gas development. It anticipated denominating the top of the Plateau a “Special Management Recreation Area” that would maintain its natural character but permit some forms of -non-motorized recreational use. The record indicates that Alternative F was supported by a significant number of interested parties, including all of the various municipalities in Garfield County- The six- alternatives were disseminated for public review and comment in late 2002. After receiving comments, BLM officials continued to discuss and refine the alternatives. In November 2002, BLM officials decided that Alternative F (as well as components of each of the other alternatives) “included unnecessary overlap, were not always clearly differentiated, and might not satisfy the specific intent of the [1242] Transfer Act without additional legislation.” As a result, “the major components of Alternatives A through F were modified and recombined” into the five alternatives addressed in the final RMP/EIS. Alternative F was deleted. The deletion of Alternative F caused concern among many who had favored it. As a result, in early 2005, some entities proposed what is referred to as the “Community Alternative.” Descriptions of the Community Alternative vary throughout the record, but its distinctive feature was a proposal designed to harmonize environmental protection and mineral recovery interests by allowing leasing of the entire Planning Area, but limiting the manner of drilling to “multilateral directional drilling” from wells situated in areas around the perimeter of the Plateau. Thus, the Community Alternative envisioned that most of the surface of the Plateau would remain undisturbed. As discussed below, the BLM did not expressly consider the Community Alternative in the final RMP/EIS. In August 2006, the BLM issued the final RMP/EIS. It selected a variation of Alternative C as its proposed plan. Like Alternative C, the proposed plan permitted leasing of 100% of the Plateau for oil and gas drilling, but provided that much of that land would be subject to various stipulations designed to provide significant environmental protection to lands atop the Plateau. The final plan differed from Alternative C in one major respect: Alternative C contemplated that drilling on the top of the Plateau would not begin until 80% of the lower-elevation (NOSR 3) area had been developed (described by the BLM as “vertical phasing”), but the final plan provided that drilling on the top of the Plateau would proceed sequentially across six “phased development areas” defined by the Plateau’s drainage basins (that is, as “horizontal phasing”). The horizontal phasing plan anticipated that drilling would commence in the first designated zone, immediately north of the cliff face (that is, the zone closest to existing oil and gas fields), but would not proceed to subsequent zones until lessees had achieved “substantial completion of diligent drilling in the preceding area” and had satisfactorily complied with “mitigation and interim reclamation requirements.”5 The BLM explained that horizontal phasing would “reduce impacts to wildlife by leaving large portions of the plateau relatively unaffected until reclamation had been completed ' in previously drilled areas.” The BLM explained that it had rejected Alternative F because it “did not comport with the Transfer Act because it would make none of the NOSR 1 area atop the plateau available for oil and gas leasing.” However, it noted that “the no-lease component was retained as part of the No Action alternative” and many of the other protective restrictions found in Alternative F were incorporated into other alternatives considered in the final RMP/EIS. The final RMP/EIS did not separately address the Community Alternative, except as part of the BLM’s responses to public comments in Chapter 6 of the document. Many of these comments requested that the BLM “fully analyze the impact of the Community Alternative and incorporate its recommendations where appropriate.” The entirety of the BLM’s response to these comments was that “the ‘Community Alternative’ was not an alternative developed and analyzed in the Draft RMPA/EIS. However, many of the basic [1243] components of the ‘Community Alternative’ were incorporated in the various combinations in the five alternatives analyzed in the Draft RMPA/EIS. Many of these components are also reflected-in the Proposed Plan.” The BLM adopted the proposed plan in June 2007. This action followed. ISSUES The Plaintiffs contend that the final RMP/EIS violated the APA, in that the BLM’s decision-making process was arbitrary, capricious, or contrary to law. The Plaintiffs raise five primary arguments: (i) the BLM violated the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321 et seq., in that it failed to consider an adequate range of alternatives when it failed to give due consideration to Alternative F; (iij it violated NEPA by failing to consider the Community Alternative; (iii) it violated NEPA, insofar as it failed to adequately assess environmental impacts occurring after the first 20 years of the plan (a time when a relatively small proportion of the total number of wells contemplated by the entire plan will be drilled); (iv) it violated NEPA by failing to assess the cumulative incremental environmental impacts of the plan on wildlife migration and air quality, caused by anticipated “explosive growth” of oil and gas development on private lands adjacent to the Planning Area; and (v) it violated both NEPA and the Federal Land Policy and Management Act (“FLPMA”), 43 U.S.C. § 1701 et seq., in that it failed to adequately consider the effects of ozone pollution. ANALYSIS The Court pauses at this point to make several observations. First, it is important to emphasize that statutes like the APA (and to the extent that the APA analysis requires consideration of additional statutes, such as NEPA and FLPMA) permit judicial review only to ensure that, in making a decision, a governmental agency follows the proper procedural process, such as obtaining appropriate information and giving a valid explanation for each aspect of the decision. The Court is not free to second guess or review the wisdom of the agency’s decision. New Mexico ex rel. Richardson v. Bureau of Land Mgt., 565 F.3d 683, 704 (10th Cir.2009) (NEPA prohibits “uninformed” rather than “unwise” decision-making). The fact that reasonable people might have reached a different decision than did the agency is not a basis to overturn a decision that is otherwise procedurally proper. Second, the Court notes that administrative agencies serve a unique role in government. Agency decisionmaking is not a political referendum. Agencies are tasked with acquiring expertise in specialized areas, and required to use that expertise in making informed decisions. Although public input with regard to such decisions is often sought and considered by agencies, that a significant, large, or even overwhelming portion of the public disagrees with an agency’s decision is not a basis for overturning it. If an agency properly follows required procedures, but reaches an unpopular decision, the public’s remedy is not found in a court challenge, but instead through the political process. With these considerations in mind, the Court turns to the Plaintiffs’ particular challenges. A. Standard of review The Plaintiffs’ substantive claims are governed by the Administrative Procedures Act (“APA”) found at 5 U.S.C. § 701, et seq. Under 5 U.S.C. § 702: A person suffering legal wrong because of agency action, or adversely affected [1244] or aggrieved by agency action within the meaning of a relevant statute, is entitled to judicial review thereof.... In turn, judicial review is governed by 5 U.S.C. § 706, which provides that a court reviewing an agency’s action shall: (2) hold unlawful and set aside agency action, findings, and conclusions found to be— (A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; ... or (D) without observance of procedure required by law; ... In making the foregoing determinations, the court shall review the whole record or those parts of it cited by a party, and due account shall be taken of the rule of prejudicial error. A court may affirm an agency’s decision only “on the grounds articulated by the agency itself.” Olenhouse v. Commodity Credit Corp., 42 F.3d 1560, 1565 (10th Cir.1994). “After-the-fact rationalization by counsel in briefs or argument will not cure noncompliance by the .agency!.]” Id. at 1575. Although the “agency’s decision is entitled to a presumption of regularity,” it is not shielded from a probing review. Id. at 1574. When the challenge is that the agency’s decision is arbitrary and capricious, a court must determine whether the agency examined the relevant data and factors, and whether it articulated a rational connection between the facts and its decision. Id. As the Supreme Court instructs in Motor Vehicle Manufacturers Association of the United States, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983): [A]n agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. As noted above, the focus is upon the rationality of the decision making process, not upon the wisdom of the decision itself. Olenhouse, 42 F.3d at 1575. A court must determine whether there was a clear error in the agency’s judgment. Id. at 1574; Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971), overruled on other grounds, Califano v. Sanders, 430 U.S. 99, 105, 97 S.Ct. 980, 51 L.Ed.2d 192 (1977). In doing so, a court does not substitute its judgment for that of the agency. Citizens to Preserve Overton Park, Inc., 401 U.S. at 416, 91 S.Ct. 814. It is not the court’s role to weigh conflicting evidence or evaluate credibility. See Pennaco Energy, Inc. v. United States Dept. of Interior, 377 F.3d 1147, 1159 (10th Cir.2004). Indeed, even when the administrative record contains evidence which arguably conflicts with the agency’s findings, it does not necessarily render the agency’s decision arbitrary and capricious. See id. Nor is it the court’s function to decide the propriety of competing methodologies. See Silverton Snowmobile Club v. United States Forest Service, 433 F.3d 772, 782 (10th Cir.2006). Review of an agency’s decision is usually deferential. See Citizens’ Committee to Save Our Canyons v. U.S. Forest Service, 297 F.3d 1012, 1021 (10th Cir.2002). The deference given “is especially strong where the challenged decisions involve technical or scientific matters within the agency’s area of expertise.” Utah Environmental Congress v. Bosworth, 443 F.3d 732, 739 (10th Cir.2006): If the agency’s exercise of discretion is truly in [1245] formed, then the court defers to it. Utah Shared Access Alliance v. United States Forest Service, 288 F.3d 1205, 1213 (10th Cir.2002). However, if the record shows that the agency prejudged the issues, then deference to the agency’s decision is diminished. See Davis v. Mineta, 302 F.3d 1104, 1112 (10th Cir.2002). The arbitrary and capricious analysis also requires the court to conduct a plenary review of the administrative record to see whether there are facts which support the agency’s decision. Olenhouse, 42 F.3d at 1575-76. “Evidence is substantial in the APA sense if it is ‘enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion to be drawn is one of fact.’ ” Id. at 1575. To be substantial, evidence ■ must be more than a scintilla; ■ “it must be such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Id. at 1581. If evidence is overwhelmed by other evidence, or if it is simply a conclusion, then it is not substantial. Id. B. NEPA claims The purpose of NEPA is to require agencies to pause before committing resources to a project to consider the likely environmental consequences of a decision, as well as of reasonable alternatives to it. Forest Guardians v. U.S. Fish and Wildlife Serv., 611 F.3d 692, 711 (10th Cir.2010). In essence, NEPA requires an agency to take two separate steps: (i) consider reasonable alternatives to the proposed action; and (ii) take a “hard look” at the environmental consequences of the decision. 42 U.S.C. § 4332(2)(C)(i)-(iii). 1. Consideration of alternatives First, an agency must “rigorously explore and objectively evaluate all reasonable alternatives” to the proposed action in comparative form, so as to provide a “clear basis for choice among the options” open to the agency. 40 C.F.R. § 1502.14. At a minimum, the agency must identify and analyze its preferred alternative, as well as a null or “no action” alternative that would occur if the agency elected to maintain the current state of affairs unchanged. Id. In addition, the agency should address all other reasonable alternatives to the proposed action. But the phrase “all other reasonable alternatives” is not entirely open-ended. To define the boundaries of the range of alternatives that must be considered, the agency must first define the objectives of the proposed action, a task in which the agency enjoys considerable discretion. Theodore Roosevelt Conservation Partnership v. Salazar, 661 F.3d 66, 72 (D.C.Cir.2011).. Once the objective of the action is established, the agency may proceed to reject or abandon alternatives that are too remote, speculative, impractical, or ineffective in achieving the stated objective. Id.; Wyoming v. U.S. Dept. of Agriculture, 661 F.3d 1209, 1244 (10th Cir.2011). The agency may also reject alternatives that are not “significantly distinguishable from the alternatives already considered” or under consideration. New Mexico ex rel. Richardson, 565 F.3d at 708-09. In reviewing an agency’s handling of alternatives, a court is guided by a “rule of reason and practicality,” examining whether the discussion of alternatives in the EIS is sufficient to permit a reasoned choice among options. Wyoming, 661 F.3d at 1244. a. Rejection of Alternative F The Court turns first to the Plaintiffs’ argument that BLM’s decision to exclude Alternative F (prohibiting leasing on the top of the Plateau and setting, that [1246] area aside for conservation and recreation purposes) violated NEPA. The BLM offers three reasons for rejecting Alternative F during the planning process: (i) it was inconsistent with the Transfer Act; (ii) it was inconsistent with the BLM’s obligation to achieve “multiple use management”; and (iii) it duplicated components of existing alternatives that remained under consideration. The Court will address each of these contentions in turn. The EIS explains that one of the BLM’s purpose was to “comply with the provisions of [the Transfer Act].” The Transfer Act expressly directed that the BLM “shall enter into leases with one or more private entities for ... the development or production of' [oil and gas] located in [NOSR 1].” 10 U.S.C. § 7439(b)(1) (emphasis added). This statutory language is unambiguous; it requires the BLM to “enter into leases ... for development or production” of oil and gas in NOSR l.6 As stated in Section 2.4.1 of the EIS, the BLM understood this language to require that a “significant portion” of NOSR 1 be made available for oil and gas leasing. It is undisputed that Alternative F “would make none of the NOSR 1 area atop the plateau available for oil and gas leasing,” and thus, the BLM properly concluded that Alternative F did not comport with the requirements of the Transfer Act. The Plaintiffs offer several arguments in response. They concede that “to satisfy-the Transfer Act, BLM was required, at most, to offer some land in NOSR 1” for leasing. However, they argue that “nothing in the Transfer Act states that ‘all’ of the NOSR areas will be leased.” This is a non-sequitur because the record does not reflect that the BLM ever construed the Transfer Act to require that all of NOSR 1 be leased. Indeed, it considered Alternative B, an option that would have prohibited leasing in some 29% of NOSR 1. It is not clear whether the Plaintiffs believe that “some” leasing is a materially different standard than the BLM’s interpretation of the Transfer Act as requiring a “significant portion” of NOSR 1 to be leased, but the Court cannot say that the BLM’s interpretation does violence to the statute; indeed, an interpretation of the Transfer Act that permitted only an insignificant portion of NOSR 1 to be leased would be far more problematic. As noted above, Alternative F did not permit any7 significant land in NOSR 1 to be offered for leasing, and thus, the BLM correctly determined that Alternative F conflicted with the requirements of the Transfer Act. Next, the Plaintiffs argue that, even if the Transfer Act required the BLM to make some of NOSR 1 available for leasing, other portions of the Transfer Act conferred the discretion on the BLM to not offer lands in NOSR 1 for leasing. This argument is based upon the sentence [1247] in the Transfer Act that directs that “[a]ny such lease shall be made in accordance with the requirements of the Mineral Leasing Act, 30 U.S.C. § 181 et seq.,” and that the BLM “manage the lands transferred under [the Act] in accordance with [FLPMA].” 10 U.S.C. § 7439(b)(1), (c). The Plaintiffs argue that both the Mineral Leasing Act and FLPMA give the BLM discretion to withhold certain public lands from leasing in certain circumstances.8 This argument is unavailing. When construing a statute such as the Transfer Act, this Court is required to avoid any reading that would render a portion of that statute meaningless. Corley v. U.S., 556 U.S. 303, 129 S.Ct. 1558, 1567, 173 L.Ed.2d 443 (2009). Although the Mineral Leasing Act and FLPMA might grant the BLM discretion to withhold lands from leasing, the express directive in the Transfer Act that the BLM lease some significant portion of NOSR 1 countermands such discretion. In other words, the correct reading of the two relevant provisions of the Transfer Act is that the BLM was required to lease some meaningful part of NOSR 1 unless the application of the Mineral Leasing Act or FLPMA would prohibit it from doing so. Any other reading of the Transfer Act would leave the decision whether to lease NOSR 1 entirely within the BLM’s discretion, vitiating that portion of the Transfer Act that expressly directed that such leasing take place. Here, the BLM made no findings that precluded the leasing of NOSR 1 under the terms of the Mineral Leasing Act or FLPMA. Thus, the BLM was required to comply with the Transfer Act by leasing portions of NOSR 1. Finally, the Plaintiffs make an abbreviated argument9 that the BLM is not precluded from considering an alternative simply because that alternative would require additional legislative action to implement. The Court understands the Plaintiffs to contend that the BLM erred in [1248] failing to consider Alternative F notwithstanding the fact that Alternative F was inconsistent with the Transfer Act. Assuming that the BLM should not have eliminated Alternative F simply because that alternative sought a course of action that was inconsistent with the dictates of the Transfer Act,10 the record reflects that the BLM gave a second justification for eliminating Alternative F — that it involved “unnecessary overlap” and was not “clearly differentiated” from other alternatives that were continuing to receive consideration. In the final EIS, the BLM points out that “the no-lease component [of Alternative F] was retained as part of the No Action alternative ...” and that other protective restrictions in the alternative were incorporated into other alternatives considered in the final RMP/EIS. Both of these observations are factually accurate. A September 18, 2002 BLM document consisting of a chart comparing the draft Alternatives A through F in a variety of respects reveals that Alternatives A and F are effectively identical with regard to matters such as how many acres in the Planning Area would be subject to “oil and gas leasing with” various restrictions and how many acres would be “closed to all oil and gas leasing.”11 In all other respects — e.g. acres set aside for heightened environmental protections — Alternative F is nearly always identical to Alternative B. Alternative F’s only unique feature was its designation of the top of the Plateau as a Special Recreation Management Area, and even that component would necessarily be conditional upon the “no leasing” component of Alternative A being selected as well. Accordingly, the Court finds that the BLM’s interpretation of the Transfer Act was reasonable and because all but one of the features of Alternative F were present in other alternatives that remained under consideration through the final EIS (and the one unique feature was contingent upon features in other alternatives being selected), the BLM’s decision to eliminate Alternative F as a stand-alone option did not improperly constrain the scope of alternatives considered under NEPA. b. Consideration of the “Community Alternative” The Plaintiffs contend that the BLM’s refusal to separately consider the “Community Alternative” also violated NEPA’s requirement that it consider “all reasonable alternatives.” Perhaps the most comprehensive discussion of the Community Alternative is contained in an April 8, 2005 letter12 to the BLM from Plaintiff Rock the Earth. As that letter explains, the salient feature of the Community Alternative was to “allow leasing of the entire NOSR subsurface gas [1249] reserves,” but to require extraction of those resources through “extended reach multilateral wells” that would be located on a “small fraction of BLM land” in designated areas around the, perimeter of the parcel, leaving “the majority of the NOSR land surface available for environmental protection and ... recreational activities.” Contrary to the BLM’s position at oral argument that the Community Alternative was a “moving target” that was “not clearly defined” so as to permit meaningful analysis, the Court finds that the April 8, 2005 letter from Rock the Earth sets forth the general contours of the (or at least “a”) Community Alternative in sufficient detail so as to permit meaningful analysis of that alternative by the BLM. The Court further finds that the Community Alternative, at least as described in Rock the Earth’s letter, was indeed a distinct and concrete “alternative” to the other courses of action being contemplated by the BLM. The Community Alternative was consistent with the BLM’s statement of the purposes of the action — most significantly, managing the lands for multiple uses while also complying with the Transfer Act, insofar as the Community Alternative would have made the top of the Plateau available for leasing (albeit mostly unavailable for surface occupancy). Moreover, the Community Alternative was sufficiently distinct from the other alternatives considered by the BLM, in that none of the other alternatives proposed a combination of nearly full exploitation of the resources under the Plateau while simultaneously preserving nearly all of the lands atop the Plateau for environmental protection. Thus, the Community Alternative was indeed an “alternative” subject to consideration under NEPA. The question then arises as to whether the Community Alternative was “reasonable,” such that NEPA required the BLM to consider it. As discussed above, the BLM is not required to consider alternatives that are, among other things, speculative or infeasible. At oral argument, the BLM argued that it did “look at directional drilling and the potential that it could achieve to access the resources” and that it concluded that the type of directional drilling contemplated by the Community Alternative was infeasible. However, careful review of the RMP/EIS fails to reveal any instance in which the BLM ever stated that it considered the Community Alternative infeasible. The only time the Community Alternative is mentioned by name13 in the RMP/EIS is in the section in which the BLM responds to commenters who advocated for the Community Alternative, and the BLM’s response to such comments never consists of more than a statement that it did .not specifically consider the Community Alternative but that some parts of it (but, notably, not the most salient feature) appeared in other alternatives.14 It appears to the Court that the BLM’s current argument — that it was not required to analyze the Community Alterna [1250] tive because that alternative was infeasible — is the type of post-hoc justification for an action that is offered in litigation, rather than a justification given by the agency at the time of the decision. Thus, it is insufficient to discharge the agency’s action under NEPA. Olenhouse, 42 F.3d at 1575. For this reason, the Court finds that the BLM’s decision did not comply with NEPA, which in turn renders it “contrary to law” under the APA. For this reason alone, the BLM’s decision must be vacated and remanded for further consideration.15 2. “Hard look” When evaluating alternatives, NEPA requires that the agency take a “hard look” at the environmental consequences that may flow from a particular action. Baltimore Gas & Elec. Co. v. Natural Res. Def. Council, 462 U.S. 87, 97, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983). The purpose of the “hard look” requirement is to ensure that the “agency has adequately considered and disclosed the environmental impact of its actions and that its decision is not arbitrary or capricious.” Id. NEPA does not require an agency to follow any particular decision-making structure in undertaking the required “hard look”; rather, the Court examines the entirety of the agency’s analysis of environmental consequences, mindful of the special expertise an agency may possess, to assess whether the failure to consider a particular matter renders the overall decision-making process arbitrary and capricious. Id. at 100-03, 103 S.Ct. 2246. It is necessary that the agency’s “hard look” be undertaken objectively and in good faith, not as a subterfuge designed to rationalize a decision already made or to purposefully minimize negative side effects. Wyoming, supra; Sierra Forest Legacy v. Sherman, 646 F.3d 1161, 1180 (9th Cir.2011). As always, the Court must be mindful that the “hard look” requirement is not an invitation for the courts to substitute their own judgments for that of the agency on environmental questions. Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 21, 96 S.Ct. 2718, 49 L.Ed.2d 576 (1976). The Plaintiffs raise two major arguments contending that the BLM failed to take a sufficiently “hard look” at the consequences of its decision: (i) the BLM improperly constrained its focus to a 20-year time horizon, rather than to the foreseeable life of the oil and gas development and production; and (ii) the BLM failed to [1251] adequately consider the cumulative incremental impacts of the new drilling under the RMP in combination with effects from existing and future private oil and gas exploration on adjacent lands. a. Planning horizon It is undisputed that the BLM constrained its effects analysis to a 20-year time period. In Section 4.1.1.2 of the EIS (and repeated, in varying formulations but with no greater specificity, elsewhere in the EIS), the BLM explains that: Beyond the 20-year timeframe, BLM believes that quantitative impact assessments are speculative and unreliable, and hence, inappropriate. This is due to a large number of economic, geopolitical, environmental, regulatory, technological, and other facts that could affect condi1 tions beyond 20 years and are themselves subject to change in unexpected ways or degrees. In general, however, it can reasonably be assumed that the Planning Area would continue to support existing multiple uses beyond the 20-year planning horizon. The Plaintiffs contend that the BLM had the ability to — and indeed did — forecast oil and gas development well beyond a 20 year planning period. They point to a Reasonably Foreseeable Development report (“RFD”), found as Exhibit H to the final RMP/EIS. The RFD is a report addressing “the level of oil and gas development activity, constrained only by technical and legal constraints, that an objective reviewer might reasonably expect to occur over a specified range of time.”16 The RFD used by the BLM contemplated oil and gas development occurring beyond the 20-year planning horizon used in the RMP/EIS. For example, the RFD states “a total of 1,026 wells will be drilled over 20 years on the upper plateau ... This is about 15% of the number of wells (6,725 wells ...) that will eventually be drilled on the upper plateau.” The Plaintiffs argue that because the BLM was capable of predicting that oil and gas development would occur outside of the 20-year period, assessment of the environmental consequences should also have exceeded 20 years.16 NEPA’s “hard look” requirement directs that an EIS address environmental effects that are “reasonably foreseeable” from the proposed action. 40 C.F.R. § 1508.7. An effect is “reasonably foreseeable” if it is “sufficiently likely to occur that a person of ordinary prudence would take it into account in reaching a decision.” Sierra Club v. Marsh, 976 F.2d 763, 767 (1st Cir.1992); City of Dallas v. Hall, 562 F.3d 712, 719 (5th Cir.2009). Agencies are expected to engage in some degree of predictive behavior, as “reasonable forecasting is implicit in NEPA, and we must reject any attempt by agencies to shirk their responsibilities under NEPA by labeling any and all discussion of future environmental effects as ‘crystal ball inquiry.’ ” Dubois v. U.S. Dept. of Agriculture, 102 F.3d 1273, 1286 (1st Cir.1996). However, an effect that “cannot be described at the time the EIS is drafted with sufficient specificity to make its consideration useful to a reasonable decision-maker” can be omitted. Id. [1252] The Court rejects the Plaintiffs’ premise that because the RFD anticipated oil and gas development more than 20 years in the future, the BLM had the ability to reasonably predict the extent of such development beyond a 20-year horizon. That argument ignores the RFD’s admittedly speculative and hypothetical nature. As the RFD itself explains (and contrary to its name), it is not a document that attempts to foresee and accurately predict the scope of future development; rather, it presents a speculative “best case” scenario of possible development, constrained only by current technical and legal considerations. In doing so, it expressly disregards a host of other factors (“field life, commodity prices, changes in technology ... ”) that any reasonable prognosticator could hardly ignore. As a result, contrary to the Plaintiffs’ argument, the Court finds that the RFD does not offer any useful suggestion as to what an appropriate planning horizon might be. The question remains, however, as to whether the BLM’s selection of a 20-year planning horizon — as opposed to a horizon of, say, 10 years or 25 years or some other figure — finds support in the record. In its briefing, the BLM makes reference to its “Planning for Fluid Mineral Resources” handbook, which states that the “life of the plan” is “generally 15 to 20 years or whatever has been determined to be the planning horizon or timeframe for the RMP.” The handbook reference is not an adequate justification for two reasons. First, it offers no independent explanation as to why planning horizons are “generally 15 to 20 years.” Without some explanation as to why the BLM selected that figure, the “general” requirement that planning horizons be 15 to 20 years is simply arbitrary. Secondly, the handbook’s language expressly contemplates that the particulars of the RMP may dictate a different planning horizon. Nothing in the RMP/EIS explains why the particular circumstances of this situation warrant applying the 20-year planning horizon, as opposed to some other figure. At oral argument, the BLM also referred the Court to an April 28, 2003 email from BLM staffer Duane Spencer to several others. Mr. Spencer was commenting to his colleagues about the draft EIS, and specifically criticizing the heavy reliance the BLM was placing on the RFD, which (as discussed above) purported to predict the number of wells drilled over the life of the extraction project, rather than an RFD assessing a shorter planning horizon. Among other things, Mr. Spencer stated that: trying to develop an RFD for ultimate development is a SWAG17 at best. We have a pretty good history of developing RFD’s for 15-20 years, but beyond those timeframes, both BLM and industry have a dismal record for forecasting development, (see the San Juan and Greater Green River Basins). There are too many variables, which results in more speculation than science. By the barest of margins, Mr. Spencer’s comments constitute sufficient evidence in the record to justify the BLM’s decision to limit its analysis to a 20-year planning horizon. Unlike the conclusory assertions in the handbook, Mr. Spencer’s comments offer an explanation of why a planning horizon in excess of 20 years would be impractical in this case (“there are too many variables” in play beyond that point), and supply a factual basis for the reasons (BLM’s “dismal record” for forecasts in excess of 20 years, as exemplified by fore [1253] casts used in projects in the San Juan and Greater Green River Basins). The observations by Mr. Spencer are consistent with the reasons given in the RMP/EIS for the BLM’s selection of a 20-year planning horizon. Accordingly, the Court rejects the Plaintiffs’ argument that the BLM’s selection of a 20-year planning horizon for assessing environmental effects of its decision was improper. b. Cumulative effects NEPA also requires that, when taking its “hard look” at environmental concerns, an agency consider not only the impact of the decision itself, but also the extent to which such impacts incrementally add to other consequences from past, present, or reasonably foreseeable future actions that may be undertaken. Center for Env. Law and Policy v. U.S. Bur. of Reclamation, 655 F.3d 1000, 1007 (9th Cir.2011). Although this analysis must be something more than generalized or conclusory, the agency is not required to specifically enumerate every project that will affect the area; assessment of cumulative effects in the aggregate is permissible. Id. The obligation to address cumulative effects includes effects that may arise indirectly from the proposed action — growth-inducing effects, changes in population density, etc. — so long as such effects are reasonably foreseeable. Id. at 1011. The Plaintiffs make two arguments with regard to their contention that the BLM failed to adequately consider the cumulative effects of its proposed plan: (i) it failed to consider how development in the Planning Area, when combined with present and future wells that would be drilled on adjacent private lands as part of an “ongoing boom” in oil and gas development in the area, would affect winter and summer habitat for elk and mule deer; and (ii) the BLM failed to consider the degree to which private oil and gas development in and around Colorado, coupled with development in the Planning Area, would affect air quality. 1. effects on elk and mule deer With regard to the contention that the BLM failed to take a sufficiently hard look at its decision’s cumulative effects on elk and mule deer habitat, the BLM responds by pointing to several parts of the record that, taken together, present a fairly broad analysis of the issue. The Court will not attempt to summarize all of these portions of the record in detail. It is sufficient to note that the BLM acknowledged the likely growth of oil and gas wells on private lands in the vicinity of the Planning Area (section 4.1.1.2), and acknowledged that elk and mule deer range extends over both the Planning Area and private lands that would be affected by such development (same). The BLM also analyzed the effects on deer and elk not just within the approximately 100,000 acres of the Planning Unit itself, but also in larger areas known as “Data Analysis Units,” comprising as much as 2.3 million acres (section 3.3.2.3). The BLM’s ultimate conclusions with regard to the project’s cumulative effects on deer and elk is set forth in Section 4.3.2.2 of the EIS. It notes that “offsite impacts (ie. on private lands within the Planning area) are proportionately greater than for Federal lands,” both because private lands typically offer greater access to drilling and because “a higher proportion of private lands within the Planning Area are in lower elevation habitats” where winter range of elk and mule deer is found. The EIS calculates certain effects on deer and elk that will occur on BLM lands, and goes on to state that “the relative impacts on winter range of development on private lands would be greater,” without actually quantifying the increase. It explains that [1254] the proposed plan would likely result in a loss of 32% of winter range in the Planning Area itself, and a 3% loss of winter range within the much larger Data Analysis Unit. The cumulative impacts assessment concludes with an observation that these habitat loss estimates “would be cumulative to the losses resulting from oil and gas development in lands outside the Planning Area,” such as from increasing drilling on private lands. The BLM states that “the winter range [in the immediate area] has already been subject to loss of winter range [from private drilling] and would probably continue to be subject to such losses at levels comparable to, or greater than, those on BLM lands in the Planning Area.” The Plaintiffs contend that only the language quoted in the previous sentence constitutes the BLM’s assessment of cumulative effects from private drilling, and that the BLM’s mere statement that range loss “would probably continue” is too general a statement to satisfy the “hard look” requirement. They cite to Oregon Natural Resources Council Fund v. Brong, 492 F.3d 1120, 1133 (9th Cir.2007), for the proposition that the BLM must “do more than merely state that past projects contributed to environmental harms.” The Court finds that the EIS does more than what was at issue in Brong. The BLM does not merely state that private drilling has contributed or will contribute to deer and elk habitat loss, but instead discusses the extent to which the BLM believed private drilling would expand over the plan’s time horizon, considers that private drilling tended to have similar or slightly greater effects on private lands, and concludes that private drilling would contribute to habitat loss “at levels comparable to ... [losses occurring] on BLM lands.” In essence, the BLM found that the cumulative effect on deer and elk habitat from additional private drilling over the plan’s lifetime would be proportional to (if not greater than) the effects that the BLM had recognized with regard to drilling inside the Planning Area. The Plaintiffs argue that the EIS should, at a minimum, “analyze whether mule deer displaced as a result of habitat losses under the plan will further overload winter range outside the Planning Area,” or “whether existing and foreseeable future wells and roads outside the Planning Area cut off migration routes for elk that also use the Roan,” among other things. To some extent, the EIS addresses questions like this directly. For example, the EIS discusses deer and elk migration corridors within the Planning Area and the steps that will be taken to preserve them. See e.g. pages 3-34, 4-54. Similarly, the EIS discusses the degree to which oil and gas development in the Planning Area results in habitat loss within that area, including the effects of drill rigs shifting to private lands in winter in response to the BLM restricting winter drilling on public lands. See e.g. page 4-55-4-57. Admittedly, the EIS does not extend these analyses to determine matters such as whether increasing private development on private lands will affect deer and elk migration patterns outside the Planning Area, but it is difficult to see how such an issue is an “effect” of the BLM’s selection of an alternative that must be analyzed. On the whole, the Court finds that while the BLM’s analysis of cumulative effects may not be comprehensive in all possible respects, it sufficiently addresses cumulative and incremental impacts on deer and elk such that the Court is satisfied that the “hard look” requirement has been met in this regard. 2. air quality effects The Plaintiffs’ second “cumulative effects” argument is that although the [1255] BLM assessed potential air quality effects from its decision, it failed to aggregate those effects-with air quality effects that would likely to result from oil and gas development on private lands in the vicinity of the Planning Area. In other words, the Plaintiffs contend that the BLM’ failed to consider the cumulative future air quality effects that would result from its decision. The factual contention that is key to this issue is a prediction by Garfield County that some 10,000-20,000 new oil or gas wells are projected to be drilled in the area over the next 20 years. The RMP/EIS expressly acknowledges that prediction in Section 4.1.1.2. However, the RMP/EIS does not quantify, model, or otherwise assess the effects that such private development will have on air quality in the region (much less the combined effects that such private development will have when aggregated with development in the Planning Area). The RMP/EIS cites to “difficulty in predicting the type, level, and frequency of emissions from private sources, over which BLM has no control,”, and that “including] such hypothetical emission sources into its air quality analysis .... would be reduced to predicting speculative impacts.” The Plaintiffs contend that the BLM’s claimed inability to assess the air quality effects that private development will create is spurious. The Plaintiffs point out that the BLM was able to analyze the air quality effects from potential wells that might be drilled on private lands located inside the Planning Area boundaries — that is, wells over which the BLM has no direct control — and thus, the Plaintiffs argue that “there was no reason the agency could not make similar estimates of such private development outside the Planning Area boundaries.” At oral argument, the BLM first argued that it declined to assess or model the potential output of wells outside the Planning Area because it did not consider the projections from Garfield County to be reliable. The Court finds this contention to be inconsistent with the record. As noted above, the RMP/EIS expressly recites the Garfield County projections, but never characterizes them as “unreliable.” To the contrary, the RMP/EIS appears to readily accept the 10,000-20,000 well projection itself. Instead, the RMP/EIS is concerned with the absence of data showing how those 10-20,000 wells would be regulated or managed: Even where an estimate of cumulative impacts due to offsite causes is available (e.g. 10,000 to 20,000 wells in Garfield County in 20 years), it is not known how much long-term surface disturbance would result, to what degree adverse impacts would be avoided or mitigated, and how the impacts would affect other resource values and land uses ... Therefore, the descriptions of cumulative impacts for the individual resources addressed [in the EIS] are necessarily qualitative. The argument in the BLM’s brief hews more closely to the statement in the RMP/ EIS. In its brief, the BLM argues that it lacked the ability to model the effects of wells drilled outside the Planning Area because it lacked any understanding of the particulars as to how (and where) those wells would be drilled and managed. On the surface, this argument has some appeal. Much of the BLM’s modeling of air quality effects from wells drilled within the Planning Area is found in a July .2005 Air Quality Assessment Report, jointly prepared for both the Glenwood Springs, CO and Vernal, UT BLM offices (apparently to be used in assessing several proposed projects being considered in each [1256] office). Table 3-11 of that report details the methodology used by the BLM to model air quality effects from wells drilled within the Planning Area. The model factors in data such as emissions from glycol dehydrators, compressor emissions, fugitive dust from roadbuilding, flaring emissions, and other specific items of information. Some similar data might arguably be available to the BLM with regard to some of the 10-20,000 wells to be drilled in Garfield County (for example, the model estimates emissions from glycol dehydrators based on an assumption of 1 dehydrator per well, and thus, the BLM could have modeled the emissions of 10-20,000 additional dehydrators). Other types of data would be more speculative, however. For example, without some understanding of where these additional wells will be drilled in Garfield County, it might be difficult for the BLM to estimate how much roadbuilding would be required, which in turn creates uncertainty as to how much fugitive dust from roadbuilding should be included in the model. The BLM implies that Garfield County’s projections are not so comprehensive as to provide this information.18 But there remains the fact that the BLM was able to fully model air quality effects for wells drilled on private lands within the Planning Area. Like the wells that will be drilled outside the Planning Area, the wells on private lands within the Planning Area fall outside BLM control and oversight. Thus, it would seem, the BLM should have had some ability to model air quality effects for private wells outside the Planning Area to a similar extent. This Court cannot speculate as to where or how the BLM was able to assemble the data necessary to model the emissions from future drilling on private lands within the Planning Area, nor can it draw any conclusions as to whether or not those same techniques or methods could be used with similar efficiency in modeling the emissions from future drilling outside the Planning Area. It is sufficient to observe that, on the record before it, the BLM has failed to adequately explain why the methods it used to collect and model air quality data from future anticipated drilling on private lands within the Planning Area could not be used to collect and model air quality data from future anticipated drilling on lands outside the Planning Area. Given the scope of Garfield County’s projections of 10,000-20,000 additional wells over the life of the RMP/EIS — a figure several times larger than the scope of the few thousand wells anticipated within the Planning Area itself over the same time frame — the BLM’s failure to either model the cumulative air quality effects or the offer a more complete explanation as to why it was impractical to do so is not inconsequential. Accordingly, the Court finds that the BLM failed to take the requisite “hard look” at the air quality effects from its decision, when accumulated with air quality effects from anticipated oil and gas development outside the Planning Area, which renders its decision arbitrary and capricious in violation of the APA. The Court will remand the matter to the BLM for further consideration on this matter.19 [1257] 3. ozone Finally, the Plaintiffs allege that the BLM violated both NEPA and FLPMA when it failed to take a “hard look” at air-quality issues that would result from ozone emissions related to the plan. Pursuant to the Clean Air Act, the federal government has established national air-quality standards for six pollutants, including ozone. As part of the EIS, the BLM concluded that the proposed plan would not result in the violation of any of these air-quality standards. The Plaintiffs contend that, at least with regard to ozone, this conclusion lacks support in the record. The BLM admits that it did not perform any specific ozone modeling or screening in conjunction with the EIS. Instead, it claims to have examined the ratio of two precursors to ozone formation — nitrogen oxide (“NOx”) and volatile organic compounds (“VOCs”) — that would be produced by oil and gas activities, and concluded that the ratio did not predict ozone formation. The BLM’s position on this issue is basically twofold: (i) it contends that it was not required to engage in ozone modeling, because there is no history of ozone issues in the Planning Area; and (ii) that the BLM nevertheless did a preliminary assessment of potential ozone issues by examining the prevalence of ozone “precursors,” and concluded that those precursors did not suggest future ozone issues. The Court will address each contention in turn. With regard to the contention that ozone has not been a problem in the Planning Area in the past, the BLM makes two related contentions: (i) that its air quality consultant confirmed no violations of ozone standards in past data (and indeed, the consultant noted to current ozone levels appeared to be on the decline); and (ii) that other environmental agencies — the Colorado Department of Public Health and Environment (“CDPHE”) and the Environmental Protection Agency (“EPA”)— agreed with the BLM that formal ozone monitoring was unnecessary. The Court finds both contentions unpersuasive. The mere fact that the area has not exceeded ozone limits in the past is of no significance when the purpose of the EIS is to attempt to predict what environmental effects are likely to occur in the future, when the development anticipated under the RMP takes place. Moreover, contrary to the BLM’s carefully-worded contention that the CDPHE and EPA did not “insist” that the BLM conduct ozone modeling, both agencies expressed concerns about potential ozone impact. In its December 20, 2002 comments on the BLM’s proposed air quality assessment protocol, the CDPHE acknowledged that the BLM would not be modeling ozone,20 but pointed out that “the area’s ozone background level ... is significant [and t]he analysis should plan to discuss this factor.” In February 2003, the EPA similarly commented on the proposed protocol, noting that “development will increase NOx emissions. Since NOx emissions are [1258] likely to be the limiting factor for ozone production, the RMP should address ozone at least qualitatively.” Thus, it appears to the Court that both agencies believed that some degree of ozone analysis was warranted.21 As to the BLM’s other argument — that its examination of ozone precursors provided it with a reliable assurance that ozone was not likely to have significant impacts — there are several problems. First, and perhaps most importantly, every citation to the record by the BLM on this issue is to post-decision statements made by the BLM in response to comments made by the public to the final RMP/EIS. For example, the BLM relies on a statement in the record that “given the high potential ration of NOx to [VOC] emissions ... air quality specialists for BLM determined at - the scoping phase that ozone creation would not be a significant impact,” but the supporting citation is to a September 29, 2008 response by the BLM to a public comment on the final RMP/EIS, rather than to the document generated during the scoping phase in which the BLM’s air quality specialists made the determination in question. For the same reason that post-hoc justifications for a decision offered by agency counsel in briefs do not constitute substantial evidence supporting an agency’s decision, the Court is extremely reluctant to treat conclusory factual assertions in postdecisional responses to comments as the equivalent of contemporaneous evidence of the agency’s actual decision-making process. Because the BLM has not pointed the Court to any evidence in the record reflecting the BLM’s assessment of precursor ratios in conjunction with its decision making, the Court cannot find that the “precursor examination” was a sufficiently “hard look” at the ozone issue. Even assuming that the BLM is correct that it assessed existing levels of ozone precursors, the Court remains troubled by the admission of the BLM’s counsel at oral argument that such an assessment was purely historical, rather than an attempt to predict whether the decision to expand drilling in the Planning Area would increase ozone levels. Specifically, the Court inquired whether the BLM did “anything to project ... what impact in light of the project there would be on potential ozone precursors.” The BLM’s counsel candidly stated that “No. If what you’re asking is if they did a projection, no.” As discussed above, the RMP/EIS necessarily has a predictive component to it, setting forth the agency’s best efforts at projecting the future effects of its decision. By confining its ozone analysis to ascer-. taining whether ozone precursors reached certain levels in the past, the BLM did nothing to ascertain whether several thou [1259] sand producing wells in the Planning Area would be likely to result in those precursor levels being reached in the future.22 Accordingly, the Court remands the matter to the BLM for further consideration of the issue.23 C. Remedy The APA provides that, where an agency action is found to be arbitrary and capricious or contrary to law, the Court must “hold [it] unlawful and set [it] aside.” 5 U.S.C. § 706(2). The Court accordingly does so. The Plaintiffs appear to request that the Court “set aside” the BLM’s decision to the fullest possible extent, requesting that the Court cancel the leases that the BLM has already issued. The Court declines to do so for several reasons. The Court has concerns as to whether such relief falls within the scope of the instant action, which is an APA challenge to the RMP/EIS (not to the decision to issues leases). Moreover, the Court has doubts as to whether the parties that would be affected by such a decision are properly before the Court. In any event, the Court notes that, although it vacates the BLM’s action and remands the matter for further consideration, it may very well be that, upon reconsideration of the issues addressed herein, the BLM may nevertheless reach the same decision (albeit upon a more complete record or more specific RMP/EIS). If reconsideration leads the BLM to conclude that the RMP/EIS needs to be modified in substantive ways that might implicate the validity of the existing leases, the question of whether and how the leases should be unwound can be addressed at that time. CONCLUSION For the foregoing reasons, the Court finds that the BLM’s decision violated the APA insofar as: (i) it failed to sufficiently address the Community Alternative; (ii) failed to sufficiently address the cumulative air quality impacts of its decision in conjunction with anticipated oil and gas development on private lands outside the Planning Area; and (iii) failed to adequately address the issue of potential ozone impacts. Accordingly, the Court SETS ASIDE the decision embodied in the RMP/ EIS and REMANDS the action to the BLM for further action in accordance with the discussion herein. The Court having resolved the dispute framed by the parties, the Clerk of the Court shall close this case. Footnotes [1] After completing substantive briefing, the parties filed multiple "Notices of Supplemental Authority” (# 94, 106, 135, 157, 158, 161, 162, 194, 196, 199, 201) and responses thereto. Most — although certainly not all — of these submissions identify new legal authori [1239] ty, which the Court has considered. The parties have also filed several motions or "notices” purporting to supplement or "complete” the Administrative Record (# 67, 109, 127, 165, 179). Although no authorization for these submissions was requested or granted, the Court has reviewed them. Unless specifically addressed herein, they do not materially alter the applicable analysis. [2] The Court observes that the initial administrative record in this case is in excess of 51,000 pages (not including the "supplements” referenced above). The primary document itself — the Resource Management Plan and Environmental Impact Statement — runs nearly 900 pages. Given the sheer size of the record (and the fact that large portions of it are inconsequential to the issues presented here), the Court has relied upon the parties’ specific' citations to the record to guide the Court to the relevant materials. 5 U.S.C. § 706(2) ("the court shall review the whole record or those parts of it cited by a party (emphasis added). [3] Preliminary draft versions of the RMP/EIS identified the alternatives by letter, while the final RMP/EIS identifies them by roman numerals. Although the substantive compo [1241] nents of each alternative changed somewhat during the process, the general thrust of each alternative did not. Thus, most of Alternative A in the draft RMP/EIS became Alternative I in the final EIS; most of Alternative B became Alternative II, etc. Except where necessary to differentiate a particular alternative's draft formulation from its final description, the Court may describe a given alternative’s general theme by referencing either its draft letter or final numeral interchangeably. [4] It is not immediately clear from the record what constituted the remaining 3% of NOSR 1. The final EIS seems to suggest that a small portion of NOSR 1 may have been leased for oil and gas development in 1999. It is sometimes difficult to assess boundaries and topog- ■ raphy from the maps in the record, but it appears that NOSR 1 includes several small and isolated tracts located near the base of the cliffs, away from the main body of the tract atop the Plateau. [5] The Plaintiffs rightly point out that it is only "drilling” that was phased, not production. In other words, drilling would only occur one zone at a time, but eventually, all six zones— i.e. the entire top of the Plateau — could simultaneously have wells producing on them. [6] Because the Court finds no ambiguity in the statutory language, it does not consider the Plaintiffs’ arguments premised upon legislative history that warrant a contrary reading. Milner v. Dept. of the Navy, - U.S. -, 131 S.Ct. 1259, 1267, 179 L.Ed.2d 268 (2011) (“[(legislative history ... is meant to clear up ambiguity, not create it’’). The Court also disregards arguments by the Plaintiffs that the BLM’s interpretation of the meaning of this portion of the Transfer Act has changed over time. An agency's interpretation of a statute cannot overcome clear statutory language to the contrary. Southeastern Community College v. Davis, 442 U.S. 397, 411-12, 99 S.Ct. 2361, 60 L.Ed.2d 980 (1979). [7] The record reflects that a minuscule portion of NOSR 1 was leased in or about 1999, but the Court is not prepared to say that such fraction — apparently 3% or less of the total area of NOSR 1, situated remotely from the main tract- — constitutes a “significant portion” of the tract. [8] The Plaintiffs point to 30 U.S.C. § 226(a), which states only that "all lands subject to disposition under [the Mineral Leasing Act] ... may be leased by the Secretary” (emphasis added), but do not point to any particular conditions that would require the Secretary to refuse to lease a particular parcel. The Plaintiffs cite generally to portions of FLPMA that "leave the BLM with considerable administrative flexibility” in managing the “multiple-use” mandate of that act. [9] This argument, presented in two sentences in a footnote in the Plaintiffs' brief, appears to rely upon a provision discussed in the Council On Environmental Quality's "Forty Most Asked Questions Concerning CEQ’s [NEPA] Regulations.” 46 Fed.Reg. 18026 (Mar. 23, 1981). See New Mexico ex rel. Richardson, 565 F.3d at 705 n. 25 (recognizing the “Forty Questions” document as "persuasive authority offering interpretive guidance regarding the meaning of NEPA and the implementing regulations”). Question 2b in that document states "Must the EIS analyze alternatives ... beyond what Congress has authorized?,” to which the response is "Alternatives that are outside the scope of what Congress has approved or funded must still be evaluated in the EIS if they are reasonable, because the EIS may serve as the basis for modifying the Congressional approval....” 46 Fed.Reg. 18027. The Court is not entirely persuaded that the type of additional legislative action described in the "Forty Questions” document — actions for which Congressional approval or funding has not yet been secured — is comparable to the type of additional legislative action — outright reversal of a prior Congressional directive — that would be necessary to effectuate Alternative F. Indeed, if the Plaintiffs are correct, an agency attempting to decide how to perform a Congressionally-mandated act would always have to consider alternatives in which that act was not performed, on the assumption that Congress could thereafter be lobbied to reverse course. Nevertheless, this Court will entertain the Plaintiffs' argument that the BLM should have considered Alternative F even though that alternative does not comply with the Transfer Act. [10] The Court notes that the BLM made compliance with the Transfer Act one of the principal objectives of the RMP process, and the Plaintiffs have not demonstrated that including that purpose exceeded the BLM's considerable discretion in defining the goals of the action. The fact that Alternative F also failed to advance the BLM’s legitimate goals for the RMP is a separate and distinct basis to find that the elimination of Alternative F was not arbitrary or capricious. [11] In some categories, such as the amount of area closed to oil and gas drilling, there are slight differences in acreage between Alternafive A and Alternative F. The record reveals no clear explanation for these relatively small disparities, and in the absence of such an explanation, this Court deems the differences to be insignificant, particularly where the remaining alternatives demonstrate far sharper degrees of difference. [12] The letter bears dates of both April 8, 2005 and April 15, 2005. The indices to the Administrative Record list the letter by the earlier date, and to maintain consistency, the Court has done so as well. [13] The parties have not pointed to, and the Court has not located, any instance in which the BLM purported to discuss the contours of anything resembling the Community Alternative by reference to its salient features or by reference to some term other than "Community Alternative.” [14] The response was invariably some form of the following text: The ‘Community Alternative’ was not an alternative developed and analyzed in the Draft RMPA/EIS. However, many of the basic components of the ‘Community Alternative' were incorporated in various combinations into the five alternatives analyzed in the Draft RMPA/EIS. Many of these components are also reflected in the Proposed Plan/Final EIS. [15] This is not to say that the BLM’s current contention that the Community Alternative was infeasible is necessarily inconsistent with the record. There is some evidence in the record that would support a conclusion that the type of directional drilling contemplated by the Community Alternative is either technically or economically impractical, given the geology of the Plateau. For example, a November 7, 2003 report from Intervenor WPX (formerly Williams) suggests (but does not expressly state) that certain technical challenges would make extended directional drilling from atop the Plateau difficult, and a February 12, 2002 e-mail from a BLM staffer recites a conversation with a Williams drilling foreman who opined that "directional drilling would be extremely more difficult on the roan plateau" due to certain geological conditions. At the same time, there is other evidence in the record suggesting that the BLM’s assumptions about drilling capability are overly-conservative and that the Community Alternative could be accomplished. For example, Rock the Earth’s April 8, 2005 letter extensively discusses and offers citations for the basis for its belief that the Community Alternative is feasible. Although the APA requires the Court to be mindful of the "rule of prejudicial error,” 5 U.S.C. § 706, because there is conflicting evidence in the record on the question of feasibility of the Community Alternative, the matter must be remanded to the BLM to weigh that competing evidence. [16] The RFD goes on to explain that "[t]he actual amount of future oil and gas development activity is dependent on many factors such as actual field life, commodity prices, changes in technology, availability of infrastructure to transport the product, inflation, availability of capital, legislation, and taxes.” [16] At oral argument, the Plaintiffs were unable to state what they believed the appropriate planning horizon should have been. [17] As the Court understands it, the term "SWAG” is an acronym commonly used in the military to disparagingly refer to a "scientific wild-ass guess.” [18] Admittedly, the document identified by the parties as embodying this projection, a September 2005 Growth Scenarios Report: Oil and Gas Employment and Population Impacts working paper, is primarily concerned with projecting how future oil and gas development will affect county populations. Thus, the BLM would be correct in stating that there is apparently no evidence in the record that details the nature, type, location, or other key information relating to the projected 10-20,000 additional wells. [19] In doing so, the Court declines to reach the Plaintiffs’ alternative argument: that the BLM failed to adequately consider the cumu [1257] lative air quality effects from oil and gas projects on nearby federal lands that have already been approved or are under consideration. The Court expresses no opinion on the merits of this argument. [20] The BLM makes much of the CDPHE's comment that it "agrees that ozone modeling is outside the scope of this project,” but the Court finds the statement to be lacking in context and somewhat cryptic. It is not clear to this Court what the "project” that the CDPHE was referencing was — i.e. the RMP itself? the air quality assessment? the request for comment on the proposed protocol?, etc. Moreover, given the context of the CDPHE’s comments, it is unclear to this Court why that agency felt that ozone modeling fell outside that project’s "scope.” [21] Indeed, far from the BLM's implication that the EPA concurred that ozone modeling was unnecessary, a comment from the EPA, quoted in the RMP/EIS, suggests the opposite. The comment concedes that "running a regulatory ozone model.... is impractical,” and notes that the BLM "may be reluctant to estimate potential ozone impacts with a conservative method such as VOC/NO[x] point source screening tables” (also known as the "Scheffe Method”). However, the EPA noted that "development in the project area includes sources of [VOCs] and [NOx] ... and the additional development proposed ... might increase these emissions significantly.” As a result, the EPA "recommend[ed] that BLM estimate potential ozone impacts using a screening table or other non-regulatory method” and - that the final RMP/EIS "disclose estimated VOC and NO[x] emissions.” The BLM's only response to this comment was that it considered the Scheffe method to be "too conservative to provide results of any real value,” but the BLM has provided no citation to contemporaneous evidence in the record that discloses the basis for this conclusion. [22] The BLM's responses to comments appear to suggest that although it believed that additional development might very well produce higher levels of NOx emissions {i.e. from the operation of drilling and compressor equipment), there would not be a corresponding increase in reactive VOCs {i.e. from hydrocarbon emissions from the wells themselves). This may be correct, and, indeed, may ultimately justify the BLM reaffirming its decision that ozone modeling is unnecessary, but the Court cannot say that the citations to the record by the BLM in this action are sufficient to support this conclusion. [23] The Court emphasizes that it is not necessarily directing that the BLM perform complete ozone modeling, apply the Scheffe Method, or otherwise take any particular action with regard to assessing ozone impacts. It may be that, upon reconsideration, the BLM concludes that all that is necessary is a more detailed explanation for its conclusions (or indeed, merely citation to more appropriate portions of the Administrative Record). The Court does not opine or intend to offer any particular direction as to how the BLM should proceed upon remand.