Ctr. For Biological Diversity v. U.S. Dep't of the Interior

Full title: CENTER FOR BIOLOGICAL DIVERSITY; LIVING RIVERS; COLORADO RIVERKEEPER; UTAH…

Court: United States Court of Appeals, Tenth Circuit

Date published: Jul 10, 2023

72 F.4th 1166 (10th Cir. 2023)

Opinion

No. 21-4098

07-10-2023

CENTER FOR BIOLOGICAL DIVERSITY; Living Rivers; Colorado Riverkeeper; Utah Rivers Council; Sierra Club, Plaintiffs - Appellants, v. U.S. DEPARTMENT OF THE INTERIOR; U.S. Bureau of Reclamation, Defendants - Appellees, and Washington County Water Conservancy District; State of Utah; Utah Board of Water Resources, Defendant Intervenors - Appellees. Ute Indian Tribe of the Uintah and Ouray Reservation, Amicus Curiae.

William John Snape, III, Center for Biological Diversity, American University Law School, Washington, D.C. (Lisa T. Belenky, Center for Biological Diversity, Oakland, California; Christine T. Greenwood, Magleby Cataxinos & Greenwood, Salt Lake City, Utah; Edward B. Zukoski, Center for Biological Diversity, Denver, Colorado, with him on the briefs), for Plaintiff-Appellants. John Emad Arbab, Attorney, Environment and Natural Resources Division, United States Department of Justice, Washington, D.C. (Todd Kim, Assistant Attorney General and Jennifer A. Najjar and Brian C. Toth, Attorneys, Environment and Natural Resources Division, United States Department of Justice, Washington, D.C. and Susannah Thomas, Of Counsel Attorney, Office of the Solicitor, United States Department of the Interior, Washington D.C., with him on the brief), for Defendants-Appellees. Erin T. Middleton, Assistant Solicitor General, Office of the Utah Attorney General, Salt Lake City, Utah (Sean D. Reyes, Utah Attorney General, Anthony L. Rampton, Kathy A.F. Davis and Kelsie L. Last, Assistant Attorneys General, Office of the Utah Attorney General, Salt Lake City, Utah, with her on the brief), for Intervenor Defendants-Appellees. Michael W. Holditch, Frances C. Bassett and Jeremy J. Patterson, Patterson Earnhart Real Bird & Wilson LLP, Louisville, Colorado, filed an Amicus Curiae Brief for the Ute Indian Tribe of the Uintah and Ouray Reservation.

TYMKOVICH, Circuit Judge.

Appeal from the United States District Court for the District of Utah (D.C. No. 2:19-CV-00636-DBB)

William John Snape, III, Center for Biological Diversity, American University Law School, Washington, D.C. (Lisa T. Belenky, Center for Biological Diversity, Oakland, California; Christine T. Greenwood, Magleby Cataxinos & Greenwood, Salt Lake City, Utah; Edward B. Zukoski, Center for Biological Diversity, Denver, Colorado, with him on the briefs), for Plaintiff-Appellants. John Emad Arbab, Attorney, Environment and Natural Resources Division, United States Department of Justice, Washington, D.C. (Todd Kim, Assistant Attorney General and Jennifer A. Najjar and Brian C. Toth, Attorneys, Environment and Natural Resources Division, United States Department of Justice, Washington, D.C. and Susannah Thomas, Of Counsel Attorney, Office of the Solicitor, United States Department of the Interior, Washington D.C., with him on the brief), for Defendants-Appellees. Erin T. Middleton, Assistant Solicitor General, Office of the Utah Attorney General, Salt Lake City, Utah (Sean D. Reyes, Utah Attorney General, Anthony L. Rampton, Kathy A.F. Davis and Kelsie L. Last, Assistant Attorneys General, Office of the Utah Attorney General, Salt Lake City, Utah, with her on the brief), for Intervenor Defendants-Appellees. Michael W. Holditch, Frances C. Bassett and Jeremy J. Patterson, Patterson Earnhart Real Bird & Wilson LLP, Louisville, Colorado, filed an Amicus Curiae Brief for the Ute Indian Tribe of the Uintah and Ouray Reservation. Before TYMKOVICH, MORITZ, and ROSSMAN, Circuit Judges. TYMKOVICH, Circuit Judge.

This appeal arises from the Bureau of Reclamation's Environmental Analysis of a proposed water contract between it and Utah involving water in the Green River Basin. The so-called Green River Block Exchange contract allows Utah to draw water, in an equal amount, from releases from Flaming Gorge Reservoir instead of depleting water from the Green River and its tributary flows to which Utah is entitled under Article XV(b) of the Upper Colorado River Basin Compact of 1948.

Conservation Groups¹ sued Reclamation and the U.S. Department of the Interior alleging violations of the National Environmental Policy Act and the Administrative Procedure Act. The district court found that Reclamation's NEPA analysis was not arbitrary and capricious, that the agency took a "hard look" at cumulative impacts, and that it properly determined that an Environmental Impact Statement was not required. In essence, the proposed action would merely change Utah's point of diversion of water to which it is already entitled; it does not change or increase the amount of water to which Utah is entitled.

1 Center for Biological Diversity, Living Rivers, Colorado Riverkeeper, Utah Rivers Council, and Sierra Club.

We affirm. The record adequately demonstrates that Reclamation took a hard look at the proposed action and provided a reasoned explanation of its decision. In short, the agency concluded the exchange contract would not significantly affect the water or fish resources in the Green River Basin, and adequately explained the methodology and scope of its decision.

I. Background

Because this case deals extensively with complex issues of water rights in the Green River and Colorado River Basin system, we first review the region's water law scheme. We then consider the specifics of the Green River, the exchange contract, and Reclamation's environmental assessment.

A. Colorado River Water Rights

The Green River is a major tributary of the Colorado River. According to Reclamation,

the Colorado River and its tributaries provide water to nearly 40 million people for municipal use, supply water to irrigate nearly 5.5 million acres of land, and is the lifeblood for at least 22 federally recognized tribes (tribes), 7 National Wildlife Refuges, 4 National Recreation Areas, and 11 National Parks.

App. 322.

Image materials not available for display.

The Colorado River Basin [App. 321]

Multiple interstate compacts, legislation, and agreements govern water use from the Colorado River. See, e.g., id. at 421-28. The Colorado River Compact of 1922, negotiated between the federal government and the seven Colorado River Basin states,² divided the Colorado River Basin into upper and lower sub-basins, apportioning 7.5 million acre-feet (AF) per year to each "in perpetuity."³ 70 Cong. Rec. at 325. The Upper Basin states (Arizona, Colorado, New Mexico, Utah, and Wyoming) later executed the Upper Colorado River Basin Compact of 1948 "to determine the rights and obligations of each signatory State respecting the uses and deliveries of the water of" the basin. Pub. L. No. 81-37, 63 Stat. 31 (1949). The 1948 Compact also apportioned these consumptive water rights "in perpetuity." Id. at 32 (Article III(a)).

2 Colorado, New Mexico, Utah, Wyoming, Arizona, California, and Nevada. See 70 Cong. Rec. 324 (1928), 46 Stat. 3000 (1929).

3 An acre-foot is equivalent to 325,851 gallons of water, or enough water to cover an acre of land 1 foot deep, and is the common unit of measurement when talking about quantities of water at scale.

Under these compacts, Utah is entitled to 23 percent of the water apportioned to and available for use in the Upper Basin. Id. at 33. The 1948 Compact states:

The provisions of this Compact shall not apply to or interfere with the right or power of any signatory State to regulate within its boundaries the appropriation, use and control of water, the consumptive use of which is apportioned and available to such State by this Compact.

Id. at 41 (Article XV(b)). Under these compacts, Utah does not need further federal authorization to regulate the appropriation, use, and control of its apportioned water rights.

B. Developments along the Green River

The 1956 Colorado River Storage Project Act allowed for comprehensive development of the water resources of the Upper Basin states. It did so by providing for long-term regulatory storage of water for beneficial purposes, thereby allowing Upper Basin states to use their compact apportionments, providing for the reclamation of arid lands, controlling floods, and generating hydroelectric power. See 43 U.S.C. § 620.

The Storage Project Act also authorized the construction of several projects along the Green River, including the Flaming Gorge Unit and the Central Utah Project. App. 152-53. In 1958 Reclamation filed a Utah application to appropriate 3.96 million AF from the Green River for storage in the newly built Flaming Gorge Reservoir and for Central Utah Project purposes including irrigation, municipal, domestic, and industrial uses. Id. at 181. Reclamation segregated 52,500 AF for other purposes, leaving 447,500 AF of undeveloped water rights with the United States. Id.

Image materials not available for display.

The Green River Downstream of Flaming Gorge Reservoir [App. 360]

The Central Utah Project was planned as a phased development, with the "Initial Phase" units being joined in time by the "Ultimate Phase" units of the project involving reservoir projects on the Green River in addition to Flaming Gorge. App. 153. In 1992 Congress defunded the undeveloped portions of the Ultimate Phase units with the passing of the Central Utah Project Completion Act. Reclamation Projects Authorization and Adjustment Act of 1992, Pub. L. No. 102-575, 106 Stat. 4600 (1992). Reclamation returned the remaining undeveloped Ultimate Phase water rights—447,500 AF—to Utah in the March 1996 "Assignment Agreement." App. 303.

The 1996 Assignment states in relevant part:

Upon release from Flaming Gorge Reservoir, said water right can be developed, diverted and perfected by the State of Utah as permitted by law. The State of Utah agrees that if it stores water in or benefits directly from Colorado

River Storage Project Facilities, the State of Utah will enter into a water service contract with the United States.

Id. This case is concerned with a portion of that assigned water right, specifically the 72,641 AF known as the Green River Block, because it was "expected that this water would be predominately developed along the Green River and its tributaries between [Flaming Gorge] and Lake Powell." Supp. App. 31.

Since the 1996 Assignment, Utah has entered into agreements with Utah water users to beneficially use portions of the Green River Block. These water users have since perfected 13,684 AF of water along the Green River and its tributaries. Id. at 31-32. These developed water rights, along with the as yet undeveloped 58,957 AF from the Green River Block, are the subject of the exchange contract.

C. The Exchange Contract

In January 2016, Utah sent a letter to Reclamation requesting two contracts for the use of its assigned water rights. App. 182. The first contract was for the Lake Powell Pipeline and represented 86,249 AF of depletion.⁴ Id. The second was for the Green River Block water assignment—the 72,641 AF depletion mentioned above—and would in effect be a one-for-one exchange of water. Id. Reclamation characterized the two contracts as "separate and distinct, each covering different blocks of water to be developed under different circumstances and wholly independent of each other." Id.

4 The Lake Powell Pipeline is a planned project to "divert water from Lake Powell and deliver it through a pipeline to Washington and Kane counties in southwestern Utah." App. 187.

For the Green River Block exchange contract, Utah agreed to "forbear" the depletion of a portion of the Green River and tributary flows to which it is entitled under Article XV(b) of the 1948 Compact. Instead, it would allow these natural flows to contribute to meeting the Endangered Species Act Recovery Program requirements in Reaches 1 and 2 of the River.⁵ App. 156. Utah's forbearance would "assist Reclamation in meeting its obligation under the 2006 [Flaming Gorge Record of Decision]." Id. "In exchange, [Utah] is authorized to deplete an equal amount of [Colorado Storage] Project Water from releases from the Flaming Gorge Unit throughout the year as water is needed for the Assigned Water Right." Id.

5 Reclamation's commitments under the ESA remain subject to the Recovery Program requirements and obligations outlined in Section 7 of the 2006 Flaming Gorge Record of Decision. Supp. App. 325-27.

The contract also specified that the one-for-one exchange of direct flow water for project water would be reconciled on an annual basis and that releases of water would be "in accordance with the flexibility in Reclamation's operations under the 2006 [Flaming Gorge] ROD," not on demand from Utah. Id. Utah also agreed to pay Reclamation a yearly fee equivalent to $19.00 per acre-foot and a proportionate share of the annual operation and maintenance costs for the Flaming Gorge Unit, $3.37 per acre-foot. Id. at 157-58.

D. The Environmental Assessment

Prior to signing the contract, Reclamation issued a draft environmental assessment (EA) for public comment in accordance with NEPA requirements. App. 177. Reclamation held a public meeting in Utah on September 26, 2018, to discuss the draft EA. Id. A 50-day comment period started on September 19 and ended November 2, 2018. Id. Reclamation included those comments and its responses to them in its final EA, released in January 2019. Id. at 175, 177.

The final EA considered two action alternatives: the no-action alternative and the proposed-action alternative. Id. at 177. Under the former, Utah and Reclamation would not enter into an exchange contract and Utah would remain entitled to deplete water from the Green River and tributaries in accordance with its water rights. Id. The latter consists of the block exchange of storage project water for Green River water. Id. at 177-78. Reclamation analyzed a total of 15 resources based on both alternatives. Id. at 179. With respect to hydrology, the agency found that "[s]mall differences were predicted during the months of July-September during drier hydrologic conditions." Id. For the remaining 14 resources, Reclamation found "no effect or [a] similar determination." Id.

Reclamation concluded that the proposed action would not have a significant impact on the quality of the human or natural environment. Id. at 176. Because of the agency's Finding of No Significant Impact (FONSI), no environmental impact statement (EIS) was required. Reclamation signed the Green River Block exchange contract on March 20, 2019. Id. at 152.

II. Analysis

Conservation Groups contend Reclamation failed to adequately support its EA conclusions in every significant respect and should have conducted an environmental impact statement.

We disagree and affirm. We conclude (1) Reclamation took a "hard look" at the environmental impacts to hydrology and fish resources; (2) Reclamation's no-action alternative used an appropriate environmental baseline to analyze the potential impacts of the contract; and (3) Reclamation reasonably concluded that the contract would not have a significant effect on the environment, thus negating the need for an EIS.

A. Standard of Review and Statutory Background

Because NEPA does not provide for private causes of action, courts review an agency's compliance with NEPA under the Administrative Procedure Act. 5 U.S.C. § 706; Utah Shared Access All. v. Carpenter, 463 F.3d 1125, 1134 (10th Cir. 2006). This court's "review of the lower court's decision in an APA case is de novo," and "owe[s] no deference to the district court's decision." N.M. Cattle Growers Ass'n v. U.S. Fish & Wildlife Serv., 248 F.3d 1277, 1281 (10th Cir. 2001) (citations omitted).

Under the APA, courts "shall . . . hold unlawful and set aside agency action, findings, and conclusions found to be arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law." 5 U.S.C. § 706(2)(A) (cleaned up). An action is arbitrary and capricious,

if the agency . . . 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 [if the decision] is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.

Utah Envtl. Cong. v. Richmond, 483 F.3d 1127, 1134 (10th Cir. 2007) (quoting Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)).

It is the duty of the reviewing court to "ascertain whether the [agency] examined the relevant data and articulated a rational connection between the facts found and the decision made." Colorado Wild, Heartwood v. U.S. Forest Serv., 435 F.3d 1204, 1220 (10th Cir. 2006). "In reviewing the agency's explanation, the reviewing court must determine whether the agency considered all relevant factors and whether there has been a clear error of judgment." Citizens' Comm. to Save Our Canyons v. Krueger, 513 F.3d 1169, 1176 (10th Cir. 2008) (quoting Olenhouse v. Commodity Credit Corp., 42 F.3d 1560, 1574 (10th Cir. 1994)). This includes a "thorough, probing, in-depth review" of the administrative record. Ron Peterson Firearms, LLC v. Jones, 760 F.3d 1147, 1162 (10th Cir. 2014) (internal quotation marks omitted).

"It is well-established that an agency's action must be upheld, if at all, on the basis articulated by the agency itself." State Farm, 463 U.S. at 50, 103 S.Ct. 2856. "Thus, the grounds upon which the agency acted must be clearly disclosed in, and sustained by, the record." Olenhouse, 42 F.3d at 1575. "[C]ourts may not accept appellate counsel's post hoc rationalizations for agency action." State Farm, 463 U.S. at 50, 103 S.Ct. 2856. Nevertheless "we will uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned." Licon v. Ledezma, 638 F.3d 1303, 1308 (10th Cir. 2011) (internal quotation marks omitted).

NEPA requires all federal agencies to take a "hard look" at the environmental impacts of their decisions before the decision is made. See 42 U.S.C. § 4332(2)(C); Wyoming v. U.S. Dep't of Agric., 661 F.3d 1209, 1263 (10th Cir. 2011). Conclusory statements regarding impacts without adequate discussion do not meet the required "hard look" under NEPA. Davis v. Mineta, 302 F.3d 1104, 1122-23 (10th Cir. 2002). The agency's "hard look" analysis must utilize "public comment and the best available scientific information." Colo. Envtl. Coal. v. Dombeck, 185 F.3d 1162, 1171 (10th Cir. 1999) (citations omitted).

NEPA obligates federal agencies "to consider every significant aspect of the environmental impact of a proposed action." Utah Shared Access All. v. U.S. Forest Serv., 288 F.3d 1205, 1207 (10th Cir. 2002) (quoting Balt. Gas & Elec. Co. v. Nat. Res. Def. Council, 462 U.S. 87, 97, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983)). Environmental impacts include direct, indirect, and cumulative effects of the action. Hillsdale Envtl. Loss Prevention, Inc. v. U.S. Army Corps of Eng'rs, 702 F.3d 1156, 1166 (10th Cir. 2012).

Agencies must prepare a "detailed statement" for all "major Federal actions significantly affecting the quality of the human environment." 42 U.S.C. § 4332(2)(C). To determine whether a proposed action significantly affects the environment—requiring preparation of an environmental impact statement—an agency may first prepare an environmental assessment. 40 C.F.R. § 1501.4(c) (2019).⁶

6 The Council on Environmental Quality updated the NEPA regulations in 2020, see 85 Fed. Reg. 43,304 (July 16, 2020), and recently amended certain of the Regulations, see 87 Fed. Reg. 23,453 (Apr. 20, 2022). Reclamation's actions are subject to the previous regulations because the actions were all completed prior to the effective date of the new regulations and because Reclamation applied the prior regulations. See Bair v. Cal. Dep't of Transp., 982 F.3d 569, 577 n.20 (9th Cir. 2020) ("Because [the agency at issue] applied the previous [NEPA] regulations to the Project, so do we."). All citations to NEPA regulations are to those in effect before September 14, 2020.

In an EA or EIS, an agency must analyze cumulative impacts from a project. 40 C.F.R. § 1508.25(c) (stating that the "scope" of an EIS includes consideration of "cumulative" impacts). "Cumulative impacts" result from the "incremental impact of the action" on the environment "when added to other past, present, and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person undertakes such other actions." 40 C.F.R. § 1508.7.

"Cumulative impacts can result from individually minor but collectively significant actions taking place over a period of time." Id. Cumulative impact analyses include private, state, and federal actions. An agency is required to analyze the environmental consequences of these future actions regardless of whether they result from action by a federal or state agency or individuals. Colo. Envtl. Coal., 185 F.3d at 1176.

"NEPA does not mandate any particular substantive result . . . it 'prescribes the necessary process' that must accompany agency action," Wild Watershed v. Hurlocker, 961 F.3d 1119, 1122 (10th Cir. 2020) (quoting Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989)), and that forces agencies to take a "hard look at the environmental consequences of its actions," Wyoming, 661 F.3d at 1264 (internal quotation marks omitted).

B. Water Availability and Hydrologic Impacts

Conservation Groups contend Reclamation's EA failed to take a hard look at (1) the effects of warming on future water availability in the Green River; (2) the impacts resulting from location, timing, and volume of new depletions from the contract on Reach 3; and (3) the cumulative impacts of multiple and significant water depletions from the Green and Colorado Rivers.

1. Future Water Availability

Conservation Groups first argue the agency failed to address scientific data and studies projecting that climate warming in the future is likely to leave the Colorado River system drier than it has been in the past. They reason that this change in the river's hydrology increases the potential harm from the contract's expected water depletions and changes in flow.

Conservation Groups' specific criticisms are two-fold. The first is that Reclamation did not use the best available science or adequately explain why it chose to ignore Conservation Groups' preferred scientific studies. The second is Reclamation's use of "trace 63,"⁷ a modeling period starting in 1969 and continuing through 2015, for its Colorado River Simulation System modeling on the effects of the contract on the Green River system. App. 190.

7 A trace is a simulation of possible future inflows derived from the historical record of natural flows. Reclamation uses a particular technique—the Indexed Sequential Method, or ISM—for sampling from the historical record. Each future hydrologic sequence is generated from the historical natural flow record by "cycling" through the record. For example, in 2007 Reclamation used the historical natural flow data from 1906 through 2005 to arrive at a set of 100 separate simulations, or traces. App. 470.

Conservation Groups point to three scientific studies—Udall & Overpeck 2017, McCabe et al. 2017, and Xiao et al. 2018—that Reclamation did not include in its EA and argue their absence means Reclamation did not use the best available science. They contend the studies show an accelerating decline in river volumes in future years. Conservation Groups also note the U.S. Fish and Wildlife Service (FWS), National Park Service (NPS), and others also referenced these studies in their comments to Reclamation's draft EA. App. 299-302.

Reclamation responded by stating that it incorporated the comments into its final EA and hydrologic modeling technical report. Id. at 299, 302. In particular, it added a drought response section to its technical appendix, which addressed the comments with a reasoned explanation justifying the use of Direct Natural Flow (DNF) hydrology sets, such as trace 63, that contain multiple periods of drought. Id. at 211. "In order to determine the impacts of continued drought," Reclamation used "the trace with the lowest elevation" of water levels in Flaming Gorge Reservoir to represent the "worst-case scenario." Id. It was Reclamation's judgment that using trace 63 as the worst-case scenario, and a scenario "statistically unlikely to occur," was an appropriate modeling data set, or surrogate, to capture any unanticipated effects of water shortages and climate-change concerns. Id. at 212.

Reclamation's response to FWS's climate concerns lacks clarity and the dissent rightly takes the agency to task over it. Nevertheless, however meritorious FWS views forward-looking data, Reclamation's response sufficiently evaluated the minimal ⁸ environmental effects of the proposed action.⁹ In response to "[c]oncerns over a changing climate," Reclamation explained in its technical appendix that its "hydrology set contains multiple periods of drought," including the drought years of 2000-2015. Supp. App. 111-12. "It is the period of operations between 2000-2015 that ha[s] the greatest impact on elevation." Id. at 112. Reclamation isolated "the trace with the lowest elevation" and included its results in its overall modeling "[i]n order to determine the impacts of continued drought." Id. at 111-12.

8 The impact of the exchange contract on the environment is discussed in greater detail in Part II.D, infra.

9 Reclamation's full response makes clear it took a "hard look" at the environmental impact of the exchange contract:

Concerns over a changing climate have been prominent in environmental and water resources. The DNF hydrology set contains multiple periods of drought, including the decades of drought that occurred in the 1930s, 1950s, 1970s and 2000 up to 2015. In order to determine the impacts of continued drought, the trace with the lowest elevation has been isolated and its results have been included. Trace 63 begins with the initial conditions and then historic year 1979 is the first hydrologic year of that trace. This trace moves through the wet years in the 1980s, but ends with the drought in 2000-2015. It is the period of operations between 2000-2015 that have the greatest impact on elevation. The impact trends of implementing the exchange agreement are seen in the worst-case scenario. The illustrations in the drought trace 63 should be considered one representation of potential possibilities of future hydrology and it is statistically unlikely that trace 63 will happen.

Supp. App. 111-12.

Reclamation found that "[t]he impact trends of implementing the exchange agreement are seen in the worst-case scenario." Id. at 112. Even in the worst-case scenario—15 consecutive years of drought—Reclamation found that implementation of the exchange contract would still result in Reservoir water levels above the minimum required level by the Flaming Gorge EIS. Supp. App. 113. The technical appendix does not mention FWS's three studies by name nor does it explicitly state that it prefers backward-looking data to forward-looking data, but that path is reasonably discernable from the record. See generally Licon, 638 F.3d at 1308-09.

FWS's comment suggests that Reclamation ignored hydrologic changes or trends associated with warming temperatures, but the record shows otherwise. Supp. App. 271. Reclamation's trace 63, which uses historical data from 1969 through 2015, incorporates the effects of these changes, i.e., increased or continued drought. While Reclamation's response to FWS's comment could have been more robust, the record confirms that Reclamation adequately incorporated in its analysis the effects of a warming climate and the likelihood of changes in hydrology. And it is worth remembering that the EA was not about changing Utah's entitlement to Upper Basin water, but changing the point of diversion of water that Utah was already using.

Reclamation's focus on the relevant geographic scale for the analysis of the exchange contract was also reasonably discernable in its decision to use trace 63 and its own models and simulations.¹⁰ While Reclamation only explicitly made this argument in response to a comment to the draft EA by Utah Rivers Council, Supp. App. 265, the refrain of geographic scale resonates throughout Reclamation's record. Utah Rivers Council commented that Reclamation's "EA ignores agency-documented risks from expected water supply shortage declarations and contains major inconsistencies between available water supplies in the Colorado River Basin" and Utah's water rights for the Lake Powell Pipeline and the Green River Block. Id. at 167. Reclamation responded that the 2012 Study analyzed "the overall Colorado River basin," spanning a geographic area of seven states and millions of square miles, whereas the EA "provides a specific and detailed look at the impacts" of the contract on the upper reaches of the Green River Basin in northeastern Utah. Id. at 265.

10 Colorado River Simulation System (CRSS), DNF models, and the ISM. App. 203-37 (Technical Appendix to the final EA).

The introduction to Reclamation's technical appendix, which the agency referenced throughout its responses, makes clear that Reclamation conducted its modeling of the exchange contract's effects within the context of its 2005 Flaming Gorge Final Environmental Impact Statement. Supp. App. 104. Commensurate with the limited scope of its modeling, Reclamation's technical assumptions were "different than standard [Colorado River Simulation System] model runs that are used in a long-term basin-wide planning context." Id. Reclamation's purpose here was simulating "the difference between diverting water out of the Green River directly below Flaming Gorge Dam (FG) and not diverting the water." Id. at 104-05 (emphasis added). Reclamation was already working within the confines of its 2005 Flaming Gorge Environmental Impact Statement and 2006 Flaming Gorge Record of Decision, not attempting a "long-term basin-wide planning context" like the 2012 Colorado River Basin Water Supply and Demand Study. Supp. App. 104.

Put simply, Reclamation was asked how moving the point of depletion of 58,957 AF per year from the Green River's tributaries to just below Flaming Gorge Reservoir would affect the environment. Reclamation did just that by analyzing the exchange contract's impacts on a discrete part of the Green River. Comments attempting to expand the scope of Reclamation's EA were not germane to the inquiry. For reference, the Colorado River Compact of 1922 apportioned 7.5 million AF per year to each of the Upper and Lower Basins. In preparation for storage in the soon-to-be-constructed Flaming Gorge Reservoir, Reclamation appropriated 3.96 million AF per year from Utah. We do not make these comparisons to slight the scope of the exchange contract or its importance, but only to illustrate that warming temperatures and the probability of decreased future volume in the Colorado River Basin were macro concerns for Reclamation's assessment, while the point of depletion for water Utah already had a right to use in the Green River Basin was the relative concern and focus of the EA. Reclamation considered the appropriate factors for the scale of its assessment.

With that in mind, we cannot say that Reclamation's choice of geographic scope was error. Nor can we fault Reclamation's choice of the best science, in its prerogative, to meet the chosen scale of analysis. Deference to the agency "is especially strong where the challenged decisions involve technical or scientific matters within the agency's area of expertise." Morris v. U.S. Nuclear Reg. Comm'n, 598 F.3d 677, 691 (10th Cir. 2010).

With respect to the three scientific studies that Reclamation chose not to use or reference by name in the final EA, we are satisfied with Reclamation's rationale that those studies' geographic scales—the entire Colorado River Basin or the entire Upper Basin¹¹—were not appropriate to evaluate the more limited hydrologic impact of the exchange contract.¹² See Kleppe v. Sierra Club, 427 U.S. 390, 414, 96 S.Ct. 2718, 49 L.Ed.2d 576 (1976) (the "determination of the extent and effect of [cumulative impacts], and particularly identification of the geographic area within which they may occur, is a task assigned to the special competency of the appropriate agencies."); San Juan Citizens All. v. Stiles, 654 F.3d 1038, 1057 (10th Cir. 2011) ("Setting the boundaries of the region to be analyzed involved technical and scientific judgments within the [agencies'] area of expertise, and their conclusion regarding which Class I sites to include in the analysis is one to which we defer.").

11 Udall & Overpeck 2017 addresses "future climate change impacts on the Colorado River" as a whole in both the Upper and Lower Basins. App. 397-400. McCabe et al. 2017 focuses on the entire Upper Colorado River Basin. App. 366. Like McCabe, Xiao et al. 2018 focuses on the entire Upper Colorado River Basin. App. 381-82.

12 This determination was consistent with Reclamation's own assessments, such as the 2012 Study's water-supply report, which explained in regard to future water shortages that while future precipitation is expected to vary within the Colorado River Basin depending on location, there will be an increase in expected precipitation "by up to 10 percent in the Upper Basin at higher elevation and toward the north (Green River Basin)." Supp. App. 317.

2. Reach 3

Conservation Groups next argue that Reclamation failed to take a hard look at the impacts of new water depletions on fish resources in Reach 3, a 246-mile stretch of the Green River between the confluences of the White and Colorado Rivers. They claim Reclamation modeled only the effects in Reaches 1 and 2.¹³ Maintaining specific water temperature and flow rates is critical for many of the endangered fishes in Reach 3, such as the bonytail, Colorado pikeminnow, humpback chub, and razorback sucker. Supp. App. 63, 66-70.

13 Reach 1 begins directly below Flaming Gorge Reservoir and extends to the confluence of the Yampa River. Reach 2 begins at the Yampa River confluence and ends at the White River confluence. App. 188.

Image materials not available for display.

Green River Reaches [Supp. App. 283]

NPS and FWS raised similar concerns over how flow recommendations and targets in Reach 3 would be met in their comments to the draft EA. App. 255, 297. Reclamation explained it modeled only the first two reaches because it "assume[d] that the 2000 Flow and Temperature objectives in Reach 3 are met whenever the flow objectives are met in Reach 2." Id. at 299 [Response to Comment 31]. Reclamation also justified its decision based on the geographic scope of their analysis, explaining that depletions "below Reach 2" are "not included in the geographical boundaries of the analysis." Id. at 206. Based on its modeling, Reclamation found that the exchange contract would have "minimal impacts on hydrology" in Reach 2, id. at 191, and that under these circumstances Reclamation would "continue[ ] to meet its commitments under the [Flaming Gorge Record of Decision]" for flow targets in Reaches 1 and 2. Id. at 301 [Response to FWS Comment].

Given Reclamation's extensive environmental impact analyses under the Flaming Gorge ROD and EIS, Reclamation's technical assumption was well-reasoned and explained by the administrative record—Reach 3 would continue to be adequately served given water temperature and flow rates were met in Reaches 1 and 2. Moreover, Reclamation's additional rationale that it was concerned with determining whether implementing the exchange contract would take the agency outside the parameters of the Flaming Gorge ROD/EIS is ample justification for its decision to not adjust the geographic scale of its analysis or change the relevant assumptions away from those used in the Flaming Gorge ROD and EIS.

Even under Reclamation's chosen modeling, the full-depletion scenario—i.e., a worst-case scenario—would have only a negligible impact on the endangered fish species because of "minimum required flows and cold-water biology of the endangered fishes." Supp. App. 72. There "would be no impact to the endangered fish habitat within the Green River." Id. Moreover, under more likely circumstances than the statistically improbable full-depletion scenario, the EA found that "[t]he additional summer flows potentially created under the Proposed Action could provide benefit to the endangered fishes," because Utah would not be depleting water from the Green River tributaries but instead drawing water directly below Flaming Gorge Reservoir. Id. at 71.

We are satisfied that the agency's technical assumption was well-reasoned and supported by the administrative record. Reclamation's choice of methodology for its site-specific analysis was reasonable and entitled to deference. Citizens' Comm. to Save Our Canyons v. U.S. Forest Serv., 297 F.3d 1012, 1027 (10th Cir. 2002) (holding that the plaintiff's attack on an agency decision failed because it "ignore[d] the general rule that courts defer to the expertise and discretion of the agency to determine proper testing methods") (internal quotation marks omitted); Colo. Envtl. Coal., 185 F.3d at 1172 (concluding that the appellants "fail[ed] to show how additional, site-specific [wildlife] data is 'essential' to reasoned decision making").

Conservation Groups additionally argue that Reclamation failed to take a hard look in its EA because it did not specifically identify the location and timing of depletions downstream of Flaming Gorge Reservoir. Because we agree with Reclamation that the exchange contract does not result in new water depletions, but rather exchanges storage project water for Green River tributary flows—discussed in greater detail below—we do not find Conservation Groups' argument persuasive that a lack of specificity as to the diversion points is indicia of a flawed EA. See generally App. 204 ("These depletions and diversions were covered in the [Flaming Gorge EIS] and are being analyzed for the purpose of signing [the exchange contract].").

3. Cumulative Impacts

Conservation Groups also contend that Reclamation failed to take a hard look at the cumulative impacts of the exchange contracts because it ignored reasonably foreseeable water depletions from the Green and Colorado Rivers in the other Upper Basin states. Reclamation evaluated the cumulative effects to hydrology in its "Full Depletion Scenario" within the EA, in which the agency described its working assumptions:

Under the Full Depletion Scenario, all assumptions from the [Green River Block] Depletion Scenario are maintained, with the addition of reasonably foreseeable depletions held constant at 2060 levels with all other depletions held constant at 2018 depletion levels.

App. 189. Under these assumptions, Reclamation included only those certain future depletions that met its definition of a reasonably foreseeable depletion. Id. at 206-07. Reclamation found that "[c]umulatively, there would not be a significant impact to hydrology based on the analysis performed in this EA." Id. at 193.

Conservation Groups' argument is unpersuasive. Reclamation did not ignore the possibility of future depletions; its technical criteria simply removed from consideration those potential depletions that were too indefinite to merit inclusion. The EA explains "a reasonably foreseeable future depletion is one which has state legislation, or a tribal resolution or federal Indian water settlement, or a FONSI or ROD." Id. at 188. The technical appendix of the EA further explains that the agency's modeling "adopts a rigorous definition of what reasonably foreseeable future depletions are in the Upper Basin," which when applied to the modeling "takes the strictest approach to defining what is included and excluded for the cumulative impacts analysis required" under 40 C.F.R. § 1508.7. Id. at 237.

Using that definition, the only reasonably foreseeable future depletions the EA identifies are 12 such depletions in Utah. App. 207. No reasonably foreseeable future depletions were identified in the remaining Upper Basin states of Wyoming, Colorado, and New Mexico. Id. Rather than challenge Reclamation's definition of reasonably foreseeable future depletions, Conservation Groups contend the EA "provides no information to show that there are no permitted water uses or other reasonably foreseeable new uses in other Upper Basin states." Aplt. Br. at 41. Notwithstanding the fact that Reclamation is not required to prove a negative, the agency is entitled to a "presumption of validity" that attaches to its action and the "burden of proof rests with the appellants who challenge such action." New Mexico ex rel. Richardson v. Bureau of Land Mgmt., 565 F.3d 683, 704 (quoting Citizens' Comm., 513 F.3d at 1176).

Conservation Groups have not rebutted that presumption by introducing anything from the administrative record that would show Reclamation overlooked future depletions in other Upper Basin states that met the agency's definition of reasonably foreseeable future depletions. Reclamation, using its expertise, made a "reasonable, good faith, objective presentation of those [cumulative] impacts sufficient to foster public participation and informed decision making." Colorado Envtl. Coal., 185 F.3d at 1177.

Accordingly, we conclude Reclamation's cumulative-impacts analysis is reasonable and supported by the record.

C. No-action Alternative

Conservation Groups argue that Reclamation's choice of a no-action alternative violated NEPA. Reclamation's choice of alternatives to its proposed action is relevant because the agency, in order "[t]o comply with [NEPA] and its implementing regulations," was "required to rigorously explore all reasonable alternatives to the Initiative, including a 'no-action' alternative." Custer Cnty. Action Ass'n v. Garvey, 256 F.3d 1024, 1039 (10th Cir. 2001). We employ the "rule of reason" when evaluating the adequacy of an agency's alternative analysis "to ensure the [EA] contained sufficient discussion of the relevant issues and opposing viewpoints to enable the [agency] to take a hard look at the environmental impacts of the proposed [action] and its alternatives, and to make a reasoned decision." Colo. Envtl. Coal., 185 F.3d at 1174 (citing All Indian Pueblo Council v. United States, 975 F.2d 1437, 1445 (10th Cir. 1992)). "The rule of reason guides both the choice of alternatives as well as the extent to which the [EA] must discuss each alternative." Custer Cnty. Action Ass'n, 256 F.3d at 1040 (quoting Am. Rivers v. FERC, 201 F.3d 1186, 1200 (9th Cir. 1999) (internal quotation marks omitted)).

Whether an EA or EIS is prepared, NEPA requires that agencies "study, develop, and describe appropriate alternatives," including a no-action alternative. 42 U.S.C. § 4332(2)(E). "In general, NEPA analysis uses a no-action alternative as a baseline for measuring the effects of the proposed action." Biodiversity Conservation All. v. U.S. Forest Serv., 765 F.3d 1264, 1269 (10th Cir. 2014). "The no action alternative may be thought of in terms of continuing with the present course of action until that action is changed. It establishes a baseline against which the proposed action and its alternatives may be measured." Id. (quoting George Cameron Coggins and Robert L. Glicksman, Discussion of Alternatives—The "No Action" Alternative, 2 Pub. Nat. Resources L. § 17:47 (2d ed., Feb. 2023 update) (internal quotation marks omitted)).

Agencies need not "analyze the environmental consequences of alternatives it has in good faith rejected as too remote, speculative, or . . . impractical or ineffective." Colo. Envtl. Coal., 185 F.3d 1162 (quoting All Indian Pueblo Council, 975 F.2d at 1444 (internal quotation marks omitted)). "[W]hat is required is information sufficient to permit a reasoned choice of alternatives as far as environmental aspects are concerned." All Indian Pueblo Council, 975 F.2d at 1444 (quoting Nat. Res. Def. Council, Inc. v. Morton, 458 F.2d 827, 836 (D.C. Cir. 1972)).

Conservation Groups object to the EA's description of the exchange contract as merely shifting water depletions to a new location instead of their preferred characterization of the contract as " 'unlock[ing]' new withdrawals of water from the Green River" that Utah would otherwise not be legally entitled to. Aplt. Br. at 42. Conservation Groups argue that the EA erroneously assumed that the same amount of water would be diverted from the Green River whether the contract was signed (proposed-action alternative) or not (no-action alternative). In their view, the EA's language that Utah would "forebear the depletion of a portion of the Green River and tributary flows to which it is entitled" mistakenly implies that Utah would discontinue a current water depletion. App. 178. (emphasis added). Conservation Groups object to this language because most of the water rights to be exchanged via the contract have never been put to use, i.e., the vast majority of the 1996 Assignment of water rights remains undeveloped and unperfected.

The parties disagree over whether Utah can develop and perfect its remaining water rights without a water service contract with Reclamation. Conservation Groups, arguing Utah may not do so, rely on language in the 1996 Assignment for their contention, which states in relevant part:

Upon release from Flaming Gorge Reservoir, said water right can be developed, diverted and perfected by the State of Utah as permitted by law. The State of Utah agrees that if it stores water in or benefits directly from Colorado River Storage Project Facilities, the State of Utah will enter into a water service contract with the United States.

App. 303 (emphasis added). In briefing before the district court, intervenor-defendant State of Utah agreed that the water subject to the contract could not be developed without the contract either as tributary flows in the Green River or under the 1996 Assignment. Id. at 128-29.¹⁴ Utah subsequently walked back this language, saying the use of "Assignment" in the first two sentences of the quote were drafting mistakes. Utah Aple. Br. at 30. In both instances Utah contends it should have referred to the exchange contract instead of the Assignment. Id.

14 Utah explained the exchange contract was necessary because:

The Assignment conditioned the water right assignment to UBWR [Utah Board of Water Resources] on: 1) satisfying the water right through release of water in Flaming Gorge Reservoir—effectively restricting the State from diverting and using an equivalent amount of water from Green River Tributaries under its Upper Basin apportionment; 2) restricting diversions for that right to the main stem of the Green and Colorado Rivers; and 3) entering into a service contracting establishing a fee payable to [Reclamation] for UBWR's pro rata share of facility operation and maintenance costs associated with diverting water below Flaming Gorge Reservoir. The Assignment effectively locks up 158,890-acre feet of diversion (47,500-acre feet of depletion) of Utah's Upper Basin apportionment in releases from Flaming Gorge Reservoir allowing that volume of water in Utah's tributaries, particularly high spring flows, to reach the Green River without diminishment from future Utah storage projects (reservoirs) or direct diversions.

App. 128-29 (emphasis added).

The district court, as we are, was nevertheless unconvinced by Conservation Groups' argument, finding their "reading of the 'forebear' statement to imply a current use [to be] untenable," as the statement was in fact "a conditional statement setting forth a future action: If the contract is implemented, the State would not deplete flows to which it is entitled." App. 136. Because the "statement did not purport to describe the current water use situation," the court found Conservation Groups' argument about the appropriate baseline, or no-action alternative, for the EA to be without merit. Id. The district court was equally unpersuaded by Conservation Groups' argument that the final EA should have analyzed the impact of a new water use because Utah would in effect be depleting water under the contract far in excess of the amount of water it had been using from the Green River tributary flows. Id.

For its part, Reclamation maintains that its no-action alternative accurately describes the status quo or "current level of activity." Biodiversity Conservation Alliance, 765 F.3d at 1269. Reclamation points to Utah's development of 13,684 AF from the 1996 Assignment as proof that Utah is not limited in developing or perfecting its water rights solely through a water service contract with Reclamation for releases from Flaming Gorge. App. 186-87. As for the remaining 58,957 AF, Reclamation argues the no-action alternative correctly recognized that "[t]he State would remain free to develop [its] apportioned water right under the 1996 Assignment"—as Utah had already done for the earlier 13,684 AF portion of the block. Supp. App. 36.

Reclamation's position is that given the development of almost 14k AF of Green River Block water rights in the previous 20 years, "[i]t is not unreasonable that the State could develop a significant portion of the remaining Green River Block WR water in the next 40 years." Id. at 265. We agree. The record confirms Reclamation ran its no-action scenario with Green River Block depletions held constant at zero for the entire model run—save for the perfected 13.7k AF Utah was already depleting. App. 189. While Conservation Groups may object to Reclamation's inclusion of Utah's contested water rights in the EA, it had no material impact on the EA and was supported by the record.

Accordingly, we hold Reclamation's choice of the no-action alternative was neither arbitrary nor capricious.

D. The Finding of No Significant Impact

Conservation Groups' final contention is that Reclamation erred by issuing a FONSI instead of conducting an EIS. "An agency may issue a FONSI only if, after reviewing the direct and indirect effects of a proposed action, it concludes that the action 'will not have a significant effect on the human environment.' " WildEarth Guardians v. Conner, 920 F.3d 1245, 1261 (10th Cir. 2019) (quoting 40 C.F.R. § 1508.13). If major federal action may significantly affect the quality of the human environment, the agency must prepare an EIS. 42 U.S.C. § 4332(2)(C). "The significance of an impact is determined by the action's context and its intensity." Hillsdale Envtl. Loss Prevention, 702 F.3d at 1166. Agencies must "consider ten factors when assessing intensity, including the proposed action's effects on public health, the unique characteristics of the geographic area, the uncertainty of potential effects, and the degree of controversy surrounding the effects on the human environment." Id. (citing 40 C.F.R. § 1508.27(b)) (emphasis added).

If the agency determines on the basis of the EA not to prepare an EIS, it must prepare a Finding of No Significant Impact. 40 C.F.R. § 1501.4(e). "An agency's decision to issue a FONSI and not prepare an EIS is a factual determination which implicates agency expertise." Utah Shared Access Alliance, 288 F.3d at 1213 (quoting Comm. to Pres. Boomer Lake Park v. Dep't of Transp., 4 F.3d 1543, 1555 (10th Cir. 1993)). The court's review of an agency's decision to issue a FONSI and not prepare an EIS requires it to determine "whether the agency acted arbitrarily and capriciously in concluding that the proposed action will not have a significant effect on the human environment." Greater Yellowstone Coal. v. Flowers, 359 F.3d 1257, 1274 (10th Cir. 2004) (quoting Davis, 302 F.3d at 1112 (internal quotation marks omitted)).

Conservation Groups argue that Reclamation's decision not to prepare an EIS in light of agency comments demonstrating a controversy violated NEPA. They contend the record establishes that NPS and FWS, along with the Utah Division of Wildlife Resources (UDWR) and the Ute Indian Tribe, expressed significant concerns about the assumptions and analysis of the draft EA that went directly to the extent and significance of the contract's impacts. App. 285, 299-302. Moreover, they say, NPS, FWS, and UDWR questioned and criticized the draft EA's reliance on past hydrological data to estimate the combined impacts of future drought on the Colorado and Green Rivers. Id. at 299-302. NPS also raised concerns about Reclamation's failure to analyze impacts to Reach 3. Id.

Because we have previously addressed these concerns in detail above, we will not belabor the point again here. Suffice it to say, Reclamation's responses to these concerns were reasoned and supported by the record. One argument requires further explanation. Conservation Groups contend that the contract's nature and effect were highly controversial within the meaning of 40 C.F.R. § 1508.27(b)(4)¹⁵ and that Reclamation's dismissal of these concerns meant the final EA was arbitrary and capricious in violation of the APA.

15 That provision states: "The degree to which the effects on the quality of the human environment are likely to be highly controversial." 40 C.F.R. § 1508.27(b)(4).

"Controversy" in the NEPA context "does not necessarily denote public opposition to a proposed action, but a substantial dispute as to the size, nature, or effect of the action." Middle Rio Grande Conservancy Dist. v. Norton, 294 F.3d 1220, 1229 (10th Cir. 2002). The presence of controversy is also not dispositive in whether an EIS is required, but is merely one of ten factors "to be weighed in deciding what documents to prepare." Hillsdale Envtl. Loss Prevention, 702 F.3d at 1181 (quoting Town of Marshfield v. FAA, 552 F.3d 1, 5 (1st Cir. 2008)). "The relevant analysis is the degree to which the proposed action affects this interest, not the fact it is affected." Id. at 1180. The decision not to prepare an EIS is only improper "[i]f the [appellants] can demonstrate substantively that the [agency's] conclusion of non-significant effect on the environment represents a 'clear error of judgment.' " WildEarth Guardians v. U.S. Fish & Wildlife Serv., 784 F.3d 677, 691 (10th Cir. 2015) (internal quotation omitted).

Reclamation responds that Conservation Groups' contention that the agency ignored relevant comments—thereby triggering the highly controversial factor—is unsupported by the record. Conservation Groups argued in their brief that Reclamation "ignored concerns expressed by the Ute Indian Tribe that the 'exchange' and the NEPA analysis failed to address potentially damaging impacts to the Tribe's reserved water rights." Aplt. Br. at 47. Reclamation, however, adequately responded to the Tribe's comments on the draft EA. Supp. App. 276. The Tribe, amicus here, disagrees with Reclamation's response and claims Reclamation violated NEPA in analyzing the impact of the exchange contract on the Tribe's water rights. Amicus Br. at 10-12, 16-17, 21-22. The nature of this dispute may well be controversial, but two considerations militate against this factor necessitating that Reclamation should have conducted an EIS.

First, the mere presence of a controversy is not dispositive in requiring an EIS. The relevant analysis is the degree to which the exchange contract affects the highly controversial factor. Hillsdale Envtl. Loss Prevention, 702 F.3d at 1180. We agree that the exchange contract implicates a controversy over water rights between Utah and the Ute Indian Tribe. The degree to which it does that, however, falls below what the record demonstrates as a material controversy for the purposes of this agency action. Reclamation responded to the Tribe's concerns over government-to-government consultation, the issue of seniority as to water rights in Flaming Gorge Reservoir, and that the EA—in the Tribe's view—failed to recognize the Tribe's reserved water rights in the Green River. Supp. App. 276. The Tribe may well remain unsatisfied with Reclamation's responses, but based on the administrative record Reclamation's decision to continue with its FONSI and not move forward with an EIS was certainly not a clear error in judgment. And, in any event, the Tribe's water entitlement will be resolved if and when it exercises its senior rights.

Second, the Tribe's concerns over its water rights—held in trust by the United States for the benefit of the Tribe—are better addressed in its pending suit against the Department of the Interior in the Utah federal district court.¹⁶ Amicus Br. at 1-2. The Tribe's pending litigation, which includes its NEPA challenge to Reclamation's final EA and FONSI, also covers broader issues of the Tribe's water rights in general. Id. That is the proper forum for the Tribe's dispute with the federal government.

16 Ute Indian Tribe of the Uintah and Ouray Reservation v. United States, No. 2:21-cv-00573 (D. Utah).

Finally, Conservation Groups argue that Reclamation's admission of a "longstanding disagreement," App. 201, with Utah over the 1996 Assignment "is precisely the type of disagreement, based on outstanding questions of water availability and usage, that necessitates a full [EIS]," Aplt. Br. at 47. We disagree with Conservation Groups' reading of the "highly controversial" intensity factor. The mere fact that the contract resolved a "longstanding disagreement" between Utah and Reclamation over water rights in the Green River does not establish the antecedent proposition that there must have been a "substantial dispute as to the size, nature, or effect of the action," Middle Rio Grande Conservancy Dist., 294 F.3d at 1229, for purposes of 40 C.F.R § 1508.27(b)(4). In sum, Reclamation properly weighed the relevant factors and reached a reasoned conclusion that an EIS was not necessary in these circumstances.

Because Conservation Groups have not carried their burden in "demonstrat[ing] substantively that the agency's conclusion represents a clear error of judgment," Greater Yellowstone Coal., 359 F.3d at 1274 (internal quotation marks omitted), we conclude that Reclamation's FONSI and decision not to conduct an EIS was neither arbitrary nor capricious.

III. Conclusion

We affirm the district court's denial of Conservation Groups' claims for relief.

ROSSMAN, J., concurring in part, dissenting in part, and dissenting in the judgment

This case asks us to review whether a federal agency complied with the National Environmental Policy Act (NEPA). Several Conservation Groups contend the Bureau of Reclamation (Reclamation) did not fulfill its obligations under NEPA when it approved the Green River Block Exchange Contract (Contract), which permits the State of Utah to deplete water below the Flaming Gorge Dam.

The majority rejects all the Conservation Groups' appellate arguments, holding Reclamation complied with NEPA. I agree with most of the majority opinion, including that "Reclamation's no-action alternative used an appropriate environmental baseline to analyze the potential impacts of the contract." Maj. Op. at 1177. I also agree Reclamation took a "hard look" at the impacts of water depletions in Reach 3 and appropriately considered the cumulative impacts of the Contract. But while the majority is certain an Environmental Impact Statement (EIS) was not warranted, I remain unpersuaded on the record before us.

I diverge from my colleagues on the narrow, but important, issue of whether Reclamation satisfied its duty under NEPA to take a "hard look" at the effects of climate warming on future water availability in the Green River. As the Conservation Groups persuasively explain, Reclamation failed to address relevant scientific data—identified in public comment during the decision-making process—projecting climate warming will leave the Colorado River system far drier in the future than it has been in the last century.

The majority excuses Reclamation's deficient analysis as to future water availability by emphasizing Reclamation's assessment was only focused on "changing the point of diversion of water that Utah was already using." Id. at 1181. Even assuming Reclamation's focus was so limited, the law does not excuse agencies from NEPA's procedural requirements. Our circumscribed role under the Administrative Procedure Act (APA) is to ensure compliance with NEPA; we may not, as the majority does, perform the agency's obligations ourselves. I would instruct the district court "to remand to the agency for additional investigation [and] explanation." Fla. Power & Light Co. v. Lorion, 470 U.S. 729, 744, 105 S.Ct. 1598, 84 L.Ed.2d 643 (1985). Because I cannot join my colleagues in affirming, I respectfully dissent as to Part II.B.1 and in the disposition. I concur in all other aspects of the majority opinion.

I.

NEPA requires federal agencies "to consider every significant aspect of the environmental impact of a proposed action." Utah Shared Access All. v. U.S. Forest Serv., 288 F.3d 1205, 1207 (10th Cir. 2002) (quoting Balt. Gas & Elec. Co. v. Nat. Res. Def. Council, 462 U.S. 87, 97, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983)). The statute directs "federal agencies to prepare a detailed statement of the environmental impact for 'major Federal actions significantly affecting the quality of the human environment.' " Id. (quoting 42 U.S.C. § 4332(C)). In doing so, NEPA "ensures that an agency will inform the public that it has considered environmental concerns in its decision-making process." Id.

"NEPA itself does not mandate particular results, but simply prescribes the necessary process." Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989). As part of that process, agencies must "take a 'hard look' at the impacts of a proposed action." Wyoming v. U.S. Dep't of Agric., 661 F.3d 1209, 1263 (10th Cir. 2011) (citation omitted); see also Balt. Gas & Elec., 462 U.S. at 100, 103 S.Ct. 2246 ("Congress intended that the 'hard look' be incorporated as part of the agency's process of deciding whether to pursue a particular federal action.").

While the parameters of the "hard look" standard have not been defined with granular precision, the aim of the requirement is clear: an agency must identify and evaluate "the adverse environmental effects of the proposed action." Robertson, 490 U.S. at 350, 109 S.Ct. 1835. This includes "utilizing public comment and the best available scientific information." Colo. Env't Coal. v. Dombeck, 185 F.3d 1162, 1171 (10th Cir. 1999).

In promulgating regulations explaining how agencies must comply with NEPA, the Council on Environmental Quality (CEQ) instructs

NEPA procedures must insure that environmental information is available to public officials and citizens before decisions are made and before actions are taken. The information must be of high quality. Accurate scientific analysis, expert agency comments, and public scrutiny are essential to implementing NEPA.

40 C.F.R. § 1500.1(b) (2019).¹⁷

17 Although CEQ promulgated new regulations in 2020, we apply the earlier regulations because Reclamation's actions were all completed prior to the effective date of the new regulations, and because the agency applied the prior regulations. See 40 C.F.R. § 1506.13 (2020). As such, any citation to CEQ regulations is to those in effect before September 14, 2020.

An agency that fails to "adequately consider[ ] and disclose[ ] the environmental impact of its actions" has not satisfied the "hard look" standard. Utah Shared Access All., 288 F.3d at 1208 (citation omitted). An agency that fails to "articulate with reasonable clarity its reasons for decision," Greater Bos. Television Corp. v. F.C.C., 444 F.2d 841, 851 (D.C. Cir. 1970), or offer a "reasoned evaluation of the available information" has not fulfilled its duty to take a "hard look," Utah Shared Access All., 288 F.3d at 1213.

While NEPA prescribes what agencies must do, the APA governs how courts review agency action. See Utah Shared Access All. v. Carpenter, 463 F.3d 1125, 1134 (10th Cir. 2006) (NEPA does not provide a private right of action so "judicial review provisions of the APA govern."). The APA instructs courts to "set aside agency action, findings, and conclusions" that are "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). As the majority correctly acknowledges, our duty as an appellate court reviewing NEPA compliance is to "ascertain whether the [agency] examined the relevant data and articulated a rational connection between the facts found and the decision made." Maj. Op. at 1177 (quoting Colo. Wild, Heartwood v. U.S. Forest Serv., 435 F.3d 1204, 1220 (10th Cir. 2006)). In conducting this assessment, "the reviewing court must determine whether the agency considered all relevant factors and whether there has been a clear error of judgment." Id. (quoting Citizens' Comm. to Save Our Canyons v. Krueger, 513 F.3d 1169, 1176 (10th Cir. 2008)); see also Diné Citizens Against Ruining Our Env't v. Haaland, 59 F.4th 1016, 1034 (10th Cir. 2023) ("[W]e 'determine simply whether the challenged method had a rational basis and took into consideration the relevant factors.' " (citation omitted)).

When "the challenged decisions involve technical or scientific matters within the agency's area of expertise," our deference to the agency is "especially strong." Morris v. U.S. Nuclear Regul. Comm'n, 598 F.3d 677, 691 (10th Cir. 2010) (citation omitted). But deference is not unlimited. Defs. of Wildlife v. Babbitt, 958 F. Supp. 670, 679 (D.D.C. 1997). Deference to the agency "may be rebutted if its decisions, even though based on scientific expertise, are not reasoned." Id. "[C]ourts should not automatically defer . . . without carefully reviewing the record and satisfying themselves that the agency has made a reasoned decision . . . ." Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 378, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989). When an agency does "not provide any reasoning or analysis for its conclusion" then "there is nothing" to which the court can defer. WildEarth Guardians v. U.S. Bureau of Land Mgmt., 870 F.3d 1222, 1238 (10th Cir. 2017). "A contrary approach would not simply render judicial review generally meaningless, but would be contrary to the demand that courts ensure that agency decisions are founded on a reasoned evaluation of the relevant factors." Marsh, 490 U.S. at 378, 109 S.Ct. 1851 (internal quotation marks omitted).

When an agency does not articulate "a rational connection between the facts found and the choice made," Balt. Gas & Elec., 462 U.S. at 105, 103 S.Ct. 2246, the court "may not supply a reasoned basis for the agency's action that the agency itself has not given," Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) [hereinafter State Farm]. It is not the job of the reviewing court to "attempt itself to make up for . . . deficiencies" in the agency's explanation. State Farm, 463 U.S. at 43, 103 S.Ct. 2856. "[T]he grounds upon which the agency acted must be clearly disclosed in, and sustained by, the record" and "[a]fter-the-fact rationalization by counsel in briefs or argument will not cure noncompliance by the agency with these principles." Olenhouse v. Commodity Credit Corp., 42 F.3d 1560, 1575 (10th Cir. 1994).

Ensuring agencies offer reasoned explanations for their course of action "promotes results in the public interest by requiring the agency to focus on the values served by its decision[s]." Greater Bos. Television Corp., 444 F.2d at 852. It also enables "the public to repose confidence in the process as well as the judgments of its decision-makers." Id. As I will explain, the application of these well-settled principles compels reversal.

II.

After publishing a draft Environmental Assessment (Draft EA) in 2018 and accepting public comment, Reclamation issued a Final Environmental Assessment (Final EA) in 2019. Shortly thereafter, the agency issued a Finding of No Significant Impact (FONSI) and signed the Green River Block Exchange Contract with Utah. The Conservation Groups unsuccessfully challenged the agency's decision in federal district court, contending Reclamation failed in its mandate under NEPA to adequately analyze the Contract's environmental impacts.

At issue here, the Conservation Groups claimed Reclamation did not "address relevant scientific data and studies projecting that climate warming in the future will leave the Colorado River system far drier than it has been in the last century," Aplt. Opening Br. at 25-26, and "ignored agency and public comment" raising concerns with its hydrology data and modeling, id. at 28-30. The Conservation Groups identified three specific problems with Reclamation's Final EA.

• Reclamation failed to address three scientific studies—raised in public comment to the agency—showing warming temperatures as the cause of future river flow declines much greater than declines in the past century.

• Reclamation did not account for reduced water flows projected by its own 2012 Basin Study.

• Reclamation's use of trace 63 modeling for projecting future hydrology was erroneous because trace 63 relies only on past drought scenarios.

A. The Three Scientific Studies

U.S. Fish and Wildlife Service (FWS) pointed to three scientific studies—Udall & Overpeck 2017, McCabe et al. 2017, and Xiao et al. 2018—in a comment submitted to the Draft EA. In the comment, FWS observed "Reclamation's modeling is based on the 1906 through 2015 hydrologic record, with no consideration of hydrologic changes or trends associated with warming temperatures." Supp. App. at 271. Citing the three studies, FWS queried whether "it [is] realistic to assume that upper Colorado River basin hydrology in the future will look like that of the past, given recent research suggesting otherwise." Id.

Reclamation responded to FWS that "[a] drought response section has been added to the Technical Appendix to further address concerns regarding potential impacts from future drought scenarios." Id. Reclamation also explained "[t]he hydrologic analysis included 110 years of historic hydrology." Id. Turning to the Technical Appendix, however, there is no mention of the three studies referenced in the comment submitted by FWS. Rather, Reclamation added the following paragraph:

Concerns over a changing climate have been prominent in environmental and water resources. The DNF hydrology set contains multiple period[s] of drought, including the decades of drought that occurred in the 1930s, 1950s, 1970s and 2000 up to 2015. In order to determine the impacts of continued drought, the trace with the lowest elevation has been isolated and its results have been included. Trace 63 begins with the initial conditions and then historic year 1979 is the first hydrologic year of that trace. This trace moves through the wet years in the 1980s, but ends with the drought in 2000-2015. It is the period of operations between 2000-2015 that have the greatest impact on elevation. The impact trends of implementing the exchange agreement are seen in the worst-case scenario. The illustrations in the drought trace 63 should be considered one representation of potential possibilities of future hydrology

and it is statistically unlikely that trace 63 will happen.

App. at 211-12.

Reclamation's response did not satisfy NEPA for at least two reasons.

First, Reclamation fails to respond to the actual concern raised in the comment. The three studies cited in FWS's comment suggest a future climate scenario far warmer than what past data shows. FWS expressed concern with Reclamation's reliance on data only from the "1906 through 2015 hydrologic record" for its modeling. Supp. App. at 271. As the Conservation Groups put it, FWS was concerned with how "the data Reclamation used is all backward-looking." Aplt. Opening Br. at 30. The majority concludes Reclamation adequately "addressed the comments with a reasoned explanation justifying the use of Direct Natural Flow (DNF) hydrology sets, such as trace 63, that contain multiple periods of drought." Maj. Op. at 1180. The majority further contends that, "[w]hile Reclamation's response to FWS's comment could have been more robust, the record confirms that Reclamation adequately incorporated in its analysis the effects of a warming climate and the likelihood of changes in hydrology." Id. at 1181. I respectfully disagree.

The main purpose of asking agencies to "utiliz[e] public comment," Dombeck, 185 F.3d at 1171, is to "use public input in assessing a decision's environmental impact," California v. Block, 690 F.2d 753, 771 (9th Cir. 1982). A key goal of NEPA is to "ensure[ ] that the agency will inform the public that it has indeed considered environmental concerns in its decisionmaking process." Balt. Gas & Elec., 462 U.S. at 97, 103 S.Ct. 2246; Utah Shared Access All., 288 F.3d at 1207. Agency engagement with public comments supports this aim.

Reclamation never explains why the use of backward-looking data is appropriate under the circumstances to evaluate the adverse environmental effects of the proposed action. Absent any explanation from the agency as to how looking at past data addresses the actual concern raised by FWS about predicted future warming conditions, I fail to see how Reclamation's response can be described as the "reasoned explanation" needed to satisfy the agency's duty under NEPA.

Second, Reclamation fails to explain why the three scientific studies cited by FWS are irrelevant to assessing the environmental impacts of the project.¹⁸ It is not our job to dictate the specific evidence an agency must rely on, but "an agency must 'examine[ ] the relevant data and articulate[ ] a rational connection between the facts found and the decision made.' " New Mexico ex rel. Richardson v. Bureau of Land Mgmt., 565 F.3d 683, 713 (10th Cir. 2009) [hereinafter Richardson] (citation omitted). NEPA also directs agencies conducting an environmental review "to consider and respond to the comments of other agencies." Custer Cnty. Action Ass'n v. Garvey, 256 F.3d 1024, 1038 (10th Cir. 2001). NEPA does not require consensus among stakeholders, but "a reviewing court 'may properly be skeptical as to whether an [agency's] conclusions have a substantial basis in fact if the responsible agency has apparently ignored the conflicting views of other agencies having pertinent expertise.' " Davis v. Mineta, 302 F.3d 1104, 1123 (10th Cir. 2002) (citation omitted), abrogated on other grounds by Diné Citizens Against Ruining Our Env't v. Jewell, 839 F.3d 1276 (10th Cir. 2016).

18 Nothing in the record suggests this data is irrelevant to evaluating the environmental impacts of the Contract. Indeed, the parties, the district court, and the majority do not contend otherwise. Thus, the appropriate inquiry must be whether Reclamation provided the necessary explanation to support its decision in light of the evidence before it, including the scientific studies advanced by FWS. See Marsh, 490 U.S. at 385, 109 S.Ct. 1851 ("It is also clear that, regardless of its eventual assessment of the significance of this information, the Corps had a duty to take a hard look at the proffered evidence.").

Here, the administrative record confirms FWS—an agency with pertinent expertise in threatened and endangered animal species in the Colorado River Basin—raised a concern during the review process about warming temperatures affecting future Colorado River basin hydrology and specifically focused Reclamation's attention on the science to support it. NEPA does not command Reclamation to agree with FWS, but the agency needed to at least acknowledge the scientific data provided and offer a "reasoned evaluation of the available information." Utah Shared Access All., 288 F.3d at 1213.

The scientific studies in the record show future water availability will be fundamentally different than past water availability. As the Conservation Groups persuasively argue, Reclamation, without even acknowledging the existence of this evidence, "bas[ed] its analysis of the Contract's depletions on the unsupported assumption that future water supply will mirror past water supply." Aplt. Opening Br. at 31. Contrary to what NEPA requires, Reclamation's assessment of the environmental consequences of the Contract relies on a premise contradicted by the information before it without explanation. "Though we do not sit in judgment of the correctness of such evidence, where it points uniformly in the opposite direction from the agency's determination, we cannot defer to that determination." Richardson, 565 F.3d at 715 (emphasis omitted).

The majority acknowledges "[t]he technical appendix does not mention FWS's three studies by name nor does it explicitly state that it prefers backward-looking data to forward-looking data" but finds "that path is reasonably discernable from the record." Maj. Op. at 1180. I cannot see any path, let alone a "reasonably discernable" one. While we may "uphold a decision of less than ideal clarity," State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (citation omitted), this principle does not excuse an agency's obligation to "cogently explain why it has exercised its discretion in a given manner," id. at 48, 103 S.Ct. 2856. Here, there is no explanation for Reclamation's preference for backward-looking data over forward-looking data. "We must know what a decision means before the duty becomes ours to say whether it is right or wrong." United States v. Chicago, M., St. P. & P.R. Co., 294 U.S. 499, 511, 55 S.Ct. 462, 79 L.Ed. 1023 (1935). Reclamation's failure to acknowledge the three studies or contend with the data presented in them means the agency has not fulfilled its duty to "supply a reasoned basis" for its actions. State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (citation omitted).

The majority also says we cannot "fault Reclamation's choice of the best science, in its prerogative, to meet the chosen scale of analysis." Maj. Op. at 1182. This framing distracts from the real problem. The question is not whether Reclamation chose the best science but rather whether it explained how an evaluation of the scientific evidence in the record supports its choice. Our "concern is for elucidation of basis, not for restriction of [the agency's] latitude." Env't Def. Fund, Inc. v. EPA, 465 F.2d 528, 541 (D.C. Cir. 1972). The agency's choice must account fully for the evidence before it. See Richardson, 565 F.3d at 714 ("[T]he State asks us to ensure that BLM's conclusion was based on the requisite 'hard look' at the evidence before it.") (emphasis added). An agency's choice of the best science could fall within the realm of deference; choice without explanation violates NEPA.

B. 2012 Basin Study

The Conservation Groups next argue Reclamation's reliance on past data to predict future water availability ignores the impacts of future drought—the subject of the agency's own 2012 Colorado Basin Study. The 2012 Study was designed "to define current and future imbalances in water supply and demand in the Basin and the adjacent areas of the Basin States that receive Colorado River water over the next 50 years (through 2060)." App. at 323. In the 2012 Study, Reclamation projected "a strong continued warming throughout the Basin" and stated, "it is tenuous to assert that the past record is predictive of future conditions." Supp. App. at 316.

A comment from the Utah Rivers Council advised Reclamation "[t]he EA ignores agency-documented risks from expected water supply shortage declarations." Id. at 265. Reclamation replied:

The 2012 Basin Study report analysis specifically detailed the overall Colorado River Basin. This EA provides a specific and detailed look at the impacts of signing a water exchange contract with the State of Utah, as required under NEPA, and pur[su]ant to water rights held by the State of Utah under the 1922 Compact.

Id. Again, I fail to see how Reclamation responded to the actual concern raised.

The 2012 Study shows "that water in the Colorado River system will be scarcer due to warming temperatures than it has been in the prior century." Aplt. Opening Br. at 29; see also App. at 326. Yet, Reclamation does not deal with this information and persists in relying only on past data. Importantly, Reclamation does not explain how the geographic scope of the 2012 Study—the overall Colorado River basin—would be irrelevant to addressing the environmental impact of the Contract, which covers a smaller section of the Colorado river system within the larger Colorado River Basin. There might be a reason—counterintuitive though it seems—why the larger geographic scope of the 2012 Study makes it irrelevant to the assessment at issue here, but the agency has not provided one.¹⁹

19 Perhaps what Reclamation meant was the 2012 Study was irrelevant to the assessment of the environmental consequences of the Contract. However, explanation, not speculation, satisfies NEPA. Courts may "uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned." State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (citation omitted). But when "we are left to spell out, to argue, to choose between conflicting inferences," as here, the agency's path is not reasonably discernable. Chicago, M., 294 U.S. at 510-11, 55 S.Ct. 462. Under these circumstances, I cannot agree Reclamation complied with the "hard look" required by NEPA.

The majority contends "[c]omments attempting to expand the scope of Reclamation's EA were not germane" to "analyzing the exchange contract's impacts on a discrete part of the Green River." Maj. Op. at 1181. This observation is not especially responsive to the Conservation Groups' argument, when correctly understood. The Conservation Groups were not trying to expand the geographic scope of Reclamation's assessment. They argue Reclamation failed to address the findings from the 2012 Study when evaluating the environmental impacts of the Contract. And, what's more, Reclamation failed to "cogently explain" why the difference in geographic scope between the 2012 Study and the focus of the Final EA rendered the 2012 Study inapplicable.

C. Trace 63

Last, the Conservation Groups contend Reclamation failed to address concerns raised in public comment about future drought conditions. Along with the FWS's previously referenced comment, the National Park Service (NPS) also expressed "concern . . . regarding the hydrology modeling['s] . . . lack of evaluation under a drier scenario." Supp. App. at 279. NPS pointed to a "growing consensus among partners and among scientific studies that the future 'new normal' may be warmer and drier years on average." Id. Likewise, the Utah Division of Wildlife Resources (UDWR) stated "it is unclear how this model accounts for future climate change and lack of inflow." Id. at 266.

Reclamation responded to these comments with modeling using past drought data in the Final EA. The agency specifically referenced the trace 63 model—which measures the 2000-2015 drought and models the "worst-case scenario" for impacts of future water availability. See App. at 190. Again, the Conservation Groups argue Reclamation failed to "explain how using a model based on past drought can project impacts of future drought that Reclamation's 2012 analysis and more recent scientific studies indicate will be far more severe than those over the last century." Aplt. Opening Br. at 30. I agree with the Conservation Groups.

To comply with NEPA and demonstrate a "rational connection between the facts found and the choice made," State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (citation omitted), Reclamation needed to explain why reliance on only historic data was justified under the circumstances. Instead, Reclamation avoided the issue by using historical data as a proxy for future hydrology without explanation.

The majority says trace 63 "was an appropriate modeling data set, or surrogate, to capture any unanticipated effects of water shortages and climate-change concerns." Maj. Op. at 1180. Respectfully, the record suggests otherwise. The studies in the record indicate future water availability will look nothing like past water availability. Yet trace 63 uses only historic data. As the Conservation Groups correctly observe, it is not clear on this record how "modeling based on past drought cycles alone" could offer "a meaningful 'surrogate' to predict future water availability." Reply Br. at 13.

The majority acknowledges our duty is to conduct "a 'thorough, probing, in-depth review' of the administrative record." Maj. Op. at 1178 (quoting Ron Peterson Firearms, LLC v. Jones, 760 F.3d 1147, 1162 (10th Cir. 2014)). We do not "depart from [our] proper function when [we] undertake[ ] a study of the record, hopefully perceptive, even as to evidence on technical and specialized matters" because doing so "enables [us] to penetrate to the underlying decisions of the agency, to satisfy [ourselves] that the agency has exercised a reasoned discretion . . . ." Greater Bos. Television Corp., 444 F.2d at 850. Thus, absent any explanation from Reclamation as to how past data can serve as a surrogate for future drought scenarios, the record here does not permit the conclusion drawn by the majority about trace 63. When the judgment of an agency appears unsupported in the record and is unaccompanied by a rational explanation, "we cannot defer to that determination." Richardson, 565 F.3d at 715.

III.

Review of agency action under the APA's arbitrary and capricious standard is "highly deferential," but "our inquiry must 'be searching and careful.' " Ecology Ctr., Inc. v. U.S. Forest Serv., 451 F.3d 1183, 1188 (10th Cir. 2006) (citation omitted). The majority correctly articulates the standard of review but fails to heed a critical limitation: we do not consider "[a]fter-the-fact rationalization by counsel in briefs or argument" in lieu of reasoned decision-making by the agency during the decision-making process. Olenhouse, 42 F.3d at 1575. "It is well-established that an agency's action must be upheld, if at all, on the basis articulated by the agency itself." State Farm, 463 U.S. at 50, 103 S.Ct. 2856. And if an agency's reasoning is inadequate, "[t]he reviewing court should not attempt itself to make up for such deficiencies." Id. at 43, 103 S.Ct. 2856.

Rather than abide these limits, the majority endorses Reclamation's evaluation of the environmental impacts of the Contract by impermissibly accepting as proxy arguments made by the agency in litigation. The majority also offers its own rationale to compensate for Reclamation's explanatory shortcomings. Courts may not satisfy the obligations the agency left unfulfilled during the decision-making process.

In addressing the three scientific studies identified in the FWS's comment, the majority says it is "satisfied with Reclamation's rationale that those studies' geographic scales . . . were not appropriate to evaluate the more limited hydrologic impact of the exchange contract." Maj. Op. at 1182. Nowhere in the Final EA does Reclamation even mention the three studies, much less provide the rationale now supplied by the majority to explain why they could be disregarded.

Reclamation's response to FWS's comment similarly makes no mention of the geographic scope of the studies, nor does the added paragraph addressing "concerns over a changing climate" reference a difference in geographic scale. As far as I can tell, the only source for the majority's invocation of geographic scale, as it applies to the three studies, is Reclamation's appellate brief, where the agency argues "the cited additional studies analyze projections of future water availability at a much larger geographic scale . . . than Reclamation determined is scientifically appropriate . . . ." Aplee. Br. at 24-25.

The majority further contends "Reclamation's focus on the relevant geographic scale for the analysis of the exchange contract was also reasonably discernable in its decision to use trace 63 and its own models and simulations." Maj. Op. at 1181. Though the majority acknowledges "Reclamation only explicitly made [the geographic scope] argument in response to a comment" about the 2012 Study, my colleagues conclude NEPA is satisfied because "the refrain of geographic scale resonates throughout Reclamation's record." Id. at 1181. This is insufficient. The "hard look" standard is not unyielding but demands more than what courts can glean from the record's ambiance.

"[I]t is the agency's responsibility, not this Court's, to explain its decision." State Farm, 463 U.S. at 57, 103 S.Ct. 2856. But the majority relies on arguments made in litigation or discerns support from a record devoid of agency explanation. Doing so violates the longstanding principle that "[w]e can only affirm agency action, if at all, on grounds articulated by the agency itself." Utahns for Better Transp. v. U.S. Dep't of Transp., 305 F.3d 1152, 1165 (10th Cir. 2002), modified on reh'g, 319 F.3d 1207 (10th Cir. 2003).

IV.

Reclamation failed to take a "hard look" at the evidence before it showing future water availability may not resemble the past. In doing so, the agency failed to assess all significant environmental effects of the Contract, as required under NEPA. Without such a complete assessment, we cannot yet determine whether an EIS is required. Therefore, I respectfully dissent from Part II.B.1 of the majority opinion and would reverse and remand to the district court with instructions to remand to the agency.