Front Range Nesting Bald Eagle Studies v. U.S. Fish & Wildlife Serv.

353 F. Supp. 3d 1115

• Court: District Court, D. Colorado
• Decided: December 13, 2018
• Docket: Civil Action No. 18-cv-0356-WJM
• Status: Published

Opinion

William J. Martinez, United States District Judge

This is a challenge to a permit issued by the United States Fish and Wildlife Service *1120("the Service") authorizing a construction company to engage in activities that may significantly disturb a pair of bald eagles that maintain a nest in the City and County of Broomfield, Colorado ("Broomfield"). Plaintiff sues under the Administrative Procedure Act ("APA"), 5 U.S.C. §§ 701 et seq. , to have the Service's actions declared unlawful under the National Environmental Policy Act ("NEPA"), 42 U.S.C. §§ 4231 et seq. ; and also under the Bald and Golden Eagle Protection Act ("Eagle Act" or "Act"), 16 U.S.C. §§ 668 et seq.

For the reasons explained below, the Court finds that the Service's action withstands scrutiny except in two respects: the failure to perform a cumulative impacts analysis, and a related failure, under the circumstances, to explain the length of a relatively short public comment period. The Court therefore vacates the permit and associated environmental analysis and remands to the Service for further consideration.

I. NEPA & APA STANDARDS

NEPA "require[s] agencies to consider environmentally significant aspects of a proposed action." Utahns for Better Transp. v. U.S. Dep't of Transp. , 305 F.3d 1152, 1162 (10th Cir. 2002). "NEPA does not, however, require agencies to elevate environmental concerns over other appropriate considerations; it requires only that the agency take a 'hard look' at the environmental consequences before taking a major action." Citizens' Comm. to Save Our Canyons v. Krueger , 513 F.3d 1169, 1178 (10th Cir. 2008) (citation and internal quotation marks omitted). Also, "NEPA dictates the process by which federal agencies must examine environmental impacts, but does not impose substantive limits on agency conduct." Utah Envtl. Cong. v. Russell , 518 F.3d 817, 821 (10th Cir. 2008). NEPA merely guards against "uninformed-rather than unwise-agency action." Robertson v. Methow Valley Citizens Council , 490 U.S. 332, 351, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989).

In conducting this analysis [under NEPA], the [agency] must prepare one of the following: (1) an environmental impact statement ['EIS'], (2) an environmental assessment ['EA'], or (3) a categorical exclusion. An environmental impact statement involves the most rigorous analysis, and is required if a proposed action will "significantly affect[ ] the quality of the human environment." 42 U.S.C. § 4332(2)(C) ; 40 C.F.R. § 1502.4.

If an agency is uncertain whether the proposed action will significantly affect the environment, it may prepare a considerably less detailed environmental assessment. 40 C.F.R. § 1508.9. An environmental assessment provides "sufficient evidence and analysis" to determine whether a proposed project will create a significant effect on the environment. Id. If so, the agency must then develop an environmental impact statement; if not, the environmental assessment results in a "Finding of No Significant Impact," and no further agency action is required. Id.

Utah Envtl. Cong. v. Bosworth , 443 F.3d 732, 736 (10th Cir. 2006).

NEPA contains no private right of action, but is enforceable through the APA, which empowers a reviewing court to set aside agency action if it is, inter alia , "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). Generally, an agency decision will be considered 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 *1121for 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.

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). A reviewing court should engage in a "thorough, probing, in-depth review," Wyoming v. United States , 279 F.3d 1214, 1238 (10th Cir. 2002) (citation omitted), with its review of the merits "generally limited to...the administrative record," Custer Cnty. Action Assoc. v. Garvey , 256 F.3d 1024, 1027 n.1 (10th Cir. 2001).

However, "[t]he scope of review under the 'arbitrary and capricious' standard is narrow and a court is not to substitute its judgment for that of the agency." Motor Vehicle Mfrs. Ass'n , 463 U.S. at 43, 103 S.Ct. 2856 ; see also Davis v. Mineta , 302 F.3d 1104, 1111 (10th Cir. 2002) (stating that the court's review is "highly deferential"), abrogated on other grounds by Diné Citizens Against Ruining Our Env't v. Jewell , 839 F.3d 1276 (10th Cir. 2016). The Court confines its review "to ascertaining whether the agency examined the relevant data and articulated a satisfactory explanation for its decision, including a rational connection between the facts found and the decision made." Colo. Wild v. U.S. Forest Serv. , 435 F.3d 1204, 1213 (10th Cir. 2006).

II. STATUTORY BACKGROUND

The parties do not dispute that the Service's actions under the Eagle Act may be reviewed under the APA. The Eagle Act prohibits any "take" of a bald eagle without a permit. 16 U.S.C. § 668(a) - (b).¹ " '[T]ake' includes...disturb[ing eagles]." Id . § 668c. "Disturb" is further defined by regulation as follows:

to agitate or bother a [bald eagle] to a degree that causes, or is likely to cause, based on the best scientific information available, (1) injury to an eagle, (2) a decrease in its productivity, by substantially interfering with normal breeding, feeding, or sheltering behavior, or (3) nest abandonment, by substantially interfering with normal breeding, feeding, or sheltering behavior.

50 C.F.R. § 22.3.

The Eagle Act gives the Secretary of Interior power to grant permits under regulations the secretary prescribes when a taking "is necessary ... for the protection of wildlife or of agricultural or other interests in any particular locality." 16 U.S.C.§ 668a. The Secretary of the Interior delegated this authority to the Service, which has promulgated additional regulations for permit applications. See 50 C.F.R. §§ 22.1 et seq. The relevant regulation directs the Service to issue what is often known as an "incidental take" permit if the following seven criteria are met:

(1) The direct and indirect effects of the take and required mitigation, together with the cumulative effects of other permitted take and additional factors affecting the eagle populations within the eagle management unit and the local area population, are compatible with the preservation of bald eagles....

(2) The taking is necessary to protect an interest in a particular locality.

(3) The taking is associated with, but not the purpose of, the activity.

(4) The applicant has applied all appropriate and practicable avoidance and minimization measures to reduce impacts to eagles.

(5) The applicant has applied all appropriate and practicable compensatory mitigation measures, when required, *1122pursuant to paragraph (c) of this section, to compensate for remaining unavoidable impacts after all appropriate and practicable avoidance and minimization measures have been applied.

(6) Issuance of the permit will not preclude issuance of another permit necessary to protect an interest of higher priority as set forth in paragraph (e)(7) of this section.

(7) Issuance of the permit will not interfere with an ongoing civil or criminal action concerning unpermitted past eagle take at the project.

50 C.F.R. § 22.26(f) ; see also id. § 13.21(b) (stating that the Service "shall issue the appropriate permit" if regulatory criteria are satisfied).

For permits to engage in construction activities that will be visible from a bald eagle nest, the Service's National Bald Eagle Management Guidelines recommend a 660-foot setback from the nest location. (R. at 1266.)

III. BACKGROUND

A. The Garrett Application

On December 5, 2017, the Service received a completed application for incidental eagle take from an entity known as the Garrett Construction Company, LLC ("Garrett"). (Administrative Record [ECF No. 37] ("R.") at 827-28.) Garrett proposed to build a 288-unit apartment complex known as the "Caliber at Flatirons" on most of an irregularly shaped parcel in Broomfield, Colorado, bounded by the Northwest Parkway, Via Varra, Del Corso Way, the property line of an existing city park, and a BNSF rail line. (R. at 831-32.)

On the other side of the rail line is open space under a conservation easement jointly administered by Broomfield and Boulder County. (R. at 833.) Within that open space, and about 530 feet from the boundary of the proposed Caliber development, is a mature cottonwood tree where a pair of bald eagles have constructed a nest about 40 feet off the ground. (R. at 833-35, 847.) The eagles first laid eggs in the nest in 2012 and have successfully fledged at least one eaglet every year since, except in 2014 and 2017. (R. at 1200, 3321-22.)

As required by 16 U.S.C. § 668a and 50 C.F.R. § 22.26(f)(2), Garrett explained that the local interests to be protected by an incidental take permit were: its ability to build a housing complex on what it characterized as a "tight site"; the needs of future residents; and Broomfield's need for housing while experiencing rapid population growth. (R. at 832-33.) Thus, Garrett intended to undertake "construction activities which may disturb or cause take of a bald eagles," including

heavy equipment and light duty traffic, excavation, building foundation and two and four-story vertical construction of multi-family residences (288 units), a clubhouse and swimming pool along with all associated parking and infrastructure including but not limited to concrete and asphalt installation, garages, domestic water distribution, sanitary sewer, storm sewer, landscaping, park construction, regional trail and other miscellaneous construction and field activities, foot traffic, and environmental safety monitoring.

(R. at 831.) These activities would be visible from the eagles' nest because there are no line-of-sight obstructions between the nest and project site. (R. at 836.) Garrett expected these activities to last through the end of 2019. (R. at 829.)

Concerning existing activities in the vicinity that might similarly disturb the eagles, Garrett reported:

• an existing apartment complex (the "Retreat at Flatirons"), which was constructed beginning in 2012, completed in 2014, comprises 374 units, and is located just south of the proposed *1123Caliber development, about the same distance from the nest;

• the BNSF rail line, through which "at least 12 trains" pass per day;

• oil and gas activity about 1/2 mile northwest, which has been in place since 2008;

• horse grazing immediately around the nest site; and

• traffic on Northwest Parkway about 1/4 mile north of the nest. (R. at 836-38.)

To avoid or minimize disturbance of the eagles, Garrett proposed numerous measures, including:

• prohibiting vertical construction within 660 feet of the nest;

• scheduling non-vertical construction within 660 feet-a hay bale wall (discussed below), a sliver of a parking lot, and a community garden-to occur outside of the most sensitive months for eagle breeding (January through July) except when a qualified biologist determines that "the nesting attempt failed or the eaglets successfully fledged";

• monitoring of the nest by a qualified biologist at least weekly;

• fencing the project area to prevent workers from approaching the eagles, and educating workers about their duty not to disturb the eagles;

• erecting a "a hay bale sound/visual barrier" measuring twelve feet high and 400 feet long "where the 660' buffer and the Project intersect...to minimize construction sounds reaching the eagle nest";

• placing the access road and the laydown yard on the western edge of the project site, as far away from the nest as possible;

• implementing waste management best practices to avoid attracting eagles (presumably meaning to avoid attracting animals that the eagles might hunt); and

• requiring construction vehicles to drive slowly, including around nearby but non-adjacent prairie dog colonies where the eagles can hunt.

(R. at 835, 839-41, 848-49.)

B. The Original Permit

On February 6, 2018, the Service issued an incidental take permit to Garrett that would expire on December 31, 2020. (R. at 1237.) That permit authorized Garrett "to disturb up to two (2) Bald eagles...including the loss of productivity (i.e., eggs or young) due to potential abandonment of the eagle nest during construction activities." (Id. )

Of the many limitation and minimization proposals in Garrett's application, the only proposal the Service adopted without change was the hay bale wall. (R. at 1238.) The Service also required monitoring, but instead of Garrett's at-least-weekly proposal, the Service required monitoring "approximately one time per month for a minimum of 1 hour of observation time. Monitoring visits may need to be extended beyond the 1 hour minimum to determine current status of the nest." (Id. ) There was no explicit 660-foot buffer requirement, although the permit said that the hay bale wall must be constructed "at the eastern edge of the project where it intersects the 660 foot buffer zone around the nest tree." (Id. )

C. This Lawsuit

Plaintiff filed this lawsuit on February 13, 2018. (ECF No. 1.) On March 13, 2018, Plaintiff filed an opening merits brief asserting, among other things, that the February 6 permit should be vacated because the Service had never prepared either an EA or EIS in connection with the decision *1124to approve Garrett's application. (ECF No. 14 at 23-35.)²

Before filing a response brief, the Service moved for voluntary remand, claiming that "[n]ew information has arisen which requires the agency to reconsider the permit's terms and possibly revisit some of the analyses conducted by the agency." (ECF No. 24 at 1.) Plaintiff did not oppose voluntary remand in principle, but asked the Court to impose certain conditions. (ECF No. 28 at 3.) While the parties litigated that issue, Plaintiff submitted to the Service what Plaintiff describes as "extensive comments raising numerous important issues," as did Dr. Al Manville, "the Service's own chief biologist in the agency's Migratory Bird Division for nearly two decades before his retirement." (ECF No. 41 at 22.)

On April 18, 2018, the Service announced its decision "to undertake additional analyses regarding the Garrett permit, including conducting an Environmental Assessment ('EA') of potential impacts of issuing a bald eagle incidental take permit to Garrett." (ECF No. 31 at 2.) The parties therefore jointly requested a stay of proceedings (id. at 3), which the Court granted (ECF No. 32).

D. The EA

On April 23, 2018, the Service issued a draft EA (R. at 1879) and announced a one-week public comment period running through April 29, 2018 (R. at 1907-08). Plaintiff, Dr. Manville, and others responded, submitting "over 440 comments" in all. (R. at 2061-2546, 3316.) In the days after the comment period closed, some potential commenters requested an extension of time because they received notice of the draft EA with insufficient time to prepare a response. (R. at 3299-3300, 3304-05, 3307-08.) The Service did not act on those requests.

On May 18, 2018, the Service issued its final EA. (R. at 3309.) The Service described Garrett's proposed activities and restrictions, and four potential alternatives to granting the permit on Garrett's proposed terms. (R. at 3316-20.)

The first alternative was to deny the permit in full. (R. at 3318.) The Service cited 50 C.F.R. § 13.21(b) for the notion that it must issue a permit if regulatory criteria are satisfied, and then announced that the relevant regulatory criteria from 50 C.F.R. § 22.26(f) were satisfied, meaning complete denial was ruled out. (Id. ; see also Part II, above (quoting 50 C.F.R. §§ 13.21(b) and 22.26(f) ).)

As for the second, third, and fourth alternatives, the Service described them as "considered" but "eliminated from further review" because "[t]hey were either not consistent with the Eagle Act and its permitting regulations, were impracticable for the applicant to carry out, or both." (R. at 3318-19.)

The second alternative was to grant the permit but require a 1/4 mile or 1/2 mile buffer, instead of a 660-foot buffer. (R. at 3319.) The 1/4 mile option was in response to a comment received during the public comment period. (R. at 3386-87.) The 1/2 mile option arose from a comment submitted by the Colorado Division of Parks & Wildlife ("CPW") noting that agency's "advisory recommendations" of a 1/4 mile buffer at all times and a 1/2 mile buffer from October 15 through July 31 of each year. (R. at 2539, 3388-89.) The Service explained that it rejected this alternative because it "negotiated with Garrett to try and implement the larger recommended *1125buffer" but "Garrett indicated that the Project was not feasible if they were required to implement larger nest buffers ... than the 660 foot buffer in the Service's guidance for bald eagles." (R. at 3319.) In its response to comments received, the Service elaborated that "a 1/2 mile nest buffer would encompass the entire Project area and a 1/4 mile nest buffer would eliminate about 75% of the Project. In either case Garrett could not implement either size nest buffer and still have a viable Project." (R. at 3387.)

The third alternative was to grant the permit but "require that the Garrett acquire permanent conservation easement lands that would offset the effects to the acres impacted by development of the Project." (R. at 3319.) But the Service deemed this a "mitigation" measure that was unnecessary under Eagle Act regulations given the expected amount of "take" that could result from Garrett's construction activities. (R. at 3319-20.)

The fourth alternative was to grant the permit only for the 2019 nesting season. (R. at 3320.) This alternative is informed by 50 C.F.R. § 22.26(e)(1), which states that the Service's first step in deciding whether to issue a permit is to evaluate "[w]hether take is likely to occur based on the magnitude and nature of the impacts of the activity." If not, the Service may deny a permit as unnecessary. (See also R. at 3318.) But, with respect to the fourth alternative, the Service "determined that there was still some risk of disturbance take...to the bald eagle pair during the remainder of the [already underway] 2018 nesting season." (R. at 3320; see also R. at 3314-15.) The Service thus eliminated this alternative.

The Service then drew on a 2016 programmatic EIS ("PEIS") that established standards for evaluating the potential population impacts of an incidental eagle take permit. (R. at 3323.) That PEIS divides the country into four "flyways" or "Eagle Management Units" ("EMUs") that, very roughly speaking, track the four time zones covering the lower forty-eight states. (R. at 3331.) Garrett's proposed project falls within what the Service has dubbed the "Central" flyway or EMU. (Id. ; R. at 3324.) The Service searched for active eagle take permits within the EMU and learned that there were currently eight, all of which were disturbance take permits (as opposed to permits to capture or kill eagles). (Id. ) According to the PEIS, each disturbance take permit is estimated to result in an "annual loss of productivity" of "1.33 bald eagles per nest per year," and the annual maximum permissible disturbance take is 70 bald eagles per year. (Id. ) Therefore, existing permits created an expectation of 10.64 takes per year, leaving significant room for additional permits. (Id. )

The Service then turned to the potential effect on the Local Area Population ("LAP"). The LAP is the number of bald eagles within 86 miles of the project for which a permit is sought. (R. at 3325.)³ The Service estimated that 65.66 bald eagles live within the relevant LAP. (Id. ) The Service, through the PEIS, set a 5% threshold of annual take within any given LAP, above which "a harder look at the specific circumstances" is required. (R. at 3324-25.) The two existing take permits within the LAP estimated annual take of 2.6 eagles, so granting the Garrett permit would push that to 3.93 eagles, or 6% per year. (R. at 3325.) The Service thus moved to a "harder look" analysis. (Id. ) The most important data in this phase of the analysis *1126were data showing that the relevant LAP is experiencing higher than average population growth and could withstand up to an 11% annual take rate, so 6% was unproblematic. (R. at 3326.)

Following all of this analysis, the Service returned to the alternative of denying the permit. The Service repeated that, in light of its analysis, Garrett's application satisfied the relevant regulatory criteria. (R. at 3327.) "Hence, the Service should not deny an eagle take permit to Garrett." (Id. )

E. The Current Permit

Consistent and concurrent with the EA, the Service issued an incidental take permit that expires on September 30, 2021 ("Permit"). (R. at 3382.)⁴ As before, the Permit authorized Garrett "to disturb up to two (2) Bald eagles...including the loss of productivity (i.e., eggs or young) due to potential abandonment of the eagle nest during construction activities." (Id. ) But in contrast to the previously issued permit, this Permit included almost all of Garrett's proposed limitations and conditions, materially unchanged. (R. at 3383.) The only substantial difference was, again, requiring monitoring only once per month, rather than at least weekly. (Id. )⁵

Plaintiff then filed a supplemental petition for review, challenging the EA and the Permit. (ECF No. 34-1.) That is the matter currently before the Court.

IV. STANDING

Plaintiff is an "organization dedicated to the study and conservation of nesting bald eagles, golden eagles, and other raptors in the Front Range region of Colorado." (ECF No. 1 ¶ 5.) Plaintiff's members have regularly observed, and plan to continue regular observations of, the eagles at issue in this lawsuit. (Id. ¶ 6.) The Service does not challenge the adequacy of Plaintiff's standing in this regard, nor does the Court see any standing defect in these claims.

However, Plaintiff's opening brief attaches a declaration from a member of its organization, who reports (as of June 22, 2018) that "preliminary stages of project construction have commenced." (ECF No. 41-1 ¶ 3.) The Service's response brief similarly announces the Service's "understanding" that Garrett began constructing the Caliber project soon after receiving the Permit. (ECF No. 45 at 16.) In this light, the Court issued a jurisdictional order to show cause addressing two subjects (ECF No. 54), which the Court will address in turn.

*1127A. Mootness

The first issue is whether the eagles continue to occupy the nest or whether they are reasonably likely to return-in other words, whether the case has become moot. As to that, Plaintiff responds with another declaration from one of its members, who reports that the heaviest construction work closest to the eagle nest took place in late July and early August 2018, and that the eagles continue to be spotted near the nest. (ECF No. 55-1 ¶ 3.) This member emphasizes, however, that the eagles are seen less frequently, and that they have begun a new nest in the same tree, but on the side of it furthest from construction activities. (Id. ) The Service responds by pointing to reports from biologists under contract with Broomfield to observe the eagles. (ECF No. 56 at 6 (citing an online repository of these reports at https://www.broomfield.org/2741/Nesting-Bald-Eagles).) The service characterizes these reports as showing "that during times of heavy construction (July-August) the eagles were present at the nest and behaving normally." (Id. ) Plaintiff disputes this characterization. (ECF No. 60 at 3.)

The Court need not resolve the parties' factual dispute over the effect of ongoing construction on the eagles.⁶ Whatever effect construction maybe having, it is clear from both sides' submissions that the eagles have not abandoned the nest tree. As a result, the case is not moot.

B. Redressability

The second issue raised in the Court's order to show cause is whether the Court can issue an order likely to redress the Plaintiff's claims. See Lujan v. Defenders of Wildlife , 504 U.S. 555, 561, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992) (federal court's ability to enter an order redressing claimed injury is essential to Article III standing). This potential stumbling block arises from the Service's account in prior rulemaking that many developers have been content to take their chances rather than seek an incidental take permit. See 81 Fed. Reg. 91494, 91495-96 (Dec. 16, 2016). Or in other words, the Eagle Act does not mandate that parties seek permits before any action that might incidentally take eagles, but simply gives the Service power to seek penalties against those parties that actually take eagles. The question, then, is whether Garrett would do anything differently if the Court vacated the Permit. Cf. Zeppelin v. Fed. Highway Admin. , 305 F.Supp.3d 1189, 1198- 99 (D. Colo. 2018) (when a plaintiff asserts procedural injury such as failure to follow NEPA, Article III standing fails unless there is a reasonable chance that a court order requiring more procedure would lead to the desired substantive outcome); Zeppelin v. Fed. Highway Admin. , 293 F.Supp.3d 1267, 1285-88 (D. Colo. 2017) (plaintiffs lacked Article III standing because the actions of a third party would lead to the undesirable substantive outcome regardless of any court order vacating and remanding agency action).

Having reviewed the parties' positions on this issue, the Court is convinced that Plaintiff has properly pleaded redressability. Assuming the Court vacates the Permit, and further assuming that Garrett does not stop construction (i.e. , it is willing to take its chances, under the circumstances), Garrett would still have an active permit application in front of the Service. This means the Service might still impose conditions on the Caliber project (potentially including longer-term conditions, as discussed below in Part V.D.2) that are reasonably likely to protect Plaintiff's interest in observing the eagles to a greater degree than the Permit currently does, *1128even if only incrementally more. See Consumer Data Indus. Ass'n v. King , 678 F.3d 898, 903 (10th Cir. 2012) (redressability component of standing does not require "complete redressability," but only that "a favorable decision would relieve [the plaintiff's] problem to some extent" (internal quotation marks omitted)).

In this light, the only true factual question is whether Garrett would be likely to withdraw its permit application. There is no evidence in the record to support such a likelihood. Garrett has already completed and filed the application and has explicitly stated that "avoiding and minimizing disturbance to the eagles is a priority." (R. at 839.) The Court accordingly finds that Plaintiff satisfies the redressability component of Article III standing. In consequence, the Court discharges its jurisdictional order to show cause.

V. ANALYSIS

Plaintiff brings several NEPA challenges, and further challenges the Service's findings as to two of the seven regulatory criteria in 50 C.F.R. § 22.26(f). The Court will address all of these arguments in the order presented by Plaintiff.

A. The Seven-Day Public Comment Period

Plaintiff initially argues that a seven-day public comment period was arbitrarily and capriciously short under the circumstances. (ECF No. 41 at 26-29.) The Service defends the length of the comment period but also asserts that Plaintiff has failed to demonstrate any prejudicial error (or in other words, the Service's error, if any, was harmless) because Plaintiff itself submitted extensive comments despite the short timeframe. (ECF No. 45 at 32.) See also 5 U.S.C. § 706 (in performing APA review of agency action, "due account shall be taken of the rule of prejudicial error"); Prairie Band Pottawatomie Nation v. Fed. Highway Admin. , 684 F.3d 1002, 1008 (10th Cir. 2012) ("even if an agency violates the APA, its error does not require reversal unless a plaintiff demonstrates prejudice resulting from the error").⁷

The Service cites Gerber v. Norton , 294 F.3d 173, 182 (D.C. Cir. 2002) ; Miami-Dade County v. EPA , 529 F.3d 1049, 1061 (11th Cir. 2008) ; and Conservation Law Foundation v. Evans , 360 F.3d 21, 30 (1st Cir. 2004), for the proposition that prejudicial error in this context requires a challenger to show specific prejudice to its own interests, such as a comment it would have made but for the agency's alleged error. (See ECF No. 45 at 32.) But these cases were largely about supplemental agency material that the challengers did not have an opportunity to comment on, or a claim that a formal public comment window *1129might have led to a different outcome than the public involvement in which the challenger actually participated and through which it actually got its concerns before the agency. These cases did not address the argument of a comment window so short that certain potential commenters could not timely submit their concerns.

It is true that Plaintiff submitted extensive comments during the one-week timeframe (R. at 2550-3289), and also sent information and comments to the Service before it had announced any public comment period (R. at 367-407, 413, 442-45, 503, 571, 573, 600-04, 1493). However, the possibility remains that the Service might have made a decision more protective of eagles, if only slightly, had the Service permitted more time for others to comment. The Court need not decide whether this alone, with no other basis to remand, would overcome the harmless error standard because, in this case, at least one commenter seeking additional time actually raised a cumulative impacts criticism. (R. at 3299-3300.) As discussed in Part V.D.2, below, the Service's failure to analyze cumulative impacts was prejudicial error because if it had engaged in such analysis, it might have imposed conditions directed at longer-term avoidance of eagle disturbance. Accordingly, the Court finds that Plaintiff's may demonstrate prejudicial error in these circumstances. The Court therefore turns to the question of whether the Service indeed erred with respect to the public comment period.

NEPA regulations do not require a public comment period before issuing an EA, but instead require the agency to "involve environmental agencies, applicants, and the public, to the extent practicable." 40 C.F.R. § 1501.4(b). "[T]his language affords an agency considerable discretion to decide the extent to which such public involvement is practicable." WildEarth Guardians v. U.S. Fish & Wildlife Serv. , 784 F.3d 677, 698 (10th Cir. 2015) (internal quotation marks omitted).

The Court need not decide whether a seven-day comment period was arbitrary and capricious in these circumstances, nor whether the Service has an obligation in every case to explain the length of time it chooses for a public comment period. This matter may be resolved on narrower grounds. Here, the Service's explanation of comments received fails to address the requests for extension of the comment period. Having received specific criticism about the length of the comment window, and in the absence of any argument from the Service that it was somehow excused from addressing those types of criticisms, the Court finds it arbitrary and capricious to have failed to explain the brevity of the comment period. The Court will vacate the EA and Permit on this basis, combined with the lack of cumulative impacts discussed below.

B. Reasonable Range of Alternatives

NEPA and its implementing regulations require federal agencies preparing an EA to consider "alternatives" to any proposed course of action. 42 U.S.C. § 4332(E) ; 40 C.F.R. § 1508.9(b). The EA at issue here analyzes Garrett's proposal and four alternatives. (See Part III.D, above.) The Court will refer to all five of these possibilities as "options," to avoid confusion between the generic concept of "alternatives" and those courses of action specifically labeled in the EA as "alternatives."

Plaintiff argues that, in reality, the Service analyzed only two options: granting a permit on the terms Garrett proposed, or denying it. (ECF No. 41 at 29-34.) According to Plaintiff, these were the only options that the Service evaluated "in detail." (Id. at 30, 31, 33.) From Plaintiff's perspective, apparently, the options to require a larger *1130buffer around the nest tree (second alternative), to require Garrett to obtain conservation easements as an offset (third alternative), and to issue a permit only for the 2019 eagle nesting season (fourth alternative) were discussed so cursorily as to not count.

Resolving this argument first requires understanding that, by regulation, an EA should be "a concise public document" that "include[s] brief discussions of the need for the proposal, of alternatives as required by section 102(2)(E), of the environmental impacts of the proposed action and alternatives, and a listing of agencies and persons consulted." 40 C.F.R. § 1508.9(a), (b) (emphasis added). This is in contrast to an EIS, which requires much more detail. See W. Watersheds Project v. BLM , 721 F.3d 1264, 1274 (10th Cir. 2013) ("Regulations require both documents to incorporate a range of reasonable alternatives, but the depth of discussion and analysis required is different depending on whether the document is an EIS or an EA.").⁸

The EA at issue here appears to satisfy these regulatory standards, but Plaintiff claims that the Service insufficiently considered potential options falling between what Plaintiff characterizes as "the two extremes of granting the permit as contemplated by Garrett and denying it." (ECF No. 41 at 31.) See also Colo. Envtl. Coal. v. Dombeck , 185 F.3d 1162, 1175 (10th Cir. 1999) (in reviewing a challenge to an EIS, stating that "agencies [must] take responsibility for defining the objectives of an action and then provide legitimate consideration to alternatives that fall between the obvious extremes"). It is not clear that Plaintiff has identified the relevant "extremes." The Court agrees that outright denial is one end of the spectrum, but, in the Court's view, the other end of that spectrum was the fourth alternative, which would have imposed conditions only in the 2019 nesting season, leaving all of the 2018 construction activities unpermitted (and therefore under no minimization requirements). (See R. at 3320.) So it is incorrect to say that the Service only analyzed two extremes.

Plaintiff also argues that an agency violates NEPA if it "examines several action alternatives that would each lead to excessively similar results." (ECF No. 41 at 31.) See also Citizens for Envtl. Quality v. United States , 731 F.Supp. 970, 989 (D. Colo. 1989) (in reviewing a forest management plan prepared with an EIS, stating that "[c]onsideration of alternatives which lead to similar results is not sufficient under NEPA and [the relevant regulation regarding forest management plans, which required a 'broad range of reasonable alternatives']"). If this argument refers to the five options actually considered, the Court does not see how it applies. Denying the permit is as dissimilar from granting it (on any terms) as can be. Granting it on Garrett's terms is distinctly dissimilar from granting it with a 1/4 mile or 1/2 mile buffer (second alternative), or from requiring conservation easements as an offset (third alternative), or granting the permit only for a later time frame (fourth alternative).

If Plaintiff means to say that the proposed action (granting Garrett's permit on Garrett's terms) and the second, third, and fourth alternatives are excessively similar because they all involve some form of granting the permit, the argument also fails. There is nothing in between denial and grant, no matter how limited the grant may be. So complete denial need only be discussed once. Every other option will necessarily address some form of grant.⁹

*1131Plaintiff further accuses the Service of "refus[ing] to consider any action alternative short of rubberstamping Garrett's permit application-any alternative that might impose at least some additional conditions designed to reduce eagle take." (ECF No. 41 at 32.) This entirely ignores the second alternative, regarding a potential 1/4 mile or 1/2 mile buffer instead of a 660-foot buffer. (R. at 3319.) To the extent Plaintiff means to say that the second alternative was never adequately considered, the Court disagrees. From a judicial review perspective, the Service would have been more helpful if it had provided a fuller discussion in the EA itself of why Garrett believed that a 1/4 mile or 1/2 mile buffer was not feasible. (See id .) But the Service's response to comments received during the public comment period adds that "a 1/2 mile nest buffer would encompass the entire Project area and a 1/4 mile nest buffer would eliminate about 75% of the Project. In either case Garrett could not implement either size nest buffer and still have a viable Project." (R. at 3387.) The truth of Garrett's representations is obvious from the maps in the record, which show just how much of the project area would be subsumed by a larger buffer. (See R. at 575, 832, 3332.) Given this, the Service's consideration of the 1/4 mile and 1/2 mile buffers was adequate in light of the record before it. See BioDiversity Conservation All. v. BLM , 608 F.3d 709, 715 (10th Cir. 2010) ("Agencies may not define a project's objectives so narrowly as to exclude all alternatives. But where a private party's proposal triggers a project, the agency may give substantial weight to the goals and objectives of that private actor." (citations and internal quotation marks omitted) ).

In short, given the "highly deferential review" required under the APA, a court "cannot set aside the agency's decision merely because the EA could have been more thorough than it was." W. Watersheds Project , 721 F.3d at 1275. The Court therefore rejects Plaintiff's argument that the Service failed to adequately consider a reasonable range of alternatives.

C. Preordained Outcome

A regulation governing EISs state that an EIS must be prepared "early enough so that it...will not be used to rationalize or justify decisions already made." 40 C.F.R. § 1502.5. This same regulation implies that the same requirement holds for EAs. Id. § 1502.5(b) (discussing *1132timing of EAs prompted by "applications to the agency"); see also Davis , 302 F.3d at 1112 (in the EA context, inquiring whether "the defendants prejudged the NEPA issues").

Plaintiff argues that the EA and Permit were a preordained outcome, in violation of NEPA, given that the Service, in Plaintiff's view, "reache[d] the exact same outcome after conducting NEPA review and receiving public input...that it previously reached without the benefit of any environmental analysis or public scrutiny." (ECF No. 41 at 34 (emphasis in original).) Plaintiff's "exact same outcome" premise is incorrect because the Permit is more restrictive than its February predecessor. (See Part III.E, above.) Regardless, Plaintiff's argument fails under the standard articulated by the Tenth Circuit in Forest Guardians v. U.S. Fish & Wildlife Service , 611 F.3d 692, 714 (10th Cir. 2010) :

A petitioner must meet a high standard to prove predetermination .... [P]redetermination occurs only when an agency irreversibly and irretrievably commits itself to a plan of action that is dependent upon the NEPA environmental analysis producing a certain outcome, before the agency has completed that environmental analysis .... [P]redetermination is different in kind from mere "subjective impartiality" ....

Cases satisfying the standard involve pre-existing contractual obligations between the agency (or its contractor) and a third party interested in the outcome of the NEPA process. See id . at 714, 718.

Plaintiff points to no contractual obligations that would satisfy this standard. In its reply brief, however, Plaintiff points to the Service's negotiations with Garrett over the 660-foot buffer. (See ECF No. 46 at 25; see also R. at 3319, 3386-87.) Plaintiff appears to equate these negotiations with the formation of a contract, but cites no record evidence to support that interpretation. Plaintiff has failed to meet the "high standard" for a NEPA violation based on a preordained outcome. Forest Guardians , 611 F.3d at 714. The Court therefore rejects this challenge.

D. Various NEPA Matters

Under a single heading, Plaintiff raises several matters that, in its view, show that the Service did not take the "hard look" that NEPA requires. (ECF No. 41 at 38-46.) The Court will discuss each matter in turn.

1. Number of Takes

Plaintiff notes that the Permit authorizes Garrett to "disturb up to two (2) Bald eagles" (R. at 3382) and assumes this means that the Service concluded that Garrett's actions "will only result in a total of two bald eagle takes" (ECF No. 41 at 38-39 (emphasis in original)). Working from this assumption, Plaintiff faults the Service because it supposedly failed to examine

the different forms of take that are likely to occur here (i.e., sub-lethal disturbance take interfering with breeding, nesting, and sheltering behaviors; lethal take of eaglets through nest failure; seasonal nest abandonment forcing the eagles to less desirable nests with lower rates of survival success). Thus, by lumping all forms of take together and failing to examine the various distinct forms of take that can occur to the same eagle before it either dies or is forced to abandon the nest for the season, the Service failed to take a hard look at the impacts of its permit under NEPA...in concluding that this project will result in no more than two eagle takes.

(Id. at 40 (emphasis in original).)

This argument starts from the wrong assumption. The Service did not authorize *1133any specific number of takes. It instead authorized Garrett to engage in a range of activities that might cause disturbance take, "including the loss of productivity (i.e., eggs or young) due to potential abandonment of the eagle nest during construction activities." (R. at 3382.) In other words, the Permit gives Garrett the right to do things that, in the end, might cause the eagles to abandon the nest, in turn leading to death of eggs or young. Again, there is no specifically authorized number of takes, in the sense of discrete acts of agitating or bothering an eagle to such a degree that it amounts to a disturbance under the Eagle Act. See 50 C.F.R. § 22.3. Nor is it clear how the Service could ever calculate such a number without estimating every swing of a hammer, every movement of a construction vehicle, and so forth. Not only would this be nearly impossible to calculate, it would also be nearly impossible to enforce without intensive round-the-clock monitoring of both the eagles and the construction site to determine which activities count against the allotted number of takes.

Plaintiff cites no authority suggesting that this is how incidental take permits are meant to work. Plaintiff also cites no authority suggesting that the Service acted arbitrarily and capriciously when it instead specified the number of eagles that Garrett was authorized to disturb, instead of the number of actions that might cause a disturbance. For these reasons, this challenge fails.

2. Cumulative Effects

NEPA regulations direct federal agencies to consider, among other things, the cumulative environmental impacts of their proposed actions. 40 C.F.R. § 1508.25(c).

Cumulative impact is the impact on the environment which results from the incremental impact of the action 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. Cumulative impacts can result from individually minor but collectively significant actions taking place over a period of time.

Id. § 1508.7.

Plaintiff argues that the Service failed to conduct any cumulative impacts analysis, which is an especially significant omission given that the outcome of the Permit will be a large new apartment complex, with all of its associated activities, reaching within 660 feet of the nest tree. (ECF No. 41 at 40-43.) The Service responds that this argument "misunderstands the scope of the proposed action" and that it only needed to analyze the construction activities themselves through the 2019 nesting season. (ECF No. 45 at 23.) "[T]he impacts of the future operation of the Garrett project, considered alone or cumulatively with the impacts of people living at the nearby Retreat [apartment complex], simply are not relevant to the agency action analyzed in the EA or authorized by the Service." (Id. )

The Service's explanation implicitly concedes error. "Regardless of whether an EA or EIS is being prepared, the agency conducting the analysis must consider the 'cumulative impacts' of the proposed action." Colo. Envtl. Coal. v. Office of Legacy Mgmt. , 819 F.Supp.2d 1193, 1212 (D. Colo. 2011). The explicit aim of a cumulative impacts analysis is to force the agency to consider the long-term effects of "reasonably foreseeable future actions" of all potential actors, including "individually minor but collectively significant actions taking place over a period of time." 40 C.F.R. § 1508.7. By confining its analysis to the effects of construction activities alone, the Service acted "not in accordance with law." 5 U.S.C. § 706(2)(A).

This error is conceivably excusable as harmless, see 5 U.S.C. § 706 ;

*1134Prairie Band Pottawatomie Nation , 684 F.3d at 1008, particularly because the EA assumes an "annual loss of productivity" of "1.33 bald eagles per nest per year" (R. at 3324) and the Permit authorizes disturbance take of two eagles up to the point of nest abandonment (R. at 3382). In other words, the Service already concluded that nest failure or abandonment are consequences that the EMU and LAP can absorb,¹⁰ and so whether that result flows from Garrett's construction activities or the operating Caliber complex, the goal of maintaining a healthy bald eagle population is not thwarted.

However, no party argues for or against harmless error in this context, and the Court does not find it harmless. In particular, proper consideration of cumulative impacts may helpfully inform, at a minimum, the sorts of minimization requirements (and, if needed, mitigation requirements) that the Service explores with or requires from the applicant. Conceivably the Service could insist upon, for example, permanent fencing and signage to prevent or deter Caliber residents from approaching the nest tree.¹¹

In short, although an EA is meant to be brief, some analysis of cumulative impacts-appropriately tailored to the presumed conciseness of an EA, the scope of proposed activities, and the Service's authority when deciding whether to grant the permit or the conditions to impose-is both necessary under NEPA and helpful in evaluating the regulatory criteria set forth in 50 C.F.R. § 22.26(f). Because the Service failed to perform a cumulative impacts analysis, the EA and Permit must be vacated.

3. The Hay Bale Wall

Plaintiff claims that the EA does not adequately analyze whether the hay bale wall would be effective, or why it was chosen over "more proven and reliable measures for reducing eagle take." (ECF No. 41 at 43.) As to the hay bale wall's effectiveness, Plaintiff is correct that the EA contains scant discussion, but Plaintiff again overlooks the Service's response to public comments. During the public comment period, one commenter urged that "a cost-benefit analysis of the proposed hay bale sound/visual barrier has yet to be completed and needs to be discussed in the Final EA." (R. at 3398.) The Service responded that it had

included the hay bale wall sound/visual barrier for the Project in the required permit conditions for the Garrett Project as a test of its effectiveness. We are not aware of any peer-reviewed scientific publications which document that use of a hay bale wall as a sound/visual barrier either reduces or eliminates the likelihood of disturbance take in relation to nesting bald eagles. We would anticipate that such a hay bale wall could reduce the amount of construction-related noise reaching the eagle nest. But as a visual screen it would likely have limited value *1135since the Stearns Lake bald eagle nest is substantially higher than the top of the hay bale wall. Hence from the standpoint of the nesting bald eagles we would expect that there would be some beneficial aspects to putting up the hay bale wall, while at the same time there are other concerns related to Project construction that this measure will not address.

(R. at 3399.)

Thus, the Service was perfectly frank about the hay bale wall's limitations and the Service's desire to test its effectiveness. Plaintiff does not argue that the Service violates any legal obligation by approving a permit with experimental minimization measures, and it is unclear what additional analysis the Service could perform in this regard before authorizing the hay bale wall.

As for choosing the hay bale wall supposedly in place of "more proven and reliable measures for reducing eagle take" (ECF No. 41 at 43), Plaintiff does not explain what measures it has in mind. Plaintiff also does not support the premise that the Service picked the hay bale wall instead of other minimization measures (as opposed to approving the hay bale wall and rejecting other minimization measures for independent reasons). To the extent Plaintiff means to refer to a buffer larger than 660 feet, the Court addresses that elsewhere in this order. (See Parts V.B & V.E.) In any event, the Court finds no error in the Service's consideration and explanation of the hay bale wall.

4. Amount of Monitoring

Plaintiff faults the EA for not explaining why the Permit imposes only monthly monitoring requirements as compared to Garrett's at-least-weekly proposal. (ECF No. 41at 44-45.) Yet again, the Service provided this explanation in its response to public comments, where it stated that

monitoring requirements for the Project...are designed to determine key outcomes for each bald eagle nesting season with regard to nest occupancy, productivity, and nest success. Hence, monitoring requirements included in any eagle incidental take permit issued for any project only need to provide a level of nest monitoring sufficient to answer these key outcomes of the eagle nesting season. Also, since an eagle incidental take permit for the Project would authorize disturbance take (i.e., the take would meet the Eagle Act's preservation standard), designing a more detailed monitoring protocol to define thresholds of disturbance, and steps to be taken when such thresholds are reached, is not needed.

(R. at 3387-88.)

Plaintiff understandably does not like this explanation because it essentially reduces to something like, "Once we decide that nest failure or abandonment is sustainable, we only need enough monitoring to confirm whether that happens." Plaintiff, of course, wants to prevent nest failure or abandonment. But, almost by definition, the fact that the Eagle Act gives the Service authorization to issue incidental take permits means that the Service may conclude, under the circumstances, that there is no reason to spare particular eagles from disturbance to the point of abandoning the locality. Given the Service's conclusion that failure or abandonment of the nest near the Caliber project would not thwart the Eagle Act's purposes,¹² its explanation regarding monitoring logically flows. The Court finds no error here.

*11365. Plaintiff's Data

During the public comment period, Plaintiff submitted data its members had collected over 29 hours of observation from October 1 through December 31, 2016, and 47 hours of observation from the same time span in 2017. (R. at 1871-72.) These data appear to show that the eagles reacted to a nearby drilling rig in operation by spending less time at an otherwise frequent perch that was relatively nearer to the rig as compared to the nest tree. (Id. ) Plaintiff argues that the Service committed error by failing to address these data, which supposedly rebut the Service's conclusion that the pair of eagles in question is relatively tolerant of human activity. (ECF No. 41 at 45.)

Plaintiff is correct that the Service does not directly analyze these data, but the Service acknowledges the existence of Plaintiff's self-generated observational data. (R. at 3391-93, 3394-95, 3401-02.) The Service found the data unhelpful when analyzing the potential effect of disturbance take as compared to EMU and LAP population estimates. (See id. ) Moreover, again, the Service evaluated Garrett's application under the assumption that disturbance take would occur, and so Plaintiff's data add nothing to that assumption. The Service did not act arbitrarily or capriciously, or contrary to law, in this respect.

E. Feasibility of a Buffer Larger than 660 Feet

Plaintiff claims that the record does not adequately support the conclusion that "a buffer larger than 660 feet would be infeasible." (ECF No. 41 at 46.) "Indeed," Plaintiff argues, "there is absolutely nothing in the record explaining why Garrett insists that a 660-foot buffer is feasible but any larger buffer, even by a single foot, renders the project infeasible...." (Id. at 46-47 (emphasis in original).) Plaintiff asserts that the Service therefore failed to justify one of the required regulatory criteria, namely, that "[t]he applicant has applied all appropriate and practicable avoidance and minimization measures to reduce impacts to eagles." 50 C.F.R. § 22.26(f)(4). The Court is not persuaded.

To begin, neither the APA, NEPA, or the Eagle Act required the Service, in these circumstances, to explain the costs, benefits, obstacles, and possibilities for every "single foot" either more or less than 660 feet. That would be an extremely onerous burden with likely little overall value to the decision-making process, and Plaintiff has cited no authority for the idea that it should be required here. See Ron Peterson Firearms, LLC v. Jones , 760 F.3d 1147, 1163 (10th Cir. 2014) ("in making policy judgments that involve line-drawing, agencies are not required to identify the optimal threshold with pinpoint precision" (internal quotation marks omitted) ).

To the extent Plaintiff believes that the record does not adequately explain why a larger buffer would be problematic, the Court has already described the documents in the record, particularly the maps, that adequately explain the Service's conclusion that Garrett was not simply being lazy by claiming infeasibility with a buffer larger than 660 feet. (See Parts III.D & V.B, above.)

Moreover, the regulatory requirement is "all appropriate and practicable avoidance and minimization measures." 50 C.F.R. § 22.26(f)(4) (emphasis added). "Appropriate" is undefined but naturally vests the Service with some amount of discretion under a normal definition of that word.¹³

*1137"Practicable" means "available and capable of being done after taking into consideration existing technology, logistics, and cost in light of a mitigation measure's beneficial value to eagles and the activity's overall purpose, scope, and scale." 50 C.F.R. § 22.3. In this light, "all appropriate and practicable avoidance and minimization measures" does not lend itself well to precise line-drawing. The Service has many factors to balance, including the financial interests of the permit applicant.

Considering all this, the Court finds that the Service's decision to impose a 660-foot buffer, as opposed to somewhere between 660 feet and, say, 1/4 mile (the next shortest radius mentioned in the record), is adequately explained and reasonable under the circumstances.¹⁴

F. Impacts Finding

Finally, Plaintiff questions the Service's conclusion under 50 C.F.R. § 22.26(f)(1) that "[t]he direct and indirect effects of the take and required mitigation, together with the cumulative effects of other permitted take and additional factors affecting the eagle populations within the eagle management unit and the local area population, are compatible with the preservation of bald eagles." This argument claims that there will likely be much more disturbance than the Permit supposedly authorized. (ECF No. 41 at 48-49.) But the argument turns entirely on the erroneous assumption that the Service underestimated the number of incidental takes that constructing the Caliber project might cause, in connection with cumulative effects. For the reasons explained above in Part V.D.1, Plaintiff misunderstands the Service's analysis on this point. This argument therefore also fails.

VI. CONCLUSION

For the reasons set forth above, the Court ORDERS as follows:

1. The Court's Order to Show Cause (ECF No. 54) is DISCHARGED;

2. The Service's May 18, 2018 final environmental assessment (R. at 3309) and May 18, 2018 incidental take permit issued to Garrett (R. at 3382) are VACATED;

3. This matter is REMANDED to the Service for further consideration of Garrett's application in light of the deficiencies described above, and consistent with this Order; and

4. The Clerk shall enter judgment accordingly and shall terminate this case. The Court finds, in its discretion, that the parties shall bear their own costs.

1 Golden eagles receive the same protection, but none are at issue here.

2 All ECF page citations are to the page number in the ECF header, which does not always match the document's internal pagination, especially in documents with separately paginated prefatory material such as a table of contents.

3 Eighty-six miles is the "median natal dispersal distance of female[ ] [bald eagles]." (R. at 954.)

4 The EA appears to confine its analysis to a timeframe lasting through the 2019 nesting season. (R. at 3320.) Thus, there appears to be a discrepancy between the scope of the EA and the expiration date of the Permit. If this truly is a discrepancy, rather than a misunderstanding on the Court's part, it is another issue the Service should address on remand.

5 Plaintiff argues that the Permit was "exactly the same" as the February permit. (ECF No. 41 at 23.) When challenged by the Service on this somewhat surprising claim (ECF No. 45 at 15-16), Plaintiff responded (see ECF No. 46 at 8) by pointing to language from the previous permit stating that it would only protect Garrett from "take that results from activities conducted in accordance with the description contained in the permit application" (R. at 1237). This language, which also appears in the later Permit (R. at 3383), plainly refers to the "description" of "activities," meaning the construction activities disclosed as potential causes of a disturbance take. It is obviously form language meant to authorize the Service to pursue remedies if disturbance take happens due to activities not disclosed in a permit application. It cannot be interpreted as incorporating every proposal in the application because, at least in this case, it would lead to a contradiction between the monitoring requirements the Service imposed compared to those Garrett proposed. Plaintiff therefore has no basis to characterize the Permit as "exactly the same" as its predecessor.

6 No party has moved to supplement the administrative record.

7 Plaintiff cites the D.C. Circuit's recent decision in Oglala Sioux Tribe v. U.S. Nuclear Regulatory Commission , 896 F.3d 520 (D.C. Cir. 2018), for the notion that there can be no harmless error under NEPA because it is a procedural statute. (See ECF No. 46 at 18.) Plaintiff quotes the D.C. Circuit's relevant language as follows: "[n]or does NEPA-or any other statute that has been called to our attention-give [courts] authority to forgive 'harmless' violations of NEPA." (Id. (certain internal quotation marks omitted; alterations in original).) This is an inexcusable distortion of what the D.C. Circuit actually said. The words that Plaintiff replaces with "[courts]" are "the NRC," referring to the Nuclear Regulatory Commission. See 896 F.3d at 533. The question before the D.C. Circuit was whether that commission could apply harmless error analysis to internal administrative appeals . The D.C. Circuit explicitly contrasted an Article III court's duty to apply harmless error analysis under 5 U.S.C. § 706 with the lack of any similar statutory authorization for the commission. In other words, the D.C. Circuit's decision confirms that harmless error exists in the federal courts when performing administrative review. It is highly troubling that Plaintiff's counsel would represent otherwise.

8 Plaintiff does not challenge the Service's choice not to prepare an EIS.

9 There are some statements in Plaintiff's brief expressing skepticism about the Service's apparent conclusion in the EA that it lacked discretion to deny Garrett's permit altogether. (See ECF No. 41at 30-31.) But these statements are subordinate clauses attached to other arguments and do not amount to an independent argument that the EA's analysis of the first alternative (denying the permit) was legally flawed. Arguments hinted at but not actually developed in an opening brief are deemed forfeited. See, e.g. , Thompson R2-J Sch. Dist. v. Luke P., ex rel. Jeff P. , 540 F.3d 1143, 1148 n.3 (10th Cir. 2008) ; Rojem v. Gibson , 245 F.3d 1130, 1141 n.8 (10th Cir. 2001). In any event, the Court does not understand the EA to be claiming that its only option when faced with a properly filled-out application is to grant a permit in some form. The EA is only saying that if the regulatory criteria in 50 C.F.R. § 22.26(f) are satisfied, the Service has no discretion to deny a permit for other reasons. This is untroubling given that the regulatory criteria require the Service to ensure, among other things, that the direct and indirect effects of the proposed take along with the cumulative effects of existing take permits would still be "compatible with the preservation of bald eagles," id. § 22.26(f)(1), which is the basic purpose of the Eagle Act. Plaintiff may be confused because the EA seems to conclude that all the regulatory criteria are satisfied before it has actually discussed them. (See R. at 3318.) But the EA returns to the first alternative at the end of the regulatory criteria discussion (see R. at 3327), showing that its author simply chose to announce the conclusion at the outset, then to present the justification, and then to restate the conclusion. This is a drafting choice of no legal significance.

10 Nest abandonment, of course, does not decrease the current eagle population, but may slow its growth depending on whether the eagles also abandon eggs or young.

11 The Court provides this example solely to explain why the Service could conclude that cumulative future incidental take is acceptable and nonetheless be guided by projected cumulative impacts when formulating the terms of a permit-or in other words, why the error is not harmless as compared to Plaintiff's feared aesthetic, scientific, and recreational injuries. The Court does not mean to suggest that the Service, on remand, must consider this example, or that it must necessarily include conditions in any new permit aimed at minimizing cumulative effects, or that the only real defect in the Permit is its lack of such conditions. The likely cumulative effects and how to address them (or not) are for the Service to consider in the first instance.

12 Plaintiff challenges this conclusion in a separate argument, addressed in Part V.F, below.

13 The Service argues that it should receive deference to its interpretation of "appropriate" (ECF No. 45 at 41), but it never actually offers an interpretation. In any event, the Court is not convinced that broad and generic words such as "appropriate," when undefined, trigger any rule of deference to an agency's interpretation, as opposed to simply creating a fair amount of discretion regarding the decisions made under the regulation.

14 In its reply brief, Plaintiff cites Utahns for Better Transportation v. U.S. Department of Transportation , 305 F.3d 1152, 1186 (10th Cir. 2002), for the notion that the "burden is on the Applicant...to provide detailed, clear and convincing information proving impracticability." (See ECF No. 46 at 32.) In that case, however, the Tenth Circuit was referring to specific Clean Water Act regulations imposing a burden and standard of proof on the applicant for justifying a project that will fill or dredge wetlands. See 305 F.3d at 1163. Obviously that is not the legal regime applicable here, given the different governing statute. Again, it is troubling that Plaintiff's counsel feels it necessary to cite cases or record materials for propositions they obviously do not support. (See also nn.5 & 7, above.)