Hillsdale Environmental Loss Prevention, Inc. v. United States Army Corps of Engineers

702 F.3d 1156, 42 Envtl. L. Rep. (Envtl. Law Inst.) 20250, 75 ERC (BNA) 1673, 2012 U.S. App. LEXIS 24531, 2012 WL 5935949

• Court: Court of Appeals for the Tenth Circuit
• Decided: November 28, 2012
• Docket: 11-3210
• Status: Published

Opinion

TYMKOVICH, Circuit Judge.

This case concerns the construction of a new Burlington Northern Santa Fe (BNSF) rail/truck terminal outside Kansas City, Kansas. Because the preferred site contained streams and wetlands protected under federal law, several groups (collectively, Hillsdale) brought challenges to a dredge and fill permit issued by the United States Army Corps of Engineers (Corps) under the Clean Water Act, 33 U.S.C. §§ 1251-1387, and the National Environmental Policy Act, 42 U.S.C. §§ 4321-4370H. The district court denied Hills-dale’s motion for an injunction and granted summary judgment for the Corps and BNSF.

On appeal, Hillsdale requests we set aside the Corps’s decision to grant the permit because the Corps inadequately considered alternatives to the selected site under the Clean Water Act and violated the National Environmental Policy Act by preparing an inadequate environmental assessment and failing to prepare a full environmental impact statement. We conclude the Corps’s decision is supported by the record, and was not an arbitrary and capricious exercise of its approval powers under federal law.

I. Background

In 2007, BNSF applied for a permit from the Corps to dredge and fill waters of the United States (§ 404 permit) as part of its plan to construct a new intermodal facility in the Kansas City area. 1 BNSF operates a transcontinental railroad, the Southern Mainline, that passes through Kansas City. BNSF sought to construct the new intermodal facility near the Southern Mainline because its current Kansas City facility, the Argentine Yard, is inadequate to handle the current volume of freight shipped through Kansas City and lacks space to expand.

BNSF identified a site near Gardner, Kansas as its preferred location for the new intermodal facility. The Gardner site consists of 490 acres of primarily agricultural land, containing 28,000 linear feet of streams and nearly 8 acres of wetlands. Construction of the intermodal facility would affect only a portion of these. The unnamed and seasonally dry streams are tributaries of Big Bull Creek, which flows into Hillsdale Lake, a Corps-operated reservoir. The Gardner site is located roughly one-half mile from a residential subdivision, and two miles from Interstate 35. BNSF anticipates businesses dependent on the intermodal facility will use an adjacent 567 acres to construct a logistics center.

After receiving BNSF’s application, the Corps issued a public notice describing the application and solicited comments from the public. The Corps worked with other federal, state, and local authorities regarding the proposal, and asked the EPA to participate in its review as a cooperating agency. BNSF and its consultants provided the Corps with information relevant to this analysis.

The Corps then prepared a draft environmental assessment based on this information and comments provided by other agencies. As part of its alternatives analysis under the Clean Water Act (CWA), the Corps considered many options, including modifications to existing BNSF facilities, seven alternative sites for a new intermodal facility, alternative designs for the facility, and a no-action alternative.

Based on this review, the Corps concluded modifications to BNSF’s existing facilities were infeasible due to space limitations. The Corps also found the no-action alternative would have detrimental impacts on regional traffic and air quality because increased shipping by trucks would be necessary to handle all growth in freight shipments in the area. In its analysis of alternative sites for the intermodal facility, the Corps compared the proposed sites to criteria provided by BNSF. For example, BNSF required the site to be close to existing rail tracks and highways, large enough to handle the projected volume of freight, and within 30 miles of BNSF’s existing intermodal facility at the Argentine Yard.

After applying BNSF’s criteria, the Corps eliminated all alternatives as impracticable except for the Gardner site and a nearby location, Wellsville North. 2 Therefore, the Corps analyzed the potential environmental impacts only of the Gardner and Wellsville North sites. The Corps determined construction at Gardner would impact 17,302 linear feet of streams and 4.61 acres of wetlands, whereas construction at Wellsville North would impact 19,594 linear feet of streams and 15.83 acres of wetlands. It also determined the streams and wetlands at Wellsville North were of a higher quality than those at Gardner. Based on this comparison, the Corps concluded construction at Gardner was the least environmentally damaging practicable alternative under the CWA.

The Corps also prepared an environmental assessment to consider the impact of BNSF’s proposal on the human environment, as required by the National Environmental Policy Act (NEPA). This analysis considered both direct and reasonably foreseeable indirect impacts to land use, air quality, noise, traffic, water quality, threatened and endangered species, and cultural resources.

Its air quality analysis was based on a report developed by a BNSF consultant in conjunction with the EPA, the Kansas Department of Health and the Environment (KDHE), and the Mid-America Regional Council. This report went through multiple rounds of review and analysis to incorporate feedback from these agencies. The report modeled emissions from on-site trains, trucks, miscellaneous heavy equipment used at the intermodal facility, and off-site trucks traveling to the intermodal facility, as well as the health risks associated with these emissions. The Corps determined air quality impacts from the project would not be significant, with the possible exception of fugitive dust emissions. Because BNSF and KDHE entered a binding agreement to monitor dust emissions at the site and adopt mitigation measures should emissions exceed specified levels, the Corps concluded the intermodal facility was unlikely to have any significant impacts on air quality. The EPA and KDHE concurred with this analysis.

The Corps’s water quality analysis considered impacts to local streams and wetlands, as well as water quality. The Corps concluded construction at Gardner would impact 17,302 linear feet of streams and 4.61 acres of wetlands. But it also found BNSF would reroute 9,100 linear feet of streams, create 7.18 acres of wetlands, and restore a large, degraded wetland on the site. And the Corps found the intermodal facility was unlikely to significantly impact local groundwater. The agency concluded the overall impacts to onsite waters would not be significant.

After the Corps released its final environmental assessment and finding of no significant impact, it issued BNSF a § 404 permit. Soon after, Hillsdale filed a complaint in federal district court challenging the Corps’s environmental analysis under both NEPA and the CWA. 3 Hillsdale also moved for a preliminary injunction to halt construction of the intermodal facility, which the district court denied.

The district court then granted summary judgment to the Corps. The court upheld the Corps’s conclusion that the Gardner site was the least environmentally damaging practicable alternative under the CWA. The court also affirmed the Corps’s NEPA analysis of air and water impacts, concluding the Corps’s reliance on KDHE’s mitigation agreement was reasonable, its air quality assessment methodologies were also reasonable, and the project was not highly controversial. This appeal followed.

II. Discussion

We review the district court’s grant of summary judgment de novo. New Mexico ex rel. Richardson v. Bureau of Land Management, 565 F.3d 683, 704-05 (10th Cir.2009) (New Mexico). Because suits alleging NEPA and CWA violations are brought under the Administrative Procedure Act (APA), we review the underlying agency • decision to determine whether it was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A); Colo. Wild v. U.S. Forest Serv., 435 F.3d 1204, 1213 (10th Cir.2006). An action is arbitrary and capricious if

the agency (1) entirely failed to consider an important aspect of the problem, (2) offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise, (3) failed to base its decision on consideration of the relevant factors, or (4) made a clear error of judgment.

New Mexico, 565 F.3d at 704 (internal quotation omitted).

Our inquiry under the APA must be thorough, but the standard of review is very deferential to the agency. Forest Guardians v. U.S. Fish and Wildlife Serv., 611 F.3d 692, 704 (10th Cir.2010). “A presumption of validity attaches to the agency action and the burden of proof rests with the parties who challenge such action.” Morris v. U.S. Nuclear Regulatory Comm’n, 598 F.3d 677, 691 (10th Cir.), cert. denied — U.S.-, 131 S.Ct. 602, 178 L.Ed.2d 436 (2010) (internal quotation and alteration omitted). We may set aside the agency’s decision “only for substantial procedural or substantive reasons.” Silverton Snowmobile Club v. U.S. Forest Serv., 433 F.3d 772, 780 (10th Cir.2006).

“Deficiencies in an [environmental assessment] that are mere ‘flyspecks’ and do not defeat NEPA’s goals of informed decisionmaking and informed public comment will not lead to reversal.” New Mexico, 565 F.3d at 704. “Furthermore, even if an agency violates the APA, its error does not require reversal unless a plaintiff demonstrates prejudice resulting from the error.” Prairie Band Pottawatomie Nation v. Federal Highway Admin., 684 F.3d 1002, 1008 (10th Cir.2012) (Prairie Band); 5 U.S.C. § 706(2)(F) (“[D]ue account shall be taken of the rule of prejudicial error.”).

A. Clean Water Act

The CWA prohibits dredging or filling “waters of the United States” without a permit from the Corps. 33 U.S.C. § 1344; Greater Yellowstone Coal. v. Flowers, 359 F.3d 1257, 1269 (10th Cir.2004). This permit, known as a § 404 permit, requires the Corps to review projects and ensure jurisdictional waters are not disturbed without an adequate study of alternatives. Neither party disputes that the streams and wetlands on the Gardner site are waters of the United States. Before the Corps may issue a § 404 permit, it must determine there is “no practicable alternative” to the proposed activity “which would have less adverse impact on the aquatic ecosystem.” 40 C.F.R. § 230.10(a). A practicable alternative is one that is “available and capable of being done after taking into consideration cost, existing technology, and logistics in light of overall project purposes.” Id. § 230.10(a)(2).

When a project is not water dependent, a presumption arises that there are “practicable alternatives that do not involve special aquatic sites” and “have less adverse impact on the aquatic ecosystem.” 4 Id. § 230.10(a)(3). In such cases, the § 404 applicant must rebut this presumption if an alternative involving the destruction of U.S. waters is chosen. Greater Yellowstone Coal., 359 F.3d at 1269 (citing Utahns for Better Transp. v. U.S. Dep’t of Transp., 305 F.3d 1152, 1186-87 (10th Cir.2002) (Utahns)). The intermodal facility is not water dependent, so § 230.10(a)(3)’s presumption applies in this case.

B. National Environmental Policy Act

NEPA “requires federal agencies to pause before committing resources to a project and consider the likely environmental impacts of the preferred course of action as well as reasonable alternatives.” New Mexico, 565 F.3d at 703. NEPA requires an agency to prepare an environmental assessment prior to undertaking any “major federal action” to determine whether the proposed action is likely to “significantly affect[] the quality of the human environment.” 42 U.S.C. § 4332(2)(C). The environmental assessment must discuss the need for the proposal, alternatives to the proposal, and the environmental impacts of the proposal and any alternatives. 40 C.F.R. § 1508.9(b). This includes not only direct but also indirect and cumulative impacts. Id. §§ 1508.7,1508.8.

If the agency concludes the action is unlikely to have a significant impact, it may issue a finding of no significant impact and proceed. 40 C.F.R. § 1508.13. If the agency reaches the opposite conclusion, it must prepare an environmental impact statement (EIS) to thoroughly analyze the action’s predicted environmental impacts, including direct, indirect, and cumulative impacts. 42 U.S.C. § 4332(2)(C); 40 C.F.R. pt. 1502 & §§ 1508.11,1508.25.

The significance of an impact is determined by the action’s context and its intensity. Middle Rio Grande Conservancy Dist. v. Norton, 294 F.3d 1220, 1224 (10th Cir.2002) (.Middle Rio Grande). Applicable regulations require agencies to 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. 40 C.F.R. § 1508.27(b).

Unlike the Clean Water Act, NEPA requires no substantive result. New Mexico, 565 F.3d at 704. NEPA imposes procedural, information-gathering requirements on an agency, but is silent about the course of action the agency should take. Id. “NEPA merely prohibits 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).

C. Mootness

Because we denied Hillsdale an injunction pending this appeal, BNSF has proceeded with construction of the intermodal facility. In a motion filed July 27, 2012, the Corps informed us construction of the facility is now 65% complete, 95% of jurisdictional waters have been filled or rerouted, and 95% of the associated mitigation is now complete. The Corps suggests this renders Hillsdale’s appeal moot and asks us to dismiss the appeal.

The Corps acknowledges the appeal is not constitutionally moot, as we still have the ability to afford Hillsdale at least partial relief. See WildEarth Guardians v. Pub. Serv. Co. of Colo., 690 F.3d 1174, 1182-83 (10th Cir.2012). But the Corps argues the appeal is prudentially moot because “the case is so attenuated that considerations of prudence and comity for coordinate branches of government counsel the court to stay its hand.” Rio Grande Silvery Minnow v. Bureau of Reclamation, 601 F.3d 1096, 1121 (10th Cir.2010) (internal quotation omitted). The Corps argues the intermodal facility’s construction, particularly as it affects the waters of the United States, is so far advanced that we should decline to grant Hillsdale relief. Unsurprisingly, Hillsdale disagrees. BNSF, for its part, suggests we avoid the question altogether and simply decide the appeal on the merits.

We are convinced, as the Corps acknowledges, that this appeal is not moot in the Article III sense. As for prudential mootness, it is within the court’s discretion to decline to address an issue on prudential mootness grounds. See S. Utah Wilderness Alliance v. Smith, 110 F.3d 724, 727 (10th Cir.1997). We have no trouble concluding that Hillsdale’s NEPA claims are not prudentially moot. Because the intermodal facility is not yet complete, there is a reasonable probability that if we find the Corps’s environmental assessment was defective and remand for further analysis, new information revealed by that analysis could motivate the Corps to revise its decision by, for example, requiring additional mitigation. Marsh v. Or. Natural Res. Council, 490 U.S. 360, 371, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989) (holding preparation of a postdecision EIS may sometimes be necessary, so long as an agency can still take environmentally significant action).

The Corps does not lack authority to impose additional mitigation, as it argues, because it may add conditions to permits even after they are granted when those conditions are necessary to satisfy legal requirements or protect the public interest. 33 C.F.R. §§ 325.4(a), 325.7. BNSF’s permit specifically states that the Corps may reevaluate its decision to issue the permit at any time, and such reevaluation may lead to the suspension, modification, or revocation of the permit. Accordingly, we decline to find Hillsdale’s NEPA claims prudentially moot.

Whether Hillsdale’s CWA claim is prudentially moot is a closer question. Nearly all of the jurisdictional waters on the Gardner site have been filled, and nearly all of the associated mitigation is now complete. We still have the theoretical power to afford Hillsdale relief by enjoining further construction on the Gardner site, or even ordering BNSF to restore the preexisting wetlands and streams. But the progress in the intermodal facility’s construction undoubtedly changes the balance of the equities. Nonetheless, whether this change is so drastic as to render Hills-dale’s appeal prudentially moot is a question we need not reach since Hillsdale’s CWA claim fails on the merits. See Winzler v. Toyota Motor Sales U.S.A., Inc., 681 F.3d 1208, 1210 (10th Cir.2012) (finding a claim is prudentially moot when “the anticipated benefits of a remedial decree no longer justify the trouble of deciding the case on the merits.”); see also 13B Charles A. Wright, et al., Federal Practice and Procedure § 3533.1 (3d ed. 2008).

D. Clean Water Act Alternatives Analysis

Hillsdale contends the Corps did not properly apply Clean Water Act regulations in concluding there was no practicable alternative to the Gardner site. The thrust of its argument is that the Corps did not rebut the presumption that a practicable alternative existed which did not involve the destruction of U.S. waters. 40 C.F.R. § 230.10(a)(3).

The Corps’s actions are presumptively valid under the APA, and Hillsdale bears the burden of proving the agency acted arbitrarily and capriciously. Forest Guardians, 611 F.3d at 704. But BNSF has a duty under the CWA to rebut the presumption that a less environmentally damaging practicable alternative to the Gardner site exists. 40 C.F.R. § 230.10(a)(3). Thus, if Hillsdale establishes BNSF did not rebut this presumption but the Corps issued BNSF a § 404 permit anyway, Hillsdale can establish the Corps acted arbitrarily and capriciously, “or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). We review the Corps’s practicability analysis and determination that the selected site was the least environmentally damaging practicable alternative under this standard of review.

1. Practicability Analysis

For an alternative to be selected under the Corps’s CWA regulations, it must be practicable, as defined by 40 C.F.R. § 230.10(a)(2), and it must be less environmentally damaging than all other practicable alternatives. Id. § 230.10(a)(3). Practicability is thus a threshold determination. “[A]n agency need not analyze the environmental consequences of alternatives it has in good faith rejected as too remote, speculative, or ... impractical or ineffective.” Prairie Band, 684 F.3d at 1011 (internal quotation omitted); see also Airport Neighbors Alliance, Inc. v. United States, 90 F.3d 426, 432 (10th Cir.1996). 5 The quéstion, then, is what information the Corps must provide to support its practicability determination.

Hillsdale contends the Corps and BNSF failed to provide “detailed, clear and convincing” information establishing the eliminated alternatives were not practicable. 6 Utahns for Better Transp., 305 F.3d at 1186. The Corps’s regulations state that “practicable alternatives ... are presumed to be available, unless clearly demonstrated otherwise.” 40 C.F.R. § 230.10(a)(3) (emphasis added). This does not require a specific level of detail to rebut the presumption, but only record evidence the agency took a hard look at the proposals and reached a meaningful conclusion based on the evidence. Under this understanding. of 40 C.F.R. § 230.10(a)(3)’s presumption, the Corps met this standard.

The Corps’s environmental assessment discusses BNSF’s site selection criteria in detail, and clearly indicates whether each proposed alternative met those criteria. The environmental assessment does not explain in detail the application of every criterion to every site, but it provides more explanation when necessary. For example, the environmental assessment explains that a proposed site near Olathe, Kansas, is close to an airport, and lighting at the site would conflict with airport operations.

Hillsdale does not clearly state what sort of additional information the Corps should have provided in rejecting alternatives. For several criteria, additional information or explanation would not be helpful. For example, one criterion states the site must have a minimum of 300 to 400 acres to accommodate ten 8,000-foot stripping tracks. Two sites — Wellsville South and Olathe — simply are not large enough to satisfy this criterion. Explaining in more detail why they are not large enough is unnecessary.

Similarly, BNSF mandated the site be within 30 miles of its existing intermodal facility at the Argentine Yard because many of its customers have facilities nearby. It would be too expensive for them to ship freight more than 30 miles from the new intermodal facility to their existing facilities. Again, several of the sites were simply too far from the Argentine Yard to meet this criterion, and explaining this in greater detail is also unnecessary.

Finally, the Corps’s analysis was appropriate given the minor-to-moderate antiei pated impact of BNSF’s project. The Corps’s CWA regulations instruct it to “recognize the different levels of effort that should be associated with varying degrees of impact and require or prepare commensurate documentation. The level of documentation should reflect the significance and complexity of the discharge activity.” 40 C.F.R. § 230.6(b). In addition, “[ajlthough all requirements in § 230.10 must be met, the compliance evaluation procedures will vary to reflect the seriousness of the potential for adverse impacts on the aquatic ecosystems posed by specific dredged or fill material discharge activities.” Id. § 230.10.

Our decision in Greater Yellowstone Coalition better explains this process. There, we upheld the Corps’s approval of a project that would destroy 1.45 acres of wetlands but improve 32.65 acres of wetlands and add new wetlands, even though the Corps failed to consider an obvious alternative: committing more of the developer’s property to the project so more of the development could be located away from wetlands and bald eagle habitat. 359 F.3d at 1270-71. We noted the impacts to jurisdictional waters were minimal and possibly even beneficial and held the Corps adequately supported its alternatives analysis in light of the project’s anticipated impact. Id.

Here, the impacts to U.S. waters also are projected to be minimal. Construction of the intermodal facility at Gardner will affect 4.61 acres of “low-quality” wetlands and 17,302 feet of similarly low-quality streams, which have low or intermittent flows. App. Vol. XIV at 4577. The largest seasonal stream, P-1, is now mostly dry, as the bulk of its flow came from a wastewater treatment facility that shut down in 2008. Based on these conditions, the Corps estimated the impact to onsite waters will be minor to moderate. In addition, BNSF’s proposal includes a conservation corridor that will create 7.18 new acres of wetlands, realign approximately 9,100 feet of the P-1 stream, and restore a large existing wetland. Given the minor-to-moderate anticipated impact to U.S. waters, the Corps was not required to extensively document why the eliminated alternatives were impracticable.

2. Practicable Alternatives

This leads to Hillsdale’s next argument: that the Corps had a duty under § 230.10(a)(3) to locate additional practicable sites that do not contain jurisdictional waters. They contend it is “virtually certain” a suitable site for the new intermodal facility that does not contain jurisdictional waters exists somewhere along the Southern Mainline, and the Corps violated its duty under § 230.10(a)(3) by failing to locate such a site. Aplt. Br. at 35.

To be practicable, an alternative site would have to satisfy BNSF’s site selection criteria, including the requirement that the site be located within 30 miles of the Argentine Yard. The intermodal facility is needed to serve the Kansas City area, so BNSF could not build the intermodal facility anywhere along the “hundreds of miles” of the Southern Mainline in the Midwest, as Hillsdale argues. Id. at 34.

In addition, the Corps considered seven alternative sites within or near this fairly limited geographical area. There is no magic number of alternatives the Corps must consider for its analysis to be acceptable, but the agency must draw the line somewhere, even when § 230.10(a)(3)’s presumption applies. “There will always be more data that could be gathered; agencies must have some discretion to decide when to draw the line and move forward with decisionmaking.” Habitat Educ. Ctr., Inc. v. U.S. Forest Serv., 673 F.3d 518, 531 (7th Cir.2012) (quoting Town of Winthrop v. FAA, 535 F.3d 1, 11 (1st Cir.2008)).

If Hillsdale had identified an obvious practicable alternative with no jurisdictional waters that the Corps failed to consider, we might conclude the Corps failed in its duty. But Hillsdale has not identified such a site, and even if it had, it should have brought the site to the Corps’s attention. See River Rd. Alliance, Inc. v. Corps of Eng’rs of U.S. Army, 764 F.2d 445, 452-53 (7th Cir.1985). Hillsdale instead identified two sites, Le Loup and Ottawa, which the Corps actually considered but found impracticable. 7 Thus Hillsdale has not established the Corps acted arbitrarily and capriciously by failing to find an alternative site that would not involve the destruction of jurisdictional waters.

3. Duty to Evaluate

Hillsdale’s final argument is that the Corps did not critically evaluate the accuracy and necessity of BNSF’s site selection criteria. In particular, it argues the 30-mile criterion was baseless and claims the Corps should have required BNSF to consider sites farther from the Argentine Yard.

The Corps has a duty to independently evaluate BNSF’s site selection criteria. Greater Yellowstone Coal, 359 F.3d at 1270. In Utahns, we held the Corps acted arbitrarily and capriciously by eliminating an alternative without verifying the applicant’s claim that it was prohibitively expensive. 305 F.3d at 1187. Utahns did not state what sort of critical assessment was required, merely that the Corps failed to assert independent verification of the applicant’s cost estimates and the record contained no evidence of such evaluation. Id. at 1165.

In contrast, there was evidence here that the Corps questioned BNSF’s criteria by requiring BNSF to seriously consider the Wellsville North site. In addition, the environmental assessment discusses each criterion in some detail, explaining the reasoning behind its application to the proposal. The 30-mile criterion, in particular, is supported by a BNSF study of shipping costs at a similar intermodal facility in Alliance, Texas. Based on this study, BNSF estimated it would cost its customers more than $1 million in additional shipping costs during the intermodal facility’s first year of operation if the facility were located at Wellsville North (34.4 miles from the Argentine Yard) versus Gardner (28.6 miles). Hillsdale contends this study only showed a de minimis difference in shipping costs and the Corps should have rejected or discounted it.

This argument fails for two reasons. First, the Corps’s CWA regulations stipulate “it will generally be assumed that appropriate economic evaluations have been completed, the proposal is economically viable, and is needed in the marketplace.” 33 C.F.R. § 320.4(q). The Corps is not entitled to reject an applicant’s determination that a particular type of development is economically advantageous. See Sylvester v. U.S. Army Corps of Eng’rs, 882 F.2d 407, 409 (9th Cir.1989).

Further, the Corps is entitled to accept a project applicant’s criteria based on information the applicant submits. Sierra Club v. Van Antwerp, 661 F.3d 1147 (D.C.Cir.2011). In Van Antwerp, the plaintiff challenged the Corps’s acceptance of an applicant’s claim that an 8% rate of return was necessary to secure financing for its project, and that alternatives to its preferred site offered a lower rate of re turn. Id. at 1152. Van Antwerp found a report submitted by the applicant sufficiently supported the Corps’s acceptance of the 8% figure as the minimum acceptable rate of return on the project. Id.

Like the applicant in Van Antwerp, BNSF submitted a study showing its customers’ costs increase the farther they have to transport goods from an intermodal facility to their places of business. BNSF’s study establishes its customers will experience real cost increases if BNSF builds its facility more than 30 miles from the Argentine Yard. The Corps was entitled to rely on this study. Id.; see also Sylvester, 882 F.2d at 409 (“In evaluating whether a given alternative site is practicable, the Corps may legitimately consider such facts as cost to the applicant and logistics.”).

As the Corps’s acceptance of the 30-mile criterion finds support in the record, the Corps did not act arbitrarily and capriciously by failing to take additional steps to verify BNSF’s claim that sites more than 30 miles from the Argentine Yard were infeasible. See River Road Alliance, Inc. v. Corps of Eng’rs of U.S. Army, 764 F.2d 445, 452-53 (7th Cir.1985) (“The Corps is not a business consulting firm. It is in no position to conduct ... a study that would have to ... evaluate [BNSF’s] business needs....).”

Second, the Corps’s analysis was appropriate in light of the anticipated environmental impact. The Corps in Greater Yellowstone Coalition also failed to “examine whether any commitment of Ranch property beyond the 359 acres” allotted by the developer would compromise the viability of the developer’s ranch. 8 359 F.3d at 1271. Despite a lack of “any evidentiary support for such a conclusion” in the record, Greater Yellowstone Coalition held this was not arbitrary and capricious in light of the project’s minor anticipated impact to jurisdictional waters. Id. The project here is also anticipated to have a minor impact on jurisdictional waters. And BNSF’s transportation cost study provides significantly more support for the 30-mile criterion than the applicant in Greater Yellowstone Coalition provided for the 359-acre limit. The Corps’s analysis of this criterion was not arbitrary and capricious in light of this anticipated impact. 9

Ip. Conclusion

Because the Corps did not act arbitrarily and capriciously by excluding all but two of the proposed alternative sites as impracticable, we affirm the Corps’s determination that Gardner was the least environmentally damaging practicable alternative of the two remaining sites. Both the Gardner and Wellsville North sites contain wetlands. The Corps examined both sites and determined construction at Wellsville North would impact more wetlands and streams than at Gardner. It also determined the wetlands and streams at Wells-ville North were of a higher quality than the wetlands and streams at Gardner. Hillsdale does not claim construction at Wellsville North would be less environmentally damaging than at Gardner. Thus the Corps successfully rebutted the presumption that less environmentally damaging practicable alternatives to the Gardner site existed, and its CWA analysis was not arbitrary and capricious.

E. Fugitive Dust Emissions

Pursuant to NEPA, the Corps prepared an environmental assessment but declined to prepare an EIS after finding the intermodal facility would not have a significant environmental impact. As part of its analysis, the Corps considered BNSF’s model of fugitive dust emissions at the intermodal facility, which concluded such emissions would be insignificant. Fugitive dust is dust that accumulates on hard surfaces like roads and is launched into the air by vehicle traffic and other disturbances. The EPA designed its own model of dust emissions at the intermodal facility and concluded they had the localized potential to exceed National Ambient Air Quality Standards (NAAQS) for particulate emissions. As a result of the EPA’s analysis, the Corps determined the inter-modal facility had the potential for localized significant impacts.

In response, BNSF entered a mitigation agreement with KDHE. The agreement requires BNSF to monitor dust emissions at the intermodal facility for two years after the facility opens, using a KDHEoperated sampling station. If dust concentrations exceed specified levels — levels lower than the applicable NAAQS — BNSF must work with KDHE to determine the cause of the elevated dust emissions and must take concrete steps to reduce those emissions. KDHE can require BNSF to implement one of the mitigation options listed in the agreement, such as spraying for dust suppression, or it can require BNSF to adopt any other mitigation practice it determines is appropriate. BNSF must submit a written compliance plan to KDHE, which the agency must approve. The agreement is enforceable under Kansas law. See Kan. Stat. Ann. § 65-3011. KDHE may extend the agreement if dust emissions exceed the specified levels at any time during the monitoring period. The Corps decided this agreement sufficiently mitigated the potential for significant fugitive dust emissions at the site. The EPA agreed. The Corps then issued a finding of no significant impact, concluding, based on this mitigation agreement, that the potential for fugitive dust emissions did not warrant preparation of an EIS.

The Corps can decline to prepare an EIS even if it finds a potentially significant impact so long as it also finds “changes or safeguards in the project sufficiently reduce the impact to a minimum.” Mich. Gambling Opposition v. Kempthorne, 525 F.3d 23, 29 (D.C.Cir.2008). Mitigation measures must be supported by substantial evidence of some kind. Nat’l Audubon Soc’y v. Hoffman, 132 F.3d 7, 17 (2d Cir.1997); Nat’l Parks & Conservation Ass’n v. Babbitt, 241 F.3d 722, 734-35 (9th Cir.2001) (Nat’l Parks), abrogated on other grounds by Monsanto v. Geertson Seed Farms, — U.S. -, 130 S.Ct. 2743, 2757, 177 L.Ed.2d 461 (2010). They also must, be imposed by statute or regulation, or submitted as part of the original proposal. Davis v. Mineta, 302 F.3d 1104, 1125 (10th Cir.2002). In general, agencies “should not rely on the possibility of mitigation as an excuse to avoid the EIS requirement.” Id. (internal quotation omitted).

Hillsdale claims the mitigation agreement here cannot support a finding of no significant impact. First, Hillsdale argues the record contains no studies or papers supporting the effectiveness of the mitigation options in the agreement. But “mitigation measures have been found to be sufficiently supported when based on studies conducted by the agency ... or when they are likely to be adequately policed.” Nat’l Audubon Soc’y, 132 F.3d at 17 (emphasis added); see also Greater Yellowstone Coal., 359 F.3d at 1276; Wyoming Outdoor Council v. U.S. Army Corps of Eng’rs, 351 F.Supp.2d 1232, 1252 (D.Wyo. 2005). Even in the absence of studies supporting the effectiveness of the listed mitigation measures, the Corps did not commit a clear error in judgment by basing its finding of no significant impact on this agreement because the agreement contains mandatory monitoring provisions and is enforceable under Kansas law.

Second, Hillsdale claims the monitoring period is too brief because it does not cover construction of the intermodal facility and will expire in two years, whereas business at the intermodal facility is projected to increase for at least twenty years, bringing with it the potential for increased dust emissions. The intermodal facility is projected to handle roughly twice the business in its twentieth year of operation as it will in its first, so it stands to reason fugitive dust emissions will be greater at that time. But the action levels specified in the mitigation agreement are below NAAQS levels. 10 EPA’s worst-case estimates are that dust levels will exceed NAAQS by four to ten times. If dust emissions will be a significant problem at the intermodal facility, they will likely trigger action under the agreement even before the facility is operating at peak capacity.

Even if dust levels during the monitoring period do not trigger extension of the agreement, KDHE may continue monitoring air quality near the intermodal facility after the agreement expires. To the extent Hillsdale insinuates KDHE may fail to perform such monitoring if circumstances warrant, or fail to extend the agreement if dust emissions exceed the specified levels, their argument is unpersuasive. KDHE has a legal duty under both the Clean Air Act and Kansas law to protect air quality. 42 U.S.C. § 7410; Kan. Stat. Ann. § 65-3003. We presume KDHE will perform that duty and either extend the mitigation agreement or continue independent monitoring, as necessary. See Pit River Tribe v. U.S. Forest Serv., 615 F.3d 1069, 1082 (9th Cir.2010).

Hillsdale also does not establish that the Corps’s failure to require fugitive dust monitoring during the intermodal facility’s construction was arbitrary and capricious. Hillsdale points to nothing in the record supporting its claim that fugitive dust emissions during construction will cause a significant impact. The intermodal facility’s construction likely will generate some dust, but the record states fugitive dust emissions primarily occur when “vehicles traveling on paved roads ... cause resuspension of dust accumulated on the roads.” App., Vol. XIV at 4636. The record contains no estimates of vehicle traffic generated by the intermodal facility’s construction. Given Hillsdale’s failure to support its claim that fugitive dust emissions during construction are likely to have a significant impact, this argument is unconvincing. See Morris, 598 F.3d at 691.

Finally, Hillsdale contends the agreement is insufficient because although it specifies mitigation measures BNSF must consider, it does not specify measures BNSF must adopt in the event mitigation is necessary. Hillsdale characterizes this as merely a commitment to do “something.” See Davis, 302 F.3d at 1125. But “[a] mitigation plan need not be legally enforceable, funded or even in final form to comply with NEPA’s procedural requirements.” Okanogan Highlands Alliance v. Williams, 236 F.3d 468, 473 (9th Cir.2000) (internal quotations omitted). In Greater Yellowstone Coalition, we upheld a mitigation plan that did not call for specific mitigation measures, other than requiring the applicant not to build within 400 meters of a bald eagle nest. 359 F.3d at 1276. That plan merely called for monitoring eagle activity and unspecified modifications to construction activities if the eagles were disturbed. Id.

Hillsdale tries to distinguish Greater Yellowstone Coalition by arguing there was greater uncertainty over how bald eagles would react to construction near their nests than there is over effective strategies to mitigate dust emissions. 359 F.3d at 1276. Hillsdale claims the mitigation plan here is like the one in Davis, which was insufficient to support a finding of no significant impact because it merely listed potential mitigation measures without any supporting data or “any basis for concluding they will occur.” 302 F.3d at 1125.

Hillsdale’s reliance on Davis is also misplaced. The plan in that case was insufficient because it made “no firm commitment to any noise mitigation measures.” Id. In other words, there was no “binding obligation to provide the proposed mitigation.” Id. n. 16. In addition, the environmental assessment in Davis actually rejected a number of the proposed mitigation measures as incompatible with the project’s purpose. Id. The plan here, by contrast, is enforceable. Hillsdale argues enforceability is insufficient without data on the measures’ effectiveness, but as we discussed, mandatory monitoring can provide such support.

And though the potential sources of fugitive dust emissions might be less uncertain than the reaction of bald eagles to construction near their nests, there is disagreement in the record about whether excessive dust emissions will even occur and, if they do, what their precise source will be. NEPA does not require a finalized mitigation plan so long as the proposed plan is supported or monitored. Okanogan Highlands Alliance, 236 F.3d at 473; Nat’l Audubon Soc’y, 132 F.3d at 17. Flexible mitigation plans are acceptable even when the harm they are designed to avert is more predictable than the behavior of bald eagles.

The Corps’s finding of no significant impact based on this plan was not arbitrary and capricious. If dust emissions exceed levels specified in the agreement, the plan requires BNSF to adopt mitigation measures, even if it does not specify which ones. The plan’s mandatory monitoring provisions are designed to ensure BNSF adopts effective mitigation measures if excessive emissions occur.

Thus, we affirm the Corps’s finding that there will be no significant impact from fugitive dust emissions at the intermodal facility.

F. Other Air Emissions

Hillsdale next raises a number of related claims concerning the adequacy of the Corps’s treatment of other potential air emissions attributable to the intermodal facility. Hillsdale contends the Corps failed to take a “hard look” at (1) emissions from off-site locomotives and (2) non-truck vehicles, (3) the cancer risks of diesel exhaust, and (4) emissions from increased traffic along Interstate 35 (1-35). Prairie Band, 684 F.3d at 1016.

We address each of Hillsdale’s claims in turn.

1. Off-Site Locomotive Emissions

The Corps considered the impacts of on-site locomotive emissions, but did not analyze locomotive emissions attributable to the intermodal facility outside the facility boundaries. Hillsdale claims this was arbitrary and capricious because train traffic will increase due to the intermodal facility, and the environmental assessment states locomotive emissions are a significant source of particulate matter (PM) and nitrous oxide (NOx) emissions. 11 Hillsdale contends increased emissions likely will be significant and should have been analyzed.

We disagree. Under NEPA, the Corps’s obligation is to take a “hard look” at information relevant to its factual determination. Forest Guardians, 611 F.3d at 710-11. The record shows the Corps reasonably considered off-site locomotive emissions in the background emissions levels. For its model, the Corps measured background emissions levels at Gardner and other sites and then added them to projected emissions at the intermodal facility, which included emissions from on-site locomotives. The Corps determined total emissions would not exceed applicable NAAQS. The Corps also noted large, EPA-mandated increases in locomotive efficiency over the next 20 years will result in lower locomotive emissions when the intermodal facility begins operations, even taking into account rail traffic growth. Thus it concluded current emissions were a worst-case scenario, and even these did not violate federal emissions limits.

Hillsdale contends this entire process was flawed. Hillsdale argues it “defies logic” to consider all off-site emissions part of existing conditions. Aplt. Br. at 39. But the record reflects that locomotive emissions are expected to decrease significantly over the next 20 years, even considering the expected increases in rail traffic. 12 Based on information provided by the EPA, BNSF’s consultant calculated that even if the number of locomotives moving through the intermodal facility increases by 250% between its opening and 2030, the most optimistic scenario in the record, locomotive emissions in the area will still decline. Thus it was not unreasonable for the Corps to conclude existing conditions — which include current locomotive emissions — are a worst-case scenario.

The Corps and BNSF then added modeled emissions at the intermodal facility— which included estimates of locomotive emissions — to the background emissions levels. These totals were still below the applicable NAAQS. Thus the Corps concluded locomotive emissions attributable to the intermodal facility — on- or off-site— were unlikely to have a significant impact. This approach was reasonable.

Hillsdale also claims the Corps should have quantified what proportion of future rail traffic — and emissions due to that traffic — is attributable to the intermodal facility. Hillsdale argues the agency must consider all direct and indirect impacts of its decision. But an agency must consider only “every significant aspect” of a proposed action. Baltimore Gas and Elec. Co. v. Natural Res. Def. Council, 462 U.S. 87, 97, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983) (emphasis added). Although some proportion of future rail traffic can logically be attributed to the intermodal facility, this by itself does not establish significance. NEPA regulations direct the Corps to consider both context and intensity when weighing the significance of an impact. 40 C.F.R. § 1508.27. Hillsdale does not argue locomotive emissions attributable to the intermodal facility implicate any of the intensity factors listed in § 1508.27(b). As Hillsdale does not carry its burden of demonstrating these emissions are likely to be significant, it does not establish the Corps erred by failing to quantify them.

Nor do the cases Hillsdale cites convince us the Corps erred. Hillsdale insists the Corps is attempting to excuse its analysis on the basis that increases in rail traffic will occur with or without the intermodal facility, an argument we rejected in Davis, 302 F.3d at 1123. This misstates the Corps’s position. 13 The Corps did not conclude locomotive emissions would increase anyway, but that emissions would not increase. This distinguishes this case from Ocean Advocates v. U.S. Army Corps of Eng’rs, 402 F.3d 846, 867-68 (9th Cir. 2005), where the Corps did not consider the potential environmental effects of increased shipping traffic. Here, the Corps considered those impacts and concluded they would not be significant even if rail traffic increased.

In sum, although there will likely be some impact from offsite locomotive emissions attributable to the intermodal facility, the Corps’s conclusion that this impact was unlikely to be significant, and its decision not to quantify this impact, was not arbitrary and capricious.

2. Non-Track Vehicle Emissions

The Corps estimated that in the intermodal facility’s first year of operation, it and nearby associated warehouse facilities are likely to generate an average of 2,185 one-way non-truck vehicle trips per weekday. By the fifth year of operation, that number is expected to grow to 9,211 one-way non-truck trips per weekday, primarily due to an increasing number of warehouse facilities. The Corps did not model emissions from non-truck vehicles. Instead, the agency determined automobile traffic in the Gardner area was projected to increase substantially in the next twenty years due to other developments and non-truck traffic attributable to the intermodal facility would be insignificant in comparison. The Corps also noted EPA regulations are expected to decrease automobile emissions significantly over the next twenty years. The Corps then declined to include non-truck vehicle emissions in its air pollution model.

Hillsdale contends this was arbitrary and capricious because non-truck vehicle emissions are a significant source of regulated air pollutants, including ozone precursors. Hillsdale points to several comments emphasizing that the Kansas City area, including the county where the Gardner site is located, has recently violated federal ozone standards and is in danger of being designated a nonattainment zone for ozone. Accordingly, Hillsdale argues the Corps should have considered these emissions per 40 C.F.R. § 1508.27(b)(10), which directs the Corps to consider whether a proposed action threatens a violation of federal environmental standards. 14

The Corps decided not to include these emissions in its model, concluding that, based on local land-use plans, traffic will increase substantially in the Gardner area regardless of the intermodal facility. The Corps also found new regulatory requirements will decrease vehicle emissions even as traffic increases. As with locomotives, the Corps claims its background emissions sampling adequately accounted for vehicle emissions because it sampled emissions in areas with much more traffic, including downtown Kansas City and Overland Park, Kansas. The Corps argues traffic near the intermodal facility will not exceed traffic in these areas within the next twenty years.

The Corps’s argument that new regulatory requirements will reduce non-truck vehicle emissions over the next 20 years finds little support in the record. In contrast to the detailed evidence discussing future decreases in locomotive emissions, the evidence for future decreases of non-truck emissions is lacking, amounting to little more than a statement that such decreases are likely to occur. And though traffic in the area may increase independently of the intermodal facility, the record specifies the intermodal facility will create some non-truck vehicle traffic. The environmental assessment even quantifies the estimated number of vehicle trips to and from the intermodal facility.

Despite this, the Corps’s decision to not analyze non-truck vehicle emissions was not arbitrary and capricious. Traffic attributable to the intermodal facility will be insignificant compared to overall traffic in the area. The record reflects that the intermodal facility will generate 9,211 vehicle trips per day by its fifth year of operation, estimated to be 2015. In contrast, total traffic in only one nearby community — Olathe—is expected -to be over 942,000 vehicle trips per day by 2015. The record also reflects that ozone levels are affected by regional emissions, indicating the proper point of comparison is not a single nearby community or even Johnson County, but the entire Kansas City region.

The Corps is only required to consider potential impacts relative to their significance. 40 C.F.R. § 1502.2(b). The small number of non-truck vehicle trips attributable to the intermodal facility is clearly insignificant in comparison to regional traffic. The Corps did not “fail[ ] to consider an important aspect of the problem” when it declined to model non-truck vehicle emissions. New Mexico, 565 F.3d at 704. Its decision to focus its analysis on fugitive dust emissions and emissions from trucks, heavy equipment, and locomotives — anticipated to be the primary sources of air pollution at the intermodal facility — was not arbitrary and capricious.

3. Cancer Risk Methodology

As part of its emissions analysis, the Corps considered potential cancer risks associated with a wide variety of pollutants. The Corps did not specifically analyze the cancer risks from diesel exhaust, although its cancer risk analysis considered many of the toxic components of diesel exhaust. 15 Hillsdale complains the Corps’s failure to separately analyze cancer risks from diesel exhaust, particularly diesel particulate matter (DPM), was arbitrary and capricious.

“Courts are not in a position to decide the propriety of competing method ologies ... but instead, should determine simply whether the challenged method had a rational basis and took into consideration the relevant factors.” Silverton Snowmobile Club, 433 F.3d at 782 (internal quotation omitted). This is particularly true when the dispute involves a technical judgment within the agency’s area of expertise. Envt’l Defense Fund v. U.S. Nuclear Regulatory Comm’n, 902 F.2d 785, 789 (10th Cir.1990) (internal quotation omitted).

We find the Corps adequately analyzed the cancer risks of DPM. Hillsdale claims deference to an agency’s chosen methodology is due only when the impact is agreed upon and the disagreement is over the appropriate methodology to assess its significance. Hillsdale characterizes the disagreement here as over whether DPM poses a cancer risk, not over the methodology to assess that risk.

This misstates the Corps’s position. The Corps does not dispute that DPM poses a cancer risk. It notes, as the EPA concluded, that DPM is a likely carcinogen. It merely chose a methodology Hills-dale dislikes to analyze that risk. The crux of Hillsdale’s claim, then, is that the Corps used the wrong methodology. Hillsdale contends the Corps should have employed a DPM-specific methodology, in particular one that has been adopted by the State of California, which the Corps has used in NEPA analyses of California-based projects.

The Corps’s prior use of California’s DPM methodology for California projects does not require it to use the same methodology here. An agency has discretion to choose a methodology, so long as it explains why it is reliable. Lands Council v. McNair, 629 F.3d 1070, 1078 (9th Cir. 2010). The Corps can rationally choose to use California’s DPM methodology when conducting NEPA analysis of a California project, and choose to use a different methodology here. Hillsdale is correct that the Corps’s prior DPM assessments were conducted to comply with NEPA, not California law, but they fail to explain why this requires the Corps to use this methodology for every future analysis of DPM emissions.

As for the reliability of the Corps’s chosen methodology, Hillsdale argues the methodology is likely to understate the cancer risk at the intermodal facility because the combination of toxic chemicals in DPM likely creates a cancer risk greater than the sum of the risks from individual toxic DPM components. But it points to nothing in the record supporting this assertion, merely to a list of the many toxic compounds in DPM.

The Corps’s chosen methodology is entitled to deference. Silverton Snowmobile Club, 433 F.3d at 782; Prairie Band, 684 F.3d at 1017. The Corps considered California’s methodology and determined it was inappropriate. The Corps based this decision on information provided by an expert agency, the EPA, including the EPA’s conclusion that it cannot currently establish an accurate dose-response relationship for DPM exposure and EPA’s concerns with California’s methodology. The Corps instead relied on EPA standards. The EPA signed off on the Corps’s air quality analysis.

We conclude the Corps’s methodology had a rational basis and considered the appropriate factors, including the carcinogenic components of DPM. Silverton Snowmobile Club, 433 F.3d at 782; Northwest Envtl. Advocates v. Nat’l Marine Fisheries Serv., 460 F.3d 1125, 1139 (9th Cir.2006). The Corps did not act arbitrarily and capriciously by utilizing its chosen methodology to assess the cancer risk from DPM and other air pollution attributable to the intermodal facility, rather than a competing methodology proposed by Hillsdale.

k. 1-85 Emissions

Finally, Hillsdale claims the Corps arbitrarily and capriciously failed to analyze DPM emissions from increased truck traffic on portions of 1-35 near the inter-modal facility. The Corps estimated 81% of the trucks traveling to the intermodal facility will pass by the town of Gardner on 1-35, coming within 122 feet of two residential areas. The Corps forecasts the intermodal facility will generate 3,000 diesel truck trips per day by year five, and 7,600 per day at full capacity. Based on these estimates, Hillsdale argues the inter-modal facility will subject the residential areas abutting 1-35 to elevated DPM emissions. The Corps did not analyze DPM or any other emissions along 1-35. It analyzed emissions only at the intermodal facility and along an access road. Hillsdale claims this failure was arbitrary and capricious.

The Corps’s conclusion is sufficiently supported by the record. The Corps’s methodology involved sampling background air quality at numerous stations in the Kansas City area. 16 The Corps then modeled emissions at or near the intermodal facility, including diesel truck emissions, and added these emissions to the background levels. Its model assumed 100% of truck emissions would occur along the access road. The Corps found these emissions were well below NAAQS levels and would quickly disperse. Consequently, the Corps concluded emissions along I-35 would be even lower and would not be significant.

We agree. Hillsdale does not challenge the Corps’s conclusion that 100% of truck emissions along the access road does not create a significant impact, and it fails to demonstrate how 81 % of the truck emissions farther from the intermodal facility would be more significant. Nor does Hillsdale argue the background emissions levels along 1-35 are higher than the background emissions levels the Corps used to model emissions closer to the intermodal facility. Instead, Hillsdale mostly repeats its arguments regarding the health risks of DPM, which we addressed previously.

Thus the Corps did not “fail[ ] to consider an important aspect of the problem,” and did not violate NEPA, when it declined to model truck emissions along 1-35. New Mexico, 565 F.3d at 704.

G. Water Pollution

The Gardner site sits in the watershed for Hillsdale Lake, a local reservoir and drinking water source for roughly 30,000 people. 17 Hillsdale Lake does not meet Kansas water quality standards for phosphorus and contains high levels of polycyclic aromatic hydrocarbons (PAHs). The Corps concluded the intermodal facility would not have a significant impact on Hillsdale Lake or local groundwater quality. The Corps relied on a BNSF study of potential impacts to water resources to reach this conclusion.

Hillsdale is correct that the environmental assessment and water quality report indicate some potential adverse effects to both Hillsdale Lake and local groundwater. The question is whether the Corps took a hard look at these potential impacts in the environmental assessment and adequately supported its conclusion that they are insignificant. Forest Guardians, 611 F.3d at 710-11. It did.

Hillsdale first argues Hillsdale Lake is “ecologically critical” under 40 C.F.R. § 1508.27(b)(3) because it is a drinking water source. Hillsdale contends this required the Corps to prepare an EIS to examine the potential impacts to the lake. It does not support its claim that all drinking water sources are ecologically critical areas under 40 C.F.R. § 1508.27(b)(3), and we fail to see how this factor is implicated. But Hillsdale also argues that because Hillsdale Lake is a drinking water source, 40 C.F.R. § 1508.27(b)(2), which pertains to effects to public health, requires preparation of an EIS. Hillsdale relies heavily on United States v. 27.09 Acres of Land, 760 F.Supp. 345, 353 (S.D.N.Y.1991), which found “the threatened introduction of contaminants into drinking water” required preparation of an EIS even if it was not projected to violate federal drinking water standards.

Regardless of which § 1508.27(b) factor is implicated, a project’s potential to affect one of these factors does not require an agency to prepare an EIS. The relevant analysis is the degree to which the proposed action affects this interest, not the fact it is affected. See Nat’l Parks, 241 F.3d at 731. Hillsdale’s attempt to create a per se rule that any potential impact to drinking water, however minor, requires preparation of an EIS is thus unconvincing. Instead, Hillsdale must show the intermodal facility has the potential to significantly impact drinking water to establish the Corps’s failure to prepare an EIS was arbitrary and capricious. It does not.

Some passages in the environmental assessment state, for example, that sedimentation may occur during construction or that runoff from the intermodal facility may contain some pollutants and could infiltrate shallow groundwater. But Hills-dale cites nothing establishing that these are “clear and well-documented threats” to local surface and groundwater, as it claims. Aplt. Br. at 55.

On the contrary, the record shows the Corps thoroughly considered the intermodal facility’s impacts to both surface and groundwater. The environmental assessment and water quality report acknowledge runoff may occur during construction and operation of the intermodal facility. Because BNSF plans a number of mitigation and water-treatment measures and must obtain a KDHE construction permit with requirements to reduce erosion and runoff, the Corps found adverse impacts during construction will be minor.

The Corps’s conclusion is supported by a study of runoff at a similar intermodal facility in Birmingham, Alabama. The study found negligible amounts of phosphorus — the major pollutant of concern at Hillsdale Lake — and suspended solids in the runoff. The only pollutants with elevated levels were oil and grease. Based on this study, the Corps concluded the only likely impacts to surface and ground water at the Gardner intermodal facility will be from oil and grease in runoff, these would be minor, and planned mitigation measures would minimize even this small risk.

Hillsdale also argues the Corps should have prepared an EIS to examine groundwater impacts because BNSF’s water quality report acknowledged a potential for contaminated runoff to infiltrate groundwater. As with surface water impacts, the Corps found “there is potential for some contaminated water to infiltrate to shallow groundwater.” App. Vol. XV at 4827. But the Corps also concluded groundwater risks are minimal for several reasons: the project area will mostly be paved or impervious, there will be no underground storage tanks onsite, and the project area consists of clay soil, which has low hydraulic conductivity. Any groundwater contamination also would be limited to groundwater at the facility, which is unlikely to be used for drinking water.

Hillsdale again points to 40 C.F.R. § 1508.27(b)(2), effects to public health, claiming there are numerous drinking water wells near the Gardner site and the impact to those wells should have been examined. But Hillsdale identifies only one comment from a landowner with a well who resides over a mile from the Gardner site. 18 Hillsdale also fails to undermine the Corps’s conclusion that groundwater impacts will be minimal and localized.

Hillsdale fails to carry its burden. The Corps’s decision is entitled to a presumption of regularity, and it finds support in the record. Accordingly, we affirm the Corps’s conclusion that water impacts from the intermodal facility are insignificant.

II. The Controversial Nature of the Intermodal Facility

The Corps’s CWA regulations require it to gauge the intensity of environmental impacts, including the extent to which these impacts are controversial, in assessing the significance of a proposal. 40 C.F.R. § 1508.27(b)(4). Hillsdale claims the intermodal facility project is highly controversial, and the Corps should have prepared an EIS on account of this controversy. See Natl Parks, 241 F.3d at 736-37; Public Citizen v. Dept. of Transp., 316 F.3d 1002, 1024 (9th Cir.2003), rev’d on other grounds by Dept, of Transp, v. Pub. Citizen, 541 U.S. 752, 124 S.Ct. 2204, 159 L.Ed.2d 60 (2004). But Hillsdale overstates the importance of this factor.

Controversy is only one of ten factors the Corps must consider when deciding whether to prepare an EIS. 40 C.F.R. § 1508.27(b)(4). Controversy in this context does not mean opposition to a project, but rather “a substantial dispute as to the size, nature, or effect of the action.” Middle Rio Grande, 294 F.3d at 1229. In addition, “controversy is not decisive but is merely to be weighed in deciding what documents to prepare.” Town of Marshfield v. FAA, 552 F.3d 1, 5 (1st Cir.2008). So even if a project is controversial, this does not mean the Corps must prepare an EIS, although it would weigh in favor of an EIS.

As support for their argument that the intermodal facility is controversial within the meaning of 40 C.F.R. § 1508.27(b)(4), Hillsdale claims 90% of the comments to the Corps’s environmental assessment either disapproved of the project or asked the Corps to prepare an EIS. See Nat’l Parks, 241 F.3d at 736-37 (finding an “outpouring of public protest” when 85% of comments objected to the agency’s chosen alternative).

This argument is without merit. When analyzing whether a proposal is controversial, we consider the substance of the comments, not the number for or against the project. Even if 90% of the comments to the environmental assessment were negative, this merely demonstrates public opposition, not a substantial dispute about the “size, nature, or effect” of the intermodal facility. Middle Rio Grande, 294 F.3d at 1229. National Parks, which Hillsdale cites, found controversy not because of the high number of negative comments but because those comments “cast substantial doubt on the adequacy of the [agency’s] methodology and data.” 241 F.3d at 736-37.

The comments here do not cast substantial doubt on the agency’s methodology and data. Hillsdale is correct that many of the comments they cite are more than mere statements of opposition; they question various aspects of the Corps’s analysis, mostly its failure to analyze the cancer risks of DPM emissions but also the inter-modal facility’s impacts on water quality, regional air quality, and so on.

But all comments Hillsdale identifies raise the same issues it raised in this appeal. As we have discussed, the Corps took the requisite “hard look” at every one of these issues, which is all NEPA requires. Forest Guardians, 611 F.3d at 711. Hillsdale cannot overcome its failure on the merits simply by pointing to comments expressing the same concerns. If Hillsdale cannot show there is some merit to opposing opinions, they cannot demonstrate controversy. Town of Cave Creek v. FAA, 325 F.3d 320, 331 (D.C.Cir.2003); see also Bering Strait Citizens v. U.S. Army Corps of Eng’rs, 524 F.3d 938, 957 (9th Cir.2008).

An additional point in the Corps’s favor is that none of the federal or state agencies the Corps consulted opposed the project or the Corps’s analysis. Although not dispositive, this is additional evidence of a lack of controversy. See Nw. Envtl. Advocates, 460 F.3d at 1139; Nat’l Wildlife Fed’n v. Norton, 332 F.Supp.2d 170, 185 (D.D.C.2004); cf. Friends of the Earth v. U.S. Army Corps of Eng’rs, 109 F.Supp.2d 30, 43 (D.D.C.2000) (finding controversy where “three federal agencies and one state agency have all disputed the Corps evaluation ... and pleaded with the Corps to prepare an EIS”).

In short, neither the nature nor the number of the comments Hillsdale cites demonstrates the intermodal facility is controversial, let alone that the Corps’s decision not to prepare an EIS was arbitrary and capricious in light of this controversy.

III. Conclusion

Having considered the record and the parties’ arguments, we AFFIRM the decision of the district court and uphold the Corps’s issuance of a § 404 permit.

1 . An intermodal facility is a facility where a railroad transfers cargo between trains and other forms of transportation, usually trucks.

2 . Wellsville North was 34 miles from the Argentine Yard, but the Corps concluded it met BNSF’s other criteria and was close enough to the Argentine Yard to warrant a closer look.

3 . Hillsdale Environmental Loss Prevention and the Natural Resources Defense Council each filed separate complaints, but the cases were consolidated.

4 . A “special aquatic site" is a geographic area protected by the CWA. 40 C.F.R. § 230.3(q-1). Wetlands are classified as special aquatic sites. 40 C.F.R. § 230.41.

5 . These cases involved challenges to a NEPA analysis, rather than a CWA analysis. But we see no principled reason why' an agency should be forced to analyze the environmental consequences of impracticable alternatives in either context.

6 . Hillsdale’s citation to Utahns does not support its position, as the portion of the opinion it quotes is a discussion of the appellant’s argument, not the court’s holding. See 305 F.3d at 1.186 (citing Aplt. Br. at 22).

7 . The Le Loup and Ottawa sites met all of BNSF's site-selection criteria save for their distance from the Argentine Yard. These two sites were both more distant than Wellsville North — 40.2 and 44 miles, respectively.

8 . One of the applicant’s criteria in Greater Yellowstone Coalition was that the proposed housing development be small enough to avoid compromising the viability of his working ranch. 359 F.3d at 1271.

9 . As for BNSF’s other site-selection criteria, Hillsdale does not explain why they are flawed or why the Corps’s acceptance of them was arbitrary and capricious. Thus it does not carry its burden with respect to these criteria. Morris, 598 F.3d at 691.

10 . The agreement requires mitigation if PM2.5 concentrations exceed 30 |xg/m3, while the NAAQS for PM2£ is 35 |xg/m3. PM10 concentrations over 100 |xg/m3 trigger mitigation, while the NAAQS for PM10 is 150 |xg/m3.

11 . PM and NOx are both criteria pollutants regulated under the Clean Air Act. 42 U.S.C. § 7408.

12 . Based on these efficiency regulations, the EPA projects that, even assuming a 50% increase in rail traffic between 2010 and 2030, nationwide NOx emissions from locomotives will decrease by roughly 66%, and PM emissions will decrease by about 80%. This means emissions from individual locomotives will decrease by at least a factor of three. The Corps was required to consider the likely impact of these regulations. Wyoming v. U.S. Dept. of Agriculture, 661 F.3d 1209, 1251 (10th Cir.2011).

13 . BNSF makes this argument, but it does not accurately reflect the Corps’s conclusions in the environmental assessment.

14 . The Corps responds that Hillsdale did not raise this issue in its comments to the draft environmental assessment, depriving it of its opportunity to respond on the record. The Corps therefore argues we should not consider this argument because Hillsdale did not exhaust its administrative remedies. Ark Initiative v. U.S. Forest Serv., 660 F.3d 1256, 1261 (10th Cir.2011).

Hillsdale did not raise this claim, but several other comments discussed Kansas City's ozone problems, including two comments that mention concerns with non-truck emissions. Claims not raised before an agency are not waived if they are "obvious, or otherwise brought to the agency’s attention.” Id. at 1262 (internal quotation omitted). These comments make it clear that concerns about non-truck emissions were otherwise brought to the agency’s attention, and so this claim is not waived.

15 . The chemicals considered in the Corps’s analysis include acetaldehyde, benzene, 1, 3-butadiene, formaldehyde, and particulate matter. These are the major toxic DPM components Hillsdale identifies. The record also reflects that DPM can contain trace amounts of various heavy metals, as well as small amounts of dioxins, although it notes diesel engines “are a minor contributor to overall dioxin emissions.” App. Vol. XII at 3968.

16 . The record states many sample stations were in areas more developed than Gardner, although it does not quantify this disparity.

17 . The Corps constructed Hillsdale Lake and continues to operate the reservoir.

18 . Hillsdale cites to one other comment, from an environmental group, but this comment merely states there are many drinking wells in the area without specifying their number or location.