KELLY, Circuit Judge.
Plaintiff-Appellants Colorado Wild, Inc. and Heartwood, Inc. (collectively, the “Conservation Groups”) appeal from a final judgment in favor of Defendant-Appel-lee, United States Forest Service (“Forest Service”) and Intervenor-Defendant-Ap-pellee Intermountain Forest Association (“Intermountain”). The Conservation Groups challenge, pursuant to the Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701-06, a categorical exclusion promulgated by the Forest Service (“Category 13”) that allows the salvage of dead and/or dying trees on up to 250 acres (with up to one-half mile of temporary road construction) to proceed without preparation of an environmental impact statement (“EIS”) or an environmental assessment (“EA”) under the National Environmental Policy Act (“NEPA”), 42 U.S.C. §§ 4321-4370f. The Shaw Lake Vegetation Project (“Shaw Lake Project”) was approved pursuant to Category 13. Our jurisdiction arises under 28 U.S.C. § 1291, and we affirm.
Statutory Background
Our review of the challenged regulation is set against the backdrop of NEPA and
the regulations and guidance created by the administrative agency charged with its implementation, the Council on Environmental Quality (“CEQ”). We therefore begin with a brief overview thereof. NEPA was enacted to regulate government activity that significantly impacts the environment and “to help public officials make decisions that are based on [an] urn derstanding of environmental consequences, and take actions that protect, restore, and enhance the environment.”' 40 C.F.R. § 1500.1(c). The CEQ administers NEPA and promulgates regulations related to NEPA that are binding on federal agencies.
See
42 U.S.C. §§ 4342, 4344(3); 40 C.F.R. §§ 1501-08. Every federal agency then drafts its own administrative regulations to implement and supplement the CEQ regulations.
See
40 C.F.R. § 1507.3.
To effectuate the goals of NEPA, the CEQ created rules requiring agencies to establish implementing procedures that facilitate the evaluation of management decisions and the environmental effects of proposed federal agency actions. Under these guidelines, an agency must identify those actions which normally require an EIS.
See id.
§ 1501.4(a)(1). An EIS is required for “major Federal actions significantly affecting the quality of the human environment.” 42 U.S.C. § 4332(2)(C).
In order to determine whether a particular proposed action requires the preparation of an EIS, agencies perform an EA. An EA is a public document (shorter than an EIS) that contains information pertaining to the need for the proposed action, other alternatives, the environmental impact of the proposal and its alternatives, and other relevant information.
See
40 C.F.R. § 1508.9(b). An agency may prepare an EA for one of several reasons: (1) to provide evidence and analysis that establish whether or not an EIS or a Finding of No Significant Impact' (“FONSI”) should be prepared; (2) to help the agency comply with NEPA when no EIS is necessary; and (3) to facilitate preparation of an EIS when one is necessary.
See id.
§ 1508.9(a)(1)-(3).
When an agency identifies certain actions that do not have any significant effect on the environment, the agency may classify those actions as categorical exclusions (“CEs”). Under NEPA and CEQ regulations, if an action falls within a particular CE, the agency need prepare neither an EIS nor an EA. The CEQ requires federal agencies to design procedures for establishing CEs. Specifically, a CE is'
a category of actions which do not individually or cumulatively have a significant effect on the human environment and which have been found to have no such effect in procedures adopted by a Federal agency in implementation of these regulations (§ 1507.3) and for which, therefore, neither an environmental assessment nor an environmental impact statement is required. An agency may decide in its procedures or otherwise, to prepare environmental assessments for the reasons stated in § 1508.9 even though it is not.required'to do so. Any procedures under this section shall provide for extraordinary circumstances in which a normally excluded action may have a significant environmental effect.
40 C.F.R. § 1508.4.
In Guidance Regarding NEPA Regulations, 48 Fed.Reg. 34,263, 34,265 (July 28, 1983), the CEQ expressed concern that some agencies were developing lists of specific activities which qualify as CEs. The CEQ discouraged this practice, noting that
“if this approach is applied narrowly it will not provide the agency with sufficient flexibility to make decisions on a project-by-project basis with full consideration to the issues and impacts that are unique to a specific project.”
Id.
The CEQ went on to encourage agencies “to consider broadly defined criteria which characterize types of actions that, based on the agency’s experience, do not cause significant environmental effects.”
Id.
In order to establish a CE, the CEQ requires that an agency publish the proposed CE in the Federal Register, provide an opportunity for public comment, and submit the CE to the CEQ for review and approval.
See
40 C.F.R. § 1507.3(a). The CEQ reviews proposed CEs at the draft stage. 48 Fed.Reg. 34,265. After reviewing comments received during the review period and prior to publication in final form, the CEQ determines whether the CEs are consistent with NEPA regulations.
Id.
Factual Background
In 1992, the Forest Service promulgated and adopted a CE for timber harvests “which remove 250,000 board feet or less of merchantable wood products or salvage which removes 1,000,000 board feet or less.” 57 Fed.Reg. 43,180, 43,209 (Sept. 18, 1992) (the “Former CE”). In 1999, a federal district court in Illinois struck down the Former CE.
Heartwood, Inc. v. U.S. Forest Serv.,
73 F.Supp.2d 962, 975 (S.D.Ill.1999).
In 2001, the Forest Service began developing a new set of CEs to cover small-scale timber harvests. As a basis for proposing a new set of CEs, the Forest Service looked at two sets of data. The Forest Service first analyzed all 306 timber harvest projects performed under the Former CE for the year 1998, the last year the Former CE was available to the Forest Service. This data was compiled and reviewed to estimate the extent to which the Former CE was used and to determine average project size (in acres) and harvest volume (in board feet).
Second, in 2001, the Forest Sendee selected, on a nationwide basis, 154 timber harvest projects that were either (1) approved under the Former CE, (2) approved after an EA or an EIS was prepared but fit within the Former CE requirements, or (3) were otherwise small in scope. This sample consisted of 101 dead timber salvage projects and 53 green timber harvest projects. None of the 154 projects reviewed predicted significant effects on the environment before the project was implemented.
Of the 154 projects reviewed, 122 were approved under the Former CE and were documented with decision memos
while 32 were documented with EAs. In addition to reviewing these documents, teams of interdisciplinary resource specialists from the Forest Service (the “Interdisciplinary Teams”) conducted on-site, post-implementation assessments of these projects’ environmental effects. The Interdisciplinary
Teams monitored and documented whether each project met project standards (e.g., Forest Plan Standards or Guidelines, state water quality standards, etc.) for “soil, water, air, vegetation, wildlife, fish, cultural and historic resources, [and] other pertinent issue related resources.” Aplt.App. at 38.
The resulting data indicated that a few of the projects showed “minor soil disturbance and compaction,” while a few others showed that “small numbers of noxious weeds or invasive plants entered the area where the trees had been removed.” 68 Fed.Reg. 44,598 (July 29, 2003). However, based on the information provided by the Interdisciplinary Teams, the respective Forest Service line officer responsible for each timber harvest project reviewed made a finding that these impacts were not significant in the NEPA context, i.e., they did not individually or cumulatively have a significant effect on the human environment.
Id.
at 44,599. Thus, the Forest Service considered pre-implementation prediction and post-implementation verification of no significance in the NEPA context.
Based on the 154 projects reviewed, the Forest Service identified three types of actions that it determined normally do not individually or cumulatively have a significant effect on the human environment. As such, the Forest Service defined the criteria which characterize these actions and proposed three new categories of CEs consistent therewith: (1) harvest of live trees not to exceed 70 acres (green timber harvests or “Category 12”), (2) salvage of dead and/or dying trees not to exceed 250 acres (salvage timber harvests or Category 13), and (3) commercial and non-commercial sanitation of trees to control insects or disease not to exceed 250 acres (“Category 14”).
See
68 Fed.Reg. 44,598. The Forest Service also defined all three CEs to permit no more than one-half mile of temporary road construction.
Id.
In addition, each CE contains the limitation that it cannot be employed for a given project where there are “extraordinary circumstances” present that would preclude its usage. Extraordinary circumstances include, but are not limited to, potentially significant effects on the following:
Federally listed threatened or endangered species or designated critical habitat, species proposed for Federal listing or proposed critical habitat, or Forest Service sensitive species; floodplains, wetlands or municipal watersheds; Con-gressionally designated areas such as wilderness, wilderness study areas, or national recreation areas; inventoried roadless areas; research natural areas; American Indian and Alaska Native religious or cultural sites; Archeological sites, or historic properties or areas.
68 Fed.Reg. 44,599; FSH 1909.15, ch. 30, § 30.3.
In accordance with CEQ regulations, the Forest Service published a draft of the proposed CEs in the Federal Register on January 8, 2003.
See
68 Fed.Reg. 1026-30 (January 8, 2003). It took public comment for the prescribed 60 days during which approximately 16,700 comment letters were received from individuals; representatives of federal agencies; tribes; state and local government agencies; environmental groups; professional organizations; and both commodity and non-commodity groups. 68 Fed.Reg. 44,599. The Forest Service responded to these comments with form letters as well as individual letters.
Id.
Thereafter, pursuant to CEQ procedures, the CEQ reviewed the proposed
CEs and in a letter dated July 22, 2003, notified the Forest Service of the following:
Based on this review, CEQ concludes that the [proposed CEs] conform with NEPA and the CEQ regulations. The [CEs] will take effect once [they are] published in final form in the
Federal Register.
Once established, the agencies will satisfy NEPA when using these [CEs] by determining whether a proposed action falls within the description of the activities and by reviewing the proposal to determine whether extraordinary circumstances exist. In the event extraordinary circumstances exist, an [EA] or an [EIS] would be prepared before proceeding with the proposed action.
ApltApp. at 28 (emphasis in original). On July 29, 2003, the Forest Service published in the Federal Register a final version of the CEs, 68 Fed.Reg. 44,598-608, which are now incorporated in FSH 1909.15, ch. 30, § 31.2.
On July 31, 2004, the Rio Grande National Forest announced the Shaw Lake Project, which proposed to treat approximately 241 acres on the Rio Grande National Forest to address a severe spruce beetle outbreak under Category 13. The Forest Service accepted public comment until August 30, 2005. After consideration of public comments, the Forest Service determined that the project involved less than 250 acres of salvage of dead or dying trees and construction of less than one-half mile of temporary road and did not involve any extraordinary circumstances which would preclude issuance of the project pursuant to Category 13. On September 28, 2004, District Ranger Tom Malecek signed a Decision Memo authorizing the project.
In late 2004, the Conservation Groups mounted both a facial challenge to Category 13 and a challenge to the Forest Service’s use of Category 13 with respect to the Shaw Lake Project.
Aplt-App. at 1-7. The Conservation Groups attacked the Forest Service’s promulgation of Category 13 arguing that it violates section 1508.4’s requirement that a CE only encompass those activities “which do not individually or cumulatively have a significant effect on the human environment.” The Conservation Groups maintained that the Forest Service’s promulgation of Category 13 constituted arbitrary and capricious agency action on two grounds. First, they challenged, for several reasons, the methodology used by the Forest Service in promulgating Category 13. Second, they contended that the Forest Service erred, in multiple ways, in its substantive conclusion that the activities included in Category 13 normally will not have significant environmental impacts. Intermountain, an organization representing forest product companies and timber purchasing sawmills, intervened unopposed as a defendant. The district court held in favor of the Forest Service finding that the Conservation Groups “failed to show that the [Forest Service’s] adoption of [Category 13] was arbitrary, capricious or contrary to law.” Aplt.App. at 135.
Discussion
This case is before us on review pursuant to the APA. We review the district
court’s decision de novo.
Defenders of Wildlife v. U.S. Environmental Protection Agency,
415 F.3d 1121, 1126 (10th Cir.2005) (citation omitted). When reviewing agency action, our scope of review is set forth in the APA. We may not overturn the decision of an administrative agency unless we determine that the decision was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A);
Utahns for Better Transp. v. U.S. Dept. of Transp.,
305 F.3d 1152, 1164 (2002).
The scope of our review under the “arbitrary or capricious” standard is narrow and we are not to substitute our judgment for that of the agency.
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). We confine our 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.
Id.; Olenhouse v. Commodity Credit Corp.,
42 F.3d 1560, 1574 (10th Cir.1994). In reviewing the agency’s explanation, we must determine whether the agency considered all relevant factors and whether there has been a clear error of judgment.
Motor Vehicle Mfrs.,
463 U.S. at 43, 103 S.Ct. 2856 (citation omitted). Agency action will be set aside if “the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”
Id.
Because the arbitrary and capricious standard focuses on the rationality of an agency’s decision making process rather than on the rationality of the actual decision, “[i]t is well-established that an agency’s action must be upheld, if at all, on the basis articulated by the agency itself.”
Motor Vehicle Mfrs.,
463 U.S. at 50, 103 S.Ct. 2856. Thus, the grounds upon which the agency acted must be clearly disclosed in, and sustained by, the record.
Olenhouse,
42 F.3d at 1575. The agency must make plain its course of inquiry, its analysis and its reasoning.
Id.
After-the-fact rationalization by counsel in briefs or argument will not cure noncompliance by the agency with these principles.
Id.
In addition to requiring a reasoned basis for agency action, the “arbitrary or capricious” standard requires an agency’s action to be supported by facts in the record.
Olenhouse,
42 F.3d at 1575. Accordingly, agency action will be set aside as arbitrary unless it is supported by “substantial evidence” in the administrative record.
Id.
“Substantial evidence is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.”
Pennaco Energy Inc. v. U.S. Dep’t of the Interior,
377 F.3d 1147, 1156 (10th Cir.2004) (quotations and citation omitted). “This is something more than a mere scintilla but something less than the weight of the evidence.”
Foust v. Lujan,
942 F.2d 712, 714 (10th Cir.1991) (discussing “substantial evidence” standard). “Evidence is generally substantial under the APA if it is enough to justify, if the trial were to a jury, refusal to direct a verdict on a factual conclusion.”
Hoyl v. Babbitt,
129 F.3d 1377, 1383 (10th Cir.1997). It is not our duty, however, to substitute our judgment for that of the agency’s on matters within its expertise.
Marsh v. Oregon Natural Resources Council,
490 U.S. 360, 378, 109 S.Ct. 1851,
104 L.Ed.2d 377 (1989). Moreover, we typically defer to the “reasonable opinions” of agency experts in matters implicating conflicting expert opinions.
See id.
Moreover, because the Conservation Groups have challenged Category 13 on its face and seek to enjoin its usage by the Forest Service, our review is also governed by the standard applicable to facial challenges.
See Pub. Lands Council v. Babbitt,
167 F.3d 1287, 1293 (10th Cir.1999). To prevail on a facial challenge, the Conservation Groups “must establish that no set of circumstances exist under which the [regulation] would be valid.”
Id.
(quotations omitted; alteration in original). With these principles of review in mind, we turn now to the challenges posed by the Conservation^ Groups.
On appeal, the Conservation Groups press the same two challenges to the Forest Service’s promulgation of Category 13 they raised below.
1. Forest Service’s Methodology
With regards to the methodology employed by the Forest Service in promulgating Category 13, the Conservation Groups object to the following: (1) the statistical analysis used in establishing the 250 acre limit; (2) the sample of projects considered in setting the length of permissible temporary road construction; and (3) the Forest Service’s reliance on the “personal observations” of the Interdisciplinary Teams in concluding that the projects reviewed did not have significant environmental impacts.
A. Acreage limit
The Forest Service set the acreage limit for Category 13 at 250 acres, slightly below the mean acreage of the 101 salvage projects reviewed, 255. The median acreage of the salvage projects reviewed was 50 acres. Some of the difference between the mean and median acreage is attributable to the inclusion of a few large salvage projects in the relevant sample subset, including one salvage project that covered 9,000 acres and four projects totaling over 6,500 acres (collectively, the “largest salvage projects”). By removing the largest salvage projects from the sample subset, the mean of the remaining salvage projects drops to 114 acres. The Conservation Groups maintain that because the inclusion of the largest salvage projects affected the results in this manner, it was arbitrary and capricious for the Forest Service to not use either (1) a mean analysis which excluded . the largest salvage projects (i.e., removing them from the sample subset as statistical outliers) or (2) a median analysis.
The Conservation Groups draw heavily on
Ass’n of Oil Pipe Lines v. Fed. Energy Reg. Comm’n,
281 F.3d 239, 246 (D.C.Cir.2002), and
American Iron & Steel Institute v. Occupational Safety and Health Admin.,
939 F.2d 975, 981 (D.C.Cir.1991), in support of their argument. Accordingly, as a preliminary matter, we address the import of
Oil Pipe Lines
and
Am. Iron & Steel
to the case here.
We conclude that neither
Oil Pipe Lines
nor
Am. Iron & Steel
compels the result urged by the Conservation Groups. In
Oil Pipe Lines,
the D.C. Circuit found that the Federal Energy Regulatory Commission failed to adequately explain
its
decision not to follow its prior method of removing, from the relevant sample set, statistical outliers, which it defined as those data points that are “so extreme as to raise a question whether they may be the result of
recording or measurement errors or some other anomaly.” 281 F.3d at 245. The D.C. Circuit was concerned that the decision to exclude outliers was outcome driven — to avoid increasing the average value: “To the extent that FERC refused to exclude outliers on the ground that doing so changed the result, it obviously missed the whole point: the object of excluding outliers is to prevent extreme and spurious data from biasing an analysis, i.e., affecting its result adversely.”
Id.
at 246.
In this case, no one has suggested that Forest Service inexplicably deviated from past practices or did not conduct an investigation into the largest salvage projects so as to preordain a particular average. In light of the evidence put forth by the Forest Service, the Conservation Groups have not carried their burden of establishing that the largest salvage projects are abnormal, will not recur, and ought to be eliminated. All 101 salvage projects in the sample set, from the smallest five to the largest five, were individually reviewed and resulted in a finding of no significant impact. The suggestion by the Conservation Groups (without benefit of citation to the record) in its reply brief that the “likely anomaly that produced the outliers here was that the [largest salvage projects] reviewed had a relatively small amount of timber removed relative to their acreages” is merely a hunch. Aplt. Rply. Br. at 9. The solution proposed by the Conservation Groups, eliminating the largest salvage projects, from the average seems arbitrary and result-oriented given the process an agency is supposed to engage in when coming up with a CE.
In
Am. Iron & Steel,
the D.C. Circuit addressed an industry challenge to OSHA’s use of a geometric mean in its technological feasibility analysis as unfair because OSHA does not use the geometric mean in its enforcement. 939 F.2d at 990-91. The D.C. Circuit decided to uphold OSHA’s preference for using a geometric mean (which basically excludes statistical outliers) rather than an arithmetic mean (which does not exclude outliers) because OSHA reasoned that the geometric mean was the best statistical method to summarize the routine exposure levels and was, therefore, a good indicator of the feasibility of compliance.
Id.
We do not read
Am. Iron & Steel
as holding (as the Conservation Groups suggest) that it would have in fact been arbitrary and capricious for the agency to have included the outliers. As the Forest Service accurately points out, the D.C. Circuit did not address this issue as it ultimately deferred to OSHA’s choice of statistical analysis.
Id.
at 991.
Accordingly, we turn to the remaining question of whether the Forest Service acted arbitrarily or capriciously in not using a median analysis when setting Category 13’s acreage limit.
But we must be
mindful that under the arbitrary and capricious standard, “our deference to the agency is greatest when reviewing technical matters within its area of expertise, particularly its choice of scientific data and statistical methodology.”
Louisiana ex rel. Guste v. Verity,
853 F.2d 322, 329 (5th Cir.1988) (noting that we do not review an agency’s decision as statisticians, but “as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality”).
As such, we look to the Forest Service’s rationale for using the mean acreage of the salvage projects reviewed when setting Category 13’s acreage limit. Explaining its choice, the Forest Service stated,
The [sample of projects reviewed] was limited to projects that were either categorically excluded [under the Former CE], could have been categorically excluded had [the Former CE] still been available, or that were otherwise small in scope. Based on this limitation, it follows that project sizes would be biased toward smaller acreages. Consequently, the median project size is ... 50 acres for salvage harvests. Since direct, indirect, and cumulative effects arise from acres of activity and not the number of projects, average acreages were used rather than median project size.
Aplee,App. at 18.
From our admittedly lay perspective and this record, the Forest Service’s rejection of the median because of the over-representation of small projects (and its focus on the acres of activity) does not seem irrational. We cannot substitute our views on statistics (including skewed data and outlier analysis) for those of the Forest Service and insist that one measure or another be used.
B. Road Length
The Conservation Groups claim that the Forest Service erred in permitting one-half mile of temporary road construction in the new CEs. Of the 154 timber harvest projects the Forest Service reviewed, 35 involved an average of one-half mile of temporary road construction. Finding no significant environmental impacts from these 35 projects, the Forest Service determined to include one-half mile of temporary road construction in the new CEs. The Conservation Groups maintain that “[g]iven that the Forest Service’s approach was to base the [CEs] on the typical [project] amongst those reviewed, the result should have been to eliminate roads from the [CEs], as the typical [timber harvest project] had no roads.” Aplt. Br. at 22-23. The Conservation Groups argue alternatively that if the Forest Service was determined to include temporary road construction in the CEs, it should have used the average road length for all 154 projects, giving road lengths of zero to the timber harvest projects it reviewed without roads.
Both these arguments are wanting. The Forest Service’s purpose for reviewing the timber harvest projects was to determine whether particular activities normally have significant environmental impacts. To determine whether timber harvest projects containing temporary road construction have significant environmental impacts, the Forest Service logically looked to the
data collected from those projects it reviewed which contained road construction, i.e., the relevant data.
Motor Vehicle Mfrs.,
463 U.S. at 43, 103 S.Ct. 2856. It would have, in fact, been against reason for the Forest Service to consider timber harvest projects without road construction because they add nothing to the determination of whether this particular activity, temporary road construction, normally has significant environmental impacts. The Forest Services’s decision to confíne its analysis to those timber harvest projects containing temporary road construction was not, therefore, arbitrary or capricious.
C. Forest Service’s Reliance on Personal Observations
The Conservation Groups claim that the Forest Service impermissibly relied on the “personal observations” of the Interdisciplinary Teams conducting the post-implementation review process as a basis for finding that the reviewed projects did not have significant environmental impacts. The Conservation Groups draw upon
Northwest Motorcycle Ass’n v. U.S. Dep’t of Agric.,
18 F.3d 1468, 1475 (9th Cir.1994), and
Heartwood,
73 F.Supp.2d at 975, in support of this proposition.
We find
Northwest Motorcycle Ass’n
and
Heartivood
distinguishable. In
Northwest Motorcycle Ass’n,
the Ninth Circuit was not concerned with the agency’s reliance on the “personal observations” of its interdisciplinary team to support its substantive conclusions, but rather the court was troubled by the lack, in the administrative record, of “any direct statements from the [interdisciplinary [t]eam that illustrates the personal experiences of the [agency] employees.” 18 F.3d at 1475. Likewise, the
Heartwood
court’s rejection of the Forest Service’s rationale for the Former CE was similarly predicated on the lack of any evidentiary support in the administrative record, not on the Forest Service’s reliance on the personal observations of its own personnel.
See Heartwood,
73 F.Supp.2d at 975.
Here there are no such evidentiary deficiencies. The record fully demonstrates the factors the Interdisciplinary Teams considered (e.g., whether the timber harvest project’s effects met Forest Plan Standards or Guidelines, state water quality standards, etc.), the monitoring techniques they employed (personal observations, measurements, photo-point, etc.), and the resulting data. Based on this field monitoring data, the respective Forest Service line officer responsible made a determination as to whether the timber harvest project caused a significant impact on the human environment. Supported by this record, we are unpersuaded that the Forest Service acted arbitrarily or capriciously in relying on the personal observations of its Interdisciplinary Teams.
The Conservation Groups further contend that because “it was in the Forest Service’s self-interest to find ‘no significance’ for [the projects reviewed], reliance on ‘personal observation^]’ of the Forest Service personnel is particularly questionable.” Aplt. Br. at 24-25. This argument must be rejected as speculative. Accordingly, we reject all claims by the Conservation Groups that the methodology employed by the Forest Service in promulgating Category Í3 was arbitrary or capricious.
2. The Forest Service’s Substantive Conclusions
In addition to challenging the methodology employed by the Forest Ser
vice in promulgating Category 13, the Conservation Groups assert, on several grounds, that the record shows the Forest Service erred substantively in concluding that projects under Category 13 will normally not have significant impact on the human environment.
A. Dead Timber Salvaging v. Green Timber Harvesting
Relying on
Heartwood,
the Conservation Groups strenuously press the proposition that dead timber salvaging and green timber harvesting have the same environmental impacts. The Conservation Groups claim, therefore, that the Forest Service acted arbitrarily and capriciously in promulgating Category 13 because it allows dead timber salvaging on 250 acres while Category 12 limits green timber harvesting to 70 acres.
The Forest Service does not maintain that there is an intrinsic difference between the environmental impacts of green timber harvesting and dead timber salvaging. The Forest Service argues, however, that
Heartwood
is distinguishable on the basis that, unlike here, in
Heartwood
the Forest Service failed to provide evidence in the administrative record to support its decision to have differing limitations applicable to green timber harvesting and dead timber salvaging. We agree.
In developing Categories 12 and 13, the Forest Service reviewed 154 timber harvest projects, 53 of which involved green timber harvesting and 101 of which involved dead timber salvaging. Based on data gathered from this review, the Forest Service found that none of these projects caused significant environmental impacts. The Forest Service then determined that the mean acreage size of the projects involving dead timber salvaging was 255 acres, while the mean acreage size of the project involving green timber harvesting was 70 acres. As such, the Forest Service concluded that the evidence supported a 70 acre limit applicable to green timber harvesting, while the evidence available for dead timber salvaging supported a larger applicable acreage limit of 255 acres. There is nothing in
Heartwood
to suggest that this approach is arbitrary or capricious. Rather, the Forest Service’s decision to set different acreage limitations applicable to Categories 12 and 13 was a rational response to the evidence before it.
Cf. Motor Vehicle Mfrs.,
463 U.S. at 43, 103 S.Ct. 2856 (agency action will be set aside if “the agency has ... offered an explanation for its decision that runs counter to the evidence before the agency
The Conservation Groups contend that the Forest Service’s rationale is a “false syllogism,” because it only shows that larger salvage projects were chosen for analysis than green timber projects. Aplt. Br. at 30. The Conservation Groups maintain that this result was preordained by the project selection process, whereby the projects reviewed were those that were, or could have been, approved under the Former CE. That is, because the Former CE allowed green timber harvests of 250,000 board feet, and dead timber salvages of one million board feet, it is logical that the latter would be almost four times larger than the former.
We are not, however, persuaded that this observation compels a different conclusion. When developing the new CEs, pursuant to CEQ Regulations and Guidance, the Forest Service was to consider types of actions that, “based on [its] experience, do not cause significant environ
mental impacts.” 48 Fed.Reg. 34,265. In accordance therewith, the Forest Service drew upon prior timber harvest projects, which based on its experience did not cause significant environmental impacts. It is true, as it were, that these projects were previously approved under the Former CE, which had differing board feet limitations applicable to green timber harvesting and dead timber salvaging, and that the resulting data was thereby affected. However, this fact does not place the Forest Service’s rationale in conflict with the proposition that the environmental impacts from these two actions are the same. Nor does it prove that dead timber salvaging normally causes significant environmental impacts. Rather, it merely demonstrates the Forest Service’s lack of experience with, and ability to provide an evidentiary record for, a CE applicable to green timber harvests of a size equal to dead timber salvages.
B. The Record Showing No Significant Impact
The Conservation Groups next forward a succession of substantive claims, through a series of quotations from public comment letters opposing the then proposed Category 13, regarding the following: (1) the reduction of collaboration between the Forest Service and states and tribes, (2) the use of an acreage limit rather than a timber volume limit, (3) the potential for significant cumulative impacts from the use of multiple CEs, and (4) the significance of environmental impacts from temporary road construction. We address each in turn.
First, the Conservation Groups contend that use of Category 13 will severely limit the opportunity for State agencies, Indian Tribes, and the public to provide meaningful input into the decision making process of the Forest Service. Aplt. Br. at 34-36. Public involvement is an important aspect of preparing an EIS.
See
40 C.F.R. §§ 1503.1, 1503.3 (requiring an agency preparing an EIS to request, and respond to, comments from federal, state, and local agencies, Indian tribes, and the public). Public involvement, of a lesser magnitude, is also required for preparing an EA.
See id.
§ 1501.4(b). CEQ regulations do not, however, require public involvement in an agency’s decision to employ a CE once that CE has been approved.
The Conservation Groups’ contention is, of course, grounded in its concern that the Forest Service will approve dead timber salvages under Category 13 without the level of public involvement that it deems necessary to making appropriate environmental decisions. However, the CEQ regulations clearly do not contemplate the level of public involvement that the Conservation Groups seek, nor do the Conservation Groups attack, in this litigation, the CEQ regulations in that regard. As such, while the Conservation Groups may feel strongly about their contention, we are certain that it has no bearing on whether the Forest Service acted arbitrarily or capriciously in promulgating category 13.
Second, the Conservation Groups argue that the Forest Service erred in limiting Category 13 by acreage rather than the total volume of timber removed, or some other un-disclosed hybrid thereof. Aplt. Br. at 37, 38, 43. Although the Conservation Groups concede that an acreage limit may be proper if small enough, they maintain that the 250 acre limit, without some kind of a volume restriction, is impermissible because it may allow a single project to salvage hundreds of thousands, even millions, of board feet of dead timber,
which could cause significant environmental impacts.
As stated previously, our role in reviewing the Forest Service’s decisions in promulgating Category 13 is not to inject our own views or pick sides,
see Motor Vehicle Mfrs.,
463 U.S. at 43, 103 S.Ct. 2856;
Marsh,
490 U.S. at 378, 109 S.Ct. 1851; rather, our role is to ascertain whether the Forest Service examined the relevant data and articulated a rational connection between the facts found and the decision made.
See Motor Vehicle Mfrs.,
463 U.S. at 43, 103 S.Ct. 2856. This review also requires us to determine whether the Forest Service’s decision was supported by facts in the record.
Olenhouse,
42 F.3d at 1574.
Our review of the record reveals that the Forest Service gave due consideration to limiting Category 13’s application by the total volume of timber removed. However, in gathering data on the 154 timber harvest projects reviewed, various problems associated with using a volume limit were revealed that made its selection less favorable, including the fact that “[h]ar-vesting a given volume of timber from one acre is likely to have different environmental impacts than harvesting the same volume from tens of hundreds of acres,” and that “timber volumes are estimated in advance of the sale, and there can be errors associated with those predications; an acreage limit is not as subject to uncertainty of estimation.” 68 Fed.Reg. 44,604. Considering the problems associated with using a volume limit, the Forest Service concluded that “potential environmental impacts are better predicted using acres treated rather than the total volume of timber removed, regardless of acreage.”
Id.
Moreover, according to the Forest Service, acreage limits are easier to control and administer in the field.
Id.
Based on the record and its consideration of the problems associated with using a volume limit, we find the Forest Service’s decision to use an acreage limitation permissible and rational.
The Conservation Groups’ concerns, while certainly not trivial, ignore Category 13’s own limitations. Each dead timber salvage project proposed under Category 13 will have different environmental consequences, some of which could pose the risk of significant environmental impacts. For example, one proposed project could significantly impact federally listed threatened or endangered species, flood plains, research natural areas, etc., while another proposed project could, in fact, involve the intense harvest of a large (e.g., 240 acre) densely populated timber area. Yet, both scenarios are adequately addressed by Category 13’s own requirement that the Forest Service consider whether extraordinary circumstances are present that would preclude its use,
see
68 Fed.Reg. 44,599, and we must presume that the Forest Service will act properly in applying it.
See, e.g., Sullivan v. Everhart,
494 U.S. 83, 94, 110 S.Ct. 960, 108 L.Ed.2d 72 (1990).
Third, the Conservation Groups argue that multiple uses of Category 13 may result in cumulatively significant impacts on the environment. The Conservation Groups’ contention in this respect is twofold: (1) there is a risk that similar salvage projects, over both time and different geographic areas, will cause significant cumulative impacts on the environment, and (2) there is a risk that the Forest Service will split up larger salvage projects into smaller ones so as to fit them within Category 13.
The Conservation Groups’ concerns are fully addressed in the CEQ regulations
adopted by the Forest Service applicable to all proposed projects, including those falling under Category 13. As to the first of the Conservation Group’s cumulative effect contentions, the Forest Service requires “scoping” on all proposed projects, including those that would appear to be categorically excluded. FSH 1909.15, ch. 30, § 30.3. In determining the “scope” of a proposed project, the responsible Forest Service officer is required to consider the cumulative impacts of connected, cumulative, and similar actions.
See, e.g.,
40 C.F.R. § 1508.25(a)(3) (the Forest Service must consider “[sjimilar actions, which when viewed with other reasonably foreseeable or proposed agency actions, have similarities that provide a basis for evaluating their environmental consequences together, such as common timing or geography”); FSH 1909.15, ch. 10, § 11.2 (referencing 40 C.F.R. § 1508.25). If the responsible Forest Service official determines, based on this scoping, that he is uncertain whether the proposed project “may have a
significant effect
on the environment,” he must prepare an EA. FSH 1909.15, ch. 30, § 30.3 (emphasis added). The scoping process therefore addresses the risk that similar salvage projects, over both time and different geographic areas, will cause significant cumulative impacts on the environment.
As to the second of the Conservation Group’s cumulative effect arguments, CEQ regulations preclude the breaking down of projects into small component parts to avoid cumulative significance.
See
40 C.F.R. § 1508.27(b)(7) (in evaluating the significance of impact, an agency must consider “[wjhether the action is related to other actions with individually insignificant but cumulatively significant impacts.... Significance cannot be avoided by ... breaking [a project] down into small component parts.”). The Conservation Groups contends that the Forest Service has shown, through other cases, that it has lost its presumption of good faith when it comes to heeding this prohibition. In support of this assertion, they Conservation Groups rely on
Blue Mountains Biodiversity Project v. Blackwood,
161 F.3d 1208, 1215 n. 6 (9th Cir.1998) (holding that a single EIS was required to address the cumulative effects of five projects that were part of a coordinated strategy);
Foundation for Global Sustainability v. McConnell,
829 F.Supp. 147, 151 (W.D.N.C.1993) (finding that the evidence that the Forest Service did not segment the operation purposely to avoid the mandatory documentation outweighed the inference of improper motive); and
Mahler v. U.S. Forest Service,
128 F.3d 578, 5881, 583 n. 6 (7th Cir.1997) (while not an issue in the litigation, providing a factual background that the Former CE was used for three dead timber salvages). These cases certainly fail, however, to the rebut the presumption that the Forest Service is entitled to here, in this facial challenge, that it will observe this CEQ regulation.
See Sullivan,
494 U.S. at 94, 110 S.Ct. 960.
Last, the Conservation Groups claim that the Forest Service erred in allowing up to one-half mile of temporary road construction under Category 13 because the record, i.e., several comment letters, shows that such road building may cause significant environmental impacts. The Forest Service counters, of course, arguing that the administrative record supports its decision in this regard. Our function in this review is not to weigh the evidence.
See Pennaco Energy,
377 F.3d at 1159. Rather, our role is confined to
ascertaining whether the Forest Service’s decision was supported by “substantial evidence.”
See Olenhouse,
42 F.3d at 1574.
As discussed
supra,
the administrative record demonstrates that the Forest Service reviewed 35 timber harvest projects that involved temporary road construction of an average of one-half mile. Each such project reviewed resulted in a finding of no significant environmental impacts from the limited road construction involved. Based on this record, the Forest Service was adequately supported in its determination that temporary road construction of an average of one-half mile does not cause significant environmental impacts. The fact that the administrative record contains some evidence pointing to a different conclusion does not render the Forest Service’s decision arbitrary and capricious.
See Pennaco Energy, 377
F.3d at 1159.
Accordingly, we reject all claims by the Conservation Groups that the substantive conclusions reached by the Forest Service in promulgating Category 13 were arbitrary or capricious.
AFFIRMED.