The Victoria Project Area (Victoria) is a 16,883 acre portion of the 1,235,780 acre Black Hills National Forest (Forest). Victoria includes 15,372 acres of National Forest System lands and 1,511 acres of private land. The Forest is located predominantly in western South Dakota, but extends into northeastern Wyoming. As required by the National Forest Management Act (NFMA), 16 U.S.C. Secs. 1604-1614, a land management plan for the Black Hills Forest (Forest Plan) was approved in 1983. The Forest Plan covers ten years and, supplemented by the analysis of its effects contained within the Environmental Impact Statement (EIS), directs the Forest Service's management of the Forest. The plan "contains the overall management direction and describes the activities necessary to achieve the desired future condition of the Forest."1

The National Environmental Policy Act (NEPA)2 requires an EIS be prepared for all "major Federal actions significantly affecting the quality of the human environment...." 42 U.S.C. Sec. 4332(2)(C). If these activities were not adequately analyzed in the Forest Plan EIS, and they constitute a "major Federal action," a project level EIS may be necessary in addition to the Forest Plan EIS. See, e.g., Salmon River Concerned Citizens v. Robertson, 32 F.3d 1346, 1356 (9th Cir.1994). If an activity is contemplated that does not automatically require an EIS, an Environmental Assessment (EA) may be conducted to determine whether a project level EIS is necessary. 40 C.F.R. Secs. 1501.3-1501.4. An EA analyzes and compares several alternative courses of action, including doing nothing, the "No Action" alternative. 40 C.F.R. Sec. 1508.9(2)(b). The purpose of the document is to assist in determining whether any of the proposed actions will significantly affect the environment. 40 C.F.R. Sec. 1508.9(a)(1).

In 1990, the Pactola Ranger District commenced planning projects to achieve the goals, or the desired future condition of the Forest, for the Victoria area. The Forest Plan has multiple goals. Timber harvesting is one, along with improving the quality and quantity of wood fiber. See 36 C.F.R. Sec. 219.27(c); Forest Plan at II-16. Other goals are to increase the biological diversity of the Forest and maintain and improve appropriate habitats for existing wildlife species. See 36 C.F.R. Sec. 219.27(g); Forest Plan at II-17. A 1990 survey of Victoria found an absence of vegetative diversity. Generally, the more diverse a habitat, the more species it will support. Forest Plan at II-17. The District considered timber sales as a way of meeting its timber production objectives3 while increasing the diversity of the habitat in Victoria by reducing the portion of the area occupied by closed-canopy and overmature timber stands.

To determine whether timber sales in Victoria would significantly affect the environment, the District undertook an EA which compared the effects of various alternative projects involving timber harvests to the effects of doing nothing, the No Action alternative. The Forest Supervisor considered the EA and issued notice of her decision. The notice indicated the Supervisor found no significant environmental impacts4 would result from the project selected. Thus, no EIS was required for the project.

The Sierra Club5 filed an administrative appeal within the agency, challenging the EA on the grounds that it was defective under NEPA because it analyzed an inadequate range of alternatives. As a result of that appeal, the Forest Service ordered a new EA to address the Sierra Club's concerns.

In January 1992, the Forest Service issued a second FONSI decision. The project selected called for two timber sales in the Victoria Area on 3,209 acres of land, for thinning timber stands on another 2,858 acres, for road work, and also included several measures to decrease the environmental impact of the timber sales.

The Sierra Club again appealed the decision within the agency. The Acting Regional Forester affirmed and the agency declined the Sierra Club's request for review at a higher agency level. In August 1992, the Sierra Club filed suit in district court claiming the second decision violated various provisions of NEPA.

The district court6 granted the Forest Service's motion for summary judgment, determining the Service did not act arbitrarily or capriciously in issuing a FONSI on the proposed timber sales. Contrary to the Sierra Club's claims, the district court found the Forest Service adequately considered the project's impacts. The court stated that NEPA did not require the Forest Service to consider the impacts of acts by private parties on private land in Victoria. The court rejected the argument that the Forest Service inappropriately relied on a computer model for assessing habitat capabilities. To the charge that the Forest Service should have prepared an additional EIS, the district court found the Sierra Club failed to prove the programmatic EIS inadequate.

On appeal, the Sierra Club claims the district court erred in affirming the FONSI and in failing to order a site specific EIS. It argues the court should not have upheld the Forest Service's EA with respect to its cumulative impact analysis as defined by 40 C.F.R. Sec. 1508.7 and as required for EIS documents by 40 C.F.R. Sec. 1508.25. It contends the Forest Service failed to: 1) consider impacts from activities on the 1,511 acres of private land in Victoria; 2) consider impacts from previous timber sales in the Area; 3) consider the effect of timber sales on habitat fragmentation, particularly with regard to species dependent on old growth forest; 4) consider the effect of changing the land designation for 1,504 acres; and 5) include an analysis of diversity unit PO7. In addition the Sierra Club charges the court also erred by interpreting the Sierra Club's request for a site specific, project level EIS as a request for a supplemental EIS and then upholding the Forest Service's decision not to prepare an EIS. We affirm.

We review the district court's grant of summary judgment de novo. As did the district court, we review the Forest Service's FONSI decision under an arbitrary and capricious standard with a concern to determining whether the Forest Service considered the relevant factors or made a "clear error of judgment." Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S. Ct. 814, 824, 28 L. Ed. 2d 136 (1971). We "must render an independent decision on the basis of the same administrative record as that before the district court; the identical standard of review is employed at both levels; and once appealed, the district court decision is accorded no particular deference." Lockhart v. Kenops, 927 F.2d 1028, 1032 (8th Cir.1991), cert. denied, 502 U.S. 863, 112 S. Ct. 186, 116 L. Ed. 2d 148 (1991) (quoting Brown v. United States Dep't of Interior, 679 F.2d 747, 748-49 (8th Cir.1982)). We must affirm if we find the Service took a "hard look" at the project, identified the relevant areas of environmental concern, and made a convincing case for its FONSI. See Audubon Society v. Dailey, 977 F.2d 428, 434 (8th Cir.1992).

The Sierra Club first attacks the EA on its failure to consider the cumulative impacts from private acts on private lands within Victoria. 40 C.F.R. Sec. 1508.7 requires an EA to consider impacts resulting from the activities of Federal or non-Federal agencies or persons.7 The Ninth Circuit has interpreted this section, and we agree, to require analysis of the impacts activities on private land have on the Forest. Resources Ltd., Inc. v. Robertson, 35 F.3d 1300, 1306 (9th Cir.1993). The district court erred in concluding otherwise.

On our review of the record however, we find this error of little consequence because the EA's consideration of the impacts of activities on private land was not so deficient as to make the Forest Service's FONSI arbitrary and capricious. The EA included private land holdings in its maps and aerial photographs. The EA reasonably assumed current land uses would continue within Victoria and expected non-commercial thinning of timber to occur on private land. The watershed report and the road density portion of the wildlife report included private lands in Victoria. The EA also included private land in its fuel/fire risk and smoke assessments, and assumed private construction would continue to occur in the urban forest interface areas. The EA did not make elaborate findings but nothing in the record suggests a need for more extensive analysis. The Sierra Club asserts an EA must address the impacts from such events on private lands as the owners deciding to clear cut all 1,511 acres of their land, but we find no statutory or regulatory mandate that an EA do so. What is required is that the EA recognize the impacts of activities reasonably expected to occur on private lands and this the EA did. See id.

The Forest Service found no significant impacts involving the private lands. The Sierra Club suggests none. This observation does not alter the Forest Service's procedural burden to study these impacts, but finding none, after the obvious inclusions in the record pertaining to private land, maps, photos, road and watershed studies, etc., the Sierra Club cannot simply doubt the FONSI determination without pointing to more than speculative impacts.

The Sierra Club's second and third criticisms of the EA are that it did not assess the impacts of previous timber sales or provide site-specific information about the effect of timber sales on habitat fragmentation and old growth dependent species. Past timber harvests in Victoria, according to an affidavit of a wildlife biologist, have resulted in decreasing the areas of closed-canopy forest and fragmenting the environment into open-forest stands.8 This has the effect of decreasing the biological diversity within the project area and made the area less well suited for deer, elk, birds, and certain small mammals.

The Forest Service contends the EA properly assessed the cumulative effects. The EA did incorporate the impacts of previous timber sales in Victoria because the timber clearing was completed when the existing conditions in Victoria were assessed.

As to habitat fragmentation, the Forest Service points out that prior to European settlement the Black Hills Forest was fragmented to a greater degree than at present.9 The EA addressed the issue of wildlife and habitat diversity and considered measures to mitigate harmful impacts. The EA showed the alternatives considered would reduce open road density, maintain cover areas, and increase the areas be devoted to grass/forb, old growth, forest openings, and forage as called for in the Forest Plan. The alternatives were expected to have favorable impacts for species requiring more open habitats but unfavorable impacts for species requiring a maturer, closed-canopy forest, such as the ruby-crowned kinglet, goshawk, and three-toed woodpecker. Although the No Action alternative allowed for more potential old growth forest, the alternatives actually designated more land for old growth. Over time, the alternative projects would increase habitat diversity by reducing closed-canopy stands, increasing open-canopy stands, and increasing grass/forb areas. The action alternatives would also prevent encroachment by pine trees into aspen, birch, and oak stands, and increase the net growth and quality of timber by reducing overstocking and overmature stands. Fisheries and riparian communities would be little affected, given the proposed mitigation measures and the recommended techniques.

We note that an EA is supposed to be "a concise public document." 40 C.F.R. Sec. 1508.9(a). It is supposed to "[b]riefly provide sufficient evidence and analysis for determining whether to prepare an environmental impact statement or a finding of no significant impact." 40 C.F.R. Sec. 1508.9(a)(1). An EA cannot be both concise and brief and provide detailed answers for every question. With regard to both the impacts of past timber sales and habitat fragmentation, we find the EA sufficient to prevent the FONSI from being arbitrary or capricious.

The fourth deficiency the Sierra Club found in the EA was its failure to assess the effects of changing of the land management designation of 1,504 acres in Victoria to permit commercial timber harvesting. The Forest Service claimed it made this change because the original designation was wrong and did not correspond to actual conditions. We do not believe an EA need consider impacts arising from an error in a Forest Plan. The purpose of an EA is to provide information for determining whether prospective projects may have significant environmental impacts.

Lastly, the Sierra Club contends the EA failed to include a complete analysis of the "diversity units" within Victoria. A "diversity unit" is an area of land quite distinct from project areas, designed for measuring the attainment of Forest Plan biological diversity goals over a large land area. Diversity unit PO6 and a small portion of unit PO7 lie within Victoria. That small portion of PO7, roughly ten percent of PO7's total area and constituting approximately ten percent of the Victoria area, was not analyzed in the community diversity section of the EA. We agree with the Forest Service that an analysis of PO7 would not have contributed much of value to the EA because so little of PO7 lies within Victoria.

Whether, as the Sierra Club claims, the district court misunderstood what kind of EIS it was seeking, and accordingly considered the issue under the wrong standard, is irrelevant. With a project-specific EIS such as the Sierra Club claims to have sought, the standard is whether the Forest Service's FONSI was arbitrary and capricious. See Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 376-78, 109 S. Ct. 1851, 1860-61, 104 L. Ed. 2d 377 (1989). Because we found the EA adequate, all we need consider is whether the Forest Service's FONSI was arbitrary or capricious. On our review, we find the Service thoughtfully reviewed the EA and reasonably concluded a FONSI was appropriate. We hold there was no requirement for the Forest Service to prepare a site specific EIS.

We affirm the order granting summary judgment to the Forest Service.