Thomas v. Peterson: The Ninth Circuit Breathes New Life into CEQ's Cumulative and Connected Actions Regulations

Peter Hapke

Editors' Summary: Federal agencies involved in complex, multi-level planning may have to prepare multiple environmental impact statements and environmental assessments to satisfy the requirements of NEPA. A recent Ninth Circuit case, Thomas v. Peterson, has shed new light on NEPA's requirements for analysis of individual actions that are connected to other activities. The case is the first appellate ruling construing the Council on Environmental Quality's connected-action and cumulative-effect regulations. The author examines Thomas against the background of previous case law on the scope and timing of environmental reviews and analyzes the case's implications, particularly for the federal land management agencies.

Mr. Hapke is an attorney with the Natural Resources Division, Office of the General Counsel, United States Department of Agriculture. The views presented in this article are solely those of the author and do not in any way reflect the views of the Office of the General Counsel of the Department of Agriculture.

[15 ELR 10289]

Determining the scope and timing of an environmental impact statement (EIS) is a recurring problem under the National Environmental Policy Act (NEPA).¹ NEPA requires federal agencies to prepare an EIS for every proposal that significantly affects the quality of the human environment.² To that end, agencies routinely prepare environmental assessments (EAs)³ to determine a proposal's likely impacts. If they are significant, as defined in the Council on Environmental Quality's (CEQ's) NEPA regulations,⁴ the agency prepares and EIS; if they are not, the agency issues a finding of no significant impact (FONSI).⁵

A proposal seldom stands alone, however. Usually it is part of a larger program, a branch in a tree of planned actions. A given action may be closely linked with concurrent actions at the same level of the tree, or linked to divergent branches (future actions). Agency actions at the top of the tree tend to be definite and site-specific, while the actions represented by the trunk are general and broad in scope, requiring a programmatic EIS.

The accepted manner to comply with NEPA in these situations involves "tiering" EISs and environmental assessments (EAs),⁶ preparing them at key points on the planning tree and having site-specific actions above rely on the series of environmental analyses below. By tiering, the agency must decide where in the planning tree it will prepare and EIS and what should be its scope. An agency may violate NEPA by failing to prepare a programmatic EIS for general actions in the trunk, preparing an EIS with a narrow scope that ignores concurrent or connected actions in the planning tree, or by relying on a programmatic EIS in the trunk to adequately cover site-specific actions in the upper branches. On the other hand, an agency can waste time and effort if it casts the scope of its EISs so broadly as to include unconnected concurrent or indefinite future actions.

Scoping and tiering are particular concerns for the Forest Service, an agency that sets national goals, looking decades into the future, and translates those goals stepwise into local action.⁷ To avoid preparing an EIS for every local proposal, the Forest Service tiers EAs on local actions to EISs broader in scope.⁸ Whether a reviewing court will uphold an agency's use of tiering is problematic because neither the CEQ nor the courts have given clear guidance in the area. NEPA itself merely calls for an EIS on "proposals" for "major Federal actions significantly affecting the quality of the human environment,"⁹ while not defining what a proposal is nor how detailed the examination of impacts must be. The courts have approached the problem under various rubrics discussing the need for programmatic EISs,¹⁰ for analysis of related actions,¹¹ for analysis of cumulative impacts,¹² or for analysis of site-specific impacts.¹³ The results have not always been consistent and, [15 ELR 10290] although tiering as a concept has been upheld, no brightline test for what constitutes proper tiering has emerged.¹⁴ Some guidance has come from CEQ's NEPA regulations,¹⁵ but questions remain.

A recent Ninth Circuit case, Thomas v. Peterson,¹⁶ has shed some new light on the problem. The case examined whether the Forest Service had to discuss the impacts of future timber sales in an EA on a logging road into a roadless area. The Forest Service argued that the timber sale proposals were too speculative to be covered in the road's EA. The Ninth Circuit disagreed. Bringing together case law and the CEQ regulations, the court wrote what may become a leading case on the issues of scope and timing, with important implications for land management agencies.

The Law Brfore Thomas

NEPA itself says little about how an agency must integrate the EIS process into its decisionmaking. It merely calls for an EIS to accompany "proposals" for "major Federal actions significantly affecting the quality of the human environment."¹⁷ It does not say how wide the scope of an EIS should be, other than requiring the EIS to discuss "the environmental impact of the proposed action," "any adverse environmental effects which cannot be avoided," and "any irreversible and irretrievable commitments of resources."¹⁸

The courts have fleshed out these statutory directives, tailoring them on a case-by-case basis to the realities of agency planning. Early on, two distinct rules emerged governing scope and timing of EISs. In Scientists' Institute for Public Information, Inc. v. Atomic Energy Commission (SIPI),¹⁹ the District of Columbia Circuit set out the irretrievable commitement of resources test for determining the timing of an EIS. SIPI concerned the Atomic Energy Commission's breeder reactor program. The Commission had planned to issue individual EISs on major aspects of the program, but none on the program as a whole. The court found that the investments required for the first steps of the program were so great, the Commission would effectively be committing itself to the whole program and would be unable to make a meaningful unbiased decision on later steps.²⁰ Because taking the first few steps would virtually ensure commitment of more resources in the future, the court ordered the Commission to prepare an EIS covering the entire breeder program.

The second distinct rule, the independent significance or utility test, which goes to the issue of scope, arose from highway construction cases. In the leading case, Indian Lookout Alliance v. Volpe,²¹ the Iowa State Highway Department had developed a 20-year plan envisioning eventual construction of nearly 2,000 miles of highway. The Department applied for federal funding for a seven-mile segment included in the plan. Plaintiffs contended that if the project received federal funding, the entire 2,000 mile plan required an EIS. The trial court concluded that an EIS covering a 14-mile segment of the highway was needed.²² On appeal, the Eighth Circuit recognized that limiting the scope of the EIS to a small segment of the system would ignore the cumulative impact of the entire plan and foreclose the consideration of alternatives.The court also recognized that an EIS on such a tentative plan would be uneconomical and provide little useful information. Therefore, the court concluded, if a proposed segment had "independent utility,"²³ then an EIS limited to that segment was appropriate. Because the 14-mile segment chosen by the trial court, had no independent utility or significance, the Eighth Circuit remanded the case.

As courts began to apply the irretrievable commitment and independent significance tests to different areas,²⁴ their subjectivity became increasingly apparent. The outcome of a case could depend on which rule the court chose to apply.²⁵ Court decisions that required a programmatic EIS often discussed the irreversible commitment involved, while courts that wanted to uphold an EIS of limited scope often directed their analysis towards a finding of independent significance.²⁶ For example in Minnesota Public Interest [15 ELR 10291] Research Group v. Butz,²⁷ the Eighth Circuit had to decide whether the Forest Service should prepare an EIS covering its timber sale program in the Boundary Waters Canoe Area. Even though each timber sale arguably had independent significance, the court held that they were interrelated and required a programmatic EIS.

Conversely, in Trout Unlimited v. Morton,²⁸ the Ninth Circuit had to decide whether the two phases of the Teton Basin Project could be covered in separate EISs. The first phase involved construction of a dam and disposal of half the resulting reservoir's capacity; the second phase involved disposal of the rest of the reservoir's water. Noting that the first phase was feasible whether or not the second phase was started, the court applied the independent significance test and limited the scope of the EIS to the first phase.

The District of Columbia Circuit began to address the weakness of the two tests in Sierra Club v. Morton.²⁹ In that case, plaintiffs sought to have the Department of the Interior prepare an EIS covering coal development in the Northern Great Plains region. The Department was preparing a national programmatic EIS on coal leasing and EISs for individual developments, but claimed that no plan existed for the region as such. In deciding the case, the court endorsed the basic test it put forth in SIPI and the importance of ensuring that planners consider cumulative effects of projects. However, it noted the "practical difficulties in [the] broad application" of an EIS requirement for related actions: "An infinite number of geographic, environmental, or programmatic interrelationships might be found among the various individual federal projects under way throughout the country. Surely, however, an infinite number of comprehensive plans, and comprehensive impact statements, are not required."³⁰ The court stressed that the scope and timing inquiry should involve a balancing of the factors set out by the SIPI court,³¹ including the likelihood a program will come to fruition, the availability of meaningful information on program impacts, the extent to which irretrievable commitments are being made, and the severity of expected impacts. Applying the SIPI balancing test, the court held that the federal government had treated coal development of the Northern Great Plains region, including leasing, such that the actions comprised, cumulatively, a major federal action significantly affecting the quality of the human environment. However, the court also stressed that the agency should make the initial finding on whether a programmatic EIS was necessary, but could not avoid preparing an EIS simply by declining to formally acknowledge the existence of a program.

The District of Columbia Circuit was not alone in moving away from the mechanistic application of the independent significance and irreversible commitment tests. By 1975, the Fifth, Eighth, Ninth, and Tenth Circuits had also begun experimenting with broader more reasoned analytic frameworks.³² However, the circuits had produced no clear guidance on the scope and timing issues. Not surprisingly, the Supreme Court agreed to review Sierra Club v. Morton.

Kleppe v. Sierra Club

The resulting Supreme Court opinion, Kleppe v. Sierra Club,³³ reversed the circuit court but did not cleanly resolve the underlying scope and timing issues. The Court held that a programmatic EIS is not required unless there is an agency proposal for the action or group of actions under review. The Court reached this conclusion through a strict reading of NEPA, which calls for an EIS "in every recommendation or report on proposals for legislation and other major Federal actions significantly affecting the quality of the human environment."³⁴ On the facts before it, the Court determined that there was no regional coal proposal at all. At best, the regional program was merely "contemplated," and contemplation did not rise to the level of a proposal. All actions that had been proposed were either local or national in scope and were accompanied by undisputed EISs.

In taking this "proposed action" approach, the Court explicitly rejected both the use of a balancing test to decide whether an action was ripe for environmental analysis and the proposition that an EIS may be required before a contemplated action is formally proposed. However, the Court did not provide a test for distinguishing proposed from contemplated actions.³⁵

But the Court further recognized that an agency could violate NEPA by deciding not to cover a series of intimately connected proposed actions in a single EIS. The Court held that the plaintiffs had failed to prove that Interior had acted arbitrarily in deciding not to prepare such an EIS for the proposals within the North Plains region. However, the Court discussed generally when NEPA might require a comprehensive EIS, suggesting several factors for the lower courts to consider, including the cumulative environmental impacts, the extent of interrelationship among the proposed actions, and practical considerations of feasibility.³⁶ Concerning cumulative effects, the Court stated:

[W]hen several proposals for coal-related actions that will have cumulative or synergistic environmental impacts upon a region are pending concurrently before an agency, their environmental consequences must be considered together. Only through comprehensive consideration of pending proposals can the agency evaluate different courses of action.³⁷

Equally important, the Court recognized that in some natural resource situations, geographical ties may set [15 ELR 10292] reasonable bounds for analysis. The Court noted with approvalthe Secretary of the Interior's plans to prepare EISs for regions determined by "basin boundaries, drainage areas, [and] areas of common reclamation problems."³⁸

Thus, Kleppe left the details of scope and timing analysis unresolved and left the door open for the evolution of new approaches to the issues. By not basing its decision on satisfaction of the irretrievable commitment and independent significance tests, the Court's decision invited development of more rational approaches that were neither mechanistic nor outcome determinative. By emphasizing both cumulative impacts and practical considerations, the Court laid the foundation for broad and rational analyses of scope and timing problems.

Post-Kleppe Developments

Though Kleppe really left more open for resolution than it resolved, the message that the lower courts seemed to read from it and other Supreme Court NEPA cases of the mid-1970s³⁹ was do not be too innovative; do not read things into the statute that are not expressly there. As a result, the post-Kleppe courts did not engage in broad experimentation, but largely kept to the primitive trails blazed by the pre-Kleppe decisions.

The force for innovation after Kleppe came not from the courts but from CEQ. In 1978, under presidential order,⁴⁰ CEQ promulgated a comprehensive, binding⁴¹ set of regulations on the implementation of NEPA. To a large degree, the CEQ regulations were a codification of the existing case law, but inevitably in restating the law, the regulations added to it.

The regulations introduced the idea of scoping as a process an agency should undertake early on in preparing an EIS to set the bounds of the study.⁴² In setting those bounds, CEQ told agencies to group together

(1) Connected actions, which means that they are closely related and therefore should be discussed in the same impact statement. Actions are connected if they:

(i) Automatically trigger other actions which may require environmental impact statements.

(ii) Cannot or will not proceed unless other actions are taken previously or simultaneously.

(iii) Are interdependent parts of a larger action and depend on the larger action for their justification.

(2) Cumulative actions, which when viewed with other proposed actions have cumulatively significant impacts and should therefore be discussed in the same impact statement.

(3) Similar actions, which when viewed with other reasonably foreseeable or proposed agency actions, have similarities that provide a basis for evaluating their consequences together, such as common timing or geography.⁴³

The scope regulation brought together many threads spun in individual court opinions and wove them into a whole more inclusive than any one court's rule. Yet the regulation is not overbroad. It links actions that rationally should be linked in environmental analyses.

As is true with many parts of the CEQ regulations, the scope regulation has not instantly become the courts' lodestar on scoping issues. It is possible to find recent opinions completely ignoring the CEQ rules and applying pre-Kleppe law. A good example is National Audubon Society v. Hartz Mountain Development Corp.,⁴⁴ where a New Jersey district court upheld a finding of no significant impact on the first half of a wetlands development project that had been intentionally split in two to avoid EIS preparation. But in one or two past instances, where a CEQ regulation has been "discovered" by a court in an important case, the regulation has quickly become the accepted standard.⁴⁵ For the scope rule, Thomas v. Peterson may prove to be the seminal case.

Thomas v. Peterson

Thomas involved a Forest Service proposal to build a road into the Jersey Jack roadless area in the Nez Perce National Forest in Idaho. Usually, when the Forest Service proposes to take such action, the proposal is part of a larger program for the development of the forest and is covered by one or more programmatic EISs. These programmatic EISs concentrate on the cumulative broader impacts of development and presuppose that site-specific EISs or EAs will be prepared to consider local specific impacts of particular projects.

In the Nez Perce National Forest, the agency had prepared a forest-wide EIS on timber management and road development in 1974, and later had prepared ten unitplan EISs covering impacts in more detail for ten subdivisions of the forest. Development of the Jersey Jack was not included in any of these EISs, however, because the area had been set aside for potential wilderness designation. It was not until 1980, when Congress drew the boundaries of the nearby River of No Return Wilderness, that the Jersey Jack was released for multiple-use development.⁴⁶

The Forest Service prepared an EA on construction of the road to determine whether to prepare an EIS. The agency concluded that the project would not cause significant environmental impacts, and thus an EIS would not be necessary. Subsequent to the decision to build the road, the Forest Service issued EAs for proposed timber sales that would use the new road for access and timber removal. Each EA covered only the effects of a single sale, and each resulted in a finding to no significant impact.

The plaintiffs raised NEPA, National Forest Management Act, and Endangered Species Act claims against the [15 ELR 10293] Forest Service. In the NEPA claim, the one of interest here, they charged that the road and sales were sufficiently related to require treatment in a single EIS.

The Ninth Circuit agreed. Citing Kleppe, the court noted that although agencies have considerable discretion in setting the scope of environmental analyses, they sometimes must consider several related actions in a single EIS.⁴⁷ In particular, the court noted an agency may not "divid[e] a project into multiple 'actions,' each of which individually has an insignificant environmental impact, but which collectively have a substantial impact."⁴⁸ To determine the constraints on agency discretion in setting scope, the court turned first to the CEQ regulations.

The court found that, under either the CEQ connected actions⁴⁹ or cumulative actions⁵⁰ definitions, the road and sales had to be analyzed together in a single EIS. The road and sales were connected because "the timber sales cannot proceed without the road, and the road would not be built but for the contemplated timber sales."⁵¹ The EA for the road stated that "the need for a transportation route in the assessment area is to access the timber lands to be developed over the next twenty years."⁵² Moreover, the Forest Service's cost-benefit analysis of the road considered timber to be the only significant benefit of the road. Finally, an agency letter indicated the close dependency between the road and sales.

The court also found evidence in the record to suggest that "the road and timber sales will have cumulatively significant impacts,"⁵³ particularly on fish and wildlife. The adjacent Salmon River supported an important salmon and steelhead fishery, and the Jersey Jack area itself included potential habitat for the endangered Rocky Mountain gray wolf. The EA itself contained a U.S. Fish and Wildlife Service comment that "[s]eparate documentation of related and cumulative potential impacts may be leading to aquatic habitat degradation unaccounted for in individual EAs" and "lack of an overall effort to document cumulative impacts could be having present and future detrimental effects on wolf recovery potential."⁵⁴

The court then turned to its own precedents to reaffirm and amplify its conclusions under the CEQ regulations. It cited Trout Unlimited⁵⁵ for the proposition that an EIS must cover all the phases of a project if "the dependency is such that it would be irrational, or at least unwise, to undertake the first phase if subsequent phases were not also undertaken."⁵⁶ Declaring this independent utility or significance principle to be the same as those embodied in the highway cases, the court held that the Forest Service had not alleged sufficient independent utility for the road to justify its construction without the sales.

But the court then also invoked a variation of the irretrievable commitment rationale to hold that the NEPA process should be integrated with agency planning at the earliest possible time.⁵⁷ The court reasoned that the cumulative environmental impacts of the road and the timber sales must be assessed in one EIS because "building the road swings the balance decidedly in favor of timber sales even if such sales would have been disfavored had road and sales been considered together before the road was built. Only by selling timber can the bulk of the expense of building the road be recovered."⁵⁸ Responding to the agency argument that the sales are too uncertain and too far in the future for their impacts to be analyzed along with the road, the court, citing City of Davis v. Coleman,⁵⁹ countered that if the sales are sufficiently certain to justify construction of the road, then they are sufficiently certain for their environmental impacts to be analyzed.⁶⁰ The court noted that the Forest Service issued EAs for two sales 16 months after issuing the road EA and concluded that "the record plainly establishes that the Forest Service, in accordance with good administrative practices, was planning contemporaneously the timber sales and the road building."⁶¹

Legal Implications

The Thomas court validated the CEQ scope regulations and acknowledged them as a logical outgrowth of case law. The opinion establishes an analytic framework for scope and timing issues, particularly those involving public lands. The decision may have far - reaching implications for land management agencies including the Forest Service and the Bureau of Land Management.

One important lesson of Thomas grows out of what the court did not do: it did not entangle itself in the semantics of "proposals" and "contemplated actions" as some courts trying to follow Kleppe have done. The Supreme Court in Kleppe placed great emphasis on not forcing agencies to prepare EISs on merely "contemplated" actions, but to wait until the actions rose to the level of a proposal. This emphasis has steered some courts towards focusing their inquiry on just what has been proposed and what is outside the proposal, or is only contemplated.⁶² The distinction between the proposed and the contemplated is artificial and only encourages agencies to keep their decisionmaking informal or to narrow the scope of a formally proposed action to avoid broader NEPA analysis.⁶³ This [15 ELR 10294] approach turns informed public decisionmaking, the primary goal of NEPA, on its head.

Courts that focus too intently on the limits of a particular proposal ignore the other major lesson of Kleppe: "there are situations in which an agency is required to consider several related actions in a single EIS."⁶⁴ Once an agency has made a proposal, nothing in Kleppe allows the agency to limit the scope of NEPA analysis to the bounds of the proposal. Kleppe suggested that an agency might have to prepare an EIS on a group of proposals if they were related closely enough, even if no super- proposal covered them all,⁶⁵ and Kleppe was expressly silent on the question of whether an analysis of a single proposal would be inadequate if it ignored related "contemplated" actions.⁶⁶

The Thomas court explored this analytical gap left by Kleppe. It acknowledged expressly the traditional independent significance test and implicitly the irretrievable commitment test. However, it broadened the conventional analysis of scope and timing. Taking its cues from the CEQ regulations, it crafted an approach that highlighted cumulative impacts as the controlling factor in determining EIS scope and timing.

So, for example, the court discussed the independent significance test as applied in the circuit and held that the road had no independent significance. However, that holding appears to be an afterthought to the court's broader connected and cumulative action analysis under the CEQ regulations.⁶⁷ Under that analysis, the court implied that an EIS on an independently significant proposal would nonetheless have to look beyond the proposal at actions triggered by the proposal and actions with cumulative effects.

Similarly, the court raised the question of whether the agency would be irretrievably committing resources before preparing its EIS, but did so in the context of whether resources would be committed in connected actions or through cumulative effects.

A recent federal district court case relying on Thomas illustrates the cogency of the cumulative effects approach as applied to a related natural resource development area, oil and gas leasing. Conner v. Burford⁶⁸ focused on the adequacy of a Forest Service EA and FONSI for issuance of leases in the Flathead and Gallatin National Forests in Montana. Plaintiffs claimed that full NEPA analysis should have been done before leases were issued. The leading case in the leasing area, Sierra Club v. Peterson,⁶⁹ held that NEPA compliance must occur at the initial leasing stage unless the lease-issuing agency reserves the authority to review all proposed activities on the lease site and preclude any activity if its environmental consequences would be unacceptable. In on-shore leasing, such a reservation of authority may take the form of a No Surface Occupancy (NSO) stipulation.

In Conner the court believed that, although the leases had NSO stipulations and the Forest Service had the option to deny development under the leases in areas that it considered unsuitable for oil and gas activity, the agencies involved might reevaluate the stipulations. Reasoning that such reevaluation could lead to modification or removal, without any assurance that the agencies involved would do a comprehensive NEPA analysis, the court went beyond Sierra Club v. Peterson by holding that the cumulative effects of all the leases in a geographic area must be evaluated in an EIS at the lease issuance stage, regardless of NSO stipulations. Without an EIS analyzing all the cumulative effects, the court concluded, "a piecemeal invasion of the two national forests will occur, possibly leading to significant and irreversible impacts."⁷⁰

Conner is being appealed; whether the Ninth Circuit will uphold the ruling is problematical. There are strong practical considerations that argue for postponing an examination of cumulative effects of oil and gas development until preliminary exploration provides some idea of the magnitude of development that is likely.⁷¹ But Conner correctly concluded that under Thomas, agencies cannot just mechanically look at whether a particular proposed action involves a commitment of resources. Consistent with NEPA's goal of informed decisionmaking, the agencies must at some early point weigh connected actions and cumulative impacts in their NEPA analysis.

[15 ELR 10295]

Policy Implications for Land Management Agencies

The Thomas court's emphasis on the CEQ cumulative impact and connected actions regulations, coupled with its reasoning that NEPA's purpose "cannot be fully served if consideration of cumulative effects of successive, interdependent steps is delayed until the first step has already been taken," sends out strong signals to the agencies concerning NEPA compliance. Agencies would do well to take a second look in their EAs and EISs to make sure they are not missing cumulative effects or connected actions, as defined in the CEQ rules. The agencies should examine not only the regulations cited in Thomas, but also the regulatory definition of "significantly," which requires the agency to consider:

[w]hether the action is related to other actions with individually insignificant but cumulatively significant impacts. Significance exists if it is reasonable to anticipate a cumulatively significant impact on the environment. Significance cannot be avoided by terming an action temporary or by breaking it down into small component parts.⁷²

Agencies should examine the regulatory definition of "cumulative impact" as well:

"Cumulative impact" is the impact on the environment which results from the environmental impact of the action when added to other past, present and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person undertakes such other actions.⁷³

In keeping their eyes open for cumulative effects, agencies will surely be forced to prepare EISs or comprehensive EAs in situations that before Thomas might have merited a short EA and FONSI.⁷⁴ However, this will not be senseless busywork. It will prevent the potentially harmful cumulative impacts of agency action from slipping through the cracks in the agency decisionmaking process.

Complying with NEPA in the Planning Process

Agencies should give some thought to how they plan for and naturally group-connected actions. Often in natural resources planning, agencies can achieve this goal by grouping actions geographically, choosing ecological rather than man-made boundaries and considering all manner of activity in a planning area. In multiple-use land management, drainages often function as natural management units. This geographical division may shield the drainage area from the visual and physical impacts of commodity development, such as mining or timber harvesting, below the drainage ridgeline. Unfortunately, drainages tend to funnel and amplify the impacts of development, rather than disperse and dilute the impacts across a broader land area. Thus, within a drainage a timber sale, a mine or a grazing allotment may have cumulatively significant effects on scenery, soil erosion, stream sedimentation, and water quality over a broader area than actual project site or grazing allotment.

Early drainage-wide assessment of environmental effects should lead to better land management. In planning site-specific projects, for example, early analysis of the cumulative environmental effects likely within a drainage will tell the managing agency at what point the impacts will reach an unacceptable level. If the analysis turns up an ecological limit to development, such as sedimentation levels, then a commitment to one project may necessarily limit the other projects that may be undertaken in the drainage. By discerning these limits early on, through a comprehensive environmental analysis, an agency can choose the best projects to achieve its goals with the least environmental cost.⁷⁵

Courts since SIPI have noted the logic of preparing comprehensive environmental analyses for geographic areas.⁷⁶ SIPI cited with approval a CEQ memorandum that called for a programmatic EIS for groups of "[i]ndividual actions that are related either geographically or as logical parts in a chain of contemplated actions."⁷⁷ in Kleppe the Supreme Court noted with approval the Secretary of the Interior's plan to prepare EISs on the cumulative effects of actions in regions described by "basin boundaries, drainage areas, [and] areas of common reclamation problems."⁷⁸ This approach has been applied to highway corridors, water resource projects, and urban renewal projects.⁷⁹

One of the best early examples of a court endorsing a geographic approach is Natural Resources Defense Council, Inc. v. Callaway.⁸⁰ In that case, the Second Circuit held [15 ELR 10296] an EIS for a Navy dumping proposal inadequate because it failed to discuss the cumulative impacts of other dumping and dredging "proposals" targeted for the same dump site in Long Island Sound. The court rejected the Navy's argument that many of the proposals were unrelated to the Navy proposal. The court required an EIS covering all the proposals for the same geographical area presenting similar pollution problems.

A recent Texas federal district court memorandum opinion, Fritiofson v. Alexander,⁸¹ struck down an Army Corps of Engineers EA/FONSI on the basis of a cumulative impact/geographical area approach. Fritiofson concerned the issuance of a wetlands fill permit for a commercial development on ecologically sensitive, relatively pristine, part of Galveston Island. The court held that the project was clearly related to other past, present and reasonably foreseeable developments in the estuary. The Corps violated NEPA by assessing the impact of the permit application within the narrow confines of the Corps' stated policy of evaluating "each permit on an individual basis."⁸² The court held that the CEQ definitions of "significantly" and "cumulative effects" compel an inquiry into whether the action is related to other actions, whereas the Corps based its finding on an interpretation requiring an inquiry into whether the action is "related to any overall plan."⁸³ The Corps "overall plan" language, said the court, is irrelevant when the issue does not concern the need for a regional EIS as in Kleppe.

More notably, a federal district court in Oregon recently determined the proper scope and timing of environmental analysis under NEPA using a geographical approach. In National Wildlife Federation v. United States Forest Service⁸⁴ (the Mapleton case), the court held, in striking down a Forest Service EA/FONSI, that the agency acted unreasonably in not preparing an EIS for a Seven Year Action Plan that set out proposed timber sales of approximately 100 million board feet per year and entailed construction of 176.4 miles of roads in the Mapleton Ranger District of the Siuslaw National Forest. In 1979 the Forest Service had adopted a Timber Resource Plan (TR Plan), with an accompanying EIS, for the entire national forest. This EIS was analogous to the forest- wide programmatic EIS in Thomas, in that it was a programmatic statement averaging the impact of proposed harvests over the entire forest. Like the Jersey Jack roadless area, the Mapleton District did not have a second-tier unit-plan EIS to which timber sale EAs could tier.

During the trial, the Forest Service argued that a second multi-level EIS was unnecessary because the general impact of the Seven Year Action Plan activities had already been considered in the TR Plan. The court found that the TR Plan EIS did not provide sufficiently specific environmental analysis under NEPA. To this the Forest Service replied that local effects would be considered in the EAs done for each timber sale. The court held that such site-specific EAs would be inadequate because they would fail to analyze cumulative effects of all timber harvesting in and around the Mapleton District.⁸⁵ The court based its analysis on a geographic approach, finding that the Mapleton District watersheds will suffer three sets of cumulative impacts: (1) seventy-five timber sales from the Seven Year Action Plan, (2) timber sales on adjoining BLM land, and (3) timber harvests on private lands. Moreover these three impacts will occur in the same geographical area, involve extensive clearcutting, and present similar threats to fish habitats.The court stated, however, thatthe agency could prepare site-specific EAs after preparing a programmatic EIS assessing these impacts for the Mapleton District."⁸⁶

Cumulative Effects and Tiering

Mapleton illustrates another implication of the Thomas analysis for land managers: the danger of tiering site-specific actions to overly broad EISs. If an agency's NEPA analysis jumps too far from the general EIS to the site-specific or local action analysis, it may fail to cover cumulative impacts, as required under the Thomas decision. By analogy, an agency's NEPA analysis is jumping from the general EIS on the trunk of the planning tree directly to site-specific actions in the upper leafy branches, without any supporting branches in between. As the geographical area of these newer agency plans becomes greater, it becomes much more difficult for the agency to fill the analytical gap between the general or programmatic EIS and the NEPA analysis accompanying the site-specific action. Such a wide gap makes tiering difficult. As an example of the gap problem the Forest Service, pursuant to the 1976 National Forest Management Act (NFMA),⁸⁷ has been preparing forest-wide land management plans with accompanying EISs.Most plans should be completed by the end of 1986, with the possible exception of the Pacific Northwest forests. One approach an agency might take to avoid tiering problems would be to add sufficient information to the plan, including detailed cumulative impact analysis to support tiering of local EAs. The addition of enough detailed information in an agency plan EIS to sustain a legal challenge, however, could make the EIS too long and complex for responsive public review.

[15 ELR 10297]

In the alternative, given the difficulty in reaching a level of specificity sufficient to allow tiering directly to local EAs, an agency could prepare a second tier of EISs. This intermediate level could be based on drainages or similar natural boundaries for environmental impacts as outlined in the geographical approach above. Then, despite the analytic burdens that the Thomas interpretation of NEPA places on an agency, its planning would be more sensitive to cumulative effects and more resistant to court challenges.

1. 42 U.S.C. §§ 4321-4361, ELR STAT. 41009.

2. NEPA § 102(2)(C), 42 U.S.C. § 4332(2)(C), ELR STAT. 41010. The Fifth and Ninth Circuits have lowered the threshold for requiring an EIS; in these circuits a determination that significant effects on the environment will in fact occur is not necessary. Plantiff need only show "that the proposed project may significantly degrade some human environmental factor." Foundation for North American Wild Sheep v. United States Department of Agriculture, 681 F.2d 1172, 1178, 12 ELR 20968, 20970 (9th Cir. 1982). See Louisiana v. Lee, 15 ELR 20609, 20611 (5th Cir. Apr. 29, 1985).

3. See 40 C.F.R. § 1508.9, ELR REG. 46032.

4. The regulations generally appear at 40 C.F.R. pts 1500-1508, ELR REG. 46001. The definition of "significantly" appears at 40 C.F.R. § 1508.27, ELR REG. 46034.

5. See 40 C.F.R. § 1508.13, ELR REG. 46032.

6. See 40 C.F.R. § 1502.20, ELR REG. 46022.

7. See the National Forest Management Act (NFMA), 16 U.S.C. §§ 1600-1676, ELR STAT. 41433, and its implementing regulations, 36 C.F.R. pt. 219.

8. For example, a national forest may have a Timber Management Plan EIS. If one timber sale is planned, then the Forest Service will do a site-specific sale-area EA, which will not duplicate the general analysis already done in the EIS. The courts have upheld this procedure. See Ventling v. Bergland, 479 F. Supp. 174 (D.S.D.), aff'd, 615 F.2d 1365 (8th Cir. 1979). Other courts have upheld the procedure in dicta while failing to uphold the agency's FONSI conclusion on the facts at issue. See Minnesota Public Interest Research Group v. Butz, 498 F.2d 1314, 4 ELR 20700 (8th Cir. 1974); Oregon Environmental Council v. Kunzman, 714 F.2d 902, 13 ELR 20901 (9th Cir. 1983).

9. NEPA § 102(2)(C), 42 U.S.C. § 4332(2)(C), ELR STAT. 41010.

10. See, e.g., Foundation for Economic Trends v. Heckler, 756 F.2d 143, 15 ELR 20248 (D.C. Cir. 1985).

11. See, e.g., National Audubon Society v. Hartz Mountain Development Corp., 14 ELR 20724 (D.N.J. Oct. 24, 1983).

12. See, e.g., Fritiofson v. Alexander, 14 ELR 20266 (S.D. Tex. Mar. 2, 1984).

13. See e.g., California v. Block, 609 F.2d 753, 13 ELR 20092 (9th Cir. 1982).

14. The problem is made worse for agencies by the strict scrutiny that courts often apply to NEPA decisions. The Eighth, Ninth, Tenth, and D.C. Circuits have all ruled that the decision whether to prepare an EIS is not committed to agency discretion because an EIS must be prepared for actions that significantly affect the quality of the human environment. Given the mandatory nature of this NEPA language, these circuits have applied the broader reasonableness standard rather than the narrower arbitrary and capricious standard in reviewing decisions not to prepare an EIS. See Foundation for North American Wild Sheep v. United States Department of Agriculture, 681 F.2d 1172, 1177 n.24, 12 ELR 20968, 20969 (9th Cir. 1982); City & County of San Francisco v. United States, 615 F.2d 498, 500, 10 ELR 20346, 20347 (9th Cir. 1980); Jette v. Bergland, 579 F.2d 59, 64, 8 ELR 20506, 20509 (10th Cir. 1978); Minnesota Public Interest Research Group v. Butz, 498 F.2d 1314, 1320, 4 ELR 20700, 20702 (8th Cir. 1974); Calvert Cliffs Coordinating Committee, Inc. v. Atomic Energy Commission, 449 F.2d 1109, 1114, 1 ELR 20346, 20349 (D.C. Cir. 1971).The Second Circuit, in contrast, has applied the arbitrary and capricious standard. See Henly v. Kleindienst, 471 F.2d 823, 2 ELR 20717 (2d Cir.), cert. denied 412 U.S. 908 (1972).

15. See, e.g., 40 C.F.R. §§ 1501.2, .3, .4, .7; 1502.4, .20; 1508.25, .27, .28, ELR REG. 46017.

16. 753 F.2d 754, 15 ELR 20225 (9th Cir. 1985).

17. NEPA § 102(2)(C), 42 U.S.C. § 4332(2)(C), ELR STAT. 41010.

18. Id.

19. 481 F.2d 1079, 3 ELR 20525 (D.C. Cir. 1973).

20. Id. at 1089 n.43, 3 ELR at 20530. The court recognized that a balancing process was involved and set out four specific factors to be considered in determining when an EIS is required: (1) whether the technology involved was more than speculative; (2) to what extent meaningful information on the impact of development of the program existed; (3) to what extent irretrievable commitments of resources were taking place in the program; and (4) whether the anticipated effects of the program on the environment were severe.

21. 484 F.2d 11, 3 ELR 20739 (8th Cir. 1973).

22. Indian Lookout Alliance v. Volpe, 345 F. Supp. 1167, 3 ELR 20051 (S.D. Iowa 1972), rev'd, 484 F.2d 11, 3 ELR 20739 (8th Cir. 1973).

23. 484 F.2d at 19, 3 ELR at 20743.

24. See, e.g., Trout Unlimited v. Morton, 509 F.2d 1276, 5 ELR 20151 (9th Cir. 1974) (an EIS need not be prepared to include both the initial construction of a dam and the later disposal of half the created reservoir capacity); Sierra Club v. Stamm, 507 F.2d 788, 5 ELR 20209 (10th Cir. 1974) (programmatic EIS was not required for the Central Utah Reclamation Project as long as an EIS was prepared for each independent unit of the project).

25. See Note, Program Environmental Impact Statements: Review and Remedies, 75 MICH. L. REV. 107 (1976).

26. For example, in Atchison, Topeka & Santa Fe Ry. v. Callaway, 382 F. Supp. 610, 5 ELR 20086 (D.D.C. 1974), the Army Corps of Engineers planned to rebuild and enlarge one of a series of locks on the Mississippi River. Applying the SIPI test of irretrievable commitment, as the Atchison court did, would require a programmatic EIS on the related network of locks and dams. Applying the independent significance test, however, would yield a contrary result. Although full use of the expanded lock would not occur unless all the locks were enlarged, substantial benefit would result from remodelling only the one lock in that river traffic could use a modern facility instead of the old outmoded lock to be replaced. See Note, supra note 25.

27. 498 F.2d 1314, 4 ELR 20700 (8th Cir. 1974).

28. 509 F.2d 1276, 5 ELR 20151 (9th Cir. 1974).

29. 514 F.2d. 856, 5 ELR 20463 (D.C. Cir 1975), rev'd sub nom. Kleppe v. Sierra Club, 427 U.S. 390, 6 ELR 20532 (1976).

30. 514 F.2d at 874, 5 ELR at 20470.

31. See note 20 supra.

32. See Cady v. Morton, 527 F.2d 786, 795, 5 ELR 20445, 20448 (9th Cir. 1975); Sierra Club v. Stamm, 507 F.2d 788, 792-93, 5 ELR 20209, 20211 (10th Cir. 1974); Sierra Club v. Callaway, 499 F.2d 982, 987, 4 ELR 20731, 20733 (5th Cir. 1974); Minnesota Public Interest Research Group v. Butz, 498 F.2d 1314, 1322-23 n.29, 4 ELR 20700, 20704 (8th Cir. 1974).

33. 427 U.S. 390, 6 ELR 20532 (1976).

34. NEPA § 102(2)(C), 42 U.S.C. § 4332(2)(C), ELR STAT. 41010.

35. The dissent noted that the statute and legislative history offer little guidance either. 427 U.S. at 420, 6 ELR at 20539.

36. 427 U.S. at 412, 6 ELR at 20537.

37. Id. at 410, 6 ELR at 20537.

38. Id. at 411, 6 ELR at 20537.

39. E.g., Aberdeen & Rockfish R.R. v. SCRAP (SCRAP II), 422 U.S. 289, 5 ELR 20418 (1975); Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 8 ELR 20288 (1978).

40. Exec. Order No. 11991, 3 C.F.R. 123 (1978), ELR REG. 45003.

41. See Andrus v. Sierra Club, 442 U.S. 347, 357, 9 ELR 20390, 20392 (1979).

42. See 40 C.F.R. § 1501.7, ELR REG. 46018.

43. 40 CFR 1508.25(a), ELR REG. 46033.

44. 14 ELR 20724 (D.N.J. Oct. 24, 1983).

45. An example is the "worst case analysis" regulation, 40 C.F.R. § 1502.22, ELR REG. 46023, which deals with how agencies should handle information gaps in an EIS. The regulation was largely ignored until the Fifth Circuit relied upon it in Sierra Club v. Sigler, 695 F.2d 957, 13 ELR 20210 (5th Cir. 1983). It has since been applied in other cases as an accepted standard. See, e.g., Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475, 14 ELR 20061 (9th Cir. 1983).

46. See Central Idaho Wilderness Act of 1980, Pub. L. No. 96-312, 94 Stat. 948 (1980).

47. Thomas v. Peterson, 753 F.2d 754, 758, 15 ELR 20225, 20226 (9th Cir. 1985).

48. Id. at 758, 15 ELR at 20226.

49. 40 C.F.R. § 1508.25(a)(1), ELR REG. 46034.

50. 40 C.F.R. § 1508.25(a)(2), ELR REG. 46034.

51. 753 F.2d at 758, 15 ELR at 20227.

52. Id.

53. Id. at 759, 15 ELR at 20227.

54. Id.

55. Discussed supra in text accompanying note 28.

56. 753 F.2d at 759, 15 ELR at 20227, citing Trout Unlimited v. Morton, 509 F.2d 1276, 1285, 5 ELR 20151, 20154 (9th Cir. 1974).

57. 753 F.2d at 760, 15 ELR at 20228, citing 40 C.F.R. § 1501.2, ELR REG. 46017.

58. 753 F.2d at 760, 15 ELR at 20228.

59. 521 F.2d 661, 667-76, 5 ELR 20633, 20636-40 (9th Cir. 1975) (EIS for road must analyze the impacts of industrial development that the road is designed to accomodate).

60. 753 F.2d at 760, 15 ELR at 20228. Though not specifically relying on the "connected action" CEQ regulation, the court added that "where agency actions are sufficiently related so as to be 'connected' within the meaning of the CEQ regulations, the agency may not escape compliance with the regulations by proceeding with one action while characterizing the others as remote or speculative." Id.

61. Id. at 760-61, 15 ELR at 20228. The court also noted that the agency's own affidavits showed that the agency was preparing the EA on at least one of the sales at the same time that it was preparing the EA on the Jersey Jack road.

62. See CEQ, Memorandum on Kleppe v. Sierra Club and Flint Ridge Dev. Co. v. Scenic Rivers Ass'n of Okla. (Sept. 1976) reprinted in Monitor, Sept. 1976, at 11, 21.

63. This is what the agency did in Hartz Mountain, discussed supra in text accompanying note 44. The Supreme Court in a Kleppe footnote implicitly approved of incremental approaches by noting that an agency could approve a pending project covered by an EIS, then take into account its impacts when preparing a programmatic EIS on the cumulative impact of the remaining proposals. 427 U.S. at 414 n.26, 6 ELR at 20538.

64. Thomas, 753 F.2d at 758, 15 ELR at 20226, citing Kleppe, 427 U.S. at 409-10, 6 ELR at 20536-37.

65. 427 U.S. at 409-10, 6 ELR at 20536-37.

66. Id. at 408, 6 ELR at 20536.

67. Only two other reported cases have relied on section 1508.25 of the CEQ regulations to require an EIS. In Foundation on Economic Trends v. Heckler, 587 F. Supp. 753, 14 ELR 20467 (D.D.C. 1984), plaintiffs sought a preliminary injunction against release of recombinant DNA into the environment until the National Institutes of Health (NIH) complied with NEPA. Citing the regulation, the court found that the authorization of separate experiments using the same novel technology are "connected" and potentially cumulative, justifying an EIS of commensurate scope. On appeal, the D.C. Circuit agreed that a programmatic EIS would be "helpful," but ruled that preparing such an EIS was not the only route to NEPA compliance open to NIH. Foundation on Economic Trends v. Heckler, 756 F.2d 143, 15 ELR 20248 (D.C. Cir. 1985).

In National Wildlife Federation v. United States Forest Service (the Mapleton case), 592 F. Supp. 931, 14 ELR 20755 (D. Or. 1984), the court held that the various environmental documents the Forest Service had prepared relating to timber sales in the Mapleton Ranger District failed to discuss cumulative impacts as required by § 1508.25. The case is discussed in detail infra in text accompanying notes 84-86.

In another district court case, Fritiofson v. Alexander, 14 ELR 20266 (S.D. Tex. Mar. 2, 1984), the court required an EIS covering cumulative effects, but relied on the CEQ definitions of "significantly" and "cumulative impact," 40 C.F.R. § 1508.7 & .27(b)(7), ELR REG. 46032, 46034. See infra text accompanying notes 81-83.

Colorado River Indian Tribes v. Marsh, 605 F. Supp. 1425 (C.D. Cal. 1985) is a recent example of a court requiring an EIS covering cumulative effects by relying on the CEQ definitions of direct, indirect and cumulative imports, 40 C.F.R. §§ 1508.8 and .7. In striking down the Army Corps of Engineers EA for a permit to place riprap along a riverbank, the court concluded that it may have been reasonably foreseeable that placement of the ripraps was just a stepping stone to major development in the area, and thus the Corps failed to consider the cumulative impacts associated with the bank stabilization project.

68. 605 F. Supp. 107, 15 ELR 20608 (D. Mont. 1985).

69. 717 F.2d 1409, 13 ELR 20888 (D.C. Cir. 1983).

70. 605 F. Supp. at 109, 15 ELR at 20608.

71. See generally Village of False Pass v. Clark, 733 F.2d 605, 14 ELR 20398 (9th Cir. 1983) (discussing the problems of early environmental analysis of off-shore leasing). Before preliminary drilling is done, geologists cannot say with better than about 15 percent accuracy whether a given site will be worth development. Remarks of Fernando Blackgoat, Rocky Mountain Oil and Gas Ass'n, PROCEEDINGS OF A SYMPOSIUM ON WORST CASE ANALYSIS 70 (L. Garrett & W. Covington, N. Ariz. U., eds. 1985).

72. 40 C.F.R. 1508.27(b)(7), ELR REG. 46034.

73. Id. at 1508.7, ELR REG. 46032.

74. As noted earlier, an EA should be used to determine whether an EIS is required. Thus, the EA must be comprehensive enough to provide the necessary information for the decisionmaker to make a reasoned choice among alternatives. And, by implication, cumulative impacts analysis and possibly worst case analysis may be necessary information. The Ninth Circuit in both Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark (SOCATS), 720 F.2d 1475, 14 ELR 20061 (9th Cir. 1983), and Save Our Ecosystems v. Clark (SOE), 747 F.2d 1240, 15 ELR 20035 (9th Cir. 1984), recognized this distinction. In SOCATS the court noted that "the EA must support the reasonableness of the agency's decision not to prepare a new or supplemental EIS," and that "the label of the [NEPA] document is unimportant. We review the sufficiency of the environmental analysis as a whole." 720 F.2d at 1480, 14 ELR at 20063. Similarly, the SOE case recognized correctly that many agencies blur the distinction between the EA and the EIS, by issuing an EA/FONSI which narrowly analyzes the environmental impacts of a project, as a time-saving device that the agencies believe complies with NEPA, while avoiding the burden of preparing an EIS. Therefore, the SOE court noted, when an EA becomes the functional equivalent of an EIS, it is subject to the same procedures as is an EIS. 747 F.2d at 1247, 15 ELR at 20039.

75. A good discussion of this premise is found in Note, supra note 25.

76. Note that the geographic approach is not coextensive with the two traditional tests for scope. Neither is it limited to actions that will result in an irreversible commitment of resources nor does it exclude actions that have an independent significance. In determining the scope of environmental analysis under NEPA, the independent significance test should not be coextensive with cumulative effects analysis under the geographic approach. It may be correct to conclude that a project which has no independent significance is too narrow to be evaluated separately, but it is erroneous to conclude that a project's environmental impacts can be adequately evaluated solely because the project has independent significance. This step-by-step analysis of related projects, which may have significant cumulative effects, is the inherent faulty logic of the independent significance test.

77. SIPI, 481 F.2d at 1087, 3 ELR at 20529, citing CEQ, Memorandum to Federal Agencies on Procedures for Improving Environmental Impact Statements (May 16, 1972), 2 ELR 46162, 46164.

78. 427 U.S. at 411, 6 ELR at 20537.

79. See Note, The Scope of the Program EIS Requirement: The Need for a Coherent Judicial Approach, 30 STAN. L. REV. 767, 798 nn. 156-58.

80. 524 F.2d 79, 5 ELR 20640 (2d Cir. 1975).

81. 14 ELR 20266 (S.D. Tex. Mar. 2, 1984).

82. Id. at 20268.

83. Id.

84. 592 F. Supp. 931, 14 ELR 20755 (D. Or. 1984).

85. The court distinguished Ventling v. Bergland, 479 F. Supp. 174 (D.S.D.), aff'd, 615 F.2d 1365 (8th Cir. 1979), which had upheld the tiering of four timber sale EAs to a programmatic EIS in the Black Hills National Forest. In that case, the forest was relatively homogeneous and the same environmental characteristics were examined in the sale areas and the overall EIS. In contrast, the Mapleton District had the greatest concentration of high-risk landslide-prone land in the Siuslaw, with the greatest potential for loss of anadromous fish habitat of any district in the forest.

86. Although the court noted that the Forest Service could comply with NEPA by preparing a programmatic EIS and site-specific EAs, the court suggested that EAs will not always satisfy NEPA. The court suggested elsewhere in its opinion that EAs might be insufficient where the individual actions are major enough to require an EIS themselves ("the preparation of a PEIS [programmatic environmental impact statement] does not obviate the necessity of preparing a particularized impact statement for individual major actions that are components of a subject program." Mapletion, 592 F. Supp. at 940 n.18, 14 ELR at 20759, quoting Natural Resources Defense Council, Inc. v. Administrator, 451 F. Supp. 1245, 1258, 8 ELR 20415, 20420 (D.D.C. 1978) modified 606 F.2d 1261, 9 ELR 20360 (D.C. Cir. 1979)) or where the programmatic statement lacks detail or is out of date ("[A] programmatic EIS will often be insufficient as it relates to site-specific actions. This may be because it does not contain sufficient detail to satisfy NEPA requirements, or because new information is revealed subsequent to its preparation." Mapleton, 592 F. Supp. at 944, 14 ELR at 20761, quoting Sierra Club v. Block, 576 F. Supp. 959, 964, 14 ELR 20009, 20011 (D. Or. 1983)).

87. 16 U.S.C. §§ 1600-1676, ELR STAT. 41433.