In Environmental Practice 12(2), I discussed the mistakes in the Council on Environmental Quality's (CEQ's) regulations. In this second installment, I discuss the gaps in CEQ's regulations. Should the CEQ's regulations once again be considered for revision or reform or amendment, I suggest these mistakes and gaps as candidates for a starting place.

A gap could be defined as a lost opportunity. There once was an opportunity to provide guidance, but that opportunity was passed. This leaves a gap. Some gaps lead to mistakes and, when they do, there would be additional justification to fill the gap. There is no particular ranking or order. This is an abbreviated list; for a longer list, please contact the author.

1. GAP: Failure to require the scope of the FONSI to encompass all alternatives in the EA

Why it's a gap: Agencies prepare an environmental assessment (EA) with two, three, four, or five alternatives. Agencies write a finding of no significant impact (FONSI) on one of them. At that point, only that one alternative is eligible to be selected by the decision maker at the time of the decision. Similarly, agencies prepare an EA or environmental impact statement (EIS) with two, three, four, or five alternatives—but consult with the listing agencies on only one, and consult with the State Historic Preservation Officer on only one. At the time of decision, only that one is really eligible to be selected.

To fix the gap: It must be clear that the alternatives considered in the National Environmental Policy Act (NEPA) process will actually be available to the decision maker at the time of decision in order to preserve the integrity of the NEPA process. Either trim the number of alternatives in the NEPA document, if they are not actually to be considered at the time of decision, or enlarge the number in the FONSI and in consultation, if they are. Either way, alternatives in the public and agency review process should match alternatives in the real decision-making process.

2. GAP: Failing to specify the process when a supplement is not required: the FONSCACI

Why it's a gap: We know what to do when there are “significant” changes in actions, circumstances, or information after a draft or final EIS is published. We prepare a “supplement … in the same fashion … as a draft and final statement” [40 CFR 1502.9(c)(4)]. But the regulations don't tell us what to do when there are changes, and the changes are found to be “not significant.” The regulations are silent on the obverse procedure.

To fix the gap: We can apply straightforward administrative law principles to fashion the kind of finding that would have to be made to fit this situation. It would be a finding of no significant changes to actions, circumstances, or information (FONSCACI,). If an agency can make this finding, then no supplement would be needed. This is parallel to the FONSI, a finding an agency would make not to prepare an EIS in the first instance.

3. GAP: Failure to account for the possibility of a supplement to an EA

Why it's a gap: The CEQ makes no reference whatsoever to a supplemental EA. But case law does. There are now a number of cases deciding the point of whether an agency should have supplemented an EA in light of new information or changed circumstances.

To fix the gap: We need to know the threshold for a supplement to an EA. This would be a threshold somewhere between zero (and thus clearly no supplement) and significance (and thus the trigger to move to an EIS). Bigger than zero, but less than significant? That would be not significant, and no supplement would be prepared. What lies between “not significant” (no supplement) and “significant” (move to an EIS)?

4. GAP: Failure to account for the possibility of a revised draft EIS

Why it's a gap: The CEQ provides for a “supplement” to either a draft or final EIS, but the Environmental Protection Agency (EPA) recognizes in their EIS rating system that a draft EIS can be so lacking in important content that only a new draft—a revised draft EIS—can satisfy public review requirements. Case law also recognizes the necessity for a revised draft EIS. This concept never appears in the CEQ's regulations.

To fix the gap: We need to know the threshold for a revision, as well as for a supplement (“significance”). Dealing with the issue directly in the regulations would make the regulations a better match to actual NEPA practice today.

5. GAP: Failure to secure and then promote the advantages of an EIS process over an EA/FONSI process

Why it's a gap: It is a common perception—but an erroneous one—that an EA/FONSI process is easier than an EIS process. No one is counting, but we see perhaps 100 EAs for every EIS. In today's practice, an EA is intrinsically no easier to write than an EIS. And then the FONSI would be an additional task. The EA/FONSI comprises two tasks, whereas the EIS is one task. What makes the EA/FONSI more attractive is the artificial advantage it has been given over the EIS, which is saddled with mandatory Federal Register notices and minimum timelines (40 CFR 1506.10). Many agencies require that an EIS be circulated through a loop in Washington, DC, whereas an EA/FONSI can be prepared and published locally. In essence, the CEQ and the agencies have made it artificially harder to prepare an EIS than an EA/FONSI—everything else being equal. Yet the EA/FONSI is the riskier procedure because of the appeals and lawsuits over failure to prepare an EIS. Thus, the CEQ and the agencies have made the riskier procedure apparently easier than the less risky one. There is nothing in the CEQ's regulations to guide us on the advantages of an EIS.

To fix the gap: Advertise the advantages of an EIS:

• • An EA and an EIS are essentially equivalent in today's practice, but with an EIS you don't have to write the FONSI (i.e., demonstrate that the impacts of the proposed action are not significant.

• • You don't have to mitigate a project into insignificance (i.e., you can actually take actions with significant consequences).

• • You won't be appealed or sued for not writing an EIS (i.e., you may gain substantial advantage in any subsequent appeal or litigation).

• • If you write an EA/FONSI and significant new information comes along, you'll have to write an EIS anyway; but if you already have an EIS, all you have to write is a supplement (i.e., you come closer to a bulletproof NEPA process).

• • The “flyspeck” cases are mostly EIS cases (i.e., courts tend not to flyspeck an EIS for adequacy, but seem plenty willing to do so for EAs and FONSIs).

6. GAP: Failure to account for operational decisions

Why it's a gap: Imagine a decision to build a house. Even if the decision includes details about the location, style, size, and major items such as roofing and siding, many more decisions remain. Scores, maybe hundreds, of decisions have yet to be made about materials, colors, features, trim, and such items as the location of electrical switches and the routing of wires and pipes. These minor decisions could be called operational decisions or implementing decisions. They are the additional decisions necessary at the operational stage even after the major decisions are made to build a house of a certain design at a certain location. They are necessary to get the job done, but are operational in the sense they merely carry out the decision to build a house. Federal projects may be no different. The decision to proceed with a timber sale, or road project, or grazing lease, etc., must necessarily be followed by operational decisions to augment or implement the major decision to proceed. So long as these operational decisions are within the scope of the major decision, they fly under the NEPA radar. But sometimes they do not. Sometimes there is an expectation that these additional minor decisions should themselves be subject to a NEPA process under a theory of tiering or supplementation. What is lacking is a clear understanding of what is operational and thus outside the ambit of NEPA, and what is supplementational and thus should be the subject of a tiered or supplemental NEPA document. There is currently no regulation an agency can cite as to why it believes a given operational decision is outside the ambit of NEPA.

To fix the gap: There should be an explicit provision that recognizes the existence of additional operational decisions that are outside the ambit of NEPA. These would be the supplemental or implementing decisions necessary to carry out the decision made pursuant to a NEPA process, within the scope of that decision but not subject to any further NEPA process. Agencies could then rely on this explicit provision to defend against allegations that operational decisions should themselves be subject to an additional NEPA process.

7. GAP: Failure to provide for a compound EA or EIS

Why it's a gap: Whether it is for perceived efficiency or streamlining or for other reasons, agencies sometimes toss a bunch of proposed actions into a single EA or EIS. These could be actions related to each other in terms of geography or timing, such as a number of projects in the same watershed or in the same fiscal year or in the same political jurisdiction. These could be called compound EAs and EISs. There could be chapters within an EA or EIS book for each of the compound actions. Each chapter would consist of a proposal for action, the need for it, and alternatives. At the conclusion of the process, the record of decision (ROD) would reflect that same compound setup and record the same number of decisions. There is nothing in the CEQ's regulations about any of this. There is advice at 40 CFR 1508.25(a)(3) that agencies should analyze similar actions (such as those with common timing or geography) when it is “the best way to assess adequately the combined impacts.” But nothing is said about how these similar actions would be packaged inside the EA or EIS. Sometimes there are no “combined impacts.” Without explicit guidance, agencies are tempted to toss together a bunch of proposed actions into a single proposal—in which case the need for action, as well as alternative actions, becomes highly problematical.

To fix the gap: There should be an explicit provision that recognizes the existence of compound EAs and EISs for when agencies choose to lump together a set of proposed actions with common timing or geography or on any grounds of streamlining or efficiency. A compound EA or EIS would separate out each of the needs to which the agency is responding with each of the proposals, and show the alternatives (including the proposed action alternative) for meeting each need. In this way, an agency would better prepare the administrative record that would better support a compound decision-making process.

8. GAP: Failure to provide for relevance as a scoping screen

Why it's a gap:Scoping is a term of art defined as the “process for determining the scope of issues to be addressed and for identifying the significant issues related to a proposed action” (40 CFR 1501.7). Among other things, the agency is to “identify and eliminate from detailed study the issues which are not significant or which have been covered by prior environmental review [1506.3], narrowing the discussion of these issues in the statement to a brief presentation of why they will not have a significant effect on the human environment or providing a reference to their coverage elsewhere” [1501.7(a)(3)].This language recognizes just two grounds to eliminate a matter during scoping: (1) it is not significant and (2) it is covered elsewhere. But relevance is surely an important screen. The regulations recognize relevance in 40 CFR 1502.9(c)(1), where changes to the proposed action or new circumstances or information may trigger a supplement if they are “relevant to environmental concerns.”

To fix the gap: Add the same provision for relevance to 1501.7 as currently exists in 1502.9.

9. GAP: Failure to be explicit about that notice-and-comment thing for EAs

Why it's a gap: When should there be a draft EA? What is the standard for notice and comment on an EA before a FONSI can be prepared? Should both the EA and FONSI first be prepared before public notice and comment? The Ninth Circuit created its own standard: “An agency, when preparing an EA, must provide the public with sufficient environmental information, considered in the totality of circumstances, to permit members of the public to weigh in with their views and thus inform the agency decisionmaking process” [Bering Strait Citizens for Responsible Resource Development v. US Army Corps of Engineers, 511 F.3d 1011, 1024–26 (9th Cir. 2008)]. When it comes to a handbook or guidance manual, and in jurisdictions where the Ninth Circuit does not make precedent, what should an agency specify? This is a gap. The closest that the regulations come to answering the question is 40 CFR 1506.6(a) (“Make diligent efforts….”) and (c) (“Hold or sponsor public hearings or public meetings whenever appropriate….”). Agencies now have to decide when it would be not appropriate to involve the public in the threshold question of whether to prepare an EIS, while at the same time it is mandatory to involve the public in the creation of a categorical exclusion and in the preparation of an EIS. It seems odd, considering how dominant the EA/FONSI process has become, that the matter would not be fully resolved in the regulations.

To fix the gap: Either way, decide the point. In the alternative, define the words “diligent” and “appropriate” in the context of a draft EA.

10. GAP: Failure to define “extraordinary circumstances”

Why it's a gap: An action that fits a category of actions that has been categorically excluded can normally be excluded from an EA or EIS procedure, unless there are extraordinary circumstances. Agency procedures for implementing NEPA “shall provide for extraordinary circumstances” (40 CFR 1508.4). What is an “extraordinary circumstance”? Some agencies have taken to defining this as, simply, a significant impact. In other words, an action that fits a categorical exclusion normally will not be subject to an EA or an EIS unless it may have a significant impact. But actions in these categories have already been found not to have significant impacts individually or cumulatively. Making this additional finding could impeach the category because it would infer that the original finding was not made reliably. Moreover, if an action may have significant impacts, an EIS is necessary so the possibility of significant impact is not a proper test to determine whether an EA should be prepared.

To fix the gap: An extraordinary circumstance is one that is not ordinary. An ordinary circumstance is one that had been taken into account when the category was established. For example, if 18 circumstances were taken into account at the time the category was established, then these 18 circumstances become ordinary. Thereafter, a proposed action having only these 18 circumstances—or fewer—will have no extraordinary circumstances. If there were a 19th relevant circumstance, that would be an extraordinary one, and an EA or EIS would have to be prepared because it is a circumstance that had not been already taken into account. This would make maximum use of the record created at the time categorical exclusions are established. And it would be a literal and plain-language application of the term “extraordinary circumstance.”

11. GAP: Failure to define “appropriate alternatives … in any proposal which involves unresolved conflicts over alternative uses of available resources”

Why it's a gap: This is NEPA 102(2)(E). It is the principal reason there are alternatives in an EA. Alternatives in an EIS derive principally from NEPA 102(2)(C)(iii). EAs are to study, develop, and describe: “alternatives as required by section 102(2)(E)” [40 CFR 1508.9(b)]. Are these the same, ultimately, as alternatives to the proposal as is required for an EIS? Are these the same, ultimately, as mitigation measures not already incorporated into the proposed action, as is required by 40 CFR 1508.25(b)(3)? Does the no-action alternative cover this requirement? Are there “conflicts” other than “unresolved conflicts”? Does the word “conflict” in this phrase have the same meaning as the word “controversy” mentioned five times [e.g., 40 CFR 1508.27(b)(4)]? It is easy to understand that NEPA 102(2)(E) is the source for alternatives in an EA. What is not so easy to understand is what it means, and there is no help in the CEQ's regulations.

To fix the gap:Option 1: A regulatory definition that requires alternatives in an EA—as would be necessary for an EIS—would cover it. This would explicitly and finally erase any possible difference between the content of an EA and an EIS, however. Option 2: The purpose of an EA is to establish the administrative record for a finding of whether the environmental consequences of the proposed action are “significant.” There are no alternative actions in an EA, except as required by NEPA 102(2)(E), for which an unresolved conflict is defined as a disagreement over resource use that is science based, not policy based, and not solely the product of opposition to the proposed use of a resource. An appropriate alternative in this context is an alternative action that would meet the same underlying need as the proposed action while also reflecting a different approach to resource use that has fewer adverse environmental consequences or adverse environmental consequences of less intensity.

12. GAP: Failure to define the no-action alternative

Why it's a gap: As originally intended, the no-action alternative was doubtless meant to be the null case to form a comparison or benchmark or baseline for the proposed action alternative. Action is proposed in order to meet a need for action, we must assume. Other action alternatives also meet the need, or they wouldn't qualify as alternatives. Thus, meeting the need (taking the proposed action or alternative action), versus leaving the need unmet (what could be the null case), would form a sharp and clear basis for comparison. If only it were kept this simple. Instead, what we have is a bare requirement for a no-action alternative [40 CFR 1508.25(b)(1)] without definition. There are two interpretations in the 40 Questions, but both of them involve action. This leaves the no-action alternative a misnomer, at the very least, because it always involves action. NEPA the statute says nothing about a no-action alternative. No action is purely an invention of the CEQ's regulations. So it could have been an invention that capitalizes on the fact that action is proposed in order to meet a need, and the null case for purposes of comparison would leave that need unmet

To fix the gap: The no-action alternative should be the null case for purposes of comparison. Action is proposed in order to meet a need. Leaving that need unmet is the null case. This would be the truest possible basis on which to compare taking action (meeting a need) to not taking action (leaving that need unmet).

13. GAP: Failure to recognize in the regulations all the good legal reasons not to prepare an EIS, instead only recognizing the lack of significant impact

Why it's a gap: The courts recognize eight distinct legal theories for not preparing an EIS, while the CEQ recognizes only one—the lack of significant impact. This is the one used for the EA/FONSI [the lack of significant impacts (40 CFR 1508.13)], as well as the categorical exclusion [a class of actions found to have no significant effects (40 CFR 1508.4)]. There are seven more legal theories. Agencies aren't getting the guidance or the tools they need; NEPA the statute is not being represented accurately.

To fix the gap: The threshold for an EIS should be reoriented to all eight legal theories: … to the fullest extent possible … federal… proposals … for plans, programs, functions, and resources … significantly affecting … the quality of the human environment … require an EIS unless Congress changes the rules … or environmental consequences are outside the United States… .

14. GAP: Failure to recognize in the regulations all the good legal reasons to scope an action out of an EA or EIS

Why it's a gap: The CEQ lists three kinds of actions that “agencies shall consider”: connected, cumulative, and similar [40 CFR 1508.25(a)]. That would be the half-full part of the glass. There is also the half-empty part that the CEQ does not mention. These are all of the rules recognized in NEPA case law for why an agency may safely leave out an action even though it may be connected, cumulative, or similar. The rules are not listed on these pages, but include rules such as independent justification, practicality, and relevance. Thus, even though an action may be connected, cumulative, or similar, an agency may still safely leave it out of the scope of an EA or EIS thanks to judge-made law. Not every action that is connected, cumulative, and similar is thereby relevant to a given decision. Practicality may require that a large EA or EIS be segmented into a smaller, more manageable scope.

To fix the gap: NEPA practitioners need the rules for both the half-full and the half-empty parts. To be a reliable guide for NEPA practice, the CEQ's regulations should include all the rules for what to put in as well as what may properly be left out.

15. GAP: Failure to recognize in the regulations all the good legal reasons to scope an alternative out of an EA or EIS

Why it's a gap: The courts recognize a number of distinct legal theories for not including a given alternative in an EA or EIS, even though the alternative may meet the same underlying need (which may be expressed as a goal or purpose or objective or even purpose and need) as the proposed action. A complete list of reasons to safely leave out such an alternative from detailed analysis and comparison is not given on these pages, but include rules such as feasibility, practicability, economy, redundancy, lack of environmental protection, and lack of legal authority. Thus, even though an alternative by some measure may be a “reasonable course of action” [40 CFR 1508.25(b), 1502.14(c)], there may be good legal reasons to safely eliminate some of these from detailed comparison. True, the CEQ's regulations provide that agencies may give such reasons [40 CFR 1502.14(a)], but the CEQ's regulations do not say what such reasons may be.

To fix the gap: NEPA practitioners need the rules for both including alternatives and safely leaving them out of detailed analysis and comparison. To be a reliable guide for NEPA practice, the CEQ's regulations should include all the rules for what to put in as well as what can safely be left out.

16. GAP: Failure to define whether it is possible to find not significant an increment to a significant problem and, if so, how

Why it's a gap: Global warming (climate change) is exhibit A. If a proposed action would yield a net change in atmospheric greenhouse gas, and thus presumably make some infinitesimal contribution to global climate change, could it still be found to be not significant? Extinction is exhibit B. If a proposed action would result in the “take” of an endangered species, and thus presumably contribute to the likelihood of its extinction, could it still be found to be not significant? The question arises because, on the one hand, the increment caused by the proposed action is very small indeed, but, on the other hand, the matter in question is very large indeed. In other words, the intensity is very small, but the context is very large. The usual practice for finding nonsignificance is to step up the scale. Yes, the housing development will eliminate habitat, but the functioning of the ecosystem remains intact (moving from the scale of the site to the scale of the ecosystem). Yes, the proposal will take individuals, but the viability of the population remains (moving from the scale of the individual to the scale of the population). But when the scale is already global, as it is for climate change and extinction, there is no reasonable larger scale to bump to. The gap is that the NEPA regulations simply do not speak to this point. Is it possible to rationalize nonsignificance in this situation, or is it not?

To fix the gap:Option 1: No, it is not possible to rationalize nonsignificance when there is a net increase in greenhouse gas emissions, for example, because this situation is inherently one where a significant consequence has come from a very large number of individually insignificant consequences. Every such proposal shall be the subject of an EIS. Option 2: Yes, not significant in this context is defined as an increment that is individually very small, even though it may be significant in the aggregate, where all practicable measures have been adopted to minimize the individual increment.