One good thing about the recent case is that because there was a dissenting judge (33% of total) we get to hear “both sides of the story” in the case. It sounds like the FS said that analyzing site specific projects impacts on fish made more sense than at the programmatic level. As I’ve said before, it’s hard to analyze impacts when you don’t know if, where, or when you will do a project, nor how it will be done. Judge Smith even quoted some similar thinking from the FEIS in his dissent, which I italicized below.
To analyze it more broadly, you have to make a host of assumptions, which are pretty much predicting the future. The best available science on our predictions of what will happen suggests that we are not too good at it. If we have a bad track record, making those assumptions and doing that analysis is not particularly enlightening or useful, and to some extent is a waste of taxpayer dollars compared to analysis when the details are known. One of the problems with having this debate- “how much and why”
about analysis – through court cases, is that it is never actually debated… judges have opinions and we move on. And as I’ve pointed out before 10th Circuit Roadless and 9th Circuit SNF seem to be going in different directions about site-specificity.
I don’t know if whether a judge is “liberal” or not affects their feelings about this case, as perhaps implied by the Bee reporter. If so, though, we could do a thought experiment about different random combinations of judges and potential outcomes.
The majority did refer to this note:
The Draft was criticized by the staff of the Forest Service’s Washington Office for Watershed, Fish, Wildlife, Air and Rare Plants. The staff wrote a letter complaining that there was no discussion of the effects of the logging and logging-related activities on fish:
Aquatic and Riparian: There needs to be a discussion of the effects of the new alternatives on riparian ecosystems, streams and fisheries. It is not sufficient to dismiss these effects as within the range of impacts discussed in the  framework ․ without further analysis, given the activities proposed in Alternative S2. If the treatments [proposed in Alternative S2] will be sufficient to have their intended effect, there is a high likelihood that there will be significant and measurable direct, indirect and cumulative effects on the environment, which need to be analyzed and disclosed in this document.
To me, the depth of discussion in a programmatic document is really a NEPA question. If every time people wanted more discussion of something in a document and wrote a letter, that was absorbed by the court as a legitimate viewpoint, that would lead us into a even more massive quagmire.As far as I can tell, few people agree about the appropriate level of discussion for any impact, ever. The toggle switch for documentation is inevitably set to more.
Here’s the link to the opinion.
Let’s look at what dissenting Judge Smith says:
Fourth, the majority incorrectly asserts that there is “no explanation” for the Forest Service’s decision to defer more in-depth analysis of individual fish species. See, e.g., Maj. Op. 1027. However, the Forest Service clearly did explain its reasons for deferring in depth analysis until more site-specific projects were identified. Specifically, in its Record of Decision, the Forest Service stated,
Our ability to strategically place fuel treatments for optimum effectiveness has been compromised by the set of complicated rules in the [2001 Framework]. The standards and guidelines in that [Framework] are applied at the stand level, rather than by land allocations․ Some of the rules are so detailed that they prescribe down to one acre what is allowed, and require measuring change in canopy to ten percent increments, which is not consistently practical with existing measurement tools. This fine-scale approach limits our ability to make significant progress. ․ [O]ur ability to strategically place fuels treatments on the landscape has been compromised by the complexity of rules [which allows] ․ more habitat [to be] lost to wildfire․ This decision is intended to reverse that trend.
Record of Decision at 8–9; see also Appellee’s Br. at 6. As a result, the agency explained that the 2004 EIS was being implemented to “assure the most efficient and appropriate use of government resources․” Record of Decision at 23–24. The Forest Service primarily argued not that providing more analysis would be entirely impossible, but rather that “there was insufficient information and analytic tools for a meaningful analysis․” Appellee’s Br. at 48 (emphasis added). Therefore, the majority should have concluded that it was well within the Forest Service’s discretion to determine that the benefits of deferring in-depth analysis of aquatic species to provide more meaningful analysis outweighed any delays in information.
If the Forest Service commits to a site-specific project in the future, without engaging in the required level of NEPA analysis, then Pacific Rivers might have a viable NEPA claim. Indeed, it is likely that “[t]he deficiencies noted by the” majority opinion (regarding analysis of fish) “are precisely the omissions the Forest Service will need to correct in order to comply fully with NEPA” at a later time. Block, 690 F.2d at 763; see also N. Alaska Envtl. Ctr. v. Lujan, 961 F.2d 886, 891 (9th Cir.1992) (approving a programmatic EIS that deferred detailed analysis until an application for a mining permit was submitted, but noting that “judicial estoppel precludes the Park Service from later arguing that it has no further duty to consider mitigation measures ․”).
Not only has the Forest Service affirmed many times that they plan to engage in further detailed analysis when specific projects are identified,7 but we have a legal duty to assume that the agency will perform that analysis. In Salmon River Concerned Citizens v. Robertson, we observed that courts should “assume that government agencies will ․ comply with their NEPA obligations in later stages of development.” 32 F.3d 1346, 1358 (9th Cir.1994) (quoting Conner, 848 F.2d at 1448).
B. The amount of programmatic, high-level analysis was sufficient to engage in informed decision-making regarding broad policies affecting all species, including fish.
The majority claims that the Forest Service “entirely failed to consider an important aspect of the problem” by not providing in-depth analysis regarding how the 2004 programmatic Framework would affect specific species of fish. Maj. Op. 1035 (citing Lands Council II, 537 F.3d at 987). But here, because the Forest Service chose to utilize a tiered NEPA analysis structure and implement a programmatic EIS, the relevant scope of “the problem” is whether the Forest Service “provide[d] ‘sufficient detail to foster informed decisionmaking.’ “ Friends of Yosemite Valley, 348 F.3d at 800 (quoting Lujan, 961 F.2d at 890–91). As discussed above, the majority is only able to claim otherwise by ignoring the proper standard of review and refusing to defer to the Forest Service’s discretion in determining the scope of its analysis. See Kleppe, 427 U.S. at 413 (agencies have discretion to “intelligently determine the scope of environmental analysis and review specific actions [they] may take”); Friends of Yosemite Valley, 348 F.3d at 800 (“[A] reviewing court [must] focus upon a proposal’s parameters as the agency defines them”) (alteration in original omitted) (quoting Block, 690 F.2d at 761). The scope of analysis in a programmatic EIS can include considerably less detail than in an EIS analyzing a site-specific project. See, e.g., Res. Ltd., Inc. v. Robertson, 35 F.3d 1300, 1306 (9th Cir.1993); Salmon River, 32 F.3d at 1357–58; Block, 690 F.2d at 761.
Thus, under the Forest Service’s tiered-analysis approach, the 2004 EIS provides sufficient high-level standards to guide future on-the-ground decisions affecting fish. These standards generally contemplate the relevant range of potential agency action and the consequences on various habitats in the Sierra Nevada. The 2004 Framework “begins by explaining that cumulative effects were analyzed in detail for the eight alternatives considered in the 2001 Framework.” Appellee’s Br. at 50. “It then identifies activities that have occurred” since the 2001 Framework, “including soil and water resource improvements, hazardous fuels reductions, wildfire suppression,” and road construction. Id.
Specifically regarding aquatic habitats (home to fish species), the Framework notes that these are one of the most “degraded of all habitats in the Sierra Nevada,” though much of the original problem was related to “lower elevation dams and diversions.” 1 SEIS at 3. The EIS observed that “[t]he greatest effects on the [a]quatic, [r]iparian and[m]eadow [e]cosystems will generally be from either mechanical fuel treatments or catastrophic wildfires.” Id. at 12, 96. “Fires can have extraordinary effects on watershed processes and, as a consequence, significantly influence aquatic organisms and the quality of aquatic habitats in many ways.” Id. at 208 (citation omitted).
These effects include “reductions in riparian shading and altered streamflows [that] can increase stream temperatures to extreme levels,” “[f]looding, surface erosion, and mass wasting ․ due to vegetation loss,” and “increases in sedimentation, debris flows, and wood inputs may occur” as well as “[c]omplete channel reorganization.” Id.
The Forest Service weighed “tradeoffs between potential aquatic ecosystem and water quality impacts from fuel management activities (mechanical treatment and prescribed fire) and risks associated with high severity wildfires.” Id. (citation omitted). It recognized that “with respect to aquatic ecosystems, there are arguments for and against the use of fuels treatments to reduce the extent and severity of future fires.” Id. (citation omitted). After providing this analysis, the EIS determined “alternatives that lower the risk of fire and have medium levels of treatment pose the least risk to aquatic and riparian system.” Id. at 12. Therefore, by allowing increased fuels treatments, the 2004 Framework would reduce the anticipated acres burned by just over 15% from the 2001 Framework. Id. at 98.
The Forest Service recognized that this approach “pose[d] higher short-term risks to aquatic resources because it prescribes larger amounts of mechanical treatments and greater treatment intensities.” Id. at 12, 97, 215. But the Forest Service concluded that this was mitigated by the expected long-term benefits to aquatic habitats resulting from reducing wildfires. Id. The Forest Service also asserted its intent to reduce any short-term threats through objectives listed in its “Aquatic Management Strategy,” best management practices, and goals related to “landscape-level conditions” and “land allocations” that would be applied during “project level analysis.” Id. at 12, 97, 207, 210, 215. It was reasonable for the Forest Service to defer more specific analysis of the proposal’s effect on aquatic species, because “[p]otential treatment effects on aquatic, riparian and meadow ecosystems are largely a function of the amounts, types, intensities, and locations of treatments and the standards by which they are implemented.” Id. at 210.
Although the majority correctly notes that the 2004 Framework anticipates considerably more logging in the forests, the majority ignores the fact that much of that logging may never occur. For example, 214 million board feet were offered for sale on average between FY 2000–2002, but only 118 million were actually sold—approximately 55%. Id. at 174–75. Similarly, only 58% of the fuel treatments projected under the 2001 Framework were carried out in the first three years of the Framework. Id.; Appellee’s Br. at 22–23. Therefore, the Forest Service reasonably concluded that it would be inefficient to perform a detailed analysis of the impact of activities that may never take place, and the 2004 EIS contains sufficient analysis of the probable consequences of increased fuel management at the programmatic level.
The 2004 Framework identified roads as another “critical component” of the risk and benefit “tradeoffs” to aquatic species, which include fish. 1 SEIS at 209. The EIS explained that roads are just behind wildfires in their potential effect on “aquatic ecosystems and water quality in forested environments.” Id. The EIS cited studies discussing how “roads can deliver more sediment to streams than any other human disturbance in forested environments.” Id. (citation omitted). However, the studies also indicated that “surface erosion from roads can be reduced through improved design, construction, and maintenance practices,” and “[p]roper road location, drainage, surfacing, and cut slope and fill slope treatments are important in limiting effects.” Id. (citation omitted). The Forest Service explained that the proposed “modest reduction in overall road miles, and improved road conditions,” subsequently adopted in the 2004 Framework, were some of “the most important aspects of reducing risks to aquatic resources.” Id. at 215.
The Forest Service determined that, because many details of actual on-the-ground activities were yet unknown, a more detailed analysis would be appropriately conducted when specific projects were identified. For example, the EIS explained that “actual locations and miles of roadwork[will] be determined through project-level planning and analysis.” 2 SEIS at 66. Changing the location of a proposed road by just a few hundred feet could make a substantial difference in the impact it had on riparian areas and on fish. A different location might have significantly different vegetation, soil type, and topography. Changing the location could even place a road in a completely different drainage basin, potentially impacting entirely different species of fish. See, e.g., Biological Assessment for SNFPA SEIS 146, July 30, 2003 (Paiute cutthroat trout found only in 14.5 miles of streams).
The EIS explained that “road management does not vary substantially between [the 2001 Framework and the 2004 Framework]. Under both alternatives, the ․ biological effects of roads, as previously described, would be reduced across the bioregion․” 1 SEIS at 212. The EIS further noted that, under the 2004 Framework, there would be a decrease in the net miles of roads. Id. (under the 2004 Framework, “1175 miles would be decommissioned and 115 miles of new road would be constructed”). Although the miles of reconstructed roads would almost double and may have short-term impacts, reconstructed roads would be expected to “improve water quality and aquatic habitat․” Id.
The 2004 EIS also provided analysis of the effects to watersheds from on-the-ground activity that the Forest Service might permit under the Framework. The Framework explained that, as a broad-based policy, future projects should remain protective of wildlife but strive for more effective reduction of hazardous fuels. See, e.g., Appellee’s Br. at 6, 9, 36, 54. It also identified activities that have occurred since the 2001 Framework, including soil and water resource improvements, hazardous fuels reductions, wildfire suppression, and road construction. Id. at 50. Based on this information, it analyzed combined or synergistic effects of the elements of the 2004 Framework on aquatic ecosystems and species, explaining that the 2001 and 2004 Frameworks are expected to have similar effects, because both alternatives are required to meet soil quality standards. Id. at 47–48.
Similarly, the EIS addressed the impacts of grazing with sufficient detail to satisfy NEPA on a programmatic level. As with logging and road construction, the Framework calls for a flexible approach based on specific conditions, rather than a full-scale analysis at this stage. The same 2001 standards will continue to be in effect and “are expected to reduce erosion of meadows and improve aquatic habitat conditions by facilitating the growth of stabilizing vegetation along streams.” 1 SEIS at 214. The 2001 and the 2004 Frameworks primarily differ in that changes to utilization and stubble heights may be allowed in the 2004 Framework when current range conditions are “good to excellent” (and after “rigorous[ ] evaluat[ion]”). Id. Monitoring requirements under this flexible approach will “minimize[ ] differences in effects on aquatic ․ ecosystems between the [2001 and 2004 Frameworks].” Id.
Thus, after recognizing the general impact that various proposals could have on the environment and the measures that could mitigate those effects in the programmatic EIS, the Forest Service reasonably deferred the detailed analysis of future site-specific projects. Based on this analysis, the Forest Service clearly did not “entirely fail[ ]” to consider an important aspect of the programmatic analysis required to provide informed decision-making. The majority may have preferred more specific analysis about individual fish species, but such preference is not a justifiable reason under NEPA to disregard the agency’s analysis as arbitrary and capricious.
These CEQ guidelines might be also considered from the recent ones put out for public comment:
Concise NEPA Documents
Agencies are encouraged to concentrate on environmental analysis in their EAs and EISs,
not to produce an encyclopedia of all applicable information.16 Environmental analysis should
focus on significant issues, discussing insignificant issues only briefly.17 Impacts should be
discussed in proportion to their significance, and if the issues are not deemed significant there
should be only enough discussion to show why more study is not warranted.18 Scoping,19
incorporation by reference,20 and integration of other environmental analyses21 are additional
methods that may be used to avoid redundant or repetitive discussion of issues.22
All NEPA environmental documents, not just EISs, should be written in plain language,23
follow a clear format, and emphasize important portions of the impact analysis over mere
background material. Clarity and consistency ensure that the substance of the agency’s analysis
is understood clearly, avoiding unnecessary confusion or risk of litigation that could result from
an ambiguous or opaque analysis. The CEQ Regulations indicate that the text of a final EIS that
addresses the purpose and need, alternatives, affected environment, and environmental
consequences should normally be less than 150 pages and a final EIS for proposals of unusual
scope or complexity should normally be less than 300 pages.24
In light of the growth of environmental requirements since the publication of the CEQ
Regulations, and the desire to use the EIS to address, via integration, those requirements, it is
recognized that there will be a range of appropriate lengths of EISs. Nevertheless, agencies
should keep EISs as concise as possible (continuing to relegate relevant studies and technical
analyses to appendices) and no longer than necessary to comply with NEPA and the other legal
and regulatory requirements being addressed in the EIS, and to provide decision makers and the
public with the information they need to assess the significant environmental effects of the action
under review. Length should vary with the number, complexity and significance of potential
Incorporation by Reference
Incorporation by reference is another method that provides efficiency and timesaving
when preparing either an EA or an EIS. The CEQ Regulations direct agencies to incorporate
material into an EIS by reference to reduce the size of the EIS and avoid duplicative effort. An
agency must cite the incorporated material in an EIS and briefly describe the content. An
agency may not incorporate any material by reference in an EIS unless the material is reasonably
available for inspection by potentially interested persons within the time allowed for comment.64
The goal should be to conduct concurrent rather than sequential processes whenever
appropriate. In situations where one aspect of a project is within the particular expertise or
jurisdiction of another agency an agency should consider whether adoption or incorporation by
reference of materials prepared by the other agency would be more efficient.