Read the comments.

The wildfire analysis establishes, from the agency's own data, that road access drives ignitions. P. 94 states that "the incidence of human-caused fires generally increases with proximity to roads," and it discloses its basis: the InFORM Fire Occurrence Data Record and the MTBS burned-area dataset, with primary comparisons focused on 2014-2024. P. 97 it reports the result - "ignition density is approximately four times greater on other NFS lands compared to the potentially affected IRAs and wilderness. Most of this difference is due to the greater density of human-caused ignitions within other NFS lands (Table 21)." Among fires with causes determined, 77 percent in wilderness and 74 percent in potentially affected IRAs were naturally ignited, against a 57 percent human-caused majority on other NFS lands. Aplet, Hartger & Dietz (2026) find ignition density lowest in designated wilderness (1.75 fires per 1,000 ha) and Inventoried Roadless Areas (1.97), and highest within 50 m of roads (7.99). The cause-and-effect link to this analysis is direct. Alternatives 2 and 3 remove the prohibition on road construction and reconstruction across the potentially affected IRAs. The DEIS's own measured relationship - human-caused ignition density rising with road proximity, by a factor of about four between unroaded and roaded NFS lands - predicts that converting unroaded acres to roaded acres increases human-caused ignitions on those acres. The DEIS has the data to estimate its magnitude. It does not do so. In the Fire Control discussion of Alternative 2 at p. 111, the DEIS analyzes only the benefit: "With increased road access, initial attack success rates would be expected to increase. Additional roads would provide safer, more effective, and more efficient containment opportunities, potentially reducing fire sizes and durations." No corresponding estimate of the ignition increase appears anywhere in the Alternative 2 effects analysis. The ignition side surfaces once, at p. 113, in the Alternative 3 discussion: "Alternative 3 has the same concern about new roads increasing ignitions, and these ignitions would be closer to communities, where they could potentially cause more damage. However, the net impact of new roads on fire occurrence is likely small or insignificant, given the potential for improved response times and increased initial attack success." That sentence carries no citation, no analysis and is a netting conclusion with one side of the ledger measured to a factor of four and the other side never estimated at all. The offset the DEIS relies on is bounded by its own finding. At p. 104 the DEIS reports that "[d]ata from the FODR for fires occurring between 2014-2024 indicate that initial attack success rates are high (more than 90 percent) for all land designations," with the differences across designations described as a trend rather than a gap (Table 24). If initial attack already succeeds on more than nine of ten fires everywhere, the improvement new roads can deliver is bounded at under ten percentage points of containment on fires that start. The DEIS performed the ignition-density comparison at the national scale, across all potentially affected IRAs, in support of a national rulemaking. Having quantified the baseline nationally, the agency cannot defer the corresponding effect to future site-specific analyses while relying on an unsupported national conclusion that the net effect is insignificant. The statement at p. 112 that "[t]he magnitude of these beneficial effects would depend on site-specific decisions regarding the amount and layout of the expanded road network" applies with equal force to the adverse effects, and the DEIS estimates neither. NEPA requires the agency to disclose the basis for the environmental conclusions it draws and to connect its data to those conclusions. 42 U.S.C. Sec. 4332(2)(C); Idaho Sporting Congress v. Thomas, 137 F.3d 1146, 1150 (9th Cir. 1998). Under 7 CFR 1b.11(a)(53) this comment bears on reasonably foreseeable impacts, on the choice among alternatives, and on compliance with applicable law. I request that the Forest Service supplement the wildfire analysis to estimate, using the InFORM FODR analysis already performed for Table 21, the expected change in human-caused ignition density and ignition count within the potentially affected IRAs under Alternatives 2 and 3, given the road mileage and road-access changes the alternatives make possible; that it state that estimate in the same units as the initial-attack benefit it credits at pp. 111 and 113 so the two can be compared; and that it either support the conclusion that "the net impact of new roads on fire occurrence is likely small or insignificant" with that comparison or withdraw it from the FEIS. If the agency retains the conclusion, it should identify where in the FEIS or the supporting record the quantitative basis for it appears.

The DEIS maintains a References Cited section at pp. 249-285, but the quantitative figures that carry its comparison of effects are not traceable to it. Three examples, each verified against the document: First, the 44.7-million-acre baseline. At p. 9 the DEIS states that "[t]he 2001 Roadless Rule applies to approximately 44.7 million acres of National Forest System lands, including in Alaska." This is the denominator for every acreage comparison in the document. The only citations on that page are 66 FR 3244, attached to the rescission of the 2001 Rule, and 7 CFR 1b, attached to which NEPA procedures govern. Neither supports the acreage, and no other source for it appears. Second, the implementation figures at p. 15. The DEIS states that Forest Service data "shows an average of 500 acres per year of timber harvest and a total of 1.8 million acres of hazardous fuels reduction within IRAs managed under the 2001 Roadless Rule," and it names its source: "The best available information on activities that have been implemented within IRAs over the last decade comes from the Forest Service Activity Tracking System." Naming the system is not disclosing the figure. The Forest Service Activity Tracking System has no entry in the DEIS's References Cited, and the passage gives no query date, no reporting period beyond "the last decade," no system version, and no list of the activity codes counted as timber harvest or as hazardous fuels reduction. As disclosed, neither figure can be reproduced or checked. This matters because the 500-acre figure is the DEIS's own measure of how little timber harvest occurred under the 2001 Rule, and it sets the baseline against which the magnitude of change under alternatives 2 and 3 is characterized. Third, the road baseline in Table 3. At p. 31 the DEIS states that "[t]here are currently about 9,500 miles of roads within the potentially affected IRAs (9,400 miles of NFS roads, 100 miles of private roads) representing about 2.5 percent of the NFS transportation system." No source is cited for the mileage, for the split between NFS and private roads, or for the 2.5-percent denominator. This is the entire quantitative basis for the transportation row of the DEIS's comparison of effects across all three alternatives. These are not incidental statistics. The acreage bounds the scope of the action; the harvest and fuels figures set the baseline against which the effects of rescission are measured; the road mileage and its 2.5-percent framing are the sole quantitative support for the DEIS's characterization of transportation-system effects. An effects comparison whose baselines cannot be traced to a disclosed source cannot support the conclusions drawn from it, and the detailed statement NEPA requires, 42 U.S.C. § 4332(2)(C), is not supplied by conclusions resting on figures neither the public nor the decisionmaker can verify. An agency decision resting on factual predicates the agency has not disclosed is not the product of reasoned decisionmaking. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983); Idaho Sporting Congress v. Thomas, 137 F.3d 1146, 1150 (9th Cir. 1998). I request that, for each of the three figures identified above, the final EIS do one of the following: (a) cite the source and add a corresponding entry to References Cited; (b) where the source is an internal Forest Service data system, state the system, the query date, the reporting period, and the activity or road codes included, so that the figure can be reproduced; or (c) correct the figure if no source supports it as stated. Because the same defect recurs, I further request that the final EIS apply the same treatment to every quantitative figure appearing in Table 3 and in the Comparison of Alternatives, and that the response to comments identify where in the final EIS or the supporting record each disclosure or correction was made, as 7 CFR 1b.7(f)(3) provides. Finally, because the reasoned-decisionmaking defect described here goes to the proposed rule as well as to the environmental impact statement, I ask that this comment be addressed in the preamble to any final rule in addition to the final EIS response to comments. A response placed only in the EIS appendix would leave the rule-side objection unanswered.

Hello, I am a resident of New York State and travel around the United States to enjoy nature in all different locations. I am writing to oppose the rescinding of the 2001 Roadless Area Conservation Rule to protect biodiversity. The Draft EIS's disclosure of Endangered Species Act effects for this rule making is inadequate. The document states that "based on preliminary draft biological assessment," the action would result in a "may affect, and is likely to adversely affect" determination for 327 ESA-listed species and 71 critical habitats, with alternative 2 findings described as "similar or reduced in alternative 3," yet nowhere does the DEIS disclose whether formal consultation with USFWS or NMFS has been initiated, is proceeding, or will conclude before a Record of Decision issues. Under 16 U.S.C. § 1536, formal consultation is mandatory once a "likely to adversely affect" determination is made, and 50 C.F.R. § 402.14(a) states "if such a determination is made, formal consultation is required," with initiation requirements specified at 50 C.F.R. § 402.14(c). Given the scale of species and habitats affected, proceeding to a decision without documented consultation status creates substantial legal exposure. I request that the agency disclose current consultation status for all 327 species and 71 critical habitats, and commit that no Record of Decision will issue before Biological Opinions are completed. In addition, the USFWS Biological Assessment concedes the limits of its own inputs. Page 26: "Species occurrence data are opportunistic and do not represent the results of exhaustive structured species surveys nor do they include negative occurrence data." Page 25: "The databases do not differentiate between DPSes of vertebrate species listed under the ESA, which causes potential inflation of exposure to these entities." Where the data runs out, the assessment substitutes assumption — page 77: "Exposure data on species occurrence are insufficient for assessment so we will assume potential exposure" — and where prediction fails, it concedes the gap outright, page 376: "currently the Forest Service does not have a reliable method to identify, describe, or analyze the likelihood and location of any future projects that may be proposed for currently inventoried roadless areas." An effects analysis assembled from opportunistic records, assumed exposure, and an admitted inability to locate the action's impacts cannot support the DEIS's conclusions about species consequences. Idaho Sporting Congress v. Thomas, 137 F.3d 1146 (9th Cir. 1998) (analysis must be supported by data, not conclusory assertions). I request the FEIS disclose, species by species, which determinations rest on assumed exposure or unvetted occurrence data, and what structured survey or modeling work would cure the gaps before irreversible commitments are made.