Read the comments.

Let's not give away the future “health” of our natural lands by rescinding the Roadless Rule. Large areas need to remain safe from human pollution and chemicals (logging and mining). The environment of western Washington has already been greatly affected by human beings. There's now a strange absence of mosquitoes at popular campgrounds. Populations of frogs and salamanders have greatly decreased. These are signals we can't ignore. We need to keep Alternative 1, the Roadless Rule. Alternative 2 has many flaws. For one thing, there is an Unquantified Magnitude Finding on Stored-Carbon Loss Under Alternative 2. The Draft EIS quantifies the carbon stock at issue and then declines to quantify the effect of the action on it, resting its finding on a comparison it never performs. At p. 74 the Draft EIS states: "Birdsey et al. (2025) estimate the IRAs contain about 5 percent of the stored forest carbon in the coterminous United States (0.9B Mg C)." At p. 87, under "Implications for Forest Vegetation, Health, and Carbon," the Draft EIS concedes the mechanism: "Under alternative 2, where future road construction, timber harvest, or non-commercial forest management remove vegetation, there would be a loss of stored carbon. Areas affected by permanent tree removal activities, such as permanent road construction, would lose capacity for carbon sequestration." It then disposes of the magnitude in a single sentence: "Given forest management objectives, these areas would be small compared to the overall carbon within the forested landscape within roadless areas." That is a quantitative comparison with only one term supplied. The Draft EIS gives the denominator - 0.9 billion Mg C - and never estimates the numerator. It does not state how many acres of the 44.7-million-acre potentially affected environment are expected to be roaded, harvested, or otherwise entered under alternative 2, what carbon density those acres carry, or over what period the loss and any recovery would occur. The finding that the loss would be "small" therefore rests on no disclosed quantity at any scale, and a reader cannot tell whether the agency is describing a fraction of a percent or a tenth of the stock it has just quantified. The Draft EIS's own source also undercuts the recovery assumption offered in mitigation. I request that the Forest Service supplement the forest vegetation and carbon analysis to disclose a quantified or bounded estimate of stored-carbon loss and foregone sequestration under each alternative, using the acreage and operability data already in Table 16 and the stock estimate already at p. 74; and, if the agency retains the finding that affected areas would be "small compared to the overall carbon within the forested landscape within roadless areas," that it disclose the quantities that comparison rests on. If no such estimate can be produced, the Draft EIS should state on the record why data the agency itself compiled will not support one. An agency may not rest a finding of no material effect on a magnitude comparison whose terms it declines to supply when it holds the data to supply them. Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 43 (1983); 5 U.S.C. 706(2)(A). We need to keep Alternative 1 and make it permanent. It's vital to preserve our natural areas now. The Roadless Rule is the major baseline to keep our nation's land healthy.

I support keeping our forests and wildlife areas roadless. Do not recind protections that are in place currently. I hike in the thousands of acres in northern Wisconsin. My mental and physical health would be degraded if these places were spoiled by roads, and or other developments. These protected places are important for the overall well-being of our communities. I travel to the forest communities from 300 miles away. I spend money in these towns and if these wild areas were degraded, I would not spend tourism dollars there. Their livelihoods would suffer. Please leave our wild places wild and roadless Unsourced FY2024 Recreation Visitation Figure and Unresolved NVUM Citation at Draft EIS p. 220 At page 220 of the Draft EIS, under "Recreation Opportunities and Setting - Effects of Alternative 3 - Modified Rule - Economic Activity from Recreation in Roadless Areas," the Draft EIS states: "The 193 million acres administered by the Forest Service received approximately 164 million recreation visits in FY2024." That sentence carries no citation, and no citation anywhere in the paragraph supplies one. The paragraph's citations are attached to other propositions. White et al. 2016 is attached to the sentence immediately preceding the figure - "Some communities located around Federal recreation lands have a high dependence on economic inputs from recreation visitor spending and may experience negative economic conditions if visitation declines or visitor spending patterns change (White et al. 2016)" - which is a statement about community dependence, not about visitation counts. USDA Forest Service 2025a (National Forest and Grassland Economic Contributions, Forest Service Publication FS-1251) is attached two sentences after the figure, to "Visitor spending in FY2023 contributed approximately $16.2 billion to U.S. GDP and supported about 167,000 jobs, both full-time and part-time." Every other statistic in that paragraph is FY2023 and is drawn from that FY2023 source. The FY2024 visitation count is the only FY2024 datum in the paragraph, and a source cited for FY2023 spending and GDP cannot supply an FY2024 visit count. The same sentence appears at page 16 of the Economic Analysis, where the sentence order is different - the figure appears first and the White et al. 2016 citation two sentences after it - which confirms that no source is attached to the figure in either document. The Draft EIS also cites the National Visitor Use Monitoring program at page 220 as "(USDA Forest Service 2026a)." The string "Forest Service 2026" appears exactly once in Volume I, at that in-text citation. The References Cited section at pages 249-285 contains no entry for USDA Forest Service 2026a; its only National Visitor Use Monitoring entry reads "U.S. Department of Agriculture, Forest Service. 2024. Forest Service National Visitor Use Monitoring Data Dashboards." The Draft EIS therefore relies on an unidentified source for the one program capable of producing a national visitation figure, and the public cannot review the data the agency used. These are not questions of citation form. The FY2024 visitation figure is the quantitative foundation of the recreation-economics discussion at pages 220-222, which the Draft EIS uses to characterize the visitor spending and employment potentially affected by the rescission. A quantity asserted with no source in the record is not a fact found, and under 5 U.S.C. 706(2)(A) and Motor Vehicle Manufacturers Association v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29, 43 (1983), the agency must articulate a rational connection between the facts found and the choice made. It cannot do so on a figure the record does not support. Because this concerns compliance with applicable law, the information is substantive under 7 CFR 1b.11(a)(53) and the issue cannot be certified as comparatively not of a substantive nature under 7 CFR 1b.7(j). I request two factual corrections under 7 CFR 1b.7(f)(2)(v), each of which the agency should account for by citation under 7 CFR 1b.7(f)(3): 1. Supply the source and data year for the FY2024 visitation estimate at Draft EIS page 220 and Economic Analysis page 16, or remove the figure from both documents. If the source is the National Visitor Use Monitoring program, identify the specific NVUM release and its publication date. 2. Add a complete References Cited entry for "USDA Forest Service 2026a," or correct the in-text citation at page 220 to the source actually relied upon. Because the same unsupported figure carries the recreation baseline in the Economic Analysis that supports the proposed rule, these corrections are owed both in the Final EIS and in the final rule's explanation of the economic effects on which it relies.

We are losing roadless areas all across the country. Once they are penetrated by roads and vehicle, they are changed for all future generations. This must stop now. Road-Proximity Ignition Finding Rests on a Designation-Level Analysis and an Undisclosed Figure 15 At Draft EIS page 98, in the Wildfire Occurrence discussion, the agency states: "Consistent with other research findings, these analyses indicate that human caused ignitions increase in abundance with proximity to roads, supporting the notion that roads facilitate human access, which can in turn result in more human caused fires (Narayanaraj and Wimberly 2012, Parisien et al., 2016, Balch et al., 2017, Chen and Jin, 2022)." The Comparison of Alternatives table in Chapter 2 carries that finding into the effects comparison, stating that under alternatives 2 and 3, "if there is an increase in public road access, there could be increase the number and frequency of wildfires as human-ignitions are more likely near roads." The Draft EIS therefore treats an increase in the number and frequency of wildfires arising from road proximity as a reasonably foreseeable effect of the proposed action. The analysis the Draft EIS presents does not support that finding at the magnitude asserted, for two reasons established by the document's own text. First, the two analyses the Draft EIS documents measure land designation, not road distance. Table 18 (p. 95) and Table 21 (p. 98) both compare other NFS lands, potentially affected IRAs, and wilderness, using Interagency Fire Occurrence Reporting Module (InFORM) Fire Occurrence Data Record ignitions with a point of origin on NFS lands for 2014-2024. A contrast among land designations cannot isolate a road-proximity effect, and the Draft EIS concedes the confounding on the same page. It states that "Other NFS lands typically have more land management activities, infrastructure (such as powerlines), and public recreation that can lead to human caused ignitions," and it reports that other NFS lands have "about twice the density of natural ignitions than IRAs or wilderness, potentially due to biophysical differences or higher detection rates" (p. 98). Natural ignitions are not caused by road access. The agency's own explanation for the natural-ignition differential - biophysical difference and detection rate - applies with equal force to the human-ignition differential in Table 21, and the Draft EIS neither tests nor excludes it before attributing that differential to roads. Second, the only analysis in the Draft EIS that measures ignitions against road distance is Figure 15, "Density of reported ignitions by distance to roads and cause within the potentially affected IRAs" (p. 98). Unlike the captions of Tables 18 and 21, the Figure 15 caption identifies no data source and no period, and the figure carries none of the documentation supplied for those tables in footnotes 38 through 40 at page 94. The narrative does not reference Figure 15 or report any value from it - no distance classes, no ignition counts by cause, no exposed area, no densities. The one within-IRA road-distance analysis in the document is presented with none of the information a reader would need to evaluate it, while the two designation-level analyses that cannot answer the road-distance question are fully documented. I request that the Forest Service supplement the wildfire occurrence analysis (7 CFR 1b.7(f)(2)(iii)) to report the data underlying Figure 15 - the distance classes, ignition counts by cause, exposed area, and resulting ignition densities - and to state whether the road-distance relationship within the potentially affected IRAs survives the confounders the Draft EIS itself identifies at page 98. If it does not, the finding at page 98 and the Fire and Fuels row of the Comparison of Alternatives require factual correction under 7 CFR 1b.7(f)(2)(v). Absent one or the other, the Final EIS would assert a reasonably foreseeable effect for which it has disclosed no supporting analysis: not a cause-effect relationship established within the meaning of 7 CFR 1b.11(a)(23), and not a reasoned explanation under 5 U.S.C. Sec. 706(2)(A). See Kern v. Bureau of Land Management, 284 F.3d 1062 (9th Cir. 2002). Forum. This concern is directed to both the Draft EIS and the proposed rule. The road-proximity ignition finding supplies a factual predicate for the wildfire rationale stated in the Purpose and Need, and the Purpose and Need is the standard against which alternatives were measured for elimination from detailed study. An answer given only in the Final EIS response-to-comments appendix would leave the concern unanswered as a comment on the rule. Protect what's left. Keep the Rule.

The draft EIS states, at page 17 and again at page 304, that "[o]f the comment letters received, approximately 99 percent were generally opposed to the proposed rule recission, while approximately 1 percent were generally in support." The same paragraph at page 17 states that "[m]ore than 90 percent of the letters received contained brief sentiments or opinions on the proposed action." Both are quantified findings about the notice of intent comment record. Neither carries a citation, and the draft EIS discloses no criteria by which any letter was assigned to any of those categories. The document says the coding occurred. Appendix 3 states at page 305 that "[a]ll NOI comments were reviewed, coded, and analyzed to develop the below overarching themes, concerns, and recommendations." That is the only appearance of the word "coded" in the draft EIS. What accompanies it is a list of four things the agency did in response — modifying the proposed action or developing alternatives, using information to inform the analysis, making factual corrections, and considering but making no change — and, at footnote 72, the definition of "substantive" the agency applied in deciding which comments to summarize. Those disclose how comments were sorted for theme and for response. Neither discloses how a letter was determined to be "generally opposed" or "generally in support," how a letter expressing mixed views was assigned, how form letters and letter-generator submissions were treated, or how petition signatures bear on the percentages, when the same paragraph reports over 220,000 letters representing over 625,000 submissions. This bears on the analysis, not only on the summary. The draft EIS states at page 17 that "[c]omments received in response to the NOI were used to inform the range of alternatives and environmental impact analyses," and Appendix 3 states that the coded comments produced the themes, concerns, and recommendations the agency responded to. The sort therefore determined which concerns reached the alternatives development described in Chapter 2. A quantified characterization of the comment record that the record itself cannot reproduce is not merely a descriptive summary; it is an input to the range of alternatives, and the draft EIS says so. I request a factual correction under 7 CFR 1b.7(f)(2)(v). The final EIS should either support the 99 percent, 1 percent, and 90 percent figures by stating the criteria used to assign a letter to each category — including the treatment of mixed and ambiguous letters, form letters, and petition signatures — and identifying where that determination is documented in the proposal record, or qualify those figures to what the record supports. Under 7 CFR 1b.7(f)(3), the final EIS should cite where that correction is accounted for. This is not a request that the agency perform new work; it is a request that a number the agency chose to publish be supported by the record that carries it, or else be stated as the approximation the record can support. Separately, and addressed to the final rule rather than to the environmental impact statement: if the preamble relies on any characterization of the volume or direction of public comment in explaining the departure from the 2001 Roadless Area Conservation Rule, that characterization must rest on something the record shows, and the agency must articulate a rational connection between the facts found and the choice made. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983); 5 U.S.C. 706(2)(A). A response to this point in the final EIS comment appendix does not answer it in the preamble.

Dear Secretary Rollins and Chief Shultz: I am writing as an avid user of the roadless areas not only in my home state of Idaho, but across the United States. Although Idaho may have its own roadless rule, I still value the over 40 million acres of roadless areas across all America's public lands. These areas are important to me because they exist to protect watersheds, promote biodiversity of fish, wildlife, and plant species, as well as provide irreplaceable recreational opportunities. With over 300,000 miles of existing roads on National Forests today that cannot be properly maintained, more roads by means of rescinding the Roadless Rule is exactly what the Forest Service should not be attempting to do. Building roads into roadless areas is likely to result in more wildfires. Roads will not stop nor lessen the increasing amount or frequency of wildfires. More roads = more people = more fires. In the Wildfire Occurrence section on page 98 of the Draft EIS, the agency states: "Consistent with other research findings, these analyses indicate that human caused ignitions increase in abundance with proximity to roads, supporting the notion that roads facilitate human access, which can in turn result in more human caused fires (Narayanaraj and Wimberly 2012, Parisien et al., 2016, Balch et al., 2017, Chen and Jin, 2022)." The Comparison of Alternatives table in Chapter 2 carries that finding into the effects comparison, stating that under alternatives 2 and 3, "if there is an increase in public road access, there could be an increase in the number and frequency of wildfires as human-ignitions are more likely near roads." The Draft EIS therefore treats an increase in the number and frequency of wildfires arising from road proximity as a reasonably foreseeable effect of the proposed action. The analysis presented in the Draft EIS does not support that finding at the magnitude asserted, for two reasons established by the document's own text. First, the two analyses of the Draft EIS documents measure land designation, not road distance. Table 18 (p. 95) and Table 21 (p. 98) both compare other NFS lands, potentially affected IRAs, and wilderness, using Interagency Fire Occurrence Reporting Module (InFORM) Fire Occurrence Data Record ignitions with a point of origin on NFS lands for 2014-2024. A contrast among land designations cannot isolate a road-proximity effect. It states that "Other NFS lands typically have more land management activities, infrastructure (such as powerlines), and public recreation that can lead to human caused ignitions," and it reports that other NFS lands have "about twice the density of natural ignitions than IRAs or wilderness, potentially due to biophysical differences or higher detection rates" (p. 98). Natural ignitions are not caused by road access. The agency's own explanation for the natural-ignition differential - biophysical difference and detection rate - applies equally to the human-ignition differential in Table 21, and the Draft EIS neither tests nor excludes it before attributing that differential to roads. Second, the only analysis in the Draft EIS that measures ignitions against road distance is Figure 15, "Density of reported ignitions by distance to roads and cause within the potentially affected IRAs" (p. 98). The Figure 15 caption identifies no data source and no period, and the figure carries none of the documentation supplied for those tables in footnotes 38 through 40 on page 94. The narrative does not reference Figure 15 or report any value from it - no distance classes, no ignition counts by cause, no exposed area, no densities. The one within-IRA road-distance analysis in the document is presented with none of the information a reader would need to evaluate it, while the two designation-level analyses that cannot answer the road-distance question are fully documented. I request that the Forest Service supplement the wildfire occurrence analysis (7 CFR 1b.7(f)(2)(iii)) to report the data underlying Figure 15 - the distance classes, ignition counts by cause, exposed area, and resulting ignition densities - and to state whether the road-distance relationship within the potentially affected IRAs survives the confounders the Draft EIS itself identifies on page 98. If it does not, the finding on page 98 and the Fire and Fuels row of the Comparison of Alternatives require factual correction under 7 CFR 1b.7(f)(2)(v). Absent one or the other, the Final EIS would assert a reasonably foreseeable effect for which it has disclosed no supporting analysis: not a cause-effect relationship established within the meaning of 7 CFR 1b.11(a)(23), and not a reasoned explanation under 5 U.S.C. Sec. 706(2)(A). See Kern v. Bureau of Land Management, 284 F.3d 1062 (9th Cir. 2002). This concern is directed to both the Draft EIS and the proposed rule. An answer given only in the Final EIS response-to-comments appendix would leave the concern unanswered as a comment on the rule.

I oppose the proposed rescission of the 2001 Roadless Area Conservation Rule (36 CFR part 294, subpart B). I request that USDA specifically address each point below in the final rule and response-to-comments, as required for reasoned decisionmaking under the Administrative Procedure Act, 5 U.S.C. 706(2)(A). ## 1. The draft EIS's alternatives analysis must not be a formality The preamble states that Tribal consultees requested specific alternatives, including a "No Action" alternative, a "Strengthened Roadless Rule Alternative," and a "Traditional Homelands Conservation Rule" incorporating Tribal co-stewardship and consent (Tribal Input Received section). It also records that Tribal participants "noted the global climate significance of the forest as a critical carbon sink," referring to the Tongass National Forest (same section). I request that the final EIS include a full, good-faith analysis of each requested alternative and disclose the carbon-sink and cumulative landscape-level effects of removing the national prohibition across all 58.5 million acres at once, not merely note that these concerns were raised. ## 2. The rule's own cost-benefit findings do not support the action The preamble states that only about 4.8 million acres (16 percent of forested inventoried roadless areas) meet both operability and plan criteria for expanded timber harvest, and that even full harvest of those areas — called "unlikely due to budgets, unforeseen limits on operability, and market conditions" — would yield only a 5 to 10 percent increase in national sawtimber harvest and $5.2–11.4 million per year in Treasury/Forest Service revenue. Against this, the same preamble estimates recreation and tourism losses of $6.1 million annually and states total additional impacts "could exceed $100 million" per year. The proposed rule does not explain why this tradeoff — modest, uncertain gains against agency-estimated costs exceeding $100 million annually — justifies rescinding a 25-year-old national protection. I request the final rule reconcile this discrepancy directly rather than restate general deregulatory policy goals. ## 3. Tribal consultation is incomplete, and majority opposition is already documented The preamble states the Forest Service received 64 requests for Tribal consultation, of which only 29 have been held, and that consultations "will be ongoing throughout the rulemaking process" — meaning the record shows consultation is incomplete as of this proposal. It also states that "the majority sentiment among Tribal governments consulted is opposition to the proposed rescission," citing threats to "inherent rights, Tribal sovereignty, cultural survival, and the ecological health of ancestral homelands." The preamble itself concludes this rescission "would have substantial direct effects on Indian Tribes ... and therefore warrants consultation under E.O. 13175." Given that determination, USDA should not finalize this rule until consultation is complete for all 64 requesting Tribes and the required ANILCA Section 810 subsistence hearings for Alaska — currently undated, with details "announced in a subsequent notice" — are held and incorporated into the record. ## Conclusion Each point above is drawn directly from USDA's own proposed rule. I request that the final rule and response-to-comments specifically address each one, consistent with the agency's obligation to engage in reasoned decisionmaking and respond to significant comments in the rulemaking record.