View PDF Version

No. 06-797

In the Supreme Court of the United States

UNITED STATES FOREST SERVICE, ET AL., PETITIONERS

v.

EARTH ISLAND INSTITUTE AND THE CENTER FOR BIOLOGICAL DIVERSITY

ON PETITION FOR A WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

APPENDIX TO THE
PETITION FOR A WRIT OF CERTIORARI

PAUL D. CLEMENT
Solicitor General
Counsel of Record

SUE ELLEN WOOLDRIDGE
Assistant Attorney General

EDWIN S. KNEEDLER
Deputy Solicitor General

JONATHAN L. MARCUS
Assistant to the Solicitor
General

RONALD M. SPRITZER
Attorney
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

APPENDIX A

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

__________

No. 05-16776

EARTH ISLAND INSTITUTE, A CALIFORNIA NON- PROFIT ORGANIZATION; CENTER FOR BIOLOGICAL DIVERSITY, A NON-PROFIT ORGANIZATION,
PLAINTIFFS-APPELLANTS

v.

7. UNITED STATES FOREST SERVICE; DALE BOSWORTH, CHIEF OF THE UNITED STATES FOREST SSERVICE; JOHN BERRY, FOREST SUPERVISOR FOR EL DORADO NATIONAL FOREST, DEFENDANTS-APPELLEES

SIERRA PACIFIC INDUSTRIES,

DEFENDANT-INTERVENOR-APPELLEE

__________

Filed: Mar. 24, 2006

__________

BEFORE: NOONAN, TASHIMA, and W. FLETCHER, Circuit Judges.

WILLIAM A. FLETCHER, Circuit Judge.

Plaintiffs Earth Island Institute and the Center for Biological Diversity (collectively, "Earth Island") appeal the district court's denial of their motion for a pre liminary injunction enjoining the implementation of two United States Forest Service ("USFS") post-fire resto ration projects in the El Dorado National Forest. Sierra Pacific Industries ("SPI") has joined defendants USFS, Dale Bosworth, and John Berry as an intervenor.

Earth Island contends that the Final Environmental Impact Statements ("FEISs") for both projects fail to meet the requirements set forth in the National Environmental Policy Act ("NEPA"), 42 U.S.C. § 4321 et seq., because the USFS used faulty scientific meth odology in developing its tree mortality guidelines, and because the FEISs failed to consider adequately the adverse impacts of the projects on the California spotted owl. Earth Island also contends that the FEISs fail to comply with the National Forest Management Act ("NFMA"), 16 U.S.C. § 1600 et seq., because the USFS did not compile sufficient population data for certain bird Management Indicator Species ("MIS").

The district court denied Earth Island's request, finding (1) that the methodology employed in the FEISs with respect to the tree mortality guidelines was not arbitrary and capricious; (2) that the FEISs took a "hard look" at the adverse impacts of the projects on the California spotted owl, and (3) that the USFS had gathered sufficient population monitoring data for certain bird species that have been categorized as MIS. For the reasons set forth below, we hold that Earth Island has shown a "strong likelihood of success on the merits" of its NEPA and NFMA claims, and that it has otherwise satisfied the requirements for a preliminary injunction. We reverse and remand to the district court for further proceedings consistent with this opinion.

I. Statutory and Factual Background

In October 2004, two substantial fires burned por tions of the El Dorado National Forest. The first, known as the Power Fire, consumed 16,993 acres of National Forest as well as additional acres of private land. The Power Fire burned at varying levels of intens ity. According to the FEIS prepared for the Power Fire Restoration Project, approximately 38% of the forest area burned at low intensity; approximately 13% burned at moderate intensity, killing 25% to 75% of the trees; and approximately 48% burned at high intensity, killing 75% to 100% of the trees as well as burning the duff and litter protecting the soil. Several Protected Activity Centers ("PACs"), Home Range Core Areas ("HRCAs"), and Riparian Conservation Areas ("RCAs") for the California Spotted Owl were located in the Power Fire area. Certain MIS cavity-nesting birds, notably the hairy woodpecker, black-backed woodpecker, and Williamson's sapsucker, were present in Power Fire area.

The second fire, known as the Freds Fire, burned 7,700 total acres, 4,600 of which were in the National Forest. Like the Power Fire, the Freds Fire burned at varying levels of intensity. According to the FEIS, approximately 12% burned at low intensity; approxi mately 11% burned at moderate intensity, killing 33% to 66% of the trees; and approximately 61% burned at high intensity, killing 66% to 100% of the trees as well as burning the duff and litter protecting the soil. An additional 16% of young plantations also burned at high intensity. California spotted owl PACs, HRCAs, and RCAs were located in the Freds fire area. The hairy woodpecker, black-backed woodpecker, and Williamson's sapsucker were also present in the Freds Fire area.

In response to the two fires, the USFS undertook the Power Fire Restoration Project and the Freds Fire Restoration Project. Both projects must comply with federal statutes as well as the relevant regional forest plans.

Under NEPA, federal agencies must prepare de tailed environmental impact statements on every pro posed action that "significantly affects the quality of the human environment." 42 U.S.C. § 4332(C). These statements must include a description and analysis of the environmental impact of the proposed action, any adverse environmental effects that cannot be avoided if the action is implemented, alternatives to the proposed action, the relationship between short-term uses and long-term productivity, and any irreversible or irre trievable commitment of resources that would be involved if the action were to be implemented. Id. In short, NEPA requires that a federal agency "consider every significant aspect of the environmental impact of a proposed action" and "inform the public that it has indeed considered environmental concerns in its de cisionmaking process." Kern v. U.S. Bureau of Land Mgmt., 284 F.3d 1062, 1066 (9th Cir. 2002) (internal quotations omitted). NEPA does not contain substan tive environmental standards but instead establishes procedural requirements designed to ensure that agencies take a "hard look" at the environmental conse quences of their actions. Id.

Under the NFMA, the USFS must develop land and resource management plans for each unit of the National Forest System. 16 U.S.C. § 1604(a). In de veloping such plans, "a systematic interdisciplinary ap proach to achieve integrated consideration of physical, biological, economic, and other sciences" must be used. The NFMA and regulations promulgated thereunder impose substantive environmental requirements. See 36 C.F.R. § 219.12. Each forest plan must also comply with NEPA. 16 U.S.C. § 1604(g)(1).

The Sierra Nevada Framework Plan Amendment (the 2001 Framework) is the relevant NFMA forest plan for the El Dorado National Forest. The 2001 Frame work is a comprehensive forest plan that establishes a "comprehensive conservation strategy" for national forests in the Sierra Nevada Mountains, including the establishment of PACs for the California spotted owl, defined as the best available 300 acres of owl habitat surrounding a known or suspected nesting site. Earth Island Inst. v. U.S. Forest Serv., 351 F.3d 1291, 1296 (9th Cir. 2003). Logging within PACs under the 2001 Framework is severely restricted, "generally to the reduction of surface and ladder fuels." Id. In addition, the 2001 Framework requires that HRCAs, defined as 1000-acre foraging grounds for the California spotted owl, be maintained surrounding each PAC. Id.

In 2004, the 2001 Framework was supplemented by a Final Supplemental Environmental Impact Statement (the 2004 Supplement) that included a revised plan to improve fire prevention and suppression, reduce fuel loads, restore fire-adapted ecosystems, and promote community assistance. The Record of Decision ("ROD") implementing the 2004 Supplement provides that "[s]ite- specific decisions [regarding timber sales] will be made on projects in compliance with NEPA, [the Endangered Species Act], and other environmental laws following applicable public involvement and administrative appeal procedures."

The El Dorado National Forest is also managed under the El Dorado National Forest Land and Re source Management Plan (the "LRMP"). The LRMP identifies certain bird Management Indicator Species ("MIS") that aid the USFS in establishing objectives for improving habitat and for evaluating the quantity and quality of habitat and species population trends, in accordance with the NFMA.

Both the Power and Freds Project must conform with the 2001 Framework, the 2004 Supplement, and the LRMP, each of which must in turn comply with NEPA and the NFMA. The USFS identified four key goals for the two post-fire restoration projects:

(1) to reduce long-term fuel loading in order to reduce future fire severity and resistance to control;

(2) improve roads and establish effective ground cover in severely burned areas to reduce erosion and sedimentation to streams in the short term, and to contribute to long term soil productivity; (3) recover the economic value of timber killed or severely injured by the fire, in an expeditious manner, for the purpose of generating funds to offset the cost of future restoration activities; and (4) reduce safety hazards to the public and forest workers.

Freds Fire Restoration, 69 Fed. Reg. 77,175-02 (Dec. 27, 2004).

In furtherance of the third goal, the Power Project was divided into six timber sales. One of these, the East Panther sale, was awarded to SPI, which contracted to remove dead trees from 1,363 acres of the Power Fire area. The Freds Project was divided into two sales. One of these, the Fred Fire Salvage sale, was also awarded to SPI, which contracted to remove dead trees from 1,363 acres of the Freds Fire area.

A Notice of Intent to prepare an Environmental Impact Statement for the Power Fire was published in the Federal Register on December 22, 2004, and for the Freds Fire on December 27, 2004. Power Fire Re storation, 69 Fed. Reg. 76,686-01 (Dec. 22, 2004); Freds Fire Restoration, 69 Fed. Reg. 77,175-01 (Dec. 27, 2004). After a comment period, Draft Environmental Impact Statements were published in the Federal Register on March 25, 2005. Environmental Impact Statements; Notices of Availability, 70 Fed. Reg. 15,315-01 (Mar. 25, 2005).

On June 16, 2005, the Forest Supervisor for the El Dorado National Forest, John Berry, requested an Emergency Situation Determination from the Regional Forester pursuant to 36 C.F.R. § 215.10. According to the request, "substantial loss of economic value to the Federal Government will occur if implementation of the [Record of Decision] were delayed. Loss of economic value will in turn jeopardize the implementation of the project[s] resulting in long term consequences to the environment." The request noted that with an Emer gency Determination, RODs for the two projects could be issued as early as August 1, 2005, but without such a determination, the RODs might not be issued until the end of October. A delay would have the consequence of postponing large portions of the logging operations until summer 2006.

Such a delay would cause deterioration of the timber to be salvaged, which would thereby reduce USFS's revenues from those sales. The Forest Supervisor anti cipated that the loss from delay would be $11.3 million for the Power Project and $800,000 for the Freds Project. On July 1, 2005, the Regional Forester granted the requests for Emergency Situation Determinations for both Projects.

On July 1, 2005, the USFS issued FEISs for both Projects. Both FEISs used mortality guidelines to predict which trees will eventually die from their fire- related injuries, and which trees should therefore be logged. The guidelines are based on a draft study by Sharon M. Hood, Sheri L. Smith, and Daniel R. Cluck (the Hood Study). The Hood Study estimated the probability of mortality for different tree species based, according to the study, "on an analysis of the largest dataset available in terms of numbers of trees and species from wild-fires in California."

On August 1, 2005, the final RODs for both projects were issued. The Power ROD chose Alternative Four; the Freds ROD chose Alternative One. Both RODs stated that the risk of cutting trees that would otherwise survive is mitigated by relying upon the data contained in the Hood Study, which provides models "that allow managers to select the desired level of predicted mortality based on land management objectives." Among the stated reasons put forth for having chosen Alternatives Four and One was the fact that each Alter native would generate the greatest revenue for the USFS: $19,056,425 for the Power Project, and $3,345,872 for the Freds Project.

Earth Island brought suit in federal district court seeking a preliminary and permanent injunction against the implementation of the Power and Freds Projects. On August 18, 2005, the district court granted Earth Island's motion for a Temporary Restraining Order ("TRO") pending that court's determination of whether a preliminary injunction should issue. On August 25, 2005, the district court vacated its August 18 TRO and denied Earth Island's motion for a preliminary injunc tion.

In this expedited appeal, Earth Island argues that the district court (1) applied an erroneous legal standard for preliminary injunctions by requiring a showing of significant irreparable harm; (2) applied an erroneous legal standard to Earth Island's claim that the USFS failed to ensure scientific integrity in its tree mortality guidelines; (3) erred as a matter of law by concluding that both FEISs met NEPA's requirement that they take a "hard look" at the adverse effects of the projects on the California spotted owl; (4) erred as a matter of law by concluding that the USFS did not violate the NFMA by failing to conduct population surveys of cer tain MIS bird species; (5) erred as a matter of law by concluding that Earth Island had not shown the possibility of irreparable harm to the California spotted owl and certain bird species; and (6) erred as a matter of law by concluding that the balance of hardships did not tip in Earth Island's favor. On September 15, 2005, Earth Island filed an emergency motion for an injunc tion pending appeal with this court, which was denied by a motions panel on September 22, 2005. After oral argument, we sua sponte reconsidered our September 22 denial and, in an order filed January 11, 2006, issued an injunction pending the issuance of this opinion.

We have jurisdiction pursuant to 28 U.S.C. § 1291(a)(1). We now reverse the decision of the district court and hold that Earth Island has met the require ments for a preliminary injunction. We remand to the district court for further proceedings consistent with this opinion.

II. Standard of Review

A district court's decision granting or denying pre liminary injunctive relief may be reversed only if the court abused its discretion. See Harris v. Bd. of Supervisors, 366 F.3d 754, 760 (9th Cir. 2004); Earth Island, 351 F.3d at 1298. A court abuses its discretion if it based its decision on an erroneous legal standard or clearly erroneous findings of fact. Earth Island, 351 F.3d at 1298. We review findings of fact for clear error and conclusions of law de novo. See Hawkins v. Comparet-Cassani, 251 F.3d 1230, 1239 (9th Cir. 2001); Brown v. Cal. Dep't of Transp., 321 F.3d 1217, 1221 (9th Cir. 2003). We "typically will not reach the merits of a case when reviewing a preliminary injunction . . . . By this we mean we will not second guess whether the court correctly applied the law to the facts of the case, which may be largely undeveloped at the early stages of litigation. As long as the district court got the law right, it will not be reversed simply because the appellate court would have arrived at a different result if it had applied the law to the facts of the case." Earth Island Inst., 351 F.3d at 1298(quoting Rucker v. Davis, 237 F.3d 1113, 1118 (9th Cir. 2001) (en banc)).

We view Earth Island's challenges through the lens of the Administrative Procedures Act ("APA"). Under the APA, agency decisions may be set aside only if "arbitrary, capricious, an abuse of discretion, or other wise not in accordance with law." 5 U.S.C. § 706(2)(A). Review under this standard is narrow, and the reviewing court may not substitute its judgment for that of the agency. See U.S. Postal Serv. v. Gregory, 534 U.S. 1, 6- 7, 122 S. Ct. 431, 151 L. Ed. 2d 323 (2001). The agency, however, must articulate a rational connection between the facts found and the conclusions reached. Midwater Trawlers Co-op v. Envtl. Def. Ctr., 282 F.3d 710, 716 (9th Cir. 2002). We reverse under the arbitrary and capricious standard only if the agency has relied on factors that Congress has not intended it to consider, has entirely failed to consider an important aspect of the problem, or has offered an explanation for that decision that runs counter to the evidence before the agency or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. See Sierra Club v. U.S. Envtl. Protection Agency, 346 F.3d 955, 961 (9th Cir. 2003), amended by 352 F.3d 1187 (9th Cir. 2003).

III. Mootness

In light of the extensive logging that has already taken place pursuant to the Power and Freds RODs, we asked the parties to advise us whether this appeal may have become moot. The USFS concedes that this appeal is not moot with respect to the Power Project. For the Power Project, the USFS reported that 100% of the timber harvest has been completed for the East Panther and Camp Creek sales, but that only 77% of harvesting has been completed on the Cole Creek sale, only 73% has been completed for the Ellis sale, only 29% has been completed for the Bear River sale, and only 15% has been completed for the Rocky Knob sale. The USFS indicated that due to winter weather conditions, logging operations on the uncompleted projects has now ceased but are expected to resume during the spring of 2006.

According to the USFS, as of January 5, 2006, the timber harvesting that had been authorized for the Freds Project was completed, and it was expected that all harvested timber would be removed from the land and transported to mills by January 9, 2006. However, Earth Island argues that even if current logging operations have been completed for the Freds Project, the logging sales contracts were issued for a term of two years, meaning that SPI has a continuing right to cut trees that meet the tree mortality guidelines contained in the FEIS.

"A case becomes moot whenever it loses its character as a present, live controversy of the kind that must exist if we are to avoid advisory opinions on abstract pro positions of law . . . [T]he question is not whether the precise relief sought at the time of the application for an injunction was filed is still available. The question is whether there can be any effective relief." Cantrell v. City of Long Beach, 241 F.3d 674, 678 (9th Cir. 2001) (internal quotations and alterations omitted). In Neigh bors of Cuddy Mountain v. Alexander, 303 F.3d 1059 (9th Cir. 2002), we concluded that challenges brought under NEPA and the NFMA were not moot simply because logging operations had already been completed. There, we noted that remedial measures continued to be available. For example, the USFS could still study and mitigate the impact of the sales on species viability, adjust future timber sales to compensate for the allegedly unlawful sale, and directly monitor bird popu lation trends. Id. at 1066. In Cantrell, we evaluated mootness under NEPA and concluded that although historic buildings with important bird habitats had already been torn down, the defendants in that case could still consider alternatives and develop ways to mitigate the damage to the birds' habitat. Cantrell, 241 F.3d at 678-79.

Similarly, we conclude in this case that the com pletion of current logging operations in the Freds Fire area does not render the controversy about the Freds Project moot. Not only is it possible that SPI might cut more trees. In addition, as in Neighbors of Cuddy Mountain and Cantrell, there are a variety of measures that could provide some effective relief, including revising the tree mortality guidelines, monitoring of the California spotted owl, and obtaining more accurate population surveys of MIS bird species.

IV. Preliminary Injunction Standard

Earth Island seeks to enjoin logging operations in the Power and Freds Project areas until its NFMA and NEPA claims are adjudicated in federal court. It argues that the district court applied an erroneous legal standard by requiring something beyond the "possibility of irreparable harm" when it denied Earth Island's request for a preliminary injunction. Specifically, Earth Island points out that in analyzing the USFS's obliga tion to monitor certain woodpecker species, the district court found that "immediate and irreparable" injury had not been shown. With respect to the California spotted owl, the district court did not discuss harm at all; instead, it simply concluded that Earth Island had shown no probability of success on the merits.

A district court "necessarily abuses its discretion when it bases its decision on an erroneous legal standard or on clearly erroneous findings of fact." Rucker v. Davis, 237 F.3d 1113, 1118 (9th Cir. 2001) (en banc), rev'd on other grounds, Dep't of Hous. & Urban Dev. v. Rucker, 535 U.S. 125, 122 S. Ct. 1230, 152 L. Ed. 2d 258 (2002). In this case, we conclude that the district court applied an improper legal standard when assessing whether a sufficient level of injury had been shown.

We recently had occasion to pass upon the proper legal standard governing preliminary injunctive relief in a case involving the same parties, the same district court judge, and a very similar set of facts. Earth Island, 351 F.3d 1291. In that case, Earth Island sought a pre liminary injunction against logging operations provided for in the USFS's Star Fire Restoration Project after the 2001 Star Fire in the Sierra Nevadas. Id. at 1295. We identified two sets of criteria for preliminary injunctive relief. Under the "traditional" criteria, a court may grant a preliminary injunction if a plaintiff shows "(1) a strong likelihood of success on the merits, (2) the possibility of irreparable injury to plaintiff if pre liminary relief is not granted, (3) a balance of hardships favoring the plaintiff, and (4) advancement of the public interest (in certain cases)." Id. at 1297 (internal quota tions omitted). Alternatively, a court may grant a pre liminary injunction if a plaintiff "demonstrates either a combination of probable success on the merits and the possibility of irreparable harm or that serious questions are raised and the balance of hardships tips sharply in his favor." Id. at 1298 (internal quotations omitted).

In Earth Island, we held that the district court had applied an improper legal standard by requiring that Earth Island demonstrate "actual harm . . . as op posed to speculation that some such harm could possibly occur." Id. In that case, the district judge noted that Earth Island had "failed to show that measures already in place . . . will not afford sufficient protection" and that Earth Island had "failed to identify any concrete probability of irreparable harm in any other respect." Id. We concluded that "[e]ach of these statements places a higher burden of proof on the plaintiff than is warranted." Id. We emphasized that a preliminary injunction "only requires plaintiffs to show probable success on the merits and the possibility of irreparable harm." Id.

In its August 25, 2005 order in this case, the district judge stated, correctly, that in order to prevail on a motion for a preliminary injunction, a party must de monstrate either "(1) a combination of probable success on the merits and the possibility of irreparable harm; or (2) that serious questions are raised and the balance of hardships tips sharply in favor of granting the requested injunction." However, the district judge continued, stating that under either standard, a party must show a "significant threat of irreparable injury." During the hearing, the district judge stated that "even if there is shown to be a probability of success on the merits by the plaintiffs, the plaintiffs have not shown at this time that there is a significant threat of irreparable injury by clear and convincing evidence, which is the standard."

The USFS argues that the district court applied the appropriate legal standard as set forth in our decision in Oakland Tribune, Inc. v. Chronicle Publ'g Co., 762 F.2d 1374 (9th Cir. 1985), because the words "significant threat of irreparable injury" are not the equivalent of the "concrete probability of irreparable harm" standard we held to be erroneous in Earth Island. While it is true that "significant threat" and "concrete probability" are different words, what matters is that both standards impose a higher burden of proof on Earth Island by going beyond the "mere possibility of irreparable harm" standard. In Oakland Tribune, we first determined that the plaintiff had shown a very low likelihood on the success of the merits of its claim, thereby justifying the higher standard of harm. Here, the district court ap plied the higher standard from the outset without first determining the probability of Earth Island's success on the merits.

We conclude that the district court applied an erroneous legal standard. As we discuss below, we conclude that Earth Island has shown a "strong likeli hood of success on the merits," and has also satisfied the other criteria of degree of injury, balance of hardships, and advancement of the public interest.

V. Discussion

A. Likelihood of Success on the Merits of Earth Island's NEPA Challenges

NEPA's procedural requirements require agencies to take a "hard look" at the environmental consequences of their actions. A hard look includes "considering all foreseeable direct and indirect impacts." Idaho Sporting Cong. v. Rittenhouse, 305 F.3d 957, 973 (9th Cir. 2002). In addition to direct and indirect impacts, NEPA also requires that agencies assess the cumulative impacts of their actions, defined as the "incremental impact of the action when added to past, present, and reasonably foreseeable future actions." 40 C.F.R. § 1508.7.

A hard look should involve a discussion of adverse impacts that does not improperly minimize negative side effects. Native Ecosytems Council v. U.S. Forest Serv., 428 F.3d 1233, 1241 (9th Cir. 2005). Thus, the USFS must "undertake a thorough environmental analysis before concluding that no significant environmental impact exists." Id. at 1239 (internal quotations omitted). We review whether the USFS has taken a hard look under the arbitrary and capricious standard. When re viewing the adequacy of an FEIS's hard look, we follow a "rule of reason" approach, which requires "a prag matic judgment whether the [FEIS's] form, content and preparation foster both informed decision-making and informed public participation." Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 960 (9th Cir. 2005); see also Dep't of Transp. v. Public Citizen, 541 U.S. 752, 767, 124 S. Ct. 2204, 159 L. Ed. 2d 60 (2004).

Under NEPA, "[a]gencies shall insure the pro fessional integrity, including scientific integrity, of the discussions and analyses in environmental impact state ments. They shall identify any methodologies used and shall make explicit reference by footnote to the scientific and other sources relied upon for conclusions in the statement." 40 C.F.R. § 1502.24. Agencies have wide discretion in assessing scientific evidence, but they must "take a hard look at the issues and respond[ ] to reason able opposing viewpoints." Earth Island, 351 F.3d at 1301. "Because analysis of scientific data requires a high level of technical expertise, courts must defer to the informed discretion of the responsible federal agencies." Id. "When specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own experts, even if a court may find contrary views more persuasive. At the same time, courts must independently review the record in order to satisfy themselves that the agency has made a reasoned decision based on its evaluation of the evi dence." Id. (quoting Marsh v. Or. Natural Res. Council, 490 U.S. 360, 378, 109 S. Ct. 1851, 104 L.Ed.2d 377 (1989)). If an agency has failed to make a reasoned decision based on an evaluation of the evidence, we may properly conclude that an agency has acted arbitrarily and capriciously. Id. at 1301.

The primary purpose of an FEIS is to allow for informed public participation and informed decision making. See Native Ecosystems, 418 F.3d at 965. In furtherance of this purpose, 40 C.F.R. § 1502.8 requires that FEISs "be written in plain language and may use appropriate graphics so that decisionmakers and the public can readily understand them." As we have interpreted this regulation, an FEIS "must be organized and written so as to be readily understandable by governmental decisionmakers and by interested non- professional laypersons likely to be affected by actions taken under the [FEIS]." Or. Envtl. Council v. Kunzman, 817 F.2d 484, 494 (9th Cir. 1987).

1. NEPA Challenges to FEISs' Analyses of Logging Activities

Earth Island challenges various aspects of the analyses of logging contained in the two FEISs. Most significantly, it challenges the analysis of tree mortality and the use of the proposed mortality guidelines con tained in the Power and Freds FEISs. It also challenges the FEISs' conclusions that there is a lack of adequate soil cover in the project areas, and that the retention of large snags (dead trees) would result in a hazardous level of surface fuels.

a. Tree Mortality Guidelines

Earth Island argues that the FEISs' guidelines for cutting burned and scorched trees substantially over predict tree mortality, with the result that many more trees will be cut than are necessary to meet the legitimate objectives of the Power and Freds Projects. Earth Island argues that by cutting many trees that will not die, the USFS will unnecessarily destroy valuable habitat for the California spotted owl and certain MIS bird species.

The tree mortality guidelines contained in the FEISs are based upon the 2005 draft Hood Study. That study predicts post-fire mortality for trees burned or scorched in forest fires. The guidelines for both projects are keyed to the percentage of the tree's green crown that has been scorched by the fires.

Based on data contained in the Hood Study, the USFS chose two different marking guidelines, one for areas to be logged by tractor and one for areas to be logged by helicopter or skyline. Table 3-5 of both of the FEISs is entitled "Mortality Guidelines for marking as applied to [various alternatives]." For simplicity of explanation, we will describe the guidelines only as they apply to yellow pine and white fir, and only as to helicopter and skyline logging. Yellow pine with 65% or more of the length of their crown scorched are con sidered dead, and are to be marked for cutting. The same crown length scorch percentages apply to white fir over 20" in diameter.

For the convenience of the reader, we reproduce Table 3-5:

 

 

_________________________________________________________

PP/JP/SP White White

[yellow Fir Fir Incense Red Douglas

pine] <20" >20" Cedar Fir Fir

_________________________________________________________

Tractor Logging System

_________________________________________________________

Crown

Scorch* 75% 95% 80% 100% ? ?

_________________________________________________________

Crown Scorch

+ RTB

or Ambrosia

Beetle ?

 

_________________________________________________________Helicopter and Skyline Logging Systems

_________________________________________________________

Crown

Scorch* 65% 85% 65% 95% ? ?

_________________________________________________________

Crown Scorch

+ RTB

or Ambrosia

Beetle ?

_________________________________________________________

* As a percentage of the original live crown. Percentage of green foliage would not be used for ponderosa or Jeffrey pine until the 2005 needle flush is visible. Any marking prior to needle flush would be limited to pines with 100% black crown.

 

Table 3-6 of the FEISs is entitled "Probability of Tree Mortality." Both FEISs state, "Table 3-6 indicates the probability of tree mortality to individual trees meeting the Power [and Freds] Fire marking guidelines (Correctly Predicted Mortality) and the predicted survival of trees with less fire damage than the minimum requirements of the marking guidelines (Correctly Predicted Survival)." According to Table 3-6, a yellow pine cut in accordance with the 65% crown length scorch guideline of Table 3-5 has a 90% "Correctly Predicted Mortality." A white fir over 20" in diameter cut in accordance with the guidelines has a 87% "Correctly Predicted Mortality."

For the convenience of the reader, we reproduce Table 3-6. We have italicized the numbers 90 and 87 to indicate the "correctly predicted mortality" percentages for yellow pine and white fir cut in accordance with the 65% crown length scorch guideline:

___________________________________________________

PP/JP/SP White Incense Red Douglas

[yellow pine] Fir Cedar Fir Fir

_________________________________________________________

Tractor Logging System

_________________________________________________________

Correctly

Predicted

Mortality (%) 96 95 100 100 -

_________________________________________________________

Correctly

Predicted

Survival (%) 51 63 88 - -

_________________________________________________________

Helicopter and Skyline Logging Systems

_________________________________________________________

Correctly

Predicted

Mortality (%) 90 87 85 100 - _________________________________________________________

Correctly

Predicted

Survival (%) 65 74 89 - -

_________________________________________________________

 

Earth Island contends that the findings contained in the draft Hood Study, as well as in other studies, have substantially different percentage estimates of tree mortality from the percentages contained in Table 3-6. Earth Island relies upon the declaration of Dr. Edwin B. Royce in support of its contention. Royce has a Ph.D. in Botany with a specialization in Forest Plant Ecology from the University of California at Davis, and a Ph.D. in Applied Physics from Harvard University. Royce has had twelve years of experience in the characterization of forest vegetation.

The USFS challenges the admissibility of Royce's declaration, as well as other expert declarations offered by Earth Island, because they were not before the agency during the administrative review process. We allow extra-record materials if necessary to "determine whether the agency has considered all relevant factors and has explained its decision." Sw. Ctr. for Biological Diversity v. U.S. Forest Serv., 100 F.3d 1443, 1450 (9th Cir. 1996). Because Earth Island's expert declarations are offered for this purpose, they were properly before the district court and are properly before this court on review.

According to Royce, yellow pine and white fir are both "thick bark trees" whose bark provides significant protection of the living tissue beneath the bark. According to Royce, "[c]rown kill is commonly the do minant source of fire-induced mortality in large trees having thick bark, such as white fir or yellow pine." However, "[f]all fires, such as the Power and Fred's fires, are least damaging to conifers. In part this is because reserves of stored products of photosynthesis are high and because new growth throughout the trees is less vulnerable to damage then, as compared to that same growth during the summer growing season." "A low-intensity surface fire commonly produces only partial crown kill and only minimal mortality in larger trees. Trees tolerate partial crown kill in a surface fire in part because only the lower part of the crown is normally killed. The lower part of the crown is less photosynthetically productive than is the upper part."

For yellow pine, Royce states, "[t]he mortality guidelines for yellow pine (ponderosa and Jeffrey pine) will permit the harvest of trees with a substantial probability of surviving if they were not harvested." The Royce declaration contains tables showing pre dicted tree mortality in the Hood Study, as well as in three other studies. Those other studies were by Ryan and Reinhardt, by Stephens and Finney, and by McHugh and Kolb. Royce states as to yellow pine, "[f]or helicopter and skyline logging, mortalities from the[Hood Study] models are 60% and 70%, as compared to a 90% mortality claimed in the FEIS and ROD." Mortalities predicted by the three other studies are even more at variance from the mortality predicted in the FEISs. Royce states,

[T]he guidelines for helicopter and skyline logging will allow the cutting of small trees that have probabilities of mortality between 12% and 57% and large trees with mortality probabilities between 11% and 32% (compared with 90% mortality claimed by the FEIS's). Even if one accepts only the largest of the probabilities of mortality-that given by the Ryan and Reinhardt paper-this still translates into a worst case probability of survival of up to 68% for trees that could be cut (meaning 68% of the trees logged would otherwise survive). (emphasis in original.)

In addition, Royce states that he personally evaluated 39 yellow pine that were part of the data base for the Hood Study, and "found evidence of measurement errors that would lead to the development of models that over predict mortality." According to Royce,

I measured the average amount of crown kill on these trees to be 88%, whereas, the [Hood Study] field workers evaluated the same damage to average 64%. A small part of this difference may be due to the fact that my measurements were crown volume measure ments, whereas, the [Hood Study] measurements were linear measurements. However, as discussed [earlier in my declaration], this should produce only a few percent difference between the two measure ments. I suggest that at least a substantial part of the difference between the two measurements is a result of the [Hood Study] field workers underesti mating crown kill. This will result in a model that over predicts mortality when correct crown kill values are used. A corrected [Hood Study] model yielding lower mortality predictions would make the proposed guidelines even less defensible.

For white fir, Royce states, "[t]he mortality guidelines for large white fir will permit the harvest of trees with a substantial probability of surviving if they were not harvested." As he does for yellow pine, Royce presents tables showing predicted tree mortality in the Hood Study, as well as in the three other studies. He states as to white fir, "[f]or the helicopter and skyline logging of large trees, mortalities from the[Hood Study] models are between 25% and 45%, as compared to an 87% mortality claimed in the FEIS and ROD." Just as with yellow pine, mortalities predicted by the three other studies are even more at variance from the mortalities predicted in the FEISs. Royce states, the guidelines for helicopter or skyline logging will allow the cutting of large trees with mortality probabilities between 7% and 50%, as compared to the 87% mortality claimed in the FEIS's. Even if one accepts only the largest of these predictions of mortality-that given by the Ryan and Reinhardt paper-this still translates into a worst case pro bability of survival (in helicopter units) of up to 50% for trees that could be cut. (emphasis in original).

In response, the USFS relies on a declaration of Sheri L. Smith. Smith is one of the three co-authors of the draft Hood Study. She has B.S. and M.S. degrees in Biology and Entomology from Utah State University. She has been an USFS employee for 15 years. She has been "involved in evaluating fire-injured trees in California since 1991."

Smith makes three points in response to the Royce declaration. First, she contends that Royce misunder stood Table 3-6. As noted above, that table is entitled "Probability of Tree Mortality." Royce understood the percentages contained in that table as indicating the probability that a tree will die. For example, Table 3-6 indicates that a yellow pine with a minimum of 75% scorched crown (the guideline for tractor logging) has a "correctly predicted mortality" of 96%, and a "correctly predicted survival" of 51%. Royce understood the table to mean that 96% of yellow pine with a minimum of 75% scorched crown will die.

Smith responds,

The marking guidelines used for the Power and Freds fire are based on models that use percent crown kill . . . . I attach a true and correct copy of an excerpt of the models which are the basis for the Power and Freds marking guidelines (Attachment A). Mr. Royce states that Table 3-6 represents that 96% of trees with 75% crown kill will die. This is simply incorrect. The 96% in Table 3-6 (classification table) shows that the marking guideline model used by the Forest Service is correct in predicting mortal ity 96% of the time when our individual study trees are run through the model. In other words, the 96% is an estimate of the accuracy of the model, not of the percentage of mortality as Mr. Royce states. Both FEISs accurately interpret the findings of our paper in light of the specific marking guideline model selected by the Forest Service.

The percentages shown in Table 3-6 (classifi cation table) of both the Power and Freds FEIS are derived from the entire [Hood Study] dataset, which, for yellow pine, is based on measurements of 1,969 trees. The models in our paper take these percent ages to arrive at a statistical probability of mortality for an individual tree. The probability of mortality (Pm) ranges from 0.0 to 1.0, with 1.0 being certain mortality. The Forest Service selected a Pm of 0.90 for all ground-based harvest units. For yellow pine, a Pm of 0.90 corresponds to a crown length kill of 75%. Using this criterion, the marking guideline model correctly predicted mortality for yellow pine trees in the [Hood Study] dataset 96% of the time and correctly predicted survival 50% of the time. In other words, using a Pm of 0.90 (75% crown kill), only 4% of the trees the model predicts will die, actually survived, but 50% of the trees predicted to survive eventually died. . . .

This means that there is a high likelihood of cor rectly predicting mortality, but a lower likelihood of correctly predicting survival using the 75% crown scorch model, which means that it is much more likely that the Forest Service is leaving trees behind that will later die, than it is taking trees that would have survived. This is directly contrary to Mr. Royce's assertions [.] . . . The same misinterpreta tion of the data is true for Mr. Royce's discussion of yellow pine on helicopter and skyline harvest units and all white fir, all of which misinterpret the percentages in Table 3-6 to be something other than what they are-a verification of the accuracy of the marking guideline model using the [Hood Study] dataset.

(emphasis in original).

Second, Smith responds that Royce estimated the probability that a tree will die based on the percentage volume of crown kill. By contrast, the Hood Study estimated the probability based on the percentage length of crown kill. Therefore, according to Smith, Royce's comparisons are "problematic."

Third, Smith responds that two of the three other studies upon which Royce relies for his comparison-the studies of Stephens and Finney, and of McHugh and Kolb-are based on "prescribed fires." According to Smith, "[m]ost often, the objective of a prescribed fire is to limit the mortality of the overstory while reducing fuel loadings and ingrowth of smaller trees. Our model is more appropriate to the type of fire represented by the Freds and Power fires, which killed not only the ingrowth of smaller trees, but also resulted in high levels of crown kill of larger trees."

We analyze Smith's three points in turn. First, Smith contends Royce has misunderstood Table 3-6. Accord ing to Smith, the percentage figures given for "correctly predicted mortality"-such as 96% for yellow pine with a minimum of 75% crown length scorch-do not predict the percentage of trees that will die. Rather, these figures predict the accuracy of the prediction of the percentage of these trees that will die. Similarly, according to Smith, the percentage figures for "cor rectly predicted survival"-such as 51% for yellow pine with a minimum of 75% crown length scorch-do not predict the percentage of trees that will live. Rather, they predict the accuracy of the prediction of the percentage of these trees that will live. Smith may well be correct. Indeed, it appears from the face of Table 3-6 that she may be, for if the table predicted the percentage of trees that will die or live, as distinct from predicting the accuracy of the predictions of these percentages, the percentages given in the table should add up to 100%. Instead, in the example given for yellow pine with 75% crown length scorch, they add up to 147% (96% plus 51%).

But even if Smith is correct, this does not solve the problem. Table 3-6 is, to say the least, misleading. Its title is "Probability of Tree Mortality," rather than "Probability that Predictions of Probability of Tree Mortality and Survival are Correct." Second, there is no other table in the FEISs providing the probability of tree mortality. The absence of such a table is signifi cant. The single most important aspect of the FEISs is their estimate of the likelihood that trees with certain amounts of fire damage will die. This is so for the obvious reason that the justification for cutting burned or scorched trees is the likelihood that they will die. Any reader of the FEISs will therefore look for a table providing probability of tree mortality. The only table in the FEISs that appears to provide that information is Table 3-6. It is not unforeseeable that a reader-even an expert reader such as Royce-would misunderstand the table. Further, the explanation for Table 3-6 pro vided by Smith's declaration in the district court is nowhere provided in the FEISs. For example, Attach ment A to Smith's declaration is not provided, or even referred to, in the FEISs. The absence of such an explanation in the FEISs obviously increases the chance that the table will be misunderstood.

Further, the Royce declaration provides tree mortality percentages given in the draft Hood Study, and he compares those percentages to those given in Table 3-6. The Hood Study percentages provided by Royce are not contested by Smith. (This is not surprising, given that Smith is a co-author of the Hood Study.) The Hood Study mortality percentages are substantially lower than the percentages given in Table 3-6. Yet those percentages are not provided in the text of the FEISs. If the USFS had been truly interested in educating the reader as to the actual percentages of trees likely to die from scorch damage, it would have provided those percentages in the FEISs. Instead, it provided Table 3-6 (entitled "Probability of Tree Mortality") giving percentages of "correctly predicted mortality." Even if the USFS could properly rely on the draft Hood Study instead of other tree mortality studies-a question we do not here decide-it should have provided the mortality percentage figures in the Hood Study rather than, or in addition to, the figures in Table 3-6.

Second, Smith responds that the comparisons used by Royce to estimate tree mortality are based on percen tage volume of crown kill rather than percentage length of crown kill. We note initially that Smith's response has no application to the comparison between the per centage numbers given in Table 3-6 and those given in the Hood Study. Both Table 3-6 and the Hood Study were based on the same methodology-correlating percentage length of crown kill with likelihood of tree death. Further, we note that with respect to studies that relied on percentage volume rather than percentage length of crown kill, Royce had explained in his declara tion why the FEISs' emphasis on the difference in the two type of crown kill measurements is misleading. He wrote:

Tables 3-4 in both the Power and Fred's fire FEIS's show a relationship between these two mea sures of crown kill in which the volume measure is much larger than the linear measure for the same amount of damage. However, these tables are valid only for young trees with a tapered shape, similar to the classic conical Christmas tree shape shown in figure F-1 of the Powers FEIS or figure A-1 of the Fred's FEIS. Large, mature trees growing in a forest surrounded by other trees have a shape more like a cylinder, tapered only in the topmost branches. . . . It is to mature trees that the guidelines [in the FEIS's] are to be applied to determine if the trees are dying and hence subject to salvage harvest. In the context of the salvage of mature trees, the inclusion of these tables and figures in the FEIS's is completely misleading.

My experience measuring mature trees on the Eldorado-Star fire site . . . was that, within the uncertainty with which these determinations can be made in the field, the two measures give the same numerical value for crown kill.

Third, Smith responds that two of the three other studies used by Royce for comparison were based on "prescribed fires." According to Smith, reliance on these two studies-the Stephens and Finney study and the McHugh and Kolb study-was inappropriate because the Power and Freds Fires "resulted in high levels of crown kill of larger trees" compared with the levels of crown kill in the prescribed fires. It may well be that, as Smith says, the prescribed fires resulted in lower levels of crown kill in larger trees. But Smith does not explain why that makes a difference. As those two studies are described by Royce, they predict tree mortality based on the actual percentage of crown kill in particular trees. The fact that fewer trees have that degree of crown kill in a prescribed fire should not make any difference in the predicted mortality of those trees that actually have a high level of crown kill.

Further, Smith does not mention the third study upon which Royce relies-the Ryan and Reinhardt study. Of the three studies, this one gives the highest probability of tree mortality. Royce explicitly discusses and relies on the Ryan and Reinhardt study in his conclusion that the FEISs overpredict tree mortality. For example, as indicated above, Royce writes with respect to yellow pine, "[e]ven if one accepts only the largest of these probabilities of mortality-that given by the Ryan and Reinhardt paper-this still translates into a worst case probability of survival of up to 68% for trees that could be cut[.]"

We recognize that the FEISs discount two of the studies-those by Stephens and Finney and by Ryan and Reinhardt. The Power and Freds FEISs both discount the Stephens and Finney study on the ground that its purpose was not to provide salvage guidelines, and that some of its data were obtained pre-rather than post-fire. The Powers FEIS discounts the Ryan and Reinhardt study as involving only one of the species (Douglas fir) that occurs in the Power Fire area. The Freds FEIS discounts that study in slightly different terms, stating, "[r]esults from the Ryan and Reinhardt 1988 study were obtained mostly for tree species not found in, and geographic regions not related to, the Sierra Nevada." But even if we must discount both of these studies-a question we do not decide-the McHugh and Kolb study remains. That study was not discussed, or discounted, in either of the FEISs.

In the end, we conclude that the USFS abused its discretion in its estimates of likely tree mortality in both the Power and Freds FEISs. We will assume, for purposes of our analysis, that the Smith declaration correctly states what the percentage numbers in Table 3-6 really mean. But even if the Smith declaration is correct, Table 3-6 is, for the reasons given above, extremely misleading. A casual, or even a careful, reader of the FEISs and of Table 3-6 could easily con clude that 96% of yellow pine with a minimum of 75% crown length scorch will die, or that 90% with a mini mum of 65% crown length scorch will die. If those were, in fact, the percentages of yellow pine with that degree of fire damage that will die, it would be easy to conclude that the USFS is justified in cutting all yellow pine that satisfy those criteria. But those are not the percentages of trees that will die.

The Hood Study itself estimates a substantially lower tree mortality than the percentage numbers provided in Table 3-6. The other three studies estimate even lower tree mortalities. It is possible that those who prepared the FEISs, and the Forest Supervisor who signed the RODs based on the FEISs, understood Table 3-6 in the way Royce understood it. If this is so, the USFS abused its discretion, for it failed to take the requisite "hard look" at the data underlying their analysis and decision. Kern, 284 F.3d at 1066. It is also possible that those who prepared the FEISs, and the Forest Supervisor, understood Table 3-6 in precisely the way Smith described it in her declaration. If this is so, the USFS also abused it discretion, for it failed to reveal the actual percentages upon which it relied and it drafted highly misleading FEISs. Native Ecosystems Council, 418 F.3d at 965. Under the first alternative, the USFS misunderstood the data; under the second, it under stood but concealed and misrepresented the data. Under either alternative, it abused its discretion.

b. Soil Cover

Both the Power and the Freds FEISs state that a goal of the restoration projects is to provide effective ground cover as a means of reducing erosion and stream sedimentation. Both FEISs specify that the average soil cover should be between 50% to 60%. The FEISs propose to achieve this amount of soil coverage by using debris from trees that are logged according to the mortality guidelines discussed above-in their words, by using "tops and limbs of dead and dying trees."

According to the Power FEIS, there was an average of 11% to 30% ground cover in areas of moderate- severity burns after the fire. There was an average of 0% to 10% ground cover in areas of high-severity burns. The average projected ground cover after needles scorched by the fire had fallen to the ground was estimated to be 51% to 70% in moderate-severity burn areas, and 20% to 30% in high-severity burn areas. According to the Freds FEIS, there was an average of 17% ground cover in moderate-severity burn areas, and an average of 9% in high-severity burn areas. Following needlecast, the projected ground cover was 46% in moderate-severity burn areas, and 19% in high-severity burn areas.

Earth Island argues that these figures for soil coverage are based on information obtained in site visits immediately after the fire, and that more recent visits show that effective soil cover already exceeds 50% to 60% in severely burned areas. In support of this argument, Earth Island offers the declaration of Jonathan J. Rhodes. Rhodes has a B.S. in Hydrology and Water Resources from the University of Arizona, and an M.S. in Hydrology and Hydrogeology from the University of Nevada-Reno. He has also finished all required academic work toward a Ph.D. in forest hydrology at the University of Washington. He has had more than twenty-two years of experience as a hydrologist.

Rhodes visited what he described as six high-severity burn areas within the Power Fire area on August 11, 2005. He found that needles, twigs, and branches that had fallen since the fire, as well as new vegetative growth, had increased soil cover substantially. He stated in his declaration:

I measured soil cover in areas where my visual estimates indicated that soil cover was lowest. Thus, my measurements include those in areas with soil cover that is well below the average within the pro posed Power project area as a whole. My measure ments and evaluation of soil cover conditions within the areas of the proposed Power logging project unequivocally demonstrate that as of August 11, 2005, soil cover is well-distributed and greater than 60% in the overwhelming majority of areas burned at high severity. In the areas that I measured soil cover, the lowest level of soil cover that I measured was 59%; soil cover in the other five areas measured ranged from 66-91%. Even this lowest level of measured soil cover exceeds the 50% target that Power and Freds FEIS state should be exceeded.

Rhodes continued, "[b]ased on conditions within the area, it is unquestionable that soil cover from [conifer regeneration, and needles, branches, and logs from burned trees] will continue to steadily increase over the next several years." He concluded, "there is not a pressing need to try to increase soil cover via logging in order to reduce soil erosion. This is especially true because logging always causes increased soil damage and elevated erosion." (emphasis in original).

The USFS responded with a declaration by Jeffrey TenPas. TenPas has a M.S. in Soil Science from the University of California, Davis. He is an employee of the USFS. He has had more than fifteen years of experience in soil science. TenPas assessed soil cover immediately after the Power Fire. He then revisited the Power Fire area on August 19, 2005, after reading the Rhodes declaration. TenPas wrote, "[c]ontrary to Mr. Rhodes findings, portions of the high severity burned areas in the Power Project area have existing ground cover much less than 50%." He suggested that Rhodes might have been examining a "more moderately burned area that received significant needle cast," rather than a high-severity burn area. TenPas stated:

The pattern of recovery [in the Power Project area] was consistent with expectations. Needle cast had provided adequate ground cover in low and moderate severity burned areas. Bear clover, where it was present, covered an estimated 0 to 70% cover at the scale of 200 transects. In high severity burned areas, bear clover was the predominant component in vegetative recovery in areas I visited. Various forbs provided a trace to 5% cover. In high severity burned areas without bear clover, ground cover was as low as 1%, that from a trace of needles.

He concluded, "[i]n sum, my observations indicate that there remain portions of high severity burned areas where soil cover is deficient . . . . Timber harvest can contribute additional cover in these areas."

As in Earth Island, we conclude that "[a]t this stage, the record does not allow us to conclude that the Forest Service acted arbitrarily and capriciously in relying on its own data and discounting the alternative evidence offered by the Plaintiffs." Earth Island, 351 F.3d at 1302. This is especially so in light of the deference given to the "reasonable opinions of [an agency's] own ex perts." Id. at 1301.

c. Fuel Loading

Both the Power and Freds FEISs state that a goal of the restoration projects is to reduce fuel loading on the ground so as to reduce the risk of future catastrophic fires, and to avoid safety hazards to workers and recrea tion visitors. Specifically, the Power FEIS rejected an alternative that would have retained four large snags per acre outside PACs for the California spotted owl. The Freds FEIS rejected an alternative that would have retained four to eight large snags per acre outside PACs. Earth Island challenges the FEISs' rejection of these alternatives, arguing that retaining this number of large snags would not impede the goals of reducing long- term fuel loading and safety hazards.

According to the Power FEIS, in areas where four large snags per acre would be retained, fuel levels 25 years after the fire would be approximately 38 to 40 tons per acre. According to the Freds FEIS, in areas where four to eight snags per acre would be retained, fuel loads would be 16 to 50 tons per acre by year 25. The USFS contends that these levels of fuel loading would provide significant fuel, contribute to severe soil heating, and impede effective fire suppression efforts.

As with soil cover, above, we cannot confidently discern from the present record whether Earth Island's contentions about fuel-loading have validity. We therefore cannot say, at this point, that the USFS has acted arbitrarily and capriciously in rejecting the alternatives that would have retained four snags per acre in the Power Fire area and four to eight snags per acre in the Freds Fire area.

2. NEPA Challenge to Analysis of Effects of Logging on California Spotted Owls

Earth Island argues that the USFS has failed to take a "hard look" at the effects of the Power and Freds Projects on the California spotted owl. The California spotted owl was identified as a "species at risk" in the 2001 Framework. "Species at risk are those with a high level of concern whose ranges are not peripheral to the Sierra Nevada and that occur in old forest ecosystems." As reported in the 2004 Supplement, the Fish and Wildlife Service ("FWS") declared in February 2003 that the California spotted owl would not be listed as an endangered species under the Endangered Species Act because there was "no definite evidence that the population is decreasing across its range, and various analytical results of the individual study areas are not wholly supportive of conclusions regarding declines in any given study area."

However, because changes in the Sierra Nevada Framework could affect the California spotted owl, the FWS has stated that it will continue to monitor the owl. Within the last year, the FWS issued a "90-day finding" under the Endangered Species Act, 16 U.S.C. § 1533(b)(2)(A), as a prelude to a possible determination that the California spotted owl should be listed under the Act. 90-Day on a Petition to List the California Spotted Owl as Threatened or Endangered, 70 Fed. Reg. 35,607 (June 21, 2005). The FWS stated that "the petition presents substantial scientific or commercial evidence that listing the species may be warranted." Id. The FWS is currently engaged in a 12-month review of the California spotted owl as required by 16 U.S.C. § 4(b)(3)(B). Id. Among the factors prompting the 12 month review was the 2004 Supplement, as well as new evidence concerning the effects of fires on the owls. Id. at 35,612.

Under the 2004 Supplement, PACs for California spotted owls must be maintained regardless of actual occupancy by owls. After a stand-replacing event such as a major fire, habitat conditions within a 1.5-mile radius around the PAC must be evaluated in order to identify opportunities for re-mapping a destroyed or diminished PAC. Only if there is insufficient suitable habitat for designating a PAC within this 1.5-mile radius can a PAC be delisted.

Earth Island argues that the FEISs allow for exces sive cutting of trees that would otherwise survive in areas used by California spotted owls, do not adequately take into account studies showing the owls' use of already-burned areas, allow the creation of isolated islands of habitat, and fail to retain sufficient numbers of large snags for use by owls. Earth Island relies upon the declaration of Monica Bond, who has a B.A. in Biology from Duke University and an M.S. in Wildlife Science from Oregon State University. She is the lead author of two peer reviewed studies of the California spotted owl published in 2002 and 2004. Bond contends that the Power and Fred Projects will have significant negative effects on the California spotted owl by substantially reducing the amount of potential foraging habitat within the project sites, by a) utilizing inaccurate mortality guidelines (see Declaration of Edwin Royce) which incorrectly categorize some areas withing the Power and Fred's project areas as high-severity burn (and thus unsuit able for owls) when they are in fact live tree areas available to the owl for foraging purposes; b) ignoring significant new scientific information which indicates that spotted owls actually utilize forests burned at even high severities; c) by allowing pockets of forest unburned or burned at low- to moderate severity within a severe burn area to become isolated islands of habitat and reducing their value to the spotted owl; and d) failing to maintain large snags throughout the areas designated to be logged, for recruitment as legacy trees when the forest regenerates in the future. In addition, these projects call for extensive clearcut logging of habitat elements important to spotted owls (large trees and snags) within currently occupied spotted owl Protected Activity Centers ("PACs") and Home Range Core Areas ("HRCAs"). All of these activities will result in the loss of potentially critical foraging grounds that are neces sary to maintain the population of California spotted owls in the Eldorado National Forest.

The USFS responds with a declaration by Chuck Loffland, a wildlife biologist employed by the USFS. Loffland does not describe his academic background. He states that he has been conducting surveys of the California spotted owl and performing analyses of effects on the owl since 1989. He concedes that "it appears from the few studies cited by Ms. Bond that owls may use burned habitat to some degree," but that "the scope and duration of that use are not well studied or well understood." He notes that Bond herself states that more scientific study is needed. He states further:

Ms. Bond claims that most or all of the PACs will be clearcut . . . . This is simply untrue. First, there will be no salvage activity within suitable habitat. For the non-core (unsuitable) portion of PACs within the Power project area, salvage will only remove hazard trees and trees that are dead with 100% crown kill in excess of the 4 largest size class snags per acre retained. This will leave behind 4 large snags/acre and all trees that have any green needles and even a remote chance of surviving . . . . Although Ms. Bond characterizes project activity in the PACs as wholly detrimental to the owl, the Forest Service decided to remove dead and hazard trees from non- core PAC areas in order to provide protection against stand-destroying fire. (emphasis in original).

Loffland also disputes some of Bond's characteri zations of the logging that will take place within speci fied PACs.

According to the Power FEIS, there were approxi mately 5,880 acres of suitable owl habitat before the Power Fire; this habitat was reduced to 2,750 acres as a result of the fire. Ten PACs were in the Power Fire area before the fire. One was not affected by the fire and no treatment was proposed; one was burned at high and moderate intensity, and the USFS determined that not enough suitable habitat remained to support a PAC; and the boundaries of the remaining eight were redrawn to encompass the best remaining habitat within a 1.5 mile radius of the center of the area. Within seven of these redrawn PACs, the USFS determined that certain areas were unsuitable for spotted owl habitat (deemed "non-core" areas) and that salvage logging in these non- core areas could proceed, but leaving standing the four largest snags (dead trees) per acre. No logging is permitted within the "core," or suitable habitat, areas of any PAC.

An analysis of the direct effects of the Power Project was limited to the area that currently remains suitable for nesting or foraging. Because no logging is allowed in core areas, the Power FEIS notes that any direct effects would be limited to hazard zone areas, meaning roads, power line corridors, and the fire flume. Thus, the potential direct effects were limited to the removal of roadside hazard trees on 295 acres of key habitat within PACs. In addition, the Power Project is esti mated to affect 660 acres of the 10,560 total suitable HRCA acres that are within or adjacent to the Power Project area.

In discussing the indirect effects, the Power FEIS states that in high- and moderate-intensity burn areas in HRCAs, the largest 5.8 snags per acre will be re tained, but that salvage harvesting will proceed according to the tractor-harvesting mortality guidelines. The cumulative effects analysis in the Power FEIS focused on logging activities proposed for 540 acres of private land within the Power Fire area and 873 acres of private land outside the area; on a project to thin trees in order to reduce fuel loads; and on the planned replanting of 700 acres of burned plantation areas.

The Power FEIS states that the effect of the Power Project may be to reduce the quality of owl habitat, but that the project would not reduce the overall amount of owl habitat. The Power FEIS predicts that to the extent the Power Project increases future fire resiliency, it may have the effect of increasing the amount of available habitat in the long run. It concludes that there would be no trend resulting in the federal listing of the California spotted owl as an endangered species.

In the Freds Fire area, all or portions of three spotted owl PACs were present before the fire. One PAC burned at high and moderate intensity, and the USFS determined that insufficient habitat remained to support this PAC. The boundaries of the remaining two PACs were redrawn to encompass the best 1.5 miles of habit surrounding the center of the area. The Freds FEIS proposed logging in non-core areas of the redrawn PACs.

According to the Freds FEIS, approximately 3,255 acres of suitable owl habitat existed on both national forest and private land before the Freds Fire. After the fire, only 285 suitable acres remained in areas of low- intensity burn, while approximately 1,848 acres remained in areas of moderate- and high-intensity burn. In addition, some 332 acres of green trees are scattered across the moderate- and high-intensity burn areas.

According to the direct and indirect effects analysis in the Freds FEIS, no logging would take place in core areas within the PACs; no logging would be proposed in areas of low-intensity burn; and hardwoods and green trees would not be harvested in areas of moderate- and high-intensity burn. The FEIS also notes that while the 332 acres of remaining scattered green trees provide some post-fire habitat for the owls, suitability in those 332 acres would "likely be reduced over time as more fire-weakened trees die. Dead trees lose foliage and therefore would reduce canopy closure to levels below what is preferred by owls."

The USFS argues that it adopted a conservative approach to logging in both FEISs in order to ensure that the California spotted owl is protected. It points to the fact that it assumed the presence of owls in PACs pending the completion of surveys. It also argues that it considered the information concerning the owl's use of post-fire habitat and determined that the findings were too inconclusive to affect its impact analysis.

We have elsewhere interpreted the "hard look" re quirement as entailing both a complete discussion of relevant issues as well as meaningful statements re garding the actual impact of proposed projects. In Native Ecosystems Council, we held that where an EIS used a calculation for determining the summer range of elk herd that was inconsistent with a specific require ment contained in the regional forest plan, the USFS had not taken a hard look because it did not "provide a full and fair discussion of the potential effects of the project . . . and did not inform decisionmakers and the public of the reasonable alternatives which would avoid or minimize adverse impacts [to the elk at issue]." 418 F.3d at 965(internal quotations and alterations omitted). Likewise, in Klamath-Siskiyou Wildlands Center v. Bureau of Land Management, 387 F.3d 989, 994-95 (9th Cir. 2004), we held that an environmental assessment ("EA") did not take a "hard look" at the cumulative impacts of a proposed action where the EA had sections devoted to discussing direct, cumulative, and foresee able actions, but gave no objective or qualified assess ment of the combined environmental impacts of the information presented. Most recently, we concluded that where an EIS stated without meaningful explana tion that a post-fire salvage project would have a negative impact on black-backed woodpeckers but would not result in a trend toward federal listing, this did not constitute a "hard look." Ecology Ctr., Inc. v. Austin, 430 F.3d 1057, 1067 (9th Cir. 2005).

In Earth Island, Earth Island claimed that the FEIS for the Star Fire Restoration Project did not adequately analyze the cumulative impact of the project's destruc tion of an HRCA on the California spotted owl. Earth Island, 351 F.3d at 1306-07. While the FEIS in Earth Island had acknowledged that spotted owls exhibit "high site fidelity" and that a pair of owls had returned to the project area, it "never assessed the potential role of the remaining suitable habitat within the former HRCA for a maintained [PAC] despite the acknowl edged presence of owls in the area." Id. at 1307. For this reason, we held that the "omission amounts to an insufficient consideration of cumulative impact under NEPA." Id.

We conclude that the Power and Freds FEISs have not taken the requisite "hard look" at the effects of the two projects on the California spotted owl. We con cluded above that the FEISs did not adequately analyze probable tree mortality. The likely consequence of the apparent overprediction of tree mortality is excessive logging. This likely excessive logging, in turn, is likely to produce adverse effects on the California spotted owl that are not adequately analyzed in the FEISs.

Further, it is likely that the projects will substantially reduce potential foraging habitat because the FEISs' designation of non-core areas, where logging will occur, is based upon the USFS's determination that because these areas were heavily burned they are not likely to be suitable owl habitat. According to Bond-both in her declaration and in her published work-the California spotted owl uses burned areas for foraging in the short- term, and these areas may also provide important benefits in the long-term.

The FEISs cannot assume that simply because the owl habitat studies are preliminary, the adverse impacts discussed therein will not occur. Rather, the FEISs must respond explicitly and directly to conflicting views in order to satisfy NEPA's procedural requirements. By removing trees that might survive in areas it assumes to be unsuitable for California spotted owl habitat, the FEISs allow logging in what could well be suitable habitat. The FEISs do not explain in any detail how their determinations that habitat was "unsuitable" were made, and do not investigate or analyze how redrawing the boundaries of the PACs and HRCAs might negatively impact the owls. We therefore conclude that the FEISs do not satisfy the requirement under NEPA that the agency take a "hard look" and that there be a "full and fair discussion" allowing informed public participation and informed decision-making.

B. Likelihood of Success on the Merits of Earth Island's Challenge under the NFMA

Earth Island argues that the FEISs' reliance on the Breeding Bird Survey ("BBS") does not satisfy its obligations under the NFMA to conduct population surveys for certain Management Indicator Species ("MIS") bird species. The NFMA requires that a forest plan "comply with substantive requirements of the [NFMA] designed to ensure continued diversity of plant and animal communities and the continued viability of wildlife in the forest . . . ." Austin, 430 F.3d at 1063; see also 16 U.S.C. § 1604(g)(3)(B). The 2001 Framework identifies certain birds as MIS species, for which increased population monitoring is required. The 2004 Supplement incorporates the 2001 Framework's population-monitoring requirements.

The El Dorado National Forest Land and Resource Management Plan ("LRMP"), as well as both the Freds and Power FEISs, list cavity-nesting birds, including the black-backed woodpecker, hairy woodpecker, and Williamson's sapsucker, as MIS species. An MIS species is a bellwether, or class representative, "for other species that have the same special habitat needs of population characteristics." Inland Empire Pub. Lands Council v. U.S. Forest Serv.,88 F.3d 754, 762 n. 11 (9th Cir. 1996). The 2001 Framework states, "[p]opulation and/or habitat monitoring will be conducted for all MIS and species at risk. Varying levels of monitoring will be conducted depending on the level of concern associated with each species; as the level of concern about a species increases, the investment in monitoring increases." The 2001 Framework allows for a very limited degree of habitat monitoring in lieu of actual population moni toring, stating that "coarse habitat relationships constitute a relatively insensitive index to the status of populations and would only be appropriate for species with a lower level of concern or for which the status of the population were also being monitored."

According to the 2001 Framework, the hairy wood pecker and Williamson's sapsucker are low-vulnerability MIS species. Low-vulnerability species are monitored to determine changes in their distribution. Distribution data consist of "changes in the presence of species across a number of sample locations" and is a "spatially explicit version of frequency of occurrence data." In addition, the 2001 Framework notes that in an area as large as the Sierra Nevadas, "changes in the distribution of species represent ecologically significant information on the status and change of populations." Appendix E of the 2001 Framework makes explicit that population data must be collected for the hairy woodpecker and Williamson's sapsucker. The black-backed woodpecker is not specifically mentioned in the 2001 Framework, but it is designated as an MIS species in the LRMP and in the two FEISs.

As a preliminary matter, we conclude that NFMA regulations promulgated in 1982 apply to the 2001 Framework and 2004 Supplement. These regulations require population monitoring. See 36 C.F.R. § 219. Because the 2001 Framework and 2004 Supplement were developed based on regulations in effect before November 9, 2000, transitional rules, now contained at 36 C.F.R. § 219.14, govern this case. The applicable regulation provides:

For units with plans developed, amended, or revised using the provisions of the planning rule in effect prior to November 9, 2000, the Responsible Official may comply with any obligations relating to manage ment indicator species by considering data and analysis relating to habitat unless the plan specifi cally requires population monitoring or population surveys for the species. Site-specific monitoring or surveying of a proposed project or activity area is not required, but may be conducted at the discretion of the Responsible Official.

36 C.F.R. § 219.14(f) (emphasis added).

The district court found that in light of the hairy woodpecker and Williamson's sapsucker's classification as "low vulnerability" species, a lower level of moni toring is envisioned for these birds, and that the use of BBS data satisfies this lower level of monitoring. Earth Island argues that the district court erred because the BBS data are insufficient and inaccurate.

The BBS is a cooperative program sponsored by the governments of the United States and Canada to monitor several North American bird species. Under the BBS, the Williamson's sapsucker and black-backed woodpecker are listed in the "red" category, meaning that the results are "very imprecise" and the data suffer from low regional abundance and small sample sizes. The hairy woodpecker is listed in the "blue" category, which reflects data of "moderate precision" and of "moderate abundance," but which still "may not provide valid results." The BBS notes other potential problems in its overall data collection, specifically that the trends do not take into account activity outside of the range of the survey; that the surveys are only conducted by roadside, and "may not be representative of regional habitat changes"; and that within the range of the survey, many habitats are not well covered and that species within those habitats are consequently "poorly sampled."

Although not controlling on this court, Sierra Club v. Eubanks, 335 F. Supp. 2d 1070 (E.D. Ca. 2004), is instructive. In Eubanks, the same district judge as in this case granted a preliminary injunction against the logging and proposed timber sale provided in the Red Star Restoration Project. Id. at 1073. The Tahoe Forest Plan expressly provided for annual population monitoring of MIS species, and it listed each species "'for which population trend data is expected to be obtained.' " Id. at 1081 (quoting the administrative record). The USFS argued that population survey information was not required as long as MIS habitats were adequately analyzed, and that because the Red Star Restoration Project would not diminish the habitat for MIS species, actual monitoring was not required. Id. The district court agreed with this premise, stating that "[h]abitat analysis is an acceptable substitute for population trend data if there is enough underlying data to support such an analysis . . . ." Id. at 1082. Nevertheless, the district court concluded that the Red Star Restoration Project failed to comply with the El Dorado LRMP, the Sierra Nevada Framework, and the NFMA because underlying data for MIS species that could provide the necessary information for a habitat analysis did not exist. Id.

In its order denying Earth Island's request for a preliminary injunction, the district court distinguished Eubanks by concluding that in this case, such under lying data, in the form of the BBS, do exist. In support of this conclusion, the district court relied upon Forest Conservation Council v. Jacobs, 374 F. Supp.2d 1187, 1207 (N.D. Ga. 2005). In Jacobs, the district court found that the BBS provided sufficient data. However in Jacobs, the USFS relied on other data in addition to the BBS. Id. at 1205.

We do allow the USFS to conduct habitat analyses in place of population monitoring under certain circum stances. In Inland Empire Public Lands Council, various environmental groups claimed that an EIS did not perform a proper population analysis under the NFMA for several sensitive species living in a project area, including the black-backed woodpecker. 88 F.3d at 759. We upheld the USFS's use of a habitat manage ment analysis where the USFS had (1) consulted field studies showing how many acres of territory an individual species needed; (2) assumed that the amount of acreage remained constant no matter the actual size of the individual species' territory; and then (3) examined the proposed alternatives to see how many acres of necessary habitat remained after the timber was harvested. Id. We also upheld the USFS's decision to not engage in a detailed analysis of one species be cause nesting and feeding habitat requirements were not available, stating that "an analysis that uses all the scientific data currently available is a sound one." Id. at 762.

The USFS relies upon Inland Empire to argue that by analyzing the amount of habitat affected, rather than direct population counts of the birds, it has satisfied the NFMA. This argument fails for two reasons. First, a plain reading of the regulations does not support the USFS's argument. The transitional rules state that the USFS may use habitat analysis in lieu of population monitoring only when a forest plan does not specifically require population monitoring. The 2001 Framework and 2004 Supplement provide that the USFS may use "population monitoring and/or habitat analyses." How ever, in discussing the hairy woodpecker and William son's sapsucker, the Framework expressly requires "population monitoring," specifically in the form of "distribution data." It is difficult to see how distribution data could effectively be gathered in the absence of actual population monitoring, and we reject the USFS's argument that it is under no obligation to determine population trends for the hairy woodpecker or Williamson's sapsucker.

Second, although we agree that the USFS could have relied on habitat monitoring for the black-backed woodpecker, the USFS has not conducted a habitat analysis on the level of that found satisfactory in Inland Empire. See Idaho Sporting Cong. v. Thomas, 137 F.3d 1146 (9th Cir. 1998) (allowing habitat analysis under the NFMA in certain circumstances). The FEISs discuss various studies of black-backed woodpeckers that confirm their preference for burned forest habitat. Table 3-52 of the Power FEIS groups cavity-nesting birds into three different species groups, lists the areas included in the project areas that are "assumed to provide high and moderate capability habitat," and then lists the number of available acres of high and moderate capability habitat. There is no indication that the USFS consulted current or accurate field studies to arrive at these numbers, and there is no identification of the methodology used in determining what constitutes suitable habitat.

As we stated in Native Ecosystems Council, "[o]ur case law permits the Forest Service to meet the wildlife species viability requirements by preserving habitat, but only where both the Forest Service's knowledge of what quality and quantity of habitat is necessary to support the species and the Forest Service's method for mea suring the existing amount of that habitat are rea sonably reliable and accurate." 428 F.3d at 1250. In Austin, we noted that the black-backed woodpecker is "particularly dependent upon post-fire landscapes." 430 F.3d at 1065. There, we concluded that the USFS had failed to provide a factual basis sufficient to satisfy the NFMA because the EIS did not indicate how the USFS determined that habitat levels were not critically low or how it planned to generate additional habitat to coun teract losses from logging. Id. at 1068. We likewise conclude here that because the USFS's habitat analysis of the black-backed woodpecker has not provided a factual basis for determining the quantity or quality of suitable habitat, its analysis does not satisfy the NFMA.

We also conclude that the USFS's reliance on BBS data to meet its population monitoring obligation was arbitrary and capricious. Table 3-53 in the Power FEIS (Table 3-54 in the Freds FEIS) is entitled "Population trend of cavity-nesting birds." Immediately beneath the table, the FEIS states, "NOTE: based upon Breeding Bird Survey routes in the Sierra Nevada physiographic province, from 1996-2003." Table 3-53 then proceeds to list the population status of the hairy woodpecker as "Decreasing" and the population status of the Williamson's sapsucker as "Unknown." The population status of the black-backed woodpecker is also listed as "Unknown." A footnote to the hairy woodpecker figure states, "Data is not statistically significant; tendency is estimated using the population trend classification system described in Siegel and deSante (1999)."

The 2004 Supplement specifically mentions the need for annual monitoring of MIS species, stating that only after a period of annual monitoring will there be "sufficient understanding of important habitat char acteristics that we can confidently monitor habitat with out annual monitoring of species' distribution and abundance." We therefore disagree with the USFS that annual monitoring is not required. The USFS has not complied with 36 C.F.R. § 219.19 because it has not sufficiently monitored the population of the hairy woodpecker and Williamson's sapsucker. We hold that the BBS alone cannot satisfy the population monitoring requirement, and the USFS has acted arbitrarily and capriciously under the NFMA in relying upon it. Cf. Colo. Wild v. U.S. Forest Serv., 299 F. Supp. 2d 1184, 1189 (D. Colo. 2004) (holding that where an FEIS stated that it "lacks qualitative data" for one species, had "no specific data" for another, and was "currently in the process of establishing a monitoring program" for a third, the USFS had not met its monitoring obligations under the 1982 regulations); Forest Guardians v. U.S. Forest Serv., 180 F. Supp.2d 1273, 1282 (D.N.M. 2001) (holding that under the 1982 regulations, the BBS did not satisfy the monitoring requirement where the district court could not tell whether population surveys had been conducted for the area at issue). With respect to the black-backed woodpecker, we also hold that the USFS has acted arbitrarily and capriciously by relying on inadequate habitat monitoring data. See Lands Council v. Powell, 395 F.3d 1019, 1036 (9th Cir. 2005) (holding that the USFS violated the NFMA where the data for a habitat analysis were outdated and featured inaccurate estimates).

The USFS's approval and implementation of both Projects without appropriate or sufficient population and habitat data is contrary to the NFMA and governing provisions of the forest plan. The district court erred in finding otherwise.

C. Overall Likelihood of Success on the Merits

For the foregoing reasons, we conclude that Earth Island has shown a "strong likelihood of success on the merits" of both its NEPA and NFMA challenges to the Power and Freds FEISs and RODs. Earth Island, 351 F.3d at 1297. We now evaluate the remaining criteria for granting a preliminary injunction.

D. Possibility of Irreparable Injury, Balance of Hardships, and Advancement of the Public Interest

Because Earth Island has shown "a strong likelihood of success on the merits," it need only show "the possibility of irreparable injury" if preliminary relief is not granted, and that the "balance of hardships" tips in its favor. Id. Further, in this case it is appropriate that it be required to show "the advancement of the public interest." Id.; Kootenai Tribe v. Veneman, 313 F.3d 1094, 1125 (9th Cir. 2002) ("[W]here the purpose of the challenged action is to benefit the environment, the public interest must be taken into account in balancing the hardships").

The "possibility of irreparable injury" has clearly been shown. We have stated that "[e]nvironmental injury, by its nature, can seldom be adequately re medied by money damages and is often permanent or at

least of long duration, i.e., irreparable." Nat'l Parks & Conservation Ass'n v. Babbitt, 241 F.3d 722, 737 (9th Cir. 2001) (internal quotations omitted). In this case, Earth Island contends that the Power and Freds Projects may result in the unnecessary cutting of trees that would otherwise survive, in harm to the California spotted owl, and in harm to several MIS bird species. Earth Island has certainly shown that there is a possibility of such injury, and that such injury is irreparable.

The balance of hardships also tips in Earth Island's favor. The USFS and SPI contend, with some reason, that they will suffer economic losses if we enjoin the timber sales. But in Babbitt, we stated that a cruise ship's "loss of anticipated revenues . . . does not outweigh the potential irreparable damage to the en vironment." Id. at 738. Further, in Earth Island we noted the importance of preserving the public's interest in "preserving precious, unreplenishable resources." Earth Island, 351 F.3d at 1309 (quoting Kootenai Tribe, 313 F.3d at 1125).

Finally, we believe that a preliminary injunction advances the public interest. The preservation of our environment, as required by NEPA and the NFMA, is clearly in the public interest.

Conclusion

We have noticed a disturbing trend in the USFS's recent timber-harvesting and timber-sale activities. See, e.g., Ecology Ctr., Inc. v. Austin, 430 F.3d 1057 (9th Cir. 2005) (holding that the USFS's post-fire treatment of old-growth forest stands in the Lolo National Forest violated both the NFMA and NEPA, and that the EIS failed to explain adequately the adverse impacts of the

proposed plan on the black-backed woodpecker); Lands Council v. Powell, 395 F.3d 1019 (9th Cir. 2005) (re versing the district court's grant of summary judgment to the USFS because its EIS did not take a "hard look" at past timber harvests or current trout habitat con ditions); Idaho Sporting Cong. v. Rittenhouse, 305 F.3d 957 (9th Cir. 2002) (remanding to the district court to enjoin two timber sales approved in violation of the NFMA and NEPA). See also Utah Envtl. Cong. v. Bosworth, 421 F.3d 1105(10th Cir. 2005) (holding that the USFS did not properly monitor MIS species and did not consider a reasonable range of alternatives in a proposed timber-harvesting project); Sierra Club v. Eubanks, 335 F. Supp.2d 1070 (E.D. Cal. 2004) (granting a preliminary injunction against salvage logging pro vided for in the USFS's post-fire Red Star Restoration Project); Sierra Club v. Bosworth, 199 F. Supp.2d 971 (N.D. Cal. 2002) (rejecting the USFS's argument that post-fire salvage burning was needed to prevent a future fire and enjoining implementation of post-fire salvage logging); Colo. Wild v. U.S. Forest Serv., 299 F. Supp.2d 1184 (D. Colo. 2004) (granting a preliminary injunction of a timber salvage project because the USFS failed to gather population data for MIS species); Forest Guardians v. U.S. Forest Serv., 180 F. Supp.2d 1273(D. N.M. 2001) (reversing authorization of a timber sale in the Cibola National Forest because of the USFS's failure to collect adequate MIS population data).

It has not escaped our notice that the USFS has a substantial financial interest in the harvesting of timber in the National Forest. We regret to say that in this case, like the others just cited, the USFS appears to have been more interested in harvesting timber than in complying with our environmental laws.

We reverse the district court's denial of Earth Island's request for a preliminary injunction, and we remand to the district court for proceedings consistent with this opinion. Our injunction pending appeal shall remain in effect for 30 days following the issuance of the mandate in order to allow the district court sufficient time to fashion such preliminary injunctive relief as it deems appropriate consistent with this opinion.

REVERSED AND REMANDED.

NOONAN, Circuit Judge, concurring:

There is no doubt that the district court articu lated the wrong standard as to the possibility of harm that the plaintiffs must show. Earth Island v. United States Forest Service, 351 F.3d 1291, 1298 (9th Cir. 2003) (Earth Island I). There is also no doubt in my mind that the financial interest of the Forest Service requires further investigation and evaluation. See id. at 1309 (Noonan, J., concurring). That the parties have not pursued this problem does not give the Forest Service a pass. If it is indeed a biased adjudicator, its deter mination is a nullity. Tumey v. Ohio, 273 U.S. 510, 522, 47 S. Ct. 437, 71 L.Ed. 749 (1927).

It is not too difficult for a court of appeals to discern the correct legal standard for an injunction and to see the problem created by a financial interest on the part of the adjudicator. It is more difficult for this court to master the massive record in an environmental case and to be confident in its discrimination between expert opinions. Recognizing the mastery of the available data that distinguishes the majority opinion, I cannot say that I am sure as to Earth Island's probable success. I would remand to the district judge (1) to obtain informa tion as to the importance of the sales to the Forest Service's operation; (2) to apply the correct legal stan dard; and (3) to make its own estimate of the probability of Earth Island's success on the merits.

APPENDIX B

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

__________

No. 05-16776

D.C. No. CV-05-01608-MCE

Eastern District of California,
Sacramento

EARTH ISLAND INSTITUTE, A CALIFORNIA PROFIT ORGANIZATION; CENTER FOR BIOLOGICAL DIVERSITY, A NON-PROFIT ORGANIZATION,

PLAINTIFFS-APPELLANTS

v.

2. UNITED STATES FOREST SERVICE; DALE BOSWORTH, CHIEF OF THE UNITED STATES FOREST SERVICE; JOHN BERRY, FOREST SUPERVISOR FOR EL DOREDO NATIONAL FOREST, DEFENDANTS-APPELLEES

3. SIERRA PACIFIC INDUSTRIES, DEFENDANT- INTERVENOR-APPELLEE

__________

[Filed: Jan. 11, 2006]

__________

ORDER

BEFORE: NOONAN, TASHIMA, and W. FLETCHER, Circuit Judges.

We reconsider our September 21, 2005 order denying plaintiff Earth Island's request for an emergency injunction pending appeal. We hereby grant the injunc tion and order the United States Forest Service and Sierra Pacific Industries immediately to cease all logging operations in the areas encompassed by the Freds Fire Restoration Project and the Power Fire Restoration Project, pending our determination of the appeal on the merits. Because of the expected short duration of this injunction, no bond or other security shall be required.

1. APPENDIX C UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT __________ No. 05-16776 D.C. No. CV-05-01608-MCE Eastern District of California, Sacramento EARTH ISLAND INSTITUTE, A CALIFORNIA PROFIT ORGANIZATION; ET AL., PLAINTIFFS-APPELLANTS v. 2. UNITED STATES FOREST SERVICE; AN AGENCY OF THE UNITED STATES DEPARTMENT OF AGRICULTURE, DEFENDANTS-APPELLEES AND 3. SIERRA PACIFIC INDUSTRIES, 4. INTERVENOR-APPELLEE __________ [Filed: Sept. 21, 2005] __________ ORDER Before: REINHARDT and RYMER, Circuit Judges. The United States Forest Service's motion to exceed opposition page limits is granted. The Clerk shall file the opposition received September 19, 2005. Appellants' motion to exceed reply page limits is granted. The Clerk shall file the reply received on September 20, 2005. Sierra Pacific Industries' motion for leave to intervene as a non-aligned appellee is granted for pur poses of this appeal only. Appellants' emergency motion for an injunction pending appeal is denied. This is a preliminary injunction appeal. The court sua sponte expedites the briefing and hearing of this appeal. The provisions of Ninth Circuit Rule 31-2.2(a) shall not apply to this appeal. Any motions to extend time to file briefs will be strongly disfavored. Each party shall serve its brief on opposing counsel by overnight mail or electronic means. The briefing schedule is set as follows: the opening brief, answering brief, and non-aligned intervenor's brief are due not later than October 11, 2005. If appellants fail to file timely the opening brief, this appeal will be dismissed automatically by the Clerk for failure to prosecute. See 9th Cir. R. 42-1. The parties are reminded of the court's preference for joint briefing. See 9th Cir. R. 28-4. This appeal and any motions pending when briefing is completed shall be referred to the next available motions panel for disposition. See 9th Cir. R. 3-3(d).

APPENDIX D

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF CALIFORNIA

__________

No. CIV. S 05-1608 MCE PAN

 

EARTH ISLAND INSTITUTE, A CALIFORNIA NON- PROFIT, AND CENTER FOR BIOLOGICAL DIVERSITY, A NON-PROFIT CORPORATION, PLAINTIFFS

v.

UNITED STATES FOREST SERVICE, DALE BOSWORTH, IN HIS OFFICIAL CAPACITY AS CHIEF OF THE U.S. FOREST SERVICE, AND JOHN BERRY, IN HIS CAPACITY AS FOREST SUPERVISOR FOR THE ELDORADO NATIONAL FOREST, DEFENDANTS

__________

[Filed: Aug. 25, 2005]

__________

MEMORANDUM AND ORDER

 

Plaintiffs, Earth Island Institute and the Center for Biological Diversity (hereinafter collectively referred to as Plaintiffs) seek to enjoin implementation, by Defendant United States Forest Service ("USFS") of two post-fire restoration projects in the Eldorado National Forest--the Freds Fires Restoration Project ("Freds Project") and the Power Fire Restoration Project ("Power Project"). Those projects, which pro vide for the logging of fire-damaged trees in the wake of two wildfires occurring in October of 2004, are claimed by Plaintiffs to violate the provisions of the 2004 Sierra Nevada Forest Plan Amendment (the "Framework"), the National Forest Management Act ("NFMA"), 16 U.S.C. § 1600, et seq., the National Environmental Policy Act ("NEPA"), 42 U.S.C. § 4321, et seq. In accordance with the provisions of the Administrative Procedures Act ("APA"), Plaintiffs now seek injunctive relief. The Court granted a temporary restraining order on August 18, 2005 pending its consideration of Plaintiffs' instant request for a preliminary injunction pending resolution of the merits of this case. Having now considered the matter further, and for the reasons set forth below, the Court denies the requested preliminary injunction.

BACKGROUND

On October 6, 2004, the Power Fire was reported in Amador County, California approximately 17 miles east of the community of Pioneer. That fire ultimately consumed nearly 17,000 acres, 14,255 of which were situated in the Eldorado National Forest. Approxi mately 48 percent of the burned area, or more than 8, 100 acres, burned with high intensity, killing 75 to 100 percent of trees and incinerating duff and litter protecting the soils. Def.'s Ex. 1 (Power ROD) at 1.

Less than ten days later, on October 13, 2004, another human-caused fire of unknown origin com busted within portions of the Eldorado National Forest near Silver Fork and Kyburz. That fire, denominated as the Freds Fire, encompassed a total of some 7,700 acres, including 4600 acres within the National Forest, 2600 of which burned at high intensity levels comparable to the Powers fire. Def's Ex. 6 (Freds FEIS) at 1.

Both the Power and Freds projects were designed to restore portions of the fire-ravaged landscape falling within the Eldorado National Forest. The USFS identified four objectives in that regard: 1) to reduce long-term fuel loading in order to reduce future fire severity and resistance to control; 2) to improve road drainage and establish effective ground cover in severely burned areas to alleviate erosion and sedi mentation to streams; 3) to remove certain dead trees while they retain economic value; and 4) to reduce safety hazards to the public and forest workers from trees falling in the future. Def's Ex. 6 at 3-6, Def.'s Ex. 1 at 1.

Final Environmental Impact Statements ("FEISs") were issued for both projects on July 1, 2005. There after, on August 1, 2005, a Record of Decision imple mented Alternative One from the FEIS for the Power Project, which permitted timber harvest on some 5574 acres of the total acreage burned by the fire within the Eldorado National Forest (13,611 acres), and which also contemplated additional harvesting on another 2,600 acres depending on ultimate tree mortality. The project incorporates snag (dead tree) retention areas of various sizes, creates no harvest zones along perennial streams, and calls for the preservation of additional dead trees in other harvest areas of various densities and configura tions in order to accommodate wildlife dependent on such habitat. Tree harvest would be accomplished using skyline and helicopter methods on steeper slopes to minimize soil impacts, and slash (debris) from har vesting operations would be treated, so as to ensure acceptable fuel loading levels while, at the same time, creating ground cover as needed to reduce erosion. A similar Record of Decision was made with respect to the Freds Project, also on August 1, 2005, which allowed timber cutting on some 2900 acres under Alternative Four of the Freds FEIS. That harvest was made under guidelines similar to those identified with respect to the Powers Project as identified above.

Because the USFS determined that a delay in imple menting the projects pending the standard adminis trative appeal period would cause deterioration of timber to be salvaged and would accordingly reduce revenues generated by timber sales for reforestation efforts within project areas, an emergency deter mination was obtained from the Regional Forester which allowed both projects to commence in August of 2005.

Through the present lawsuit, Plaintiffs have chal lenged both the Powers and Freds Projects on grounds 1) that the respective FEISs failed to ensure scientific integrity in predicting mortality rates of trees to be cut and in estimating post-fire soil cover levels; 2) that the FEISs failed to compile required population monitoring data for certain Management Indicate Species ("MIS") with the Eldorado National Forest; and 3) that the FEISs failed to adequately assess impacts to the California Spotted Owl. Plaintiffs claim that irreparable injury may result if the projects are allowed to proceed in the face of those inadequacies, which they claim run afoul of the provisions of the Framework, the NFMA, and NEPA. A preliminary injunction is sought on that basis.

STANDARD

A preliminary injunction is an extraordinary remedy and Plaintiff have the burden of proving the propriety of such a remedy by clear and convincing evidence. See Granny Goose Foods, Inc. v. Teamsters, 415 U.S. 423, 442 (1974). In order to warrant issuance of a pre liminary injunction, a party must demonstrate either: 1) a combination of probable success on the merits and the possibility of irreparable injury; or 2) that serious questions are raised and the balance of hardships tips sharply in favor of granting the requested injunction. Stuhlbarg Int'l Sales Co., Inc. v. John D. Brush & Co., Inc., 240 F.3d 832, 839-40 (9th Cir. 2001). These two alternatives represent two points on a sliding scale, pursuant to which the required degree of irreparable harm increases or decreases in inverse correlation to the probability of success on the merits. Roe v. Anderson, 134 F.3d 140-0, 1402 (9th Cir. 1998); United States v. Nutri-cology, Inc., 982 F.2d 1374, 1376 (9th Cir. 1985). Under either formulation of the test for granting a preliminary injunction, however, Plaintiffs must demon strate a significant threat of irreparable injury. Oak land Tribune, Inc. v. Chronicle Publishing Co., 762 F.2d 1374 (9th Cir. 1985).

Here a preliminary injunction is sought in the con text of a lawsuit charging the USFS with violations of both NEPA and NFMA as well as provisions of the Administrative Procedure Act, 5 U.S.C. § 701 et seq. NEPA imposes a procedural requirement that an agency contemplate the environmental impacts of its actions (Idaho Sporting Congress v. Thomas, 137 F.3d 1146, 1149 (9th Cir. 1998), with a detailed environmental impact statement required for actions significantly affecting the quality of the human environment. 42 U.S.C. § 4332(c). An agency must take a "hard look" at the consequences, environmental impacts, and adverse environmental effects of a proposed action within an environmental impact statement, when required. Kleppe v. Sierra Club, 427 U.S. 390, 410, n. 21 (1976). If an EIS adequately discloses such effects, NEPA's goal is satisfied. Inland Empire Public Lands Council v. United States Forest Serv., 88 f.3d 754, 758 (9th Cir. 1996). As the Ninth Circuit stated in Inland Empire, "NEPA exists to ensure a process, not to ensure any result." 88 F.3d at 758 (emphasis in original).

NEPA does not mandate that an EIS be based on a particular scientific methodology, nor does it require a reviewing court to weigh conflicting scientific data. Friends of Endangered Species, Inc. v. Jantzen, 760 F.2d 976, 986 (9th Cir. 1985). An agency must be permitted discretion in relying on the reasonable opinions of its own qualified expert, even if the court might find contrary views more persuasive. See, e.g., Kleppe, 427 U.S. at 420, n. 21. NEPA does not allow an agency to rely on the conclusions and opinions of its staff, however, without providing both supporting analysis an data. Idaho Sporting Congress, 137 F.3d at 1150. Credible scientific evidence that contraindicates a proposed action must be evaluated and disclosed. 40 C.F.R. § 1502.9(b).

NFMA requires that "resource plans and permits, contracts, and other instruments for the use and occupancy of National Forest System lands shall be consistent with the land management plans." 16 U.S.C. § 1604(i). Consequently all activities in USFS forests, including timber sales, must be determined to be consistent with the governing forest plan, which is a broad, programmatic planning document. See, e.g., Wilderness Society v. Thomas, 188 F.3d 1130, 1132 (9th Cir. 1999). A site specific project may proceed only if it is not only consistent with the forest plan but also has been analyzed under NEPA as discussed above. See, e.g., Idaho Conservation League v. Mumma, 956 F.2d 1508, 1512 (9th Cir. 1992).

Because neither NEPA nor NFMA contains provisions allowing a private right of action (see Lujan v. National Wildlife Federation, 497 U.S. 871, 882 (1990) and Ecology Center Inc., v. United States, 192 F.3d 922 for this proposition under NEPA and NFMA, respectively), a party can obtain judicial review of alleged violations of NEPA and NFMA only under the waiver of sovereign immunity in the Administrative Procedure Act ("APA"), 5 U.S.C. §§ 701-706. Under the APA, the court must determine whether, based on a review of the agency's administrative record, the EIS was "arbitrary and capricious," outside the scope of the agency's statutory authority, or otherwise not in accordance with the law. Salmon River Concerned Citizens, 32 F.3d at 1356. In conducting this review, the standard to be employed is decidedly deferential to the agency's expertise. Id.

ANALYSIS

A. Tree Mortality and Soil Cover Guidelines

Plaintiffs take issue with the methodology employed as well as the scientific accuracy of the FEISs for both projects with respect to estimates of tree mortality and post-fire soil cover. In addition, with respect to trees within the project area that are indeed dead, Plaintiffs argue that Defendant's fuel concern about retaining a given number of snags per acre is unfounded and unwar ranted. Examination of the FEIS, however, shows that the Forest Service identified the methodologies it used in reaching conclusions in these areas, provided sup porting data for those conclusions, and cited the scien tific literature relied upon for said conclusions. In addi tion, in opposing the instant request for preliminary injunctive relief, the USFS has submitted additional expert declarations from Sheri L. Smith and from Jeffrey D. Tenpas which buttress the validity of the scientific methodology employed in the FEISs.

While Plaintiffs have submitted declarations from their own experts, and specifically from Dr. Edwin Royce (attacking the accuracy of the mortality guide lines chosen by the USFS) and from Jon Rhodes (questioning the extent of effective soil cover), an agency like the USFS must have the discretion to rely on the reasoned opinions of its own experts, even in the face of opposing opinion from other sources. Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 378 (1989). As indicated above, NEPA does not mandate that an EIS be based on a particular scientific methodology, and does not require a reviewing court to weigh conflicting scientific data. Friends of Endangered Species, Inc. v. Jantzen, 760 F.2d at 986. Instead, courts must "defer to agency expertise on questions of methodology unless the agency has completely failed to address some factor, consideration of which was essential to a truly informed decision." Bear Lake Watch, Inc. v. Federal Energy Regulatory Comm., 324 F.3d 1071, 1077 (9th Cir. 2003), quoting Inland Empire Pub. Lands Council v. Schultz, 992 F.2d 997 (9th Cir. 1983). Review of FEISs like those involved here is consequently limited, and decidedly deferential to the agency's expertise. Salmon River Concerned Citizens v. Robertson, 32 F.3d 1346, 1356 (9th Cir. 1994). An agency's determination may conse quently be reversed under the APA only if it acted in a manner deemed "arbitrary, capricious, an abuse of discretion or otherwise not in accordance with the law." 5 U.S.C. § 706(2); see Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416 (1971).

Because this Court cannot say that the methodology utilized by the USFS with respect to tree mortality/ retention and soil cover is unreasonable and/or un founded, it finds that there is no reasonable probability of success as to those claims that would justify a pre liminary injunction. Specifically, Plaintiffs have not shown, as they must, that the actions of the USFS were arbitrary and capricious. Moreover, as discussed in more detail below, even if Plaintiffs were to raise serious questions as to the validity of the USFS methodology, which the Court does not believe they have accom plished, a balance of hardships analysis still does not weigh sharply in favor of granting the relief requested by Plaintiffs, a prerequisite for granting injunctive relief in this instance. Stuhlbarg, 240 F.3d at 839-840. Con sequently Plaintiffs have not demonstrated an entitle ment to a preliminary injunction by questioning the approach utilized by the USFS with respect to tree mortality and soil cover issues.

B. MIS Monitoring Obligations

Plaintiffs argue that the FEISs for both projects fails to include population monitoring data for two types of woodpeckers, the Hairy woodpecker and the Williamson's sapsucker. According to Plaintiffs, be cause both of those species are denominated as MIS, and because the provisions of the 2001 Framework (as incorporated within its 2004 successor) call for popula tion monitoring, the requirements of both the Frame work and the NFMA have necessarily been violated. While Plaintiffs appear to contend that such violations, on their face, warrant issuance of a preliminary injunc tion, in this Court's view the analysis is not quite so simple.

First, the Framework provisions upon which Plain tiffs rely are equivocal in terms of just what data must in fact be obtained. Although the list of MIS and Species at Risk identified in the 2001 Framework (Pl.'s Ex. J., Appendix E-64, as adopted by the 2004 Frame work, see Pl.'s Ex. K, p. 70) do contain a checkmark indicating that population monitoring is required next to the categories for the Hairy woodpecker and the Williamson's sapsucker, the same table reflects that the vulnerability level for both species is "low". With respect to that vulnerability rating, the Framework goes on to say that "varying levels of monitoring will be conducted depending on the level of concern associated with each species; as the level of concern about a species increases, the investment in monitoring increases." (Pl.'s Ex. J, Appendix E-62). The Framework may consequently be read as envisioning a lower level of monitoring, and some flexibility, for birds with low vulnerability ratings like the woodpeckers at issue herein.

Significantly, unlike the situation confronted by this Court in Sierra Club v. Eubanks, 335 F. Supp. 2d 1070 (E.D. Cal. 2004), in this case there was available wood pecker population data available that was considered within the project FEISs. Specifically, the USFS used "Breeding Bird Survey" (BBS) population data to provide information on population trends of cavity- nesting birds within the Sierra Nevada region like the woodpecker. (Def.'s Ex. 2 (Power EFIS at 198, 242-243, 522; Def.'s Ex. 7 (Freds FEIS) at 217). The BBS is an ongoing cooperative program sponsored by the United States and Canadian governments to monitor may North American bird species. (See North American Breeding Bird Survey Homepage, http://www.pwrc.usgs.gov/bbs).

In Forest Conservations Council v. Jacobs, 374 F. Supp. 2d 1187, 1207 (N.D. Ga. 2005), the court approved the use of BBS surveys as "quantitative data" of an MIS species population. Moreover, and in addition, the Power and Fred project documents also examined scientific data on the population densities of another woodpecker at issue, the black-backed woodpecker, in pre- and post-fire habitats within the Central Sierra Nevada. Def.'s Ex. 2 (Power FEIS) at 243; Def.'s Ex. 22 (FREDS MIS Report) at 44.1 The Forest Service also examined relevant scientific literature assessing popula tion trends for cavity nesting species in general. See Id. at 15-17, 19, 44-45. Based on this data, the Court cannot say that the FEISs failed to take population into account, particularly in view of the "low vulnerability" rating assigned to the Hairy woodpecker and Williamson's sapsucker. Consequently the Court cannot say that Plaintiffs have demonstrated a probability of success on their claim that the USFS violated its obligations under the 2004 Framework with respect to population monitoring.

Even were the Court to determine that a violation of the Framework, and hence also the NFMA, occurred because proper monitoring was not performed for the Hairy woodpecker and the Williamson's sapsucker, that does not end the analysis. In Amoco Prod. v. Village of Gambell, 480 U.S. 531, 544 (1987), the Supreme Court rejected the notion that any violation of an environ mental statute (in that case, NEPA) raises a presump tion of irreparable injury. Instead, a court must balance the competing interests and consider the effect on each party and the public in determining whether injunctive relief is warranted.

Looking at the bigger picture, even if the Power and Freds Projects proceed without proper population monitoring for the two woodpecker species at issue, that does not necessarily mean that any immediate and irre parable injury will occur in the absence of that moni toring. Any preference by those birds for severely burned habitat may be more than satisfied by the untreated/unharvested portions of the project areas, which total some 36 percent for Freds and 40 percent for Power.2 Additionally, even within areas to be harvested some snags will be retained, and acreage specifically for cavity-nester habitat is set aside in both projects (some 750 acres of burned forest within the area to be logged in the Power project is set aside for that purpose, and the Freds project similarly sets aside 129 acres).3 In light of these factors, Plaintiffs have not demonstrated irreparable harm even if they can show a population monitoring violation.

C. California Spotted Owl Habitat

Prior to the 2004 fires, the area encompassed by the Power project contained nine owl protected activity centers ("PACs"). The acreage attendant to the Freds project contained three PACs, which are defined as 300 acres of the best owl habitat surrounding a known nest tree for the California Spotted Owl. It appears undis puted that much of the area occupied by these PACs was severely burned.4 Nonetheless, in an effort to accommo date the needs of the CASPO, the USFS not only re tained as much suitable owl habitat that remained within the project area, but also attempted to retain many of the previously constituted PACs by including some burned habitat. Under the 2004 Framework, a stand- replacing event like the project fires in this case are a valid reason for remapping affecting PACs. 2004 Framework ROD at 37.

In taking issue with the USFS' decision not to leave all prior PACs intact, Plaintiffs point to a December 2004 study suggesting that spotted owls, radio-tracked following a fire in southwestern Oregon, actually spent a significant amount of roosting and foraging time in moderate to severely burned forest areas. (Bond Dec'1., ¶ 19 and Ex. B). Plaintiffs consequently argue that neither FEIS considered that owls might actually use and benefit from burned patches within their own terri tories. Plaintiffs contend that some of the previously designated PACs were in fact occupied after the fires.

In response, the USFS points out that in assessing impact on the California Spotted Owl in both projects, it assumed that previously designated owl habitat was occupied. (Def.'s Ex. 2 (Power FEIS) at 185; Def's Ex. 7 (Freds FEIS) at 386; Loffland Decl., ¶ 2). The Forest Service further commenced surveys beginning in the Spring of 2005 to track owl return to PACs. Def's Ex. 2 at 185; Def.'s Ex. 7 at 202. Additionally, the USFS demonstrates that prevailing scientific evidence sup ports the notion that preferred spotted owl habitat entails two canopy layers and between 40 and 70 percent canopy cover, depending on whether the use in question is for nesting or for foraging. See Def.'s Ex. 13 (Power BA/BE) at 55, Loffland Decl., ¶ 3. Both FEISs also identified the study referenced by Plaintiffs and ex plained why that study was preliminary and not dis positive as to suitability. Def.'s Ex. 2 (Power FEIS) at 201-202; Def.'s Ex. 7 (Freds FEIS) at 196; Loffland Decl., ¶ 4). An agency like the USFS is entitled to wide discretion in assessing scientific evidence, as long as it takes the requisite "hard look" at the issues and re sponds to reasonable opposing viewpoints. Earth Island Institute v. U.S. Forest Service, 351 F.3d 1291, 1301 (9th Cir. 2003). Court review is limited to determining whether EISs like those involved here contain "a rea sonably thorough discussion of the significant aspects of the probable environmental consequences" of a proposed action. Oregon Envtl. Council v. Kunzman, 817 F.3d 484, 492 (9th Cir. 1987).

Given the above, Plaintiffs have not demonstrated entitlement to a preliminary injunction based on any shortcoming in either FEIS as to the California Spotted Owl. The FEISs for both projects appear to contain a reasonably thorough analysis, and consequently the requisite "hard look", as to owl impact.

D. Balancing of Interests

Inasmuch as this case involves logging of timber in a national forest following a severe wildfire, as stated previously (in the Court's August 18, 2005 Order grant ing a temporary restraining order) the issues presented are directly related to various public interests. Because such interests are implicated, the Court must determine whether the balance of public interests tips sharply in favor of granting injunctive relief in this matter. Caribbean Marine Serv. v. Baldridge, 844 F.2d 668, 674 (9th Cir. 1988). Even if Plaintiffs were successful in demonstrating some environmental harm, in balancing the relative hardships there is no presumption that environmental harm should outweigh other potential harm to the public interest. See Fund for Animals, Inc. v. Lujan, 962 F.2d 1391, 1400 (9th Cir. 1992).

As indicated above, the objectives of both the Power and Freds projects include fuel reduction designed to protect the forest and its surrounding communities from the threat of future devastating fire. The projects are also designed to protect the public and forest workers from falling trees, and revenue generated by the timber sales will contribute to reforestation activities that may be threatened in the absence or reduction of such revenue.5 Given the force of these interests, the Court cannot say that the balance of interests tips sharply in favor of granting a preliminary injunction in this case. This is particularly true given the fact, as already indicated, that much of the charred landscape created by the Power and Freds fires (to which Plaintiffs assign considerable environmental value) will remain untreated in any event.

CONCLUSION

This Court's review of the USFS' actions in approving the Power and Freds projects is governed by an "arbitrary and capricious standard. Biodiversity Legal Found. v. Badgley, 309 F.3d 1166, 1176-77 (9th Cir. 2002). Given that standard, and following con sideration of the factors to be considered in determining the propriety of injunctive relief, the Court finds that Plaintiffs have not demonstrated entitlement to a preliminary injunction in this case. Plaintiffs' motion for such relief is accordingly denied, and the Court's Temporary Restraining Order dated August 18, 2005 is vacated.

IT IS SO ORDERED.

DATED: August 25, 2005

/s/ MORRISON C. ENGLAND, JR.
MORRISON C. ENGLAND, JR.
UNITED STATES DISTRICT JUDGE

 

1 Plaintiffs have not demonstrated that this data fails to apply to the other two woodpecker species at issue in this case, the Hairy wood pecker and the Williamson's sapsucker.

2 These percentages only encompass untreated areas of the project areas themselves, which of course do not account for the total acreage burned by the fires. Adding the total burned areas to the equation indicates that some 52 percent of the acreage burned by the Power fire will not be treated by the Power project. The comparable figure for the Freds fire is approximately 62 percent.

3 It should also be noted that, according to Plaintiffs, any woodpecker preference for burned forest habitat as living and breeding habitat is usually only for four to six years post-fire, a relatively short period of time. See Pl.'s Opening Points and Authorities, p. 7.

4 Within the Powers project area, suitable owl habitat was reduced from 5,880 acres pre-fire to 2,750 acres post-fire. Def's Ex. 2, (Power FEIS) at 201. The reduction for the Freds project was from approxi mately 2,100 acres to 600 acres. Def.'s Ex. 7 (Freds FEIS) at 201, 226.

5 According to the USFS, delay in implementing the projects may result in timber deterioration amounting to $11.3 million in lost proceeds as to the Power project and $800,00 for the Freds project. Def.'s Ex. 3 (Power ESD Analysis); Def.'s Ex. 8 (Freds ESD Analysis).

1. APPENDIX E

RECORD OF DECISION
POWER FIRE RESTORATION
Eldorado National Forest
100 Forni Road
Placerville, CA 95667
El Dorado and Amador Counties, California

Introduction

The Power Fire was reported the afternoon of October 6, 2004, on the north side of the Salt Springs Reservoir Dam located approximately 17 air miles east of Pioneer, California in Amador County. The fire burned up an extremely steep, south facing, brush covered slope, spreading into timber and heavy fuels. The fire subsequently burned west to Panther Creek drainage, fueled by strong east winds. The Power Fire burned approximately 16,993 acres on the Eldorado National Forest (ENF) and on private timberlands. The project area is the approximately 12,237 acre portion of the Power Fire on National Forest lands within the Amador Ranger District administrative boundary outside of the Mokelumne Wilderness. The project area includes portions of the Cole Creek, Bear River, and Panther Creek watersheds.

The fire burned with varying intensity. Approxi mately 38 percent burned with low intensity, as mea sured using aerial photos and field sampling. Many areas of the fire (approximately 48%) burned at high intensity, killing 75 to 100 percent of the trees and burning the duff and litter that protect the soil. Another 13 percent burned with moderate intensity, killing 25 to 75 percent of the trees. In the high and moderate in tensity areas the fire resulted in high rates of soil erosion, sedimentation to streams, destruction of wildlife habitat for sensitive species, and loss of old forest. The fire killed tens of thousands of trees that, if left untreated, will contribute to extremely high fuel loading over time. As these dead trees fall and fuel accumulates, future fires will be even more severe. Without treatment to begin to restore the fire area, significant additional impacts to soil, water quality, cultural resources, and wildlife habitat are likely over the short and long term.

The Sierra Nevada Forest Plan Amendment Record of Decision (SNFP ROD) provides for ecosystem re storation following catastrophic events. These restora tion activities are included in all land allocations and call for managing disturbed areas for long term fuels pro files, restoring habitat, and recovering the value of some dead and dying trees (SNFP ROD, pg. 6). SNFP land allocations within the Power Fire boundary are 58 percent old forest emphasis, 27 percent threat zone, 4 percent defense zone, and 11 percent wilderness. Pro tected activity centers (PACs) for spotted owls and goshwaks, spotted owl home range core areas (HRCAs), and riparian conservation areas (RCAs) adjacent to perennial, seasonal and ephemeral, streams overlay the other land allocations. A portion of the Mokelumne Wilderness and the Salt Springs State Game Refuge are within the fire area. The Mokelumne River, Bear River, Beaver Creek, Cole Creek, and Green Creek, having outstandingly remarkable cultural resource values, are eligible for possible inclusion in the National Wild and Scenic River system. Of the 15,398 acre South Fork of the Mokelumne Archaeological District, 11,131 acres are located within the Power Fire. This District has been determined to be eligible for inclusion in the National Register of Historic Places. Due to its high archaeologi cal significance, the Mokelumne Canyon is designated as a Special Interest Area with the Eldorado's Land and Resource Management Plan. The statement of signifi cance concluded that the "Mokelumne Canyon is one of the most significant archaeological areas within the ENF.

The goal of this project is to move the project area toward desired future conditions as defined by the Sierra Nevada Forest Plan Amendment (SNFP ROD, pgs. 36-48). These desired conditions by land allocation are described in detail in Table 1-1 of the FEIS, pgs. 8- 9. Desired conditions for defense zones include surface and ladder fuel conditions such that crown fire ignition is unlikely. Desired conditions for threat zones include flame lengths less than four feet at the head of a fire, reductions in rate of spread and hazards to firefighters, and a doubling of fire line construction rates. In RCAs, desired conditions include meeting the goals of the Clean Water Act and Safe Drinking Water Act; species composition and structural diversity providing suitable habitat for aquatic and riparian associated species; soils with favorable infiltration characteristics; and the physi cal structure of streambanks and shorelines that mini mize erosion and sustain desired habitat diversity.

Purpose and Need for Action

1. There is a need to reduce long term fuel loading for the purpose of reducing future fire severity and resistance to control: Meeting the desired conditions for old forest and old forest dependent wildlife requires survival and growth of individual trees and forested stands over the next 100 to 250+ years without the occurrence of another stand replacing fire. Preventing another stand replacing fire involves a combination of fuel treatments to modify fire behavior and effective suppression. In areas where the fire intensity was moderate to high, surface fuel loading is very low. As dead trees fall over time surface fuels will increase significantly, affecting future fire behavior and suppression. Excessive large woody debris can lead to difficulty in suppressing wildfires. Hand fire line construction is significantly slowed where fire lines intersect numerous large logs. This is referred to as "resistance to control" and can lead to larger fires since

[Picture Omitted]

fire lines have to be relocated to areas of less woody debris or where tractor fire lines can be built. Large woody debris may also increase fire severity, further impacting impaired watersheds, soils, and archeological sites. Excessive small woody debris, from small trees and limbs of larger trees, increases a fire's rate of spread and fire line intensity affecting the ability to suppress the fire and the ultimate fire size. Dead trees that aren't removed, will contribute to extremely high fuel loading within 5 to 10 years. Many of the snags that were left standing following the Cleveland fire in 1992 have fallen over in the intervening 12 years. The potential for a wildfire start is high due to proximity to a Pacific Gas and Electric Company (PG&E) distribu tion line that runs through the canyon, recreational use, and lightning. Predicted fire behavior modeling of timber stands and fuel types that are representative of current conditions, indicates that high intensity fire with rapid rates of spread would be likely under moderate weather conditions1 [sic]. Without additional treatment to begin to reduce fuel loads, wildfires are likely over the short and long term.

2. There is a need to reduce sedimentation to streams and erosion from roads and hillslopes for the purpose of protecting soil productivity and water quality for bene fical uses: Portions of the project area that burned at high intensity are void of effective ground cover (limbs, twigs, small boles) and soil organic material as a result of the fire, leading to soil erosion and potential loss of productivity. Runoff and erosion rates are expected to

[Picture omitted]

be high, and many areas have steep slopes, increasing the hazard for soil erosion and transport of sediment to streams. Native surfaced roads have the potential to channel runoff and increase sedimentation to streams. Sediment delivered to streams impairs water quality and has the potential to impact downstream beneficial uses of water, including municipal water supplies. The Mokelumne River is the source of drinking water for the East Bay Municipal Utility District (EBMUD), which serves 1.3 million customers in the eastern San Francisco Bay Area.

[Picture omitted]

3. There is an urgent need to recover the volume and value of timber killed or severely injured by the fire for the purpose of generating funds to offset the cost of restoration activities: Dead trees deteriorate rapidly relative to wood quality, volume, and value. By the first year following the fire, most trees have significant worm holes and cracks, culling about one third of the cubic volume on trees between 11 and 24 inches diameter (Bodenhausen 2002). By the second year, 47 to 74 percent of the volume of trees less than 24 inches diameter is lost (Lowell et al. 1992). Larger trees deteriorate more slowly, but even trees as large as 40 inches diameter will lose nearly half their volume to decay by the second year (Lowell et al. 1992). This volume loss corresponds to significant value loss. The first year after a fire, the value of trees removed will cover the cost of their removal and associated fuel treat ment and can be used to pay for restoration work. Op portunities for restoration work that can be funded by this timber value include reducing erosion from roads, treatment of fuels, future reforestation, and wildlife and watershed improvement projects. By the second or third year after a fire, the value and volume loss may preclude any economic return, thereby jeopardizing the availability of funds for restoration and meeting the other purposes of the project.

[Picture omitted]

4. There is a need to reduce safety hazards to the public and forest workers: Dead and defective trees pose a significant safety hazard to forest visitors and workers.

Decision and Rationale for Decision

Decision

Based upon my review of the alternatives, I have decided to implement Alternative 4 as described in the Final Environmental Impact Statement (FEIS).

My decision is based upon a careful consideration of the information and analysis contained in the FEIS and all of the supporting documentation, including the public comments on the Draft Environmental Impact State ment (DEIS). This environmental analysis process was conducted in accordance with the National Environ mental Policy Act (NEPA) and the direction provided in the Forest Service Manual. Alternative 4 is consistent with the goals and objectives of the Eldorado National Forest LRMP as amended by the SNFPA.

Rationale for Decision

When compared to the other alternatives, Alterna tive 4 meets the purpose for the project and is the alternative that best meets most of the needs identified for this project. I considered each of the needs as a factor in my decision and provide the rationale for my evaluation of each alternative against those needs.

Needs:

1. The need to reduce long term fuel loading for the purpose of reducing future fire severity and resistance to control.

Given the estimated fire return interval for the Power Fire area of 9.1 years (FEIS pg 79) and the estimated length of time to achieve old forest conditions (100 to 250 years), it is inevitable that fires will burn within the project area before the old forest structure and function can return. Actions taken now can in fluence the extent and intensity of future fires by deter mining the amount and configuration of fuel load in the Power Fire area. Comparing the total fuel loading, mea sured in tons of down woody fuel over time, is one way to evaluate the alternatives (FEIS pg 58 comparison of alternatives chart). Alternative 4 best meets this need by retaining the least total fuel loading over most of the project area. This will help reduce the intensity of future fires and will reduce the "resistance to control" when fires do occur (FEIS pgs 93-100 and 109-110). Direct attack of small fires within the 732 acres of snag retention clumps and the PAC non-core areas may be precluded due to fire fighter safety. The areas outside of snag clumps and PAC non-core would have a low resistance to control, would be safe for firefighters to construct fire lines along the fire's edge, would contri bute to more effective fire suppression and contain fires to a smaller size.

The no action alternative, Alternative 1, results in high fuel loadings over time, particularly in the 1000- hour fuels (>3" diameter). There is abundant scientific evidence that increased fuel loads can result in in creased fire intensity and severity. The severity of fire effects and difficulty in fire suppression are primarily associated with the total amount of fuel available and consumed (Martin and Brackebusch 1974). In other words, given the same weather and topographic condi tions, areas with high fuel loads will release more energy (burn hotter), exhibit longer flame lengths, have greater potential to convert to crown fires, be more difficult to contain, pose greater risks to firefighters, kill more vegetation, and damage soils more severely than areas with lower fuel loads. Fuel loading contributes to high resistance to control. High snag and log levels impede fire line construction, increase safety hazards, increase spotting potential, and increase fire severity. Containment lines must be constructed far from the fireline where it is safe and practical to do so, ultimately increasing fire size. Increased soil heating from burning logs kills soil microorganisms and reduces soil produc tivity.

Alternatives 2 and 5 reduce long-term fuel loading; however, in areas where four to six snags per acre are retained it would be unsafe for effective firefighting and fuel levels by year 25 would be high enough to seriously impede fire suppression and contribute to severe soil heating. Portions of the threat zone and the ridgetop fuelbreaks would be the only locations available for safe fire line construction. Direct attack on small fires in spotted owl protected activity centers (PACs), general forest, and old forest emphasis areas may be precluded due to fire fighter safety. Fires would therefore become larger and more intense as indirect methods are employed.

Alternative 3 increases fuel loading substantially in areas where there are many fire damaged and dying trees. As these trees die over time, they will be retained as snags in areas that are not economical to reenter, such as steep ground (approximately 75 percent of the treated area). These higher snag and log levels would have similar effects as described under Alternative 1.

2. The need to reduce sedimentation to streams and erosion from roads and hillslopes for the purpose of protecting soil productivity and water quality for beneficial uses.

The fire has resulted in a significant increase in sediment delivery to some streams. This will continue for the next several years under all alternatives, which in turn may affect downstream beneficial uses of water. Alternatives 2, 3, and 4 may add slightly to the risk of impacts to downstream beneficial uses of water; alterna tive 5 would result in a negligible increase in such a risk. Alternatives 2, 3, 4, and 5 will improve the existing road system, reducing chronic sources of sedimentation. Snag retention will provide recruitment of in-stream large woody debris. Hydrologically sensitive areas and steep slopes will be protected through equipment ex clusion zones and a number of other protection mea sures. The extent that each alternative provides for short-term effective ground cover was also used as a measure of meeting this need. Alternative 1 would result in higher soil erosion than the other alternatives because the amount of effective ground cover provided from natural needle and limb fall is unlikely to be sufficient to reduce erosion rates in the short term in high severity burn areas lacking vegetation. Alter natives 2, 3, 4, and 5 are similar in reducing erosion as mechanical treatment of dead trees will provide limbs from trees for ground cover. Alternative 5 best meets this need because ground cover is increased while ground disturbance is minimized by the use of helicopter logging.

Over the mid to long term, Alternative 4 will provide for greater watershed protection than other alternatives because Alternative 4 reduces the risk of severe impacts to soils and beneficial uses of water from another wildfire. Alternative 4 reduces fuel loading and im proves resistance to control, thereby increasing the po tential to suppress fires during initial attack when they are small, with less impact to soil cover; and less po tential for destruction of soil organisms and organic matter from lethal temperatures (FEIS pgs. 90-91, 102, 106, 109-110, 124, 131, 133)

3. The urgent need to recover the volume and value of timber killed or severely injured by the fire for the purpose of generating funds to offset the cost of restoration activities.

This element highlights the need to make a decision about removing dead trees within the Power Fire in a timely manner. This factor also highlights the relation ship between the economic value of the dead trees and meeting the other ecological and social needs for action. The economic value, in terms of wood quality, decreases rapidly in dead trees with the potential for substantial loss within 3 years (FEIS pgs 255-256). Trees that were immediately killed by the fire will have 10 months worth of decay by the time this project is implemented and some of the smaller diameter dead trees may have already lost all economic value. By the second year following the fire, about 65% of the wood volume will be lost to decay (FEIS pg. 255; Emergency Situation Analysis for the Power Fire Restoration Project). De terioration of the fire-killed and damaged trees has a number of impacts. Lumber quality is reduced, mer chantable volume of wood is reduced, and most import antly, value is reduced. This reduction in value impacts the revenue the Federal Government receives from stumpage and it also impacts the ability to cost effec tively remove the dead trees at all. Ultimately, after trees have significantly deteriorated, the cost of removal far exceeds the value of the trees, and the Government is faced with the dilemma of an increasing fuel load and no funds available to mitigate the impact. In California, there are numerous examples of recent fires where delay caused no-bid timber sales and the timber was not harvested because of deterioration rendering the trees worthless in the marketplace and unsafe to harvest. On the Gap Fire of 2001 on the Tahoe National Forest, one year after the fire the value of the trees dropped below the cost of removing them with helicopters. The conse quence was an estimated $400,000 loss of timber value that impacted the ability to treat fuels (refer to the Emergency Situation Analysis prepared for the Power Fire Restoration Project). On the Star Fire on the Tahoe National Forest, delay caused an estimated $2,600,000 loss in timber value. This amount of money could have paid to hand pile at least 4,300 acres of fuel on steep terrain (refer to the Emergency Situation Analysis prepared for the Power Fire Restoration Project).

Delay will diminish the opportunities to utilize dead trees to benefit human needs, lessen the ability to leverage the timber value to treat fuels, repair and im prove roads, and fund other restoration treatments, and reduce our ability to meet the long term goals and achieve the purpose and need for the project. For these reasons, I requested, and the Chief, Forest Service ap proved, an emergency situation as provided by 36 CFR 215.10

Alternative 4 generates the greatest revenue: $19,056,425 (expected stumpage payments) which is sufficient to cover the costs of road improvements, fuel treatments, and future reforestation if any. Alternative 1 does not meet this need because no revenue would be generated and future restoration costs such as fuel treatment and potential reforestation would be entirely dependent upon appropriated or other funding sources, which is unlikely to be available. Alternative 2 gen erates an estimated $13,916,229; Alternative 3 generates an estimated $11,924,641; and Alternative 5 cost is esti mated to exceed revenue by $1,728,823. All action alternatives, except Alternative 5, generate sufficient funds to pay for fuel treatments and road improvements.

4. The need to reduce safety hazards to the public and forest workers from failing trees.

All action alternatives reduce the risk of falling snags to some extent, but no alternative completely eliminates the risk. Alternative 4 best reduces the risk by pre scribing the most acres with lower snag levels. Clump ing snags into retention patches further minimizes risks to the public and forest workers within the project area. Alternative 1 has the highest risk to public and forest worker safety since all dead trees will remain. In the short-term, the risk to the public entering the Power Fire area is great as smaller trees will tend to be sus ceptible to falling soon and the tops of larger trees will begin to fall, especially in the winter. The risk of falling trees will eventually preclude safe fire fighting and future restoration work due to the potential to injure or kill a person working in the fire area.

I also considered the public comments and the signi ficant issues in making my decision. Alternative 4 is responsive to the public comments and addresses most of the significant issues (FEIS Chapter 1, pages 19-20).

Issues:

1. Whether use of the Pacific Southwest Region Forest Health Protection Staff's mortality guidelines presents an unacceptable risk of cutting trees that would have survived their fire injuries: There is a trade off between minimizing the risk of cutting a live tree and the need to repeatedly log the same ground to remove trees as they die. There is a high likelihood that repeat entries would be made in ground based logging units, and a lower probability in skyline and helicopter logged areas. The latter systems are more costly and depen dent on higher volumes per acre for economic feasibility. Repeat entries increase the risk of soil disturbance, erosion, and compaction. Lack of repeat entries in creases the number of snags retained and correspond ingly increases the long-term fuel accumulation. In the Star Fire, where mortality guidelines were not used, 22 percent additional cubic volume of timber was removed in the second and third re-entry on tractor units. No reentry occurred on skyline and helicopter logged areas. Alternative 3 does not use mortality guidelines, except for hazard trees, and consequently there is a higher risk of significantly increasing the number of snags retained and associated fuel accumulation. The other alter natives, including Alternative 4, mitigate the risk of cut ting trees that would otherwise survive by utilizing the latest and most comprehensive scientific assessment of tree mortality following wildfire in California (Hood et al. 2005). The latest science assessing fire related mortality provides models that allow managers to select the desired level of predicted mortality based on land management objectives. Tree marking guidelines are derived from the probabilities, as described in the FEIS on pages 75 and 76 and in Appendix F. For Alternatives 2, 4, and 5, I selected a more conservative probability of mortality for areas where ground based logging was proposed, and a slightly less conservative probability of mortality where aerial logging systems were proposed as displayed in the table below.

Table 1. Probability of Tree Mortality as applied to Alternatives 2, 4, and 5

 

PP/ JP/ SP

White

Fir

Incense Cedar

Red

Fir

Douglas

Fir*

Tractor Logging System

Correctly Predicted Mortality (%)

96

95

100

100

-

Correctly Predicted Surival (%)

51

63

88

-

-

Helicopter and Skyline Logging System

Correctly Predicted Mortality (%)

90

87

85

100

-

Correctly Predicted Survival (%)

65

74

89

-

-

* Marking guidelines for Douglas fir are based on Ryan and Reinhardt (1988).

As displayed in the table, the guidelines for tractor logged areas reduce the chance of trees being cut which might otherwise have survived, while accepting a rela tive high probability of many additional trees which do not meet the guidelines subsequently dying. Marking guidelines for helicopter and skyline logging result in a slightly higher potential of trees being cut which might otherwise have survived, and a lower probability of additional trees, which do not meet the guidelines, subsequently dying.

The rationale for a more conservative mortality marking guide in tractor logged areas stems from ex perience in previous fire salvage sales as described above. Since tractor logged areas are typically re- logged when significant numbers of additional trees die, fuel loading from additional dead trees can be mitigated. In order to mitigate the fuel accumulation from delayed mortality on steep slopes, more of the dying trees need to be removed initially because high yarding costs, setup time, and limited equipment availability make it unlikely that helicopter or skyline units would be re-logged to capture mortality occurring after the initial logging.

2. Whether ground based and "cable" logging would result in unacceptable impacts on soil and downstream beneficial uses of water: Alternative 4 will improve the existing road system, reducing chronic sources of sedimentation. Snag retention will provide recruitment of in-stream large woody debris. Hydrologically sensi tive areas and steep slopes will be protected through use of aerial logging methods, equipment exclusion zones, application of Best Management Practices and a number of other protection measures described in the FEIS on pages 41-44, which I believe will be effective based on monitoring on other projects and my background and experience as a hydrologist. While Alternative 5 was designed to address this issue, and would minimize ground disturbance, it is very expensive to implement and I do not believe the perceived benefits justify the high cost.

3. Whether the leaving of four to six of the largest snags per acre and all snags in some riparian areas would result in excessive fuel loading, contribute to an unacceptable safety hazard for recreationists and retains too much economic value as snags: I agree that the leaving of four to six snags over the landscape would result in excessive fuel loading as I discussed above under Need 1. Alternative 4 provides high levels of snags for wildlife but arranges them in patches. This reduces fuel loading outside the patches, and improves safety for workers and the public.

4. Whether the snag retention proposed provides enough suitable habitat for black-backed woodpeckers, hairy woodpeckers, and other birds that utilize high levels of snags in burned forests: Alternative 1 retains all snags, thereby maximizing habitat for cavity-nesting species, but fails to meet the needs identified for the project as articulated above and in the FEIS. Cavity- nesting birds are management indicator species for the ENF, but are not threatened, endangered, or sensitive. There are many cavity-nesting bird species that find habitat in the Power Fire, and under the ENF Land Management Plan I am charged with providing medium capability habitat for a suite of species. Alternative 4 provides high levels of snags for cavity-nesting birds potentially using the fire area, but arranges them in patches as well as distributed in PACs and RCAs. The patches are located in areas of large trees (WHR size class 4 and 5), but contain both large and small trees. Alternative 4 meets the needs of species that are associated with burned forest habitat, such as hairy and black-backed woodpeckers, as well as species that are snag habitat generalists and species associated with forest environments. In addition, high quality snag habitat is provided in the Mokelumne Wilderness Area, which is untouched by this project. In total, all dead trees on more than half of the fire area are retained under Alternative 4.

Alternatives 2, 3, and 5 provide for large numbers of large snags as well. These alternatives rely on retaining an overall average of four snags per acre in the largest size classes, so the distribution of snags would likely be somewhat dispersed since it is based on the distribution of large trees. Species like the hairy and black-backed woodpeckers that are associated with burned forests would benefit less from these alternatives than Alterna tive 4.

Public Involvement

The Notice of Intent to prepare an Environmental Impact Statement was published in the Federal Register December 22, 2004. A brief description of the location and type of project was included in the ENF Schedule of Proposed Actions (SOPA) in January 2005. Approximately 134 letters were mailed out to adjacent property owners, potentially affected businesses, tribes, federal, state, and local agencies, and special interest groups. The letter contained the detailed proposed action, map, and methods for participation. The mailing list is included in the project record. An article was published in the Mountain Democrat describing the proposed action. As a result of scoping, 50 comment letters were received from 93 individuals and groups. The Notice of Availability of the draft EIS was published in the Federal Register on March 25, 2005 and the 45-day comment period commenced. The DEIS and/or summary was sent to 90 individuals, groups and government agencies. On several occasions, field trips were taken to the project area for interested individuals and groups. A public meeting was held in April, 2005 to discuss the Draft EIS. Presentations were made to a number of organizations and groups, including East Bay Municipal Utility District, Amador County Board of Supervisors, Amador Watershed Council, Sierra Nevada Native American Council, Amador County Resource Conservation District, California Forestry Association, Amador County Fire Safe Council, Amardor-Eldorado Forest Forum, and the Mokelumne River Association. The 45-day comment period ended on May 9, 2005. Twenty comment letters were received (FEIS Appendix L). Comments were assessed and considered. Re sponses were given and the following actions were taken: alternatives were modified; an additional alterna tive was developed, considered, and eliminated from detailed study; factual corrections were made; and the environmental consequences were supplemented (FEIS Appendix M).

After reviewing the public scoping comments, I approved the following significant issues to generate alternatives:

o Whether use of the Pacific Southwest Region Forest Health Protection Staff's mortality guide lines presents an unacceptable risk of cutting trees that would have survived their fire injuries: During scoping, concerns were expressed about the use of these guidelines and the effects on live trees. We believe these guidelines to be the best available science applicable to the project area; they were developed based on scientific literature on this subject, extensive post-fire monitoring in the Sierra Nevada, and professional judgment. However, predictions of mortality are not an exact science. The probabilities are high that trees meeting the mortality guidelines are dead and dying, however, there is a small probability that trees meeting the guidelines would survive as well as trees not meeting the guidelines would die. The concern by some respondents is that the probabilities are not accurate enough and any risk of cutting a tree that would otherwise survive is too much. These respondents point out that retention of all live trees, particularly large trees, is important for seed sources, shading, and wildlife habitat, and requested an alternative that does not utilize mortality guidelines.

o Whether ground based and "cable" logging would result in unacceptable impacts on soil and down stream beneficial uses of water: In some scoping comments, disagreement related to effects of the Proposed Action on soil compaction, erosion, and sedimentation were expressed. Several studies were cited supporting these concerns. In parti cular, some respondents were concerned that ground based and "cable" logging in moderate to high severity burn areas would exacerbate erosion and sedimentation. Technically speaking, cable logging refers to dragging logs with the use of cables suspended from a tower sitting on a road, and may not necessarily imply full or partial suspension of the logs above the ground. The pro posed action utilizes "skyline" logging techniques which require full or partial suspension of the logs as they are brought up the hill. For pur poses of this issue however, we assumed that "cable" logging and "skyline" logging were sy nonymous. These respondents requested an alternative that does not utilize ground-based or cable logging techniques.

o Whether the leaving of four to six of the largest snags per acre and all snags in some riparian areas would result in excessive fuel loading, contribute to an unacceptable safety hazard for recreationists and retains too much economic value as snags: There was also concern that concentrations of woody material left adjacent to stream courses could predispose these areas to catastrophic re burn and contribute to debris jams that block cul verts and divert watercourses. The issue was addressed by developing an alternative to the proposed action.

o Whether the snag retention proposed provides enough suitable habitat for black-backed wood peckers, hairy woodpeckers, and other birds that utilize high levels of snags in burned forests: Some respondents assert that the proposed action would reduce the viability of these species, since studies indicate these birds preferentially use unsalvaged burned forests for foraging. This issue was addressed by developing an alternative to the proposed action.

Alternatives Considered

In addition to the selected alternative, I considered 4 other alternatives, which are discussed briefly below and in more detail on pages 21 to 54 of the FEIS. A detailed comparison of these alternatives can be found in Chaper 2, pages 55 through 62 of the FEIS.

Alternative 1 is the No Action alternative. Under this alternative no dead tree removal or treatment would occur. Road maintenance and repair and projects associated with the burn area emergency response (BAER) would continue.

Alternative 2 is the original Proposed Action. Under this alternative all dead and dying trees in the Mo kelumne Wilderness, PAC "core" areas, areas within the distance limitations prescribed for perennial streams, oak stands, and large contiguous low intensity burn areas would be retained. All dead and dying trees would be retained in patches of suitable spotted owl or go shawk habitat outside of PACs unless significant additional mortality develops rendering the habitat unsuitable. All dead and dying trees in low intensity burn areas with mortality of less than 25 percent would be retained, unless mortality increases sufficiently to meet the snag retention guidelines and is economically feasible to treat. All existing down logs and cull logs greater than or equal to 20 inches small end diameter and over 10 feet long generated from logging would be retained, and standing snags would be retained in variable numbers and sizes in riparian conservation areas, threat zones, defense zones, general forest, old forest emphasis, and unsuitable burned habitat in PACs. Roadside hazard trees would be removed. Logging would occur using helicopter, skyline, and ground-based machinery. Protection of RCAs, sensitive plants, cultural resources, and wildlife would occur. Treatment of fuels and manipulation of ground cover would occur. Roads would be reconstructed, surfaced, and maintained to improve watershed condition.

Alternative 3 was developed in response to the issue concerning use of mortality guidelines and is similar to Alternative 2 except mortality guidelines that predict which trees will die of their injuries would not be used to mark trees for harvest. Protection measures, treatment of fuels, and road improvements are the same as for Alternative 2.

Alternative 4 was developed in response to the third and fourth significant issue dealing with snag retention for wildlife and is the preferred alternative. Alternative 4 modifies the proposed action to retain dead trees in 732 acres of patches over the landscape. Outside of PACs and RCAs, this alternative was designed to leave snags for wildlife in varying sized patches instead of as individual trees spread over the landscape, avoiding ridges, roads, powerline, flume, and private land. As first priority, patches are located in WHR size class 5 stands where mortality exceeds 50 percent. Second priority is WHR size class 4 where mortality exceeds 50 percent. Snag retention in perennial and seasonal RCAs and PACs and protection measures, treatment of fuels, and road improvements are the same as for Alternative 2.

Alternative 5 was developed in response to the second issue concerning impacts of ground based logging systems and modifies the proposed action to replace skyline and ground based logging with helicopters except that ground based logging would be permitted on slopes generally less than 35 percent to remove roadside hazard trees. In all other respects, this alternative is the same as Alternative 2.

All action alternatives employ design criteria to prevent erosion, improve soil infiltration, protect sensi tive plants, protect sensitive wildlife, protect archeo logical resources, protect stream courses, prevent the introduction of noxious weeds, reduce the potential for adverse cumulative watershed effects, protect beneficial uses of water, protect public safety, and protect trails, as discussed in the FEIS on pages 41 to 44.

Monitoring of best management practice imple mentation and effectiveness, fish and herptofauna occu pancy, spotted owl and goshawk occupancy, stream tur bidity, and noxious weed populations will be conducted as part of all action alternatives (FEIS pg. 44).

Alternative Considered but Eliminated from Detailed Study

Federal agencies are required by NEPA to rigor ously explore and objectively evaluate all reasonable alternatives and to briefly discuss the reasons for eliminating any alternatives that were not developed in detail (40 CFR 1502.14). Public comments received in response to the Proposed Action provided suggestions for alternative methods for achieving the purpose and need.

An alternative that defers action until 15 years post fire was proposed by some members of the public. Under such an alternative the surface fuels would be piled and burned and felling of small snags for ground cover could occur. This alternative was eliminated from detailed study because it does not accomplish the purpose and need for the project. Specifically, this alternative does not generate any revenue that could be used to offset the costs of the fuel treatment or any other restoration activity. This alternative would retain nearly all dead material on site, increasing safety hazards and fuel loads over time to unacceptable levels. By 15 years post fire, dead trees would be significantly deteriorated to a point that it would be extremely hazardous to perform any of the treatments this alter native would require (FEIS pg. 45). Even if the work could be done safely, piling and burning of surface fuels is expensive on large areas of steep slopes, exceeding four million dollars as displayed on page 46 of the FEIS. Additionally, this alternative would not generate any effective ground cover in the short term in high intensity burn areas.

An alternative that retains all dead trees greater than 15 inches in diameter was proposed. This alterna tive would fell and lop smaller trees for ground cover, and treat small sized fuels through piling and burning. This alternative would preclude skyline logging due to the prevalence of trees greater than 15 inches that would interfere with skyline corridors. It is assumed that all areas proposed for skyline logging would be helicopter logged instead. This alternative responds to the issue of whether there is a need to remove large dead trees (over 15") to reduce future fire severity and intensity. Some respondents assert that "by the time most of the large snags have fallen and substantially decayed, the small snags, and branches and tops from large snags, will have long since fallen and decayed into soil." These respondents requested an alternative that retains all dead trees over 15-inch diameter. Based on fuels modeling displayed in the FEIS on pages 47 and 48, and the effects on fire behavior and resistance to control, this alternative fails to meet the purpose and need for the project. This alternative fails to reduce future fire intensity and would lead to significant tree mortality in the event of a fire, hindering or reversing the growth of future forest. In addition, numerous snags would continue to inhibit or preclude safe fire fighting practices within the project area; the alterna tive is not practical or capable of being implemented because there is little economic value in trees less than 15 inches to generate sufficient funds for fuel treatment and other restoration treatments; and future fires would likely have significant and long-lasting damage to soil productivity.

An alternative was proposed in response to com ments about the DEIS. The alternative would retain all dead trees greater than 24 inches diameter, triple the number of snag patches identified for Alternative 4, and remove only dead trees with no green needles. The purpose of this alternative was to increase the habitat available for black-backed, and other woodpeckers, and provide additional snags for legacy structure for future forest. Following discussions with proponents, the proposed alternative was amended to retain four of the largest snags per acre (as in Alternative 2), retain one half of the high mortality (>75% mortality) areas outside the Mokelumne Wilderness in CWHR size class 4 and 5; or 40 percent of the high-mortality areas in size class 4 and 5 plus two-thirds of the available size class 3, in snag patches greater than 50 acres in size (approxi mately 2,620 acres), and remove only dead trees with no green needles. This alternative was eliminated from detailed study because it does not accomplish the pur pose and need for the project to reduce long-term fuel loading, reduce safety hazards to forest workers, and increase ground cover in the short term in high intensity burn area.

Environmentally Preferable Alternative

The NEPA implementing regulations (Section 1505.2) require that the alternative(s) that best pro motes the national environmental policy as expressed in NEPA, Section 101, be identified in the Record of Decision as the "environmentally preferable Alter native" or alternatives. This is ordinarily "the alterna tive that causes the least damage to the biological and physical environment and best protects, preserves, and enhances historical, cultural, and natural resources" (FSH 1909.15, 05). For the Power Fire Restoration project, I believe Alternative 5 is the environmentally preferred alternative in the short term. This is because alternative 5 results in the least ground disturbance while improving roads and providing effective ground cover to help reduce soil erosion and sedimentation to streams. Alternative 4 is the environmentally pre ferable alternative in the mid to long term because it reduces fuel loading and improves resistance to control in the event of another wildlife. Alternative 4 will improve the chances of keeping future wildfires small and reduce the severity of future fires. For this project, all of the action alternatives provide protection to the environment by their design, management require ments, and built-in mitigations. All action alternatives comply with NEPA's purpose and spirit.

Findings Required by Other Laws and Regulations

All management practices and activities of Alterna tive 4 are consistent with the management direction of the ENF Land and Resource Management Plan (LRMP) as amended.

Clean Air Act

The Clean Air Act makes it the primary responsi bility of States and local governments to prevent air pollution and control air pollution at its source. States must have a plan that provides for implementation, maintenance, and enforcement of the primary ambient air quality standard. The State of California has a plan. This project meets the Clean Air Act.

Clean Water Act

Federal agencies are required by the Clean Water Act to cooperate with State agencies in preventing, reducing, and eliminating pollution in concert with pro grams for managing water resources. This project meets this requirement through the incorporation of best management practices and other design features (FEIS pgs. 41-44, Appendix B). This project meets the Clean Water Act.

National Environmental Policy Act (NEPA)

The NEPA requires that Federal agencies complete detailed statements on proposed actions that signifi cantly affect the quality of the human environment. The Act's requirement to prepare an environment impact statement is designed to provide decision makers with a detailed account of the likely environmental effects of a proposed action prior to adoption and to inform the public of, and allow it to comment on, such effects. The FEIS does a comprehensive job of analyzing the al ternatives and displaying the alternatives and displaying the environmental effects. The procedural requirements of the NEPA have been followed.

National Forest Management Act (NFMA)

Projects occurring on NFS lands must meet mini mum specific management requirements under 36 CFR 219.27 (1982). This project and the FEIS address each as follows:

o The management prescriptions discussed in the FEIS meet all of the resource protection require ments of the CFR.

o The requirement for vegetative manipulation is not applicable to this project because this project involves salvage of fire-killed trees and does not involve the manipulation of tree cover.

o The project was reviewed by a certified silvicul turist and found to be in compliance with all aspects of the silvicultural practices requirement.

o The even-aged management requirement does not apply.

o This project is consistent with the requirements for riparian areas.

o This project meets the requirements for soil and water.

o Biological Evaluations considered effects to Forest Service sensitive species. The Biological Evaluations determined that the project was not likely to result in a loss of viability for any sensi tive species.

Endangered Species Act (ESA)

Section 7(a)(2) of the Endangered Species Act requires that Federal agencies consult with the United States Fish and Wildlife Service and National Marine Fisheries Service, as appropriate, to ensure that their actions do not jeopardize the continued existence of species listed as threatened or endangered under ESA, or destroy or adversely modify their critical habitat.

A biological assessment was prepared for federally proposed, threatened or endangered wildlife species and their critical habitat. Implementation of the project would have no effect on any threatened or endangered species including bald eagle, red-legged frog, valley elderberry long-horn beetle, central valley spring run Chinook salmon, central valley steelhead, delta smelt, or Lahontan cutthroat trout. Therefore, no consultation was necessary with the United States Fish and Wildlife Service. No botanical species are federally proposed, threatened or endangered.

National Historic Preservation Act

The ENF has determined that this project will have an adverse effect on The Mokelumne Archeological District, determined to be eligible for inclusion in the National Register of Historic Places (historic district) according to the criterion in 36 CFR 60.4(d), and pursuant to 36 CFR § 800.5(d)(2). In consultation with the State Historic Preservation Officer (SHPO), the ENF has resolved any potential adverse effects by executing a Memorandum of Agreement (MOA) pur suant to 36 CFR 800.6. This agreement was signed July 29, 2005.

The procedures and stipulations of this MOA include professional standards, Native American Consultation, and the identification and treatment of cultural resource values. As stipulated in the MOA, the cultural resource values will be protected throughout the project using Standard Resource Protection Measures within the "Programmatic Agreement Among the U.S.D.A. Forest Service, Pacific Southwest Region, California State Historic Preservation Officer, and Advisory Council on Historic Preservation Regarding the Identification, Evaluation and Treatment of Historic Properties Managed by the National Forests of The Sierra Nevada, California" (1996).

The primary protection measure will include avoid ance of ground disturbing activities, project modifi cation, padding and rocking of any road reconstruction, and the removal of trees posing a hazard to cultural re source values. Removal of these trees will be accom plished using methods appropriate to the site consti tuents, ground, and harvest conditions. These methods may include hand bucking and carrying; rubber tired loader using full suspension or skyline; crane/self loader; and helicopter.

Implementation Date

The Chief of the Forest Service has determined that an emergency situation exists for 7006 acres of the project area as provided for in 36 CFR 215.10. Imple mentation may begin immediately for that portion of the decision determined to be an emergency.

Administrative Review or Appeal Opportunities

This decision is subject to administrative review (appeal) pursuant to 36 CFR Part 215. The appeal must be filed (regular mail, fax, email, hand-delivery, or express delivery) with the Appeal Deciding Officer at: Bernard Weingardt, Regional Forester, USDA Forest Service, Regional Office R5, 1323 Club Drive Vallejo, CA 94592, fax: (707) 562-9229.

The office business hours for those submitting hand- delivered appeals are: 8:00 am to 4:30 pm Monday through Friday, excluding holidays. Electronic com ments must be submitted in a format such as an email message, plain text (.txt), rich text format (.rtf), or Word (.doc) to appeals-pacificsouthwest-regional- office@fs.fed.us. In cases where no identifiable name is attached to an electronic message, a verification of identity will be required. A scanned signature is one way to provide verification.

Appeals, including attachments, must be filed within 45 days from the publication date of this notice in the Mountain Democrat, the newspaper of record. Attach ments received after the 45 day appeal period will not be considered. The publication date in the Mountain Demo crat, newspaper of record, is the exclusive means for calculating the time to file an appeal. Those wishing to appeal this decision should not reply upon dates or timeframe information provided by any other source. Individuals or organizations who submitted substantive comments during the comment period specified at 215.6 may appeal this decision. The notice of appeal must meet the appeal content requirements at 36 CFR 215.14.

Contact Person

For additional information concerning this decision or the Forest Service appeal process, contact Patricia Ferrell, Interdisciplinary Team Leader, Eldorado National Forest Supervisor's Office, 100 Forni Road, Camino, CA 95709, (530) 642-5146.

 

/s/ JOHN D. BERRY
JOHN D. BERRY

Forest Supervisor

08/01/2005

Date

 

APPENDIX F

RECORD OF DECISION
FREDS FIRE RESTORATION
Eldorado National Forest
100 Forni Road
Placerville, CA 95667
El Dorado County, California

Introduction

The Freds Fire was reported in the late afternoon of October 13, 2004, on the north side of Highway 50 approximately 1_ miles east of the communities of Silver Fork and Kyburz, in EI Dorado County. After ignition, the fire quickly spread across extremely steep slopes, burning through timber and heavy fuels. The fire burned rapidly in a westerly direction, parallel to Highway 50, driven by strong east winds. Highway 50 was closed immediately; the communities of Silver Fork and Kyburz were evacuated, and suppression efforts focused on protecting the towns and their infrastruc ture. No structures were lost within the communities, but the water systems for Kyburz and Silver Fork were severely damaged. Several outbuildings and vehicles on inholdings within the Forest boundary were destroyed. Hot, dry, and windy weather conditions allowed the fire to burn rapidly for several days, until a major winter storm on October 17 assisted firefighters to contain the blaze. The highway closure and evacuations continued for several more days due to the threat of flash flooding, slope failure, and landslides.

The Freds fire burned approximately 7,700 acres, including about 4,600 acres of the Eldorado National Forest (ENF), about 2,830 acres owned by Sierra Pacific Industries (SPI), and about 280 acres owned by five private landowners within the National Forest. The fire consumed the last section of land with a southern exposure along Highway 50 in the South Fork American River Canyon, which had not burned in recent memory. It burned into the Cleveland Fire on the west and into the Wrights Fire on the east.

The fire burned with varying intensity as displayed in the photo below.

[Picture omitted]

Approximately 12 percent (about 525 acres) burned with low intensity, as measured using aerial photos and field sampling. Many areas of the fire (approximately 61%, 2606 acres) burned at high intensity, killing 66 to 100 percent of the trees and burning the duff and litter that protect the soil. Another 11 percent (about 419 acres) burned with moderate intensity, killing 33 to 66 percent of the trees. In the high and moderate intensity areas the fire resulted in high rates of soil erosion, sedimentation to streams, destruction of wildlife habitat for sensitive species. An additional 694 acres of young plantations burned at high intensity (about 16 %). The fire killed tens of thousands of trees that, if left un treated, will contribute to extremely high fuel loading over time. As these dead trees fall and fuel accumulates, future fires will be even more severe. Without treatment to begin to restore the fire area, significant additional impacts to soil, water quality, cultural resources, and wildlife habitat are likely over the short and long term.

The Sierra Nevada Forest Plan Amendment Record of Decision (SNFP ROD) provides direction for eco system restoration following catastrophic events. These restoration activities are included in all land allocations and call for managing disturbed areas to achieve long- term fuels profiles, to restore habitat, and recover the value of some dead and dying trees (SNFP ROD, pg. 6). SNFP land allocations within the Freds Fire boundary are threat zone, defense zone, and general forest (See Map 1-1). There are three protected activity centers (PACs) for spotted owls; spotted owl home range core areas (HRCAs); and riparian conservation areas (RCAs) adjacent to perennial, seasonal, and ephemeral streams. Highway 50 is a state designated Scenic Highway, and the South Fork American River was found to be eligible as a Wild and Scenic Recreation River in 1990. A suit ability study has not been completed for the river and it has not been proposed for congressional designation. In addition, the Pony Express Trail, a National Recreation and Historic Trail, bisects the project and is a linear feature that parallels Highway 50.

The goal of this project is to move the area toward desired future conditions as defined by the Sierra Nevada Forest Plan Amendment (SNFP ROD, pgs. 36 to 48). These desired conditions by land allocation are described in detail in Table 1-1 on page 5 of the Final Environmental Impact Statement (FEIS) for the Freds Fire Restoration Project. Desired conditions for threat zones (ROD pg. 46) are geared to reduce wildland fire behavior under high fire weather conditions (hot, dry summer days). This includes flame lengths of less than 4 feet at the head of a fire; reductions in rate of spread at the head of the fire; reduction of hazards to fire fighters by removing snags from locations likely to be used for fire suppression; and a doubling of fire line construction rates. Desired conditions for defense zones (ROD pg. 45) are fairly open stands, dominated by larger, fire-tolerant trees; surface and ladder fuel conditions such that crown fire ignition is unlikely; and discontinuity of crown fuels that result in very low probability of sustained crown fire. Desired conditions for general forest areas (ROD pg. 48) include high levels of horizontal and vertical diversity; trees of varied sizes, ages, and species composition; and enough dead trees, standing and fallen to meet habitat needs of old forest associated species, while allowing for successful establishment of early-seral stage vegetation. Desired conditions for PACs and HRCAs (ROD pg. 45 to 46) include at least two tree canopy layers; dominant and co- dominant trees that average at least 24 inches diameter at breast height (dbh); 50 to 70 percent canopy cover; some very large snags; and higher than average levels of snags and down woody material. In RCAs (ROD pg. 42), the desired condition is to meet the water quality goals of the Clean Water Act and Safe Drinking Water Act, with streams that are fishable, swimmable, and suitable for drinking after normal treatment.

Purpose and Need for Action

1. There is a need to reduce long-term fuel loading for the purpose of reducing future fire severity and resis tance to control: The Freds Fire burned approxi mately 7,700 acres of mixed conifer forest. The fire damaged forest resources such as soil, riparian areas, and wildlife habitat, and killed thousands of trees. In areas where the fire intensity was moderate to high, surface fuel loading is now very low. Over time as dead trees fall surface fuels will increase significantly, affecting future fire behavior and suppression capabilities. Additionally, it is expected that shrub species will resprout rapidly in many areas of the fire. Excessive large woody debris leads to difficulty in suppressing wildfires. Fire line construction is significantly slowed where fire lines intersect numerous large logs. This is referred to as "resistance to control" and can lead to larger fires since fire lines have to be relocated to areas of less woody debris or where tractor fire lines can be built (on flatter ridgetops for example). Large woody debris also increases fire severity; further impacting impaired watersheds, soils, and archeological sites. Excessive small woody debris, from small trees and limbs of larger trees, increase a fire's rate of spread and fire line intensity, reducing the ability of fire fighters to suppress the fire and increasing the ultimate fire size. Dead trees that aren't removed, will contribute to extremely high fuel loading within five to ten years. Many of the snags that were left standing following the Cleveland fire in 1992 have fallen over in the intervening 12 years.

[Picture omitted]

The threat of another large wildfire occurring along Highway 50 in the South Fork American River corridor within five to ten years is high. The Highway 50 corridor has had four large wildfires within the past 31 years, Pilliken Fire (1973), Wrights Fire (1981), Cleveland Fire (1992) and Freds Fire (2004) (refer to Freds FEIS, Fire History Map 3-1, Chapter 3, Fire and Fuels). In addition, there have been 40 fires (under 10 acres) since 1908 and 399 fires less than 10 acres since 1970. The potential for a wildfire start is high due to proximity to the large number of travelers along Highway 50, a Pacific Gas and Electric Company (PG&E) distribution line that runs through the canyon, residential develop ment, recreational use, and lightning. Predicted fire behavior modeling of timber stands and fuel types representative of current conditions, indicates that high intensity fire with rapid rates of spread as likely under moderate weather conditions . Without additional treatment to begin to reduce fuel loads, wildfires are likely over the short and long term, once again threatening the residents of Silver Fork and Kyburz, and the other private landowners in this area.

2. There is a need to improve road drainage and establish effective ground cover in severely burned areas for the purpose of reducing erosion and sedimentation to streams in the short term and for contributing to long term soil productivity: Large portions of the project area burned at high and moderate intensity, killing 75 to 100 percent of the trees and burning the duff and litter that protects the soil. These areas are devoid of effective ground cover, leading to soil erosion and potential loss of productivity. Runoff and erosion rates are expected to be high, and many areas have steep slopes, increasing the hazard for soil erosion and transport of sediment to streams. The Highway 50 corridor has experienced several large slope landslides and debris flows/torrents within the large fire areas listed above. Over 600 landslides have been mapped, using air-photo interpretation, along the South Fork of the American River/Highway 50 corridor between Pacific House and Echo Summit. Table 1-2, on page 15 of the FEIS,

[Picture omitted]

displays the slope failures and associated fires as well as all landslides identified by the California Geologic Survey that have the potential of reaching Highway 50 within and near the Freds Fire area.

The landslide inventory/history displayed in Table 1- 2 shows a trend for slope movement to occur during the winter months and well into the spring months. This is, of course, the time of year when ground water recharge rates are at the highest. This table also includes non-fire related landslides that have the possibility of moving across Highway 50. In some cases (for example Landslide # 8, # 9, # 10, # 15 and #16) landslides have dammed or have the probability of damming the South Fork American River temporarily. This shows that slope movement can and does occur, even without fires, indicating that the rate of ground water recharge is the driving process in slope movement.

The influence of proposed salvage logging operations to the overall slope stability within the project area is directly related to the link between surface and ground waters (i.e., ground water recharge). If surface water is collected and then routed to potential or existing unstable slopes where it recharges the ground water, there is the possibility of reactivating some of the landslides and/or initiating slope movement by raising the ground water levels. The collecting and routing of surface water may occur along skid and tractor trails as well as along truck roads.

Native surfaced (dirt) roads have the potential to channel runoff and increase sedimentation to streams. Sediment delivered to streams impairs water quality and has the potential to impact downstream beneficial uses of water, including municipal water supplies. The South Fork of the American River provides drinking water for customers of the EI Dorado Irrigation District (EID).

3. There is a need to act rapidly to remove some of the dead trees while they retain economic value that can offset the cost of other restoration work: Dead trees deteriorate rapidly relative to wood quality, volume, and value. By the first year following the fire, most trees have significant worm holes and weather checking, deteriorating about one third of the cubic volume on trees between 11 and 24 inches diameter (Bodenhausen 2002). By the second year, 47 to 74 percent of the volume of trees less than 24 inches diameter is lost (Lowell et al. 1992). Larger trees deteriorate more slowly, but even trees as large as 40 inches diameter will lose nearly half their volume to decay by the second year (Lowell et al. 1992). This volume loss corresponds to significant value loss. The first year after the fire, the value of trees removed will cover the cost of their removal and associated fuel treatments and can be used to pay for restoration work. Restoration work that can be funded by the timber value includes road repair and road surfacing to reduce erosion, road closures, treatment of additional fuels, reforestation, and watershed improvement projects. Much of the fire area is very steep, requiring expensive helicopter or cable equipment to remove undesirable levels of dead trees. By the second or third year after a fire, the value and volume loss may preclude any economic return, thereby jeopardizing the availability of funds for restoration and meeting the other purposes of the project.

4. There is a need to reduce safety hazards to the public and forest workers from falling trees: Dead and defective trees pose a significant safety hazard to forest visitors and workers. Areas within or adjacent to the fire area where falling snags must be removed include: along roads that are used by the public or by forest workers, along the Pony Express Trail, around the Sugarloaf trail system, around the Phantom Spires, around recreation residences, and in dispersed camping sites. Additionally, snags must be removed adjacent to the communities of Silver Fork and Kyburz, around their water treatment systems, and adjacent to the Silver Fork School.

DECISION AND RATIONALE FOR DECISION

Decision

Based upon my review of the alternatives, I have decided to implement Alternative 1 as described in the Final Environmental Impact Statement (FEIS).

My decision is based upon a careful consideration of the information and analysis contained in the FEIS and all of the supporting documentation, including the public comments on the Draft Environmental Impact State ment (DEIS). This environmental analysis process was conducted in accordance with the National Environ mental Policy Act (NEPA) and the direction provided in the Forest Service Manual. Alternative 1 is consistent with the goals and objectives of the Eldorado National Forest Land and Resource Management Plan (LRMP) as amended by the SNFP.

Rationale for Decision

When compared to the other alternatives, Alterna tive 1 meets the purpose for the project and is the alternative that best meets the needs identified for this project. I considered each of the needs as a factor in my decision and provide the rationale for my evaluation of each alternative against those needs.

1. There is a need to reduce long-term fuel loading for the purpose of reducing future fire severity and resistance to control.

Given the estimated fire return interval for the Freds Fire area of 4 to 18 years (FEIS pg. 64) and the estimated length of time to achieve old forest conditions (100 to 250 years), it is inevitable that fires will burn within the project area before the old forest structure and function can return. Actions taken now can influence the extent and intensity of future fires by determining the amount and configuration of fuel load in the Freds Fire area. Comparing the total fuel loading, measured in tons of down woody fuel over time, is one way to evaluate the alternatives (FEIS pg. 51 comparison of alternatives chart, Fire and Fuels section pgs. 63 to 93). Alternative 1 best meets this need by retaining the least total fuel loading over most of the project area. This will help reduce the intensity of future fires and will reduce the "resistance to control" when fires do occur (FEIS pgs. 73 to 77). Direct attack of small fires within the 165 acres of snag retention clumps may be precluded due to fire fighter safety. The areas outside of snag clumps would have a low resistance to control, would be safe for firefighters to construct fire lines along the fire's edge, would contribute to more effective fire suppression and contain fires to a smaller size.

Alternative 2 results in high fuel loadings over time, particularly in the 1000-hour fuels (>3" diameter). There is abundant scientific evidence that increased fuel loads can result in increased fire intensity and severity. The severity of fire effects and difficulty in fire sup pression are primarily associated with the total amount of fuel available and consumed (Martin and Brackebusch 1974). In other words, given the same weather and topographic conditions, areas with higher fuel loads will release more energy (burn hotter), exhibit longer flame lengths, have greater potential to convert to crown fires, be more difficult to contain, pose greater risks to fire fighters, kill more vegetation, and damage soils more severely than areas with lower fuel loads. Fuel loading contributes to high resistance to control. High snag and log levels impede fire line construction, increase safety hazards, increase spotting potential, and increase fire severity. Containment lines must be constructed far from the fireline where it is safe and practical to do so, ultimately increasing fire size. Increased soil heating from burning logs kills soil microorganisms and reduces soil productivity.

Alternative 3 increases fuel loading substantially in areas where there are many fire damaged and dying trees. As these trees die over time, they will be retained as snags in areas that are not economical to reenter (approximately 72 percent of the area), such as steep ground. These higher snag and log levels would have similar effects as described under Alternative 2.

Alternative 4 is similar to Alternative 1 except additional snags would be left that do not currently meet the marking guidelines because it would be to expensive to enter a second time based on the helicopter logging costs.

Alternative 5 reduces long-term fuel loading; however, in snag retention clumps (475 acres) and areas where four snags per acre are left it would be unsafe for effective firefighting, fuel levels by year 25 would be high enough to seriously impede fire suppression and contribute to severe soil heating. Portions of the defense zone and the ridgetop fuel breaks would be the only locations available for safe control line construction. Direct attack on small fires in snag retention clumps, portions of the threat zone, and general forest may be precluded due to fire fighter safety. Fires would therefore become larger and more intense as indirect methods are employed.

2. There is a need to improve road drainage and establish effective ground. cover in severely burned areas for the purpose of reducing erosion and sedimentation to streams in the short term and for contributing to long term soil production.

The fire has resulted in a significant increase in sediment delivery to some streams. This will continue for the next several years under all alternatives, which in turn may affect downstream beneficial uses of water. Alternatives 1, 3, and 5 may add slightly to the risk of impacts to downstream beneficial uses of water; alternative 4 would result in a negligible increase in such a risk. Alternatives 1, 3, 4, and 5 will improve the existing road system, reducing chronic sources of sedi mentation. Snag retention within riparian conservation areas will provide recruitment of in-stream large woody debris. Alternative 2 provides high levels of snags that will contribute to in-stream woody habitat for aquatic species. Hydrologically sensitive areas and steep slopes will be protected through equipment exclusion zones and a number of other protection measures.

Many areas of the Freds Fire currently lack enough effective ground cover resulting in substantially in creased erosion rates. The extent that each alternative provides for short-term effective ground cover was used as a measure of meeting this need. Alternative 2 does not meet this need as the amount of effective ground cover provided from natural needle and limb fall is unlikely to be sufficient to reduce erosion rates in the short term. Alternative 2 perpetuates the existing degraded condition resulting from the fire. Sediment loads to streams are expected to remain high and turbidity levels elevated following storm events.

Alternatives 1, 3, 4, and 5 are similar in meeting this need as mechanical treatment of dead trees will provide limbs from tree tops for ground cover. Alternative 1 best meets this need by treating both small and larger sized dead trees over the largest area.

Over the mid to long term, Alternative 1 will provide for greater watershed protection than other alternatives because Alternative 1 reduces the risk of severe impacts to soils and beneficial uses of water from another wildfire. Alternative 1 reduces fuel loading and improves resistance to control, thereby increasing the potential to suppress fires during initial attack when they are small, with less impact to soil cover; and less potential for destruction of soil organisms and organic matter from lethal temperatures (FEIS pgs. 73 to 77, 113 to 120, 127 to 153).

3. There is a need to act rapidly to remove some of the dead trees while they retain economic value that can offset the cost of other restoration work.

This element highlights the need to make a decision about removing dead trees within the Freds Fire in a timely manner. This factor also highlights the relationship between the economic value of the dead trees and meeting the other ecological and social needs for action. The economic value, in terms of wood quality, decreases rapidly in dead trees with the potential for substantial loss within 3 years (FEIS pg. 157). Trees that were immediately killed by the fire will have 10 months worth of decay by the time this project is implemented and some of the smaller diameter dead trees may have already lost all economic value. By the second year following the fire, about 65% of the wood volume will be lost to decay (FEIS pg. 157; Emergency Situation Analysis for the Freds Fire Restoration Project). Deterioration of the fire-killed and damaged trees has a number of impacts. Lumber quality is reduced, merchantable volume of wood is reduced, and most importantly, value is reduced. This reduction in value impacts the revenue the Federal Government receives from stumpage and it also impacts the ability to cost-effectively remove the dead trees at all. Ultimately, after trees have significantly deteriorated, the cost of removal far exceeds the value of the trees, and the Government is faced with the dilemma of an increasing fuel load and no immediate funds available to mitigate the impact. In California, there are numerous examples of recent fires where delay caused no-bid timber sales and the timber was not harvested because of deterioration rendering the trees worthless in the marketplace and unsafe to harvest. On the Gap Fire of 2001 on the Tahoe National Forest, one year after the fire the value of the trees dropped below the cost of removing them with helicopters. The consequence was an estimated $400,000 loss of timber value that impacted the ability to treat fuels (refer to the Emergency Situation Analysis prepared for the Freds Fire Restoration Project). On the Star Fire on the Tahoe National Forest, delay caused an estimated $2,600,000 loss in timber value. This amount of money could have paid to hand pile at least 4,300 acres of fuel on steep terrain (refer to, the Emergency Situation Analysis prepared for the Freds Fire Restoration Project).

Delay will diminish the opportunities to utilize dead trees to benefit human needs, lessen the ability to leverage the timber value to treat fuels, repair and improve roads, and fund other restoration treatments, and reduce our ability to meet the long term goals and achieve the purpose and need for the project. For these reasons, I requested, and the Chief, Forest Service approved, an emergency situation as provided by 36 CFR 215.10.

Alternative 1 generates the greatest revenue, $3,345,872 (estimated gross revenue), which is sufficient to cover the costs of road improvements, fuel treat ments, and future reforestation if any. Alternative 2 does not meet this need because no revenue would be generated and road costs and potential future refore station costs would be entirely dependent upon appro priated or other funding sources. Alternative 3 gen erates an estimated $2,676,564; Alternative 4 generates an estimated $1,586,256; and Alternative 5 generates an estimated $3,280,582. All action alternatives generate sufficient funds to pay for fuel treatments and road improvements except Alternative 4.

4. There is a need to reduce safety hazards to the public and forest workers from falling trees.

All action alternatives reduce the risk of falling snags to some extent, but no alternative completely eliminates the risk. Alternative 1 best reduces the risk by pre scribing the most acres with lower snag levels. By clumping snags into 3 large patches, there will be the potential to create areas with fewer snag risks. Alternative 2 has the highest risk to public and forest worker safety since all dead trees will remain. Some level of risk reduction has already occurred along roads within the fire area. Under Alternative 1, as additional hazards are identified, they will require a separate decision to remove or retain them. Alternative 2 best meets this need by retaining lower numbers of snags across more acres in the project area. In the short-term, the risk to the public entering the Freds Fire area is great as smaller trees will tend to be susceptible to falling soon and the tops of larger trees will begin to fall, especially in the winter. There are additional risks due to timber harvest activities that are occurring along the roadside on private land.

I also considered the public comments and the signi ficant issues in making my decision. Alternative 1 is responsive to the public comments and addresses most of the significant issues (FEIS Chapter 1, pgs. 22 to 24). These issues are as follows:

Whether use of the Pacific Southwest Region Forest Health Protection Staff's mortality guidelines presents an unacceptable risk of cutting trees that would have survived their fire injuries. There is a tradeoff between minimizing the risk of cutting a live tree and the need to repeatedly log the same ground to remove trees as they die. There is a high likelihood that repeat entries would be made in ground based logging units, and a lower probability in skyline and helicopter logged areas. The latter systems are more costly and dependent on higher volumes per acre for economic feasibility. Repeat entries increase the risk of soil disturbance, erosion, and compaction. Lack of repeat entries increases the number of snags retained and correspondingly increases the long term fuel accumulation. In the Star Fire, where mortality guidelines were not used, 22 percent additional cubic volume of timber was removed in the second and third reentry on tractor units. No reentry occurred on skyline and helicopter logged areas. Alternative 3 does not use mortality guidelines and consequently there is a higher risk of significantly increasing the number of snags retained and associated fuel accumulation. The other alternatives, including Alternative 1, mitigates the risk of cutting trees that would otherwise survive by utilizing the latest and most comprehensive scientific assessment of tree mortality following wildfire in California (Hood et al. 2005). The latest science assessing fire related mortality provides models that allow managers to select the desired level of predicted mortality based on land management objec tives. Tree marking guidelines are derived from the probabilities, as described in the FEIS on page 61 and in Appendix A. For Alternatives 1, 4, and 5, I selected a more conservative probability of mortality for areas where ground based logging was proposed, and a slightly less conservative probability of mortality where aerial logging systems were proposed as displayed in the table below.

 

Table 1. Probability of Tree Mortality as applied to Alter natives 1, 4, and 5.

PP/JP/SP White Incense Red Douglas

Fir Cedar Fir Fir

Tractor Logging System

Correctly 96 95 100 100 -

Predicted

Mortality (%)

Correctly 51 63 88 - -

Predicted

Survival (%)

Helicopter and Skyline Logging System

Correctly 90 87 85 100 -

Predicted

Mortality (%)

Correctly 65 74 89 - -

Predicted

Survival (%)

Marking guidelines for Douglas fir are based on Ryan and Reinhardt (1988).

As displayed in the table, the guidelines for tractor logged areas reduce the chance of trees being cut which might otherwise have survived, while accepting a relatively high probability of many additional trees which do not meet the guidelines subsequently dying. Marking guidelines for helicopter and skyline logging result in a slightly higher potential of trees being cut which might otherwise have survived, and a lower probability of additional trees, which do not meet the guidelines, subsequently dying.

The rationale for a more conservative mortality marking guide in tractor logged areas stems from ex perience in previous fire salvage sales as described above. Since tractor logged areas are typically re- Iogged when significant numbers of additional trees die, fuel loading from additional dead trees can be mitigated. In order to mitigate the fuel accumulation from delayed mortality on steep slopes, more of the dying trees need to be removed initially because high yarding costs, setup time, and limited equipment availability make it unlikely that helicopter or skyline units would be re-logged to capture mortality occurring after the initial logging.

Whether ground based and "cable" logging would result in unacceptable impacts on soil and downstream beneficial uses of water. Alternative 1 will improve the existing road system, reducing chronic sources of sedimentation. Snag retention will provide recruitment of in-stream large woody debris. Hydrologically sensi tive areas and steep slopes will be protected through use of aerial logging methods, equipment exclusion zones, application of Best Management Practices and a number of other protection measures described in the FEIS on page 32, which I believe will be effective based on monitoring on other projects and my background and experience as a hydrologist. While Alternative 4 was designed to address this issue, and would minimize ground disturbance, it is very expensive to implement and I do not believe the perceived benefits justify the high cost.

Whether snag retention levels proposed are seriously deficient to maintain viability and diversity for the array of snag dependent and snag associated species in the project area. Alternative 2 retains all snags, thereby maximizing habitat for cavity nesting species, but fails to meet the needs identified for the project as articu lated above and in the FEIS. Cavity nesting birds are management indicator species for the Eldorado National Forest, but are not threatened, endangered, or sensitive. There are many cavity nesting bird species that find habitat in the Freds Fire, and under the Eldorado National Forest Land Management Plan I am charged with providing medium capability habitat for a suite of species.

Alternative 1 provides snags for cavity nesting birds potentially using the fire area, but arranges them in patches as well as distributed in Riparian Conservation Areas (RCAs). The patches are located in areas of large trees (Wildlife Habitat Relationships (WHR) size class 4 and 5), but contain both large and small trees. Alter native 1 provides for species that are associated with burned forest habitat, such as hairy and black-backed Woodpeckers, as well as species that are snag habitat generalists and species associated with forest environ ments.

Alternative 5 enhances the snag habitat of Alterna tive 1 by retaining additional large clumps of snags and retains snags in the largest size class within RCAs. Alternative 5 provides more habitat than the other action alternatives for the short-term needs (5 years) of the species that are associated with the burned forest habitat. Although this Alternative 5 provides the most habitat for the short term, it is the least responsive to the purpose and need to reduce long-term fuel loading, reduce safety hazards to forest workers, and increase ground cover in the short term in high intensity burn areas.

PUBLIC INVOLVEMENT

The Notice of Intent to prepare an Environmental Impact Statement was published in the Federal Register December 27, 2004. A brief description of the location and type of project was included in the Eldorado National Forest Schedule of Proposed Actions (SOPA) in January 2005. Approximately 92 letters were mailed out to adjacent property owners; potentially affected businesses; federal, state, and local agencies; and special interest groups. The letter contained the detailed proposed action, map, methods for partici pation, and an invitation to the Freds Fire Restoration public meeting on January 13, 2005. The mailing list is included in the project record. An article was published in the Mountain Democrat on January 19, 2005 de scribing the proposed action and public meeting. Approximately 36 people attended the public meeting, including local residents and adjacent property owners. Meeting notes are included in the project record. Ten individuals responded to the scoping. Significant issues were raised and alternatives to the proposed action were developed. Appendix B of the FEIS contains the scoping comments received and a summary of how the comments were addressed is in the project file.

After reviewing the public scoping comments, I approved the following significant issues to generate alternatives:

o Whether use of the Pacific Southwest Region Forest Health Protection Staff's mortality guide lines (Appendix A) presents an unacceptable risk of cutting trees that would have survived their fire injuries: During scoping, concerns were ex pressed about the use of these guidelines and the effects on live trees. While we believe these guidelines to be the best available science appli cable to the project area and they were developed based on scientific literature on this subject, extensive post-fire monitoring in the Sierra Nevada, and professional judgment, predictions of mortality are not an exact science. The pro babilities are high that trees meeting the mortality guidelines are dead or dying, however there is a small probability that trees meeting the guidelines would survive, or that trees not meet ing the guidelines would die. The concern by some respondents is that the guidelines are not accurate enough, and any risk of cutting a tree that would otherwise survive, is too much. These respondents point out that retention of all live trees, particularly large trees, is important for seed sources, shading, and wildlife habitat, and requested an alternative that does not utilize mortality guidelines.

o Whether ground based and "cable" logging would result in unacceptable impacts on soil and down stream beneficial uses of water: In some scoping comments, disagreement related to effects of the Proposed Action on soil compaction, erosion, and sedimentation were expressed. Several studies were cited supporting these concerns. In parti cular, some respondents were concerned that ground based and "cable" logging in moderate to high severity burn areas would exacerbate erosion and sedimentation. Technically speaking, cable logging refers to dragging logs up or down hill with the use of cables suspended from a tower sitting on a road, and may not necessarily involve full or partial suspension of the logs above the ground. The proposed action utilizes "skyline" logging techniques which require full or partial suspension of the logs as they are brought up the hill. For purposes of this issue however, we assumed that "cable" logging and "skyline" log ging were synonymous. These respondents re quested an alternative that does not utilize ground based or cable logging techniques.

o Whether snag retention levels proposed are seriously deficient to maintain viability and diversity for the array of snag-dependent and snag- associated species in the project area: There was also concern that the proposed action does not adequately address the ten percent retention standard to leave burned areas with large trees standing for the benefit of wildlife species depen dent on pulses of large, and significant numbers of snags, such as the black backed and Hairy woodpeckers. This issue was addressed by de veloping an alternative that leaves large clumps of snags within the high mortality portions of the project area (see FEIS pg. 11, Map 1-2).

The Notice of Availability of the Draft Environ mental Impact Statement (DEIS) was published in the Federal Register on March 25, 2005 and copies of the DEIS/project summary were mailed to 37 individuals, organizations, tribes, and government agencies. The comment period ended on May 9, 2005. Eleven in dividuals responded during the comment period. Appendix C of the FEIS contains the comment letters and Appendix D contains the response to comments. An open house was held on April 7, 2005 to discuss the DEIS. In addition, a field trip was held on April 25, 2005 to visit the project area and answer questions.

The 45-day comment period started on March 25, 2005 and ended on May 9, 2005. Comments were asses sed and considered and the following actions were taken: alternatives were modified; an additional alternative was developed, considered, and eliminated from detailed study; factual corrections were made; and the environ mental consequences were supplemented.

ALTERNATIVES CONSIDERED

In addition to the selected alternative, I considered 4 other alternatives, which are discussed below. A more detailed comparison of these alternatives can be found in Chapter 2, pages 47 to 54 of the FEIS.

Alternative 1 is the Proposed Action and Preferred Alternative. Under this alternative all dead and dying trees in the areas within the distance limitations pre scribed for perennial streams, oak stands, and large contiguous low intensity burn areas would be retained. All existing down logs and cull logs greater than or equal to 20 inches small end diameter and over 10 feet long generated from logging would be retained, and standing snags would be retained in variable numbers and sizes in riparian conservation areas, threat zones, defense zones, and general forest. Roadside hazard trees would be removed. Logging would occur using helicopter, skyline, and ground-based machinery. Pro tection of RCAs, sensitive plants, cultural resources, and wildlife would occur. Treatment of fuels and manipu lation of ground cover would occur. Roads would be reconstructed and maintained to improve watershed condition. Less than 1 mile of new road would be constructed to access tractor units.

Alternative 2 is the No Action alternative. Under this alternative no dead tree removal or treatment would occur. Projects associated with the fire suppression rehabilitation and Burn Area Emergency Response (BAER) would continue.

Alternative 3 responds to the concern that use of the Pacific Southwest Region Forest Health Protection Staffs mortality guidelines presents an unacceptable risk of cutting trees that would have survived their fire injuries. Alternative 3 is similar to Alternative 1 except mortality guidelines that predict which trees will die of their injuries would not be used to mark trees for harvest. Protection measures are the same as for Alternative 1.

Alternative 4 was designed to address the issue that ground based and "cable" logging would result in unacceptable impacts on soil and downstream beneficial uses of water. Alternative 4 is similar to Alternative 1 except that helicopter logging would replace skyline and ground based logging and there would be no new road building.

Alternative 5 was designed to address the issue that snag retention levels proposed are too low to maintain viability and diversity for the array of snag-dependent and snag-associated species in the project area. Alter native 5 is similar to Alternative 1 except that additional dead trees (equal to 4 snags per acre across the project area) would be retained in large clumps over the landscape. Outside of RCAs, this alternative was designed to leave snags for wildlife in varying sized clumps instead of as individual trees spread over the landscape, avoiding ridges, roads, powerline, and private land. As first priority, clumps are located in WHR size class 5 stands where mortality exceeds 66 percent. Second priority is WHR size class 4 where mortality exceeds 66 percent. Clumps would be located to avoid areas strategic to fire suppression. Where possible, clumps would be located on steeper ground where logging costs tend to be high and where people are less likely to recreate. Snag retention in perennial and seasonal RCAs is the same as for Alternative 1.

Alternatives Considered But Eliminated

From Detailed Study

Federal agencies are required by NEPA to rigorously explore and objectively evaluate all reason able alternatives and to briefly discuss the reasons for eliminating any alternatives that were not developed in detail (40 CFR 1502.14). Public comments received in response to the Proposed Action provided suggestions for alternative methods for achieving the purpose and need.

An alternative that defers action until 15 years post fire was proposed by some members of the public. Under such an alternative the surface fuels would be piled and burned and felling of small snags for ground cover could occur. This alternative was eliminated from detailed study because it does not accomplish the purpose and need for the project. Specifically, this alternative does not generate any revenue that could be used to offset the costs of the fuel treatment or any other restoration activity. This alternative would retain nearly all dead material on site, increasing safety hazards and fuel loads over time to unacceptable levels. By 15 years post fire, dead trees would be significantly deteriorated to a point that it would be extremely hazardous to perform any of the treatments this alternative would require (FEIS pgs. 42 to 43). Even if the work could be done safely, piling and burning of surface fuels is expensive on large areas of steep slopes, exceeding four million dollars as displayed on page 43 of the FEIS. Additionally, this alternative would not generate any effective ground cover in the short term in high intensity burn areas.

An alternative that retains all dead trees greater than 15 inches in diameter was proposed. This alterna tive would fell and lop smaller trees for ground cover, and treat small sized fuels through piling and burning. This alternative would preclude skyline logging due to the prevalence of trees greater than 15 inches that would interfere with skyline corridors. It is assumed that all areas proposed for skyline logging would be helicopter logged instead. This alternative responds to the issue of whether there is a need to remove large dead trees (over 15") to reduce future fire severity and intensity. Some respondents assert that "by the time most of the large snags have fallen and substantially decayed, the small snags, and branches and tops from large snags, will have long since fallen and decayed into soil." These respondents requested an alternative that retains all dead trees over 15" diameter. Based on fuels modeling displayed in the FEIS on page 45, and the effects on fire behavior and resistance to control, this alternative fails to meet the purpose and need for the project. This alternative fails to reduce future fire intensity and would lead to significant tree mortality in the event of a fire, hindering or reversing the growth of future forest. In addition, numerous snags would continue to inhibit or preclude safe firefighting practices within the project area; the alternative is not practical or capable of being implemented because there is little economic value in trees less than 15 inches to generate sufficient, funds for fuel treatment and other restoration treatments; and future fires would likely have significant and long lasting damage to soil productivity.

An alternative was proposed by two environmental groups in response to the DEIS. The alternative would retain 640 acres of California Wildlife Habitat Relation ships (CWHR) class 4 and 5 habitat (within high mortal ity areas) and 402 acres of CWHR class 3 habitat (within high mortality areas) in snag retention clumps larger than 30 acres across the fire. This alternative would include removal of dead trees with no green limbs, retention of four of the largest snags outside of the snag retention clumps, and removal of snags less than 6 inches dbh within the 1,042 acres of snag retention clumps to reduce fuel build up. The purpose of this alternative was to increase the habitat available for black backed woodpeckers (as well as other wood peckers) and provide additional snags for legacy struc ture for future forest. This alternative was eliminated from detailed study because it does not accomplish the purpose and need for the project. Specifically, this alternative does not meet the purpose and need to reduce longterm fuel loading, reduce safety hazards to forest workers, and increase ground cover in the short term in high intensity burn areas.

Environmentally Preferable Alternative

The NEPA implementing regulations (Section 1505.2) require that the alternative(s) that best pro motes the national environmental policy as expressed in NEPA, Section 101, be identified in the Record of Decision as the "environmentally preferable Alterna tive" or alternatives. This is ordinarily "the alternative that causes the least damage to the biological and physical environment and best protects, preserves, and enhances historical, cultural, and natural resources" (FSH 1909.15, 05). For the Freds Fire Restoration project, I believe Alternative 1 is the environmentally preferred alternative for reasons discussed above.

For this project, all of the action alternatives provide protection to the environment by their design, management requirements, and built-in mitigations. All action alternatives comply with NEPA's purpose and spirit.

Findings Required by Other Laws and Regulations

All management practices and activities of Alterna tive 1 are consistent with the 'management direction of the Eldorado National Forest Land and Resource Management Plan (LRMP) as amended.

Clean Air Act

The Clean Air Act makes it the primary responsi bility of States and local governments to prevent air pollution and control air pollution at its source. States must have a plan that provides for implementation, maintenance, and enforcement of the primary ambient air quality standard. The State of California has a plan. The procedures outlined in the FEIS are sufficient to minimize air quality concerns. This project meets the Clean Air Act.

Clean Water Act

Federal agencies are required by the Clean Water Act to cooperate with State agencies in preventing, reducing, and eliminating pollution in concert with programs for managing water resources. This project meets this through the incorporation of Best Manage ment practices listed in the project file. This project meets the Clean Water Act.

National Environmental Policy Act (NEPA)

The NEPA requires that Federal agencies complete detailed statements on proposed actions that signifi cantly affect the quality of the human environment. The Act's requirement to prepare an environment impact statement is designed to provide decision makers with a detailed accounting of the likely environmental effects of a proposed action prior to adoption and to inform the public of, and allow it to comment on, such effects. The FEIS does a comprehensive job of analyzing the alter natives and displaying the alternatives and displaying the environmental effects. The procedural requirements of the NEPA have been followed.

National Forest Management Act (NFMA)

Projects occurring on NFS lands must meet mini mum specific management requirements under 36 CFR 219.27. This project and the FEIS address each as follows:

o The management prescriptions discussed in the FEIS meet all of the resource protection require ments of the CFR.

o The requirement for vegetative manipulation is not applicable to this project because this involves salvage of fire killed trees and does not involve the manipulation of tree cover.

o The project was reviewed by a certified silviculturist and found to be in compliance with all aspects of the silvicultural practices require ment.

o The uneven requirement does not apply.

o This project is consistent with the requirements for riparian areas.

o This project meets the requirements for soil and water.

o Biological Evaluations considered effects to for Forest Service sensitive species. The Biological Evaluations determined that the project was not likely to result in a loss of viability for any sensi tive species.

Endangered Species Act (ESA)

Section 7(a)(2) of the Endangered Species Act re quires that Federal agencies consult with the United States Fish and Wildlife Service and National Marine ' Fisheries Service, as appropriate, to ensure that their actions do not jeopardize the continued existence of species listed as threatened or endangered under ESA, or destroy or adversely modify their critical habitat.

A biological assessment was prepared for federally proposed, threatened or endangered wildlife and botany species and their critical habitat. Implementation of the project would have no effect on American bald eagle, great gray owl, valley elderberry longhorn beetle, western red bat, willow flycatcher, California redlegged frog, foothill yellow-legged frog, mountain yellow-legged frog, California wolverine, Arctostaphylos nissenana, Botrychium spp., Draba asterophora var. asterophora, Draba asterophora var. macrocarpa, Epilobium howellii, Eriogonum tripodium, Horkelia parryi, Lewisia longi petala, Lewisia serrata, Lomatium stebbinsii, Meesia spp., Navarretia prolifera ssp. Lutea, Phacelia steb binsii, and Senecio layneae. Implementation of the project may effect individuals or habitat, but is not likely to result in a trend toward Federal listing or loss of viability for the following species: California spotted owl, Pacific fisher, pallid bat, northern goshawk, Town send's big-eared bat, American peregrine Falcon, American marten, northwestern pond turtle, Sierra Nevada red fox, Calochortus clavatus var. avius and Cypripedium montanum.

National Historic Preservation Act

It was determined under the Programmatic Agree ment for Compliance with Section 106 of the National Historic Preservation Act for Undertakings on the National Forests of the Pacific Southwest Region, that the proposed undertakings will have no direct, indirect or cumulative effect on cultural properties and values.

Implementation Date

The Chief of the Forest Service has determined that an emergency situation exists for the entire project area as provided for in 36 CFR 215.10.

Implementation may begin immediately for that portion of the decision determined to be an emergency.

ADMINISTRATIVE REVIEW or APPEAL OPPORTUNITIES

This decision is subject to administrative review (appeal) pursuant to 36 CFR Part 215. The appeal must be filed (regular mail, fax, email, hand-delivery, or express delivery) with the Appeal Deciding Officer at: Bernard Weingardt, Regional Forester, USDA Forest Service, Regional Office R5, 1323 Club Drive Vallejo, CA 94592, fax: (707) 562-9229.

The office business hours for those submitting hand- delivered appeals are: 8:00 am to 4:30 pm Monday through Friday, excluding holidays. Electronic com ments must be submitted in a format such as an email message, plain text (.txt), rich text format (.rtf), or Word (.doc) to appeals-pacificsouthwest-regional-office. In cases where no identifiable name is attached to an electronic message, a verification of identity will be required. A scanned signature is one way to provide verification.

Appeals, including attachments, must be filed within 45 days from the publication date of this notice in the Mountain Democrat, the newspaper of record. Attach ments received after the 45-day appeal period will not be considered. The publication date in the Mountain Democrat, newspaper of record, is the exclusive means for calculating the time to file an appeal. Those wishing to appeal this decision should not rely upon dates or timeframe information provided by any other source. Individuals or organizations that submitted substantive comments during the comment period specified at 215.6 may appeal this decision. The notice of appeal must meet the appeal content requirements at 36 CFR 215.14.

Contact Person

For additional information concerning this decision or the Forest Service appeal process, contact Laura Hierholzer, Interdisciplinary Team Leader, Placerville Ranger District, 4260 Eight Mile Road, Camino, CA 95709, (530) 647-5382.

08/01/2005

Date

Record of Decision

/s/ JOHN D. BERRY
JOHN D. BERRY
Forest Supervisor

APPENDIX G

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

__________

No. 05-16776

D.C. No. CV-05-01608-MCE

EARTH ISLAND INSTITUTE, A CALIFORNIA PROFIT ORGANIZATION; CENTER FOR BIOLOGICAL DIVERSITY, A NON-PROFIT ORGANIZATION,
PLAINTIFFS-APPELLANTS

v.

UNITED STATES FOREST SERVICE; DALE BOSWORTH, CHIEF OF THE UNITED STATES FOREST SERVICE; JOHN BERRY, FOREST SUPERVISOR FOR EL DOREDO NATIONAL FOREST,
DEFENDANTS-APPELLEES

SIERRA PACIFIC INDUSTRIES,

DEFENDANT-INTERVENOR-APPELLEE

__________

[Filed: Jul. 12, 2006]

__________

Before: NOONAN, TASHIMA, and W. FLETCHER,

Circuit Judges.

The panel has voted to deny the petition for re hearing. Judge Fletcher has voted to deny the petition for rehearing en banc; and Judges Noonan and Tashima so recommend.

The full court has been advised of the petition for rehearing en banc and no judge of the court has requested a vote on whether to rehear the matter en banc. Fed. R. App. P. 35.

The petition for rehearing and the petition for rehearing en banc, filed June 7, 2006, are DENIED.