Environmental Protection Information Center v. California Department of Forestry & Fire Protection

Opinion MORENO, J. This case arises from the “Headwaters Agreement” consummated by the Pacific Lumber Company and the state and federal governments. The agreement was intended to settle matters of litigation and public controversy surrounding the intensive logging of old growth redwoods and other trees on Pacific Lumber’s property in Humboldt County. In addition to the state and federal governments’ purchase of a relatively small portion of Pacific Lumber’s property for conservation purposes, it was agreed that Pacific Lumber could log the rest of its property, provided that it obtain certain regulatory approvals from state and federal agencies. The deadline for obtaining these approvals was March 1, 1999. The approvals were timely obtained, in some cases right at the March 1 deadline. Shortly thereafter, various environmental and labor groups challenged the validity of the regulatory approvals on numerous grounds. The trial court resolved the issues mostly in favor of the environmental and labor groups in 2003, and the Court of Appeal, at the end of 2005, reversed the trial court on almost every point and upheld each of the regulatory approvals at issue. We granted review in February 2006 to consider a number of issues, many of them of first impression, in this important case. The case was put on hold due to a stay resulting from Pacific Lumber’s filing for chapter 11 bankruptcy in February 2007. The stay was lifted in August 2007. We conclude that one of the challenges to Pacific Lumber’s Sustained Yield Plan (SYP), which, as explained below, is a kind of master plan for logging a large area, is valid, inasmuch as an identifiable plan was never approved. We also conclude, as explained below, that any resubmitted SYP should have an adequate analysis of individual planning watersheds, which the plan as originally approved did not contain. We further conclude that the state Incidental Take Permit, authorizing the capturing and killing of endangered and threatened species incidental to lawful activity, was deficient *471 because it included overly broad “no surprises” clauses limiting in advance Pacific Lumber’s obligation to mitigate the impacts of its logging operations. In all other respects, we affirm the Court of Appeal opinion, and remand the matter for further proceedings consistent with this opinion. I. Statutory and Regulatory Framework One of the obstacles to the proper understanding of this complex case is that it concerns a myriad of regulatory approvals, each approval supported by a document or documents that are to some degree interrelated with the others. Before discussing the facts of this case, an overview of the regulatory approvals required and the governing statutes is in order. A Sustained Yield Plan is a kind of master plan for logging a large area, authorized by statute (Pub. Resources Code, § 4551.3) and regulation (Cal. Code Regs., tit. 14, §§ 1091.1-1091.14), 1 designed to achieve the Forest Practice Act’s objective of obtaining the maximum timber harvest consistent with various short- and long-term environmental and economic objectives. (Z’berg-Nejedly Forest Practice Act of 1973; Pub. Resources Code, § 4511 et seq.) As explained below, the SYP does not replace the more specific timber harvest plan (THP), but inasmuch as the SYP adequately analyzes pertinent issues, a THP may rely on that analysis. Although SYP’s are usually voluntary at the option of the landowner, in this case the SYP was required by the Headwaters Agreement. Also required in this case under federal law was a Habitat Conservation Plan. Although the “taking” of a federally listed endangered species, i.e., the killing, capturing or harming of such species (16 U.S.C. § 1532(19)), is generally unlawful (id., § 1538), a permit for the taking of a species incidental to an otherwise lawful activity, known as an Incidental Take Permit, may be issued when an applicant submits to the Secretary of the Interior a Habitat Conservation Plan. (16 U.S.C. § 1539(a)(2)(A).) The plan is to specify, among other things, the impacts that will likely result from the taking and the steps the applicant intends to employ to minimize and mitigate those impacts. (Ibid.) Although the federal Incidental Take Permit is not challenged in this appeal, the Habitat Conservation Plan (HCP) was combined with the SYP for purposes of environmental review, and is critical to supporting various other approvals at issue in this case. In addition to a federal Incidental Take Permit, Pacific Lumber in this case was required to obtain a state Incidental Take Permit for species listed as endangered or threatened under the California Endangered Species Act (Fish & G. Code, § 2050 et seq.). *472 In conjunction with approval of the HCP, the United States Fish and Wildlife Service, Pacific Lumber, and various state agencies entered into an Implementation Agreement for the HCP, defining the obligations of each party under the HCP. Because the state SYP and federal HCP contained overlapping and interrelated analyses and provisions, a decision was made to prepare for both of these documents a single joint environmental impact report (EIR) under the California Environmental Quality Act (CEQA) (Pub. Resources Code, § 21000 et seq.) and an environmental impact statement (EIS) under the National Environmental Policy Act of 1969 (42 U.S.C. § 4321 et seq.). Thus, the EIS/EIR also overlaps and is interrelated with the SYP and the HCP, each of these documents considering among other things the impact of proposed logging activities on wildlife and wildlife habitat. In fact, as explained below, substantial portions of the draft and final EIS/EIR were incorporated into and became part of the SYP. Finally, Pacific Lumber was required to apply for a Streambed Alteration Agreement, pursuant to Fish and Game Code former section 1603. As will be further explained, that statute imposed on Pacific Lumber and the Department of Fish and Game (DFG) the obligation to negotiate an agreement that would mitigate impacts on fish and wildlife caused by the obstruction or diversion of streams and other watercourses. With this framework in mind, we turn to the facts of this case. II. Factual and Procedural Background Real parties in interest Pacific Lumber Company, Scotia Pacific Company, LLC, and Salmon Creek Corporation (hereinafter collectively referred to as Pacific Lumber) own approximately 211,000 acres of timberland in Humboldt County that have been used for commercial timber production for some 120 years. In 1986 Pacific Lumber was acquired by Maxxam Incorporated, and in order to pay off Maxxam’s debt for the buyout, Pacific Lumber began cutting down old growth redwoods at a faster rate than ever before. The deforestation led to litigation and considerable local protest. In the 1990’s, as a result of federal and state litigation, Pacific Lumber was enjoined from harvesting a particular stand of old growth timber that served as the habitat for the marbled murrelet, an endangered bird. Pacific Lumber, in turn, filed lawsuits alleging an unlawful taking by the state and federal governments of the land declared unusable for timber production and harvesting. To resolve the existing controversies, Pacific Lumber entered into the Headwaters Agreement of 1996 with the State of California and the United *473 States. The agreement provided for the sale of some 7,000 acres of Pacific Lumber’s timberland to the federal government and the State of California, and for Pacific Lumber to obtain the various regulatory approvals discussed above for its remaining 211,000 acres. What follows is a brief history of Pacific Lumber’s process of gaining these approvals—a history necessary to understanding the issues in this case. Pacific Lumber submitted a draft of the SYP to California’s Department of Forestry and Fire Protection (CDF) 2 for its consideration in December 1996. The CDF forwarded the draft SYP to numerous state and federal agencies seeking their comments on February 20, 1997. As will be elaborated on below, the agencies’ comments were critical of, among other things, the draft SYP’s decision to employ large watershed assessment areas that would not accurately register the impacts of logging on individual watersheds and the fish and wildlife they contained. Pacific Lumber provided responses to those comments on April 1, 1997. The state and federal governments and Pacific Lumber entered into a “Pre-permit Application Agreement in Principle” on February 27, 1998, setting forth the regulatory framework governing the approval of further logging on Pacific Lumber’s property. Under the terms of the agreement, Pacific Lumber agreed that its Incidental Take Permit and HCP would be for a term of 50 years. The agreement set forth various specifications for the Implementation Agreement for the Incidental Take Permit and the HCP and agreed to incorporate various conservation measures. The agreement also provided that Pacific Lumber would submit an SYP to CDF and that “[u]pon receipt from Pacific Lumber of an SYP incorporating CDF’s request for timber growth estimates, CDF will find the SYP sufficient for public review.” The agreement further provided that “[t]he SYP will be evaluated by the [DFG] and CDF” under the California Endangered Species Act (CESA; Fish & G. Code, § 2050 et seq.), the Forest Practice Act, “and other applicable state statutes to ensure that it satisfies applicable statutory requirements.” In June 1998, Pacific Lumber submitted a draft combined SYP/HCP, referred to in the litigation as the Public Review Draft. The draft contained descriptions of existing or “baseline” forest conditions, projections of long-term sustained yield, impacts of anticipated logging on habitat and wildlife, and proposed mitigation measures and management practices to minimize those impacts. The draft was released to the public on July 14, 1998. *474 The HCP proposed three primary conservation strategies to protect endangered and threatened species. The first is the establishment of a series of reserves, i.e., contiguous areas of second growth and old growth redwoods called marbled murrelet conservation areas, after the endangered sea bird whose habitat is partly within Pacific Lumber’s land. These conservation areas are to total approximately 8,446 of the 211,000 acres, including 1,522 acres of uncut old growth redwoods and 3,174 acres of second growth redwoods. These reserves would be for the most part, though not entirely, protected from timber harvesting. The second strategy was the establishment of a series of riparian management zones around streams, establishing no-cut buffers around the streams varying in size depending on the extent to which wildlife was found in the stream, and taking various measures to reduce the amount of sediment that accumulates in the streams. The third strategy would include various timber harvesting or “silvicultural” practices that would protect wildlife, including use of “best management practices” to monitor the forest fish and wildlife and ensure that their populations do not fall below baseline levels. At about the same time, Pacific Lumber officially submitted an application for an Incidental Take Permit to the United States Fish and Wildlife Service and the National Marine Fisheries Service. In order for the Headwaters Agreement to become a reality, Congress and the Legislature were required to approve funding for the purchase of the Headwaters Forest. Congress authorized $250 million in October 1997, conditioned on the approval of all required regulatory permits on or before March 1, 1999. On September 1, 1998, the Legislature passed Assembly Bill No. 1986 (1997-1998 Reg. Sess.) (Stats. 1998, ch. 615 (hereafter Assembly Bill 1986)), which approved expenditures of $245.5 million. Assembly Bill 1986 conditioned the appropriation of funds on modification of the HCP to provide additional measures to mitigate the impacts of Pacific Lumber’s logging on threatened and endangered species. {Ibid.) In October 1998, a draft EIS/EIR was released analyzing the Headwaters acquisition and the Public Review Draft SYP/HCP. The EIS/EIR evaluated five separate alternative harvest levels ranging from 86.9 million board feet (mmbf) to 233.5 mmbf per year on average for the first decade. 3 Pacific Lumber’s proposal for timber operations (alternative 2) had the highest projected harvest volume of 233.5 mmbf per year on average in the first *475 decade. With each of these alternatives, projections were made over a 120-year period and harvest levels were projected to decline in the middle decades and rise again in the later decades. The public draft EIS/EIR also noted that Pacific Lumber’s SYP/HCP application had not been modified in response to the restricting provisions of Assembly Bill 1986, but that the draft EIS/EIR would include assessments of the environmental effects of implementing the legislation. On January 22, 1999, the final EIS/EIR was issued, after an intensive period of public comment. Additional mitigation measures, which were added to the final HCP, reduced the land available for harvest and therefore reduced the estimated long-term sustained yield. The final HCP also called for the analysis of the impacts of logging on individual planning watersheds within five years of the SYP/HCP’s approval. The EIS/EIR was critical of the high harvest levels contemplated by Pacific Lumber and the methodology it used. As the EIS/EIR states, the harvest level projections were based on intensive management methods, “such as site preparation, planting improved stock, herbicide application to control competing vegetation, and thinning to concentrate growth,” which were expected to increase harvest yields. The EIS/EIR commented that Pacific Lumber “has not managed its land using these intensive management practices until recently. Therefore, there is no record to judge [Pacific Lumber’s] likely success at achieving the projected growth increases. If the higher harvest during the first two decades are not followed by a continuing and successful intensive management program, there will be a considerable decrease in timber available for harvest in the following decades.” As a result of this skepticism, as well as the recognition that the restrictions imposed by Assembly Bill 1986 and the final HCP would mean reduced harvest levels, the EIS/EIR proposed to reduce harvest levels from the 233.5 mmbf per year on average in the first decade. Appendix Q to the EIS/EIR, which purported to set forth the contents of the final SYP, proposed a new long-term sustained yield of about 196.5 mmbf per year, with an estimated harvest volume for maximum sustained production of 176.1 mmbf per year on average for the first decade. 4 Appendix Q also contained a “crosswalk,” or *476 index, that purported to identify where all the components of a final SYP were to be found. The EIS/EIR also included a listing of changes to the draft EIS/EIR in response to public comments. On October 29, 1998, Pacific Lumber officially applied for a state Incidental Take Permit and notified DFG that it would be seeking a Streambed Alteration Agreement pursuant to Fish and Game Code former section 1603. The United States Fish and Wildlife Service, in approving the federal Incidental Take Permit and HCP on February 26, 1999, acknowledged that the necessary watershed analysis had not yet occurred but would be done within the next five years, and that various protective measures would be taken in the interim. CDF requested, and Pacific Lumber supplied, additional information regarding the SYP in February 1999. On February 25, 1999, DFG made the required CEQA findings and approved the final EIS/EIR. The CDF director (Director) also approved the SYP, specifically approving the long-term sustained yield estimate identified as alternative 25a, set at 196.1 mmbf per year with a projected conifer harvest level of 136.6 5 mmbf per year for the first decade. The Director stated that “the Department has determined that alternative 25 a is the only alternative with constraints and timber harvesting that are consistent with the interim mitigations required by the [HCP] and the EIS/EIR.” On February 27, 1999, Pacific Lumber wrote to the Director, disagreeing with some assumptions and advocating alternative 25. Under alternative 25, the long-term sustained yield was set at 190 mmbf per year, and the projected conifer harvest level in the first decade was 178.8 mmbf per year—similar to the projections found in appendix Q to the final EIS/EIR. Both alternatives 25 and 25a proposed to implement the mitigation measures found in the final HCP, but the former was based on assumptions about results of the required watershed analysis that were more optimistic than the latter, i.e., that future watershed analysis would result in fewer restrictions on logging than was assumed under alternative 25a. Officials of the United States Fish and Wildlife Service, the National Marine Fisheries Service, and the DFG also wrote to express their support for alternative 25. They argued *477 that the HCP provided for an “adaptive management” approach that would allow for greater flexibility as conditions in the field were evaluated, which would lead to a relaxation of some of the interim restrictions contained in the HCP, and that therefore the higher harvest level estimate was more likely to be accurate. On March 1, 1999, right at the deadline imposed by federal legislation for obtaining the necessary regulatory approvals, the Director of CDF approved the SYP, selecting alternative 25, allowing for the higher harvest levels. The approved harvest level was substantially lower than Pacific Lumber’s harvest levels for 1987-1997, after Maxxam Corporation had taken over the company, of 250 mmbf per year on average, but was substantially higher than the historic level of logging prior to that time of approximately 120 mmbf per year. On February 26, 1999, DFG executed a Streambed Alteration Agreement pursuant to Fish and Game Code former section 1603. On March 1, 1999, the DFG approved a state Incidental Take Permit authorizing the take of various species incidental to Pacific Lumber’s timber harvesting. On March 31, 1999, an administrative mandamus action was filed by the Environmental Protection Information Center and the Sierra Club (hereafter collectively EPIC). The lawsuit challenged (1) the approval of the SYP by CDF, (2) the issuance of the state Incidental Take Permit by DFG, (3) the approval of the Streambed Alteration Agreement by DFG, and (4) the findings issued by both state agencies under CEQA concerning the Headwaters Forest project. Simultaneously, the United Steelworkers of America (Steelworkers) also petitioned for administrative mandamus to challenge only the SYP on similar but not identical grounds. The trial court proceedings involved an extensive preliminary dispute over the contents of the administrative record. The court then held several days of evidentiary hearings on whether certain materials had been excluded from the administrative record—i.e., whether documents existed that should have been considered by the agencies. EPIC and the Steelworkers were granted leave to amend their complaints to allege a failure by the state agencies to provide an accurate administrative record. The trial court ruled on EPIC’s and the Steelworkers’s petitions in two separate statements of decision, issued June 22, 2003. The trial court ruled that petitioners had failed to sufficiently demonstrate that any of the challenged agency decisions were not based on substantial evidence, because they had not sufficiently identified the evidence claimed to be insufficient. In virtually all other respects, the trial court agreed with petitioners that the public agencies had not proceeded according to law. Specifically, the trial *478 court held that the SYP was deficient on a number of grounds, and that the state Incidental Take Permit, Streambed Alteration Agreement and CEQA findings were all inadequate and represented a failure to comply with the law on the part of CDF and DFG. The trial court then held a further hearing to decide whether Pacific Lumber’s timber operations should be enjoined. The court, weighing the balance of harms, concluded that although Pacific Lumber’s past and current timber operations had resulted in water quality degradation and reduction in fish population, enjoining all of Pacific Lumber’s timber operations would cause excessive hardship to the company, its employees, and the community. The court concluded that timber operations being conducted pursuant to THP’s approved prior to the court’s July 22, 2003 statement of decision would not be enjoined but that logging under any THP approved after that date that relied upon the now-vacated Sustained Yield Plan would be enjoined. Separate judgments were entered in the lawsuits filed by the environmental plaintiffs and by the Steelworkers, and the trial court issued a peremptory writ of mandate in each case. Pacific Lumber and both state agencies appealed from each judgment. The Court of Appeal consolidated the appeals. For reasons discussed at greater length below, the Court of Appeal reversed the trial court on every point that had been decided adverse to the state agencies and Pacific Lumber, upholding the validity of each of the regulatory approvals and reversing the granting of a peremptory writ of mandate. EPIC and the Steelworkers separately petitioned for review and we granted both petitions. The Steelworkers’s petition raises several issues related to the SYP. EPIC raises some of these same issues and, in addition, raises issues with respect to the Incidental Take Permit, the Streambed Alteration Agreement, and the adequacy of CEQA findings. The issues raised by the parties before this court, while numerous, are somewhat fewer than were raised below. III. Discussion A. Standard of Review The standard of review, and the related question of what constitutes prejudicial error, will be discussed in more detail below. For now, we state these general principles. First, the standard for review of agency decisions in connection with regulatory approvals is generally one of abuse of discretion. “ ‘Abuse of discretion is established if the respondent [agency] has not proceeded in the manner required by law, the order or decision is not supported by the findings, or the findings are not supported by the evidence.’ [Citations.]” (Sierra Club v. State Bd. of Forestry (1994) 7 Cal.4th 1215, 1236 [32 Cal.Rptr.2d 19, 876 P.2d 505] (Sierra Club).) *479 In this case, we are reviewing for errors of law, and will not engage in substantial evidence review. As the Court of Appeal correctly stated: “In the present case, the trial court rejected the allegations in [EPIC’s] writ petition that the administrative findings were unsupported by the evidence. The trial court found that [EPIC] failed to present a summary of the material evidence or any argument on the sufficiency of the evidence. In essence, the trial court found that [EPIC] waived or abandoned [its] challenges to the factual bases for the administrative decisions. [EPIC has] not cross-appealed, nor [does it] dispute that the focus of our review is whether the state agencies committed legal, not factual, error. Hence, for purposes of our review, we will accept that the administrative findings were supported by the evidence and we will confine our review to determining whether the state agencies failed to proceed in a manner required by law.” In determining whether the agency complied with the required procedures and whether the agency’s findings are supported by substantial evidence, the trial court and the appellate courts essentially perform identical roles. We review the record de novo and are not bound by the trial court’s conclusions. (Bixby v. Pierno (1971) 4 Cal.3d 130, 149, fn. 22 [93 Cal.Rptr. 234, 481 P.2d 242]; Sierra Club v. California Coastal Com. (1993) 19 Cal.App.4th 547, 557 [23 Cal.Rptr.2d 534].) B. Challenges to the Sustained Yield Plan 1. Standing of the Steelworkers As a threshold matter, Pacific Lumber contends that the Steelworkers have no standing to bring this writ of mandate action to challenge the SYP. We disagree. Generally speaking, in order to have standing to sue, “a party must be ‘beneficially interested’ (Code Civ. Proc., § 1086), i.e., have ‘some special interest to be served or some particular right to be preserved or protected over and above the interest held in common with the public at large.’ ” (Associated Builders & Contractors, Inc. v. San Francisco Airports Com. (1999) 21 Cal.4th 352, 361-362 [87 Cal.Rptr.2d 654, 981 P.2d 499].) There is nonetheless a well-established exception to the beneficial interest rule for citizen suits. “ ‘ “ ‘[W]here the question is one of public right and the object of the mandamus is to procure the enforcement of a public duty, the relator need not show that he has any legal or special interest in the result, since it is sufficient that he is interested as a citizen in having the laws executed and the duty in question enforced (Common Cause v. Board of Supervisors (1989) 49 Cal.3d 432, 439 [261 Cal.Rptr. 574, 777 P.2d 610].) The trial court found that in this case, which involves the proper enforcement of administrative regulations governing a plan for logging over 200,000 *480 acres of timberland highly valued both for environmental and economic reasons, a public right and a public duty were at stake. Pacific Lumber did not contest that finding on appeal. Pacific Lumber argues rather that an exception to the rule of citizen standing should be recognized for labor unions like the Steelworkers, along the lines of the exception recognized for corporations in Waste Management of Alameda County, Inc. v. County of Alameda (2000) 79 Cal.App.4th 1223 [94 Cal.Rptr.2d 740] (Waste Management). In that case, in considering whether a corporation had standing to bring a CEQA action under the citizen suit doctrine, the court reasoned that “where a corporation attempts to maintain a citizen suit, it is appropriate to require the corporation to demonstrate it should be accorded the attributes of a citizen litigant, since it generally is to be expected that a corporation will act out of a concern for what is expedient for the attainment of corporate purposes . . . .” (79 Cal.App.4th at p. 1238.) In giving effect to this principle, the court articulated a number of factors that may be considered, including “whether the corporation has demonstrated a continuing interest in or commitment to the subject matter of the public right being asserted [citations]; whether the entity is comprised of or represents individuals who would be beneficially interested in the action [citations]; whether individual persons who are beneficially interested in the action would find it difficult or impossible to seek vindication of their own rights [citations]; and whether prosecution of the action as a citizen’s suit by a corporation would conflict with other competing legislative policies [citation].” (Ibid.) We need not decide whether the corporate exception to citizen suits articulated by the Waste Management court is a correct statement of the law, nor whether and to what extent that exception applies to labor unions. In this case, the trial court found that the Steelworkers qualified as a citizen litigant under Waste Management. The court concluded that the Steelworkers had shown a continuing interest in and commitment to issues related to this case, including that of sustainable economic development and environmental quality and specifically issues regarding timber harvesting. The court also found that the union had over 12,000 members in California who had sufficient interest in the proper enforcement of timber harvest laws, that interested individuals would have trouble participating in the litigation due to its size and complexity, and that the Steelworkers’ participation presented no conflict with competing legislative policies. Pacific Lumber did not contest those findings on appeal and does not discuss the findings before this court. It does quote a statement in the record that the Steelworkers’ participation was motivated by a labor dispute with Pacific Lumber’s parent company, Maxxam Incorporated. But the record also *481 contains ample evidence the Steelworkers have long-standing involvement in environmental and economic sustainability issues. We will review the trial court’s factual determinations that bear upon the issue of standing under a substantial evidence standard. (Daro v. Superior Court (2007) 151 Cal.App.4th 1079, 1092 [61 Cal.Rptr.3d 716].) We conclude that substantial evidence supports the trial court’s conclusion that the Steelworkers have standing in this case. 2. What Are Sustained Yield Plans? A proper understanding of the nature and purpose of Sustained Yield Plans for timber harvesting begins by placing them in the context of the Forest Practice Act (Pub. Resources Code, § 4511 et seq.). “The Act’s provisions, together with implementing rules and regulations promulgated by the State Board of Forestry (board) ([Pub. Resources Code,] §§ 4521.3, 4551), provide a comprehensive scheme regulating timber operations in a way which promotes the legislative ‘goal of [achieving] maximum sustained production of high-quality timber products . . . while giving consideration to values relating to recreation, watershed, wildlife, range and forage, fisheries, regional economic vitality, employment, and aesthetic enjoyment’ ([Pub. Resources Code,] §§4513, subd. (b), 4512, subd. (c)). The heart of the scheme is its requirement that logging be carried out only in conformance with a timber harvesting plan (THP or plan) submitted by the timber owner or operator and approved by the department after determining, with an opportunity for input from state and county agencies and the general public, that the proposed operations conform to the Act and rules and regulations. ([Pub. Resources Code,] §§ 4581-4582.75, 4583; [citations].) [f] Since 1976, the THP preparation and approval process developed under the Act has been certified as the functional equivalent to, and hence an adequate substitute for, the full environmental impact report (EIR) process required by CEQA. [Citations.]” (T.R.E.E.S. v. Department of Forestry & Fire Protection (1991) 233 Cal.App.3d 1175, 1180 [285 Cal.Rptr. 26].) As part of fulfilling the Forest Practice Act’s goals, the Legislature has authorized the Board of Forestry and Fire Protection to create rules and regulations for the development of Sustained Yield Plans. (Pub. Resources Code, § 4551.3, subd. (a).) The SYP is intended to serve as a kind of master plan for timber harvesting a large geographic area. The board’s regulations, adopted as article 6.75 of title 14 of the California Code of Regulations, declares: “This Article carries out the Legislature’s direction that the Board adopt regulations to assure the continuous growing and harvesting of commercial forest tree species and to protect the soil, air, fish and wildlife, and water resources in accordance with the policies of the . . . Act .... Those policies include creating and maintaining a system of timberland regulations *482 and use which ensures that timberland productivity is maintained, enhanced and restored where feasible and the goal of maximum sustained production of high-quality timber products ... is achieved while giving consideration to environmental and economic values. The Sustained Yield Plan (SYP) may be submitted at the option of the landowner and is intended to supplement the THP process by providing a means for addressing long-term issues of sustained timber production, and cumulative effects analysis which includes issues of fish and wildlife and watershed impacts on a large landscape basis.” (FP Rules, § 1091.1, subd. (b).) Under the Forest Practice Rules, an SYP “shall not replace a THP. However, to the extent that sustained timber production, watershed impacts and fish and wildlife issues are addressed in the approved SYP, these issues shall be considered to be addressed in the THP; that is the THP may rely upon the SYP.” (FP Rules, § 1091.2, italics added.) Forest Practice Rules section 1091.45, subdivision (a) further elaborates on the SYP requirements: “Consistent with the protection of soil, water, air, fish and wildlife resources a SYP shall clearly demonstrate how the submitter will achieve maximum sustained production of high quality timber products while giving consideration to regional economic vitality and employment at planned harvest levels during the planning horizon. The average annual projected harvest over any rolling 10-year period, or over appropriately longer time periods for ownerships which project harvesting at intervals less frequently than once every 10 years, shall not exceed the long-term sustained yield estimate for a SYP submitter’s ownership.” Forest Practice Rules section 1091.3 defines “Planning Horizon” as the “100 year period over which sustained timber production, watershed, and fish and wildlife effects shall be evaluated.” The Forest Practice Rules also require “[a]n estimate of the long-term sustained yield.” (FP Rules, § 1091.45, subd. (c)(2).) Thus, the SYP is a kind of master plan for timber harvesting over a long time period that supplements but does not replace the THP process, and individual THP’s may rely on the SYP to the extent it analyzes the pertinent issues. 6 3. Omitted Public Comments The Steelworkers contend that certain comments submitted by the public regarding the draft SYP were not taken into account by CDF, which amounts to prejudicial error. *483 It is first undisputed that none of the comments in question were placed in the administrative record. CDF certified the administrative record. A certified record in an action challenging the sufficiency of an EIS/EIR under CEQA is supposed to include all public comments and supporting documentation. (Pub. Resources Code, § 21167.6, subd. (e)(6)-(8), (10)—(11).) Moreover, as the Court of Appeal stated: “The record does suggest that the missing documents were not taken into account. The trial court explained that the order for preparation of the administrative record required the Department of Forestry to prepare a record of all documents that were before the agency and taken into account—not just the documents from the agency’s file compiled post hoc. Trial counsel for the Department (the Attorney General) conceded at trial that what was not in the certified administrative record was not taken into account. The question, then, is whether the failure of the Department of Forestry to consider the missing documents rendered the Sustained Yield Plan invalid.” The trial court found that three types of public comments were not considered by CDF. First, there were documents submitted prior to the November 16, 1998 deadline for receiving public comments on the SYP. Second, there were written documents submitted by members of the public at public hearings. Third, there were a number of letters and public comments submitted after November 16, 1998, which, for reasons discussed below, the Steelworkers contend and the trial court found were timely submitted. Each of these categories will be discussed in turn. As to the first category of comments, CDF characterizes them 7 as “cover memos written on behalf of [EPIC], which transmitted reference materials such as scientific articles cited by other members of the public in their comment letters.” CDF asserts that these materials “contain no substantive comments.” An examination of the record reveals that the exhibits in question consist of scholarly articles about various subjects generally related to the kind of subjects addressed in an SYP; for example, an article entitled “Forest Vegetation Removal and Slope Stability in the Idaho Batholith.” One of the omitted exhibits in this category, submitted by Cynthia Elkins, contains documents pertaining to Pacific Lumber’s previous THP’s. As the Court of Appeal observed: “The articles themselves are not comments on the Sustained Yield Plan but are reference materials that were cited in comment letters that had been previously submitted. Those comment letters are in the certified administrative record and were responded to in the final EIS/EIR.” *484 We agree with the implicit distinction drawn by the Court of Appeal. Although CDF has a duty to consider comments by members of the public under the Forest Practice Rules, that duty does not necessarily extend to considering all of the non-project-specific secondary materials submitted in support of the comments. Whether and to what extent CDF reviews such material cited in the comments is a matter to be left to its sound discretion and professional judgment. This deferential standard does not change when scholarly articles are not only cited in the comments but reproduced and submitted along with the comments. There is no indication CDF did not consider the comments themselves. The second category of excluded documents is written comments apparently submitted at public hearings, in conjunction with oral comments, some opposing and some supporting the SYP. CDF contends that this material was not included in the administrative record because it was duplicative. The third category of documents is comments submitted after CDF’s comment period closed on November 16, 1998, up to February 22, 1999, comments mainly critical of the SYP. The trial court concluded that the public comment period had been extended and that whether or not it had been extended, CDF should have considered these documents. The Court of Appeal did not dispute this factual conclusion, but held that the failure to consider these documents was nonprejudicial. The record discloses that CDF announced that the public comment period would end on November 16, 1998, “unless the public review period is extended by mutual consent of the SYP submitter and the [CDF].” The United States Fish and Wildlife Service posted a notice in the Federal Register on January 22, 1999, announcing that public comments would be received on the SYP/HCP and the EIS/EIR until February 22, 1999. The notice included the address of the United States Fish and Wildlife Service persons who would be receiving the comments, and also stated that “comments on the SYP may be mailed to John Munn” of CDF. The notice further explained that during the initial comment period, CDF and other government agencies had received approximately 18,000 comments on the SYP/HCP and draft EIS/EIR and that numerous changes had been made in response to those comments and to the enactment of Assembly Bill 1986. The new public comment period was intended to address these changes. We therefore agree with the trial court and Court of Appeal that the Federal Register notice effectively reopened the public comment period for the SYP until February 22, 1999. The question, then, is whether the error in failing to consider the second and third category of comments is prejudicial. In order to address this *485 question, we first consider what constitutes prejudicial error in cases involving environmental review. As previously noted, “Only if the manner in which an agency failed to follow the law is shown to be prejudicial, or is presumptively prejudicial, as when the department or the board fails to comply with mandatory procedures, must the decision be set aside . . . .” (Sierra Club, supra, 7 Cal.4th at p. 1236.) In Sierra Club, we found prejudicial abuse of discretion when the Board of Forestry and Fire Protection approved a THP notwithstanding the fact that real party in interest Pacific Lumber had failed to provide information requested by CDF and DFG. “The failure of the board to proceed as required by law was prejudicial. The absence of any information regarding the presence of the four old-growth-dependent species on the site frustrated the purpose of the public comment provisions of the Forest Practice Act. ([Pub. Resources Code,] §§ 4582.6, 4582.7.) It also made any meaningful assessment of the potentially significant environmental impacts of timber harvesting and the development of site-specific mitigation measures impossible. In these circumstances prejudice is presumed.” (Sierra Club, supra, 7 Cal.4th at pp. 1236-1237.) In coming to this conclusion, we cited with approval Rural Landowners Assn. v. City Council (1983) 143 Cal.App.3d 1013 [192 Cal.Rptr. 325] (Rural Landowners Assn.). (Sierra Club, supra, 7 Cal.4th at p. 1237.) That case considered the approval of an FIR for the annexation and development of certain agricultural land by the Lodi City Council, when the draft FIR had not been timely submitted to the Governor’s Office of Planning and Research, as required by law. The city council had therefore failed to consider that agency’s substantive comments before approving the FIR. (Rural Landowners Assn., supra, at pp. 1017-1018.) The trial court found that because the state agency’s comments were incorporated into an addendum after the approval, and the city council had not changed its decision, failure to include the comments in the FIR was harmless error. (Id. at p. 1019.) The Court of Appeal in Rural Landowners Assn, disagreed with this line of reasoning: “Were we to accept respondent’s position that a clear abuse of discretion is only prejudicial where it can be shown the result would have been different in the absence of the error, we would allow ... a subversion of the purposes of CEQA. Agencies could avoid compliance with various provisions of the law and argue that compliance would not have changed their decision. Trial courts would be obliged to evaluate the omitted information and independently determine its value. . . . We conclude that where that failure to comply with the law results in a subversion of the purposes of CEQA by omitting information from the environmental review process, the error is prejudicial. The trial court may not exercise its independent judgment on the omitted material by determining whether the ultimate decision of the lead agency would have been affected had the law been followed. The decision is for the discretion of the agency, and not the courts.” (Rural *486 Landowners Assn., supra, 143 Cal.App.3d at pp. 1022-1023.) The remedy for this deficiency was for the trial court to have issued a writ of mandate compelling the city to prepare a supplemental FIR. (Id. at p. 1025.) The above rule emerges out of the difficulty courts have in assessing the effects of the omitted information, much of it generally highly technical, on the ultimate decision. A trial court’s “independent judgment that the information was of ‘no legal significance’ amounts to a ‘post hoc rationalization’ of a decision already made, a practice which the courts have roundly condemned.” (Rural Landowners Assn., supra, 143 Cal.App.3d at p. 1021.) On the other hand, errors in the CEQA or THP process which are insubstantial or de minimis are not prejudicial. (Environmental Protection Information Center, Inc. v. Johnson (1985) 170 Cal.App.3d 604, 623, fn. 11 [216 Cal.Rptr. 502],) 8 The Forest Practice Rules require the director to “review public input” at the close of a public comment period prior to approval of an SYP. (FP Rules, § 1091.10, subd. (e).) Public comments are therefore an integral part of the SYP approval process, as they are in the FIR approval process, and such comments, like the comments from state agencies at issue in Rural Landowners Assn., may contain information critical to that process. “Public review is essential to CEQA. The purpose of requiring public review is ‘ “ ‘to demonstrate to an apprehensive citizenry that the agency has, in fact, analyzed and considered the ecological implications of its action.’ ‘[P]ublic review and comment . . . ensures that appropriate alternatives and mitigation measures are considered, and permits input from agencies with expertise in timber resources and conservation.’ [Citation.] Thus public review provides the dual purpose of bolstering the public’s confidence in the agency’s decision and providing the agency with information from a variety of experts and sources.” (Schoen v. Department of Forestry & Fire Protection (1997) 58 Cal.App.4th 556, 573-574 [68 Cal.Rptr.2d 343].) *487 If it is established that a state agency’s failure to consider some public comments has frustrated the purpose of the public comment requirements of the environmental review process, then the error is prejudicial. (See Sierra Club, supra, 7 Cal.4th at pp. 1236-1237; Rural Landowners Assn., supra, 143 Cal.App.3d at pp. 1022-1023.) As the case law establishes, courts are generally not in a position to assess the importance of the omitted information to determine whether it would have altered the agency decision, nor may they accept the post hoc declarations of the agencies themselves. (Rural Landowners Assn., supra, 143 Cal.App.3d at p. 1021.) 9 On the other hand, an agency’s failure to consider public comments is not necessarily prejudicial. For example, when the material not considered was, on its face, demonstrably repetitive of material already considered, or so patently irrelevant that no reasonable person could suppose the failure to consider the material was prejudicial, or when the omitted material supports the agency action that was taken, then such omissions do not subvert the purpose of the public comment provisions and are nothing more than technical error. Short of these showings, which the agency that failed to consider the comments would have the burden to make, the omission of the information must be deemed prejudicial. 10 *488 With these principles in mind, we turn to the present case. The Court of Appeal stated that “[t]he Steelworkers do not dispute that the missing comments were duplicative, raising objections to the Sustained Yield Plan that were covered by over 16,000 written comments made by others during the public comment period and responded to in the final EIS/EIR.” The Steelworkers did not contest the accuracy of that statement in its rehearing petition to the Court of Appeal nor in its briefing before this court. 11 Rather, the Steelworkers argue only that under Rural Landowners Assn., the question whether the comments are duplicative is irrelevant, because a court “may not exercise its independent judgment on the omitted material by determining whether the ultimate decision of the lead agency would have been affected had the law been followed.” (Rural Landowners Assn., supra, 143 Cal.App.3d at p. 1023.) But a determination of whether omitted information would have affected an agency’s decision is significantly different from a determination of whether the omitted material is duplicative of information already considered. The former determination is highly speculative, an inquiry that takes the court beyond the realm of its competence. The latter determination—whether omitted evidence is duplicative or cumulative—is an inquiry courts commonly make. (See, e.g., People v. Keehley (1987) 193 Cal.App.3d 1381, 1386-1387 [239 Cal.Rptr. 5].) To be sure, the question whether public comments were duplicative, particularly when these comments involve, as they do here, highly technical material, may not be obvious to a reviewing court. As stated above, when an SYP or FIR is challenged for failing to consider comments alleged to contain significant new information, it is the burden of the agency that erroneously omitted the comments to establish they are merely duplicative. When, however, their duplicative nature essentially is not contested, as in the present case, no further inquiry is necessary. We conclude CDF’s failure to consider these comments was not prejudicial. 4. Consideration in Sustained Yield Plan of Long-term Regional Economic Vitality and Employment The Steelworkers contend that the SYP failed to consider issues of regional economic vitality and employment over a 100-year period, as required in the Forest Practice Rules. As Forest Practice Rules section 1091.45, subdivision (a) states: “Consistent with the protection of soil, water, air, fish *489 and wildlife resources a SYP shall clearly demonstrate how the submitter will achieve maximum sustained production of high quality timber products while giving consideration to regional economic vitality and employment at planned harvest levels during the planning horizon.” (Italics added.) As noted ante, Forest Practice Rules section 1091.3 defines “Planning Horizon” to mean “the 100 year period over which sustained timber production, watershed, and fish and wildlife effects shall be evaluated”—although in the present case a 120-year planning horizon was chosen. Thus, the Forest Practice Rules require “consideration” of regional employment and economic vitality over a 100-year period in the SYP’s demonstration of how Pacific Lumber will achieve “maximum sustained production of high quality timber products.” This is consistent with the primary objective of the SYP regulations: to address “long-term issues of sustained timber production,” such that the “goal of maximum sustained production of high-quality timber products ... is achieved while giving consideration to environmental and economic values.” (FP Rules, § 1091.1, subd. (b).) The Steelworkers point to the statement in the SYP that the first decade of the 120-year planning period “is the only period appropriate for [analysis of] economic and social effects. Too many variables, including economic diversity of the local economy, strain of the local timber industry, and timber-related tax revenue, would not be constant over a longer-term analysis period. Thus, a discussion of social and/or economic effects beyond 2012 would be very uncertain, if not speculative, and would not be appropriate in either an FIS or FIR.” It contends that this limitation of economic analysis to the first decade contravenes the injunction of section 1091.45 of the Forest Practice Rules that the SYP consider economic and employment effects for the entire planning horizon. In rejecting the Steelworkers’ claim, the Court of Appeal relied on a brief portion of the Public Review Draft of the SYP/HCP that in fact assessed the employment impacts of logging over a 120-year period. In that section, Pacific Lumber estimated jobs per decade in relation to millions of board feet of timber per year, using a multiplier of six jobs per year for every million board feet harvested. On this basis, Pacific Lumber projected a decline in employment as timber harvesting tapered off, going from a high of 1,401 jobs in the first decade to a low of 844 jobs in the fifth decade and then steadily rising thereafter as newer growth timber matured and was harvested. As the Steelworkers point out, however, although there is some confusion about the contents of the final SYP (as discussed below), the above draft section was superseded and was not incorporated in the final SYP. CDF does not dispute that the employment portion of the Public Review Draft of the SYP was superseded. Rather, it states that “the evolution of the discussion of *490 jobs and economic vitality from draft to final indicates that the issue is analyzed over twelve decades but ultimately CDF found any discussion beyond ten years to be speculative. This represents evidence of the consideration of the issue, not a failure to consider.” Moreover, the final SYP does contain projections of harvest levels for each decade for a 120-year period that would be the basis for further economic analysis of the effects of timber harvesting. The SYP projects a conifer harvest level of 178.8 mmbf per year for the first decade, declining in each subsequent decade to a low of 113.8 mmbf per year on average for the fifth decade, and then gradually increasing to 166.2 mmbf per year on average for the final decade. The projections also specify the kind of timber to be harvested, with old growth timber making up a large portion of the harvest in the first decade and giving way increasingly to younger growth timber in subsequent decades. It is unclear from the Forest Practice Rules how much detail is required in “giving consideration” to economic issues over the planning horizon. The rules do state that in an SYP, “the accuracy of, and therefore the need for, detailed future projections becomes less as the time horizon lengthens” and that “[i]t is not the intent of this Article that speculation shall be promoted such that analyses shall be undertaken which would produce only marginally reliable results or that unneeded data would be gathered. ... It is the intent of this Article that the requirements for informational or analytical support for a SYP shall be guided by the principles of practicality and reasonableness; no information or analysis shall be required which in the light of all applicable factors is not feasible. However, it is the intent of this Article that all potential adverse environmental impacts resulting from proposed harvesting be described, discussed and analyzed before such operations are allowed. Should such analysis not be included in the SYP, it must be contained in those THPs which rely on the SYP, including any impact discovered after the SYP is approved.” (FP Rules, § 1091.1, subd. (b).) As a general matter, courts will be deferential to government agency interpretations of their own regulations, particularly when the interpretation involves matters within the agency’s expertise and does not plainly conflict with a statutory mandate. (See Yamaha Corp. of America v. State Bd. of Equalization (1998) 19 Cal.4th 1, 12-13 [78 Cal.Rptr.2d 1, 960 P.2d 1031].) In the present case, the question of how much economic and employment analysis over how long a period of time is feasible, and at what point it becomes speculative, is a judgment call, and we will not disturb the agency’s determination without a demonstration that it is clearly unreasonable. Here, the SYP contains information regarding (1) projected harvest levels for 12 decades; (2) a credible estimate of the employment effects of such harvesting *491 and projection of timber-related employment over 12 decades in the draft SYP; (3) a detailed analysis of economic and employment impacts of timber harvesting in the first decade; and (4) a reasoned decision to omit detailed analysis of the effects of timber harvesting on employment and the economy over the subsequent decades of the planning horizon. Under these circumstances, we conclude that CDF did not abuse its discretion in determining that the SYP had adequately followed the Forest Practice Rules by “giving consideration” to the economic and employment consequences of timber harvesting during the period of the planning horizon (EP Rules, § 1091.45), while at the same time not engaging in overly speculative analysis (FP Rules, § 1091.1). 12 5. Is There a Valid Sustained Yield Plan Document? Petitioners 13 contend there was no single, agreed-upon SYP that has been approved, and that the CDF director’s approval of the SYP must therefore be invalid. We agree. As discussed, various federal and state agencies approved several interrelated documents: an EIS/EIR, an HCP and an SYP. The final EIS/EIR was circulated in January 1999, and contained an appendix Q, which purported to identify the final SYP. Appendix Q states: “This [f]inal EIS/EIR constitutes the final HCP/SYP. To reduce the volume of paper associated with finalizing the six-volume proposed HCP/SYP, it is incorporated here by reference. To ensure that all requirements of the SYP are met, and that key components can be located easily, the following crosswalk is provided. It indicates the primary location where information may be found; it is not all-inclusive, and relevant information may be found in other sections. Except as noted, volume and part references refer to [Pacific Lumber’s] July 1998 draft SYP/HCP.” The crosswalk then references the topics that are required to be addressed in the SYP together with the volume and section in which the topic is addressed in the draft SYP/HCP. For example, “[sustained timber production assessment” is found in “Volume I, Part C, . . . and Part E,” and “Fish and wildlife assessment” is found in Volumes I, II, and IV. *492 Pacific Lumber argues that appendix Q and the documents to which it refers constitute the final SYP. As petitioners point out, however, there are several problems with relying on appendix Q to definitively set forth the contents of the final SYP. First, by its own terms, it “is not all-inclusive, and relevant information may be found in other sections.” Second, appendix Q purports to incorporate the six-volume draft SYP circulated for public review. Yet in a document that was prepared for the trial court below, CDF made clear that substantial portions of the Public Review Draft SYP had been superseded, noting in the margins of the table of contents of the Public Review Draft SYP those portions that had been replaced by the final EIS/EIR. Thus, the appendix Q crosswalk, in referencing a Public Review Draft SYP that had been substantially superseded, failed to give an accurate picture of the document’s contents at the time the SYP was approved by the CDF director on March 1, 1999. Third, appendix Q is included in a January 1999 document. Additional information was provided by Pacific Lumber in February 1999 that the CDF Director relied on for his March 1, 1999 approval of the SYP. On February 16, 1999, Pacific Lumber presented CDF with a lengthy document entitled “Updated Sustained Yield Planning Information,” with extensive supplemental information pertaining to the long-term sustained yield estimate. Pacific Lumber on February 23, 1999, provided further extensive information on alternative 25A, which contemplated a conifer harvest of approximately 136.6 mmbf per year for the first decade in response to a CDF request. Along with providing that information, Pacific Lumber made clear it believed that this alternative was infeasible inasmuch as it contemplated a lower harvest than Pacific Lumber found economically viable. On February 25, 1999, the Director approved the SYP with alternative 25A. On February 28, 1999, Pacific Lumber again supplied extensive additional information, this time targeted to alternative 25, which contemplated the higher conifer harvest of 178.8 mmbf per year for the first decade. The Director eventually chose alternative 25. None of the voluminous supplemental information on which the Director partly based his decision is included in appendix Q. CDF, in contrast to Pacific Lumber, does not contend that appendix Q represents the definitive SYP. Rather, it claims that the CDF Director’s March 1, 1999 and February 25, 1999 letters approving the SYP contain “a description of the location of the substantive information which comprises the various components of the Sustained Yield Plan required by the Forest Practice Rules.” As the CDF Director stated in the March 1, 1999 letter, his determination that the SYP was in conformance with Forest Practice Rules was “[b]ased upon analysis of the revised draft of [the SYP] submitted by [Pacific Lumber] in July of 1998 in combination with provisions of the HCP, EIS/EIR, supplemental information received from [Pacific Lumber] on February 16, 1999, responses from Pacific Lumber to watershed questions received *493 on February 23, 1999, and with additional information provided by the National Marine Fisheries Service, the U.S. Fish and Wildlife Service, and the California Department of Fish and Game . . . .” Yet the Director’s terse statement in this approval letter cannot be regarded as setting forth a definitive SYP. First, the letter refers to the Public Review Draft SYP, a substantial portion of which, as discussed above, had been superseded. The Director’s approval does not specify which portions of the draft SYP are to be included in the final SYP, which parts of the final EIS/EIR are to be included, or how the draft SYP dovetails with the February 16, 1999 and February 23, 1999 documents to which the Director’s approval also refers. Second, the document refers nonspecifically to “additional information provided by the National Marine Fisheries Service, the U.S. Fish and Wildlife Service, and the California Department of Fish and Game . . . .” That the contents of the draft SYP were unsettled at the time of its approval is further evidenced by a communication on March 15, 1999, two weeks after the SYP was approved. CDF project manager John Munn requested “within a relatively short time frame” “supplemental SYP materials,” in order “to meet the requirements of the SYP,” including “[a] consolidated version of the material submitted by [Pacific Lumber] in support of Alternative 25 and related responses to CDF questions, Appendix Q in the EIS/EIR, and information from the July 1998 public review draft of the SYP/HCP that is still applicable to the approved SYP and Habitat Conservation Plan.” (Italics added.) There is nothing in the record indicating that Pacific Lumber ever complied with this request. It is noteworthy, then, that even Pacific Lumber and CDF do not appear to agree on what constitutes the final SYP—the former would find it in appendix Q, the latter in the February 25 and March 1, 1999 letters of approval. As explained at greater length below, the SYP is intended to be relied on by Pacific Lumber and CDF and other government agencies in determining whether Pacific Lumber’s logging activities, as described in its timber harvest plans, are lawful. As also discussed below, Public Resources Code section 4551.3 contemplates a role for the public in monitoring compliance with an SYP after it has been approved. As we recently reaffirmed in the analogous case of an EIR: “The data in an EIR must not only be sufficient in quantity, it must be presented in a manner calculated to adequately inform the public and decision makers, who may not be previously familiar with the details of the project, ‘[information “scattered here and there in EIR appendices,” or a report “buried in an appendix,” is not a substitute for “a good faith reasoned analysis.” ’ ” (Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova (2007) 40 Cal.4th 412, 442 [53 Cal.Rptr.3d 821, 150 P.3d 709].) Similarly, basic confusion about the *494 contents of an unconsolidated SYP scattered over a voluminous administrative record does not allow the public and decision makers to readily know those contents and use the SYP for the purposes for which it was intended. And the fact that the information and analysis contained in the various environmental documents Pacific Lumber submitted is so extensive makes the need for an easily identifiable document all the greater. Moreover, the Steelworkers convincingly argue that the Director improperly delegated to Pacific Lumber the task of determining the final contents of the SYP. As noted, the Director charged Pacific Lumber, in the March 1, 1999 letter approving the SYP, with preparing “an updated report based on alternative 25 that contains the SYP information contained in Appendix Q to the EIS/EIR and incorporates information from the July, 1998 public review draft of the SYP/HCP,” and would include unspecified information from various documents provided by Pacific Lumber and by certain government agencies in February 1999. In effect, the Director was giving Pacific Lumber the task of revising appendix Q, in light of the new information about alternative 25. This revision was to incorporate unspecified sections of the Public Review Draft SYP, and which sections were to be incorporated was to be apparently left, at least initially, to Pacific Lumber’s judgment. John Munn’s March 15, 1999 postapproval letter discussed above also refers to Pacific Lumber’s assembling portions of documents “still applicable” in a final SYP. As the Steelworkers state: “Whether or not an agency may delegate to a private party the duty of consolidating various identified documents into a final plan . . . , there should be no question that an agency cannot delegate to a private party the responsibility of determining what it is that the agency approved. This is a core agency function.” We agree. CDF and Pacific Lumber argue that any confusion about what constitutes a final SYP can be rectified in administrative proceedings pursuant to Public Resources Code section 4551.3. This was the position taken by the Court of Appeal, which reasoned that “[a]n integrated document was not a condition precedent to approval of the Sustained Yield Plan; it was a condition subsequent,” and if that condition was not met, petitioners could avail themselves of the remedies set forth in Public Resources Code section 4551.3, which provides for “ ‘continuing monitoring’ of an approved sustained yield plan by the Department of Forestry, including a hearing whenever an interested party comes forth with evidence of potential noncompliance with the terms and conditions of the approval of a sustained yield plan.” The Court of Appeal therefore concluded that “the assertion by [petitioners] to the trial court in the administrative mandamus proceedings that [Pacific Lumber] failed to provide the integrated document was misdirected and premature. When an administrative remedy is provided by statute, relief must be sought from the administrative body and exhausted before the courts will act. [Citations.] The remedy available to [petitioners] was to *495 request a hearing by the Department of Forestry pursuant to section 4551.3 of the Public Resources Code. Having failed to exhaust their administrative remedies, the environmental plaintiffs and the Steelworkers were not entitled to assert that [Pacific Lumber] failed to comply with the condition for approval of the Sustained Yield Plan.” Public Resources Code section 4551.3, to which the Court of Appeal opinion refers, states in pertinent part: “(b) As part of the continuing monitoring process for an approved sustained yield plan . . . , the department shall hold a public hearing on the plan if requested by an interested party who submits, in writing, a request based on substantial evidence of potential noncompliance with any of the following: [][] (1) The terms and conditions of the original sustained yield plan approval, [ft] (2) The applicable provisions of the rules or regulations adopted by the board that were in effect on the date the sustained yield plan was originally approved, [ft] (3) Other requirements that have been imposed on the sustained yield plan by operation of law. [ft] (c) The request shall identify specific issues in the plan to be addressed at the public hearing. To be considered, a request shall be made to the department within six months after the midpoint of the effective term of a sustained yield plan described in subdivision (a). The department shall hold the public hearing within 120 days after the date of the close of the six-month request period. A sustained yield plan shall be effective for the remainder of its term unless the director makes written findings, based on a preponderance of evidence, that implementation of the sustained yield plan is not in compliance with any material provision of paragraph (1), (2), or (3) of subdivision (b).” (Italics added.) It is difficult to fathom how the procedures and remedies set forth in Public Resources Code section 4551.3 address petitioners’ objections to the SYP. That statute, which contemplates continued monitoring of the manner in which the SYP is implemented, presupposes an SYP in its final form that can be monitored, i.e., a clear, written plan that can be compared to the plan as executed. If there is uncertainty about what constitutes the final SYP, then it is difficult to see how Public Resources Code section 4551.3’s monitoring provisions can address this shortcoming. This point is underscored by the provisions in subdivision (c) that the shortcomings be raised in a hearing approximately midway through the term of the SYP’s operation. Subdivision (a) of the statute provides that an SYP may be effective for a period of up to 10 years, and that is the effective period for the SYP in the present case. Section 4551.3 was plainly not meant to be used to cure inadequacies in an SYP present at the time the document was approved, but rather to remedy deficiencies in implementing the document that have become clear over time. CDF and Pacific Lumber also argue the shortcoming identified by petitioners amounts to merely a formatting problem, that there is nothing in the *496 Forest Practice Rules that require an SYP to be a consolidated document, and that, in any case, there is a lack of prejudice from not producing such a consolidated document. In support of this argument, CDF cites an e-mail from project manager John Munn on January 13, 1999, “indicating that the usability of the final document was ‘primarily a matter of formatting.’ ” But that quotation, placed in its proper context, does not support CDF’s argument. As Munn wrote: “There is still some question about what constitutes the final document. The HCP included as Appendix P to the EIS/EIR appears to be self-contained. The SYP discussion contained in Appendix Q, however, relies heavily on reference to the draft SYP/HCP. Does this mean that the final package consists of the new EIS/EIR, Responses to Comments, the draft SYP/HCP and an additional addendum (or appendix?) that includes the updated SYP information? This is primarily a question of formatting. Would it be possible to give conditional approval based on the company preparing a consolidated document containing SYP information? If not, I assume that this could be accomplished by preparing a working document following approval. Somehow, we have to end up with a usable document.” Thus, the above quotation indicates that the CDF soil erosion studies project manager made clear that in its then-current state, the document was not usable. There is no indication that the shortcomings identified in Munn’s e-mail were ever corrected. On the contrary, the subsequent information that CDF received and considered in approving the SYP in February 1999, made the identification of a single, usable document even more problematic. Munn’s postapproval letter of March 15, 1999, quoted above, continues the same theme, asking Pacific Lumber to expeditiously update and supplement various documents into a complete SYP. There can be no question that approval of a final document that is usable by the government agencies and by the public in monitoring the SYP is required. Indeed, the fact that the Forest Practice Rules contemplate an SYP that can be “filed” (FP Rules, § 1091.10) strongly supports the idea that there must be a specific document that CDF, and the public, can turn to for the purposes served by the SYP. As explained, CDF and Pacific Lumber have yet to identify or agree upon a definitive SYP. For these reasons, we reject CDF and Pacific Lumber’s argument that the lack of an identifiable SYP was not prejudicial. Although minor ambiguities in what constitutes a final SYP may be harmless, here the ambiguity as to the SYP’s contents was sufficiently substantial that CDF staff did not consider the document to be readily usable. Moreover, the fact that the Director of CDF improperly delegated to Pacific Lumber the task of finalizing the contents of the SYP after it was approved, abdicating the agency’s basic function of making that determination itself, appears to be the kind of error that is not amenable to harmless error analysis. And even if it were proper for CDF to promulgate a condition subsequent to approval of the SYP that it be finalized *497 by the consolidation of various unspecified documents, there is no indication that this condition was ever met. We conclude that in failing to approve an identifiable final SYP, CDF failed to proceed according to law, and that such error was prejudicial. The parties have not briefed the remedy for this deficiency. That question, and the related question of the procedures appropriate for resubmitting an adequate, identifiable SYP for approval, should be addressed on remand. 6. The Sufficiency of the Sustained Yield Plan for Public Review Petitioners contend that CDF failed to obtain sufficient information to authorize the SYP for public review because the Public Review Draft SYP/HCP failed to analyze individual planning watersheds and the cumulative impacts of the proposed logging on those watersheds. The process of approving an SYP is described in section 1091.10 of the Forest Practice Rules: First, within 20 days after receipt of the SYP, the CDF director reviews the document to ensure that it “is in proper order, and meets the informational requirements of the rules, and if so, the SYP shall be filed.” (Id., subd. (a).) Otherwise the Director is to return the document with noted deficiencies. Once filed, the Director has a 45-day or longer period to review the SYP to determine if it “contains sufficient and complete information to permit further review by the public and other agencies.” (Ibid.) After a 90-day or longer period of public review, the Director has a 30-day period to review and respond to public input and determine whether the SYP should be approved. If not, the reasons must be in writing. The Forest Practice Rules also require that an SYP contain analysis of the impacts of proposed logging on individual planning watersheds. Section 1091.6 states in part: “The following watershed issues shall be addressed in a SYP: H] (a) Assessment Area. The minimum assessment area shall be no less than a planning watershed. The assessment area may include multiple watersheds within a Management Unit, and areas outside the ownership may be included. [j[] (b) Impacts Analysis and Mitigation. The Assessment shall include an analysis of potentially significant adverse impacts, including cumulative impacts, of the planned operations and other projects, on water quality, fisheries and aquatic wildlife, [f] (c) The SYP shall contain a description of the individual planning watersheds in sufficient detail to allow a review of the analysis of impacts.” 14 *498 Petitioners contend that the SYP failed to provide sufficient information in their watershed analysis for public review. In conducting watershed analysis, Pacific Lumber used five watershed assessment areas (sometimes WAA’s) ranging in size between 55,000 and 426,000 acres, each of which consisted of a number of planning watersheds 15 —from approximately seven for the smallest WAA to approximately 45 for the largest. The record reflects that CDF staff found Pacific Lumber’s treatment of watershed analysis and the cumulative impacts of logging on individual watersheds to be inadequate throughout the SYP preparation process. Ross Johnson, CDF’s Chief of Forest Practices, requested in an April 25, 1997 letter that Pacific Lumber explain “how the watershed assessment based on the very large WAAs can identify cumulative watershed effects related to timber operations, distinguish between natural and man-caused event effects, and identify the location of sensitive areas for project planning and mitigation.” A letter by CDF’s Deputy Director Craig Anthony written in November 1997 to Pacific Lumber, months before public review began, stated that the SYP in its then form “must address the watershed assessment issues described below before it is sufficient for public review.” Anthony continued, “CDF concurs with other reviewing agencies that the watershed assessment areas (WAAs) are so large that potentially significant impacts from intensive management in one or more of the smaller subwatersheds” may occur without those impacts being detected by the proposed monitoring system. *499 Pacific Lumber prepared and submitted a revised draft SYP/HCP in June 1998. The Director then released this draft in July 1998 for public review. But a letter from John Munn, CDF’s soil erosion studies project manager, the following year in November 1998, at a time when public review was almost over, stated that the issues described in Anthony’s November 1997 letter “have not been addressed.” The record further indicates that Pacific Lumber and the state and federal governments had entered on February 27, 1998, into a Pre-Permit Application Agreement in Principle which, as noted, set forth a procedural framework for processing the required environmental documents, including the SYP. The agreement provided in part that Pacific Lumber was to submit to CDF an SYP that incorporated a range of timber growth estimates employing various timber management strategies. Upon the receipt of those estimates, “CDF will find the SYP sufficient for public review.” No mention was made in this agreement of the watershed analysis issues. As John Munn stated in a December 18, 1998 letter to Pacific Lumber, explaining his continued pursuit of the watershed analysis issues after the initial period of public review: “The Pre-Permit Application Agreement in Principle . . . simply says that [CDF] will find the SYP sufficient for public review. Although the Department would like to have major concerns addressed prior to public review, this is not a requirement of the Forest Practice Rules.” Moreover, whether or not the public draft SYP was sufficient for public review under the Forest Practice Rules, it appears clear that the Director did not abuse his discretion when his actions are viewed in light of Assembly Bill 1986, the state legislation authorizing the Headwaters Agreement. (Stats. 1998, ch. 615.) Sections 3, subdivision (a)(1), and 4, subdivision (c) of the statute specifically contemplate that the watershed analysis process will be completed after the approval of the SYP/HCP, and that until the process is completed and site-specific prescriptions emerging from that process have been implemented by the relevant government agencies, interim measures such as 100-foot no-cut buffers for class I watercourses, 16 will be adopted. Although Assembly Bill 1986’s September 1998 enactment postdated the July 1998 circulation of the Public Review Draft SYP/HCP, the statute certainly appears to legislatively ratify the decision of CDF and other government agencies to circulate a public review draft before completion of individual planning watershed analysis. We therefore conclude that the circulation of the Public Review Draft was not error. *500 7. Insufficiency of the Sustained Yield Plan Approval EPIC contends that the SYP should not have been approved because it lacked the information identified above not only at the public review stage but also in its final form. As indicated above, a number of CDF officials pointed out the inadequacy of Pacific Lumber’s watershed analysis. This inadequacy was not remedied before the SYP was approved. The dissatisfaction was expressed at the executive level by Douglas Wheeler, Director of the Resources Agency of California, who sent a letter to Pacific Lumber on December 8, 1998, a few months before the SYP was approved, stating: “Specifically, the watershed assessment areas should be described and reduced in size”; and “the assessment must then consider past, present, and future impacts.” EPIC also points to the comments during the public review process by Robert Hrabes, a forester and resource economist in its employ, that explain the significance of the lack of planning watershed analysis: “Planning watersheds, which averaged [10,000] to 20,000 acres, have been delineated by state water resource personnel and are correlated with topographic and drainage patterns across the landscape. At the scale of the planning watershed, it is possible to ascertain the potential contributory effects of plan ground disturbing activities in conjunction with other activities as well as whether resource sensitivity is within a geographic area united by common watershed drainage patterns.” In contrast, the Pacific Lumber watershed assessment areas “range in size from 55,000 acres to 426,000 acres and each [watershed assessment area] encompasses numerous planning watersheds. At the highly aggregated scale of a [watershed assessment area], it is impossible to assess the extent to which individual planning watersheds are being cumulatively impacted by [Pacific Lumber’s logging activities] and other industrial timber harvesting and road building activities.” As discussed, Assembly Bill 1986 specifically contemplates deferred watershed analysis to be completed after the SYP and HCP are approved. That statute and the HCP prescribe a five-year period after the SYP’s and HCP’s approval in which the watershed analysis will be accomplished. But this fact does not entirely resolve the issue before us. As noted, under the Forest Practice Rules, an SYP “shall not replace a THP. However, to the extent that sustained timber production, watershed impacts and fish and wildlife issues are addressed in the approved SYP, these issues shall be considered to be addressed in the THP; that is the THP may rely upon the SYP.” (FP Rules, § 1091.2, italics added.) In approving the SYP, the CDF director also approved a conifer harvest level of an average of 178.8 mmbf per year for the first decade—and specifically found that Pacific Lumber may rely on that estimate in its future timber harvest plans. *501 EPIC argues that, apart from the question whether substantial evidence supports that estimate, CDF failed to proceed according to law because it approved that estimate before it had gathered critical information necessary to understand the effects of Pacific Lumber’s timber harvesting on the environment, and therefore necessary to arrive at an accurate long-term sustained yield estimate. It points to the provision of Forest Practice Rules and the Forest Practice Act itself, that the achievement of “maximum sustained production of high-quality timber products” (FP Rules, § 1091.1, subd. (b)) that is the goal of the act must be “consistent with the protection of soil, water, air, fish and wildlife resources.” (FP Rules, § 1091.45(a); see Pub. Resources Code, § 4513, subd. (a).) It also points to Forest Practice Rules section 1091.6, subdivision (c): “The SYP shall contain a description of the individual planning watersheds in sufficient detail to allow a review of the analysis of impacts.” Without sufficient information about the environmental impacts of Pacific Lumber’s contemplated intensive logging, EPIC argues, there can be no reliable long-term sustained yield estimate which, as discussed, signifies a timber harvest that is, among other things, environmentally sustainable. All parties appear to agree that the long-term sustained yield estimate is at the core of a sustained yield plan, and EPIC argues that in the absence of a reliable estimate, the SYP itself must be invalidated. Moreover, EPIC argues, in essence, that the issue of this insufficiency is not excused or addressed by Assembly Bill 1986. CDF contends that the watershed planning and assessment was adequate to comply with the Forest Practice Rules. It points to Forest Practice Rules section 1091.6, subdivision (a), which provides that “[t]he minimum assessment area shall be no less than a planning watershed. The assessment area may include multiple watersheds . . . .” Section 1091.6, subdivision (d) further provides: “The SYP submitter shall utilize any one or a combination of methods to assess adverse watershed impacts including but not limited to: FU]... [f] (3) Other methods proposed in the SYP and approved by the Director.” Yet the fact that section 1091.6, subdivision (a) of the Forest Practice Rules refers to “assessment area” and provides that the “minimum assessment area shall be no less than a planning watershed” but may include “multiple watersheds” does not modify the obligation found in section 1091.6, subdivision (c) to describe “individual planning watersheds in sufficient detail to allow a review of the analysis of impacts.” An “assessment area” generally refers to the total geographic area over which environmental review must be conducted, and the controversy surrounding such areas generally concerns whether a government agency and the plan submitter have selected areas that are too small to fully encompass the environmental impacts of a project or logging activity on an endangered or threatened species. (See Ebbetts Pass Forest Watch v. California Dept. of Forestry & Fire Protection (2008) 43 *502 Cal.4th 936, 945-951 [77 Cal.Rptr.3d 239, 183 P.3d 1210].) Here, the question is not whether the overall assessment area referenced in section 1091.6, subdivision (a) was sufficiently large in scope, but whether watershed assessment areas were too large to permit the individual watershed analyses required by the Forest Practice Rules. Although Pacific Lumber contends that “the watershed assessment contained information for individual planning watersheds consistent with the [Forest Practice Rules],” it cites to a portion of the SYP that merely lists the individual planning watersheds within each watershed assessment area. This is plainly insufficient to meet the descriptive requirements of section 1091.6, subdivision (c). CDF also points to the definitional section of the Forest Practice Rules, section 895.1, defining “planning watershed” (see fn. 15, ante) and in particular to the language that “[timber harvest] Plan submitters may propose and use different planning watersheds, with the director’s approval.” But nothing in the record suggests that the Director approved any “different planning watershed” in this case, or that the permitted use of watershed assessment areas at the SYP stage displaced Pacific Lumber’s obligation under section 1091.6, subdivision (c) to assess impacts on individual planning watersheds. CDF further seeks to justify its manner of proceeding by pointing to the fact that the SYP is “a large scale planning documents] similar to a programmatic environmental impact report.” The CDF contends that the relationship between an SYP and a THP “is analogous to the relationship between a programmatic FIR and a site-specific FIR.” In other words, CDF and Pacific Lumber argue, echoing the Court of Appeal, that the SYP engaged in the common practice in environmental analysis of “tiering.” Tiering is a process “by which an agency prepares a series of EIRs or negative declarations, typically moving from general, regional concerns to more site-specific considerations with the preparation of each new document.” (Remy et al., Guide to CEQA (11th ed. 2006) p. 601.) We recently articulated the appropriate role of tiering: “While proper tiering of environmental review allows an agency to defer analysis of certain details of later phases of long-term linked or complex projects until those phases are up for approval, CEQA’s demand for meaningful information ‘is not satisfied by simply stating information will be provided in the future.’ [Citation.] As the CEQA Guidelines explain: ‘Tiering does not excuse the lead agency from adequately analyzing reasonably foreseeable significant environmental effects of the project and does not justify deferring such analysis to a later tier FIR or negative declaration.’ (Cal. Code Regs., tit. 14, § 15152, subd. (b).) Tiering is properly used to defer analysis of environmental impacts and mitigation measures to later phases when the impacts or *503 mitigation measures are not determined by the first-tier approval decision but are specific to the later phases. For example, to evaluate or formulate mitigation for ‘site specific effects such as aesthetics or parking’ (id., § 15152 [Discussion]) may be impractical when an entire large project is first approved; under some circumstances analysis of such impacts might be deferred to a later-tier FIR.” (Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova, supra, 40 Cal.4th 412, 431, fn. omitted.) Stated another way, CEQA contemplates consideration of environmental consequences at the “ ‘ “earliest possible stage, even though more detailed environmental review may be necessary later.” ’ [Citation.] The requirements of CEQA cannot be avoided by piecemeal review which results from ‘chopping a large project into many little ones—each with a minimal potential impact on the environment—which cumulatively may have disastrous consequences.’ ” (Rio Vista Farm Bureau Center v. County of Solano (1992) 5 Cal.App.4th 351, 370 [7 Cal.Rptr.2d 307].) On the other hand, “ ‘ “[W]here future development is unspecified and uncertain, no purpose can be served by requiring an FIR to engage in sheer speculation as to future environmental consequences.” [Citation.]’ ” (Id. at p. 372.) In the present case, there is no indication that analysis of planning watershed assessments was infeasible under the principles of tiering cited above, i.e., that the lack of specific details about Pacific Lumber’s projected activities made it infeasible to do individual watershed planning analysis. In fact, the completion of the watershed analysis within five years was not tied to any particular THP and was not contingent on Pacific Lumber formulating the siting and other details of its logging activity more precisely. Rather, as Pacific Lumber admits, “the deferral of a more specific analysis of smaller ‘planning watersheds’ was because more detailed site-specific information was not readily available at that smaller scale by the conclusion of the administrative review process on March 1, 1999 . . . .” As discussed above, the March 1, 1999 deadline was imposed by federal funding legislation, and did not mark a natural stopping point in the environmental analysis. What was done in this case is best characterized not as tiering of environmental analysis but rather as deferring a portion of the analysis in order to approve the SYP by a statutory deadline. As noted, the Forest Practice Rules provide that “to the extent that sustained timber production, watershed impacts and fish and wildlife issues are addressed in the approved SYP, these issues shall be considered to be addressed in the THP; that is the THP may rely upon the SYP.” (FP Rules, § 1091.2.) The position of CDF and Pacific Lumber has been that future THP’s may not rely on the SYP’s watershed impacts analysis, because it is admittedly incomplete, but that it may rely on its analysis of long-term *504 sustained yield. But the above categories of environmental analysis, although distinct, are interrelated, and the substantial informational and analytic gap in the analysis of watershed impacts, which directly affect fish and wildlife issues, may also call into question the reliability of the long-term sustained yield estimate, which depends in part on an assessment of watershed and wildlife impacts. In any case, whether or not there was adequate justification in 1999 for deferring individual watershed planning analysis, we perceive no justification for further delay. As discussed, we hold that an identifiable SYP was never properly approved and must be resubmitted for approval. We hold also that the document must include individual planning watershed analyses, which CDF agrees is necessary to address the cumulative effects of Pacific Lumber’s logging practices on the 211,000 acres in question. In considering whether to approve the resubmitted SYP, moreover, CDF must decide whether the information on individual planning watersheds complies with the Forest Practice Rules and is adequate to support Pacific Lumber’s long-term sustained yield estimate. 8. Sustained Yield Plan’s Demonstration of the Maximum Sustained Production of High Quality Timber The Steelworkers claim that the SYP violated the provision in Forest Practice Rules section 1091.45, subdivision (a) that an SYP must demonstrate how sustained production of “high quality timber products” will be achieved. In support of this claim, the Steelworkers point to what they contend are “several undisputed factual findings on this issue” by the superior court. First, that “old-growth trees are high-quality timber; in fact, the highest quality,” and produce the “most desirable commercial timber.” Second, “that the majority of old-growth trees projected to be logged over 120 years will be felled in the first decade and more than 80 percent in the first 20 years.” Third, that such a “rate of logging does not balance growth and harvest over time with respect to old-growth timber.” The success of this claim depends upon the Steelworkers’s equating “high-quality timber products” in the Forest Practice Rules with old growth trees. In making this equation, the Steelworkers cite two pieces of evidence in the record. The first is a reference in section 3.9 of the final EIS/EIR singling out the unique attributes of old growth forests. These include that old-growth redwood stands “may have 10 to 20 times the wood volume of an entire acre of trees in the deciduous forests of eastern North America [citation],” that the “volume and the quality of the wood . . . make such redwood trees extremely valuable,” and that old growth forests “provide important habitat for many plant and animal species not provided by younger forests.” The Steelworkers *505 also cite to a table found in the Public Review Draft of the SYP demonstrating that old growth redwoods, and to a lesser degree old growth Douglas firs, are significantly more valuable economically than younger growth species. We conclude that these citations fail to demonstrate that CDF violated Forest Practice Rules section 1091.45, subdivision (a). The fact that old growth timber is of the highest quality, and that 80 percent will be logged over the first 20 years does not mean that other, remaining timber is not of “high quality” within the meaning of the Forest Practice Rules. These rules provide that maximum sustained production is demonstrated in an SYP “by providing sustainable harvest yields established by the landowner which will support the production level of those high quality timber products the landowner selects while at the same time” meeting the various other requirements. (FP Rules, § 913.11, subd. (b), italics added.) As noted, we defer to an agency’s interpretation of its own regulations, particularly when the interpretation implicates areas of the agency’s expertise. (Yamaha Corp. of America v. State Bd. of Equalization, supra, 19 Cal.4th 1, 12-13.) Although heavily logging old growth timber in the early decades may cause economic or ecological repercussions, the Steelworkers have failed to demonstrate that such heavy logging, by itself, violates any Forest Practice Rule. 9. Confusion of Late Succession Forests with Late Serai Habitat Section 919.16, subdivision (a) of the Forest Practice Rules states that “[w]hen late succession forest stands are proposed for harvesting and such harvest will significantly reduce the amount and distribution of late succession forest stands or their functional wildlife habitat value so that it constitutes a significant adverse impact on the environment,” then “[t]he THP, SYP, or NTMP[ 17 ] shall include a discussion of how the proposed harvesting will affect the existing functional wildlife habitat for species primarily associated with late succession forest stands in the plan or the planning watershed, as appropriate, including impacts on vegetation structure, connectivity, and fragmentation.” As the Court of Appeal opinion explained, EPIC contends “that the Sustained Yield Plan here does not include such information. The Public Review Draft supplies an evaluation of ‘late serai forests,’ a classification that includes but is not limited to late succession^ forests. The category of ‘late serai forests’ is also used in the Habitat Conservation Plan and in the EIS/EIR.” A “late serai forest” is defined in the public draft SYP as “stands with overstory trees that on average are larger than generally 24 [inches diameter breast height] and may have developed a multi-storied structure. It *506 occurs in stands as young as 40 years old but more typically in stands about 50 to 60 years old and older.” Late succession forests, on the other hand, are dominated by large, old growth trees. So late serai forests may consist largely of trees younger than those found in late succession forests, with features less suitable to certain species than the latter forests. EPIC contends that Pacific Lumber was not authorized to unilaterally change the definition of what constituted a late succession forest, and that this altered definition amounted to noncompliance with Forest Practice Rules section 919.16. They point to a statement by CDF in response to comments on the public draft EIS/EIR: “We are aware that there is a gap in [Pacific Lumber’s] serai stage classification: that it does not take into account the lengthy transition from even-age stands that are relatively young and weakly stratified (including [Pacific Lumber’s] late serai stage) to relatively old, complex, and highly stratified stands that would be considered old-growth. Monitoring efforts and agency considerations in the watershed analysis process will be focused on actual stand attributes.” The Court of Appeal concluded that Forest Practice Rules section 919.16, subdivision (a) was not violated because the regulation called for analysis of late succession or forest impacts at either the SYP or THP stages. It further concluded: “In any event, the variant classification used by [Pacific Lumber] was harmless. . . . [EPIC has] made no assertion that the habitats of any particular wildlife species were overlooked or omitted by the analysis of late serai forests, rather than late succession forests.” We agree that deferring the analysis of late succession forests to the THP stage, although it creates an analytical gap in assessing impacts on wildlife, does not violate the Forest Practice Rules, when, as here, the relevant environmental documents contain substantial analysis of the impacts of timber operations on wildlife associated with late succession forests. On remand, the parties may address whether inclusion of any omitted information related to late succession forests in the resubmitted SYP would be appropriate. C. Challenges to the Incidental Take Permit EPIC makes several challenges to the validity of the state 50-year Incidental Take Permit. Each of these will be considered in turn. 1. The Validity of the “No Surprises Clauses” EPIC contends that the DFG violated CESA, the California Endangered Species Act (Fish & G. Code, § 2050 et seq.), in agreeing to what are called *507 “no surprises clauses” that would limit in advance the obligation of Pacific Lumber to mitigate various impacts on endangered and threatened species. An overview of CESA is useful for addressing these claims. The Legislature has declared that “[I]t is the policy of the state to conserve, protect, restore, and enhance any endangered species or any threatened species and its habitat.” (Fish & G. Code, § 2052.) “Under CESA, a native species of bird, mammal, fish, amphibian, reptile, or plant is considered ‘endangered’ when it ‘is in serious danger of becoming extinct throughout all, or a significant portion, of its range’ (Fish & G. Code, § 2062), and ‘threatened’ when it ‘is likely to become an endangered species in the foreseeable future in the absence of . . . special protection and management efforts.’ (Fish & G. Code, § 2067.)” (Mountain Lion Foundation v. Fish & Game Com. (1997) 16 Cal.4th 105, 114 [65 Cal.Rptr.2d 580, 939 P.2d 1280].) Central to CESA is its prohibition on the taking of an endangered or threatened species. (Fish & G. Code, § 2080.) To “take” in this context means to catch, capture or kill. (Fish & G. Code, § 86.) Nonetheless, CESA allows the DFG to authorize a “take” that is incidental to an otherwise lawful activity if certain conditions are met. (Fish & G. Code, § 2081, subd. (b); see also Cal. Code Regs., tit. 14, § 783 et seq.) At the heart of CESA is the obligation to mitigate such takes. “The impacts of the authorized take shall be minimized and fully mitigated. The measures required to meet this obligation shall be roughly proportional in extent to the impact of the authorized taking on the species. Where various measures are available to meet this obligation, the measures required shall maintain the applicant’s objectives to the greatest extent possible. All required measures shall be capable of successful implementation. For purposes of this section only, impacts of taking include all impacts on the species that result from any act that would cause the proposed taking.” (Fish & G. Code, § 2081, subd. (b)(2); hereafter section 2081(b)(2).) In this case, a state Incidental Take Permit was issued to Pacific Lumber authorizing the incidental take of the marbled murrelet, an endangered bird, and the bank swallow, a threatened bird. The taking of two other fully protected species was not permitted under the permit. 18 EPIC contends that the state Incidental Take Permit was issued with unlawful no surprises clauses. As explained in the HCP, the no surprises *508 provision consists of two major components. First, if there are changed circumstances that were anticipated in the HCP, and mitigation measures were prescribed to meet the adverse impacts of those changed circumstances, then, if and when those circumstances occur, the landowner will be expected to implement those measures and no others. As the HCP’s Implementation Agreement makes clear, this is the case even if “additional conservation and mitigation measures are deemed necessary by [DFG] to respond to a Changed Circumstance.” Second, in the case of unforeseen circumstances, the government will not require the commitment by the landowner of additional land, water, or financial compensation, or additional restrictions on the use of land, water or other natural resources unless the landowner consents. “Unforeseen circumstances” are defined as “those changes in circumstances affecting a species or geographic area covered by an HCP, that could not reasonably be anticipated by a landowner and the wildlife agencies at the time of the HCP development and that result in a substantial and adverse change in the status of a species covered by the HCP.” Particular types of “changed circumstances” and “unforeseen circumstances” are defined in the HCP. For example, “fire changed circumstances” are wildfires, including “those originating from timber operations and prescribed burning” that are 5,000 acres or less. “Fire unforeseen circumstances” are defined as all such wildfires that are over 5,000 total acres. Changed and unforeseen circumstances for wind, landslides, and flooding are similarly defined in terms of the magnitude of the events. EPIC argues that these kinds of advanced assurances that additional mitigation measures will not be required even when the measures are deemed necessary by DFG is contrary to that agency’s statutory mandate. EPIC bases the argument on the language of section 2081(b)(2), as discussed above, that the impact of the authorized take must be “fully mitigated.” Pacific Lumber and DFG have several responses to this argument. First, they note, as the Court of Appeal did, that the no surprises rule is the established policy of federal wildlife agencies, as adopted by federal regulation (50 C.F.R. § 17.22). They contend the authority to make regulatory assurances likewise resides in DFG. They also point to a provision of the Natural Community Conservation Planning Act (NCCPA; Fish & G. Code, § 2800 et seq.), which allows for similar regulatory assurances in the context of the development of a Natural Community Conservation Plan. EPIC counters that the existence of a provision within the NCCPA authorizing a “no surprises” provision undermines rather than supports Pacific Lumber’s argument. It argues that this statute demonstrates that when the Legislature has intended to authorize an agency to give a landowner regulatory *509 assurances that no further mitigation measures will be required in the case of changed or unforeseen circumstances, it has done so explicitly, and that we should infer from the lack of such explicit authorization in CESA that the Legislature did not intend such authorization. (See Dyna-Med, Inc. v. Fair Employment & Housing Com. (1987) 43 Cal.3d 1379, 1395 [241 Cal.Rptr. 67, 743 P.2d 1323] [when the Legislature intends to authorize an agency to award damages, it does so expressly, as evidenced by pertinent statutes].) In order to evaluate that argument, it is useful to understand the background and scope of the NCCPA. As originally enacted in 1991, the act provided that DFG “may enter into agreements with any person for the purpose of preparing and implementing a natural community conservation plan to provide comprehensive management and conservation of multiple wildlife species” (Fish & G. Code, former § 2810), and that such planning “may be undertaken by local, state and federal agencies independently or in cooperation with other persons.” (Fish & G. Code, former § 2820; Stats. 1991, ch. 765, § 2, pp. 3424-3425.) The former statute further provided that the Fish and Game Commission, on recommendation from DFG, “may authorize . . . the taking of any candidate species whose conservation, protection, restoration, and enhancement is provided for in a [DFG] approved natural community conservation plan” that ensured compatibility with the federal Endangered Species Act. (Id., former § 2830; see also id., former § 2825, subd. (a)(6).) The NCCPA was amended in 2002 (Stats. 2002, ch. 4, § 2) to define in much greater detail the kind of provisions that are to be included in a natural community conservation plan, including public participation in the development of the plan (Fish & G. Code, § 2815), and an extensive set of findings required for plan approval (id., § 2820, subd. (a)). These findings are to include that “[t]he plan integrates adaptive management strategies that are periodically evaluated and modified based on the information from the monitoring program . . .” (id., subd. (a)(2)) and that “[t]he plan provides for the protection of habitat, natural communities, and species diversity on a landscape or ecosystem level through the creation and long-term management of habitat reserves or other measures that provide equivalent conservation of covered species” (id., subd. (a)(3)). Section 2820 also includes detailed provisions for implementation agreements (id., subd. (b)) and provisions for monitoring and enforcement (id., subds. (b), (c)). Section 2820, subdivision (f)(2) provides: “If there are unforeseen circumstances, additional land, water, or financial compensation or additional restrictions on the use of land, water, or other natural resources shall not be required without the consent of plan participants for a period of time specified in the implementation agreement, unless the department determines that the plan is not being implemented consistent with the substantive terms of the implementation agreement.” *510 DFG argues that CESA and the NCCPA are distinct statutory schemes that never have been amended together, and that therefore the explicit provision for regulatory assurances in the latter statute does not imply a lack of authority to grant regulatory assurances under the former statute. We find DFG’s argument unpersuasive. First, although CESA and the NCCPA are distinct statutes, they share a common objective—they authorize the incidental taking of threatened and endangered species in a way that minimizes impacts on those species. The statutes take different routes to that objective, CESA through the imposition of “roughly proportional” mitigating measures on landowners, the NCCPA through a comprehensive agreement incorporating various mitigation measures, including the creation of habitat reserves. Although in practice these lines may be blurred, the Legislature clearly contemplated distinct statutory paths to the same objective. Moreover, CESA has been amended several times either contemporaneously with or subsequent to the 2002 amendment of the NCCPA. (See Stats. 2004, ch. 614, § 1; Stats. 2003, ch. 62, § 96; Stats. 2002, ch. 32, § 2.) Where as here the Legislature has established alternative statutory schemes for authorizing and minimizing the taking of endangered species, but has provided a particular benefit to landowners—regulatory assurances—in only one of those schemes, the natural inference is that it did not intend the same assurances to be provided in the other scheme. Nor does the language of CESA assist DFG’s position. Pacific Lumber and DFG point out that although the act speaks of “fully mitigat[ing]” the impacts of the authorized take, it also has significant limiting language. The statute provides that the landowner’s obligation only be “roughly proportional in extent to the impact ... on the species.” (§ 2081(b)(2).) As amici curiae California Building Industry Association et al. point out, the “roughly proportional” language mirrors the constitutional standard for what constitutes the taking of property set forth in Dolan v. City of Tigard (1994) 512 U.S. 374 [129 L.Ed.2d 304, 114 S.Ct. 2309]. In that case, the court held under Fifth and Fourteenth Amendment takings jurisprudence that when a government requires a dedication of land in exchange for a development permit, it must be guided by the principle of “rough proportionality,” i.e., it must “make some sort of individualized determination that the required dedication is related both in nature and extent to the impact of the proposed development.” (512 U.S. at p. 391.) As we stated in a case that applied Dolan’s rationale to development fees, Dolan was “concerned with implementing one of the fundamental principles of modem takings jurisprudence— ‘to bar Government from forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.’ ” (Ehrlich v. City of Culver City (1996) 12 Cal.4th 854, 880 [50 Cal.Rptr.2d 242, 911 P.2d 429].) *511 Thus, to require that mitigation measures be roughly proportional to a landowner’s impact on a species means that the landowner is required to mitigate only its own impacts on the species. If the no surprises provisions applicable to Pacific Lumber did no more than guarantee this kind of proportionality, then they would be unquestionably within DFG’s purview. But these provisions go further. For example, included in the changed and unforeseen circumstances pertaining to fire are fires “originating from timber operations.” Furthermore, in defining “landslide” or “flood,” changed and unforeseen circumstances are cast solely in terms of magnitude, and do not differentiate between those events partially caused or exacerbated by timber harvesting and those that are not. Nor do the regulatory assurances permit DFG to require additional mitigation measures when changed and unforeseen circumstances have rendered previously prescribed mitigation measures insufficient. Inasmuch as the language categorically exempts Pacific Lumber from mitigating impacts of its own activities on listed species and their habitat, it goes further than the language section 2081(b)(2) contemplates. In other words, reading the “roughly proportional” language together with the “fully mitigate” language leads to the conclusion the Legislature intended that a landowner bear no more—but also no less—than the costs incurred from the impact of its activity on listed species. To the extent that the changed and unforeseen circumstances provisions of the Incidental Take Permit exempt landowners from this obligation, they exceed DFG’s statutory authority under CESA. The language in the last sentence of section 2081(b)(2) stating that “impacts of taking include all impacts on the species that result from any act that would cause the proposed taking” (§ 2081(b)(2)) further supports our construction of the statute. 19 *512 Pacific Lumber and DFG in support of their argument also point to the language providing that “[w]here various measures are available to meet this obligation [to fully mitigate], the measures required shall maintain the applicant’s objectives to the greatest extent possible.” (§ 2081(b)(2).) This language does not diminish the extent of a landowner’s obligation under CESA, however, but merely provides that when that obligation can be met in several ways, the way most consistent with a landowner’s objectives should be chosen. It does not relieve the landowner of the obligation to fully mitigate its own impacts. With respect to the changed circumstances portion of the no surprises provisions, Pacific Lumber and DFG endorse the Court of Appeal’s conclusion: “The required responses to changed circumstances are designed to mitigate the impact of physical processes (such as fire, flood, earthquake) that can be anticipated in the course of the underlying activities. Insofar as [EPIC contends] that the responses will not in fact fully mitigate the adverse impacts, their contention is a challenge to the sufficiency of the evidence to support the Department’s finding on full mitigation, and that challenge is foreclosed.” But as noted, the Implementation Agreement to the HCP provides that even “[i]f additional conservation and mitigation measures are deemed necessary by [DFG] to respond to a Changed Circumstance and such measures were not provided for pursuant to the HCP, [DFG] will not require any new, additional or different conservation and/or mitigation measures from [Pacific Lumber] in addition to those provided for pursuant to the HCP without the consent of [Pacific Lumber].” Thus, we do not understand EPIC to be mounting a sufficiency of the evidence challenge to the mitigation measures proposed in the HCP in response to certain anticipated changed circumstances, but rather to be challenging a provision stating that even when DFG itself concludes the prescribed mitigation measures are not adequate in light of changed circumstances, it will not impose new measures without Pacific Lumber’s consent. As discussed, this provision cannot be reconciled with Pacific Lumber’s duty to fully mitigate the impacts of its take. *513 Nor do we agree with the Court of Appeal’s approach in addressing the unforeseen circumstances issue. As the court stated: “With respect to unforeseen circumstances, the full mitigation requirement does not apply. The focus of the full mitigation requirement is on adverse impacts that result from an ‘act’—i.e., a purposeful activity. (Fish & G. Code, § 2081, subd. (b); see Department of Fish & Game v. Anderson-Cottonwood Irrigation Dist. (1992) 8 Cal.App.4th 1554, 1561 [11 Cal.Rptr.2d 222].) Adverse impacts that result from unforeseen circumstances are impacts that cannot reasonably be anticipated, not impacts from purposeful activities.” We agree that the focus of the full mitigation requirement is on adverse impacts that result from purposeful activity. But as discussed above, “unforeseen” circumstances, as defined in the HCP, includes impacts resulting from purposeful activity. A catastrophic event such as a fire or flood is classified as unforeseen when it reaches a certain magnitude, whether or not Pacific Lumber’s timber operations contributed to that event. Moreover, when natural disasters change baseline conditions, then logging activities that previously would not have had a significant impact on endangered species may now have such an impact, and therefore fall within the scope of the CESA obligation to fully mitigate impacts. To be sure, there is no obligation for a permit holder to mitigate the impacts of the natural disasters themselves when it did not contribute to them. But when these impacts are exacerbated by the permit holder’s own subsequent purposeful activities, then section 2081(b)(2) mandates the full mitigation of the impacts of a take, guided by the principle of rough proportionality. Particularly in light of the 50-year duration of the permit, provisions that freeze Pacific Lumber’s obligations to mitigate in the face of changing circumstances, even when these circumstances are labeled “unforeseen,” cannot comply with the statutory mandate. 20 Moreover, the term “unforeseen circumstance” is a misnomer. Obviously, events identified in the HCP, such as fires over 5,000 acres and 100-year floods, are not unforeseen. They may be rare events, but if Pacific Lumber’s timber operations contribute to causing such events to occur more frequently, or if the events themselves change conditions in such a way as to necessitate additional mitigation measures, there is no reason under section 2081(b)(2) that additional measures cannot be required. Pacific Lumber further argues that Assembly Bill 1986, the legislation that authorized the Headwaters Agreement, implicitly approved the no surprises *514 clause. The statute “made compliance with the Implementation Agreement [for the HCP] a condition of the SYP and other permits.” Section 3 of that act states: “Notwithstanding any other provision of law, funds appropriated by this act shall only be encumbered by the board if the final habitat conservation plan (hereafter ‘final HCP’), implementing agreement, and permits to allow the incidental take of threatened and endangered species, . . . incorporate, at minimum, the following additional conditions and the final HCP is no less protective of aquatic or avian species than the draft HCP, as amended by those conditions . . . .” (Stats. 1998, ch. 615, § 3.) The Act then goes on to prescribe certain specific conditions and restrictions on Pacific Lumber’s timber harvesting. (Ibid.) This argument is not persuasive. Assembly Bill 1986’s reference to the draft HCP established the minimum protective measures to be included in the final HCP—it was to serve as a floor, not a ceiling. The citation to a draft HCP that was then undergoing public review and possible revision obviously did not signify legislative approval of the contents of that draft beyond its use as a baseline. Moreover, an implied amendment of a statute is generally disfavored. (Lesher Communications, Inc. v. City of Walnut Creek (1990) 52 Cal.3d 531, 540-541 [277 Cal.Rptr. 1, 802 P.2d 317].) Here, the general reference to complying with the conditions imposed by the draft HCP does not evince any legislative intent to alter the scope of DFG’s statutory authority to issue incidental take permits under Fish and Game Code section 2081(b)(2), nor indicate any consideration of the use of regulatory assurances. There is no evident legislative intent to grant an exception to section 2081(b)(2)’s full mitigation and rough proportionality requirements. Pacific Lumber and DFG, as well as several amici curiae, also emphasize the important policy promoted by such regulatory assurances, endorsing the Court of Appeal’s statement that “the ‘no surprises’ rule is . . . intended to encourage landowners to factor into their day-to-day activities measures to protect endangered species. By bringing in an element of certainty, the no-surprise rule removes a disincentive a landowner might have to obtaining an incidental take permit and submitting to the mitigation measures.” As discussed above, however, the Legislature has already provided a means for DFG to validly provide the types of regulatory assurances at issue here to landowners pursuant to the NCCPA. The expansion of the circumstances under which such assurances can and should be given is a matter best addressed by the Legislature. 21 *515 2. Violations of the Public Trust Doctrine EPIC contends that the Incidental Take Permit constituted abandonment of the DFG’s public trust obligation to protect the natural resources of this state by virtue of the no surprises clauses, discussed above, and because of improper delegation to Pacific Lumber to determine which northern spotted owl sites will receive protection and which will be eliminated. As the Court of Appeal recognized, there are two distinct public trust doctrines invoked by EPIC. First is the common law doctrine, which involves the government’s “affirmative duty to take the public trust into account in the planning and allocation of water resources . . . .” (National Audubon Society v. Superior Court (1983) 33 Cal.3d 419, 446 [189 Cal.Rptr. 346, 658 P.2d 709].) The second is a public trust duty derived from statute, specifically Fish and Game Code section 711.7, pertaining to fish and wildlife: “The fish and wildlife resources are held in tmst for the people of the state by and through the department.” (Id., subd. (a).) There is doubtless an overlap between the two public tmst doctrines—the protection of water resources is intertwined with the protection of wildlife. (See National Audubon Society, supra, 33 Cal.3d at p. 447.) Nonetheless the duty of government agencies to protect wildlife is primarily statutory. Fish and Game Code section 1801, which declares that it is “the policy of the state to encourage the preservation, conservation, and maintenance of wildlife resources under the jurisdiction and influence of the state,” also declares in subdivision (h) that “[i]t is not intended that this policy shall provide any power to regulate natural resources or commercial or other activities connected therewith, except as specifically provided by the Legislature.” Generally speaking, therefore, we will look to the statutes protecting wildlife to determine if DFG or another government agency has breached its duties in this regard. In the previous part of this opinion we concluded that DFG breached its duty to require full mitigation of the impacts of an authorized take of a listed species under section 2081(b)(2) by the no surprises provisions in the HCP and Implementation Agreement. Its violation, therefore, is not of some general public tmst duty, but of a specific statutory obligation. Moreover, we find no support in the record for EPIC’s second claim, that in the Incidental Take Permit DFG improperly delegated to Pacific Lumber *516 which northern spotted owl sites should be preserved. Rather, the relevant documents reveal that DEG has maintained its authority to review Pacific Lumber’s site-specific decisions regarding preservation of northern spotted owl habitat. We therefore conclude the Incidental Take Permit did not violate a common law public trust duty. 3. Inadequate CESA Findings EPIC contends that there were inadequate CESA findings to support the Incidental Take Permit. Although the findings leave something to be desired, we disagree there is prejudicial error. Administrative agency decisions in which discretion is exercised may generally be challenged by a writ of administrative mandamus pursuant to Code of Civil Procedure section 1094.5. In Topanga Assn, for a Scenic Community v. County of Los Angeles (1974) 11 Cal.3d 506, 515 [113 Cal.Rptr. 836, 522 P.2d 12] (Topanga), we considered the meaning of subdivision (b) of that statute, defining “ ‘abuse of discretion’ to include instances in which the administrative order or decision ‘is not supported by the findings, or the findings are not supported by the evidence’ ” and subdivision (c), wherein “ ‘abuse of discretion is established if the court determines that the findings are not supported by substantial evidence in the light of the whole record.’ ” We concluded “that implicit in section 1094.5 is a requirement that the agency which renders the challenged decision must set forth findings to bridge the analytic gap between the raw evidence and ultimate decision or order. ... By focusing . . . upon the relationships between evidence and findings and between findings and ultimate action, the Legislature sought to direct the reviewing court’s attention to the analytic route the administrative agency traveled from evidence to action. In so doing, we believe that the Legislature must have contemplated that the agency would reveal this route. Reference, in section 1094.5, to the reviewing court’s duty to compare the evidence and ultimate decision to 'the findings’ ... we believe leaves no room for the conclusion that the Legislature would have been content to have a reviewing court speculate as to the administrative agency’s basis for decision.” (Topanga, supra, 11 Cal.3d at p. 515.) The findings do not need to be extensive or detailed. “ ‘[W]here reference to the administrative record informs the parties and reviewing courts of the theory upon which an agency has arrived at its ultimate finding and decision it has long been recognized that the decision should be upheld if the agency “in truth found those facts which as a matter of law are essential to sustain its . . . [decision].” ’ ” (Sierra Club v. California Coastal Commission, supra, *517 19 Cal.App.4th at p. 556.) On the other hand, mere conclusory findings without reference to the record are inadequate. (See Village of Laguna Beach, Inc. v. Board of Supervisors (1982) 134 Cal.App.3d 1022, 1035 [185 Cal.Rptr. 41].) EPIC contends that the CESA findings are inadequate. Under DEG regulations promulgated pursuant to CESA, the director of DEG must make findings that the take authorized by the Incidental Take Permit is consistent with the statutory requirements in Fish and Game Code section 2081, subdivision (b). (FP Rules, § 783.4, subd. (a).) EPIC claims that DFG’s CESA findings merely recited statutory criteria without any supporting rationale linking the evidence to the ultimate conclusion. The record discloses that the March 1, 1999 document containing the CESA findings recites the language of Fish and Game Code section 2081, subdivision (b) and affirms compliance with its provisions, referring to specific documents in the record: for example, that the “Take of Covered Species as defined in the ITP [Incidental Take Permit] will be incidental to the otherwise lawful activities covered under the ITP,” that the impacts will be “minimized and fully mitigated through the HCP’s Operating Conservation Program and [Implementation Agreement]” and that “the conservation and mitigation measures required pursuant to the HCP’s Operating Conservation Program are roughly proportional in extent to the impact of Pacific Lumber’s take.” Thus, the findings refer to a specific document—the HCP’s operating conservation program. This portion of the HCP describes conservation plans for each of the critical species expected to be impacted by Pacific Lumber’s activities, setting forth for each species specific management objectives, conservation measures, and a monitoring program. The findings also refer to the Implementation Agreement, where Pacific Lumber’s obligations are further delineated. The CESA findings were made in conjunction with findings for the final EIS/EIR. In the final EIS/EIR, the HCP’s conservation programs were analyzed, and it was concluded that these programs would mitigate the adverse effects of incidental take on various species. Although the better practice would have been for the CESA findings to have referred more specifically to those portions of the final EIS/EIR that support the conclusion that the impacts of the take will be minimized and fully mitigated, we have no trouble under the circumstances discerning “the analytic route the administrative agency traveled from evidence to action.” (Topanga, supra, 11 Cal.3d at p. 515; see No Slo Transit, Inc. v. City of Long Beach (1987) 197 Cal.App.3d 241, 260 [242 Cal.Rptr. 760].) We find no prejudicial error. *518 D. Challenges to the Streambed Alteration Agreement 1. Failure to Negotiate Lawful Agreement EPIC claims that DFG and Pacific Lumber did not enter into a proper Streambed Alteration Agreement pursuant to Fish and Game Code former section 1603, 22 and that agreement is therefore invalid. We disagree. Fish and Game Code former section 1603, subdivision (a) provided during the relevant period that “[i]t is unlawful for any person to substantially divert or obstruct the natural flow or substantially change the bed, channel, or bank of any river, stream, or lake designated by the department, or use any material from the streambeds, without first notifying the department of that activity, except when the department has been notified pursuant to Section 1601. The department, within 30 days from the date of receipt of that notice, or within the time determined by mutual written agreement, shall, when an existing fish or wildlife resource may be substantially adversely affected by that activity, notify the person of the existence of that fish or wildlife resource together with a description of the fish or wildlife, and shall submit to the person its proposals as to measures necessary to protect fish and wildlife. . . . The department’s description of an existing fish or wildlife resource shall be specific and detailed and the department shall make available upon request the information upon which its conclusion is based that the resource may be substantially adversely affected.” (Stats. 1996, ch. 825, § 3.5, p. 4327.) Subdivision (b) dictated that the parties are to enter into an agreement about the appropriate measures to adopt and provides a framework for resolving disagreements. The evident purpose of Fish and Game Code former section 1603 was to protect existing fish and wildlife resources, and it accomplished that purpose by imposing on DFG and private persons a set of interlocking obligations. A private person is obliged to notify DFG before it diverts or obstructs streams or other watercourses. This notice triggers DFG’s duty to determine if the obstruction or diversion “may” substantially adversely affect fish and wildlife. If that determination is made, then DFG has the duty to submit “proposals as to measures necessary to protect fish and wildlife,” and to conduct an appropriate investigation. DFG’s description of existing fish and wildlife resources must be “specific and detailed.” The person may then either accept the proposal or negotiate with DFG, and if agreement is not reached, both parties are obliged to follow the dispute resolution mechanism set forth in subdivision (b). Each of these obligations is to be performed pursuant to prescribed statutory deadlines. *519 With these rules in mind, we review the factual background behind the Streambed Alteration Agreement in this case. Despite earlier announcements of an intention to seek a Streambed Alteration Agreement, Pacific Lumber did not officially notify the department of an intention to engage in streambedaltering activity until very late in the regulatory approval process, on February 24, 1999. Rather than discuss specific streams that would be impacted by Pacific Lumber’s activities, the resulting agreement was instead a “master” document that encompasses the entire 211,000 acres without identifying the location of specific streams or activities. The notice referred to tire final SYP/HCP for the location of all streams and watercourses affected and the measures taken to protect fish and wildlife. This notice was filed with a Final Streambed Alteration Agreement, dated February 25, 1999, that had already been negotiated with DFG. The agreement was structured as follows. Exhibits A and B fist certain “covered activities” that are expected to occur on the property in question. Exhibit A consists of activities that DFG has determined “may substantially divert or obstruct the natural flow” of streams or other enumerated bodies of water, “depending on the location and/or impacts of the covered activities.” These covered activities “would be addressed under separate notifications and agreements pursuant to” section 1603. These activities include timber harvesting, site preparation, thinning, fire suppression, and road construction. A second set of covered activities, listed under exhibit B, is the subset of activities in exhibit A that are the subject of the present Streambed Alteration Agreement, and the agreement adopts in exhibit C specific measures necessary to protect fish and wildlife resources from these activities. Activities listed under exhibit B include construction of road crossings within class I and class II watercourses, water drafting, and operating conservation programs. Exhibit C lists various measures to protect against the detrimental effects of activities listed in exhibit B, including that any structure or culvert placed within any class I watercourse is to be designed and constructed so as not to constitute a barrier to the upstream or downstream movement of fish, and various prescriptions for constructing bridges across watercourses. Thus, the Streambed Alteration Agreement at issue here responds to the statutory mandate to protect fish and wildlife that may be adversely affected by streambed alteration in three ways: (1) by referencing mitigation measures put in place by the HCP/SYP filed in conjunction with the agreement; (2) by adopting certain conservation measures in addition to those required under the HCP/SYP with regard to some of Pacific Lumber’s anticipated activity; and (3) by expressly providing that most activities in which Pacific Lumber plans to engage, including timber harvesting and road construction, will require Pacific Lumber to enter into additional Streambed Alteration Agreements. *520 EPIC argues that Pacific Lumber and DFG failed to follow the mandatory procedures set forth in Fish and Game Code former section 1603—that Pacific Lumber failed to give timely notice and DFG failed to identify for Pacific Lumber the wildlife to be affected by the proposed stream altering activity. We disagree. Although Pacific Lumber and DFG may not have followed the precise procedures contemplated by section 1603, they appear to have substantially complied with that statute. Because the Streambed Alteration Agreement was undertaken in conjunction with a massive regulatory approval process that included an integrated HCP/SYP and an EIS/EIR, both Pacific Lumber and DFG had ample notice through this process—DFG that Pacific Lumber would engage in streambed-altering activity, and Pacific Lumber of the wildlife that would be affected and the mitigation measures that DFG and other government agencies would require to mitigate adverse impacts on fish and wildlife. EPIC also contends that the agreement here is not sufficiently specific with respect to particular streams, and that there is nothing in the statutes or regulations that authorizes DFG or Pacific Lumber to enter into a “master” Streambed Alteration Agreement, as they did here. We disagree. Statute and regulation neither specifically authorize nor forbid this type of master agreement. Of course, were such agreements used to circumvent the substantive requirements of Fish and Game Code section 1603 to identify with specificity the stream-altering activities and negotiate particular mitigating measures, they would obviously not pass muster. But there is no indication that the present agreement would do so. This “master” agreement is extremely limited in scope, adopting standard mitigating or protective measures for some of Pacific Lumber’s activities ancillary to timber harvesting, while deferring most of the measures to be adopted to future agreements, when Pacific Lumber’s plans for particular streambeds will be more concretely formulated. In light of DFG’s expertise and its statutory authority to formulate Streambed Alteration Agreements, we cannot say that this manner of proceeding violated the statutory duties to which either it or Pacific Lumber are subject. 2. Lack of Finding for Streambed Alteration Agreement EPIC also contends that the lack of any findings related to the Streambed Alteration Agreement makes that agreement invalid. DFG and Pacific Lumber respond that no findings are required. We agree. Findings are required in support of administrative decisions when such decisions are reviewable under Code of Civil Procedure section 1094.5 (see Topanga, supra, 11 Cal.3d at pp. 514-515) or are otherwise required by statute or regulation. Code of Civil Procedure section 1094.5, subdivision (a), provides administrative mandamus is available to review a decision made by *521 an agency as a result of a proceeding in which by law (1) a hearing is required to be given, (2) evidence is required to be taken, and (3) discretion in determining the facts is vested in the agency. The hearing and evidence requirements are met when a statute or regulation provides an opportunity for public input and requires a public agency to respond to that input, such as is the case with an EIR or THP. (Friends of the Old Trees v. Department of Forestry & Fire Protection (1997) 52 Cal.App.4th 1383, 1391-1392 [61 Cal.Rptr.2d 297].) On the other hand, an administrative decision that does not require a hearing or a response to public input is generally not reviewable under Code of Civil Procedure section 1094.5 but by traditional mandamus pursuant to Code of Civil Procedure section 1085, under an abuse of discretion standard, and no findings are required. (See Association for Protection etc. Values v. City of Ukiah (1991) 2 Cal.App.4th 720, 730-732 [3 Cal.Rptr.2d 488].) We conclude that a Streambed Alteration Agreement under section 1603 did not require findings, because the statute did not require that a hearing be held or public input be taken. Nor did any implementing regulation impose a findings requirement. On the other hand, an activity or project that necessitates a Streambed Alteration Agreement may require environmental review under CEQA. (See DEG, Lake and Streambed Alteration Program, Questions and Answers, No. 4 <http://www.dfg.ca.gov/habcon/1600/qa.html> [as of July 17, 2008] [“The Department must comply with . . . CEQA . . . before it may issue a final Lake or Streambed Alteration Agreement.” (Italics added.)].) CEQA requires findings under certain circumstances. (Pub. Resources Code, § 21081.) EPIC’s challenge to the final EIS/EIR’s CEQA findings are discussed below. We therefore conclude that the lack of separate findings supporting the present Streambed Alteration Agreement was not error. E. Challenges to the EIS/EIR 1. Inadequate Findings EPIC contends that DFG’s CEQA findings were insufficient. Under CEQA, in Public Resources Code section 21081, “no public agency shall approve or carry out a project for which an environmental impact report has been certified which identifies one or more significant effects on the environment that would occur if the project is approved or carried out unless ...[]□••• [t]he public agency makes one or more of the following findings with respect to each significant effect: [][] (1) Changes or alterations have been required in, or incorporated into, the project which mitigate or avoid the significant effects on the environment, [f] (2) Those changes or alterations are within the *522 responsibility and jurisdiction of another public agency and have been, or can and should be, adopted by that other agency. H] (3) Specific economic, legal, social, technological, or other considerations, including considerations for the provision of employment opportunities for highly trained workers, make infeasible the mitigation measures or alternatives identified in the environmental impact report.” EPIC contends that the final EIS/EIR identified several significant environmental impacts but failed to make one of the three findings set forth in Public Resources Code section 21081. More specifically, it contends that the EIS/EIR concluded there would be long-term and short-term adverse impacts on the northern spotted owl, red tree vole, Pacific fisher, and other late serai habitat species, but that DEG failed to make the required findings regarding each of these significant impacts, as required by Public Resources Code section 21081 and the CEQA guidelines. (Cal. Code Regs., tit. 14, § 15000 et seq. (hereafter CEQA Guidelines); see CEQA Guidelines, § 15091.) In addressing this claim, it first must be kept in mind that the project for which the EIS/EIR was prepared was the HCP/SYP, and that the HCP was specifically designed to mitigate significant impacts on wildlife. In the part of the EIS/EIR devoted to the northern spotted owl, for example, the EIS/EIR concludes that the project will have less than significant effects on the species, stating that although “effects may be significant in short and long-term due to potential substantial decline in population,” HCP mitigation and monitoring was “expected to minimize and mitigate effects to less than significant.” The HCP incorporates extensive conservation measures including the selection of at least 80 “activity sites” that will maintain suitable spotted owl habitat. Similar conclusions were reached as to the other species EPIC identifies in its brief as being inadequately addressed. Moreover, in the case of the coho salmon, also singled out by EPIC, the EIS/EIR found that the Aquatics Conservation Plan in the HCP would fully mitigate impacts on that species. Therefore, because the EIS/EIR was for an HCP the purpose of which was to mitigate the effect of Pacific Lumber’s activities on wildlife to a less than significant level, it was not error for the EIS/EIR to conclude that HCP did not create significant wildlife impacts. We therefore find no merit in EPIC’s argument that the CEQA findings were inadequate. 2. Cumulative Impacts EPIC contends that the EIS/EIR failed to analyze or address the project’s cumulative impacts to the marbled murrelet, northern spotted owl, and coho salmon by failing to identify past projects, including Pacific Lumber’s *523 previous intensive logging. We conclude that on the record before us, EPIC has failed to identify prejudicial error. Public Resources Code section 21083, subdivision (b), provides that the CEQA guidelines prepared by the Office of Planning and Research should address a situation in which “[t]he possible effects of a project are individually limited but cumulatively considerable. As used in this paragraph, ‘cumulatively considerable’ means that the incremental effects of an individual project are considerable when viewed in connection with the effects of past projects, the effects of other current projects, and the effects of probable future projects.” (Pub. Resources Code, § 21083, subd. (b)(2), italics added.) Pursuant to this statutory mandate, the Office of Planning and Research has promulgated section 15130 of the CEQA Guidelines, which states in subdivision (b), in pertinent part: “The discussion of cumulative impacts shall reflect the severity of the impacts and their likelihood of occurrence, but the discussion need not provide as great detail as is provided for the effects attributable to the project alone. The discussion should be guided by the standards of practicality and reasonableness .... The following elements are necessary to an adequate discussion of significant cumulative impacts: [f] (1) Either: [][] (A) A list of past, present, and probable future projects producing related or cumulative impacts, including, if necessary, those projects outside the control of the agency, or [f] (B) A summary of projections contained in an adopted general plan or related planning document, or in a prior environmental document which has been adopted or certified, which described or evaluated regional or areawide conditions contributing to the cumulative impact. Any such planning document shall be referenced and made available to the public at a location specified by the lead agency.” The CEQA Guidelines further elaborate on the use of prior environmental documents in section 15130, subdivision (d): “Previously approved land use documents such as general plans, specific plans, and local coastal plans may be used in cumulative impact analysis. A pertinent discussion of cumulative impacts contained in one or more previously certified EIRs may be incorporated by reference pursuant to the provisions for tiering and program EIRs. No further cumulative impacts analysis is required when a project is consistent with a general, specific, master or comparable programmatic plan where the lead agency determines that the regional or areawide cumulative impacts of the proposed project have already been adequately addressed, as defined in section 15152[, subdivision] (f), in a certified EIR for that plan.” DEG and Pacific Lumber concede that there is no “list of past, present, and probable future projects producing related or cumulative impacts.” They contend that they employed the second approach to cumulative impacts: “A *524 summary of projections contained in an adopted general plan or related planning document. . . .” (CEQA Guidelines, § 15130, subd. (b)(1)(B).) They do not identify any specific document or documents containing the information about cumulative impacts. Rather, they contend that the EIS/EIR itself has an adequate analysis of “current population status” of the various species that “necessarily entails consideration of the effects of past projects.” Thus, although the EIS/EIR does not refer to earlier planning documents, it contains within itself a great deal of information regarding current conditions of critical species and their habitat equivalent to what would be contained in general plans or similar planning documents. As DFG explained at oral argument, the SYP/HCP for which the EIS/EIR was prepared was the first master planning document for Pacific Lumber’s holdings, and so no previous planning document could be relied on in making its projections. EPIC argues that the lack of discussion of past projects means that the EIS/EIR ignores the reality that logging, and in particular logging by Pacific Lumber, is responsible for the substantial loss of suitable habitat for various species. EPIC’s argument is that placing current population and habitat conditions in the historical context of Pacific Lumber’s and other timber companies’ role in causing those conditions puts that information, and information regarding projections of future habitat and population loss, in a different perspective. Inasmuch as an EIS/EIR is primarily an informational document (see Pub. Resources Code, § 21000), the public and the decision makers informed by that document would be more critical of Pacific Lumber’s planned logging activities, and more skeptical of the probable success of its mitigation activity, were it informed in the EIS/EIR of the extent of Pacific Lumber’s and other timber companies’ responsibility for current environmental conditions. We agree with EPIC that the statutory injunction to assess “the incremental effects of an individual project ... in connection with the effects of past projects, the effects of other current projects, and the effects of probable future projects” (Pub. Resources Code, § 21083, subd. (b)(2), italics added) signifies an obligation to consider the present project in the context of a realistic historical account of relevant prior activities that have had significant environmental impacts. Such historical accounting assists, for example, in understanding development trends. (See Governor’s Off. of Planning & Research, General Plan Guidelines (1990) pp. 44-46 [need to understand population, environmental and economic trends, including historical data, to guide development].) This historical information also may help to identify previous activities that have caused intensive environmental impacts in a given area, the full effects of which may not yet be manifested, thereby disclosing potential environmental vulnerabilities that would not be revealed merely by cataloging current conditions. (See Environmental Protection *525 Information Center v. Johnson, supra, 170 Cal.App.3d at p. 624 [analysis of past clearcutting may reveal extent of present danger of hillside erosion].) We review an agency’s decision regarding the inclusion of information in the cumulative impacts analysis under an abuse of discretion standard. “The primary determination is whether it was reasonable and practical to include the projects and whether, without their inclusion, the severity and significance of the cumulative impacts were reflected adequately.” (Kings County Farm Bureau v. City of Hanford (1990) 221 Cal.App.3d 692, 723 [270 Cal.Rptr. 650].) Although courts have grappled with the abuse-of-discretion issue with respect to the inclusion of pending and possible future projects (see id. at pp. 721-724), none have addressed the adequacy of an analysis of projects that have already been completed. As the above discussion indicates, an EIS/EIR must reasonably include information about past projects to the extent such information is relevant to the understanding of the environmental impacts of the present project considered cumulatively with other pending and possible future projects. Although such historical context is somewhat muted in the EIS/EIR, it is present to some degree. The EIS/EIR does acknowledge population declines and degradation of habitat, including increased water temperature and sediment buildup in streams and loss of habitat for various species. For example, the EIS/EIR contains detailed information about the current population and distribution and loss of suitable habitat for the marbled murrelet, the northern spotted owl, and the coho salmon. The report also acknowledges, albeit somewhat obliquely, that past logging practices are at least in part responsible for this loss and degradation. EPIC argues in effect that the EIS/EIR substantially understates the effects of past timber harvest practices on various species, and that a more realistic account of those effects can be found in various public comments made to the draft EIS/EIR and draft SYP/HCP. As noted, the discussion of cumulative impacts should be guided by the standards of practicality and reasonableness. Although there are conflicting views about whether the EIS/EIR’s discussion of past logging activity was adequate, on the record before us we cannot say that this discussion of the effects of previous logging activity was unreasonable. EPIC also claims that the EIS/EIR fails to consider cumulative impacts of future activities in the marbled murrelet conservation areas. These are the dozen or so areas of marbled murrelet habitat ranging from 300 to 1,400 acres that are protected for the most part from logging and certain other activities and in which various conservation activities will occur. Such activities will be implemented in consultation with, and reviewed by, DEG and the United States Fish and Wildlife Service. *526 EPIC contends that notwithstanding these restrictions, there is no cumulative assessment of the impact of the activities that will be taking place within these areas, including some mining and road construction. We disagree. Given the extensive analysis of the impacts of the project on the marbled murrelet and other wildlife noted above, and the adoption of these conservation areas as part of the HCP to mitigate the environmental impacts of Pacific Lumber’s activities, we do not believe that CEQA requires separate cumulative impact analysis in connection with the adoption of these conservation areas. The final EIS/EIR concludes that creating these areas will on balance be beneficial to the marbled murrelet and other wildlife. Absent a successful challenge to this conclusion based on the lack of substantial evidence, a challenge that is not before us, we will defer to the government agencies’ implicit conclusion that no additional environmental analysis of this measure is required. IV. Disposition For the reasons explained above, we conclude: (1) that CDF did not properly approve an identifiable Sustained Yield Plan; (2) that any newly submitted : Sustained Yield Plan must include an adequate analysis of the cumulative impacts of Pacific Lumber’s timber harvesting activities at the individual planning watershed level consistent with the Forest Practice Rules and sufficient to support Pacific Lumber’s long-term sustained yield estimate; and (3) that the Incidental Take Permit was deficient inasmuch as it included “no surprises” clauses inconsistent with Pacific Lumber’s statutory duty to fully mitigate the impacts of its incidental take. As noted in the statement of facts, the trial court issued a peremptory writ of mandate, which among other things set aside the Director of CDF’s approval of the SYP and the DFG’s approval of the state Incidental Take Permit. In conjunction with the issuance of a peremptory writ of mandate, the trial court’s order enjoined logging pursuant to any THP’s approved in reliance on the SYP after June 22, 2003, which is designed to preserve the status quo and balance the hardships. The Court of Appeal reversed the judgment granting the peremptory writ. We therefore reverse the judgment of the Court of Appeal and remand to that court with directions to reinstate the judgment of the trial court insofar as the latter concluded that the SYP and state Incidental Take Permit approvals were invalid, and to remand the matter to the trial court for remediation of these approvals in a manner consistent with the views expressed in this opinion. The question whether the no surprises clauses, to the extent they are unlawful, can be severed, and the rest of the Incidental Take Permit reinstated, was not specifically addressed below. This question should be addressed by the trial court on remand. *527 The parties have not briefed in this court the question of interim remedies. Because this opinion concludes that the SYP was not properly approved, we hold that the interim remedy imposed by the trial court was proper. Arguments about whether the injunction should be modified due to changed circumstances or for any other reason should be addressed to the trial court. In all other respects, we affirm the Court of Appeal judgment, including, inter alia, its rulings that the EIS/EIR and Streambed Alteration Agreement had been properly approved. Each party is to bear its own costs. George, C. J., Kennard, J., Baxter, J., Werdegar, J., Chin, J., and Corrigan, J., concurred. Footnotes [1] We will refer to these and related rules as the “Forest Practice Rules,” and will follow the Court of Appeal’s practice of parenthetically citing them as “FP Rules,” e.g., “(FP Rules, § 1091.1).” All these rules are to be found in title 14 of the California Code of Regulations. [2] California’s Department of Forestry and Fire Protection’s acronym was recently changed from CDF to CAL FIRE. We use CDF here, because that is the acronym the agency was known by during the administrative review process and throughout this litigation. [3] There is potential for confusion in the discrepant numbers that are cited by the parties, and that are found in the record and in the Court of Appeal opinion. We will express numerical values for harvest levels and sustained yield estimates as x mmbf (million board feet) per year, and, when expressing harvest levels for a decade, for example, the first decade, we will use “x mmbf per year on average for the first decade.” [4] The difference between long-term sustained yield and maximum sustained production can be explained as follows. According to Forest Practice Rules section 895.1, long-term sustained yield “means the average annual growth sustainable by the inventory predicted at the end of a 100 year planning period,” in other words, the amount of timber that will be produced in the last decade of the planning horizon in accordance with the projected inventory, growth, and harvest levels. (See FP Rules, § 913.11, subd. (b)(4).) The long-term sustained yield is a means of demonstrating and quantifying that logging activity in the near future will not exhaust the timber supply in the long term. Maximum sustained production, perhaps the core concept of the Forest Practice Act, is in quantifiable terms the average annual projected harvest over any rolling 10-year period (FP *476 Rules, § 1091.45; see Pub. Resources Code, § 4513, subd. (a)) and must be “Consistent with the protection of soil, water, air, fish and wildlife resources.” (FP Rules, § 1091.45, subd. (a).) Maximum sustained production “shall not exceed the long-term sustained yield estimate for a SYP submitter’s ownership.” {Ibid.) Thus, projected maximum sustained production, as an average expected yearly harvest level, will be somewhat lower than the long-term sustained yield estimate. [5] In the record below, harvest levels are sometimes expressed in terms of conifer harvest, which comprises most of the harvest, and sometimes in terms of conifer and hardwood harvest. We will adopt the former practice. [6] We note that SYP’s have not been commonly used and, according to the briefing, the SYP at issue here was only the second one ever done. Consequently, most of the issues raised with respect to SYP’s are ones of first impression. [7] CDF and DFG submitted a common brief, prepared by the Attorney General. For the sake of clarity and convenience, we will attribute a contention or argument to the agency most involved in the regulatory approval being challenged—here, in the case of the SYP, CDF. [8] Pacific Lumber suggests that our standard of review for what constitutes prejudicial error should be particularly deferential in the present case, because the Legislature, through Assembly Bill 1986, the statute authorizing the Headwaters Agreement, in effect endorsed the project. As Pacific Lumber states: “[Assembly Bill] 1986 was adopted after the draft SYP/HCP had been prepared and circulated, and after the essential terms for issuance of the stay permits had been agreed upon. [Assembly Bill] 1986 continued specific legislative authorization of the requirements for final state permits and imposed additional material constraints on [Pacific Lumber’s] timber operations beyond those contained in the draft SYP/HCP.” Although, as will be discussed below, Assembly Bill 1986 is important for resolving some of the legal questions before us, we do not construe such legislation to alter in any way well-established rules regarding the standard of review. The Legislature can, and has, exempted various projects from environmental review. (See, e.g., Pub. Resources Code, §§ 21080.14, 21080.16 [CEQA does not apply to certain seismic retrofit projects].) Assembly Bill 1986 did not exempt Pacific Lumber from any environmental review requirements and, except for certain specific matters discussed below, did not alter the manner in which the various environmental review procedures were to be conducted. [9] We emphasize that the claim here—the failure to consider public comments on the draft SYP—is distinct from the claim that an agency did not adequately respond to such comments. Agencies generally have considerable leeway regarding such response. When an agency adequately addresses an environmental issue in response to one commenter, it may refer to the prior response when addressing other commenters, and a failure to respond to a particular comment is not prejudicial error when the issue raised by the comment is adequately addressed elsewhere. (Twain Harte Homeowners Assn. v. County of Tuolumne (1982) 138 Cal.App.3d 664, 681-685 [188 Cal.Rptr. 233].) The instant case presents the rarer situation of comments not being considered altogether. [10] We note that this case law is consistent with the standard of prejudice found in Public Resources Code section 21005: “(a) The Legislature finds and declares that it is the policy of the state that noncompliance with the information disclosure provisions of this division which precludes relevant information from being presented to the public agency, or noncompliance with substantive requirements of this division, may constitute a prejudicial abuse of discretion within the meaning of Sections 21168 and 21168.5 [regarding actions to set aside CEQA determinations], regardless of whether a different outcome would have resulted if the public agency had complied with those provisions, fj] (b) It is the intent of the Legislature that, in undertaking judicial review pursuant to Sections 21168 and 21168.5, courts shall continue to follow the established principle that there is no presumption that error is prejudicial.” The Court of Appeal in Environmental Protection Information Center, Inc. v. Johnson, supra, 170 Cal.App.3d 604, construed Public Resources Code section 21005’s provision that courts reviewing CEQA decisions “ ‘shall continue to follow the established principle that there is no presumption that error is prejudicial’ ”: “Judicial decisions indicate that the ‘established principle’ in CEQA cases was not one of presumed prejudice from any error, but one involving the determination of prejudice from the violation of a fundamental regulatory provision. Absent additional guidance from the Legislature, and in light of the policy expressed in the cases . . . , we assume that the enactment of section 21005 was simply a reminder of the general rule that errors which are insubstantial or de minimis are not prejudicial.” (170 Cal.App.3d at p. 623, *488 fn. 11.) We note that the Legislature has not amended section 21005 since the above case, except to add a subdivision (c) not relevant to the issue of prejudicial error. (See Stats. 1994, ch. 1230, § 2, p. 7681.) [11] At oral argument, the Steelworkers, in response to a question, denied they were making any such concession. We do not regard this belated, conclusory assertion as sufficient to disavow their earlier position. [12] The Steelworkers point to various documents of legislative history of the Forest Practice Act, which it contends support the proposition that “regional economic vitality” and “employment” are distinct considerations, and that analysis of the latter did not relieve CDF and Pacific Lumber of the obligation to analyze the former. Without disputing the above, we note nonetheless that the analysis of the effects of timber harvesting on employment and on regional economic vitality are interrelated. Nothing in the legislative history persuades us that CDF abused its discretion under the particular circumstances of the present case in not requiring more economic analysis. [13] When the Steelworkers and EPIC make the same or similar arguments, they will be referred to collectively as “petitioners.” [14] The meaning, function and significance of watersheds has been described as follows: “The watershed—an area or region draining into the same watercourse—is the fundamental building block of the landscape, and thus, natural resource systems. Watersheds can be scaled up or down, aggregated or disaggregated, to analyze and address problems or opportunities of *498 varying scope. For example, from the 14,000 square mile watershed of the San Joaquin River, we can focus down to the 700-square-mile Mokelumne River watershed, to the 75-square-mile Middle Fork of the Mokelumne watershed, or to the 22-square-mile Forest Creek watershed. «]- • • HD “Forest watersheds integrate the water quality impacts of land management activities. Sediment generated by land management moves from the hillslopes to the intermittent draws to the small creeks, and on to the main stem of the river. If you want to assess the potential water quality impact of a proposed activity, you must look at the whole watershed—upstream and downstream—to see what’s already being put into the stream system. Add a time dimension to this spatial analysis—what’s been moving through the stream in the recent past, what’s going to be moving through the stream in the future—and you’ve completed, in the professional lingo, a water quality cumulative effects analysis.” (Wilson, Director of CDF, California Watersheds: Natural Resource and Community Integrators (Aug. 1993) CDF Comment <http://www.frap.cdf.ca.gov/publications/califomia_watersheds.html> [as of July 17, 2008].) [15] Forest Practice Rules section 895.1 defines “planning watershed” as “the contiguous land base and associated watershed system that forms a fourth order or other watershed typically 10,000 acres or less in size. Planning watersheds are used in planning forest management and assessing impacts. The Director has prepared and distributed maps identifying planning watersheds plan submitters must use. Where a watershed exceeds 10,000 acres, the Director may approve subdividing it. Plan submitters may propose and use different planning watersheds, with the director’s approval. Examples include but are not limited to the following: when 10,000 acres or less is not a logical planning unit, such as on the Eastside Sierra Pine type, as long as the size in excess of 10,000 acres is the smallest that is practical.” [16] Class I watercourses are those in which fish are continuously, or seasonally present, and class II watercourses are those that contain nonfish aquatic species. [17] NTMP stands for Nonindustrial Timber Management Plan. (FP Rules, § 895.) [18] The state Incidental Take Permit also authorized in advance the take of 13 “unlisted” species should they become listed in the future under CESA. The Court of Appeal held that DFG erred in issuing a permit in advance for unlisted species, concluding that Pacific Lumber must seek new permits if and when the species become listed. It concluded that this provision must be severed from the Incidental Take Permit. Pacific Lumber and DFG do not challenge this ruling. [19] Amici curiae California State Association of Counties and League of California Cities and the California Building Industry Association et al., call our attention to the legislative history of section 2081(b)(2), and argue that this history demonstrates that regulatory assurances were contemplated. We disagree that the legislative history supports their position. Section 2081 was enacted in response to the Court of Appeal opinion in Planning and Conservation League v. Department of Fish and Game (Apr. 10, 1997, A074048), review granted June 18, 1997, S061521, review dism. and cause remanded to the Court of Appeal, First Appellate District, Division Two, Nov. 25, 1997), in which the court held that CESA does not give DFG the authority to issue Incidental Take Permits. (Sen. Com. on Nat. Resources & Wildlife, Analysis of Sen. Bill No. 879 (1997-1998 Reg. Sess.) as amended Sept. 9, 1997, p. 2.) Amici curiae point to a letter from Senator Tom Hayden to Senator Patrick Johnson expressing concern that the bill that eventually became section 2081 “gives unprecedented assurances to private parties limiting their responsibility to mitigate damage to species (their cost is limited to rough proportionality, the mitigation must be economic, it must be assuredly successful, etc.” (Sen. Tom Hayden, letter to Sen. Patrick Johnson, Sept. 3, 1997.) Even assuming that such a letter is relevant to the determination of legislative intent (but see Kaufman & Broad Communities, Inc. v. Performance Plastering, Inc. (2005) 133 Cal.App.4th 26, 37-38 [34 Cal.Rptr.3d 520] [letters to and from individual legislators not judicially noticeable on issue of legislative intent]), it does not advance the argument of amici curiae. *512 The fact that various legislators or environmental groups believed the bill went too far in limiting the obligation of private parties to mitigate impacts on endangered species by imposing proportionality requirements and the like does not mean the Legislature contemplated the kind of categorical assurances included in the current Implementation Agreement. In other words, although the language and legislative history reveals that the Legislature was unquestionably attempting to strike a balance between competing interests in passing section 2081(b)(2), it does not disclose that the regulatory assurances at issue here were part of that balance. [20] We recognize that the HCP also contains various “adaptive management” programs designed to protect wildlife in response to changing circumstances. The relationship between these programs and the no surprises provisions is unclear. The parties may address on remand the extent to which these programs fulfill Pacific Lumber’s obligation to fully mitigate the impact of its take of listed species, notwithstanding the regulatory assurances found in the Incidental Take Permit. [21] As noted, the regulatory assurances provision of the NCCPA in Fish and Game Code section 2820, subdivision (f) was added by statute in 2002. (Stats. 2002, ch. 4, § 2.) It is unclear whether the NCCPA prior to that date impliedly authorized such assurances. This is not a question we need to address here because DFG made the finding that the present HCP did not constitute a natural community conservation plan under the NCCPA as defined at the time *515 of the HCP’s approval in 1999. In any case, the validity of those pre-2002 plans or any regulatory assurances given within them does not appear to be open to question. (See Fish & G. Code, § 2830, subd. (a) [authorizing incidental take pursuant to a natural community conservation plan approved prior to 2002].) [22] Fish and Game Code section 1603, together with the entire statutory scheme for streambed alteration agreements, was substantially amended in 2003. (Stats. 2003, ch. 736, §2.)