OPINION

KENNARD, J.

California's two largest rivers, the Sacramento and the San Joaquin Rivers, meet to form a delta (California Delta or Delta) near the City of Sacramento, and their combined waters, if not diverted, flow through the Delta, Suisun Bay, and San Francisco Bay, to the Pacific Ocean. The flow of water through this region, commonly known as the Bay-Delta, forms the largest estuary on the West Coast of the United States. It is also the hub of California's two largest water distribution systems, supplying drinking water for two-thirds of California's residents and irrigation water for seven million acres of agricultural land.

Competition for the Bay-Delta's resources, pollution of Bay-Delta water, draining and filling of tidal marshes and other wetlands, and diversion of Bay-Delta water for urban and agricultural uses throughout the state have, however, resulted in a decline in Bay-Delta wildlife habitat, the threatened extinction of plant and animal species, an increasing risk of failure of Bay-Delta levees, and degradation of the Bay-Delta as a reliable source of high quality water.

In 1994, to address the Bay-Delta's problems, 18 federal and state agencies formed a consortium, known as CALFED, to design and implement a long-term and comprehensive plan (the CALFED Program or Program), to restore the Bay-Delta's ecological health and to improve management of Bay-Delta water for the various beneficial uses that depend on it. The CALFED Program was intended to reduce conflicts and provide solutions that competing interests could support. Because of the plan's comprehensive and long-range nature, CALFED decided to proceed in stages and to begin by preparing a program environmental impact statement/environmental impact report (EIR; together PEIS/R). Under state law, a program environmental impact report is one that "may be prepared on a series of actions that can be characterized as one large project" and are related in specified ways. (Cal. Code Regs., tit. 14, § 15168, subd. (a).)

Here, we must determine whether, as the Court of Appeal concluded, the final PEIS/R for the CALFED Program (CALFED Final Programmatic EIS/EIR (July 2000)) failed to comply with the California Environmental Quality Act (CEQA; Pub. Resources Code, § 21000 et seq.) because it did not examine in detail a program alternative requiring reduced water exports from the Bay-Delta; because it did not identify with adequate specificity the potential sources of water required for the proposed projects or analyze in sufficient detail the environmental impacts of taking water from those specific sources; and because it did not provide sufficient detail about the proposed "Environmental Water Account" (a specific project within the CALFED Program). Disagreeing with the Court of Appeal, we conclude that the CALFED program environmental impact report is not legally defective in any of these ways.

I. FACTS, BACKGROUND, AND PROCEDURAL HISTORY

California has a long history of conflict over its water resources. "The history of California water development and distribution is a story of supply and demand. California's critical water problem is not a lack of water but uneven distribution of water resources." ( United States v. State Water Resources Control Bd. (1986) 182 Cal.App.3d 82, 98 [ 227 Cal.Rptr. 161].) Approximately 75 percent of the state's natural water runoff occurs north of Sacramento, while about 75 percent of the net water demand, for both agricultural and urban uses, occurs south of Sacramento. (See ibid.) The Bay-Delta has been the focal point of the most ambitious projects to resolve this mismatch of supply and demand.

The Bay-Delta's watershed encompasses 37 percent of the state's surface area, and its average annual inflow is 22 million acre-feet of water, of which 17.9 million acre-feet comes from the Sacramento River region. Covering over 738,000 acres in five counties, the Bay-Delta is a haven for plants, fish, and wildlife, supporting over 750 native and introduced plant and animal species. Home to residential and business communities supported by major transportation networks, the Bay-Delta is also the hub of the state's major water distribution networks. Currently an average of 5.9 million acre-feet of water is exported south each year from the Bay-Delta, of which about 60 percent is taken for agriculture and the remainder for urban uses. Two-thirds of California households receive at least some of their domestic water from the Bay-Delta, and over seven million acres of highly productive land are irrigated from the same source. (See United States v. State Water Resources Control Bd., supra, 182 Cal.App.3d at p. 97.)

As a result of the uneven distribution of water resources in California, the Bay-Delta has long been the focus of competing interests making conflicting demands. As the PEIS/R explains, "conflicting demands have resulted in several resource threats to the Bay-Delta: the decline of wildlife habitat; the threat of extinction of several native plant and animal species; the collapse of one of the richest commercial fisheries in the nation; the degradation of Bay-Delta water quality; the continued land subsidence on Delta islands; and a Delta levee system faced with a high risk of failure." (PEIS/R, supra, Technical Appen., Phase II Rep., p. 11.) The CALFED Program was developed to address these issues and to reduce conflicts in the system. A brief history of Bay-Delta water use and related legal issues will aid in understanding the CALFED Program and the issues presented here.

A. Historical Background of Bay-Delta Water Use and Related Legal Developments

Due to limited water supplies and rapid population growth, the Southern California area began to experience a need for imported water in the early part of the 20th century. In 1928, the Metropolitan Water District of Southern California (Metropolitan) was created to combine the financial resources of cities and communities in Southern California to import water from distant sources. ( Metropolitan Water Dist. v. Imperial Irrigation Dist. (2000) 80 Cal.App.4th 1403, 1415 [ 96 Cal.Rptr.2d 314].) Metropolitan constructed aqueducts to bring water from the Colorado River to Southern California. ( Id. at p. 1417.) In a normal year, California's rights to Colorado River water are limited to 4.4 million acre-feet. By using Nevada's and Arizona's underused entitlements and surplus water, however, California has historically used more than its normal year's entitlement. Because both Arizona and Nevada are approaching full use of their entitlements, California's overuse of the Colorado River cannot continue, and the United States Secretary of the Interior has directed California to devise a plan to live within its annual 4.4 million acre-feet entitlement.

In 1940, the City of Los Angeles obtained a permit to appropriate virtually the entire flow from four of the five streams supplying water to the state's second largest lake, Mono Lake, near the eastern entrance to Yosemite National Park. ( National Audubon Society v. Superior Court (1983) 33 Cal.3d 419, 424 [ 189 Cal.Rptr. 346, 658 P.2d 709].) Mono Lake began shrinking due to this diversion. ( Ibid.) In 1983, this court found that under the public trust doctrine "[t]he state has an affirmative duty to take the public trust into account in the planning and allocation of water resources, and to protect public trust uses whenever feasible." ( Id. at p. 446.) Accordingly, this court issued a decision effectively limiting the amount of water that can be exported from Mono Lake. ( Id. at p. 452.) As a result of that decision, and also to protect affected trout populations (see California Trout, Inc. v. Superior Court (1990) 218 Cal.App.3d 187, 195 [ 266 Cal.Rptr. 788]), the State Water Resources Control Board (SWRCB) has restricted diversions from Mono Lake's tributary creeks.

In 1933, primarily to control flooding in the Central Valley, the California Legislature approved the Central Valley Project (CVP), which is the nation's largest water reclamation project and California's largest water supplier. (See County of San Joaquin v. State Water Resources Control Bd. (1997) 54 Cal.App.4th 1144, 1147 [ 63 Cal.Rptr.2d 277].) Originally a state project, the CVP was turned over to the federal Bureau of Reclamation, which operates the CVP under rights granted by the SWRCB. ( Ibid.; United States v. State Water Resources Control Bd., supra, 182 Cal.App.3d at p. 97.) The CVP annually exports around 3.5 million acre-feet of water from the Bay-Delta and its tributaries and watershed. In 1992, in what was seen as a victory for environmentalists, Congress passed the Central Valley Project Improvement Act (Pub.L. No. 102-575 (Oct. 30, 1992) 106 Stat. 4706), which elevated fish and wildlife protection and restoration to the status of a primary purpose of the CVP, reserved 800,000 acre-feet of CVP water for environmental and wildlife protection purposes, and prohibited new water contracts.

In 1951, the Legislature approved what became known as the State Water Project (SWP), another water storage and delivery system and the other major exporter of Bay-Delta water. ( Planning Conservation League v. Department of Water Resources (2000) 83 Cal.App.4th 892, 898 [ 100 Cal.Rptr.2d 173]; United States v. State Water Resources Control Bd., supra, 182 Cal.App.3d at pp. 99-100.) Construction of the SWP did not begin, however, until the Legislature passed the California Water Resources Development Bond Act (also known as the Burns-Porter Act) (Wat. Code, § 12930 et seq.) in 1959, and the electorate approved the corresponding bond measure in 1960. (See Metropolitan Water Dist. v. Marquardt (1963) 59 Cal.2d 159, 170 [ 28 Cal.Rptr. 724, 379 P.2d 28]; Planning Conservation League, supra, at p. 898; Antelope Valley-East Kern Water Agency v. Local Agency Formation Com. (1988) 204 Cal.App.3d 990, 993 [ 251 Cal.Rptr. 593].) The SWP serves the domestic water needs of approximately two-thirds of all Californians, with Metropolitan receiving about half of the SWP's water delivery. ( Metropolitan Water Dist. v. Imperial Irrigation Dist., supra, 80 Cal.App.4th at pp. 1411, fn. 8, 1418; see also State Water Resources Control Bd. Cases (2006) 136 Cal.App.4th 674, 693 [39 Cal.Rplr.3d 189].) Due to environmental concerns, however, construction of the entire SWP project has never been completed, resulting in the annual delivery of only about half of the 4.2 million acre-feet of water projected. ( Santa Clarita Organization for Planning the Environment v. County of Los Angeles (2007) 157 Cal.App.4th 149, 152 [ 68 Cal.Rptr.3d 449].)

The problem of insufficient water supplies was intensified by persistent drought between 1987 and 1992. ( Santa Clarita Organization for Planning the Environment v. County of Los Angeles, supra, 157 Cal.App.4th at p. 153; Planning Conservation League v. Department of Water Resources, supra, 83 Cal.App.4th 892, 900.) In 1991, the state Department of Water Resources organized a drought water bank to relieve shortages. ( Planning Conservation League, supra, at pp. 900-901.) In 1994, the department renegotiated the allocation of SWP water with urban and agricultural water contractors, resulting in an agreement known as the Monterey Agreement. ( Planning Conservation League v. Department of Water Resources (1998) 17 Cal.4th 264, 267 [ 70 Cal.Rptr.2d 635, 949 P.2d 488]; Planning Conservation League, supra, 83 Cal.App.4th at pp. 901-902.)

In 1978, the SWRCB adopted a water quality control plan for the Delta and Suisun Marsh, which led to years of litigation that ended in 1986 when the Court of Appeal decided that "the Board failed to carry out properly its water quality planning obligations." ( United States v. State Water Resources Control Bd., supra, 182 Cal.App.3d at p. 120.) In 1987, the SWRCB began holding hearings to revise the plan, resulting in a report that was criticized by both the northern and southern regions. In 1991, after revisions were made to address those criticisms, the SWRCB issued a final report that the federal Environmental Protection Agency (EPA) rejected. It was not until 1995 that the SWRCB adopted a final water quality control plan for the Delta. (See State Water Resources Control Bd. Cases, supra, 136 Cal.App.4th at pp. 699-701.)

By 1993, two fish species — the winter-run Chinook salmon and the Delta smelt — were listed as threatened or endangered under the Endangered Species Act of 1973 ( 16 U.S.C. § 1531 et seq.), resulting in further restrictions on the operations of the CVP and the SWP and the amount of water exported from the Delta. ( O'Neill v. U.S. (9th Cir. 1995) 50 F.3d 677, 681.) In 1994, the federal EPA concluded that during the preceding 20 years, largely as a result of water diversions by the CVP and SWP, the Bay-Delta's fish and wildlife resources had "deteriorated drastically." ( 60 Fed.Reg. 4665 (Jan. 24, 1995).) The California Department of Fish and Game reached the same conclusion.

B. The CALFED Program

As noted, in 1994, against a backdrop of mounting concerns over water shortages, the ecological deterioration of Bay-Delta estuary, the decline in water quality, and the risk of levee system failure, eight state agencies and 10 federal agencies with management or regulatory responsibility over the Bay-Delta formed CALFED to develop a long-term solution to the Bay-Delta's problems.

In June 1994, the CALFED agencies signed an agreement, known as the Framework Agreement, in which they pledged to coordinate the operation of the SWP and the CVP, to coordinate implementation of water quality standards, and to develop a process to establish a long-term solution to the problems of ecosystem quality, water quality, water supply reliability, and levee system vulnerability. In December 1994, the CALFED agencies signed a statement of "Principles for Agreement on Bay-Delta Standards" (the Bay-Delta Accord), which contained detailed interim measures for environmental protection and regulatory stability in the Bay-Delta.

The CALFED Program, which was to be administered over the next 30 years, arose out of the Framework Agreement and the Bay-Delta Accord. The PEIS/R describes the CALFED Program as "a general description of a range of actions that will be further refined, considered, and analyzed for site-specific environmental impacts as part of second- and third-tier environmental documents prior to making a decision to carry out these later actions." (PEIS/R, supra, p. 3-5.) The Resources Agency serves as the lead agency under CEQA. The Program is divided into three phases.

1. Phase I

During phase I, which ran from May 1995 to August 1996, CALFED used a series of public workshops to define the Bay-Delta's problems and to develop a range of potential alternative solutions. This process resulted in CALFED's adoption of a mission statement, program objectives, and solution principles, which CALFED used to create and then to narrow a list of program alternatives.

The Program's mission statement reads: "The mission of the CALFED Bay-Delta Program is to develop a long-term comprehensive plan that will restore ecological health and improve water management for beneficial uses of the Bay-Delta system." (PEIS/R, supra, at p. 1-5.)

CALFED identified these four primary objectives for the Program: "(1) ' Ecosystem Quality — Improve and increase aquatic and terrestrial habitats and improve ecological functions in the Bay-Delta to support sustainable populations of diverse and valuable plant and animal species.' [¶] (2) ' Water Supply — Reduce the mismatch between Bay-Delta water supplies and the current and projected beneficial uses dependent on the Bay-Delta system.' [¶] (3) ' Water Quality — Provide good water quality for all beneficial uses.' [¶] (4) " Vulnerability of Delta Functions — Reduce the risk to land use and associated economic activities, water supply, infrastructure, and the ecosystem from catastrophic breaching of Delta levees.'" (PEIS/R, supra, at p. 1-5.)

CALFED determined that "[e]ach of the four primary objectives for the Program . . . must be met to achieve the project purpose." (PEIS/R, supra, p. 1-6.) To this end, each of the alternatives that CALFED examined in the PEIS/R was "designed to meet these objectives in a comprehensive, integrated manner." ( Ibid.) The PEIS/R states: "The purpose of the Program is to develop and implement a long-term comprehensive plan that will restore ecological health and improve water management for beneficial uses of the Bay-Delta system. To practicably achieve this program purpose, CALFED will concurrently and comprehensively address problems of the Bay-Delta system within each of four resource categories: ecosystem quality, water quality, water supply reliability, and levee system integrity. Important physical, ecological, and socioeconomic linkages exist between the problems and possible solutions in each of these categories. Accordingly, a solution to problems in one resource category cannot be pursued without addressing problems in the other resource categories." ( Id. at pp. 1-6 to 1-7.)

CALFED's six solution principles were adopted to "provide an overall measure of the acceptability of alternatives and guide the design of the institutional part of each alternative." (PEIS/R, supra, p. 1-5.) The solution principles are: " Reduce conflicts in the system. Solutions will reduce major conflicts among beneficial uses of water. [¶] Be equitable. Solutions will focus on solving problems in all problem areas. Improvement for some problems will not be made without corresponding improvements for other problems, [¶] Be affordable. Solutions will be implementable and maintainable within the foreseeable resources of the Program and stakeholders. [¶] Be durable. Solutions will have political and economic staying power and will sustain the resources they were designed to protect and enhance. [¶] Be implementable. Solutions will have broad public acceptance and legal feasibility, and will be timely and relatively simple to implement compared with other alternatives. [¶] Pose no significant redirected impacts. Solutions will not solve problems in the Bay-Delta system by redirecting significant negative impacts, when viewed in their entirety, within the Bay-Delta or to other regions of California." ( Ibid.)

Fifty categories of potential action, including hundreds of individual actions within these categories, were identified to achieve the Program's objectives. (PEIS/R, supra, p. 1-13.) These "action categories" became the building blocks of the alternatives — that is, each alternative was a combination of action categories reflecting differing approaches to achieving Program objectives. ( Ibid.) To narrow the alternatives, CALFED defined approaches "to resolve four 'critical conflicts' among beneficial users: fisheries and diversions, habitat and land use/flood protection, water supply availability and beneficial uses, and water quality and land use." ( Id., pp. 1-13 to 1-14.)

The conflict between fisheries and diversions results primarily from fish mortality attributable to water diversions, including direct losses at pumps, reduced survival when young fish are drawn out of river channels into the Delta, and reduced spawning success of adult fish when migratory cues are altered. The conflict between habitat and land use or flood protection arises because Bay-Delta wildlife habitat has been destroyed by land development and the construction of flood control facilities to protect developed land, and because the needs of wildlife habitat now constrain both land development and levee maintenance and sometimes require that agricultural land be dedicated to habitat. The conflict between water supply availability and beneficial use reflects both increased competition among beneficial users and increased conflict between in-stream and out-of-stream needs at particular times within the annual hydrological cycle. Water quality and land use conflict because water returned to the Bay-Delta after urban and agricultural use contains pollutants and contaminants that degrade water quality. (PEIS/R, supra, pp. 1-13 to 1-14.)

The process of narrowing alternatives yielded 32 approaches for resolving these conflicts, which resulted in a list of 100 alternatives that were later reduced to 10. (PEIS/R, supra, pp. 1-14 to 1-15.)

To assess the 10 alternatives, CALFED held eight public meetings, one workshop, and a meeting of the Bay-Delta Advisory Council. Through this public process, CALFED staff identified four common components (water quality, levee system integrity, ecosystem quality, and water use efficiency) and two variable components (storage and conveyance) and determined that each alternative should include each of these components. The staff further determined that the four common components "were necessary in each of the alternatives to achieve the Program's purpose and needed to be composed of the same actions in all alternatives." (PEIS/R, supra, p. 1-16.) Accordingly, the alternatives were structured around the variable components of storage and conveyance, with the description of the common components not varying among the alternatives.

In alternative number one (existing system conveyance), the Delta's channels would remain in their existing configuration with the addition of some new facilities in the South Delta, including a new pumping station and an inter-tie connecting the SWP and CVP facilities. In alternative number two (modified through-Delta conveyance), North Delta channel modifications, including a diversion facility from the Sacramento River to the Mokelumne River and widening the Mokelumne River channel, would be added to the South Delta alterations contemplated by the first alternative. In alternative number three (dual-Delta conveyance), in addition to many of the modifications contemplated by the first and second alternatives, a new canal or pipeline would be constructed connecting the Sacramento River north of the Delta to the SWP and CVP south of the Delta.

2. Phase II

During phase II, which ran from August 1996 to December 2000, two additional program elements (watershed and water transfer) were added to each alternative, a preferred program alternative was identified, the PEIS/R and the federal Record of Decision (ROD) were certified, and a plan was developed for the ensuing seven years.

In March 1998, CALFED released a first draft PEIS/R evaluating 12 variations or configurations of the three basic alternatives. In June 1999, after further study and public input, CALFED released a second draft PEIS/R that analyzed a preferred alternative employing a through-Delta conveyance with specific facility improvements, and three other alternatives (each with and without additional water storage) that involved few or no facility improvements or sent water around the Delta, and also a "no action" alternative. CALFED held 15 workshops on this draft PEIS/R at which 760 individuals testified, and it considered several thousand letters and postcards.

In July 2000, CALFED issued the final PEIS/R. The preferred alternative and the other alternatives were generally as described in the second draft PEIS/R. On August 28, 2000, the Resources Agency certified that the final PEIS/R complied with CEQA. The CALFED agencies then adopted the ROD for the Program.

3. Phase III

During phase III, the preferred program alternative identified in the final PEIS/R is to be fully implemented. The first seven years of implementation are referred to as "Stage 1 actions." Among the planned Stage 1 actions is implementation of "an Environmental Water Account that acquires water for critical ecosystem and species recovery needs, substantially through voluntary purchases in the water transfer market in its first few years and developing additional assets over time." (PEIS/R, supra, Technical Appen., Implementation Plan, p. 2-8.)

C. The Litigation

In September 2000, a petition for writ of administrative mandate was filed in Sacramento County Superior Court alleging, as here relevant, that the CALFED PEIS/R did not comply with the requirements of CEQA. Filing the action as petitioners were Regional Council of Rural Counties (RCRC), a nonprofit corporation representing 28 (now 30) rural counties; the Central Delta Water Agency, a public agency; the South Delta Water Agency, also a public agency; and three owners of agricultural land in the Delta (R. C. Farms, Inc., Zuckerman-Mandeville, Inc., and Rudy Mussi). Named as respondents were the State of California, the state Resources Agency, the California Environmental Protection Agency, and the secretaries of those agencies. Named as real parties in interest were the United States of America, the state Department of Water Resources, and various heads of federal agencies.

In December 2000, a petition for writ of mandate was filed in Fresno County Superior Court also challenging the CALFED PEIS/R for noncompliance with CEQA. Filing the action as petitioners were the California Farm Bureau Federation (Farm Bureau; a nonprofit corporation representing local farm bureaus, individual farmers, and others with agricultural interests in the state) and three owners of agricultural land in San Joaquin County (Don Laub, Debbie Jacobsen, and Ted Sheely). Named as respondents were the Governor and various CALFED agency officials.

In April 2001, these two actions were coordinated in Sacramento County Superior Court under the title Bay-Delta Programmatic EIR Cases (JCCP No. 4152). (Code Civ. Proc., § 404 et seq.) In April 2003, the trial court ruled that the CALFED PEIS/R satisfied the requirements of CEQA. Separate judgments were then entered denying both petitions. The petitioners in each action appealed, and the Court of Appeal consolidated the appeals.

D. Court of Appeal Opinion and Petitions for Review

In a 224-page opinion, the Court of Appeal reversed the judgments and remanded the matter to the trial court with directions to grant the petitions for writ of mandate vacating both the certification of the PEIS/R and the adoption of the ROD. Although it rejected most of the CEQA challenges, the Court of Appeal concluded that the PEIS/R was defective (1) in failing to discuss an alternative to the CALFED Program requiring reduced water exports from the Bay-Delta, (2) in failing to adequately discuss the environmental impacts of diverting water from various potential sources to meet the Program's goals, and (3) in failing to include certain information relating to the Environmental Water Account.

Four petitions for review were filed, and this court granted each petition.

II. STANDARD OF REVIEW

"Section 21168.5 [of the Public Resources Code] provides that a court's inquiry in an action to set aside an agency's decision under CEQA 'shall extend only to whether there was a prejudicial abuse of discretion. Abuse of discretion is established if the agency has not proceeded in a manner required by law or if the determination or decision is not supported by substantial evidence.' As a result of this standard, 'The court does not pass upon the correctness of the EIR's environmental conclusions, but only upon its sufficiency as an informative document.' [Citation.]" ( Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376, 392 [ 253 Cal.Rptr. 426, 764 P.2d 278], fn. omitted; see also id. at p. 407.) "We may not set aside an agency's approval of an EIR on the ground that an opposite conclusion would have been equally or more reasonable." ( Citizens of Goleta Valley v. Board of Supervisors (1990) 52 Cal.3d 553, 564 [ 276 Cal.Rptr. 410, 801 P.2d 1161] ( Goleta).)

"An appellate court's review of the administrative record for legal error and substantial evidence in a CEQA case, as in other mandamus cases, is the same as the trial court's: The appellate court reviews the agency's action, not the trial court's decision; in that sense appellate judicial review under CEQA is de novo." ( Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova (2007) 40 Cal.4th 412, 427 [ 53 Cal.Rptr.3d 821, 150 P.3d 709].)

III. REDUCED EXPORTS ALTERNATIVE

The Court of Appeal concluded that the CALFED PEIS/R was defective for failing to discuss an alternative to the CALFED Program requiring reduced exports of Bay-Delta water. The Court of Appeal explained: "CALFED appears not to have considered, as an alternative, smaller water exports from the Bay-Delta region which might, in turn, lead to smaller population growth due to the unavailability of water to support such growth. Taking an assumed population as a given and then finding ways to provide water to that population overlooked an alternative that would provide less water for population growth leaving more for other beneficial uses." The Court of Appeal declared that a reduced exports alternative "would also appear to be feasible, at least in the long term as population growth adjusts to the new realities of water availability." Although it conceded that a planned reduction of water exports was inconsistent with the Program's water supply objective, the Court of Appeal agreed with the parties challenging the PEIS/R (the objecting parties) that "CALFED's rejection of a reduced exports alternative is premised on the false assumption that, for an alternative to be feasible, it must meet all of the Program's goals."

For the reasons set forth below, we conclude that the Court of Appeal erred.

The purpose of an EIR is to give the public and government agencies the information needed to make informed decisions, thus protecting "'not only the environment but also informed self-government.'" ( Goleta, supra, 52 Cal.3d at p. 564.) The EIR is the heart of CEQA, and the mitigation and alternatives discussion forms the core of the EIR. ( Ibid.)

The basic framework for analyzing the sufficiency of an EIR's description of alternatives is set forth by the Legislature in CEQA, by the Governor's Office of Planning and Research in the CEQA Guidelines (Cal. Code Regs., tit. 14, § 15000 et seq.), and by this court in Goleta, supra, 52 Cal.3d 553. CEQA requires that an EIR, in addition to analyzing the environmental effects of a proposed project, also consider and analyze project alternatives that would reduce adverse environmental impacts. (Pub. Resources Code, § 21061; see also id., §§ 21001, subd. (g), 21002, 21002.1, subd. (a), 21003, subd. (c); Goleta, supra, 52 Cal.3d at pp. 564-565.) The CEQA Guidelines state that an EIR must "describe a range of reasonable alternatives to the project . . . which would feasibly attain most of the basic objectives of the project but would avoid or substantially lessen any of the significant effects of the project. . . ." (Cal. Code Regs., tit. 14, § 15126.6, subd. (a).) An EIR need not consider every conceivable alternative to a project or alternatives that are infeasible. ( Ibid.; see also Goleta, supra, at p. 574.)

"In determining the nature and scope of alternatives to be examined in an EIR, the Legislature has decreed that local agencies shall be guided by the doctrine of 'feasibility.'" ( Goleta, supra, 52 Cal.3d at p. 565.) CEQA defines "feasible" as "capable of being accomplished in a successful manner within a reasonable period of time, taking into account economic, environmental, social, and technological factors." (Pub. Resources Code, § 21061.1; see also Cal. Code Regs., tit. 14, § 15364.)

"There is no ironclad rule governing the nature or scope of the alternatives to be discussed other than the rule of reason." (Cal. Code Regs., tit. 14, § 15126.6, subd. (a).) The rule of reason "requires the EIR to set forth only those alternatives necessary to permit a reasoned choice" and to "examine in detail only the ones that the lead agency determines could feasibly attain most of the basic objectives of the project." ( Id., § 15126.6, subd. (f).) An EIR does not have to consider alternatives "whose effect cannot be reasonably ascertained and whose implementation is remote and speculative." ( Id., § 15126.6, subd. (f)(3).)

The process of selecting the alternatives to be included in the EIR begins with the establishment of project objectives by the lead agency. "A clearly written statement of objectives will help the lead agency develop a reasonable range of alternatives to evaluate in the EIR and will aid the decision makers in preparing findings. . . . The statement of objectives should include the underlying purpose of the project." (Cal. Code Regs., tit. 14, § 15124, subd. (b).)

Here, CALFED identified four objectives and an underlying purpose for the CALFED Program. (See ante, p. 1157.) The four objectives are "(1) ' Ecosystem Quality — Improve and increase aquatic and terrestrial habitats and improve ecological functions in the Bay-Delta to support sustainable populations of diverse and valuable plant and animal species.' [¶] (2) ' Water Supply — Reduce the mismatch between Bay-Delta water supplies and the current and projected beneficial uses dependent on the Bay-Delta system.' [¶] (3) ' Water Quality — Provide good water quality for all beneficial uses.' [¶] (4) ' Vulnerability of Delta Functions — Reduce the risk to land use and associated economic activities, water supply, infrastructure, and the ecosystem from catastrophic breaching of Delta levees.'"

The underlying purpose of the CALFED Program, as noted (see ante, p. 1157), is "to develop and implement a long-term comprehensive plan that will restore ecological health and improve water management for beneficial uses of the Bay-Delta system." The PEIS/R further explains: "In the past two decades, disagreements regarding the use and management of the Delta have increasingly taken the form of protracted litigation and legislative battles. These disagreements have not yielded solutions to the water-related conflicts centering in the Delta. The CALFED Program was established to reduce the conflicts and provide a solution that competing interests could support. . . . Because both of the purposes composing the CALFED mission are essential to the success of the CALFED Program, only alternatives that would both restore ecological health and improve water management for beneficial uses of the Bay-Delta system were carried forward for detailed consideration." (PEIS/R, supra, p. 1-13, italics added.) Accordingly, the PEIS/R describes its integrated approach to achieving all four objectives concurrently as "the very foundation of the Program." ( Id., Technical Appen., Response to Comments (vol. I), p. IA-2-2.) Nothing less can achieve the underlying fundamental purpose of reducing conflicts by providing a solution that competing interests can support.

During phase I, CALFED studied the feasibility of limiting or reducing exports of Bay-Delta water by means of a "demand reduction approach." Techniques considered for reducing demand included water conservation, water reclamation, water pricing, and retirement and fallowing of agricultural land. A reduced exports alternative employing these techniques was one of the 10 alternatives submitted for public comment during phase I. (See, ante, p. 1159.) The public comments revealed substantial opposition to the strategy of reducing demand by retiring agricultural land, however, and CALFED staff concluded that actions to implement that strategy would exacerbate rather than reduce the conflicts that the CALFED Program was seeking to address. The strategies of water conservation and reclamation, on the other hand, were so popular that CALFED staff decided that an aggressive water use efficiency program incorporating those strategies should form part of each of the program alternatives.

The PEIS/R summarized this process and explained why a reduced export alternative was not further studied: "A primary objective of the CALFED Program is to reduce the mismatch between Bay-Delta water supplies and current and projected beneficial uses dependent on the Bay-Delta system. The CALFED objectives that were developed to meet this primary objective are described in Section 1.2 in the [PEIS/R]. Among these objectives are to improve export water supplies to help meet beneficial use needs and to improve the adequacy of Bay-Delta water to meet Delta outflow needs. These objectives and the alternatives designed to meet these and other CALFED Program objectives are based on the alternatives and Program goals developed during Phase I. Among these were alternatives that emphasized water use efficiency and de-emphasized or eliminated action to improve export water supplies and improve the adequacy of Bay-Delta water to meet Delta outflow needs. Based on input from public workshops, scoping meetings, the BDAC, and the CALFED agencies, CALFED concluded that these actions would not achieve the primary objective for water supply reliability. Water use efficiency is an important element of the CALFED Program. (See the Water Use Efficiency Program Plan.) However, water use efficiency alone will not suffice to reduce the mismatch between Bay-Delta water supplies and current and projected beneficial uses dependent on the Bay-Delta system. Similarly, an alternative that would achieve water quality objectives by reducing or capping exports would prevent the CALFED Program from achieving its objectives regarding water supply reliability." (PEIS/R, supra, Technical Appen., Response to Comments (vol. I), p. CR-30.)

As the Court of Appeal correctly pointed out, an EIR should not exclude an alternative from detailed consideration merely because it "would impede to some degree the attainment of the project objectives." (Cal. Code Regs., tit. 14, § 15126.6, subd. (b).) But an EIR need not study in detail an alternative that is infeasible or that the lead agency has reasonably determined cannot achieve the project's underlying fundamental purpose. ( Goleta, supra, 52 Cal.3d at p. 574 ["a project alternative which cannot be feasibly accomplished need not be extensively considered"].) In the CALFED program, feasibility is strongly linked to achievement of each of the primary program objectives. Past experience has shown that piecemeal efforts to address the Bay-Delta's problems have failed because those problems are interrelated and because conflicting interest groups and stakeholders can block actions that promote some interests at the expense of others. Accordingly, CALFED determined that the four primary project objectives had to be addressed concurrently in an integrated manner if the project was to be successful and therefore feasible. CALFED determined that a reduced export alternative would seriously compromise the water supply objective, and for this reason would not achieve this basic underlying goal of reducing conflicts and providing a solution that competing interests could support.

Petitioner RCRC disagrees with CALFED's determination and argues in this court that the reduced exports alternative could meet the water supply reliability objective. RCRC maintains that the water supply reliability objective encompassed multiple water supply goals, including in-Delta beneficial use needs and demands by upstream populations and riparian users. It asserts that CALFED never equated water supply reliability with increasing Delta exports, nor did this program objective focus on supplying water to south-of-Delta users. Therefore, according to RCRC, even a reduced export alternative could satisfy the water supply reliability objective if other water supply goals, such as the ones for in-Delta and north-of-Delta uses, were met.

We are not persuaded that improving water supply reliability for in-Delta and north-of-Delta users alone can satisfy the water supply reliability objective. In the PEIS/R, CALFED proposed to reduce the mismatch between water supply and demand by addressing a series of objectives that " collectively reduce the conflict among beneficial water users. . . ." (PEIS/R, supra, p. 1-7, italics added.) These objectives include, among other things, in-Delta beneficial uses as well as improvement of export water supplies. RCRC correctly points out that CALFED's water supply reliability goal does not focus on water supply to export users in the south. Nonetheless, the PEIS/R also clearly states that "improv[ing] export water supplies" ( Ibid.) is one of several objectives that must be collectively met to accomplish the overall water supply reliability goal. Because CALFED's goal of water supply reliability encompasses all beneficial uses of Delta water, it cannot be achieved by an alternative that benefits some groups of water users at the expense of other users.

Therefore, CALFED properly exercised its discretion when it declined to carry the reduced export alternative over for study into the final PEIS/R after concluding that such an alternative would not achieve the CALFED Program's fundamental purpose and thus was not feasible.

Although a lead agency may not give a project's purpose an artificially narrow definition, a lead agency may structure its EIR alternative analysis around a reasonable definition of underlying purpose and need not study alternatives that cannot achieve that basic goal. For example, if the purpose of the project is to build an oceanfront resort hotel ( Goleta, supra, 52 Cal.3d at p. 561) or a waterfront aquarium ( Save San Francisco Bay Assn. v. San Francisco Bay Conservation etc. Com. (1992) 10 Cal.App.4th 908, 924-925 [ 13 Cal.Rptr.2d 117]), a lead agency need not consider inland locations. (See also Sequoyah Hills Homeowners Assn. v. City of Oakland (1993) 23 Cal.App.4th 704, 715 [ 29 Cal.Rptr.2d 182] [lead agency need not consider lower density alternative that would defeat primary purpose of providing affordable housing].)

CALFED's determinations that an integrated solution was necessary to the success of the program, and that the water supply objective could not feasibly be achieved with a reduced exports alternative, are supported by substantial evidence and consistent with the rule of reason. Because each alternative included in the PEIS/R (except the no action alternative) requires balanced progress in achieving each of the four primary objectives, improvement in water supply (which may entail increased exports) will occur only if accompanied by improvement in Bay-Delta ecosystem restoration. The program design thus includes a built-in safeguard to ensure that any increase in exports does not result in further deterioration of the Bay-Delta's ecological health. As the PEIS/R explains, "Improvements in ecosystem health will reduce the conflict between environmental water use and other beneficial uses, and allow more flexibility in water management decisions." (PEIS/R, supra, p. 2-7.)

The Court of Appeal erred also in failing to sufficiently distinguish between preexisting environmental problems in the Bay-Delta, on the one hand, and adverse environmental effects of the proposed CALFED Program. Under CEQA, the range of alternatives that an EIR must study in detail is defined in relation to the adverse environmental impacts of the proposed project. An EIR must include a description of feasible project alternatives that would substantially lessen the project's significant environmental effects. (Pub. Resources Code, § 21061; Cal. Code Regs., tit. 14, § 15126.6, subds. (d), (f).) The project's environmental effects, in turn, are determined by comparison with the existing "baseline physical conditions." (Cal. Code Regs., tit. 14, § 15125, subd. (a); see County of Amador v. El Dorado County Water Agency (1999) 76 Cal.App.4th 931, 952 [ 91 Cal.Rptr.2d 66].)

Here, the Court of Appeal gave this explanation for its conclusion that the PEIS/R should have included a reduced export alternative: "An alternative with reduced exports of water may well be environmentally superior to one that requires redirection of water from existing streams or construction or expansion of water storage facilities. Water exported south of the Delta must come from sources flowing into the Delta. Where one of the objectives of the ecosystem restoration component of the Program is to increase stream flows for the benefit of fish and wildlife, an alternative that does not require diversion of stream flows into the Delta would obviously benefit the environment. And, for the reasons stated earlier, an alternative that does not require construction or expansion of reservoirs will avoid the negative environmental impacts of dam construction."

The main thrust of the Court of Appeal's reasoning was that reducing Bay-Delta water exports would "be environmentally superior" because it would facilitate achievement of the ecosystem restoration component of the CALFED Program and thereby more effectively address the Bay-Delta's existing environmental problems. But those problems would continue to exist even if there were no CALFED program, and thus under CEQA they are part of the baseline conditions rather than program-generated environmental impacts that determine the required range of program alternatives.

Insofar as the Court of Appeal identified "expansion of water storage facilities" by means of "dam construction" as a source of negative environmental effects resulting from the CALFED Program, the PEIS/R provided a reasonable range of alternatives to avoid or substantially lessen those effects. The PEIS/R analyzed each of the program alternatives both with and without additional storage. Although the PEIS/R did not analyze a reduced exports alternative, it did analyze no-additional-storage alternatives that would avoid any adverse environmental consequences of constructing new dams or enlarging existing ones. Under CEQA, this was sufficient. (Pub. Resources Code, § 21061; Cal. Code Regs., tit. 14, §§ 15125, subd. (a), 15126.6, subds. (d), (f).) Also, as explained above, the decision to concurrently pursue each of CALFED Program's objectives means that no additional storage will be built, no new stream diversions will occur, and Bay-Delta water exports will not increase, unless accompanied by measurable progress in restoring the Bay-Delta ecosystem.

As the CALFED PEIS/R itself recognizes, Bay-Delta ecosystem restoration to protect endangered species is mandated by both state and federal endangered species laws, and for this reason water exports from the Bay-Delta ultimately must be subordinated to environmental considerations. The CALFED Program is premised on the theory, as yet unproven, that it is possible to restore the Bay-Delta's ecological health while maintaining and perhaps increasing Bay-Delta water exports through the CVP and SWP. If practical experience demonstrates that the theory is unsound, Bay-Delta water exports may need to be capped or reduced. At this relatively early stage of program design, however, we conclude that CALFED properly applied the rule of reason when it decided to consider in the PEIS/R only alternatives that have the potential to both achieve ecosystem restoration goals and meet current and projected water export demands, and that will provide balanced progress in all four of the program areas. Failure to include a reduced exports alternative thus was not an abuse of discretion.

IV. PROGRAM WATER SOURCE IDENTIFICATION

The Court of Appeal found the CALFED PEIS/R lacking in sufficient detail regarding the sources of water that would be used to implement the CALFED Program. The court asserted that "[i]n light of the overarching importance of water to the success of the CALFED Program, merely listing potential sources of water, indicating that the ultimate source determination will be made later, and deferring CEQA analysis of the need to provide water to the Program violates the PEIS/R's basic informational purpose. 'Water is too important to receive such cursory treatment.' [Citation.]"

Although the Court of Appeal conceded that "[t]he PEIS/R may not be able to provide a precise determination of the sources for Program water," it concluded that "the PEIS/R must include an analysis of the impacts of supplying such water, from whatever source." The court stated: "CALFED has approved a Program requiring large amounts of water to fulfill its objectives without analyzing the environmental impacts of supplying such water."

We conclude that the Court of Appeal erred on both points — the need to more specifically identify potential water sources and the need for additional analysis of the impacts of supplying water from each identified potential source. As we explain, CALFED's PEIS/R is a first-tier program EIR, and CEQA does not mandate that a first-tier program EIR identify with certainty particular sources of water for second-tier projects that will be further analyzed before implementation during later stages of the program. Rather, identification of specific sources is required only at the second-tier stage when specific projects are considered. Similarly, at the first-tier program stage, the environmental effects of obtaining water from potential sources may be analyzed in general terms, without the level of detail appropriate for second-tier, site-specific review. The CALFED PEIS/R satisfies these requirements.

A program EIR, as noted, is "an EIR which may be prepared on a series of actions that can be characterized as one large project" and are related in specified ways. (Cal. Code Regs., tit. 14, § 15168, subd. (a).) An advantage of using a program EIR is that it can "[a]llow the lead agency to consider broad policy alternatives and program wide mitigation measures at an early time when the agency has greater flexibility to deal with basic problems or cumulative impacts." ( Id., § 15168, subd. (b)(4).) Accordingly, a program EIR is distinct from a project EIR, which is prepared for a specific project and must examine in detail site-specific considerations. ( Id., § 15161.)

Program EIR's are commonly used in conjunction with the process of tiering. (See Laurel Heights Improvement Assn. v. Regents of University of California, supra, 47 Cal.3d at p. 399, fn. 8.) Tiering is "the coverage of general matters in broader EIRs (such as on general plans or policy statements) with subsequent narrower EIRs. . . ." (Cal. Code Regs., tit. 14, § 15385.) Tiering is proper "when it helps a public agency to focus upon the issues ripe for decision at each level of environmental review and in order to exclude duplicative analysis of environmental effects examined in previous environmental impact reports." (Pub. Resources Code, § 21093, subd. (a); see also Cal. Code Regs., tit. 14, § 15385, subd. (b).)

In addressing the appropriate amount of detail required at different stages in the tiering process, the CEQA Guidelines state that "[w]here a lead agency is using the tiering process in connection with an EIR for a large-scale planning approval, such as a general plan or component thereof . . ., the development of detailed, site-specific information may not be feasible but can be deferred, in many instances, until such time as the lead agency prepares a future environmental document in connection with a project of a more limited geographic scale, as long as deferral does not prevent adequate identification of significant effects of the planning approval at hand." (Cal. Code Regs., tit. 14, § 15152, subd. (c).) This court has explained that "[t]iering is properly used to defer analysis of environmental impacts and mitigation measures to later phases when the impacts or mitigation measures are not determined by the first-tier approval decision but are specific to the later phases." ( Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova, supra, 40 Cal.4th at p. 431.)

The text of the CALFED PEIS/R itself explains its scope and purpose in the tiering scheme: "The Program currently consists of multiple possible actions that are diverse, geographically dispersed, and described in general terms. . . . [¶] [The PEIS/R] provides a broad and comprehensive overview of the potential actions that could be taken by the Program. It describes, in a broad sense, the overall and long-term environmental consequences of all the potential proposed actions at the end of the Program's 30-year time span. This [PEIS/R] is structured to be used as a tiering document. Individual, second-tier projects can use this analysis as a basis from which to supplement and refine the level of detail and can incorporate by reference relevant provisions in the [PEIS/R], such as the cumulative impacts." (PEIS/R, supra, p. 4-2.) Because it is a first-tier, program EIR, the CALFED PEIS/R "does not analyze site-specific impacts of future projects at proposed locations." ( Id., p. 3-5.)

Consistent with its function as a first-tier document, the PEIS/R identifies potential sources of water — including purchases from willing sellers, water conservation by agricultural and urban users, and new or expanded surface or underground storage — that will be needed for the CALFED Program's components, and it includes tables estimating, for each program alternative, potential water acquisitions from willing sellers along various rivers in the Program areas. Further, the PEIS/R addresses the significant impacts of taking water from the identified potential sources in discussing the anticipated environmental effects of the Program's common components. Environmental impacts of the Program are analyzed in the PEIS/R by resource topic (including water supply and water management, water quality, fisheries and aquatic systems, vegetation and wildlife, agricultural land and water use). These impacts are then discussed in general terms for the five CALFED geographic regions (the Sacramento Valley, the San Joaquin Valley, the Delta, the San Francisco Bay region, and Southern California). Although it does not identify specific future water sources with certainty, the PEIS/R does evaluate in general terms the potential environmental effects of supplying water from potential sources. This was sufficient.

To support its conclusion that the PEIS/R lacked appropriate detail regarding the sources of water that would be used to implement the CALFED Program, and the environmental effects of obtaining water from those sources, the Court of Appeal relied on Stanislaus Natural Heritage Project v. County of Stanislaus (1996) 48 Cal.App.4th 182 [ 55 Cal.Rptr.2d 625], and Santiago County Water Dist. v. County of Orange (1981) 118 Cal.App.3d 818 [ 173 Cal.Rptr. 602]. In each of those decisions, an EIR for a project was found defective for failing to identify the source of water needed for the project and the environmental effects of obtaining the needed water. ( Santiago, supra, at p. 829; Stanislaus, supra, at pp. 205-206.) Unlike the CALFED Program at issue here, however, those projects involved proposed commercial land developments, with readily quantifiable water requirements, on identified sites. ( Santiago, supra, at p. 822 [sand and gravel mining operation]; Stanislaus, supra, at p. 186 ["29,500-acre, 5,000-residential-unit destination resort and residential community"].) Although the project in Stanislaus was to be developed "in four overlapping phases over twenty-five years" ( Stanislaus, supra, at p. 188), it was in no relevant sense comparable to the broad, general, multiobjective, policy-setting, geographically dispersed CALFED Program.

More relevant here is Rio Vista Farm Bureau Center v. County of Solano 5 Cal.App.4th 351 [ 7 Cal.Rptr.2d 307], which concerned the validity of a final EIR for a county's hazardous waste management plan. ( Id. at p. 362.) The plan did not select any specific sites for hazardous waste disposal facilities, but instead merely designated certain areas within the county as being potentially consistent with stated criteria for such a facility. ( Id. at p. 364.) At issue was whether the EIR was defective for failing to provide a sufficient project description or to sufficiently analyze the environment impacts of, possible mitigation measures for, and project alternatives to, constructing hazardous waste disposal facilities at identified potential sites. ( Id. at p. 369.) Rejecting the claim, the Court of Appeal stated: "The flaw in appellant's argument is that the Plan makes no commitment to future facilities other than furnishing siting criteria and designating generally acceptable locations. While the Plan suggests that new facilities may be needed by the County, no siting decisions are made; the Plan does not even determine that future facilities will ever be built. Both the Plan and the [final EIR] consistently state that no actual future sites have been recommended or proposed. For that reason, the [final EIR] is intended to be a 'program EIR' or 'tiering EIR,' with subsequent 'project EIR's' to follow in the event specific, identified facilities are proposed in the future." ( Id. at p. 371, fn. omitted.) The Court of Appeal added: "Where, as here, an EIR cannot provide meaningful information about a speculative future project, deferral of an environmental assessment does not violate CEQA." ( Id. at p. 373.) The Rio Vista court concluded: "Considering the speculative nature of any secondary effects from an uncertain future facility, which will be subject to its own separate environmental review, we conclude that no further findings on environmental impacts or the rationale for such findings was reasonably required from the [final EIR]." ( Id. at p. 375.)

Similarly here, the description of potential water sources for the CALFED Program's future projects and the environmental effects of obtaining water from those sources must be appropriately tailored to the current first-tier stage of the planning process, with the understanding that additional detail will be forthcoming when specific second-tier projects are under consideration. (See Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova, supra, 40 Cal.4th at p. 434 ["the burden of identifying likely water sources for a project varies with the stage of project approval involved"].) The PEIS/R has complied with this requirement.

The CALFED Program is to be implemented over a 30-year period and the sources of water actually used depend on future decisions between willing buyers and sellers. It is therefore impracticable to foresee with certainty specific sources of water and their impacts. Furthermore, water supply plans must remain flexible as they are subject to changing conditions, such as changes in population projections, demographics, new or revised environmental restrictions, pollution of sources, or water supply effects from prolonged droughts. As a result, one cannot be certain that a particular future water source identified at the first-tier stage will ever materialize, or that the source will even be suitable 10 or 20 years later as changed conditions may make another source more advantageous.

Given the uncertain nature of water acquisitions over a 30-year period, the PEIS/R provided region-by-region analysis of the general impacts of water acquisitions. For example, it noted that acquiring water from agricultural lands for ecosystem restoration would likely result in significant fallowing of the land or shifting of crops. Such region-by-region identification of potential impacts allows decision makers to intelligently consider the consequences of water acquisitions before approving it, while leaving more site-specific details for later project-level EIR's. (See Cal. Code Regs., tit. 14, §§ 15151, 15152.)

The purpose of tiering is to allow a lead agency to focus on decisions ripe for review. (Pub. Resources Code, § 21093, subd. (a); Cal. Code Regs., tit. 14, § 15385, subd. (b).) An agency that chooses to tier may provide analysis of general matters in a broader EIR, then focus on narrower project-specific issues in later EIR's. (Cal. Code Regs., tit. 14, § 15152, subd. (a).) Future environmental documents may incorporate by reference general discussions from the broader EIR, but a separate EIR is required for later projects that may cause significant environmental effects inadequately addressed in the earlier report. ( Id., § 15152, subds. (a), (f).)

The PEIS/R complied with CEQA by identifying potential sources of water and analyzing the associated environmental effects in general terms. The level of detail contained in the PEIS/R's impact analysis was consistent with its first-tier programmatic nature. Although later project-level EIR's may not simply tier from the PEIS/R analysis and will require an independent determination and disclosure of significant environmental impacts (see Cal. Code Regs., tit. 14, § 15152, subd. (f)), this stage of program development did not require a more detailed analysis of the Program's future water sources, nor did it appear practicable. By compelling CALFED at the first-tier stage to provide greater detail about potential sources of water for second-tier projects, the Court of Appeal's decision undermined the purpose of tiering and burdened the program EIR with detail that would be more feasibly given and more useful at the second-tier stage. Such details were properly deferred to the second-tier of the CALFED Program, when specific projects can be more fully described and are ready for detailed consideration.

V. ENVIRONMENTAL WATER ACCOUNT

The Environmental Water Account (EWA) is a second-tier project that the CALFED agencies proposed in conjunction with the ecosystem restoration program. The EWA is based on the idea that "flexible management of water operations" could effectively balance the competing demands of fishery, restoration and recovery needs with the need to improve supply reliability and quality for water users. Without reducing deliveries to water users, the EWA provides water for fish by authorizing the state and federal governments to acquire, bank, transfer and borrow water beyond that available through existing regulatory actions.

CALFED disclosed the idea of an EWA in its December 1998 revised phase II report, released about nine months after CALFED's first draft PEIS/R. The revised phase II report explained a proposal for how an EWA might operate. It set forth the possibility of using "transfers, options and acquisitions" to create a water account that fisheries could draw on to provide additional protection. CALFED's final PEIS/R also addressed the EWA in general terms, but the Court of Appeal held that the PEIS/R provided inadequate detail on the EWA.

The Court of Appeal emphasized the PEIS/R's omission of certain additional details regarding the EWA's anticipated initial project-level actions. These details were contained in a document entitled "California's Water Future: A Framework for Action" (Action Framework) that was released shortly before certification of the PEIS/R. The CALFED agencies had continued to refine this proposed project during the time between the release of the draft PEIS/R and the final PEIS/R. The Action Framework differed from CALFED's prior EWA disclosures primarily by specifying actual sources for the EWA's initial assets. These sources included water available from "State Water Pumping of (b)(2)/ERP Upstream Releases," "Export/Inflow Ratio Flexibility" and water purchases north of the Delta and south of the Delta.

The Court of Appeal found this information should have been included in the PEIS/R: "Use of a programmatic EIR is not an excuse to defer analysis of the significant impacts of the program. (Guidelines [Cal. Code Regs., tit. 14], § 15152, subd. (b).) To the extent CALFED is able to resolve issues regarding the structure of the EWA before the PEIS/R is issued, that information should be disclosed in the PEIS/R." The State of California argued that this information was more appropriately included in a project-level EIR and that "CALFED should not 'be faulted for providing other agency decision makers and members of the public with as much information as possible about the developing concept of an EWA, how it related to the CALFED plan, and how a second-tier EWA project would be structured and implemented during the first seven years after the ROD [Record of Decision].'"

Under CEQA's tiering principles, it is proper for a lead agency to use its discretion to focus a first-tier EIR on only the general plan or program, leaving project-level details to subsequent EIR's when specific projects are being considered. (See Cal. Code Regs., tit. 14, § 15152, subd. (b).) This type of tiering permits a lead agency to use a first-tier EIR to adequately identify "significant effects of the planning approval at hand" while deferring the less feasible development of detailed, site-specific information to future environmental documents. (See id., § 15152, subd. (c).) In determining the adequacy of an EIR, the CEQA Guidelines look to whether the report provides decision makers with sufficient analysis to intelligently consider the environmental consequences of a project. (Cal. Code Regs., tit. 14, § 15151.) The CEQA Guidelines further provide that "the sufficiency of an EIR is to be reviewed in the light of what is reasonably feasible. . . . The courts have looked not for perfection but for adequacy, completeness, and a good faith effort at full disclosure." (Cal. Code Regs., tit. 14, § 15151.) The Court of Appeal erred in finding the CALFED PEIS/R's level of analysis of the EWA inadequate because (1) the PEIS/R adequately identified the significant environmental effects of an EWA, and (2) EWA details disclosed in the Action Framework properly belonged in a second-tier CEQA document.

The PEIS/R fulfills the function of a first-tier document because it analyzes the environmental impacts of the mechanisms that will establish and develop the EWA — water transfers (including purchases from willing sellers), reservoirs, groundwater storage, and more flexible operations of water projects. Unlike a project with localized or site-specific water requirements, the EWA is based on flexible water management and its water assets may exist anywhere in the state. The EWA's water requirements are met through the establishment of what CALFED's 1998 revised phase II report describes as a "portfolio of assets." This portfolio of assets may include "water, entitlement to capacity in water diversion facilities, aqueducts, storage and money." Also, "an EWA could use transfers, options and acquisitions to obtain water" and "[v]ariances in export standards could be granted in the interest of generating additional EWA water." As the Court of Appeal noted, CALFED's response to comments in the PEIS/R further explains that many of the initial EWA assets "will come from access to existing [p]roject flexibility, new changes in project flexibility (for example, joint point of diversion and export/inflow ratio flexibility) and through voluntary purchases . . . on the water transfer market." (PEIS/R, supra, Technical Appen., Response to Comments (vol. I), p. CR-80.)

Although the Farm Bureau recognized the "numerous distinct and separate water acquisition projects" necessary to develop EWA assets, it nonetheless argued that the PEIS/R was additionally required to provide a "big-picture" impact analysis of EWA's overall need to acquire one million acre-feet of water. What the Farm Bureau fails to take into account is that the impacts of the separate water acquisition projects used to develop EWA assets are the impacts of the program. Because the EWA is not a localized program, it is not reasonably feasible to require quantification of the "big picture" impacts of its water needs. Impacts result from each water acquisition project launched to develop EWA assets. Therefore, when the PEIS/R analyzed the impacts of using water transfers, groundwater storage, and other mechanisms to develop the EWA, it did not unjustifiably defer analysis of the significant impacts of the EWA, but rather it adequately identified them. (See Cal. Code Regs., tit. 14, § 15152, subd. (c).) Such analysis allows decision makers to intelligently consider the environmental consequences of an EWA before approving it. (See id., § 15151.)

The Court of Appeal also erred when it held that specific EWA details in the Action Framework that preceded the PEIS/R certification should have been included in the PEIS/R. The PEIS/R contained a level of detail appropriate to its first-tier, programmatic nature. In determining the degree of specificity required in an EIR, the CEQA Guidelines provide that the "degree of specificity required in an EIR will correspond to the degree of specificity involved in the underlying activity which is described in the EIR." (Cal. Code Regs., tit. 14, § 15146.) For example, an EIR on a construction project will necessarily be more specific than an EIR on the adoption or amendment of a comprehensive zoning ordinance or a local general plan. ( Id., subds. (a), (b).)

The analysis in Al Larson Boat Shop, Inc. v. Board of Harbor Commissioners (1993) 18 Cal.App.4th 729 [ 22 Cal.Rptr.2d 618] ( Al Larson) is instructive for this case. At issue in Al Larson was the propriety of deferring analysis to future project EIR's for a city's port development plan. ( Id. at p. 743.) The plan proposed the use of six anticipated projects to develop the port to meet increased demand for commercial cargo handling. ( Id. at p. 742.) The Long Beach Board of Harbor Commissioners chose, however, to defer approval on specific sites for those six projects to second-tier project EIR's, two of which were considered nearly concurrently with the final first-tier EIR. ( Id. at p. 743.) The Court of Appeal upheld the board's decision to tier, stating: "The concept of tiering supports allowing the agency and the public to first decide whether it is a good idea to increase Port capacity in a given five-year period at all. . . . If that decision is made in the affirmative then each individual project can be reviewed in-depth on its merits in a project EIR. . . ." ( Id. at p. 744.) In Al Larson, the board had committed itself to "'conduct individual environmental assessments in accordance with CEQA on a project-by-project basis for each of the indicated projects.'" ( Id. at p. 742.)

The CALFED program analyzed in the PEIS/R is as broad if not broader in scope than the port development plan analyzed in the first-tier EIR at issue in Al Larson, supra, 18 Cal.App.4th 729. As stated earlier, the text of the CALFED PEIS/R itself states: "The Program currently consists of multiple possible actions that are diverse, geographically dispersed, and described in general terms." Similarly the Action Framework describes the CALFED Program as the "largest, most comprehensive water management program in the world." The 30-year CALFED Program establishes a complex water management system that seeks to concurrently achieve the primary program objectives of ecosystem restoration, water supply reliability, levee system integrity, and water quality improvement. The PEIS/R's description and analysis of the EWA, like the analysis of Program water sources, was therefore appropriately tailored to the first-tier planning stage with its general discussion of the impacts of EWA water transfers, water storage, and flexible operations of water management. (See Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova, supra, 40 Cal.4th at p. 434.)

In contrast with the broad programmatic nature of the PEIS/R, the EWA was designated a second-tier project from its inception. In its 1998 revised phase II report, CALFED set forth a series of actions for implementing Stage 1 (the first seven years) of the Program, making clear that these actions were subject to and could be altered as a result of second-tier environmental review. The EWA is listed as a Stage 1 action. CALFED's 2000 Action Framework reiterated the EWA's status as a second-tier Stage 1 action, dependent upon "CALFED concluding its programmatic environmental review and subsequent site-specific analyses." Thus, CALFED intended the EWA to be a second-tier project, subject to later, project-specific environmental analysis.

CALFED worked out some of the EWA details while it was completing the final PEIS/R, and it properly released those details in the second-tier Action Framework in June 2000, one month before it released the final PEIS/R. The Action Framework set out specific details regarding the EWA project components whose general impacts were analyzed in the PEIS/R. For example, the PEIS/R generally analyzed the impacts of water transfers while the Action Framework specifically established that some initial EWA assets would be acquired through south-of-Delta and north-of-Delta water purchases. These second-tier project details were not, as the Court of Appeal asserted, "significant information" that should have been included in the first-tier, final PEIS/R. The PEIS/R therefore complied with CEQA in analyzing the impacts of the EWA in general terms and deferring project-level details to subsequent project-level EIR's.

VI. CONCLUSION AND DISPOSITION

The Court of Appeal's judgment is reversed.

George, C. J., Baxter J., Werdegar J., Chin J., Moreno J., and Corrigan J., concurred.