Opinion

STEIN, Acting P. J.

On October 20, 1998, the Napa County (the County) Board of Supervisors adopted a resolution certifying a final subsequent *352 environmental impact report (the FSEIR) and adopting an updated specific plan (the Updated Specific Plan) for the development of an unincorporated area surrounding the Napa County Airport. Two citizens groups and the City of American Canyon, a city located within the County, responded to the County’s actions by filing complaints and petitions for writ of mandate challenging the County’s general plan (General Plan), the Updated Specific Plan and the certification of the FSEIR. (The challengers will be referred to, collectively, as Petitioners.) The superior court sustained demurrers to the causes of action challenging the General Plan, ruling that they were time-barred. It later entered judgment granting the petition for writ of mandate, ruling that the FSEIR was inadequate and that the Updated Specific Plan was invalid.

The County has filed an appeal from the judgment. Petitioners have filed a cross-appeal from the order sustaining the demurrer to the causes of action challenging the County’s General Plan.

We do not agree with every detail of the trial court’s rulings, but we will find that the demurrers properly were sustained, the FSEIR was inadequate and the Updated Specific Plan was invalid. We therefore will affirm the judgment, although without completely adopting the trial court’s reasoning.

Background

The Legislature has declared a policy “to protect California’s land resource, to insure its preservation and use in ways which are economically and socially desirable in an attempt to improve the quality of life in California.” (Gov. Code, § 65030.) To further this policy, each of the state’s counties is required to adopt a comprehensive, long-term, general plan for the physical development of that county. (Gov. Code, § 65300.) The county may then, if it chooses, adopt one or more specific plans for the systematic implementation of the general plan for all or part of the area covered by the general plan. (Gov. Code, § 65450.)

In 1986, in accordance with Government Code section 65450, the County adopted a specific plan (the 1986 Specific Plan) to develop the airport industrial area (the Project or Project area). The Project area is made up of approximately 2,945 acres, including the Napa County Airport, immediately south of the City of Napa City. It is bordered by the area’s main north-south transportation route, State Route 29. The area’s main east-west transportation route, Highway 12, crosses State Route 29 and runs through the Project area to the Napa County Airport. In 1986, the area consisted mainly of flat grasslands, crossed by several creeks. It was used primarily for agricultural *353 purposes such as grazing and growing forage crops. The 1986 Specific Plan contemplated developing the area to accommodate 1,923 acres of industrial development, including 1,354 acres designated business/industrial park and 569 acres designated general industrial.

The 1986 Specific Plan was a “project” within the purview of the California Environmental Quality Act (CEQA), Public Resources Code section 21000 et seq. This meant, among other things, that the County was required to prepare and circulate an environmental impact report, or EIR, and was encouraged to hold public hearings prior to certifying the EIR and adopting the 1986 Specific Plan. (Pub. Resources Code, §§ 21065, 21080, subd. (a), 21151.) An EIR was prepared. It identified a number of adverse environmental effects that would result from the adoption of the Project, identified various mitigation measures that might alleviate some or all of those effects, and made recommendations. The County adopted several of the proposed mitigation measures, incorporating them into the 1986 Specific Plan.

The County found, however, that no feasible mitigation measures or alternatives had been proposed that would fully mitigate the adverse effects of the Project on traffic, road access, hydrology, water quality, vegetation, wildlife, land use, and visual, noise and air quality. The County nonetheless certified the EIR, finding that identified economic and social factors justified approval of the Project notwithstanding that the Project would have unmitigated adverse effects on the environment. The County thereafter approved the Project, and adopted the 1986 Specific Plan.

The County began to update the 1986 Specific Plan in 1994. As part of this process, the County caused a draft Updated Specific Plan and a draft subsequent EIR to be prepared and circulated, and again conducted hearings, reviews and related proceedings. After receiving comments and criticisms, the County caused a supplement to the draft subsequent EIR (the Supplement) to be prepared and circulated, and after receiving comments and criticisms to the Supplement, it caused an addendum to the draft subsequent EIR (the Addendum) to be prepared. At the completion of these proceedings, the County, on October 20, 1998, adopted the Updated Specific Plan after certifying the FSEIR, which was comprised of the draft subsequent EIR as modified by the Supplement and the Addendum.

As in 1986, the County found that no feasible mitigation measures or alternatives had been proposed that would fully mitigate the adverse effects of the project. Indeed, the County concluded that a number of the measures approved or adopted as part of the 1986 Specific Plan were not feasible. The *354 County nonetheless certified the FSEIR and adopted the Updated Specific Plan, deleting the mitigation measures adopted in 1986 but found infeasible in 1998. The County found that with the adoption of such mitigation measures as were feasible, the specified impacts would be reduced to a “less than significant” level, and/or that additional mitigation measures were infeasible and the benefits of the Updated Specific Plan sufficiently overrode and outweighed the significant impacts it would have on the environment.

On November, 19, 1998, Napa Citizens for Honest Government and North Bay Citizens for Responsible Transportation filed a complaint and petition for writ of mandate, challenging the FSEIR and Updated Specific Plan. American Canyon was permitted to intervene in the proceedings, and filed its own complaint and petition for writ of mandate challenging the FSEIR and Updated Specific Plan.

The County successfully demurred to both complaints and petitions insofar as they attacked the County’s General Plan, on the basis that such an attack was time-barred. The court, however, granted the petition for writ of mandate, and entered judgment in favor of Petitioners on their remaining claims, finding (1) that the Updated Specific Plan was inconsistent with the General Plan’s circulation element; (2) that the FSEIR failed adequately to analyze identified traffic problems and failed to provide adequate mitigation measures for identified traffic and circulation impacts; (3) that the Updated Specific Plan was inconsistent with the goals and policies of the General Plan’s housing element; (4) that the FSEIR failed adequately to evaluate and mitigate the Updated Specific Plan’s impact on housing; (5) that the FSEIR failed adequately to analyze and mitigate identified significant impacts as to wastewater treatment and water distribution; and (6) that the FSEIR failed to investigate and make findings as to the impact the Updated Specific Plan would have on steelhead trout.

The Appeal

Introduction

Two interrelated bodies of law govern the County’s actions. The first is the state’s planning and zoning laws, Government Code section 65000 et seq., which, as mentioned above, are designed to protect California’s land resource, mandate that a county such as Napa develop and adopt a general plan, and authorize counties to adopt specific plans.

The planning and zoning laws require each general plan to incorporate certain elements including, as relevant here, a land use element, a circulation *355 element (which must be coordinated with the land use element), a housing element, and a conservation element. (Gov. Code, §§ 65302, subds. (b), (c) & (d).) “ ‘The general plan has been aptly described as the “constitution for all future developments” within the city or county. . . . “[T]he propriety of virtually any local decision affecting land use and development depends upon consistency with the applicable general plan and its elements.” . . .’ [Citations.] ‘The consistency doctrine has been described as “the linchpin of California’s land use and development laws; it is the principle which infuse[s] the concept of planned growth with the force of law.” . . .’ [Citation.]” (Families Unafraid to Uphold Rural etc. County v. Board of Supervisors (1998) 62 Cal.App.4th 1332, 1336, [74 Cal.Rptr.2d 1]) (hereafter FUTURE) quoting from Corona-Norco Unified School Dist. v. City of Corona (1993) 17 Cal.App.4th 985, 994 [21 Cal.Rptr.2d 803].)

The Updated Specific Plan, therefore, is valid only to the extent that it is consistent with the County’s General Plan; i.e., to the extent that it is compatible with the General Plan’s objectives, policies, general land uses and programs.

The second body of law at issue is CEQA, a comprehensive scheme designed to provide long-term protection to the environment. (Mountain Lion Foundation v. Fish & Game Com. (1997) 16 Cal.4th 105, 112 [65 Cal.Rptr.2d 580, 939 P.2d 1280].) The Public Resources Code codifies CEQA at section 21000 et seq. Its provisions are supplemented by the CEQA Guidelines, set forth in the California Code of Regulations, title 14, section 15000 et seq. (CEQA Guidelines). 1

“ ‘The foremost principle under CEQA is that the Legislature intended the act “to be interpreted in such manner as to afford the fullest possible protection to the environment within the reasonable scope of the statutory language.” ’ [Citations.] [ft] The EIR has been aptly described as the ‘heart of CEQA.’ [Citations.] Its purpose is to inform the public and its responsible officials of the environmental consequences of their decisions before they are made. Thus, the EIR ‘protects not only the environment but also informed self-government.’ [Citation.]” (Citizens of Goleta Valley v. Board of Supervisors (1990) 52 Cal.3d 553, 563-564 [276 Cal.Rptr. 410, 801 P.2d 1161], fn. omitted.) “ ‘[T]he ultimate decision of whether to approve a project, be that decision right or wrong, is a nullity if based upon an EIR that *356 does not provide the decision-makers, and the public, with the information about the project that is required by CEQA.’ [Citation.] The error is prejudicial ‘if the failure to include relevant information precludes informed decisionmaking and informed public participation, thereby thwarting the statutory goals of the EIR process.’ [Citation.]” (San Joaquin Raptor/Wildlife Rescue Center v. County of Stanislaus (1994) 27 Cal.App.4th 713, 721-722 [32 Cal.Rptr.2d 704].)

The validity of the FSEIR, therefore, depends in large part upon whether it provides the information necessary for the County and the public to understand the nature and environmental consequences of the Project. There is no requirement that the FSEIR itself be consistent with the County’s General Plan, but it is required to identify any inconsistencies between the Project and the General Plan. (CEQA Guidelines, § 15125, subd. (d).)

Finally, although in regulating EIR’s, CEQA describes the information that must be provided before an agency can approve a project, it also, to a limited degree, restricts the power of the agency to approve a project. Public Resources Code section 21002.1, subdivision (b), thus prohibits an agency from approving a project without requiring the implementation of any feasible mitigation measures, providing: “Each public agency shall mitigate or avoid the significant effects on the environment of projects that it carries out or approves whenever it is feasible to do so.” Public Resources Code section 21002, and CEQA Guidelines section 15093, subdivisions (a) and (b), permit an agency to approve a project even though it will have significant impacts on the environment that cannot be fully mitigated, but only if the agency finds that specific economic, legal, social, technological, or other benefits of a proposed project outweigh the unavoidable adverse environmental effects it will have. In the present case, the FSEIR reported that the Project would cause some adverse effects that could not be feasibly mitigated. The County nonetheless adopted the Updated Specific Plan, finding that the significant effects that could not.be mitigated would be outweighed by the Project’s benefits. Petitioners do not claim that this finding was an abuse of discretion; their complaints are with the process leading up to that finding.

Standard of Review

An agency’s certification of an EIR is subject to judicial review, but “[i]n reviewing agency actions under CEQA, Public Resources Code section 21168.5 provides that a court’s inquiry ‘shall extend only to whether there was a prejudicial abuse of discretion. Abuse of discretion is established if the *357 agency has not proceeded in a manner required by law or if the determination or decision is not supported by substantial evidence.’ Thus, the reviewing court “’does not pass upon the correctness of the EIR’s environmental conclusions, but only upon its sufficiency as an informative document.” ’ [Citations.] 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. ‘Our limited function is consistent with the principle that “The purpose of CEQA is not to generate paper, but to compel government at all levels to make decisions with environmental consequences in mind. CEQA does not, indeed cannot, guarantee that these decisions will always be those which favor environmental considerations.” ’ [Citations.] We may not, in sum, substitute our judgment for that of the people and their local representatives. We can and must, however, scrupulously enforce all legislatively mandated CEQA requirements.” (Citizens of Goleta Valley v. Board of Supervisors, supra, 52 Cal.3d at p. 564.)

Similarly, a governing body’s conclusion that a particular project is consistent with the relevant general plan carries a strong presumption of regularity that can be overcome only by a showing of abuse of discretion. “An abuse of discretion is established only if the [governing body] has not proceeded in a manner required by law, its decision is not supported by findings, or the findings are not supported by substantial evidence.” (Sequoyah Hills Homeowners Assn. v. City of Oakland (1993) 23 Cal.App.4th 704, 717 [29 Cal.Rptr.2d 182].)

On appeal, “[i]n applying the substantial evidence standard, ‘the reviewing court must resolve reasonable doubts in favor of the administrative finding and decision.’ ” (Laurel Heights Improvement Assn. v. Regents of University of California, supra, 47 Cal.3d at p. 393.) The role of the appellate court’s rule is precisely the same as the trial court’s, and the lower court’s findings are not conclusive on appeal. (Bowman v. City of Petaluma (1986) 185 Cal.App.3d 1065, 1076 [230 Cal.Rptr. 413].)

Deletion of Mitigation Measures Adopted in 1986

A preliminary question is presented by the trial court’s determination that the County lacked the authority to delete the mitigation measures adopted as part of the 1986 Specific Plan. Logic dictates against a conclusion that a mitigation measure once adopted may not be deleted. A county’s needs necessarily change over time in light of such matters as the development of surrounding communities, state and federal action, and/or natural disasters. It follows that a county must have the power to modify its land use *358 plans as circumstances require. The Government Code recognizes this need and permits counties to modify their general plans if they deem the amendment to be in the public interest, except that (with exceptions not present here) any mandatory element of a general plan may not be modified more frequently than four times in any calendar year. (Gov. Code, § 65358, subds. (a) & (b).) A specific plan may be amended in the same manner as a general plan, except that a specific plan may be amended as often as deemed necessary by the legislative body. (Gov. Code, § 65453, subd. (a).) Of course, the specific plan, as amended, still must be consistent with the general plan (Gov. Code, § 65454), and any EIR submitted in connection with the modified plan still must be adequate, but there is no statutory authority for the proposition that an amendment may not include the deletion of an earlier adopted mitigation measure.

The claim that once a mitigation measure is adopted it never can be deleted is inconsistent with the legislative recognition of the need to modify land use plans as circumstances change. It also is true that mistakes can be made and must be rectified, and that the vision of a region’s citizens or its governing body may evolve over time. In light of all these considerations, we conclude that there are times when mitigation measures, once adopted, can be deleted.

Petitioners cite Rio Vista Farm Bureau Center v. County of Solano (1992) 5 Cal.App.4th 351 [7 Cal.Rptr.2d 307] as providing support for the argument that a mitigation measure, once adopted, cannot be deleted. The scope of the project at issue there was broad and somewhat nebulous, with the result that the county could not formulate mitigation measures with any precision. It was held that certain mitigation measures, although imprecise, were sufficient because “a firm commitment has been made to future mitigation of significant impacts. Where, as here, devising more specific mitigation measures early in the planning process is impractical, ‘ “the agency can commit itself to eventually devising measures that will satisfy specific performance criteria articulated at the time of project approval. . . .” ’ ” (Id. at p. 377.) Rio Vista, accordingly, establishes that a firm commitment to devise an effective mitigation measure can itself be a mitigation measure. It does not, however, establish that a particular mitigation measure, once adopted, is a commitment that may never be modified or deleted.

Petitioners also cite Federation of Hillside & Canyon Associations v. City of Los Angeles (2000) 83 Cal.App.4th 1252 [100 Cal.Rptr.2d 301], where the court recognized that CEQA requires an agency, such as the County here, to take steps to ensure that any mitigation measures “will actually be implemented as a condition of development, and not merely adopted and then *359 neglected or disregarded.” {Id. at p. 1261, italics omitted.) Although the court found that an agency cannot simply ignore mitigation measures required by an EIR, nothing in that case compels the conclusion that a mitigation measure, once adopted, is binding for all time.

In short, we find nothing in established law or in logic to support the conclusion that a mitigation measure, once adopted, never can be deleted. Nonetheless, when an earlier adopted mitigation measure has been deleted, the deference provided to governing bodies with respect to land use planning decisions must be tempered by the presumption that the governing body adopted the mitigation measure in the first place only after due investigation and consideration. We therefore hold that a governing body must state a legitimate reason for deleting an earlier adopted mitigation measure, and must support that statement of reason with substantial evidence. If no legitimate reason for the deletion has been stated, or if the evidence does not support the governing body’s finding, the land use plan, as modified by the deletion or deletions, is invalid and cannot be enforced.

Assuming a valid reason for the deletion is stated, and the evidence supports the governing body’s finding that the stated reason exists, the land use plan, as modified, and the supporting EIR, should be subjected to the same scrutiny as would be given any land use plan and supporting EIR. The fact that a mitigation measure had been adopted in an earlier plan, but has been deleted, will be relevant to the question of the adequacy of the modified EIR, because it identifies a mitigation measure that the modified EIR then must address. The modified EIR also must address the decision to delete a mitigation measure. In other words, the measure cannot be deleted without a showing that it is infeasible. In addition, the deletion of an earlier adopted measure should be considered in reviewing any conclusion that the benefits of a project outweigh its unmitigated impact on the environment.

In the present case, the EIR discussed the 1986 mitigation measures, and concluded, in essence, that they were infeasible. The County concluded that the Project could not go forward if it was conditioned on implementing the 1986 mitigation measures, and deleted them as having been ill-advised. This conclusion was based on findings that the traffic resulting from Project-related development would be only a minor contributing factor to the region’s traffic problems, that the County lacked the funds to implement the measures recognized in 1986 and that the County had little control over improvements to the state’s highways, which improvements fall under the authority of the state itself, through the California Department of Transportation (Caltrans). The County, accordingly, stated a legitimate reason for *360 deleting the 1986 measures. As will be discussed further, below, substantial evidence supports the EIR’s findings that the 1986 measures were infeasible, and the conclusion of the County that the Project could not go forward unless those measures were not carried into the Updated Specific Plan.

Adequacy of the FSEIR’s Identification of Significant Effects and Analysis of Mitigation Measures

In order to fulfill its purpose as an informational document, an EIR is required, among other things, to identify the “significant effects” that a proposed project will have on the environment. (Pub. Resources Code, § 21100, subd. (b)(1); CEQA Guidelines, § 15126, subd. (a).) A “ ‘[significant . . .’ effect means a substantial, or potentially substantial, adverse change in the environment.” (Pub. Resources Code, § 21068.) Whether a project will have a significant effect is a matter of judgment, and it is recognized that an “ironclad definition of significant effect is not always possible.” (CEQA Guidelines, § 15064, subd. (b).)

Once a significant effect has been identified, the EIR must propose and describe mitigation measures that will minimize the significant environmental effects that the EIR has identified. (Pub. Resources Code, § 21100, subd. (b)(3); CEQA Guidelines, § 15126, subd. (e).) Mitigation measures must be feasible and enforceable. (CEQA Guidelines, § 15126.4, subd. (а) (1), (2).) “ ‘Feasible’ means 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.) Any mitigation measure must be “ ‘roughly proportional’ to the impacts of the project.” (CEQA Guidelines, § 15126.4, subd. (a)(4)(B).) “[A]n EIR need not analyze ‘ “ ‘every imaginable alternative or mitigation measure; its concern is with feasible means of reducing environmental effects.’ ” ’ [Citation.] Under the CEQA statute and guidelines a mitigation measure is ‘feasible’ if it is ‘capable of being accomplished in a successful manner within a reasonable period of time, taking into account economic, environmental, social, and technological factors.’ [Citations.] flQ In keeping with the statute and guidelines, an adequate EIR must respond to specific suggestions for mitigating a significant environmental impact unless the suggested mitigation is facially infeasible. [Citations.] While the response need not be exhaustive, it should evince good faith and a reasoned analysis. [Citations.]” (Los Angeles Unified School Dist. v. City of Los Angeles (1997) 58 Cal.App.4th 1019, 1029 [68 Cal.Rptr.2d 367].)

In addition, as noted above, while there is no requirement that an EIR itself be consistent with the relevant general plan, it must identify and *361 discuss any inconsistencies between a proposed project and the governing general plan. (CEQA Guidelines, § 15125, subd. (d).)

The failure to provide enough information to permit informed decision-making is fatal. “When the informational requirements of CEQA are not complied with, an agency has failed to proceed in ‘a manner required by law’ and has therefore abused its discretion. [Citations.]” (Save Our Peninsula Committee v. Monterey County Bd. of Supervisors (2001) 87 Cal.App.4th 99, 118 [104 Cal.Rptr.2d 326].)

A. Traffic

1. Identification of Significant Effects

Traffic congestion is described by “Levels of Service,” or “LOS.” LOS A represents a high level of service. LOS D represents a problematic level of service. The County’s Congestion Management Agency (CMA) permits levels of service as low as LOS E, notwithstanding that LOS E represents delays indicating poor progression and long cycle lengths. LOS F represents any service level below LOS E. In other words, once the level of service has deteriorated to F it can deteriorate further, but the service level will continue to be described as LOS F.

The FSEIR reports that traffic along the region’s highways will increase with or without the Project. It ultimately determined that traffic generated by the Project alone would have a significant effect on certain intersections, but that determination evolved in light of public comments and criticisms. At first—and before it was decided to delete the 1986 traffic mitigation measures—the drafters calculated the levels of service that would result if the Specific Plan went forward including the 1986 mitigation measures, such as widening portions of the state highways and creating means of bypassing key intersections. The drafters contrasted these projected levels of service with those that would occur if the Project was not developed and if none of the mitigation measures was adopted. The drafters found that traffic would be less congested if development went forward and the mitigation measures were adopted, than it would be if development did not go forward and the mitigation measures were not adopted, and concluded that “the project overall is deemed not to have significant adverse impacts upon traffic and circulation.”

The first draft became obsolete when it was decided to delete the 1986 mitigation measures. The drafters then took the position that the Project’s effect on traffic congestion and circulation would not be significant if future *362 congestion would cause an intersection to deteriorate to below acceptable levels whether or not the development went forward. 2 Thus, if an intersection would deteriorate to LOS F even without the Project, the fact that the Project would cause it to deteriorate even further was not deemed a significant effect. The only exception recognized by the drafters resulted if it could be concluded that the Project alone would cause an intersection’s level of service to deteriorate to below an acceptable level. Under such circumstances, the drafters calculated the amount of traffic that would be generated by the development by the year 2015. That figure was added to existing conditions. In light of these calculations, the drafters found that traffic generated by the Project alone would cause only one intersection to deteriorate to an unacceptable level, and thus concluded that the Project would have a significant effect only on that intersection.

This approach was criticized for underrating the effect the Project would have on traffic congestion by emphasizing the fact that traffic congestion would increase whether or not the Project went forward. In addition, it was pointed out that this approach failed to analyze the actual effect of the Project on any intersection that would deteriorate to LOS F irrespective of the Project’s impact on traffic.

The drafters responded by taking a third approach. In brief, the drafters found that development of the Project area would have a significant effect on traffic by the year 2015 if traffic generated by the Project alone would cause an intersection to deteriorate to an unacceptable level. They also found, however, that the Project would have a significant effect on other intersections where the traffic generated by the Project increased the “computed volume/capacity ratio by more than 10%” 3 and the intersection would deteriorate to an unacceptable level irrespective of Project-generated traffic. The Addendum then concluded that the Project would have a significant effect on traffic congestion at three intersections.

Although the methodology adopted by the drafters is difficult to follow, and the fact that their approach evolved causes additional confusion, the *363 drafters did explain their theories, supported them by calculations and made no attempt to hide the fact that the 1986 mitigation measures were being deleted. To the contrary, the drafters were careful to explain that the earlier measures had been deleted, and explained in detail their calculations of the impact the Project would have on traffic within the Project area and on regional traffic along the highways adjacent to the Project area. The FSEIR, accordingly, contained an adequate explanation of the drafters’ reasoning, and of the data underlying that reasoning. We conclude that it fulfilled its informational purpose as to identification of the significant effects the Project will have on traffic and circulation.

2. Identification and Analysis of Mitigation Measures

As discussed above, the FSEIR determined that traffic generated by the Project would have a significant effect on three key intersections. The FSEIR deleted the mitigation measures adopted in 1986, reciting that they were considered but rejected as infeasible in part because of a lack of funding, in part because they would require right-of-way takings from the adjacent properties and, as to one intersection, because of concerns expressed about the visual impact of the proposed improvement.

Petitioners do not complain about the FSEIR’s findings that the 1986 mitigation measures would require significant takings, or that they would have a negative visual impact on one intersection. They note, however, that the County has raised slightly over $2 million dollars through an “Airport Industrial Area Mitigation Fee,” 4 and suggest that the FSEIR erroneously concluded that the County cannot fund the 1986 mitigation measures. Fee-based infrastructure can be an adequate mitigation measure under CEQA (Save Our Peninsula Committee v. Monterey County Bd. of Supervisors, supra, 87 Cal.App.4th at p. 140), and can be particularly useful where, as here, traffic congestion results from cumulative conditions, and not solely from the development of a single project. (Ibid.) It therefore is somewhat surprising that the FSEIR contains no discussion of the Airport Industrial *364 Area Mitigation Fee, and only a very limited discussion of the possibility of using that or other fees as a means of mitigating the Project’s future effect on traffic.

Petitioners, however, do not contend that the FSEIR should have included a discussion of a traffic mitigation fee as a potential mitigation measure. They argue that the money the County has raised, or reasonably can be expected to raise, by means of the Airport Industrial Area Mitigation Fee, will be sufficient to fund the needed roadway improvements. The cost of the highway improvements, however, is far greater than $2 million; indeed, the County estimates that the improvements will cost $70 million. In addition, because the Project will cause only a small percentage of the projected traffic congestion, the County cannot insist that developers within the Project area shoulder the bulk of the expense for the needed highway improvements as a means of alleviating that congestion. Mitigation measures must be roughly proportional to the impacts of a project. (CEQA Guidelines, § 15126.4, subd. (a)(4)(B).) Although the existing mitigation fee appears to be a reasonable attempt to have developers pay their proportionate share of the cost of needed highway improvements, and the continued use of such fees undoubtedly would be useful, it cannot reasonably be argued that the funds that the County already has raised or that it reasonably can expect to raise in the future, will be enough to mitigate the effect on traffic that will result from cumulative conditions. 5

In addition, the record discloses that a number of entities, most particularly the City of Napa, criticized the drafters’ first attempt to determine significant impacts (which still assumed that the 1986 mitigation measures would be implemented) because local funding was inadequate to cover the costs of those measures, Caltrans’s money was committed to other projects, and there simply was no reason to assume that funding was or would be available. The record therefore fully supports the conclusion that the mitigation fee will not, cannot, and should not pay for the roadway improvements needed to obtain acceptable levels of service along the highways adjacent to the Project area.

For similar reasons it is unpersuasive that the record contains evidence that other localities are funding state highway improvements through local *365 tax measures. 6 Petitioners have made no showing that the County has the capability to raise taxes for purposes of highway improvements.

We find that the record supports the FSEIR’s conclusion that the 1986 mitigation measures could not be accomplished in a successful manner within a reasonable period of time, and thus were infeasible.

The FSEIR declined to analyze other possible improvements to the state highways as mitigation measures because, although they could reduce congestion, they would not reduce congestion to an acceptable level, and “they would not really solve the problem facing the corridor as a whole; and it is thus not possible to say that they are projects that are worth pursuing, even as interim measures.” These findings also are supported by the record.

After rejecting all proposed highway improvements, the FSEIR recited: “The greater global problem is that there is currently no consensus within the County on how to deal with expected congestion [along the major County corridor]. Without an overall concept for the corridor, it is difficult, and, in the view of the authors, unwise to propose stand-alone mitigation measures. This is largely because traffic from [the Project] is an important but still minority portion of the traffic growth expected within the corridor in the next 20 years. The overall growth in traffic dominates the future demand for improved traffic service, and any improvements due to the [Project] must fit within a logical framework of response to future growth, ffl] In the view of the authors of this document, a better approach would be to recognize that Napa County faces significant problems throughout the length of this corridor. . . . We believe that a better approach than attempting to assign responsibility via the standard type of impact analysis discussed above would be for the County to initiate a corridor study to develop acceptable solutions and to develop a funding plan.”

The FSEIR, accordingly, set forth a single mitigation measure, reciting: “A corridor-wide study of traffic impacts and potential solutions, including a *366 review of potential funding sources and potential legislative remedies to the impacts of non-Napa-County traffic growth, should be initiated by the County with the support of all jurisdictions neighboring the Airport Specific Plan Area. The study should cover SR 29 from the Napa/Solano County border to the SR 12/29/121 intersection and all of SR 12 from the Solano County Line to the Sonoma County Line. Development projects within the Napa Airport Specific Plan should be assessed a fair-share fee to be contributed to the cost of this study. Depending on the eventual results of the study, development within the airport may also be assessed fees for fair-share costs of improving the roadway system itself. The proposal for a study of the corridor is not intended as a replacement for fair-share contributions to eventual physical mitigation measures.”

A study, such as that proposed by the County, may be a mitigation measure if there is a definite commitment both to produce the study and to take such mitigation measures as are recommended by it. (See Federation of Hillside & Canyon Associations v. City of Los Angeles, supra, 83 Cal.App.4th at pp. 1261-1262; Rio Vista Farm Bureau Center v. County of Solano, supra, 5 Cal.App.4th 351, 377; and Sacramento Old City Assn. v. City Council (1991) 229 Cal.App.3d 1011, 1026-1030 [280 Cal.Rptr. 478].) As will be discussed further in connection with the issue of consistency between the Updated Specific Plan and the County’s General Plan, the County has made only a limited commitment to participate in the study recommended by the FSEIR. The study, accordingly, cannot be deemed a true mitigation measure. That a proposal was mislabeled a mitigation measure, however, does not by itself invalidate the FSEIR.

Petitioners make no claim that there were other measures for mitigating the impact the Project would have on traffic that should have been analyzed or should have been adopted as feasible. 7

We conclude that the FSEIR adequately identifies the Project’s significant effects on traffic. It could have discussed the efficacy of imposing a mitigation fee or of using the existing mitigation fee as means of providing some funding for highway improvements, but the available evidence indicates that it would be unreasonable to view such fees as a potential solution to the region’s traffic congestion and circulation problems. In short, the FSEIR adequately identified and analyzed mitigation measures, and adequately stated its reasons for rejecting the mitigation measures adopted in 1986.

*367 B. Housing

1. Identification of Significant Effects

CEQA Guidelines section 15126, subdivision (d), requires an EIR to discuss the “Growth-Inducing Impact of the Proposed Project.” Guidelines section 15126.2, subdivision (d), elaborates: “. . . Discuss the ways in which the proposed project could foster economic or population growth, or the construction of additional housing, either directly or indirectly, in the surrounding environment. . . . Increases in population may tax existing community service facilities, requiring construction of new facilities that could cause significant environmental effects. Also discuss the characteristic of some projects which may encourage and facilitate other activities that could significantly affect the environment, either individually or cumulatively. It must not be assumed that growth in any area is necessarily beneficial, detrimental, or of little significance to the environment.”

The Updated Specific Plan does not call for the construction of any housing in the Project area, and its proximity to the airport renders it unsuitable for housing. The County’s position was and is that because there is no provision for housing units within the Specific Plan area, the proposed development has no direct impact on housing, and thus no significant effect on the environment requiring discussion in the FSEIR. Petitioners’ position is that because the Project will create jobs, and the creation of jobs will bring people into the area, the FSEIR was required to analyze the resulting housing needs. We will find that the FSEIR was indeed required to discuss such housing needs as reasonably might be generated by the Project, but not in great detail. We also will find that the FSEIR, including an attached “Market and Jobs/Housing Analysis,” contains an adequate discussion of these housing needs.

In arguing that the Project will have no significant effect on housing, the County relies, in part, on CEQA Guidelines section 15131. Subdivision (a) of CEQA Guidelines section 15131 recognizes that a project may have economic and social effects that do not themselves cause a physical change in the environment, or act only indirectly to cause a physical change in the environment. It recognizes that CEQA applies only if a project causes a physical change, and accordingly provides that the EIR need not include a discussion of economic or social effects that do not cause a physical change. If the anticipated economic or social effects will create a chain of cause and effect that will result in a change in the environment, the intermediate economic or social changes need not be analyzed in any detail; “[t]he focus *368 of the analysis shall be on the physical changes.” (Ibid.) The County argues that because the Project calls for no construction of housing units, the Project’s impact on housing is merely an intermediate step on the chain of cause and effect, and therefore need not be discussed in the FSEIR.

CEQA Guidelines section 15131, does not, however, preclude the necessity for EIR review of a project simply because the project itself does not include action that will have a significant effect on the environment. It distinguishes between a project’s effects on the social or economic environment and its effects on the physical environment, providing that EIR review is not required absent some resulting effect on the physical environment. Thus, for example, a “social” impact, such as overcrowding in a classroom is not a “significant effect” requiring analysis unless the overcrowding will be so great as to have an effect on the physical environment by requiring the construction of additional classrooms. (Goleta Union School Dist. v. Regents of University of California (1995) 37 Cal.App.4th 1025, 1032 [44 Cal.Rptr.2d 110].)

It also is settled that the EIR must discuss growth-inducing impacts even though those impacts are not themselves a part of the project under consideration, and even though the extent of the growth is difficult to calculate. The case law supports this distinction. The court in City of Antioch v. City Council (1986) 187 Cal.App.3d 1325 [232 Cal.Rptr. 507] found that a project required an EIR notwithstanding that the project itself involved only the construction of a road and sewer project which did not in and of themselves have a significant effect on the environment. The court recognized that the sole reason for the construction was to provide a catalyst for further development in the immediate area. It held that because construction of the project could not easily be undone, and because achievement of its purpose would almost certainly have significant environmental impacts, the project should not go forward until such impacts were evaluated in the manner prescribed by CEQA. (Id. at pp. 1337-1338.)

In Stanislaus Audubon Society, Inc. v. County of Stanislaus (1995) 33 Cal.App.4th 144 [39 Cal.Rptr.2d 54], the court considered a proposed construction of a country club and golf course and attendant facilities. It was contended there that an EIR was not required because the growth-inducing impacts of the proposed project were too remote or speculative, and EIR’s would be prepared in connection with any application for a housing development. The court responded, “The fact that the exact extent and location of such growth cannot now be determined does not excuse the County from preparation of an EIR. . . . [Rjeview of the likely environmental effects of *369 the proposed country club cannot be postponed until such effects have already manifested themselves through requests for amendment of the general plan and applications for approval of housing developments.” (Id. at pp. 158-159, fn. omitted.)

It follows that an agency cannot avoid the EIR process simply because a project does not itself call for the construction of housing or other facilities that will be needed to support the growth contemplated by the project. It does not follow, however, that an EIR is required to make a detailed analysis of the impacts of a project on housing and growth. Nothing in the Guidelines, or in the cases, requires more than a general analysis of projected growth. The detail required in any particular case necessarily depends on a multitude of factors, including, but not limited to, the nature of the project, the directness or indirectness of the contemplated impact and the ability to forecast the actual effects the project will have on the physical environment. In addition, it is relevant, although by no means determinative, that future effects will themselves require analysis under CEQA.

We also do not believe that EIR review can be avoided simply because the project’s effect on growth and housing will be felt outside of the project area. Indeed, the purpose of CEQA would be undermined if the appropriate governmental agencies went forward without an awareness of the effects a project will have on areas outside of the boundaries of the project area. That the effects will be felt outside of the project area, however, is one of the factors that determines the amount of detail required in any discussion. Less detail, for example, would be required where those effects are more indirect than effects felt within the project area, or where it is be difficult to predict them with any accuracy.

The issue in City of Antioch and Stanislaus Audubon Society was whether it was necessary to prepare an EIR at all in connection with the contemplated projects. Those cases, however, also illustrate situations requiring more detail in the EIR than we believe is required in the present case. Each case involved a project that was simply the first step in the development of a particular area. To permit them to go forward without analyzing the effect of the ultimate development as a whole, therefore, would result in a “piecemeal review in which ‘environmental considerations . . . become submerged by chopping a large project into many little ones—each with a minimal potential impact on the environment—which cumulatively may have disastrous consequences.’ [Citations.]” (City of Antioch v. City Council, supra, 187 Cal.App.3d at p. 1333.) Here, in contrast, the Project contemplates not just the first step of the development of the Project area, but the area’s full *370 development, and the FSEIR contains an analysis of the full extent of the contemplated development of the Project area. In addition, although the Project-induced growth inevitably will have an effect on the physical environment, unlike the situations in City of Antioch, and Stanislaus Audubon Society, this effect will be diffused over a large, undefined area. It also is true that there are a greater number of variables in the present case than existed in City of Antioch, Stanislaus Audubon Society, and similar cases, rendering it even more difficult to predict the location of anticipated housing with any accuracy.

Nonetheless, in order to fulfill its purpose as an informational document, the FSEIR should, at a minimum, identify the number and type of housing units that persons working within the Project area can be anticipated to require, and identify the probable location of those units. The FSEIR also should consider whether the identified communities have sufficient housing units and sufficient services to accommodate the anticipated increase in population. If it is concluded that the communities lack sufficient units and/or services, the FSEIR should identify that fact and explain that action will need to be taken to provide those units or services, or both. Because it cannot be known if the Project will cause growth in any particular area, and because the Project most likely will not be the sole contributor to growth in any particular area, it is not, however, reasonable to require the FSEIR to undertake a detailed analysis of the results of such growth.

With this in mind, we turn to the FSEIR’s actual discussion of housing impacts. The FSEIR found that the Project would create a need and opportunity for employment, concluding that it would result in a need for additional housing units at locations outside of the Project area. It reasoned, however, that it was impossible to discuss the environmental effects such units would have because “the nature and extent of future, indirect development is not known at this time.” The FSEIR therefore simply declined to consider the possible effects the Project might have on housing in surrounding communities. This discussion, in and of itself, is inadequate.

CEQA Guidelines section 15131, subdivision (c), however, provides: “Economic, social, and particularly housing factors shall be considered by public agencies together with technological and environmental factors in deciding whether changes in a project are feasible to reduce or avoid the significant effects on the environment identified in the EIR. If information on these factors is not contained in the EIR, the information must be added to the record in some other manner to allow the agency to consider the factors in reaching a decision on the project.” In the present case, the Market *371 and Jobs/Housing Analysis is appended to the Updated Specific Plan. This plan seems to make exactly the projections that the FSEIR declares cannot be made, detailing the types of businesses that can be expected to locate within the Project area, the rate at which acreage within the area can be expected to be “absorbed” into each type of projected land use and the number of employees that each type of business can be expected to add. It concludes that by the year 2015, an additional 9,881 employees will be added to area, requiring construction of an additional 5,457 housing units. It projects the number of units required by income (“very low,” “low,” “moderate” and “above moderate”), and it explores the possibility of locating the majority of the units in the City of Napa and/or in American Canyon.

We find that the FSEIR, including the Market and Jobs/Housing Analysis, adequately fulfills the FSEIR’s informational purpose, and that the FSEIR, accordingly, contains an adequate discussion of housing.

2. Identification and Analysis of Mitigation Measures

As the FSEIR concluded that the Project would not have any significant effect on housing, it also did not discuss any mitigation measures directed at the effect the Project would have on housing. As discussed above, we disagree with the first conclusion, but find that the FSEIR, together with the Market and Jobs/Housing Analysis, provides an adequate discussion of the impact the Project can be expected to have on growth and housing in outlying areas. The question now is whether the FSEIR is deficient in failing to discuss measures that might mitigate that impact.

Neither CEQA itself, nor the cases that have interpreted it, require an EIR to anticipate and mitigate the effects of a particular project on growth on other areas. In circumstances such as these, we think that it is enough that the FSEIR warns interested persons and governing bodies of the probability that additional housing will be needed so that the they can take steps to prepare for or address that probability. The FSEIR need not forecast the impact that the housing will have on as yet unidentified areas and propose measures to mitigate that impact. That process is best reserved until such time as a particular housing project is proposed.

C. Treatment of Wastewater and Sources of Water

1. Identification of Significant Effects

The FSEIR provides a detailed analysis of the amount of wastewater that will be generated by users within the Project area, and the amount of *372 water that will be consumed by those users. 8 It reports that, for the most part, main pipelines and related infrastructure for both wastewater and water are in place. It assumes that wastewater will be treated, as it is now, by the Soscol Treatment Plant. It reports that the Soscol Treatment Plant is close to capacity, and will not be able to treat additional wastewater generated by the Project. The FSEIR further reports, however, that American Canyon is expected to enter into an agreement with the City of Vallejo that should allow American Canyon to send its wastewater to Vallejo for treatment, rather than to the So’scol Treatment Plant, which then will have the capacity to treat wastewater from the Project area. It also reports that there are plans to expand the Soscol Treatment Plant to provide capacity for all wastewater generated within the Napa Sanitation District (NSD) service area through the year 2020, and that funds are available for that expansion. The FSEIR points out, however, that “[wjhile new facilities are being planned by the NSD and financing methods are available, there would be an interim period prior to the completion of improvements in which the biological treatment capacity of the existing facility may be inadequate, resulting in a significant impact upon the adequacy of wastewater service.”

The FSEIR does not discuss the possibility that it might become necessary to treat wastewater from the Project at some facility other than the Soscol Treatment Plant, and therefore does not address the impacts of any alternative means of treatment.

As to water use, the FSEIR assumes that water to the Project area will be supplied in the future, as it is supplied now, by American Canyon. American Canyon receives water from the State Water Project via the North Bay Aqueduct. The FSEIR reports that at present, American Canyon uses less than one-half of the amount of water allocated to it, but it appears that by the year 2015, the combined needs of the city and the Project will exceed American Canyon’s aqueduct allotment. The FSEIR further reports that American Canyon is in the process of reaching an agreement with the City of Vallejo that will permit American Canyon to purchase additional water from a water treatment facility in that nearby town. The FSEIR assumes that this water will prevent the anticipated shortfall. It therefore concludes that the Project’s demand for water will not result in a significant effect.

It has been held that an EIR is inadequate if it fails to identify at least a potential source for water. In Stanislaus Natural Heritage Project v. County *373 of Stanislaus (1996) 48 Cal.App.4th 182 [55 Cal.Rptr.2d 625], for example, the failure to identify a source of water beyond the first five years of development rendered the EIR inadequate, although the developer was pursuing several possible sources. It also has been held that an EIR is inadequate if the project intends to use water from an existing source, but it is not shown that the existing source has enough water to serve the project and the current users. (Santiago County Water Dist. v. County of Orange (1981) 118 Cal.App.3d 818 [173 Cal.Rptr. 602].) On the other hand, it has been held that an EIR is not required to engage in speculation in order to analyze a “worst case scenario.” (Towards Responsibility in Planning v. City Council (1988) 200 Cal.App.3d 671 [246 Cal.Rptr. 317] (hereafter TRIP).) In that case, the court held that an EIR was not required to analyze the effects that would result from the construction of a sewage treatment facility, when (1) all indications suggested that the facility would never be needed, and (2) the facility—if it was constructed—would be subjected to its own environmental review.

The present situation falls somewhere between that at issue in TRIP on the one hand, and those in Stanislaus and Santiago, on the other. In TRIP, affected cities had entered into agreements designed to provide service sufficient to meet the project’s needs. In the present case, the necessary agreements have not yet been reached, and as the Project has no control over those agreements, it cannot ensure that they will be reached. Unlike the EIR in Santiago, the FSEIR does consider the impact of the Project’s needs on the area’s resources and the ability of those resources to meet the demands of other users. Unlike the situation in Stanislaus, the FSEIR has identified sources for water and facilities for the treatment of wastewater, although their availability has not been absolutely established. Moreover, the FSEIR analyzes the capacities of the existing systems and concludes that the anticipated resources, if available, will be able to handle the Project area’s needs for water and disposal of wastewater.

It follows that a compromise between the positions adopted in those cases is in order. We conclude that the FSEIR need not identify and analyze all possible resources that might serve the Project should the anticipated resources fail to materialize. Because of the uncertainty surrounding the anticipated sources for water and wastewater treatment, however, the FSEIR also cannot simply label the possibility that they will not materialize as “speculative,” and decline to address it. The County should be informed if other sources exist, and be informed, in at least general terms, of the environmental consequences of tapping such resources. Without either such information or a guarantee that the resources now identified in the FSEIR *374 will be available, the County simply cannot make a meaningful assessment of the potentially significant environmental impacts of the Project. (See Sierra Club v. State Bd. of Forestry (1994) 7 Cal.4th 1215, 1237 [32 Cal.Rptr.2d 19, 876 P.2d 505].)

2. Identification and Analysis of Mitigation Measures

The FSEIR set forth the following measures to mitigate the impact the Project can be expected to have on treatment of wastewater at the Soscol Treatment Plant during the “interim period” before the plant completes the planned expansion and before American Canyon diverts its wastewater to Vallejo: (1) “Prior to permit approval, the NSD should review all new large development proposals to determine their ability to service the site”; (2) “[Industries that produce large volumes of sewage should provide on-site pretreatment and storage of wastes for a controlled release into the sewer system”; and (3) “Temporary treatment and/or storage ponds should be provided to increase the biological treatment capacity of the existing facilities pending the implementation of the planned Soscol Treatment Plant expansion.”

Once the FSEIR concluded that the Project would have no significant effect on water resources, it became unnecessary for it to analyze measures that would mitigate the Project’s effect on water sources. The FSEIR nonetheless set forth the following general policies, characterizing them as mitigation measures: (1) “The County should coordinate future projects and associated water demand in the Plan Area with American Canyon and the NSD to further ensure that adequate water supplies and treatment are provided in a timely manner”; (2) “The County should promote the design, utilization and construction of water conservation devices within proposed buildings and landscape plans”; and (3) “The County should promote the utilization of reclaimed water to minimize potable water use.”

These measures and policies would be adequate if the identified sources for water and treatment of wastewater were certain. As they are not certain, however, and as we have found that the FSEIR is inadequate in failing either to identify new sources or to report that none is available, the FSEIR also is inadequate in failing to identify and analyze appropriate mitigation measures related to the alternative sources, if any. In theory, at least, the FSEIR also could state a mitigation measure that would prevent development if the identified sources fail to materialize. As it stands, however, the FSEIR’s discussion of wastewater and water is inadequate.

*375 Conclusion

In conclusion, the FSEIR’s identification of significant effects, and its discussion of mitigation measures relating to traffic, is adequate. Its identification of significant effects on growth and housing is adequate. It need not include a discussion of mitigation measures related to future growth outside of the Project area, notwithstanding that the Project may contribute to that growth. The FSEIR, however, is inadequate in failing to identify and analyze alternative sources for water and resources for treatment of wastewater. As no such resources and sources have been identified, we cannot and do not here determine what kind of analysis of them might be required or whether there are mitigation measures relating to them that need to be discussed.

Consistency of Updated Specific Plan with County’s General Plan

A. Relevant Provisions of the County’s General Plan

The circulation element of County’s General Plan recognizes six levels of traffic congestion, LOS A through LOS F. The General Plan recites that roadways operating at LOS D or lower are candidates for improvement. It estimates the increase in number of trips over the county’s roadways as can be expected to occur as the County and surrounding areas are developed and their populations grow, concluding that by the year 2000, assuming no major changes are made in the highway system, the average peak hour traffic conditions will have reached or exceeded LOS D along a number of roadways in the Project area.

The General Plan includes “Goals” of developing a comprehensive circulation system for the County and of improving the county roadway system so as to increase the area’s ability to handle current and future traffic. It includes supplementary “Policy Guidelines” designed to “[cjontinue or commence planning and engineering activities to improve levels of service” on designated “critical links in the highway system.” It then describes a number of projects that would improve the levels of service on these highways, and sets forth a recommended improvement program for the County through the year 2000. Included in these improvements are the improvements adopted in the 1986 Specific Plan but not carried forward into the 1998 Updated Specific Plan. 9

The General Plan’s housing element recognizes the County’s “responsibilities under Measure A [the County’s ‘Slow-Growth Initiative’],” Measure *376 J (the “Agricultural Lands Preservation Initiative”) “and State Housing laws to address its housing needs independently and in concert with other local, regional, state and national private and public sector organizations.” It acknowledges the objectives of the Association of Bay Area Governments “a. To increase the housing supply in accord with the Region’s needs,” “b. To maintain and improve existing housing so that it can better fill the Region’s needs,” and “c. To expand and conserve housing opportunities for lower income people.” The General Plan sets forth eight “Goals,” which can be summarized as: (1) facilitate the implementation of housing element programs for all economic segments of the population residing in the unincorporated areas of the County; (2) ensure that the County housing stock is continually maintained or upgraded for quality, safety and livability; (3) ensure that the designated County residential areas are continually maintained or improved for quality, safety and livability; (4) encourage choice and economic integration and eliminate discrimination based on improper factors; (5) ensure that lower and middle income housing in unincorporated County areas is maintained and developed in ways that increases the level of home ownership of low-income families and minimizes the need for employment-related private transportation; (6) encourage coordination between private and public parties to regulate, develop and make available housing stock; (7) work with cities, government, citizens, and the private sector to plan for housing services, facilities and accommodations, including housing; and (8) encourage energy efficiency.

The General Plan states many “Policies” and “Objectives” designed to further these goals. The General Plan does not require the County to build any housing units or to fund any housing, but it does require the County to continue programs that will promote housing, and requires the County to take action such as working with its cities, or with public and private groups, to promote the creation of affordable housing and to obtain funding for low-income households. Under Goal 5, concerning housing location, density and timing, the General Plan states policies of concentrating housing in urban areas. Policy 5.2 provides: “The County will assume that the density of urban development in the American Canyon Area precludes extensive future subdivision activity based on septic tanks and wells.” Policy 5.8 provides: “The County will encourage the construction of residential units in commercial and industrial development to service jobs created by such *377 development.” Under Goal 6, concerning urban facilities and services, the General Plan states policies of ensuring that sufficient services are available to the County’s housing units. Again, the General Plan does not define any particular action that the County should take; it simply provides that the County will discourage some kinds of development, encourage other kinds of development and will work with the owners of utilities.

B. Relevant Provisions of Updated Specific Plan

The circulation element of the Updated Specific Plan does not include any specific highway improvements. It also does not include any detailed statements of goals or policies comparable to those set forth in the General Plan. It recites that “[a] primary purpose of the circulation and transportation system is to improve regional access and to facilitate efficient local access throughout the Plan Area in a cost-effective manner.” It includes two “Roadway System Objectives”: “a) Where feasible, improve regional access to the Plan Area, b) Coordinate the local and regional roadway system into a logical, integrated circulation network.” Finally, it provides: “Design, size, and improve roadways to adequately meet future traffic demands, consistent with County standards and additional standards set forth herein.”

The Updated Specific Plan contains no housing element. Its only discussion of housing points out that the plan “does not change the amount of land designated for housing or increase or reduce the existing opportunities for housing development. Consequently, the Specific Plan is consistent with the existing Napa County Housing Element.” The Updated Specific Plan nonetheless seeks to discourage “in-commuting” from areas outside the County, and therefore proposes that the General Plan be amended to add the following programs and measures: “(a) The County will seek to increase the housing opportunities within the County’s existing urban areas, given the limitations of Measures ‘A’ and T, by: [authorizing the development of residential uses within urban areas designated for industrial uses in the General Plan, on such designated parcels and areas on the east side of State Highway 221, south of and including the State Hospital, and HQ [wjorking with the City of American Canyon to redesignate existing urban areas, adjacent to the City but outside of its adopted Sphere of Influence as additional housing opportunity sites. ft[] (b) The County will work with the cities of American Canyon and Napa to develop and adopt agreements . . . to jointly designate appropriate housing sites which could be developed for housing.”

*378 C. Effect of Failure to Adopt Affirmative Measures to Mitigate Adverse Effects on General Plan’s Goals and Policies

“A project is consistent with the general plan ‘ “if, considering all its aspects, it will further the objectives and policies of the general plan and not obstruct their attainment.” ’ [Citation.] A given project need not be in perfect conformity with each and every general plan policy. [Citation.] To be consistent, a [project] must be ‘compatible with’ the objectives, policies, general land uses and programs specified in the general plan. [Citation.]” (FUTURE v. Board of Supervisors, supra, 62 Cal.App.4th atp. 1336, quoting from Corona-Norco Unified School Dist. v. City of Corona, supra, 17 Cal.App.4th at p. 994.)

General plans ordinarily do not state specific mandates or prohibitions. Rather, they state “policies,” and set forth “goals.” The County’s General Plan is no exception, setting forth “goals” and supplemental “policy guidelines,” and outlining improvements that would further these goals and policy guidelines. In finding that the Updated Specific Plan was inconsistent with the County’s General Plan, the trial court, accordingly, found that it was fatally inconsistent with goals and policies set forth in the County’s General Plan.

The County first contends that the goals and policies stated in its General Plan should not be viewed as directives, and the General Plan therefore should be read as advisory rather than mandatory. This contention conflicts with the recognition that consistency requires compatibility with the general plan’s “ ‘objectives, policies, general land uses, and programs.’ ” (Corona-Norco Unified School Dist. v. City of Corona, supra, 17 Cal.App.4th at p. 994, fn. 5.) The question is not whether there is a direct conflict between some mandatory provision of a general plan and some aspect of a project, but whether the project is compatible with, and does not frustrate, the general plan’s goals and policies.

The County next acknowledges that the Updated Specific Plan’s circulation element does not actually implement the goals and policies identified in the General Plan. The County also recognizes that the Updated Specific Plan requires no specific action that will further the General Plan’s housing goals, policies or objectives. The County argues, however, that it does not follow that the Updated Specific Plan is inconsistent with the General Plan. It points out that the Updated Specific Plan includes roadway objectives that echo the General Plan’s goals and policies, that the Updated Specific Plan does not harm the quality or availability of existing housing stock or designated *379 residential areas, and that it does not conflict with the General Plan’s goal of working with cities, other governmental units, developers and citizens to plan for housing. The County also points out, correctly, that the cases that have struck down a specific plan for inconsistency with a general plan, have concerned more than a failure to implement the general plan’s goals and policies. In FUTURE v. Board of Supervisors, supra, 62 Cal.App.4th 1332, for example, the general plan specified that the designation “Low Density Residential” would be restricted to lands contiguous to the identified areas. It was held that a plan to develop housing designated “Low Density Residential” in an area that was not contiguous to any identified area was inconsistent with the general plan. (Id. at pp. 1340-1341.)

We are of the opinion that the consistency doctrine requires more than that the Updated Specific Plan recite goals and policies that are consistent with those set forth in the County’s General Plan. We also are of the opinion that cases such as FUTURE v. Board of Supervisors, supra, 62 Cal.App.4th 1332, do not require an outright conflict between provisions before they can be found to be inconsistent. The proper question is whether development of the Project Area under the Updated Specific Plan is compatible with and will not frustrate the General Plan’s goals and policies. If the Updated Specific Plan will frustrate the General Plan’s goals and policies, it is inconsistent with the County’s General Plan unless it also includes definite affirmative commitments to mitigate the adverse effect or effects.

We find support for this conclusion in Concerned Citizens of Calaveras County v. Board of Supervisors (1985) 166 Cal.App.3d 90 [212 Cal.Rptr. 273], The court in that case considered a possible conflict between the circulation and land use elements of a general plan. The land use element recognized the likelihood that the area’s population would grow, and stated a goal of encouraging commercial development to support that growth. The circulation element, like the circulation element in the County’s General Plan here, recognized the limitations of the area’s roadways, finding that the roadways would not be able to handle a substantial increase in traffic. The circulation element included no specific means of increasing the circulation of traffic should growth occur, reciting that there were no funds available for any major projects on the highways. The circulation element stated a “goal” of encouraging the improvement of the highways, a “policy” of supporting the state in any plans to improve state highways traversing the county, and an “implementation measure” of lobbying for increased state funding for state highway improvements. {Id. at pp. 100-102.) The court found that the circulation and land use elements were internally inconsistent and contradictory. It also held that “the general plan cannot identify substantial problems *380 that will emerge with its state highway system, further report that no known funding sources are available for improvements necessary to remedy the problems, and achieve statutorily mandated correlation with its land use element (which provides for substantial population increases) simply by stating that the county will solve its problems by asking other agencies of government for money.” (Id. at p. 103.)

The question in Concerned Citizens was whether the general plan itself was flawed because it included inconsistent provisions, while the question here is whether the County’s General Plan and the Updated Specific Plan contain inconsistent provisions. Nonetheless, the essential holding of the court in Concerned Citizens was that an inconsistency was created if the implementation of one provision will frustrate a policy stated in a second provision and there is no affirmative commitment to mitigate that adverse effect. The same principle applies here. The County cannot state a policy of reducing traffic congestion, recognize that an increase in traffic will cause unacceptable congestion and at the same time approve a project that will increase traffic congestion without taking affirmative steps to handle that increase. It also cannot state goals of providing adequate housing to meet the needs of persons living in the area, and at the same time approve a project that will increase the need for housing without taking affirmative steps to handle that increase.

The County suggests that even if it is not enough that the Updated Specific Plan states goals and policies similar to those stated in the General Plan, it should be enough that the Updated Specific Plan recites that the County will work towards improving the roadways and will work with nearby communities to provide suitable housing. As we noted earlier, in connection with our discussion of the EIR’s analysis of mitigation measures for traffic congestion and circulation, an actual commitment to study a problem and to prepare and implement a plan to mitigate that problem may be a valid mitigation measure. (See Federation of Hillside & Canyon Associations v. City of Los Angeles, supra, 83 Cal.App.4th at pp. 1261-1262; Rio Vista Farm Bureau Center v. County of Solano, supra, 5 Cal.App.4th 351, 377; Sacramento Old City Assn. v. City Council, supra, 229 Cal.App.3d atpp. 1026-1030.) The Updated Specific Plan, however, makes no binding commitment to do anything to alleviate the impact the Project will have on traffic and housing.

Finally, the County, while acknowledging that the Project will create a need for housing for persons working within the Project area, also claims that the record discloses that development in the Project Area will not create *381 a need for housing units. This claim is based on the Jobs and Housing Impact Analysis appended to the Updated Specific Plan. The analysis discloses that as of the time it was prepared there were more housing units than jobs in American Canyon and Napa City. It predicts that the same kind of jobs/ housing imbalance will exist in those cities in the future. The analysis also, however, points out that the opposite is true for the three small Upvalley cities (St. Helena, Calistoga and Yountville) and for the unincorporated County. It further predicts that by the year 2015 there will be an overall housing shortage in the County of 6,650 units. As discussed earlier, the analysis estimates that 5,457 housing units will be needed to support the employees that can be expected to work in the Project area. The record, accordingly, does not support the County’s claim.

D. Water and Wastewater

The parties did not argue in the trial court that the Updated Specific Plan’s discussion of water and wastewater was inconsistent with the County’s General Plan, and have not furnished us with a copy of the General Plan’s conservation element which, presumably, includes a discussion of those topics. The County’s conclusion that the Updated Specific Plan is consistent with the General Plan’s conservation element is, however, invalidated by the fact that the FSEIR does not contain an adequate discussion of these topics.

Steelhead Trout

CEQA Guidelines section 15065, subdivision (a), provides that an agency shall find that a project has a significant effect on the environment if the project has the potential to “reduce the number or restrict the range of an endangered, rare or threatened species.” The trial court ruled that the FSEIR was inadequate in that it did not investigate and make findings as to the impact the Updated Specific Plan would have on steelhead trout, an endangered species. The County contends that the trial court was not entitled to make this ruling because the question of the effect of the Project on steelhead trout had not been adequately raised at the administrative level.

The 1986 Specific Plan, and the various drafts of the Updated Specific Plan, called for a number of measures designed to protect the integrity of the area’s creeks and wetlands, including a 150-foot setback along Soscol Creek and a 75-foot setback along Fagan Creek. The draft EIR generally discussed the effect of the Updated Specific Plan on aquatic habitat and special status species, concluding that measures incorporated into the Updated Specific Plan—including, presumably, the creek setbacks—would reduce the effect of the Project to an acceptable measure. The draft EIR, however, was silent *382 as to steelhead trout, a species that at that time was not listed as endangered. The County certified the FSEIR on April 21, 1998. The FSEIR, like the draft EIR, was silent as to steelhead trout, notwithstanding that by that time the steelhead trout had been identified as an endangered species.

Although it had certified the FSEIR, the County did not adopt the Updated Specific Plan until October 20, 1998. Instead, it continued to hold public hearings and receive letters and comments from interested persons, later amending the FSEIR as a result of this process. Public Resources Code section 21177 provides, as relevant here, “(a) No action or proceeding may be brought. . . unless the alleged grounds . . . were presented to the public agency orally or in writing by any person during the public comment period provided by this division or prior to the close of the public hearing on the project before the issuance of the notice of determination.” The County accordingly concedes that comments made and letters received after the FSEIR was certified, but before the Updated Specific Plan was adopted, may be considered in determining if the issue of steelhead trout was raised at the administrative level.

The subject of steelhead trout first was mentioned on April 28, 1998, during a discussion of creek setbacks. An attorney representing a property owner complained that the Updated Specific Plan allowed no development within a 150-foot setback along Soscol Creek, stating that the setback deprived the owner of the use of his property. He sought a statement in the plan that a 50-foot setback would be permitted if the applicant could convince the Fish and Game Department that water quality would be protected. The same attorney, representing another property owner, also sought a provision in the Updated Specific Plan that would permit a property owner to trade open space for a reduction of a setback, referring in this case to Fagan Creek.

The County, the attorney representing the landowners, and several other persons, discussed the possibilities of creating a variance procedure for setbacks along one or more creeks within the Specific Plan area, or for reducing the setbacks to 50 feet. Several commentators spoke in support of maintaining a 150-foot setback in order to preserve water quality. One of the County supervisors pointed out that Soscol Creek is a known fishery that includes steelhead spawn. This supervisor later pointed out that Fagan Creek is the fresh water supply for the Fagan Marsh Ecological Reserve, a valuable wetland habitat, urging the County to recognize that there are significant ecological reasons for preserving setbacks. A representative of the Friends of the Napa River also spoke at length about the importance of preserving the 150-foot setback limits, pointing out, among other things, that developers *383 already were “asking for variances to these riparian corridors in order to place industrial buildings on these valuable watershed land[s].”

A letter from American Canyon, dated May 11, 1998, complains that the FSEIR fails to designate California coast steelhead trout as an endangered species, asserting that “This endangered species designation relates directly to an issue raised at the last hearing . . . wherein it was requested that to avoid a ‘takings claim’ that the riparian set back along . . . Soscol and Fagen Creeks should be modified from . . . 150 to [no more than] 50 feet.” American Canyon continued, “Should this occur, not only is this a substantive change in the Specific Plan, but also an impact bearing directly on the habitat of the designated endangered species which is not acknowledged or assessed presently in the FSEIR.” In addition, American Canyon’s vice-mayor complained at a later hearing, “I have a problem with Fagan Creek. Nothing seems to be being done about the steelhead trout that seem[] to [exist] in the Napa River and its tributaries.”

As a result of these proceedings, the Updated Specific Plan was amended to permit variances when “the required creekside setbacks do not substantially advance a legitimate government interest or [deny] a property owner economically viable use of the land (or whatever is the current legal standard in effect for a ‘takings’ claim under the Fifth Amendment at the time the applicant seeks a variation from the requirements of this Specific Plan).” The amendment did not change the existing setbacks of 150 feet for Soscol Creek and 75 feet for Fagan Creek. The FSEIR was not amended to list steelhead trout as an endangered species, and it did not consider the effect of the Project, including the setbacks, on steelhead trout. It also did not consider whether the variances permitted under the amended Updated Specific Plan might have an effect on aquatic habitat.

The record does not indicate that any further challenge was made at the administrative level as to the failure of the FSEIR to address issues relating to steelhead trout.

In arguing that the effect of the Project on steelhead trout was not adequately raised at the administrative level, the County claims that the only point considered in the administrative proceedings was whether creekside setbacks should be reduced or whether variances should be granted. The County therefore contends that the trial court had the power to consider if the FSEIR was inadequate in failing to discuss setbacks and variances, but did not have the power to consider if the FSEIR was inadequate in its discussion of steelhead trout. We find, to the contrary, that the issue of steelhead trout was timely and adequately raised at the administrative level.

*384 The purpose of the rule of exhaustion of administrative remedies is to provide an administrative agency with the opportunity to decide matters in its area of expertise prior to judicial review. (Browning-Ferris Industries v. City Council (1986) 181 Cal.App.3d 852, 859 [226 Cal.Rptr. 575].) The decisionmaking body “ ‘is entitled to learn the contentions of interested parties before litigation is instituted. If [plaintiffs] have previously sought administrative relief ... the Board will have had its opportunity to act and to render litigation unnecessary, if it had chosen to do so.” [Citations.]’ [Citation.]” (Corona-Norco Unified School Dist. v. City of Corona, supra, 17 Cal.App.4th 985, 997.) The record reflects that the County was made aware that steelhead trout is an endangered species, that at least one creek in the Project Area is a spawning area for steelhead trout and that care had to be taken to ensure that the trout were protected. CEQA Guidelines, section 15065, subdivision (a) provides that an agency shall find that a project has a significant effect on the environment if the project has the potential to “reduce the number or restrict the range of an endangered, rare or threatened species.” The comments made during the administrative proceeding, together with the letter from American Canyon, should have alerted the County that the FSEIR needed to consider the impact of the Project on steelhead trout. The issue thus was adequately raised at the administrative level.

City of Walnut Creek v. County of Contra Costa (1980) 101 Cal.App.3d 1012 [162 Cal.Rptr. 224], cited by the County, is distinguishable. The project in that case was the construction of an apartment building. The City of Walnut Creek argued during the administrative proceedings that the project was inconsistent with the city’s general plan, and that its population density would exceed the density permitted on adjacent city land. In response to this argument, the county lowered the project density so that it was more compatible with the city’s general plan. In the trial court, however, the city for the first time contended that the project violated the county’s own general plan. The appellate court characterized the new argument as having “the effect of turning the administrative hearings into ‘shadow-play,’ in the sense that a key issue in the case [violation of the county’s general plan] was not specifically dealt with.” (Id. at p. 1021.) While it is true that the issue of steelhead trout was not specifically dealt with at the administrative level, unlike the situation in City of Walnut Creek, the issue was definitely identified at the administrative level.

It is irrelevant that it was never shown that recognizing the status of steelhead trout as an endangered species would necessitate major revisions to the FSEIR. As the trial court pointed out, the burden is not on the *385 objectors to show that a project will cause a significant effect on the environment. The burden is on the EIR to consider and decide if a project will cause a significant effect.

Therefore, once the issue of steelhead trout was identified, the burden shifted to the County to investigate the effect of the Project on that species. If the investigation resulted in a finding that the Project’s effect on steelhead trout would be significant, the FSEIR had to set forth the basis for that determination, had to analyze that effect and had to identify and analyze measures that could mitigate that effect. Thus, depending on the results of the investigation, it would be necessary to draft and circulate an amendment or supplement to the FSEIR, or to draft an addendum sufficient to inform the public of that conclusion and the reasons for it. 10

The County suggests that there was no need to consider the necessity of adopting measures to mitigate any effect of the Project on steelhead trout because the Updated Specific Plan simply readopted the setback requirements in the 1986 Specific Plan, which had been analyzed in the 1986 EIR. As the County argued in connection with the traffic issues, however, the mitigation measures adopted in 1986 are not binding. Indeed, the FSEIR does not simply modify the 1986 EIR, it replaces it. It therefore is subject to the same review as would be required of an original EIR. In addition, the need for an effective review is particularly compelling here, as steelhead trout were not listed as an endangered species in 1986, and it is not at all clear that steelhead trout were considered in the drafting of the 1986 EIR.

*386 Conclusion

The body that adopts general plan policies in its legislative capacity has unique competence to interpret those policies when applying them in its adjudicatory capacity. It follows that a reviewing court gives great deference to an agency’s determination that its decision is consistent with its general plan. (Save Our Peninsula Committee v. Monterey County Bd. of Supervisors, supra, 87 Cal.App.4th at p. 142.) “Because policies in a general plan reflect a range of competing interests, the governmental agency must be allowed to weigh and balance the plan’s policies when applying them, and it has broad discretion to construe its policies in light of the plan’s purposes. [Citations.] A reviewing court’s role ‘is simply to decide whether the [public] officials considered the applicable policies and the extent to which the proposed project conforms with those policies.’ [Citation.]” (Id. at p. 142.)

Nonetheless, “[w]hen the informational requirements of CEQA are not complied with, an agency has failed to proceed in ‘a manner required by law’ and has therefore abused its discretion. [Citations.]” (Save Our Peninsula Committee v. Monterey County Bd. of Supervisors, supra, 87 Cal.App.4th at p. 118.) In addition, “although the agency’s factual determinations are subject to deferential review, questions of interpretation or application of the requirements of the CEQA statute are matters of law. [Citations.] While we may not substitute our judgment for that of the decision makers, we must ensure strict compliance with the procedures and mandates of the statute. [Citation.]” (Ibid.)

We have concluded that the FSEIR failed to provide sufficient information as to the effects development of the Project might be expected to have on the region’s water supply and the need for treatment of wastewater. That conclusion necessarily invalidates the finding that the Updated Specific Plan is consistent with the General Plan’s conservation element. We have concluded that the Updated Specific Plan is inconsistent with the General Plan in failing to take affirmative steps to mitigate the impact the Project will have on traffic and circulation, including the failure to make an affirmative commitment to implement the traffic study. We have concluded that the Updated Specific Plan also is inconsistent with the General Plan in failing to take affirmative steps or make an affirmative commitment to mitigate the effect the Project will have on housing and growth, notwithstanding that no housing units will be built within the Project area. Finally, we find that the FSEIR is inadequate in its discussion of steelhead trout.

The inadequacies of the FSEIR necessarily invalidate the County’s certification of the FSEIR and adoption of the Updated Specific Plan. The *387 inconsistencies between the County’s General Plan and Updated Specific Plan also require a finding that the County abused its discretion in adopting the Updated Specific Plan. The judgment granting the petition for writ of mandate, therefore, is affirmed.

The Cross-appeal

Timeliness of Cross-appellants’ Attack on General Plan

Petitioners contended that the deletion of the traffic mitigation measures in the 1986 Specific Plan not only rendered the Updated Specific Plan inconsistent with the County’s General Plan, but invalidated the circulation element of the General Plan, creating an inconsistency between its circulation element and its land use element. The trial court ruled that any attack on the County’s General Plan was time barred by Government Code section 65009, subdivision (c)(1).

Government Code section 65009, subdivision (c)(1) provides:

“(c)(1) Except as provided in subdivision (d) [not applicable here], no action or proceeding shall be maintained in any of the following cases by any person unless the action or proceeding is commenced and service is made on the legislative body within 90 days after the legislative body’s decision:

“(A) To attack, review, set aside, void, or annul the decision of a legislative body to adopt or amend a general or specific plan. This paragraph . . . does apply to an action attacking a general plan or mandatory element thereof on the basis that it is inadequate.”

Petitioners interpret these provisions as providing a 90-day “window of opportunity” to attack a general plan on inadequacy grounds when there has been an action challenging a specific plan.

First, Petitioners are not attacking the County’s General Plan on inadequacy grounds. Their claim, rather, appears to be that the plan is internally inconsistent.

Second, Petitioners misconstrue Government Code section 65009. The express purpose of section 65009 is “to provide certainty for property owners and local governments regarding decisions made pursuant to this division.” Decisions made on the basis of the provisions of a general plan would provide no security if that plan becomes subject to change anytime a *388 specific plan is either adopted or amended. The relevant legislative history, although not unambiguous, also fails to support Petitioners’ interpretation. Section 65009, as it currently exists, was passed despite heated opposition claiming that it unreasonably limited the time for challenges to general plans notwithstanding that changes in circumstances over time could affect the adequacy of a general plan.

We therefore interpret Government Code section 65009 as providing that the decision to adopt a general plan or a specific plan may be attacked no later than 90 days after that decision was made. Similarly, a decision to amend a general plan or a specific plan may be brought no later than 90 days after that decision was made. Section 65009, however, should not be interpreted as permitting an attack on a general plan 90 days after a decision was made to adopt or amend a specific plan. The 90-day period is attached to the decision under attack, and to no other decision.

The cases cited by Petitioners in support of their argument are unpersuasive. In Garat v. City of Riverside (1991) 2 Cal.App.4th 259 [3 Cal.Rptr.2d 504], overruled on other grounds in Morehart v. County of Santa Barbara (1994) 7 Cal.4th 725, 743, fn. 11 [29 Cal.Rptr.2d 804, 872 P.2d 143], the court rejected a claim that the adoption of a measure amending a portion of a general plan permitted a challenge to the plan as a whole on grounds of inadequacy. The court found, in part, the under Government Code section 65009, “actions ‘attacking a general plan or mandatory element thereof on the basis that it is inadequate’ must be brought in the context of an action ‘[t]o attack, review, set aside, void, or annul the decision of a legislative body to adopt or amend a general or specific plan.’ ” (2 Cal.App.4th at p. 289.) Petitioners argue that the court thereby recognized that an action to amend a specific plan creates a window of opportunity to attack a general plan. The issue in Garat had nothing to do with specific plans and, at best, the cited language was nothing other than dicta. Moreover, the court in that case rejected a claim that an amendment of a general plan permitted an attack on unamended portions on the grounds of inadequacy, finding, “this limitation on actions challenging general plans is entirely consistent with the expressed legislative desire for finality and certainly set forth in section 65009, subdivision (a)(3).” {Ibid.) Petitioner’s contention that an amendment to a specific plan permits an attack on a general plan on grounds of inadequacy is completely inconsistent with this holding.

The Court in Kings County Farm Bureau v. City of Hanford (1990) 221 Cal.App.3d 692 [270 Cal.Rptr. 650] found only that defects in a general plan invalidated the decision to approve a project. Because the action involved an *389 attack on the decision to approve the project, as opposed to an attack on the general plan itself, the controlling statute of limitations was not Government Code section 65009, but “Government Code section 65907 which requires commencement of suit within 90 days after the date of the decision to approve the permit. [Citation.]” (221 Cal.App.3d at pp. 741-742, original italics.) Flavell v. City of Albany (1993) 19 Cal.App.4th 1846 [25 Cal.Rptr.2d 21] similarly involved a challenge to an ordinance as it related to the City of Albany’s general plan. Although the argument was made that the housing element of the general plan was legally inadequate, the challenge was to the ordinance; i.e., it was argued that the inadequacy of the general plan invalidated the ordinance. In the present case, the challenge has been brought against the County’s General Plan itself. Government Code section 65009, therefore, is the controlling statute of limitations.

We conclude, therefore, that the trial court correctly found Petitioners’ attack on the County’s General Plan to be time-barred. Moreover, even if Petitioners had the power to challenge the County’s General Plan, they cannot do so by arguing—as they do—that the adoption of the Updated Specific Plan renders the General Plan invalid. If the Updated Specific Plan is inconsistent with the General Plan, the Updated Specific Plan is invalid; but the General Plan is unaffected. The court in Lesher Communications, Inc. v. City of Walnut Creek (1990) 52 Cal.3d 531, 541 [277 Cal.Rptr. 1, 802 P.2d 317] established this point in connection with the claim that a zoning ordinance essentially amended a general plan, rendering it invalid: “The Planning and Zoning Law itself precludes consideration of a zoning ordinance which conflicts with a general plan as a pro tanto repeal or implied amendment of the general plan. The general plan stands. A zoning ordinance that is inconsistent with the general plan is invalid when passed [citations] and one that was originally consistent but has become inconsistent must be brought into conformity with the general plan. [Citation.] The Planning and Zoning Law does not contemplate that general plans will be amended to conform to zoning ordinances. The tail does not wag the dog. The general plan is the charter to which the ordinance must conform.”

Neighborhood Action Group v. County of Calaveras (1984) 156 Cal.App.3d 1176 [203 Cal.Rptr. 401], cited by Petitioners, similarly establishes only that an inadequate general plan may have the effect of invalidating a decision lower on the hierarchy of land use planning. “[T]he scope of authority of the agency to enact a general plan and zoning ordinances and to apply them is governed by the requirements of state law. A permit action taken without compliance with the hierarchy of land use laws is ultra vires as to any defect implicated by the uses sought by the permit.” (Id. at p. 1184, *390 original italics.) Petitioners, again, are not arguing that an alleged inadequacy in the General Plan renders the Updated Specific Plan invalid. They are arguing that the adoption of the Updated Specific Plan renders the General Plan invalid. They are indeed seeking to have the tail wag the dog.

The trial court correctly ruled that Petitioners’ attack on the County’s General Plan was time-barred.

Disposition

The judgment is affirmed. The order sustaining the County’s demurrer to petitioners’ complaints is affirmed. Each party will bear its own costs on appeal.

Strankman, J., * and Marchiano, J., concurred.

A petition for a rehearing was denied September 4, 2001, and on August 7, and September 4, 2001, the opinion was modified to read as printed above. The petition of appellant Napa County Board of Supervisors for review by the Supreme Court was denied October 17, 2001.

It has not been decided if the CEQA Guidelines are regulatory mandates or merely aids to interpreting CEQA; nonetheless, “[a]t a minimum, . . . courts should afford great weight to the Guidelines except when a provision is clearly unauthorized or erroneous under CEQA.” (Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376, 391, fn. 2 [253 Cal.Rptr. 426, 764 P.2d 278].)

The drafters’ approach is authorized by CEQA Guidelines, section 15130, subdivision (a)(4): “An EIR may determine that a project’s contribution to a significant cumulative impact is de minimus and thus is not significant. A de minimus contribution means that the environmental conditions would essentially be the same whether or not the proposed project is implemented.”

The “volume/capacity ratio” measures the ability of a roadway to handle the volume of traffic. If the amount of traffic on a roadway is equal to its capacity, the ratio will be 1.00 That figure is reduced as the ability of a roadway to handle the volume of traffic passing on it is reduced. A volume/capacity ratio between 0.80 and 0.89, for example indicates service at LOS D.

We take judicial notice of Napa County Board of Supervisors Resolutions Nos. 90-152 and 98-117, describing the mitigation fee. The County adopted the mitigation fee in 1990, “to provide for the orderly development of the area around the Napa County Airport and to provide sufficient road improvements to accommodate the traffic which will be generated by the new development.” As of February 28, 2000, $2,000,389.71 had been raised, including interest. These funds are earmarked for improvements to the Project’s internal roadways at such time as the development required them to be constructed. It also is to be used to pay for the area’s proportionate share of future improvements to two state highway intersections. Other improvements discussed in the 1986 EIR were not included in the fee program either because it was concluded that they were not needed, or because it was concluded that the benefit to the public outweighed the associated responsibility of the airport area developers.

CEQA Guidelines section 15130 requires a discussion of cumulative impacts on the environment when a project’s incremental effect is cumulatively considerable, and subdivision (a)(3) of that guideline provides that a project’s contribution to a significant cumulative impact may be rendered less than “cumulatively considerable if the project is required to implement or fund its fair share of a mitigation measure or measures designed to alleviate the cumulative impact.” The FSEIR, therefore, could have considered if the Project’s impact on traffic and circulation could be reduced to “less than cumulatively considerable,” by the adoption of a fee-based infrastructure program.

Petitioners cite to a memorandum from the County’s Department of Public' Works, asserting that it provides evidence that other localities are funding state highway improvements through local tax measures. In fact, the memorandum points out the difficulties trying to solve highway congestion by means of locally financed improvements, particularly when the congestion is the result of regional, rather than local, traffic patterns. The memorandum suggests that alternative routes for access be developed for those times when the state highways are congested. It recognizes that “Most of the work being done today on the inner Bay Area highways is being funded by locally approved sales tax measures. If Caltrans is not forthcoming with an appropriate solution to funding the capacity demands of their highway system in the future, the County may also need to consider similar alternative funding mechanisms to solve congestion in a more timely way.”

Petitioners, however, did complain that the FSEIR does not indicate how or when the proportional fair share is to be computed, if ever. They do not renew this complaint in their appellate briefs, it is not addressed by the County and it will not be addressed here other than to mention that such a computation reasonably might be a part of a discussion of fee-based mitigation measures.

The FSEIR’s analysis does not extend to the residential needs for water and treatment of wastewater tied to any housing units that may be constructed in other areas as a result of the Project. As discussed, ante, such an analysis is best left to the time when housing is proposed.

For example, Policy Guideline 2a(2) concerns the construction of three grade separated interchanges at the intersections of State Route 29 and State Route 12, State Route 221 and American Canyon Road. As to that project, the General Plan’s transportation system improve *376 ment program recommends that preliminary plans, cost estimates and an EIR be prepared from 1982-1987, that final construction documents be prepared from 1988-1992 and that the improvements be constructed from 1993-2000. The 1986 Specific Plan, similarly, set forth a suggested capital improvement of widening State Route 29 from four to six lanes between State Route 221 and American Canyon Road. The 1998 Updated Specific Plan does not include such an improvement.

Under CEQA, a lead agency may require the preparation of a subsequent EIR when “(a) Substantial changes are proposed in the project which will require major revisions of the environmental impact report. [H] (b) Substantial changes occur with respect to the circumstances under which the project is being undertaken which will require major revisions in the environmental impact report. ft[] (c) New information, which was not known and could not have been known at the time the environmental impact report was certified as complete, becomes available.” (Pub. Resources Code, § 21166; CEQA Guidelines, § 15162.) In addition, the lead agency may require the preparation of a supplemental EIR when the conditions which would require the preparation of a subsequent EIR are present but only minor additions or changes would be necessary to make the previous EIR adequate for the project as revised. (CEQA Guidelines, § 15163, subd. (a).) Both a subsequent EIR and a supplemental EIR must be circulated and reviewed in the same manner as the original EIR. (CEQA Guidelines, §§ 15162, subd. (d), 15163, subds. (c) & (d).) The public entity also may prepare an addendum to an EIR when changes to an EIR are necessary but none of the conditions described in CEQA Guidelines, section 15162, in connection with subsequent EIR’s, have occurred. (CEQA Guidelines, § 15164, subd. (a).) An addendum need not be circulated for public review, but can be included in or attached to the final EIR. (CEQA Guidelines § 15164, subd. (c).)

Retired Presiding Justice of the Court of Appeal, First Appellate District, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.