Opinion

WILLARD, J.

The real parties in interest, Wallover, Inc., and Hyatt Corporation (hereinafter sometimes jointly called Hyatt), seek to develop a resort hotel on 73 acres of beachfront land in Santa Barbara County. Respondent board of supervisors certified an environmental impact report (EIR) as complete, rezoned the site to permit the proposed use, and approved a preliminary development plan. Appellants, groups who oppose the development, sought a writ of mandate to set aside the certification and approval. They appeal from the judgment of the trial court denying their petition. We reverse.

Factual Background

Project Site: The project site, generally known as “Haskell’s Beach,” is located on the western edge of Goleta. It is bounded on the north by U.S. Highway 101 and the Southern Pacific Railway. To the east is the ARCO Ell wood oil processing facility and Sandpiper golf course. Westerly of the site is the Aminoil Dos Pueblos oil field and Ellwood pier (used by ARCO and Exxon for oil-related activities). The Pacific Ocean lies to the south. Directly across U.S. 101 from the project site is the 176-lot Rancho Embarcadero residential subdivision.

The project site has a central valley bounded by eastern and western terraces, each rising over 100 feet. Tecolote Creek runs southerly to the ocean through the western portion of the valley floor. It supports a narrow band of riparian vegetation and a small marsh. Eucalyptus trees border the northern edge of the property and screen the site from U.S. 101.

Site History: Haskell’s Beach was for many centuries the site of a major Chumash settlement. At least five archaeological sites remain, including three recorded Native American cemeteries, containing over two hundred known Chumash burials.

Portions of the site were used for oil development and subsequently dismantled in the 1950’s. Some remnants of the former development remain, such as old steel pilings visible on the beach during some months, and old concrete abutments and pipelines in other areas.

The area has a long history of informal recreational use by the general public, although public use rights have never been formally established.

Wallover acquired the property in 1968, along with the balance of the Embarcadero Ranch. A Wallover proposal to develop some 153 residential units at Haskell’s Beach was rejected by the county in 1979 as inconsistent with the county’s local coastal program. The issue of defining an appropriate use for the property soon became an area of major controversy during review of the county’s proposed coastal land use plan by the regional and state coastal commissions. The state coastal commission ultimately refused to approve the county’s proposed land use plan designation of the site as an “urban” area suitable for planned residential development, recommending instead that the site be designated for new visitor-serving facilities. The county refused to accept this designation and zoned the area for planned development, with the result that the site became a “white hole” in the county’s local coastal program—that is, an island remaining subject to concurrent county and coastal commission planning authority.

Hyatt and Wallover jointly filed an application for development of the site in 1983. The application resulted in certification of an EIR as being complete, approval of a preliminary development plan, rezoning of the land from PRD (planned residential development) to CV (visitor serving commercial), and major amendments to the county’s land use plan designation and development policies for the site.

The Project: The project, as amended by the applicant and reviewed in a supplemental EIR, is planned for development in two phases. Phase I is a resort hotel of 400 rooms and related restaurants, a conference center complex, parking areas, tennis courts, swimming pools, maintenance and service facilities, an artificial “lagoon,” and beachside snack bar and shower facilities. It also includes water wells and tanks to be constructed on Wall-over’s ranch property north of U.S. 101, water pipelines to the hotel site, and extensive landscaping. The access road for the project will cross two coastal streams—Tecolote Creek, on the property, and Bell Canyon Creek, along the property’s eastern border. There will be a new access road, requiring a substantial cut through one of the terraces, but resulting in the removal of the currently dangerous access road across Highway 101. Phase II is a future addition of 100 hotel rooms and 24 villas. It would require future discretionary approval.

The EIR: An environmental impact report was prepared for the project and certified as complete in September 1984. A supplement was prepared and certified the same month. The supplemental report was addressed to changes in the project to conform to the design described above, which deleted a tennis facility and 50 hotel units originally proposed to be located north of U.S. 101. A minor addendum to the EIR, listing the purported benefits of the project, was also certified. The EIR identified some 16 significant environmental effects which could possibly result from the project, including adverse impacts on visual aspects, archaeology, ethnic concerns (“cultural impacts”), transportation, water resources, air quality, biology, and others. The EIR proposed various mitigation measures intended to lessen the environmental effects of the project and examined four development alternatives for the project site, including a “No Project Alternative.” Among the alternatives considered was a scaled-down project consisting of 340 rooms. This alternative was rejected on the ground that it was not economically feasible. There was no detailed study made of the relative mitigation that could be provided by the 340-room alternative as compared to the 524-room plan or its first phase of 400 rooms. It was found, however, that the 340-room alternative would provide greater flexibility to avoid sensitive archaeological and biological resources.

Administrative Review of the Project: After public hearings, the planning commission voted to deny approval of the project. This decision was appealed to respondent board of supervisors, which held public hearings. On June 3, 1985, respondent granted the appeal and adopted an ordinance rezoning the site and a resolution approving preliminary plans for its development, subject to approximately 130 conditions imposed to mitigate adverse environmental impacts. It also adopted written findings and a statement of overriding considerations with respect to remaining impacts.

On December 19, 1985, the California Coastal Commission, after reviewing the project for consistency with the policies of the California coastal act, granted a coastal development permit subject to 14 special conditions. Subsequently, the coastal commission adopted findings in support of the permit.

Among the many conditions imposed by respondent and by the coastal commission are the following: The developers must dedicate a public easement for equestrian and hiking trails. The beach frontage which extends from the top of the bluff to the mean high tide line, and is approximately one-half mile long, is to be dedicated for public beach use and public access is to be provided. Hyatt will contribute a minimum of $100,000 to a regional hostel for low cost facilities available to the general public. Buildings must be in the California Mission style, not exceeding three stories. Improvements are to be located so as to interfere as little as possible with the Chumash archaeological sites and Tecolote Creek and its riparian habitat.

Additional facts are stated in connection with our discussion of the issues.

The Basic Issues

Some of the issues on appeal relate to requirements of the California Environmental Quality Act (CEQA). Others relate to the Santa Barbara County local coastal program, which has two components, the coastal zoning ordinance, and the land use plan. The issues, as framed by the parties, are summarized as follows:

Standard of Review

The parties agree that review of the preliminary development plan is subject to the provisions of Public Resources Code section 21168, which prescribes review pursuant to Code of Civil Procedure section 1094.5. (City of Carmel-by-the-Sea v. Board of Supervisors (1986) 183 Cal.App.3d 229, 239 [ 227 Cal.Rptr. 899 ].) The basic issue is whether respondent has committed a prejudicial abuse of discretion. This can be established by either of two showings: (1) that respondent failed to prepare an adequate EIR as required by statute (Rural Landowners Assn. v. City Council (1983) 143 Cal.App.3d 1013, 1018 [ 192 Cal.Rptr. 325 ]), or (2) that its findings were not supported by substantial evidence (Topanga Assn. for a Scenic Community v. County of Los Angeles (1974) 11 Cal.3d 506, 514-515 [ 113 Cal.Rptr. 836 , 522 P.2d 12 )].

(1) The legal duties imposed by CEQA are to be strictly enforced. (Mira Monte Homeowners Assn. v. County of Ventura (1985) 165 Cal.App.3d 357, 366 [ 212 Cal.Rptr. 127 ].) “Only by requiring the County to fully comply with the letter of the law can a subversion of the important public purposes of CEQA be avoided . . . .” (People v. County of Kern (1974) 39 Cal.App.3d 830, 842 [ 115 Cal.Rptr. 67 ].) The requirements of CEQA also must be “interpreted in such manner as to afford the fullest possible protection to the environment within the reasonable scope of the statutory language.” (Friends of Mammoth v. Board of Supervisors (1972) 8 Cal.3d 247, 259 [ 104 Cal.Rptr. 761 , 502 P.2d 1049 ].)

Technical perfection, however, is not required. Rather, EIR’s are subject to a court-developed standard for measuring their adequacy, known as the “rule of reason”: “An EIR should be prepared with a sufficient degree of analysis to provide decisionmakers with information which enables them to make a decision which intelligently takes account of environmental consequences. An evaluation of the environmental effects of a proposed project need not be exhaustive, but the sufficiency of an EIR is to be reviewed in the light of what is reasonably feasible. . . . The courts have looked not for perfection but for adequacy, completeness, and a good faith effort at full disclosure.” (Cal. Code Regs., tit. 14, § 15151.)

(2) In determining whether substantial evidence supports a finding, the court may not reconsider or reevaluate the evidence presented to the administrative agency. (Pub. Resources Code, § 21168.) All conflicts in the evidence and any reasonable doubts must be resolved in favor of the agency’s findings and decision. (Topanga Assn. for a Scenic Community v. County of Los Angeles, supra, 11 Cal.3d at p. 514 .)

In applying that standard, rather than the less deferential independent judgment test, “the reviewing court must resolve reasonable doubts in favor of the administrative findings and decision.” (Topanga Assn. for a Scenic Community v. County of Los Angeles, supra, 11 Cal.3d at p. 514 .)

Discussion

I

Alternative Site

The project as originally proposed encompassed 574 units, 50 of which, together with a tennis court, would be located north of U.S. 101. The EIR considered four alternatives: (1) no project; (2) clustered high-density residence development on the area south of the highway; (3) a low-density, 340-unit hotel and conference center on the area south of the highway; and (4) the revised proposed hotel development, limited to the area south of the highway. Alternative 4 was ultimately approved with the number of hotel rooms being 400, subject to subsequent discretionary approval of 100 additional hotel rooms and 24 villas. There was no in-depth consideration of an alternative site.

Consideration of alternatives is required by CEQA. The range of alternatives is governed by the “rule of reason,” which requires only an analysis of those alternatives necessary to permit a reasoned choice. An EIR need not consider an alternative, the effect of which cannot be reasonably ascertained and the implementation of which is remote and speculative. “The statute does not demand what is not realistically possible given the limitation of time, energy, and funds. ‘Crystal ball’ inquiry is not required. . . . An agency need not devote itself to an extended discussion of the environmental impact of alternatives remote from reality such as those which are of speculative feasibility . . . .” (Residents Ad Hoc Stadium Com. v. Board of Trustees (1979) 89 Cal.App.3d 274, 286-287 [ 152 Cal.Rptr. 585 ].)

Two California appellate court decisions have involved the question as to whether failure to consider an alternate site as an alternative was reasonable.

City of Lomita v. City of Torrance (1983) 148 Cal.App.3d 1062, 1069-1070 [ 196 Cal.Rptr. 538 ], rejected claims that an EIR on a master plan was insufficient because it failed to include, as an alternative, relocation of an airport to án alternate site. Applying a reasonableness standard, the court noted that the petitioners failed to explain how the alternative was reasonable or how it could feasibly attain the basic objectives of the project. This decision, in effect, placed the burden of proof upon the objector, a concept that is questionable. CEQA requires that governmental agencies consider reasonable alternatives. It is not limited to alternatives proposed and justified by objectors.

The other decision (San Bernardino Valley Audobon Society, Inc. v. County of San Bernardino (1984) 155 Cal.App.3d 738 . [ 202 Cal.Rptr. 423 ]) involved a proposed public cemetery consisting of 19,600 grave sites to be privately developed on 14 acres of land zoned for agriculture and designated by the general plan for residential development. The EIR stated that the Forest Service, in conjunction with the Nature Conservancy, had purchased nearby property to preserve sensitive plant habitat, and that the Forest Service desired to acquire the property proposed for the cemetery by trading Forest Service property of less biological and archaeological sensitivity. The report did not discuss the attributes of the alternate site or why it would or would not be a feasible alternative. The EIR was held to be inadequate. “Although EIRs are not required to be perfect or to discuss project alternatives beyond what is realistically possible (Foundation for San Francisco’s Architectural Heritage v. City and County of San Francisco (1980) 106 Cal.App.3d 893, 910 [ 165 Cal.Rptr. 401 ]), an EIR must produce information sufficient to permit a reasonable choice of alternatives so far as environmental aspects are concerned. (Residents Ad Hoc Stadium Com. v. Board of Trustees [supra] 89 Cal.App.3d 274, 287 .) . . . ffl] The EIR does not discuss whether there actually are other sites within the Big Bear area which would be suitable for such a project and do not contain the sensitive plant species habitat but simply that development at another site may result in similar impacts. Further, the EIR merely states that the Forest Service had proposed a land trade. It does not discuss the location or attributes of that property or why it would or would not be a feasible alternative. In short, the EIR does not contain the required sufficient degree of analysis to provide decisionmakers with information to allow them to intelligently take account of environmental consequences. (Santiago County Water Dist. v. County of Orange (1981) 118 Cal.App.3d 818, 831 .)” (San Bernardino Valley Audubon Society, Inc., supra, 155 Cal.App.3d at pp. 750-751 [ 173 Cal.Rptr. 602 ].)

We find no authority or rationale for an inflexible rule that the availability of other sites always must be considered or that it never need be considered. Situations differ; what is reasonable in one case may be unreasonable in another. It is necessary to examine the particular situation presented to determine whether the availability of other feasible sites must be considered in the EIR.

Hyatt contends that because it owns the Haskell’s Beach site and no other feasible site in the general area, it would be unreasonable to require consideration of another site as an alternative. (Cf. Horn v. International Bus. Mach. Corp. (1985) 110 App.Div.2d 87 [ 493 N.Y.S.2d 184 ].) Further, it suggests that its development of Haskell’s Beach as a hotel will not foreclose hotel development of any other feasible site.

Both of these contentions are questionable. From the viewpoint of the public interest, a visitor-serving development in the general area has been found to be desirable. Whether its location should be Haskell’s Beach or elsewhere depends upon the relative merits and demerits remaining after maximum amelioration of environmental impacts. Serving the public purpose at minimal environmental expense is the goal of CEQA. Ownership of the land used and the identity of the developer are factors of lesser significance.

It is also questionable to assume that development of two large hotels in the general area is feasible. Among environmental impacts presented by the proposed development are air pollution, water requirements, and vehicular traffic. One development might exhaust the tolerance. It would not be reasonable to fail to consider an alternative site based on the assumption that it, too, could be developed for the same general purpose. Reason requires that the agency charged with the duty to protect the environment compare impacts at feasible alternative locations.

We conclude that here, omission from the EIR of consideration of whether there was a feasible alternate site or sites was unreasonable and rendered the EIR inadequate, so as to make the respondent’s actions with regard to it a prejudicial abuse of discretion.

II

340-Room Alternative

The second issue for determination is whether respondent abused its discretion in not approving the alternative of a 340-unit project.

Substantial Evidence: Such a scaled-down project was considered as an alternative in the EIR but was found to be economically infeasible. The trial court found that the record failed to contain substantial evidence to support such a finding. Our review of the administrative record leads us to agree with the trial court’s statement: “Nowhere in the record can the Court find any evidence which analyzes that alternative in terms of comparative costs, comparative profit or losses, or to the extent appropriate, comparative economic benefit to the Respondent, nearby communities, or the public at large.”

Hyatt contends that substantial evidence of economic infeasibility is presented by estimates of annual revenues, infrastructure costs and overall project costs. None of the figures purports to relate to estimated costs, projected income, or expenses for the 340-unit alternative. They provide no basis for a comparative analysis between the project actually approved and the 340-unit alternative. In the absence of such comparative data and analy sis, no meaningful conclusions regarding the feasibility of the alternative could have been reached. (Cf. Burger v. County of Mendocino (1975) 45 Cal.App.3d 322, 326-327 [ 119 Cal.Rptr. 568 ].)

It is also suggested that economic infeasibility may be inferred from the fact that construction costs per room may go up as project size is reduced. Assuming that this conclusion is valid, it is not sufficient to show that further reductions in project size would render the project infeasible. CEQA defines “feasible” as “capable of being accomplished in a successful manner within a reasonable period of time, taking into account economic, environmental, social, and technological factors.” (Pub. Resources Code, § 21061.1.) The fact that an alternative may be more expensive or less profitable is not sufficient to show that the alternative is financially infeasible. What is required is evidence that the additional costs or lost profitability are sufficiently severe as to render it impractical to proceed with the project. The scant figures contained in the administrative record are not sufficient to support such a conclusion.

Hyatt also contends that substantial evidence supports respondent’s subfinding that implementation of the 340-unit would require complete redesign and revision of the project. Assuming this to be true, it does not follow that the project would be rendered economically infeasible.

Another contention is that there is evidence in the record that the 340-unit project would be infeasible due to “social or other,” as opposed to purely economic, reasons. The reasons suggested are that reductions in project size would reduce the “beneficial” aspects of the project, such as property taxes received by the county, provision of employment, and the number of rooms created for hotel visitors, or otherwise make the project less desirable. This argument misses the point. The only finding made by respondent was that the alternative was economically infeasible. Judicial review extends to determining whether this finding is supported by evidence, not to fabricate new findings to support a decision. (Sierra Club v. City of Hayward (1981) 28 Cal.3d 840, 859 [ 171 Cal.Rptr. 619 , 623 P.2d 180 ].)

It is also asserted that respondent did in fact find the 340-unit alternative infeasible for other than economic reasons. The language referred to is simply a recital in the preamble to respondent’s actual findings regarding project alternatives. By its terms the language simply directs the reader to “following” specific reasons; the only such specific reason cited for the 340- unit alternative is “economic” infeasibility. It is obvious from its context that this language in itself was not intended as a finding.

Another suggestion is that statements in respondent’s statement of overriding considerations should be deemed to be the equivalent of findings that the 340-unit alternative was infeasible for “social or other” reasons. The statement of overriding considerations, however, makes no attempt to compare the costs and benefits of the 340-unit project with those of the project actually approved; it does not explain why the smaller project is infeasible. It merely recites the reasons why the project was deemed acceptable in spite of its major impacts, not why no smaller project would be acceptable. This is not an adequate substitute—nor was it ever intended to be—for findings addressing the feasibility of the 340-unit alternative. (Cf. Resource Defense Fund v. Local Agency Formation Com. (1987) 191 Cal.App.3d 886, 896-898 [ 236 Cal.Rptr. 794 ].)

Substantial Mitigation: The trial court found that respondent’s finding that the 340-room alternative was economically infeasible lacked the support of substantial evidence. It also ruled that CEQA did not require such a finding of infeasibility because adverse environmental impacts of the 524-room project had been “substantially lessened” through mitigation measures.

Section 21002 of the Public Resources Code adopts the policy that proposed projects should not be approved “if there are feasible alternatives or feasible mitigation measures available which would substantially lessen . . . significant environmental effects . . . .” It goes on to state that CEQA procedures are designed to assist in the identification of “feasible alternatives or feasible mitigation measures which will avoid or substantially lessen such significant effects.” Relevant guidelines prohibit approval of a project for which an EIR was prepared unless the agency has “(A) Eliminated or substantially lessened all significant effects on the environment where feasible . . . , and [ft] (B) Determined that any remaining significant effects on the environment found to be unavoidable . . . are acceptable due to overriding concerns . . . .” (Cal. Code Regs., tit. 14, § 15092.)

The trial court construed this language to authorize any one of three alternatives: (1) to require avoidance, (2) to require mitigation, or (3) to require substantial lessening of environmental impacts, without comparing the merits of these alternatives and without selecting the one that best protected the environment. It held that substantial lessening of environmen tal impacts by the conditions respondent had imposed satisfied the statutory requirement, and therefore consideration need not be given to an alternate plan even if that plan would go farther in lessening or mitigating the impacts, or even avoiding them altogether.

The language of CEQA quoted above—that projects should not be approved “if there are feasible alternatives or feasible mitigation measures available which would substantially lessen . . . environmental effects”— does not expressly state that either alternative is acceptable regardless of its relative effectiveness in mitigating environmental impacts. We are asked by appellants to consider the legislative intent and from it to find that CEQA prohibits approval of a plan that has mitigating features when a feasible plan with less environmental impact is available as an alternative.

Such a finding, however, is not necessary to a determination of this appeal. As explained in the next section of this opinion, Santa Barbara County’s local coastal program does expressly require adoption of the feasible plan with the least substantial environmental impacts. Respondent board, in rejecting the 340-room alternative as economically infeasible, did not compare the environmental impacts of the 524-room and 400-room plans, as modified by conditions to mitigate impacts, with the 340-room plan as it likewise could have been modified by mitigating conditions. Respondent should have considered the feasibility of the smaller development. If found to be feasible, then the plans should have been compared to determine their relative environmental impacts. Under the local coastal program respondent’s decision to approve the preliminary plan was not justified.

Ill

Local Coastal Program

Pursuant to the mandate of the coastal act, Santa Barbara County adopted a local coastal program, consisting of a land use plan and a coastal zoning ordinance. The land use plan was approved by the coastal commission, except for details relating to Haskell’s Beach, which, as indicated above, was left as a “white hole,” subject to concurrent county and coastal commission planning authority.

Maximum Feasible Mitigation-. The coastal zoning ordinance provides that a development plan shall be approved only if “adverse impacts are mitigated to the maximum extent feasible.” (§ 35-174.7, subd. 1b.) This requirement is free from any ambiguity present in CEQA. Imposition of conditions to partially ameliorate adverse environmental impacts of the proposed project does not excuse failure to evaluate the alternative scaled-down alternative. Inasmuch as there was no substantial evidence to support respondent’s finding that the alternate design was economically infeasible, further consideration at the administrative level is required.

Tecolote Creek Buffer Zone: The zoning ordinance contains the following provision relating to stream habitats: “1. The minimum buffer strip for streams in rural areas, as defined by the Coastal Land Use Plan, shall be presumptively 100 feet, and for streams in urban areas, 50 feet. These minimum buffers may be adjusted upward or downward on a case-by-case basis. The buffer shall be established based on an investigation of the following factors and after consultation with the California Department of Fish and Game and California Regional Water Quality Control Board in order to protect the biological productivity and water quality of streams: [fl] a. Soil type and stability of stream corridors, [fl] b. How surface water filters into the ground. fl[] c. Slope of land on either side of the stream, [fl] d. Location of the 100-year flood plain boundary. [^|] Riparian vegetation shall be protected and shall be included in the buffer. Where riparian vegetation has previously been removed, except for channelization, the buffer shall allow for the re-establishment of riparian vegetation to its prior extent to the greatest degree possible. . . .” (§ 35-97.19, subd. 1.)

The site is within urban territory, and the presumptive 50-foot minimum buffer zone applies. To meet this requirement, the preliminary development plan provided a minimum buffer zone of 50 feet, intruded by a tennis court, pedestrian bridge, paths leading to the bridge, and existing bridge abutments. A condition of approval of the preliminary plan required removal of the tennis court from the 50-foot zone. In certain areas the buffer was 100 feet and in others 200 feet. Respondent’s staff recommended a minimum buffer of 100 feet. The preliminary development plan was approved, notwithstanding the staff recommendation.

Our attention has not been called to anything in the voluminous record that shows consideration of adjustments in the zone width based on investigation of factors mandated by the ordinance. We construe the ordinance to require such investigation and adjustment of the presumptive width for buffer zones where appropriate.

Stream Crossings: The preliminary development plan proposed that access road crossings of Bell Canyon and Tecolote Creeks be provided by placing culverts in the stream beds and importing fill material for approaches. The land use plan provides: “No structures shall be located within the stream corridor except: public trails, dams for necessary water supply projects, flood control projects . . . ; and other development where the primary function is for the improvement of fish and wildlife habitat. Culverts, fences, pipelines, and bridges (when support structures are located outside the critical habitat) may be permitted when no alternative route/location is feasible. All development shall incorporate the best mitigation measures feasible.” (§ 3.9-38.)

By way of mitigation, respondent imposed the following condition to its approval: “The project sponsor would be required to obtain a Streambed Alteration Agreement from the California Department of Fish and Game for any alteration of the Tecolote Creek or Bell Canyon Creek channels. The Department of Fish and Game has direct jurisdiction under Section 1601-03 of the Fish and Game Code over any proposed activities that would substantially divert or obstruct the natural flow or substantially change the bed, channel, or bank of any creek. The Streambed Alteration Agreements may impose conditions to partially mitigate the project’s adverse effects on the creeks. The provisions of this section of the Code are intended to protect and conserve fish and wildlife resources.”

The revised EIR stated that the culvert plan, together with related features of the proposed access road, would result in a moderately adverse impact which could be partially mitigated by conditions that might be imposed by the Department of Fish and Game.

The county staff recommended that the stream crossings be by means of bridges with supporting structures located outside the critical areas. The trial court found no violation of the local coastal program because the adverse effects would be insignificant. Such finding lacks evidentiary support.

Respondent should have considered these alternatives. If both were found to be feasible, respondent should have required adoption of the alternative that provided the greater mitigation of adverse effects, assuming there was any significant difference.

Archaeological Sites

The land use plan adopts policies concerning development of sites of cultural significance. “All available measures . . . shall be explored to avoid development on significant historic, prehistoric, archaeological, and other classes of cultural sites. fl[] When developments are proposed for parcels where archaeological or other cultural sites are located, project design shall be required which avoids impacts to such cultural sites if possible, [fl] When sufficient planning flexibility does not permit avoiding construction on archaeological or other types of cultural sites, adequate mitigation shall be required.” (§ 3-10.4.) In addition, the coastal act provides, “Where development would adversely impact archaeological. . . resources as identified by the State Historic Preservation Officer, reasonable mitigation measures shall be required.” (Pub. Resources Code, § 30244.)

Haskell’s Beach contains several such sites of varying sensitivity. In this connection, respondent imposed numerous conditions to its approval. These included the preparation of a cultural resources management plan for the implementation of specified conditions. Major areas of cultural significance where burials exist are to be avoided. Some such areas will be filled and paved for parking lot use.

Appellants’ basic contention in this regard is that the policies require avoidance of such sites, if possible, not just mitigation, and that only if such avoidance is infeasible is “mitigation” permitted. They point out that a smaller development, the 340-room alternative, would provide greater flexibility for avoidance of such sites. They also complain that construction of a vehicle parking lot over a burial site, while it may protect against physical degradation, does not mitigate the religious and cultural concerns of present day Native Americans. Hyatt replies that it has designed the improvement to minimize impact on the sites, particularly the important and sensitive ones, to the maximum extent consistent with the development.

The local coastal program requires reduction of adverse impacts to the maximum feasible extent. It requires that project design avoid such impacts, if possible. It follows that the requirement to consider the feasibility of a design involving a 340-room development applies to cultural as well as environmental impacts. The economic feasibility of such a design should have been studied. Without such a study the preliminary plans for the development run afoul of the local coastal program.

Collateral Estoppel-. Inasmuch as the coastal commission had not approved Santa Barbara County’s land use plan insofar as it dealt specifically with Haskell’s Beach, the coastal commission’s approval of Hyatt’s proposed development was required. (Pub. Resources Code, §§ 30600, subd. (c), 30601; Del Mar v. California Coastal Com. (1984) 152 Cal.App.3d 49, 52 [ 199 Cal.Rptr. 225 ].) The coastal commission held a hearing on the plan and found the project, subject to 14 new conditions, to be in full compliance with coastal act policies and Santa Barbara County’s local coastal program. It also found that the project would not have any significant adverse impacts within the meaning of CEQA. Respondent claims that this collaterally estops appellants from litigating the issues arising under the local coastal program.

Public Resources Code section 30604, subdivision (a), which governs the commission’s independent permit authority in such situations, provides that the commission shall issue a coastal development permit for new development if the commission finds that the project is 1) “in conformity” with the general development provisions of chapter 3 of the coastal act (Pub. Resources Code, § 30200 et seq.) and 2) “will not prejudice the ability of the local government to prepare a local coastal program that is in conformity with the provisions of Chapter 3” of the act. We find nothing in the coastal act that requires or permits the commission to approve or disapprove a project on the basis of consistency with local regulations.

The question of whether the proposed project will “prejudice the ability of the local government to prepare a local coastal program” is distinguishable from those issues in which Hyatt here claims collateral estoppel, for example, whether Hyatt should be permitted to place culverts and fill instead of bridges across the Tecolote and Bell Canyon streams and whether the project encroached more than necessary upon Native American archaeological sites. These issues are of considerable importance on the specific site planning level, but they obviously have no bearing on the “ability of the local government to prepare a local coastal program.” The notion that this provision of section 30604, subdivision (a) mandates detailed conformity with anticipated local regulations was expressly rejected in Billings v. California Coastal Com. (1980) 103 Cal.App.3d 729, 744-745 [ 163 Cal.Rptr. 288 ].

The commission’s finding regarding the alleged consistency of the project with the county’s local coastal program and its finding that the project would not have any significant adverse impacts within the meaning of CEQA were not necessary to the decision before the commission. Collateral estoppel may be applied only to an issue “ ‘. . . necessarily decided at the previous [proceeding which] is identical to the one which is sought to be relitigated ....’” (People v. Sims (1982) 32 Cal.3d 468, 484 [ 186 Cal.Rptr. 77 , 651 P.2d 321 ], fn. omitted.)

Appellants have discussed a number of other factors which they contend foreclose collateral estoppel. Inasmuch as the doctrine is inapplicable for the reason discussed above, it is unnecessary to consider the additional factors.

Disposition

The judgment of the trial court is reversed. The matter is remanded for issuance of a mandatory writ of mandate consistent with our discussion.

Appellants are entitled to costs on appeal.

Stone (S. J.), P. J., and Abbe, J., concurred.

A petition for a rehearing was denied February 19, 1988, and the opinion was modified to read as printed above. The petition of real parties in interest and respondents for review by the Supreme Court was denied April 27, 1988. Mosk, J., and Kaufman, J., were of the opinion that the petition should be granted.

Retired judge of the Superior Court sitting under assignment by the Chairperson of the Judicial Council. The EIR states: “Development of a project similar to the proposed project but located at an alternative site is not addressed because the project sponsor has no holdings of similarly situated property susceptible to development.” “The Legislature further finds and declares that it is the policy of the state to:. . . [][] (g) Require governmental agencies ... to consider alternatives to proposed actions affecting the environment.” (Pub. Resources Code, § 21001, subd. (g).) “All . . . boards . . . shall . . . certify the completion of ah environmental impact report.... Such a report shall include a detailed statement setting forth ...:...[]]] (d) Alternatives to the proposed project.” (Pub. Resources Code, § 21100, subd. (d).) Guidelines to CEQA state that the EIR shall describe “a range of reasonable alternatives to the project, or to the location of the project . . . .” (Cal. Code Regs., tit. 14, § 15126, subd. (d).) Hyatt contends that the $10 million “infrastructure” costs will remain the same regardless of project size, and moreover appears to assume that the entire estimated $75 million cost of construction will remain the same regardless of project size. Neither of these suppositions is supported by evidence in the record. There is evidence that the county never interpreted its ordinance to require more than is required by CEQA. Such interpretation is entitled to great weight. (Morris v. Williams (1967) 67 Cal.2d 733, 748 [ 63 Cal.Rptr. 689 , 433 P.2d 697 ]; Sanchez v. Unemployment Ins. Appeals Bd. (1977) 20 Cal.3d 55, 67 [ 141 Cal.Rptr. 146 , 569 P.2d 740 ].) Nevertheless, final responsibility for interpretation rests with the courts. (Ibid.) Administrative interpretation must give way to plain and unambiguous statutory wording. (Sutton v. Industrial Acc. Com. (1956) 46 Cal.2d 791, 797 [ 298 P.2d 857 ]; Sacramento Typographical Union No. 46 v. State of California (1971) 18 Cal.App.3d 634, 638 [ 96 Cal.Rptr. 194 ].) The land use plan contains identical provisions. (§ 3.9-37.) An identical provision is contained in the coastal zoning ordinance. (§ 35-97.19, subd. 2.)