The Pocket Protectors v. City of Sacramento

*908 Opinion SIMS, Acting P. J. Plaintiff The Pocket Protectors, an unincorporated association, appeals from the denial of its petition for writ of mandamus under the California Environmental Quality Act (CEQA) (Pub. Res. Code, § 21000 et seq.; undesignated section references are to the Public Resources Code). The Pocket Protectors seek the preparation of an Environmental Impact Report (EIR) for a residential project called The Islands at Riverlake proposed by real party in interest Regis Homes of Northern California, Inc. (Regis). Defendant Sacramento City Council voted to approve the project with a mitigated negative declaration (MND), overriding rejection of the project by the city planning commission. 1 Agreeing with The Pocket Protectors that substantial evidence exists to support a fair argument the project may have a significant effect on the environment, we shall reverse. FACTUAL AND PROCEDURAL BACKGROUND The history of the project site The proposed project occupies 20.6 acres of undeveloped land in the Pocket area, a region of 4.5 square miles within the City of Sacramento (City). The project site consists of separate narrow parcels running roughly a mile altogether on both sides of Pocket Road (north and south), between East and West Shore Drives; however, the bulk of the site is on the north side of Pocket Road. Forty-foot-wide landscaped parkways (including a 15-foot-wide parkway easement and a 25-foot-wide landscape easement) create greenbelts adjacent to Pocket Road the entire length of the site. The Pocket area was developed residentially beginning in the 1960’s in accordance with a general development plan adopted in 1965. Specific plans and a South Pocket Area Community Plan were adopted in 1976. In 1985, the City Council approved the “L & P—Pacific Teichert Planned Unit Development” (LPPT PUD) to cover 373 acres within the scope of the South Pocket Area Community Plan, including the project site. The resolution approving the PUD, which declared itself binding on all persons intending to develop any portion of the property, stated that all development should conform to the attached LPPT Development Guidelines. The LPPT PUD incorporated a variety of housing types, including “Single Family,” zoned R-l, and “Townhouse (or similar development),” zoned *909 R-l A. 2 (The project site is within the R-lA-zoned part of the PUD.) The LPPT Development Guidelines stressed the importance of developing all the proposed housing types as part of “an interrelated total environment” throughout the PUD. Before the current project was proposed, the City Council had approved two unconsummated plans to develop the site. The first, submitted in 1987, would have constructed 155 clustered townhouse units; the second, submitted in 1994, would have constructed 167 clustered townhouse units. The record does not reveal why neither project was built. The developer and the City executed a development agreement for the original proposed project, which was extended until August 25, 2002, for the second proposed project. The development agreement stated in part: “If Developer wishes to develop as single family residential one or more portions of the project zoned R-l A or for multifamily use, it may do so, in which case the portion or portions shall be rezoned R-l ... .” By the time the current project was proposed, the surrounding area was fully developed with housing. All the housing types called for in the PUD and its development guidelines had been built, except for townhouses. The proposed project The project proposal and the MND approved by the City Council differed only in points of detail from those previously rejected by the planning commission. We first discuss the project’s history before the planning commission, then its subsequent history before the City Council. THE PLANNING COMMISSION The original project According to Regis’s application to the City, submitted on October 19, 2001 (while the development agreement for the previous unbuilt project was still in effect), Regis would construct 143 units of 3- and 4-bedroom detached single-family housing, ranging from 1,800 to 2,500 square feet, at a density *910 of 6.68 units per net acre. 3 A special permit would be required to build this type of housing within the LPPT PUD. (Regis did not request a rezoning to R-l.) Each part of the project site (north and south of Pocket Road, respectively) would be developed with two rows of wide but shallow lots, bisected by a private street 25 feet wide running the parcel’s full length. As the City’s standard right-of-way street width is 41 feet, a subdivision modification would be required. The houses would have minimal setbacks. In front, there would be zero setbacks from the property line. In back, the houses in the rear rows would be set back as little as five feet from the fence lines of the existing homes abutting the project. Staff response City staff responded supportively to the first version of Regis’s submittal in April 2001, but noted certain problems. In a communication to Regis, staff observed that the project “does not fulfill the intent of the LPPT PUD Townhouse land-use designation insofar as it does not incorporate the landscaping and open space concepts embraced by the remainder of the LPPT PUD”; however, it was within the allowed density, and “[t]he unusual shape of the subject site presents a number of design challenges/opportunities which the applicant is willing to address.” Staff also observed that, as designed, the project proposed long expanses of similar building massing, creating a “[c] any on” effect which needed to be minimized with varying heights and facades and a “thematic landscaping plan.” In keeping with this point, staff recommended the planting of one shade tree per 30 lineal feet of street frontage. In addition, staff noted that “[t]he necessity of shallow lots for the proposed project limits the amount of privacy afforded adjacent property owners” and recommended configuring second-story windows to minimize views into the rear yards of existing homes. 4 After Regis had revised its original submittal and produced the proposal shown in its October 2001 application to the City, staff continued to support *911 the proposal. On June 25, 2002, the City issued a notice of availability/intent to approve a negative declaration for the project. The Initial Study/MND The City circulated a draft initial study/MND (IS/MND) from June 25, 2002, through July 25, 2002, and allowed public comments through July 29, 2002. The City published responses to the comments on August 7, 2002. The IS/MND found the project could cause significant impact on the environment only as to air quality, biological resources, and cultural resources, which could all be mitigated to a level of insignificance. As to all other relevant variables, including “land use/planning” and aesthetics, the IS/MND found no potential for significant impact on the environment. On these two variables, the IS/MND said the following, inter alia: LAND USE/PLANNING The Pocket Area Community Plan—South Pocket Specific Plan (PACP-SPSP) sets goals, objectives, and policies for the area within its scope. The LPPT PUD designates the project area for “townhouses and similar development”—i.e., cluster and row housing. Because the proposed project would build single-family detached housing instead, its consistency with overall community growth goals and policies, and the goals and policies for single-family and townhouse and related development of the PACP-SPSP, must be assessed. The Rl-A zone is a “Single-Family Alternative Zone,” “intended to permit the establishment of single-family, individually owned, attached or detached residences where lot sizes, height, area and/or setback requirements vary from standard single-family. This zone is intended to accommodate alternative single-family designs which are determined to be compatible with standard single-family areas and which might include single-family attached or detached units, townhouses, cluster housing, condominiums, cooperatives or other similar projects.” PUD’s are intended to permit flexibility in the design of integrated developments and to encourage “creative and imaginative planning.” Special permits shall be granted at the discretion of the planning commission or city council “upon sound principles of land use,” provided that a special permit use does not cause detriment to the public health, safety, or welfare, and that it “complies] with the objectives of the general or specific plan for the area in which it is to be located.” *912 The project does not conflict with general plan designation, zoning, or PACP-SPSP land use goals and policies. “The PUD designation allows some flexibility in the configuration of parcels [, and t]he Planning Commission will decide if the proposed lot configurations are acceptable.” The planning commission will also decide if the proposed narrower-than-standard private road is acceptable, and the relevant City departments will review the plans to ensure access and convenience. A project of detached single-family homes on this site is not inconsistent with the LPPT PUD or incompatible with neighboring homes. “The project site was identified in the LPPT [PUD] Guidelines ... for development of townhouses. In July 1993, the City of Sacramento approved a project to develop the planned townhouses. The approvals for the 1993 townhouse project expired and the City is considering the development of single-family homes in place of the townhouse concept. In addition to City review by the Planning and Building Department, Public Works Department, Utilities Department, Planning Commission, and State review through CEQA, the River-lake Community Association and the Architectural Review Committee and other effected [sic] neighbors have been reviewing the proposed project.[ 5 ] [f] . . . [][] By approving the LPPT PUD, the City determined that the townhouse and similar developments planned for the Islands at Riverlake Project site are compatible with the single-family residential homes on the surrounding parcels. The City is working with the Riverlake Community Association and Architectural Control Committee to review plans to ensure that the proposed development conforms to the minimum design standards set in the LPPT PUD Development Guidelines. To ensure that the homes are not hazardous to the residents and neighbors and are constructed safely, the Islands at Riverlake Project would be constructed in accordance with the City Public Works, Utilities, and Planning and Building departments standards and specifications provided in the Sacramento City Code. To reduce potential privacy impacts to the residents bordering the site, the Project proposed design features that eliminate second-story windows and minimize site-line visibility from first-story windows.” AESTHETICS The project would not “[ajffect a scenic vista or scenic highway!],” “[h]ave a demonstrable negative aesthetic effect[],” or “[c]reate light and glare[].” R-l zoning requires maximum heights of 35 feet, maximum lot coverage of 40 percent, front yard setbacks of 25 feet, back yard setbacks of 15 feet, *913 interior sides setbacks of five feet, and street side setbacks of 12.5 feet. However, R-1A zoning “allows flexibility in setback and lot coverage requirements.” “The existing character of the site is a seasonally disced vacant lot, bordered on one side by residential development and by a 40-foot wide linear parkway on the other. The character of the Islands at Riverlake Project is congruous with the neighboring residential development. Passers-by of the Project on Pocket Road would not consider the development visually disruptive because urban residential development is a common and accepted part of the landscape in the City. However, the Islands at Riverlake Project could conflict with the visual expectations of the residents living in the homes adjacent to the Project. The Project would figure prominently in the foreground of the view shed from the rear of the neighboring houses. “The LPPT PUD utilizes two tools to minimize the extent of visual impacts: 1) The LPPT [PUD] Guidelines, and 2) The Riverlake Community Association and Architectural Control Committee. . . . The City of Sacramento determined that the proposed project is consistent with the LPPT PUD Development Guidelines designation. The Riverlake Community Association and Architectural Control Committee reviewed initial design plans to verify that the proposed development conforms to the minimum design standards set in the LPPT PUD Development Guidelines. The Riverlake Community Association provided comments, which were incorporated into the project. “The Planning Commission has discretion over design through the Special Permit required by development within a PUD and the Rl-A zone. Staff evaluates design through the Special Permit process. City review of the design plans helps to minimize negative visual impacts.” Public comment Many neighbors of the proposed project objected to it before the IS/MND appeared. A petition eventually signed by 486 neighborhood residents and forwarded to the City alleged: “[T]he following issues have not been satisfactorily resolved: housing proposed is too dense, appropriate set-backs [ric] (at least 30 feet) have not been met, traffic issues have not been resolved, strain on community services has not been assessed, use of common greenbelt as front yards for new homes, why the cluster home concept was abandoned.” By the end of the public comment period, the protesters had organized as The Pocket Protectors. Their comments on the IS/MND included the following points, among others: *914 The project needed to be rezoned to R-l and to meet all R-l standards, including 15-foot rear setbacks as specified in the development agreement; it should also be required to meet the standards of the Riverlake community, which might be even more restrictive. Two-story houses with five-foot setbacks from existing homes would create serious problems of noise, privacy, and visual impact. The MND acknowledged the major visual impact of the project on neighboring homes. Yet it asserted this major impact need not be mitigated. Neighbors had bought homes or land near the project site in reliance on the approved PUD, including its plan for cluster homes fronting Pocket Road, and the development agreement for this site, which was still in force. Although the IS/MND, in keeping with the PUD, spoke of the need for alternatives to conventional housing, the proposed homes were entirely conventional except that they would be built on substandard lots with a substandard private street and substandard setback requirements. The project would have major impacts on the preservation of open space and trees, which the MND did not adequately address. The MND identified planting shade trees along the private streets as a mitigation measure, but the project did not allocate enough land alongside the private streets to make this possible. Furthermore, if they were required to have four-foot sidewalks, this would either cause intrusion into the greenbelt in front or cut across driveways in the rear and eat up needed parking space. The new homes would have virtually no setbacks along the greenbelt, and any new sidewalk in front would encroach onto the greenbelt. The project’s appearance—“two rows of uninterrupted houses, fronting on a highly traveled and dangerous street, divided by a substandard width road, within arms [sic] distance of existing homes”—was “subpar design” and “a poor attempt to fit 10 pounds of nails into a 5-pound bag.” Contrary to the IS/MND, the project was not consistent with the intended purpose of the land or with proper land use and design for the City. It had no internal open spaces. The subdivided parcels would be too small for landscaping and tree planting. There would be a major parking problem because driveways on one side would be too narrow to use as parking spaces and no street parking would be possible. Contrary to the IS/MND, the neighborhood did not support the project. In voting to approve it, the Riverlake Community Association board had ignored the overwhelming majority’s express wishes. *915 The Planning Commission hearing The planning commission held a public hearing on the project on August 8, 2002. Regis and City staff made presentations, and many members of the public (Pocket Protectors and others) testified. Tom Pace, a senior City planner, testified that the City did not believe downzoning was required. The R-1A zone was flexible and used for all types of small lot development, not necessarily attached housing. And when the development agreement stated rezoning would be required to develop “single family residential,” this meant only “standard single family development,” i.e., “single family homes developed on lots with 5200 square foot [sic] which is our standard R-l requirement.” Since the lots in this project would be smaller, it could be built under the current zoning. 6 Pace described a number of changes to the project and new conditions imposed by the City since Regis’s application was originally submitted. The number of lots had been reduced from 143 to 139. Two-story houses abutting existing homes would be allowed only at the ends of cul-de-sacs, partly in order to limit the number of houses with five-foot rear setbacks adjacent to the property line; if two-story, such houses would have no upper-story windows. Houses built side by side would have to be of different elevations. Conditions intended to improve parking and driveway designs had been imposed. Contrary to the developer’s original intent, staff had insisted on a four-foot-wide sidewalk (one foot narrower than the standard residential sidewalk) beside the private street, which would remove half of the front yard setback for homes facing that street. Of the 23 trees proposed to be removed (out of 503 in the project area), only two were protected under the City’s tree ordinance and the City Arborist had required mitigation for their removal. Pace testified that the setback issue was troubling, but given the project’s design the problem could not be significantly alleviated. Regis wanted to create a “single family detached appearance from the . . . public street” while maintaining “something close to the approved density.” The only way to achieve this end within the “very narrow strip” of developable land between the greenbelt and the existing homes (only 120 feet altogether) was to group houses in two tiers with a “common driveway” in between (much narrower than the standard 41-foot-wide city street). 7 If the houses were built smaller to save open space, neighbors might become concerned about the new *916 residents’ income levels and the project’s effect on property values. 8 Pace acknowledged the site was narrow because the PUD had designated it for development with a single tier of townhouses or “manor houses” (three or four attached units designed to look like one large house). Bill Heartman, representing Regis, testified that its design for “mini-mansions” was “more in keeping . . . with the neighborhood” than either single-family homes on narrow lots or attached housing, but could be executed only by double-tiering the houses. Heartman also noted that attached housing generally has a lower market value than detached housing. Members of The Pocket Protectors and other citizens then spoke. Allan Lind testified that, in the group’s opinion, the project, “particularly the traffic circulation assumptions, the allowances for parking, the building set backs [sic], the open space provisions,” is “stunningly ill conceived.” The “bizarre mile-long alley way with homes on either side of it” is “an assault on the neighborhood. It’s squeezing too many units into too small ... a space.” Even if there were no second-story windows, there would be too many blank walls and rooftops five feet away from neighboring homes. Alan Hockitson, who claimed experience with CEQA going back to the 1970’s as a state employee working on power plants and transmission lines, criticized the MND’s identified mitigation measures as “generic.” He also spoke to the removal of trees and to the impact of the new project on the local schools’ enrollment. Christopher Gáneles testified that the project must be rezoned R-l because it proposed single-family homes by any definition. Roger McCardle, a licensed architect with 25 years’ experience in planning and building designs for institutions and in private consulting, testified that he had reviewed plans and drawings for the project and found it deeply flawed. City staff’s April 2001 suggestions for improving the project had not been implemented; the developer had not significantly reduced the number of homes, reconfigured the project into a single row of houses on deeper lots, or integrated common open space areas into the design. It was also questionable whether a fire truck or moving van could get in and out of the development. Gary Hartwick testified that the development agreement for the prior (unbuilt) project noted townhouses were the only building type called for in *917 the PUD that had not yet been built, and the developer’s plan to build them was consistent with the community plan and the general plan. Although that project had a higher density than the current project, it preserved far more open space (35-foot rear setbacks and 30 feet between units) and did not intrude into the rear yards of the property Une, as half the units in the present project would do. “[T]o allow development of two rows of houses basically that looks like a monopoly board is not prudent and it’s certainly not creative or sound land use planning.” Chris Briggs testified about the potential problems with parking, especially with a sidewalk on the private street. Briggs also testified that it would be impossible to comply with the City’s condition that at least two shade trees be planted in each front yard if there was only a four-foot setback in front and guest parking spots in back; only ornamental trees could fit the available space. The current proposal crammed too many houses into too little space. Bill McElroy, The Pocket Protectors’ chairman, testified that the project violated sound land use principles and the City’s original vision for the development of the Pocket area. The Pocket Protectors supported the Development Agreement for the prior unbuilt project because it fit as “the final piece of this jigsaw puzzle.” The current project, on the other hand, was simply a “cram as much as you can development proposal” which had been “unworkable and flawed since it’s [sic] inception with its disrespect for compatibility with the existing homes in our community.” Finally, the commissioners commented. All but one opposed the project. Commissioner Kennedy stated that the proposed project was very nice in many ways, but “it doesn’t work here”—a five-foot setback in an existing neighborhood was unacceptable. Commissioner Valencia stated that it appeared the development agreement required a rezoning for the project, but staff and counsel should have more time to assess this issue. Commissioner Jones, an engineer, stated that after inspecting the site she was concerned about transportation problems in relation to the private street; she was also concerned about other issues on which an adequate paper trail under CEQA had not yet been laid and staff had not yet provided all the documentation the commission needed. 9 Commissioner Taylor-Carroll expressed opposition based on “the configuration” and “the environmental aspects that haven’t been addressed.” Commissioner McKeany expressed opposition based on “the concerns of the community.” Finally, commission chairman Waste opposed the project for several reasons: (1) “[I]t’s a bad land use where it’s proposed. . . . this is a project *918 that would be phenomenal in so many places across this town that are legitimate in-fill candidates and ... I would welcome it in any one of probably 35 places that I could think of around town. ... I would not add this location as a 36.” (2) A mitigated negative declaration was “just not an appropriate call” due to issues concerning traffic, population, and other environmental impacts that needed CEQA study. (3) The project did not work as infill because it was not connected to public transit. The commission thereupon voted six to one to deny the staff recommendation for approval of the project and the MND and to deny Regis’s application. The planning commission’s findings of fact On August 23, 2002, the planning commission adopted the following written findings of fact: “The Special Permit to develop detached single family dwellings within the LPPT PUD is denied based on the following Findings of Fact: “1. The project is not based upon sound principles of land use in that: “a. the shallow depth of the existing parcels does not afford sufficient area to develop the proposed lotting plan with adequate setbacks from adjacent properties; “b. the massing of the houses creates crowded conditions along the narrow interior private drive; “c. adequate play yards for children have not been provided; “d. small front yards prevent the planting of large shade trees; “e. the ability to provide guest parking adjacent to each dwelling is impeded by the narrow street which does not afford on-street parking and by the shallow front setbacks, which do not allow for parking in the driveways of many lots. “2. The proposed use will adversely affect the general welfare of the surrounding residential neighborhood in that the height and bulk of the proposed dwellings are not sufficiently mitigated by the proposed five foot rear yard setbacks, impacting negatively on the privacy of the neighboring property owners. *919 “3. The design of the proposed subdivision is not consistent with the Sacramento General Plan Update, Pocket Community Plan and LPPT [PUD] in that the proposed single-family detached units are not consistent with the attached townhouse-style housing previously anticipated for the site. “. . . The Tentative Map subdividing 20.6 vacant acres into 139 single family lots in the R-1A PUD zone is denied based on the following Findings of Fact: “1. [Repeats finding 3 above.] “2. [Repeats finding l.a. above.] “. . . The Subdivision Modification to reduce the standard 41’ right-of-way width for a private street is denied based on the following Findings of Fact: “1. The modification would be detrimental to the public health, safety or welfare or be injurious to other properties in the vicinity, in that it would facilitate a substandard lot configuration; and “2. Granting the modification is not in accord with the intent and the purpose of the Subdivision regulations and is not consistent with the General Plan and with all other applicable specific plans of the City in that: “a. Reduced street widths previously were approved for attached housing, not detached single family residences; and “b. The proposed street design does not include sidewalks, requiring walkways to be constructed in the front yard setbacks of certain lots which reduces the useable [jzc] landscaping area to a less than acceptable width.” THE CITY COUNCIL Before the first hearing on Regis’s appeal, there were four community meetings to discuss the project further. After these, Regis requested the following changes to its application and tentative map: 1) Increase the minimum rear setback to 10 feet. 2) Reduce points of vehicular access to the site (to lessen tree removal and traffic problems). 3) Provide an additional single-story plan and eliminate the existing “quasi-two-story” plan (to lessen adjacent neighbors’ privacy concerns). 4) Provide multiple “mini-parks” (to increase play areas). 5) Remove the sidewalk along the private street (to increase guest parking). The Pocket Protectors continued to oppose the project. *920 The first City Council hearing On March 11, 2003, the City Council held its first hearing on the project. 10 City planner Pace testified that staff supported the planning commission’s position. However, staff also believed the MND was correctly prepared, and some points raised by objectors, such as the project’s impact on schools, were legally beyond the MND’s purview. Staff still recommended a sidewalk for the private street, although they did not object to Regis’s revisions (which omitted the sidewalk). 11 However, there would inevitably be a tradeoff as to three issues: installing a sidewalk, providing full-length driveways, and maximizing rear yard setbacks. The City Fire Department had approved the width of the private street, which Pace thought was the narrowest with which they would be comfortable. Pace opined that the project was not an infill project because it was not in an area targeted by City policy for infill-type incentives. Gregory Thatch, attorney for Regis, opined that the project was an infill project. He also stated: “Out of all of the products that Regis Homes could put on this property, this ... is the one that will have the highest selling point and the one that will achieve we think the highest property values.” Townhouses, on the other hand, “have less market acceptance.” Moreover, the Riverlake Community Association would not accept a townhouse project now. On behalf of The Pocket Protectors, Christopher Gáneles commented that his group supported the development of this property. However, Regis’s changes to the project had not cured its fundamental flaws. Gary Hartwick commented that the project would have a detrimental impact on 50 percent of the adjoining residences, whose owners had been told an entirely different kind of development would occur on this site. Roger McCardle commented that the April 2001 recommendations of the City’s long-range planner to mitigate the “tunnel or canyoning effect of wide houses on small lots” and to “break up . . . the visual monotony” of long rows of driveways had not been heeded. Furthermore, it would be difficult to get adequate landscaping into the project, especially with a sidewalk; the proposed houses had a footprint 40 percent larger than the buildings previously planned for the site. Half the houses in the project would open their *921 front doors onto the greenbelt, a publicly dedicated space on which previous plans for the site had not encroached. As designed, the project required concrete patios and sidewalks in that area. 12 Claudia Bonsenior commented that under the latest proposal Regis was no longer required to plant two or more shade trees per unit, but only to plant “trees.” If shade trees were not planted, many of the units would get very hot. This was bad planning in terms of energy and livability. Alan Hockitson cited a planning commissioner’s comment that “schools, public services, transportation, shade impacts” needed to be addressed under CEQA. Hockitson said the MND relied on outdated data and did not identify or mitigate cumulative impacts to which the project would contribute, such as those on school populations. Furthermore, Regis had never considered alternatives to its “double-stacked concept.” Cassandra Hockitson commented that the Riverlake Community Association did not speak for the community: the board had voted to approve the project at a meeting after almost 200 people in attendance had signaled their opposition. Moreover, the board had conditioned its approval on the absence of sidewalks; thus, including a sidewalk negated the board’s approval. Allan Lind stressed the importance of sticking to the PUD plan and the housing diversity it had fostered; it should not be abandoned just because Regis thought attached housing might be harder to sell. Lind objected to the “mile-long alleyway . . . with two rows of houses on it” which Regis had proposed, and called it “a meaningless exercise” to increase the rear setback to 10 feet. Cluster development would preserve open space and avoid “this shoebox effect mnning back-and-forth down the street,” while removing the weeds that now colonized the undeveloped site and ensuring that it was properly cared for. City Councilmember Cohn indicated that he saw the project as infill development of the kind the City should encourage. Councilmember Tretheway, though also supportive, suggested increasing the rear setbacks further and incorporating smaller homes to open up more room for trees. Councilmember Sheedy saw the project as infill and “a quality project [that] promotes the City’s anti-sprawl objectives.” Councilmember Pannell called it “a great project.” Councilmember Jones indicated that it would be desirable to put in sidewalks, smaller homes, and more shade trees, and to increase the setbacks. Mayor Fargo spoke on similar lines. None of the council members specifically addressed CEQA or the MND. *922 The Council voted to continue the matter until May 27, 2003, directing Planning Director Gary Stonehouse to meet with representatives of the parties and report back to the City Council in the interim. The second City Council hearing Before the City Council heard the matter again, Stonehouse filed a report recommending approval of the project and the MND. The report stated that Regis had revised the project further as follows: To reduce “hardscape,” the private street had been narrowed from 25 to 22 feet, with the concurrence of public works staff. The three feet gained had been added to the front and rear yards. Regis had reduced the size of two house models from 1,650 to 1,580 square feet and from 1,950 to 1,750 square feet; it had also added a new plan for a 1,440-square-foot model. All of these changes together had permitted the addition of a four-foot-wide sidewalk along the interior lots, while maintaining driveways long enough to accommodate off-street parking. Regis also proposed to plant larger street trees in the project’s seven mini-parks. (However, the front yards would still not support larger shade trees.) The City Council heard the matter again on May 27, 2003. In addition to Stonehouse’s report, the City Council had before it a letter from counsel for The Pocket Protectors, asserting that under CEQA’s “fair argument” standard an EIR needed to be prepared for the project. The City Council had also received an alternative project proposal from The Pocket Protectors. 13 Regis’s attorney admitted that the Riverlake Community Association had withdrawn its support. Association board members testified that the latest changes had reduced the private street width, front yard sizes, and house sizes unacceptably, while still not permitting shade trees to be planted. The community association president commented that with the extremely narrow private street running one-third mile, the close-together houses, and the small yards, “architectural canyoning” had become a problem in the current design; furthermore, the reduced house sizes created a possible architectural incompatibility with the surrounding neighborhood. The Pocket Protectors’ spokespersons testified, reiterating their previous positions. *923 Regis’s attorney, Gregory Thatch, responded that neither the former development agreement, which had now expired, nor the R-1A zoning precluded the project. He asserted that the previous proposals had not been built because they would not have made money, and no one could make money by building attached housing on this site. He denied that the project would encroach onto the greenbelt: front porches would come up to the line, but doors would open onto the porches, not out into the easement. Councilmember Jones asked Thatch if Regis could build and market The Pocket Protectors’ proposal. Thatch replied that this was “an interesting question” and he could not “say unequivocally no,” nor could he say that it would be impossible to build or sell. It simply was not the project Regis wanted to build. Councilmembers Cohn, Tretheway, Pannell, and Sheedy spoke in support of the project, as did Mayor Fargo; Councilmembers Yee and Hammond spoke against it. 14 Councilmember Yee opined that the setbacks were still too small on all sides and that the lots were too small for the houses proposed to be built on them; he also questioned putting in a sidewalk only four or five feet from houses. Councilmember Hammond felt the setbacks were insufficient in relation to the greenbelt. Mayor Fargo asked City Attorney Jackson whether staff agreed that no further CEQA study needed to be done. Jackson replied affirmatively. The City Council then voted six to two to approve the project. The City Council’s adoption of the project On June 17, 2003, the Council adopted a resolution approving the project and the MND. The resolution made the following findings of fact, among others: “The City Council approves the [MND] based upon the following findings: “1. The [MND] was prepared and circulated for the above-identified project pursuant to the requirements of CEQA; “2. The proposed [MND] and comments received during the public review process were considered prior to action being taken on the project; *924 “3. Based upon the record as a whole, including the Initial Study and the comments received during the public review process, and in the City Council’s exercise of its independent judgment, there is no substantial evidence, either individually limited or cumulatively considerable, that the project will have a significant effect on the environment, [f] . . . [][] “. . . The Special Permit to develop detached single family dwellings within the LPPT PUD is approved based on the following Findings of Fact and subject to the Conditions of Approval: “1. The project is based upon sound principles of land use in that the proposed project is consistent with the General Plan, the Pocket Community Plan, the Single Family Residential Design Principles, and the underlying Single-family Alternative (R-1A) zoning; “2. The proposed use will not adversely affect the public health, safety or general welfare of the surrounding residential neighborhood in that: “a. The project is designated for single-family alternative residential development; “b. The project proposes to apply design compatible with the surrounding area, will use quality construction materials, and has implemented numerous design components consistent with the Single Family Residential Design Principles, and; “c. The proposed project is within the Zoning Ordinance’s requirements for allowable setbacks within the Single-family Alternative (R-1A) zone. “3. The proposed project is consistent with the General Plan and the Pocket Community Plan land use designations and density requirements of Low Density Residential (4-15 du/na) and Residential (7-15 du/na), respectively; “4. The proposed project is within the LPPT PUD Development Guidelines maximum density allowed for the parcels . . ., and; “5. The project complies with the Sacramento] G[eneral] P[lan] U[pdate] Housing Element that encourages the promotion of a variety of housing types within neighborhoods to encourage economic diversity and housing choice ....[f]... [j[]. *925 “. . . The Subdivision Modification to reduce the standard 4V right-of-way width for a private street is approved based on the following Findings of Fact and subject to the Conditions of Approval: “1. The property to be divided is of such size or shape, or is affected by such topographic conditions, or there are such special circumstances or conditions affecting the property that it is impossible, impractical, or undesirable in this particular case to conform to the strict application of these regulations. The narrow width of the project site requires that in order to develop the site as proposed the lot widths be shallower than the City standard and that the roadway be a narrower width than standard in order to provide the project’s two tiers of shallow lots access; “2. The cost to the subdivider, of strict or literal compliance with the regulation, is not the sole reason for granting the modification; “3. The modification will not be detrimental to the public health, safety or welfare or be injurious to other properties in the vicinity in that the Department of Public Works, Department of Utilities, and the Fire Department reviewed the project and the private drive for safe ingress and egress, adequate room for utility placement, and sufficient room for emergency vehicle access. Each department has approved the project as proposed, subject to the attached conditions; “4. Granting the modification is in accord with the intent and the purposes of these regulations and is consistent with the General Plan and with all other applicable specific plans of the City. The Single-family Alternative (R-1A) zoning is consistent with the General Plan and Community Plan land use designations of Low Density Residential (4-15 du/na) and Residential 7-15, respectively.” The writ petition The Pocket Protectors filed a petition for writ of mandamus in Sacramento County Superior Court, naming the City and the City Council as respondents. They asserted substantial evidence existed in the administrative record to support a “fair argument” that an EIR should be performed. (Cf. Cal. Code Regs., tit. 14, § 15064, subd. (f).) The Pocket Protectors’ briefing cited four areas of potential significant environmental impact: (1) inconsistencies with city land use polices and *926 regulations, including the LPPT PUD, the development agreement, the City’s development standards, and “impacts relating to land use”; (2) aesthetic impacts; (3) traffic and parking; (4) biological impacts. 15 After oral argument, the trial court denied the writ petition. The court found specifically: The site had been subjected to prior planning and zoning, which included environmental review, and the proposed project was consistent with the intended use of the site. The aesthetic issues raised by the project opponents did not rise to a level of substantial evidence constituting a fair argument for an EIR: “Indeed it may well be that inherent in the kind of issue here is not appropriate for an environmental review on the kind of configuration issues which have been joined.” Although some other project might have been preferable, the City Council had acted within its discretion to approve this one, given the “rule of reasonableness” that applied to all environmental law. 16 This appeal followed. DISCUSSION PRINCIPLES OF CEQA LAW APPLICABLE TO THIS CASE “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.’ (Friends of Mammoth v. Board of Supervisors (1972) 8 Cal.3d 247, 259 [104 Cal.Rptr. 761, 502 P.2d 1049].) More than a decade ago, we observed that, ‘It is, of course, too late to argue for a grudging, miserly reading of CEQA.’ (Bozung v. Local Agency Formation Com. (1975) 13 Cal.3d 263, 274 [118 Cal.Rptr. 249, 529 P.2d 1017] [hereafter Bozung].)” (Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376, 390 [253 Cal.Rptr. 426] (Laurel Heights I).) “We have repeatedly recognized that the EIR is the ‘heart of CEQA.’ (Citizens of Goleta Valley v. Board of Supervisors (1990) 52 Cal.3d 553, 564 [276 Cal.Rptr. 410, 801 P.2d 1161], [hereafter Goleta Valley II]; Laurel Heights I, supra, 47 Cal.3d at p. 392; see also Guidelines, § 15003, subd. (a).) ‘Its purpose is to inform the public and its responsible officials of the environmental consequences of their decisions before they are made. Thus, *927 the EIR “protects not only the environment but also informed self-government.” (Lau rel Heights [I], supra, 47 Cal.3d at p. 392.)’ (Goleta Valley II, supra, 52 Cal.3d at p. 564.) To this end, public participation is an ‘essential part of the CEQA process.’ (Guidelines, § 15201; see also Concerned Citizens of Costa Mesa, Inc. v. 32nd Dist. Agricultural Assn. (1986) 42 Cal.3d 929, 936 [231 Cal.Rptr. 748, 727 P.2d 1029].) “With certain limited exceptions, a public agency must prepare an EIR whenever substantial evidence supports a fair argument that a proposed project ‘may have a significant effect on the environment.’ (§§ 21100, 21151, 21080, 21082.2 [fair argument standard]; Guidelines, §§ 15002, subd. (f)(1), (2), 15063; No Oil, Inc. v. City of Los Angeles (1974) 13 Cal.3d 68, 75 [118 Cal.Rptr. 34, 529 P.2d 66] [fair argument standard of review] [(No Oil)].) ‘ “Significant effect on the environment” means a substantial, or potentially substantial, adverse change in the environment.’ (§ 21068; see also Guidelines, § 15382.)” (Laurel Heights Improvement Assn. v. Regents of University of California (1993) 6 Cal.4th 1112, 1123 [26 Cal.Rptr.2d 231, 864 P.2d 502], fn. omitted.) If there is substantial evidence in the whole record supporting a fair argument that a project may have a significant nonmitigable effect on the environment, the lead agency shall prepare an EIR, even though it may also be presented with other substantial evidence that the project will not have a significant effect. (§21151, subd. (a); Cal. Code Regs., tit. 14, § 15064, subd. (f)(1), (2); 17 No Oil, supra, 13 Cal.3d 68, 75; Architectural Heritage Assn. v. County of Monterey (2004) 122 Cal.App.4th 1095, 1109 [19 Cal.Rptr.3d 469] (Architectural Heritage Assn.); Communities for a Better Environment v. California Resources Agency (2002) 103 Cal.App.4th 98, 111-112 [126 Cal.Rptr.2d 441].) “May” means a reasonable possibility. (§§ 21082.2, subd. (a), 21100, 21151, subd. (a); League for Protection of Oakland’s etc. Historic Resources v. City of Oakland (1997) 52 Cal.App.4th 896, 904-905 [60 Cal.Rptr.2d 821] (League for Protection).) “Substantial evidence” means “enough relevant information and reasonable inferences from this information that a fair argument can be made to support a conclusion, even though other conclusions might also be reached.” (Guidelines, § 15384, subd. (a).) Substantial evidence “shall include facts, reasonable assumptions predicated upon facts, and expert opinion supported by facts.” (Guidelines, § 15384, subd. (b).) “Argument, speculation, unsubstantiated opinion or narrative, evidence which is clearly erroneous or inaccurate, or evidence of social or economic impacts which do not contribute to or are *928 not caused by physical impacts on the environment does not constitute substantial evidence.” (Guidelines, § 15384, subd. (a).) The fair argument standard is a “low threshold” test for requiring the preparation of an EIR. (No Oil, supra, 13 Cal.3d 68, 84; League for Protection, supra, 12 Cal.App.4th at p. 905; Sierra Club v. County of Sonoma (1992) 6 Cal.App.4th 1307, 1316-1317 [8 Cal.Rptr.2d 473]; Oro Fino Gold Mining Corp. v. County of El Dorado (1990) 225 Cal.App.3d 872, 881 [274 Cal.Rptr. 720] (Oro Fino).) It is a question of law, not fact, whether a fair argument exists, and the courts owe no deference to the lead agency’s determination. Review is de novo, with a preference for resolving doubts in favor of environmental review. (Architectural Heritage Assn., supra, 122 Cal.App.4th 1095, 1110; San Joaquin Raptor/Wildlife Rescue Center v. County of Stanislaus (1996) 42 Cal.App.4th 608, 617-618 [49 Cal.Rptr.2d 494] (San Joaquin Raptor); Stanislaus Audubon Society, Inc. v. County of Stanislaus (1995) 33 Cal.App.4th 144, 151 [39 Cal.Rptr.2d 54] (Stanislaus Audubon Society); Quail Botanical Gardens Foundation, Inc. v. City of Encinitas (1994) 29 Cal.App.4th 1597, 1602-1603 [35 Cal.Rptr.2d 470] (Quail).) Although our review is de novo and nondeferential, however, we must “ ‘giv[e] [the lead agency] the benefit of [the] doubt on any legitimate, disputed issues of credibility.’ ” (San Joaquin Raptor, supra, 42 Cal.App.4th 608, 617, quoting Quail, supra, 29 Cal.App.4th 1597, 1603, first and third brackets added.) The lead agency has discretion to determine whether evidence offered by the citizens claiming a fair argument exists meets CEQA’s definition of “substantial evidence.” (Citizens for Responsible Development v. City of West Hollywood (1995) 39 Cal.App.4th 490, 499, fn. 2 [45 Cal.Rptr.2d 917]; Citizens’ Com. to Save Our Village v. City of Claremont (1995) 37 Cal.App.4th 1157, 1170-1171 [44 Cal.Rptr.2d 288] (Claremont).) Relevant personal observations of area residents on nontechnical subjects may qualify as substantial evidence for a fair argument. (Ocean View Estates Homeowners Assn, Inc. v. Montecito Water Dist. (2004) 116 Cal.App.4th 396, 402 [10 Cal.Rptr.3d 451] (Ocean View Estates); Arviv Enterprises, Inc. v. South Valley Area Planning Com. (2002) 101 Cal.App.4th 1333, 1347 [125 Cal.Rptr.2d 140] (Arviv).) So may expert opinion if supported by facts, even if not based on specific observations as to the site under review. (Friends of the Old Trees v. Department of Forestry & Fire Protection (1997) 52 Cal.App.4th 1383, 1398-1399 & fn. 10 [61 Cal.Rptr.2d 297] (Friends) [expert testimony for fair argument purposes need not meet standard required of such testimony at trial].) Where such expert opinions clash, an EIR should be done. (Guidelines, § 15064, subd. (g).) On the other hand, mere argument, speculation, and unsubstantiated opinion, even expert opinion, is not substantial evidence for a fair argument. *929 (§ 21082.2, subd. (c); Guidelines, § 15384, subd. (a); Santa Monica Chamber of Commerce v. City of Santa Monica (2002) 101 Cal.App.4th 786, 798 [124 Cal.Rptr.2d 731] (Santa Monica Chamber of Commerce); Apartment Assn. of Greater Los Angeles v. City of Los Angeles (2001) 90 Cal.App.4th 1162, 1176 [109 Cal.Rptr.2d 504] (Apartment Assn.); Pala Band of Mission Indians v. County of San Diego (1998) 68 Cal.App.4th 556, 580 [80 Cal.Rptr.2d 294] (Pala Band).) “The existence of public controversy over the environmental effects of a project shall not require preparation of an environmental impact report if there is no substantial evidence in light of the whole record before the lead agency that the project may have a significant effect on the environment.” (§ 21082.2, subd. (b); see San Joaquin Raptor, supra, 42 Cal.App.4th 608, 622.) 18 Neither is the mere possibility of adverse impact on a few people, as opposed to the environment in general. (Association for Protection etc. Values v. City of Ukiah (1991) 2 Cal.App.4th 720, 734 [3 Cal.Rptr.2d 488] (Ukiah)) Analysis As we have recounted, the trial court opined that the issues tendered by the Pocket Protectors were “not appropriate for an environmental review on the kind of configuration issues which have been joined.” The Pocket Protectors contend that substantial evidence exists to support a fair argument for potential significant effects on the environment as to City land use policies and regulations (including City development standards) and aesthetic impacts. We agree with The Pocket Protectors. For reasons that follow, we conclude the trial court erred in ruling that the issues tendered by The Pocket Protectors were immune from environmental review in an EIR. CITY LAND USE POLICIES AND REGULATIONS The CEQA Initial Study Checklist, used to determine whether a project may have significant environmental impacts, includes the question whether a project may “[c]onflict with any applicable land use plan, policy, or regulation . . . adopted for the purpose of avoiding or mitigating an environmental effect.” (Guidelines, appen. G, § IX, subd. (b).) The LPPT PUD governs the development of the project site. This is why Regis had to apply for and obtain a special permit to develop detached *930 single-family dwellings on the site: as we have shown, the PUD’s drafters intended the site for a different type of housing (“townhouse and similar development”). The PUD’s development guidelines define the PUD’s objectives as follows: “1. To provide adequate natural light, pure air and safety from fire and other dangers, [f] 2. To enhance the value of land and structures within and adjacent to the project, [f] 3. To minimize congestion due to vehicular and pedestrian circulation within the project area. [j[] 4. To preserve and enhance the aesthetic values throughout the project, [f] 5. To promote public health, safety, comfort, convenience and general welfare.” It is clear in light of these objectives (especially numbers 1 and 3) that the City adopted the PUD in part “for the purpose of avoiding or mitigating an environmental effect.” (Guidelines, appen. G, § IX, subd. (b).) Furthermore, in adopting the PUD the City Council found that it “meets the purposes and criteria stated in City Zoning Ordinance Sections 8A and 8B in that the PUD facilitates a variety of housing types and site plans, accessible open ‘green spaces,’ recreation areas and other features of substantial benefit to a viable and balanced community. [rj[] . . . [f][] . . . [T]he PUD [e]nsures that development will be well-designed, and that non-residential uses will be adequately buffered from residential uses by landscaping and setbacks.” This statement confirms that in adopting the PUD the City Council sought to avoid or mitigate the environmental effects that might arise from unplanned development. In addition to their general objectives, the PUD’s development guidelines specifically stress the importance of landscaping. In a section headed, “'Landscape Requirements (Excluding Single Family Residential),” the guidelines state: “The role of landscaping as a common element to unify the overall PUD cannot be overstated.” They go on to prescribe specific rules, including 25 percent landscape coverage for any project within the PUD and a minimum 25-foot landscaped setback for all public road frontages. These conditions apply to the project site: the PUD designates only sites zoned R-l as “single family” residential, not sites zoned R-l A and reserved for townhouses, such as the project site. Given all of the above, if substantial evidence supports a fair argument that the proposed project conflicts with the policies of the PUD, this constitutes grounds for requiring an EIR. Whether a fair argument can be made on this point is a legal question on which we do not defer to the City Council’s determination. (San Joaquin Raptor, supra, 42 Cal.App.4th at pp. 617-618; Stanislaus Audubon Society, supra, 33 Cal.App.4th at p. 151; Quail, supra, 29 Cal.App.4th at pp. 1602-1603.) Furthermore, if substantial *931 evidence supports the existence of a fair argument, the existence of contrary evidence does not excuse a lead agency from its duty to prepare an EIR. (Guidelines, § 15064, subd. (f); League for Protection, supra, 52 Cal.App.4th at pp. 904-905.) The Pocket Protectors have adduced substantial evidence that the project conflicts with the objectives of the PUD. Not only did the PUD require “townhouses and similar development” for the site, but the site’s unusually narrow shape dictated that only such housing could be built at the desired density without violating the PUD’s objectives. Even the City planning staff admitted this fact. In April 2001 a staffer informed Regis that the project “does not fulfill the intent of the LPPT PUD Townhouse land-use designation insofar as it does not incorporate the landscaping and open space concepts embraced by the remainder of the LPPT PUD.” (In other words, Regis’s plan to construct as many large detached houses as possible side by side on minimal lots violated the PUD’s intent to preserve greenery and open space while building out the site.) Staff also pointed out the “canyon” effect of putting so many houses of similar scale so close together along the whole length of the site. 19 Staff recommended mitigating this effect by, among other things, planting one shade tree per 30 lineal feet of street frontage, but the approved project did not include this mitigating measure. Nor could it have done so: after every possible adjustment had been made to increase the setbacks, they remained too small to permit large shade trees. Similarly, City planner Pace told the planning commission that the project’s setback problem could not be solved perfectly because the only way to develop the “very narrow strip” available with single-family housing at “something close to the approved density” was to double-tier rows of houses along a narrow private street. (Bill Heartman, speaking for Regis, testified to the same effect.) Pace also told the City Council of an inevitable “trade off” between providing a sidewalk, providing standard-length driveways, and providing generous rear yard setbacks. The site’s physical properties did not cause these problems. What caused them was Regis’s plan to build a type of housing that the site could not easily accommodate at the proposed density. Furthermore, the planning commission expressly found that the first version of the project did not comply with the policies and objectives of the *932 PUD, which had anticipated “attached townhouse-style housing” for the site. (See Stanislaus Audubon Society, supra, 33 Cal.App.4th at p. 155 [planning commissioner’s fact-based opinions, stemming from commission’s experience in planning and development, are substantial evidence for a fair argument].) Based on the site’s configuration as determined by the PUD, the planning commission also found more broadly that the project failed to comply with “sound principles of land use” (impliedly including the objectives of the PUD) due to inadequate setbacks and front yards, insufficient possibilities for landscaping as a result, excessive massing of houses along the interior drive, and encroachment on neighboring owners’ privacy, inter alia. (Even after the project was modified to increase setbacks, two City Council members still found them inadequate for essentially the same reasons.) Finally, the abundant testimony by Pocket Protector members (and disenchanted Riverlake Community Association members) on this issue— neighbors familiar with the site and with the PUD, some of whom had moved to the neighborhood in reliance on the promise that the PUD would control its development—also counts as substantial evidence for purposes of the fair argument test. Relevant personal observations by area residents are properly considered for this purpose. (Ocean View Estates, supra, 116 Cal.App.4th at p. 402; Arviv, supra, 101 Cal.App.4th at p. 1347.) While some planning issues are inherently technical, the potential adverse environmental effects of minimizing the open space and landscaping required by the PUD are not. Many of the objectors, both Pocket Protectors members and others, offered detailed factual observations on these points, not mere general opinions. Moreover, some of The Pocket Protectors’ spokespersons, such as architect Roger McCardle, have specific expertise on issues going to planning and design. 20 It is true that the MND found the project consistent with the PUD. However, its findings are devoid of reasoning and evidence. After observing that the PUD identified the project site for townhouse development, the MND merely states: “The approvals for the 1993 townhouse project expired and the City is considering the development of single-family homes in place of the townhouse concept.” It asserts measures will be taken to ensure that the project conforms to “the minimum design standards set in the LPPT PUD Development Guidelines,” but does not explain how this can be done without building the type of housing the guidelines mandate for the site. In short, the MND does not support its finding of consistency with the PUD, but simply accepts the City’s decision to disregard the PUD as a fait accompli. *933 The City Council’s findings of fact on this point are equally open to dispute. The City Council approved detached single-family housing on the project site partly because R-lA zoning generally permits detached housing as a “Single-family Alternative” housing type. But as the planning commission and City staff pointed out, the PUD specifically designates sites zoned R-l A within the PUD for townhouse or other clustered housing development. Furthermore, the development agreement for the prior unbuilt project, which the City Council presumably executed with the PUD’s objectives in mind, stated that a rezoning to R-l would be required to build “single family residential” housing on the site. 21 Regis asserts its project fits within the PUD’s townhouse designation because a dictionary gives an alternate definition of “townhouse” as “a house in a compact planned group in a town.” But even if the PUD had incorporated this definition (which Regis does not show), it would not mean that an entire development of detached houses is a townhouse development. It is clear that this is not what the drafters of the PUD, or the City Council in approving the PUD, had in mind. Regis asserts its project is within the PUD’s approved density for the site. This fact does not advance Regis’s argument. Maximum density is only one of the PUD’s conditions for development of R-l A sites. Relying on Sequoyah Hills Homeowners Assn. v. City of Oakland (1993) 23 Cal.App.4th 704, 717 [29 Cal.Rptr.2d 182] (Sequoyah Hills), Regis asserts: “Consistency with any local agency’s land use plan or policies ... is evaluated under the ‘substantial evidence’ standard, not the ‘fair argument’ test.” But Sequoyah Hills is not a “fair argument” case: the appellant challenged the certification of an EIR, which a reviewing court will overturn only for abuse of discretion. (Id. at pp. 709, 712.) 22 Under the fair argument test, the appellant has a much lower threshold to meet and we do not defer to the lead agency’s exercise of discretion. (No Oil, supra, 13 Cal.3d at p. 84; San Joaquin Raptor, supra, 42 Cal.App.4th at pp. 617-618; League for Protection, supra, 12 Cal.App.4th at p. 905; Sierra Club v. County of *934 Sonoma, supra, 6 Cal.App.4th at pp. 1316-1317; Oro Fino, supra, 225 Cal.App.3d at p. 881.) Because the land use policies at issue were adopted at least in part to avoid or mitigate environmental effects, we consider their applicability under the fair argument test with no presumption in favor of the City. Regis accuses The Pocket Protectors of turning ordinary planning and zoning issues into CEQA issues to avoid the substantial evidence test. This argument fails for the reasons already given. Because the issues raised are genuine CEQA issues to which the fair argument test applies, Regis’s cited authorities, which hold that the test is merely whether substantial evidence supports the agency’s determination, are inapposite. Regis asserts: “Unsupported opinions, even those of Planning Commissioners, do not rise to the level of ‘substantial evidence,’ and do not create a de facto presumption of inconsistency with existing land use plans.” This proposition is correct but inapposite. Regis relies heavily on a decision of this court which stated: “The commission’s conclusions from the evidence presented to it do not themselves constitute evidence of such effects.” (Perley v. Board of Supervisors (1982) 137 Cal.App.3d 424, 435 [187 Cal.Rptr. 53] (Perley).) But in Perley, neither the planning commission nor the plaintiff cited facts to support the commission’s conclusions. (Id. at pp. 429, 434-435.) Here, the planning commission made findings of fact, specifying the elements of the proposed project which clashed with the policies of the PUD and other land use principles. 23 (See Architectural Heritage Assn., supra, 122 Cal.App.4th at p. 1115 [similarly distinguishing Perley].) Relying on Claremont, supra, 37 Cal.App.4th 1157, Regis asserts that the City Council properly weighed the credibility of the opponents’ evidence and found it wanting. The lead agency’s weighing of legitimate, disputed credibility questions is indeed entitled to deference even under the fair argument test. (San Joaquin Raptor, supra, 42 Cal.App.4th at p. 617.) However, since fair argument review is generally nondeferential and prefers resolving doubts in favor of maximizing environmental review (id. at pp. 617-618), before accepting Regis’s argument we would have to find that the City Council actually resolved disputed factual questions going to credibility. But the City Council’s findings of fact do not discuss any opposing evidence: they merely recite generally that substantial evidence of a significant effect on the *935 environment does not exist. Thus, we see no specific credibility call by the City Council which requires deference. We also note that Claremont, supra, 37 Cal.App.4th 1157, must be read with great care and caution. Citing no authority, Claremont said: “The determination of whether or not evidence is ‘substantial’ is in itself a weighing process. The court does not look only to the evidence relied upon by appellants to the exclusion of all contrary evidence. Evidence that rebuts, contradicts or diminishes the reliability or credibility of appellants’ evidence is properly considered. The absence of supporting evidence is properly considered.” (Id. at pp. 1168-1169; italics added.) This passage is as slippery as a ball bearing sprayed with WD-40. We agree with Claremont, supra, 37 Cal.App.4th 1157, that a lead agency or a court may weigh evidence on the whole record in determining the preliminary issue of whether evidence is “substantial” and thus deserving of consideration. Thus, for example, if an expert purporting to hold a Ph.D. testifies as to the environmental effect of a project, a lead agency or a court may properly consider and “weigh” evidence in the record showing the expert never attended college and his Ph.D. is phony. But this limited weighing of evidence to determine admissibility in an environmental debate must not be confused with a weighing of some substantial evidence against other substantial evidence. Unlike the situation where an EIR has been prepared, neither the lead agency nor a court may “weigh” conflicting substantial evidence to determine whether an EIR must be prepared in the first instance. Guidelines section 15064, subdivision (f)(1) provides in pertinent part: “if a lead agency is presented with a fair argument that a project may have a significant effect on the environment, the lead agency shall prepare an EIR even though it may also be presented with other substantial evidence that the project will not have a significant effect. (No Oil[, supra,] 13 Cal.3d 68).” Thus, as Claremont itself recognized, “Consideration is not to be given contrary evidence supporting the preparation of a negative declaration. (City of Carmel-by-the Sea v. Board of Supervisors (1986) 183 Cal.App.3d 229, 244-245 [227 Cal.Rptr. 899]; Friends of “B” Street v. City of Hayward (1980) 106 Cal.App.3d 988 [165 Cal.Rptr. 514].” (Claremont, supra, 37 Cal.App.4th at p. 1168.) It is the function of an EIR, not a negative declaration, to resolve conflicting claims, based on substantial evidence, as to the environmental effects of a project. (See No Oil, supra, 13 Cal.3d at p. 85.) In the instant case, there is no basis in the record to find the testimony of the Pocket Protectors’ spokesmen, certainly including architect/planner Roger *936 McCardle, insubstantial. Their testimony constituted substantial evidence (that could not be “weighed” with contrary evidence) that mandated preparation of an EIR. To the extent the City Council may have found this evidence insubstantial, it abused its discretion. Regis asserts that the previously approved projects and the neighbors’ expectations carry no special weight in determining planning consistency. 24 While this might be true as a general proposition, here the prior projects and the expectations based on them derived squarely from the policies and requirements of the PUD. Finally, Regis cites authority holding that effects on particular persons are not significant environmental impacts under CEQA. (Santa Monica Chamber of Commerce, supra, 101 Cal.App.4th at p. 799; Friends of Davis v. City of Davis (2000) 83 Cal.App.4th 1004, 1019 [100 Cal.Rptr.2d 413] (Davis)] Ukiah, supra, 2 Cal.App.4th at p. 734; Topanga Beach Renters Assn. v. Department of General Services (1976) 58 Cal.App.3d 188, 195 [129 Cal.Rptr. 739] (Topanga).) This rule is inapposite, and the cases are factually distinguishable. In Ukiah, the “project” was a single house, whose impact (if any) could affect only a few immediately adjacent residents. (Ukiah, supra, 2 Cal.App.4th at pp. 724—725.) In Davis and Santa Monica Chamber of Commerce, the only alleged environmental impacts were adverse economic effects on a few persons or businesses, which are not cognizable harms under CEQA. (Santa Monica Chamber of Commerce, supra, 101 Cal.App.4th at p. 799; Davis, supra, 83 Cal.App.4th at p. 1019; Guidelines, § 15064, subd. (e).) In Topanga, the project entailed demolishing private structures on a public beach and evicting their dwellers to restore the land to its natural state, a clear environmental boon with no potential to cause an adverse environmental effect on persons in general. (Topanga, supra, 58 Cal.App.3d at p. 195.) Here, by contrast, the proposed mile-long project facially conflicts with a PUD established by the City to mitigate the possible environmental effects of uncontrolled development, and has the potential to cause an immediate adverse environmental impact to hundreds of nearby residents. It may be, as Regis told the City Council, that a developer could not now make money by building the kind of housing on this site which the PUD intended for it. However, this possibility does not justify skirting CEQA by ignoring the PUD’s intent or finding consistency with the PUD where there is none. AESTHETIC IMPACTS Under CEQA, it is the state’s policy inter aha to “[t]ake all action necessary to provide the people of this state with . . . enjoyment of aesthetic, *937 natural, scenic, and historic environmental qualities.” (§ 21001, subd. (b); italics added.) The CEQA initial study checklist asks four questions as to aesthetic impact, including whether a project will “[substantially degrade the existing visual character or quality of the site and its surroundings.” (Guidelines, appen. G, § I, subd. (c).) 25 Thus, courts have recognized that aesthetic issues “are properly studied in an EIR to assess the impacts of a project.” (Mira Mar Mobile Community v. City of Oceanside (2004) 119 Cal.App.4th 477, 492 [14 Cal.Rptr.3d 308]; see Ocean View Estates Homeowners Assn., Inc. v. Montecito Water Dist., supra, 116 Cal.App.4th 396, 401; National Parks & Conservation Assn. v. County of Riverside (1999) 71 Cal.App.4th 1341, 1360 [84 Cal.Rptr.2d 563].) As on other CEQA topics, the opinions of area residents, if based on direct observation, may be relevant as to aesthetic impact and may constitute substantial evidence in support of a fair argument; no special expertise is required on this topic. (Ocean View Estates, supra, 116 Cal.App.4th at p. 402.) We need not repeat here the extensive evidence offered by The Pocket Protectors and other area residents, including that of professional architect and planner Roger McCardle, based on their personal observations, as to the potential aesthetic impacts of the proposed project. We need only reiterate the specific concerns they expressed: the “tunneling” or “canyoning” effect of long double rows of houses flanking a narrow private street, the insufficient use of shade trees and other landscaping, the possibility of intrusions into the greenbelt along Pocket Road, and the overall degradation of the existing visual character of the site from the excessive massing of housing with insufficient front, rear, and side yard setbacks. These observations—which pertain even to the revised project approved by the City Council, not merely to its initial version as Regis suggests—suffice to raise the potential of a significant aesthetic impact from the proposed project. As Regis points out, City staff and City Council members have disagreed with the neighbors on these points. This disagreement does not reduce the neighbors’ evidence to insubstantiality, however. At most, the opposing views are substantial evidence going the other way, which is insufficient to refute the claim of a fair argument. (Guidelines, § 15064, subd. (f).) Regis cites Ukiah, supra, 2 Cal.App.4th 720 at page 734, for the proposition that “ ‘height, view and privacy objections’ are properly considered by an agency in the context of a site development permit approval, not under CEQA, because they do ‘not affect the environment of persons generally.’ ” Regis misreads the case. As mentioned above, the so-called project in Ukiah *938 was the construction of a single house. (Id. at p. 724.) The court found that this “project” was categorically exempt from CEQA and no unusual circumstances existed which could create an exception to that categorical exemption. (Id. at pp. 734-736.) The “height, view and privacy objections” at issue impacted only a few neighbors; therefore, they were properly considered in the context of the City’s site development permit approval. (Id. at. p. 734.) The holding of Ukiah on its unique facts does not amount to a rule that “height, view and privacy concerns” can never constitute substantial evidence of a potentially significant aesthetic impact on the environment. Regis also asserts: “The purported aesthetic impacts of this [pjroject are far less dramatic, and more subjective, than those at issue in the reported CEQA cases dealing with aesthetic issues.” This argument is a non sequitur. Regis cites cases that deal with impacts on scenic vistas and scenic resources as if only such aesthetic impacts were cognizable under CEQA. (Ocean View Estates, supra, 116 Cal.App.4th at pp. 401-403; Eller Media Co. v. Community Redevelopment Agency (2003) 108 Cal.App.4th 25, 35-36 [133 Cal.Rptr.2d 324]; Riverwatch v. County of San Diego (1999) 76 Cal.App.4th 1428, 1447 [91 Cal.Rptr.2d 322]; Quail, supra, 29 Cal.App.4th at pp. 1605-1606.) 26 But the question from the CEQA Initial Study Checklist quoted above makes clear that the potential harm need not be so “dramatic” as in these cases to give rise to a fair argument for doing an EIR: any “substantial[] degradation of] the existing visual character” is sufficient. (Guidelines, appen. G, § I, subd. (c).) Regis asserts: “[I]t is difficult to imagine how” the proposed project “could create the kind of objectively significant aesthetic impacts contemplated by Appendix G.” Appendix G does not speak of “objectively significant aesthetic impacts.” Regis’s inability to “imagine” an aesthetic impact is merely its own opinion, which cannot trump those of the objectors at this stage of CEQA review. And so far as Regis relies on the opinions stated in the MND on this subject, they are also at best substantial evidence contrary to the objectors’ evidence, which cannot refute the claim of a fair argument. (Guidelines, § 15064, subd. (f).) *939 In a supplemental letter brief, Regis cites the very recent decision, Bowman v. City of Berkeley (2004) 122 Cal.App.4th 572 [18 Cal.Rptr.3d 814] (Bowman). This decision is also unhelpful to Regis. In Bowman, the court rejected a claim of aesthetic impact sufficient to raise a fair argument for performing an EIR where scenic vistas and scenic resources were not at stake; however, the facts are fundamentally different from those in our case. The project in Bowman, supra, 122 Cal.App.4th 572, is a single four-story building intended to provide low-income housing for senior citizens, with retail on the ground floor. The building is to be located on a heavily trafficked thoroughfare on a site zoned for mixed-use residential and commercial development, now occupied by parking lots and a vacant, graffiti-scarred, one-story commercial building of no architectural value. (Id. at pp. 572-573.) As the court characterizes the objectors’ aesthetic arguments, they amount to the claim that the building should be one story lower, so as to fit in better with the scale of the surrounding residential neighborhood. (Id. at pp. 576-577.) Unsurprisingly, the court concludes that the difference between a three-story building and a four-story building does not amount to a significant environmental impact even under the fair argument standard. (Id. at p. 579.) The court relies in part on a federal case construing the National Environmental Policy Act, the federal counterpart to CEQA. {Bowman, supra, 122 Cal.App.4th 572, 578.) However, this part of its discussion is not necessary to the court’s conclusion, which would follow logically under the CEQA Guidelines from the objectors’ failure to show how a four-story building would be more likely than a three-story building to substantially degrade an already degraded visual environment. (So far as the court finds that CEQA was never intended to require an EIR “where the sole environmental impact is the aesthetic merit of a building in a highly developed area” (ibid.), we think that finding is also dictum on the facts before the court. In any event, we are not dealing in our case with the impact of a single building; thus, whether dictum or otherwise, the proposition is irrelevant.) In short, Bowman is distinguishable. On the issue of potential aesthetic impacts, as on that of land use policies, The Pocket Protectors have shown sufficient substantial evidence to support a fair argument that the project may have a significant effect on the environment. An EIR must be done on the proposed project. *940 DISPOSITION The judgment is reversed and the matter is remanded to the trial court with directions to enter a new judgment directing the City to undertake an EIR on the proposed project. The stay of on-site construction activity, previously entered by the court and as modified, shall remain in effect pending further order of this court. The Pocket Protectors shall receive their costs on appeal. (Cal. Rules of Court, rule 27(a).) Davis, J., and Nicholson, J., concurred. A petition for a rehearing was denied January 3, 2005, and the petition of both respondents and real parties in interest for review by the Supreme Court was denied March 30, 2005. Footnotes [1] The city and city council join in Regis’s brief on appeal. [2] According to the initial study for The Islands at Riverlake prepared by City planning staff, “Mousing projects similar to townhouses include cluster and row housing.” In other words, the term “[flownhouse (or similar development)” in the LPPT PUD did not mean detached single-family housing. [3] Permitted densities for the site were four to 15 dwelling units per net acre (du/na) according to the City’s General Plan, seven to 15 du/na according to the Pocket Community Plan, and eight du/na according to the PUD Guidelines for R-1A zoning. The townhouse units proposed in the previously approved project would have ranged from 1,100 to 1,500 square feet. [4] Another staff comment at this stage suggested that Regis consider including duplexes or townhouses in keeping with the townhome development plan of the PUD, or reducing the number of units and reconfiguring the houses into a single row on deeper lots. [5] The Riverlake Community Association originally approved the proposed project. By the time it reached the City Council, however, the association had withdrawn its approval. [6] Pace had not consulted with the City Attorney about this point. However, staff who were present when the original project was approved had told him that was their understanding. [7] Pace noted that the previously approved project had also included a 25-foot-wide private street, but conceded that houses would have been built on only one side of that street. [8] The commission’s chairman interjected that the commission had received many letters that talked about property values, but none that talked about new residents’ income levels. [9] Commissioners Jones and Valencia observed that they had received the final IS/MND, including responses to public comments, only that afternoon and had not had the chance to peruse it. [10] By this time, the development agreement for the prior unbuilt project, still in force during the planning commission proceedings, had expired. [11] Pace stated that City policy requires new private streets to be built to City standard, including sidewalks on both sides of the street. [12] Staff confirmed that the project included a three-foot sidewalk connecting the fronts of the homes and encroaching into the greenbelt. [13] Architect Roger McCardle, who drafted the alternative proposal, explained that it combined the houses in the proposed project into “halfplexes” to create greater setbacks and open space and avoid intrusion into the greenbelt without reducing the project’s density, in keeping with suggestions by the City’s long-range planner in April 2001. Councilmember Jones expressed a preference for the alternative over Regis’s proposal and moved that the matter be continued in order to obtain Regis’s and the Riverlake Community Association’s comments on the alternative. The motion died without a second. [14] Councilmember Cohen speculated that project opponents did not want anything built on the site. Councilmember Pannell speculated that the planning commission had rejected the project under duress. Planning Commissioner Taylor-Carroll, who was in attendance, took strong exception to Councilmember Pannell’s remark. [15] On appeal, The Pocket Protectors have abandoned the latter two issues. [16] The trial court also found, though deeming the issue moot, that the planning commission did not have sufficient evidence before it to justify denying the MND. [17] CEQA’s implementing regulations, the Guidelines, are found in California Code of Regulations, title 14, section 15000 et sequitur. All subsequent regulatory citations to the Guidelines are to title 14 of the Code of Regulations. [18] This language was added to section 21082.2 in 1984, some 10 years after No Oil, supra, 13 Cal.3d 68, was decided by our Supreme Court. (See Stats. 1984, ch. 1514, § 6, p. 5339.) This statutory language apparently nullifies No Oil’s remark that “the existence of serious public controversy concerning the environmental effect of a project in itself indicates that preparation of an EIR is desirable.” (No Oil, supra, 3 Cal.3d at pp. 85-86, fn. omitted.) [19] Regis notes that it adopted some of staff’s suggestions for ameliorating this effect, such as varying the heights and facades of adjacent houses. However, the fundamental plan to pack as many houses as possible on lots as small as possible along both sides of a long straight private street did not change. Thus, substantial evidence exists to support a fair argument that “canyoning” is still a feature of the approved project. [20] Regis attacks McCardle’s credentials as an expert. However, Regis fails to show that his expertise is insufficient under the fair argument standard. (See Friends, supra, 52 Cal.App.4th 1383, 1398-1399 & fn. 10.) [21] We recognize the development agreement has expired. Nevertheless, it tends to show that the City Council’s findings of fact as to zoning are arguably inconsistent with the PUD and the City Council’s original directives for developing the site. We also recognize that City planners (without input from the City Attorney) interpreted “single family residential” in the development agreement inconsistently with the express terms of the PUD. So far as the City Council may have relied on that interpretation, that merely creates a further basis for finding substantial evidence exists to support a fair argument that the City’s approval of the project and the MND conflicts with the PUD’s policies. [22] The same is true of Defend the Bay v. City of Irvine (2004) 119 Cal.App.4th 1261 at pages 1266 through 1268 [15 Cal.Rptr.3d 176], cited by Regis in a supplemental letter brief. That case is therefore likewise inapposite. [23] Regis asserts that some of the commissioners “admitted that they were not familiar with the whole record, including the responses to comments on the MND.” As mentioned above, two commissioners stated that they had only received the final MND, including responses to comments, shortly before the public hearing. (See fn. 9, ante.) But those responses at most constitute substantial evidence in favor of the MND’s approval, which cannot outweigh substantial evidence to the contrary. [24] Regis correctly observes that the trial court took this view in oral argument. [25] The other questions pertain to scenic vistas, scenic resources, and light or glare. [26] Regis also cites Woodward Park Homeowners Assn. v. Garreks, Inc. (2000) 77 Cal.App.4th 880 [92 Cal.Rptr.2d 268], but the only issue addressed in the published part of the decision was mootness. In addition, Regis cites yet another recent case which is inapposite because it deals with a challenge to a certified EIR. (Mira Mar Mobile Community v. City of Oceanside, supra, 119 Cal.App.4th 477, 484.)