*652 This case involves the California Environmental Quality Act (CEQA; Pub. Resources Code, § 21000 et seq. ; unless otherwise set forth, statutory references that follow are to the Public Resources Code) and CEQA Guidelines (Cal.Code Regs., tit. 14, § 15000 et seq. ) which are entitled to great weight. ( In re Bay-Delta etc. (2008) 43 Cal.4th 1143 , 1163, fn. 7, 77 Cal.Rptr.3d 578 , 184 P.3d 709 ( Bay-Delta ).)

California Department of Food and Agriculture (CDFA) prepared and certified a programmatic environmental impact *563 report (EIR) for a seven-year *653 program to eradicate an invasive pest-the light brown apple moth (LBAM), aka Epiphyas postvittana-but "at the last minute" approved instead a seven-year program to control LBAMs based on new information that eradication was no longer attainable. The EIR did not evaluate control as a reasonable alternative to eradication, and there was no supplemental environmental review in connection with the last-minute change. CDFA's position is that the change from pest eradication to pest control reduced the scope of the seven-year program. The opponents maintain the change actually expanded the scope of the program, because pest control activities will necessarily have to continue indefinitely after expiration of the seven-year period, as expressly acknowledged in the EIR.

These two appeals challenge judgments entered after the trial court denied two petitions for writ of administrative mandamus (Code Civ. Proc., § 1094.5 ) asserting CEQA violations and challenging the program approved by the California Department of Food and Agriculture (CDFA) and its former Secretary A.G. Kawamura. On our own motion, we ordered the appeals consolidated for purposes of oral argument and decision.

One group of appellants (case No. C072067) is comprised of North Coast Rivers Alliance, Stop the Spray Marin, California Alliance to Stop the Spray, Frank Egger, Whitney Merchant, Loralie Cioffi, Helen Kozoriz, Gayle McLaughlin, Robert Lieber, Tony Madrigal, Larry Bragman, Paulina Borsook, Sharon Luehs, Mike De Lay and Janice De Lay (collectively NCRA). They name as real parties in interest Aberdeen Road Company, Pacific Biocontrol Corporation, and ISCA Technologies Incorporated.

The other group of appellants (case No. C072617) is comprised of Our Children's Earth Foundation, Mothers of Marin Against the Spray, Stop the Spray East Bay, City of Albany, City of Berkeley, City of Richmond, City and County of San Francisco, Center for Environmental Health, Californians for Pesticide Reform, Pesticide Watch, Pesticide Action Network North America, Citizens for East Shore Parks, and Stop the Spray San Francisco (collectively OCEF).

Appellants contend the EIR violated CEQA by making assumptions unsupported by substantial evidence and by inadequately addressing environmental impacts, a reasonable range of alternatives, and cumulative impacts. Appellants also contend CDFA's "last-minute" approval of a control program instead of the eradication program rendered the environmental review deficient in failing to provide an accurate and stable project description, inadequately discussing alternatives, and improperly "segmenting" the project by reviewing a seven-year program but approving a seven-year program that will have to continue beyond seven years.

*654 We deny, as immaterial to resolution of this appeal, CDFA's July 16, 2013, and August 1, 2013, requests for judicial notice of federal orders (1) removing Santa Barbara County from the LBAM quarantine area on the ground no additional moths have been detected there, and (2) exempting additional crops from restrictions on interstate movement due to pest mitigation provided through industry standards of production, harvesting, and packaging practices.

We conclude that, even before the new information came to light that eradication was no longer attainable, the EIR violated CEQA by giving the project's "objective" an artificially narrow definition ("eradication of LBAMs") and thereby omitting analysis of pest control as a reasonable *564 alternative to the eradication program. The EIR acknowledged the project's "purposes" included protecting California native plants and agricultural crops from damage. While a control program may have achieved these "purposes" to some extent (as evidenced by the ultimate approval of a control program), the EIR declined to evaluate a control program as an alternative to an eradication program on the ground that a control program would not achieve the "objective" of eradication. The EIR did not even address in its cumulative impacts discussion the incremental effect of the reasonably foreseeable future need to continue anti-LBAM measures after expiration of the seven-year period. The EIR's omissions leave the record devoid of evidence to prove CDFA's claim that the last-minute change was legally acceptable because the adopted control program was narrower than the EIR's eradication program.

Our finding of CEQA violations as to some issues does not relieve us from reviewing appellants' other contentions. (§ 21005, subd. (c) [any court that finds a CEQA violation "shall specifically address each of the alleged grounds for noncompliance"]; but see, Communities for a Better Environment v. City of Richmond (2010) 184 Cal.App.4th 70 , 101-102, 108 Cal.Rptr.3d 478 [§ 21005 does not require appellate court to address additional alleged defects that may be addressed in a completely different and more comprehensive manner upon subsequent CEQA review following remand].)

We reverse the judgments and remand with directions.

FACTS AND PROCEEDINGS

The LBAM is an invasive moth native to Australia. Though appellants view LBAM as "innocuous," the record contains evidence supporting a contrary conclusion. The LBAM is a leaf-roller that makes a nest by rolling up plant leaves. The larvae feed on leaves and buds, reducing the plant's photosynthetic ability and deforming its growth patterns. LBAM is polyphageous, meaning it eats a variety of plants and trees. The larvae can feed *655 directly on some fruits, rendering them unmarketable. LBAMs multiply rapidly and can adapt quickly to feeding on new kinds of plants. The insects do not generally travel great distances on their own but spread quickly by inadvertent human transfer through the nursery trade.

The first LBAM reported in California was in Berkeley in February 2007. CDFA initiated a pheromone-baited trapping project, found additional LBAMs in other counties, and in April 2007 quarantined 182 square miles in Alameda, Contra Costa, San Francisco, Marin, and Santa Clara counties. In May 2007, the United States Department of Agriculture (USDA) issued a federal quarantine order requiring trapping, inspection, and certification of all nursery stock and host commodities in eight California counties.

The USDA Animal and Plant Health Inspection Service (APHIS) convened a technical working group (TWG) of experts from California, Australia, and New Zealand, to develop a plan to combat the pest. Although no one had ever succeeded in completely eradicating the LBAM (because the insect was either native to the country or deeply embedded after a number of years), the TWG determined complete eradication was possible in California due to the relatively small LBAM population in California at that time.

The Legislature in September 2007 enacted emergency legislation creating and funding a temporary Light Brown Apple Moth Program in the CDFA, authorizing CDFA to take action until January 1, 2010, *565 to combat the spread of the LBAM. (Former Food & Agr.Code, §§ 6050, 6050.1 ; Stats. 2007, ch. 190 (Sen. Bill No. 556), repealed by own terms on Jan. 1, 2010.) The Legislature found in former section 6050 that "[i]ntroduction of the light brown apple moth represents a clear, present, significant, and imminent danger to California's natural environment and agricultural industry. This is an insect species that feeds on over 250 species of native and ornamental plants, fruits, and vegetables. [¶] ... [¶] ... [and also] will feed on alder, ceanothus, columbine, cottonwood, cypress, ferns, fir, hawthorn, honeysuckle, lupine, madrone, oak, pine, poplar, redwood, spruce, and willow." (Former § 6050 ; Stats. 2007, ch. 190, § 2.) "There is an imminent threat for adverse effect and ultimate extinction to some of these sensitive species if this pest becomes permanently established in California." ( Ibid. ) At that time, LBAM was "currently found in the urban and natural areas in all parts of nine California counties and could move into agricultural croplands." ( Ibid. ) The state has a "great interest in protecting its native species and agricultural products from further harm caused by [LBAMs]." ( Ibid. ) The legislation required CDFA's Secretary to provide support staff and logistical support "for eradicating the [LBAM]." (Former § 6050.1 ; Stats. 2007, ch. 190, § 2.) CDFA was required to "annually review the progress made by each local agency in eradicating the [LBAM], and *656 make recommendations, as needed, to improve individual local agency eradication efforts." ( Ibid. )

In late 2007, CDFA and USDA started emergency treatment of isolated LBAM populations.

The LBAM nevertheless spread rapidly. The numbers of LBAMs trapped by CDFA were 10,285 in 2007; 62,346 in 2008; and 71,867 as of March 2009. By 2009, the quarantine area had expanded to 13 counties in California.

In February 2008, CDFA filed a CEQA Notice of Preparation of a "Programmatic [EIR] (PEIR) for the eradication of the [LBAM]."

The draft EIR (DEIR) which was released for public review at the end of July 2009, stated the "goal" was to eradicate LBAMs from California by 2015 (later reset at 2017). "The CDFA's objective is to eradicate LBAM from the state of California by 2015. Initial efforts will be to contain the LBAM population and not allow LBAM to spread into adjacent counties. The CDFA will then suppress heavily infested areas around production nurseries, suppress and eradicate populations in outlying counties to remove these counties from regulation, and finally eradicate populations throughout the state. Eradication is to be accomplished using a variety of tools with reliance primarily on biological control methods applied in an effective and environmentally safe manner. [¶] The Program is needed to protect the state's native plants, forest species, agronomically important crops, and ornamental plants from damage by this invasive pest species. The CDFA must protect agriculture from invasive pests to protect the nation's food supply, the environment, and the economy. Furthermore, eradication of the pest within the state is needed to prevent its expansion to other states and other countries."

The DEIR's chapter on Alternatives reveals a failure to appreciate CEQA's requirement to study alternatives to the program. (§§ 21001, subd. (g), 21100, subd. (b)(4); Guidelines, § 15126.6.) It studied as "alternatives" seven tools (five of which were approved) to be used in whatever combination was best suited for particular sites to be treated. The DEIR said: "This chapter summarizes the analysis of alternatives for LBAM eradication. It presents *566 a series of potential alternatives or 'tools' and screening criteria to produce a 'toolbox' of options to support the LBAM Eradication Program. " (Italics added.)

The five approved tools mislabeled as "alternatives" were:

(1) sterile insect technique (SIT), releasing sterile moths into the environment to mate with wild moths;

*657 (2) mating disruption pheromone (to attract males and prevent them from mating without killing them) using twist-ties (MD-1), placing plastic twist-ties infused with LBAM pheromone in places of small isolated LBAM infestations;

(3) mating disruption pheromone applied using ground-based equipment (MD-2), applying LBAM pheromones to trees and shrubs in residential yards and to telephone poles and trees on public property along roadways;

(4) inundative parasite wasp release (Bio-P) releasing an egg parasitoid, which is a native stingless wasp-a predator to the moths-near foliage where LBAM have been detected; and

(5) foliar ground treatments with approved insecticides (Btk and S). Btk is the biopesticide Bacillus thuringiensis kurstaki, a live bacterium that invades the host organism. "S" is the chemical pesticide Spinosad.

DEIR's Appendix H-"Process Used to Select Tools for Use in the [LBAM] Eradication Program" by LBAM Eradication Program Director Robert Dowell dated April 5, 2009-stated: "The goal of the [LBAM] Eradication Program is the elimination of breeding populations of the moth from California. This is fundamentally different [from] controlling the pest. [Italics added.] In control programs the goal is to protect a specified area such as a crop from the damage caused by the moth. Only a portion of the pest population is treated; that which threatens the area to be protected. The control measures are applied to the area to be protected and it is assumed that some damage is acceptable. Control measures are generally not applied outside the area to be protected. Thus, if an exotic pest becomes permanently established in California, control measures will be needed forever. [Italics added.] [¶] Eradication programs treat the entire pest population with the goal of eliminating it. If successful, the pest is gone and additional, permanent control measures are no longer needed. [¶] LBAM has never been the target of a previous eradication program. Therefore there are no successful model programs for this program to follow. In these circumstances [CDFA] and [USDA] have relied on the use of successful control measures to eradicate the pest. The difference is that instead of simply lowering the pest population in a selected portion of its population (control) we treat the entire population to elimination (eradication). [¶] The CDFA used a step-wise process to evaluate a number of potential tools for use in the LBAM Eradication Program. The process involved an initial screen to determine which tools merited further evaluation and then a second evaluation to determine which tools will actually be used in the program.... This information was used to select the tools to be evaluated further for use in the eradication program."

*658 The DEIR stated: "The Program anticipates using all of the chemical and nonchemical alternatives (and options) in combination as part of an integrated pest management Program."

However, the DEIR also spoke of an apparently different Integrated Pest Management (IPM), characterized as a "control" strategy that was therefore "not evaluated further in the process to determine *567 which tools would be used in the LBAM Eradication Program. IPM as an approach to pest eradication provides the framework in which individual tools were evaluated."

The DEIR discussed a "No-Program Alternative" (NPA), whereby ongoing activities of quarantine, detection, and inspection would continue, but without application of the tools. Restrictions on domestic and foreign trade would increase. The DEIR anticipated that, absent eradication, crops would be damaged and Californians would increase private use of pesticides to control LBAM in yards and agricultural areas.

CDFA held public meetings and received numerous comments from the public.

On February 26, 2010, CDFA published the final EIR (FEIR). The FEIR noted it had received comments that some experts opined eradication may not be feasible, but CDFA was going to follow its own experts who opined the available tools should be sufficient to accomplish the goal of eradication. (Guidelines, § 15151 ["Disagreement among experts does not make an EIR inadequate, but the EIR should summarize the main points of disagreement among the experts"].)

The FEIR stated: "The difference between control and eradication does not lie in the tools used but rather in the scope and intensity with which the tools are used, as explained below:

"In control programs the goal is to lower pest numbers below the economic injury level in the crop areas. In control programs it is assumed that the growers can tolerate some damage and that the pest will never be eliminated from the area. The tools are applied only as long as necessary [emphasis added] to lower pest numbers, and they are applied only to the crop to be protected. Pest populations outside the croplands are not treated.

"Eradication uses the same tools but in a different fashion. In eradication programs, the tools are applied to the entire pest population, not just part of it, and the tools are consistently applied until the pest is entirely *659 eliminated not just reduced to a number below the chosen economic threshold. This approach has been successfully used to eradicate numerous pests including but not limited to, Mexican bean beetle, Japanese beetle, Hall scale, tulip tree scale, Mediterranean fruit fly, Mexican fruit fly, kharpa beetle, and Caribbean fruit fly. International guidelines (ISPM # 9 1998) pertain to how eradication programs are conducted and what conditions must be met to declare successful eradication. The programs conducted by USDA and CDFA meet these guidelines.

"Eradicating a population of an exotic pest from California does not mean the pest cannot reinvade the state. The Program has stated that it will conduct an annual review each December of its progress and evaluate whether the goal of eradication still appears to be attainable (CDFA Report to Legislature 2008). The latest available review was conducted in December 2008."

The public comments complained the DEIR did not analyze a reasonable range of alternatives to the program, including an "Integrated Pest Management" (IPM) control program, classic biological control, removing quarantine restrictions, and monitoring.

The final EIR responded IPM is "not a tool but an approach to controlling pests.... '[It] first assesses the pest situation, evaluates the merits of pest management options and then implements a system of complementary management actions within a defined area. The goal of IPM is to mitigate pest damage while protecting *568 human health, the environment and economic viability.' The components of an IPM program include setting action thresholds, monitoring and detection, proper identification, and action/implementation.... The tools used in an IPM program include biological control, cultural controls such as clean culture and habitat manipulation, chemical controls (including insect pheromones, pesticides, and biologically produced toxins), and genetic control with resistant plants and quarantines (University of California IPM Program 2008; Ciborowski 2007).

"All CDFA insect eradication programs represent an integrated approach in which action thresholds are set, detection and monitoring programs are established, target organisms are properly identified, available tools are identified and evaluated for their effectiveness and environmental impacts, and the program is then implemented. For LBAM, the action threshold has changed from the detection of a single moth to the detection of two or more moths within 3 miles of each other within a time period equal to a single lifecycle. Detection and monitoring is done using over 40,000 traps statewide baited with LBAM pheromone. Suspect LBAMs are sent to a CDFA expert for identification. The first LBAM detected in each county is sent to USDA Systematic Entomology Laboratory for confirmation. Like IPM programs, the *660 LBAM Eradication Program evaluated all available tools including insecticides for possible inclusion in the LBAM Program.

"The main difference between classical IPM and CDFA eradication programs is the desired endpoint. For IPM programs, the goal is to use one or more control measures to lower the pest populations within the defined area below economically damaging levels. It is assumed that some damage can be tolerated and that these measures will be needed into the foreseeable future. For CDFA eradication programs, the goal is the elimination of the pest so that control measures will not be needed at all. It is notable, though, that the Program includes components, such as organically approved insecticides and SIT, which many laypeople identify as 'IPM' techniques. While the identification of these components as 'IPM' is not technically correct, the Program components do contain environmentally sensitive, species-specific features that many people mistakenly identify as 'IPM.' "

In response to public comments about alternatives of removing quarantine restrictions and monitoring, the final EIR responded, "neither are measures to physically control LBAM populations. Rather, it simply provides some relief from the economic harm that will be suffered by growers who would not be able to market their products due to the quarantines." The response further stated that reclassification to remove quarantines would allow for a greater rate of spread; USDA was evaluating a petition for reclassification; reclassification is outside the scope of CDFA's authority; and "[b]ased on the information available to CDFA regarding the potential threat to the environment (including California's native plants) and economy from LBAM, CDFA does not consider reclassification a feasible alternative." The response added that the current trapping program was monitoring LBAMs' spread; the county agricultural commissioners were monitoring for damage caused by leafrollers and requesting identification of larvae to determine if the damage was caused by LBAMs. "Because monitoring will not accomplish the Program objective of eliminating the threats to the economy and the environment posed by LBAM, CDFA does not consider it a feasible alternative."

*569 On March 12, 2010, CDFA received advance notice of a new release issued by USDA's APHIS three days later, stating eradication of LBAMs was no longer feasible: "[S]ince the pest was detected, LBAM has spread to such an extent in California that eradication in the state is not feasible until new tools, such as sterile insect technology [SIT], are fully developed and ready for widespread use in concern with currently available methods. Therefore, APHIS is shifting to a control and suppression strategy and will maintain its survey and regulatory framework, including enforcement of phytosanitary measures, to ensure that LBAM is not spread to other states or trading partners."

*661 On March 22, 2010, CDFA's Secretary certified as complete the EIR for the seven-year eradication program, declared it complied with CEQA, yet adopted findings and approved a seven-year control program. The Findings approved a program for containment, control, suppression, and eradication where feasible (i.e., small, discrete infestations in outlying areas). Although the DEIR had stated eradication was "fundamentally different" from control, CDFA now found the new objectives merely "differ somewhat" from the objectives in the draft. CDFA stated that, while the scope of the project was being reduced, all of the five remaining approved tools had been fully analyzed in the draft and final EIRs and were applicable to the reduced program because, as explained in the FEIR and Findings, "[t]he difference between control and eradication does not lie in the tools used but rather in the scope and intensity with which the tools are used." "In control programs the goal is to lower pest numbers below the economic injury level in the crop areas.... The tools are applied only as long as necessary to lower pest numbers, and they are applied only to the crop to be protected.... [¶] Eradication uses the same tools but in a different fashion. In eradication programs, the tools are applied to the entire pest population ... and the tools are consistently applied until the pest is entirely eliminated not just reduced to a number below the chosen economic threshold."

CDFA's Findings stated, under a heading of "Other Alternatives Analyzed, Considered and Rejected" that " 'Integrated Pest Management (IPM) is an approach which first assesses the pest situation, evaluates the merits of pest management options and then implements a system of complementary management actions within a defined area. The goal of IPM is to mitigate pest damage while protecting human health, the environment and economic viability.' The components of an IPM program include setting action thresholds, monitoring and detection, proper identification, and action/implementation [citation]. Classical IPM programs may include biological control, cultural controls such as clean culture and habitat manipulation, chemical controls (including insect pheromones, pesticides, and biologically produced toxins), and genetic control with resistant plants and quarantines [citation]. [¶] The Program incorporates many of these features in that it includes an integrated approach in which action thresholds are set, detection and monitoring programs are established, target organisms are properly identified, available tools are identified and evaluated for their effectiveness and environmental impacts, and the program is then implemented. However, for IPM programs, in contrast to the proposed Program, control measures are used to lower the pest populations within the defined area below economically damaging levels. For IPM programs, [it] is assumed that some damage can be tolerated and that these measures will be needed into the foreseeable future. [Italics added.] [¶] These features of IPM are inconsistent *570 with *662 the Program's objectives of containing, controlling, suppressing, and eradicating LBAM populations within California. [Fn. omitted.] Moreover, to the extent IPM programs include applications of pesticides and release of biologically produced toxins, they may have potentially significant environmental impacts, particularly on biological resources and water quality. Accordingly, IPM would conflict with the Program objective of implementing the Program primarily through biological control and other methods that can be applied in an effective and environmentally safe and responsible manner, and IPM is environmentally inferior to the alternatives proposed for implementation in the proposed Program. Consequently, this alternative is rejected."

This foregoing passage in the Findings included a footnote that "As noted in Master Response C in the Final EIR, the elements of a classical IPM program include biological control and use of insect pheromones and pesticides. Likewise, the alternatives evaluated for potential implementation under the Program include elements of biological control (... Bio-P), use of insect pheromones (... MD-1, MD-2, MD-3 and MMA), and use of pesticides (... Btk and S). This overlap caused many members of the public to be confused about whether the Program is an IPM program. The Program is not an IPM program because the Program aims to use available tools to eradicate LBAM populations where it is feasible to do so. However, the Program's ... tools do include environmentally sensitive, species-specific features that many lay people mistakenly identify as 'IPM.' "

CDFA rejected the "No-Program Alternative," because LBAM populations would spread. Absent a CDFA-led effort to contain, control, and suppress LBAMs, the existing quarantine, detection, and inspection activities mandated by regulations of CDFA, Canada, and Mexico-which are part of the environmental baseline-would be continued, and restrictions on domestic and foreign trade would increase as LBAM populations increased. Without a coordinated control program, a five to 20 percent crop yield loss would be expected based on New Zealand's experience. Spread of the pest could impact grapes, citrus, peaches, plums, cherries, apricots, as well as trees such as pine, cypress, and oak. If LBAM caused extensive damage to plants, it may have adverse impacts on sensitive animals dependent on those plants for breeding, feeding, or shelter. Almost every type of ornamental plant would be in danger of being damaged by LBAM larvae. Given the risk of damage, it was "highly likely" that private individuals (residential and commercial) would use insecticides to control LBAM.

CDFA concluded potential adverse impacts associated with the program could be mitigated to a less-than-significant level through implementation of *663 mitigation measures (e.g., restricting pesticides during spawning seasons), and the program would not result in any significant, unavoidable environmental impacts.

CDFA's findings declared the program's objectives to be to "protect the nation's food supply, and protect and promote California's agricultural economy and the environment by keeping LBAM from attaining damaging levels; [¶] contain, control and suppress LBAM in areas where current LBAM population densities and the extent of the contiguous spread of the pest make eradication infeasible at this time; [¶] eradicate small, discrete LBAM populations within California; [¶] ... [¶] protect California's natural resources ... from environmental damage by LBAM; [¶] protect the larger environment by preventing the expansion of *571 LBAM to other states and areas; and [¶] avoid the expense, environmental risks, and administrative difficulties of implementing permanent quarantines and an agricultural inspection program to address persistent LBAM populations statewide."

Appellants filed their petitions for writ of administrative mandamus. The two petitions were argued together, and the trial court issued a detailed decision denying the petitions and entered judgments.

DISCUSSION

I

Standard of Review

" 'In reviewing an agency's compliance with CEQA in the course of its legislative or quasi-legislative actions, the courts' inquiry "shall extend only to whether there was a prejudicial abuse of discretion." (Pub. Resources Code, § 21168.5...; see Western States Petroleum Assn. v. Superior Court (1955) 9 Cal.4th 559 , 568, 38 Cal.Rptr.2d 139 , 888 P.2d 1268 ; Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376 , 392-393, 253 Cal.Rptr. 426 , 764 P.2d 278 ( Laurel Heights I ).)

" 'An appellate court's review of the administrative record for legal error and substantial evidence in a CEQA case, as in other mandamus cases, is the same as the trial court's: The appellate court reviews the agency's action, not the trial court's decision; in that sense appellate judicial review under CEQA is de novo.... We therefore resolve the substantive CEQA issues ... by independently determining whether the administrative record demonstrates any legal error by the Department and whether it contains substantial evidence to support the Department's factual determinations.' ( *664 Vineyard Area Citizens for Responsible Growth, Inc. v. City of Rancho Cordova [ (2007) ] 40 Cal.4th [412,] 426-427, 53 Cal.Rptr.3d 821 , 150 P.3d 709 , fns. omitted.)" ( Center for Biological Diversity v. Dept. of Fish and Wildlife (2015) 234 Cal.App.4th 214 , 231-232, 183 Cal.Rptr.3d 736 ( Center for Biological Diversity , orig. brackets omitted).)

II

Distinction Between Program EIR and Project EIR

The EIR in this case was a program EIR, as opposed to a project EIR. A CEQA "project" can be a program. A project EIR is typically used for a specific development project. (Guidelines, § 15161.) In contrast, "[a] program EIR is an EIR which may be prepared on a series of actions that can be characterized as one large project and are related" in specified ways, including "[a]s individual activities carried out under the same authorizing statutory or regulatory authority and having generally similar environmental effects which can be mitigated in similar ways." (Guidelines, § 15168, subd. (a).)

A program EIR has advantages for the public agency, in that it is possible to conduct subsequent activities without preparing a new EIR if the agency finds that no new effects could occur or no new mitigation measures would be required. (Guidelines, §§ 15168(c); Center for Biological Diversity, supra, 234 Cal.App.4th at p. 233 , 183 Cal.Rptr.3d 736 .)

A program EIR should contain a sufficient degree of analysis, in the light of what is reasonably feasible, to provide decisionmakers with information that enables them to make a decision which intelligently takes account of environmental consequences. ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 234 , 183 Cal.Rptr.3d 736 .) " 'The courts have looked not for perfection but for adequacy, completeness, and a good faith effort at full *572 disclosure.' " ( Ibid. citing Guidelines, § 15151.)

III

Last-Minute Change

Appellants argue the last-minute change from eradication to control violated CEQA in various ways, such as by rendering the project description unstable and inaccurate, and by improperly "segmenting" the matter by studying a program of limited duration but approving a program of unlimited duration. However, the approved program was of limited duration of seven years.

*665 On the other hand, appellants' fear that the pest control activities may continue after seven years without a new EIR is justified. CEQA Guidelines allow use of an earlier EIR for later activities in some circumstances. (E.g., Guidelines, §§ 15153(a) ["agency may use an earlier EIR prepared in connection with an earlier project to apply to a later project, if the circumstances of the projects are essentially the same"], 15168(c) [program EIR may be used for later activities, "and no new environmental document would be required," if agency finds no new effects would occur or no new mitigation measures would be required].)

Here, we cannot accept the assurances of CDFA's attorney at oral argument that a new EIR will be prepared for any activity post-2017. The trial court asked for a stipulation or stipulated order to that effect. (RT 17-20, 37-39, 56-58) CDFA drafted a stipulation (North CoastAA 373) that it "hereby confirms that-consistent with paragraph 34 of [its] March 22, 2010 findings of fact [citation to administrative record] and paragraph 6 of [its] March 22, 2010 certification of the certified EIR [citation to administrative record]-the LBAM program approved by [CDFA] on March 22, 2010, will not continue beyond seven (7) years from the date the certified EIR was certified (that is, beyond March 22, 2017), without the Department first undertaking further environmental review under [CEQA]." (NCRA-AA 370) The referenced paragraph 34 stated, "The risk assessments for the PEIR analyzed the potential impacts associated with implementing the Program for seven years. Since the starting date of the Program will be 2010, the Program could be implemented through 2017 within the scope of the analysis of the risk assessments." (AR 13-14) The referenced paragraph 6 stated, "Should CDFA wish to continue implementing the Program's alternative tools beyond the seven-year period analyzed in the existing risk assessments, additional CEQA review may be required." (Italics added.) (AR 48)

Appellants opposed the proposed stipulation. (AA 375) Despite CDFA's equivocation, the trial court stated in its written ruling on the merits that additional review after seven years is required by CDFA's confirmation of the seven-year duration of the approval. (AA 387-388)

CDFA continues to equivocate on appeal. Its appellate brief states, "if in 2017 CDFA determines the LBAM Program is still needed for the protection of the state's economy and environment, CDFA will conduct additional environmental review under CEQA prior to continuing with the LBAM Program." (NRB 12) Notably, CDFA does not say such additional environmental review would take the form of a new EIR. Additional review could mean CDFA considers whether a new EIR is needed and decides no. This is apparent in CDFA's further statement in its appellate brief, that it "represented that at the end of seven years, it would re-evaluate the need for continued action and any further environmental review." (NRB 29)

*573 *666 We need not belabor appellants' arguments that the last-minute change violated CEQA; we will conclude that, even without the last-minute change, the EIR violated CEQA, requiring reversal of the judgments. The last-minute change to a seven-year program does not save CDFA from reversal, because there is no assurance that a CEQA-compliant EIR will be prepared in the event of post-2017 activity.

IV

EIR Failed To Address Reasonable Range of Alternatives

Appellants complain the EIR violated CEQA by failing to consider a reasonable range of alternatives in the DEIR, particularly a control program as an alternative to an eradication program, and CDFA violated CEQA by failing to reconsider alternatives after its last-minute change from an eradication program to a control program. We conclude that, even before the last-minute change, the EIR was fatally defective in failing to analyze a control program as a reasonable alternative to an eradication program.

A. CEQA Law Regarding Alternatives

The Legislature has declared it the policy of the state to "[r]equire governmental agencies at all levels ... to consider alternatives to proposed actions affecting the environment." (§ 21001, subd. (g).) "The Legislature finds and declares that it is the policy of the state that public agencies should not approve projects as proposed if there are feasible alternatives or feasible mitigation measures available which would substantially lessen the significant environmental effects of such projects, and that the procedures required by this division are intended to assist public agencies in systematically identifying both the significant effects of proposed projects and the feasible alternatives or feasible mitigation measures which will avoid or substantially lessen such significant effects. The Legislature further finds and declares that in the event specific economic, social, or other conditions make infeasible such project alternatives or such mitigation measures, individual projects may be approved in spite of one or more significant effects thereof." (§ 21002; see also, Bay-Delta, supra, 43 Cal.4th at p. 1169 , 77 Cal.Rptr.3d 578 , 184 P.3d 709 .)

The Guidelines provide: "An EIR shall describe a range of reasonable alternatives to the project ... which would feasibly attain most of the basic objectives of the project but would avoid or substantially lessen any of the significant effects of the project, and evaluate the comparative merits of the alternatives. An EIR need not consider every conceivable alternative to a project. Rather it must consider a reasonable range of potentially feasible alternatives that will foster informed decisionmaking and public participation.

*667 An EIR is not required to consider alternatives which are infeasible. The lead agency is responsible for selecting a range of project alternatives for examination and must publicly disclose its reasoning for selecting those alternatives. There is no ironclad rule governing the nature or scope of the alternatives to be discussed other than the rule of reason. [Citation to case law.]" (Guidelines, § 15126.6, subd. (a), italics added.) "The process of selecting alternatives to be included in the EIR begins with the establishment of project objectives by the lead agency. 'A clearly written statement of objectives will help the lead agency develop a reasonable range of alternatives to evaluate in the EIR and will aid the decision makers in preparing findings.... The statement of objectives should include the underlying purpose of the project. ' " ( *574 Bay-Delta, supra, 43 Cal.4th at p. 1163 , 77 Cal.Rptr.3d 578 , 184 P.3d 709 , italics added, citing Guidelines, § 15124, subd. (b).)

"The range of alternatives required in an EIR is governed by a 'rule of reason' that requires the EIR to set forth only those alternatives necessary to permit a reasoned choice. The alternatives shall be limited to ones that would avoid or substantially lessen any of the significant effects of the project. Of those alternatives, the EIR need examine in detail only the ones that the lead agency determines could feasibly attain most of the basic objectives of the project. The range of feasible alternatives shall be selected and discussed in a manner to foster meaningful public participation and informed decision making." (Guidelines, § 15126.6, subd. (f).)

"[I]nfeasible alternatives that do not meet project objectives need not be studied even when such alternatives might be imagined to be environmentally superior. Tasked with the study of a proposal to build a new shopping center, a public agency need not study a fruit stand as an alternative." ( Saltonstall v. City of Sacramento (2015) 234 Cal.App.4th 549 , 556-557, 576-577, 183 Cal.Rptr.3d 898 ( Saltonstall ) [no need to study alternative of remodeling existing basketball arena, because it would not meet city's objective to create an attraction to revitalize previously-blighted areas].)

B. Failure to Consider Control as an Alternative to Eradication

As indicated, the EIR confusingly mislabeled various tools for attacking LBAMs as "alternatives" to the program. And the EIR rejected out of hand anything that would not achieve complete eradication of the LBAMs, e.g., a control program of integrated pest management.

The EIR did not decline to study control as an alternative to eradication on the ground a control program would not lessen the environmental effects. Rather, the EIR declined to evaluate a control program as an alternative on the ground it would not achieve the stated goal of eradication. However, this position confuses the CEQA project, objectives, and purposes.

*668 The objective of the program was to protect California's native plants and agricultural crops from damage. That the DEIR defined the program "objective" as "eradication" was an improper "artificially narrow" definition ( Bay-Delta, supra, 43 Cal.4th at pp. 1163-1166, 77 Cal.Rptr.3d 578 , 184 P.3d 709 ), as is apparent from the DEIR's distinction between "objective" and "purpose" and relegation of plant/crop protection to a "purpose" as opposed to an "objective." Thus, the DEIR stated:

" PROGRAM OBJECTIVES AND PURPOSE

"The CDFA's objective is to eradicate LBAM from the state of California by [2017]. Initial efforts will be to contain the LBAM population and not allow LBAM to spread into adjacent counties. The CDFA will then suppress heavily infested areas around production nurseries, suppress and eradicate populations in outlying counties to remove these counties from regulation, and finally eradicate populations throughout the state. Eradication is to be accomplished using a variety of tools with reliance primarily on biological control methods applied in an effective and environmentally safe manner. [¶] The Program is needed to protect the state's native plants, forest species, agronomically important crops, and ornamental plants from damage by this invasive pest species. The CDFA must protect agriculture from invasive pests to protect the nation's food supply, the environment, and the economy. Furthermore, eradication of the pest within *575 the state is needed to prevent its expansion to other states and other countries."

Thus, the EIR purports to view eradication as the "objective" and protection of crops/plants as the "purpose." However, " 'The statement of objectives should include the underlying purpose of the project.' " ( Bay-Delta, supra, 43 Cal.4th at p. 1163 , 77 Cal.Rptr.3d 578 , 184 P.3d 709 , citing Guidelines, § 15124, subd. (b).)

Moreover, "[a] lead agency may not give a project's purpose an artificially narrow definition." ( Bay-Delta, supra, 43 Cal.4th at p. 1166 , 77 Cal.Rptr.3d 578 , 184 P.3d 709 .) An agency "may structure its EIR alternative analysis around a reasonable definition of underlying purpose and need not study alternatives that cannot achieve that basic goal. For example, if the purpose of the project is to build an oceanfront resort hotel [citation] or a waterfront aquarium [citation], a lead agency need not consider inland locations. [Citation.]" ( Bay-Delta, supra, 43 Cal.4th at p. 1166 , 77 Cal.Rptr.3d 578 , 184 P.3d 709 .)

Bay-Delta involved a long-term, comprehensive plan to restore the Bay-Delta's ecological health and improve management of Bay-Delta water for the various beneficial uses that depend on it. ( Id. 43 Cal.4th at pp. 1151-1152, 77 Cal.Rptr.3d 578 , 184 P.3d 709 .) The Supreme Court held the failure to examine in detail a program alternative requiring reduced water exports from the Bay-Delta was *669 not an abuse of discretion, because that alternative would not achieve the program objective of water supply reliability, i.e. reduction of the mismatch between Bay-Delta water supplies and the current and project beneficial uses dependent on the Bay-Delta system. ( Id. at pp. 1163-1166, 77 Cal.Rptr.3d 578 , 184 P.3d 709 .)

Here, in contrast, protection of plants and crops were clearly objectives and the underlying purpose of the eradication program.

Moreover, the goal of eradication was always known to be tenuous, because LBAMs kept spreading despite the emergency treatments authorized by the Legislature, spreading to more counties during the DEIR process, and the Legislature required CDFA to monitor LBAMs on an annual basis. Additionally, the DEIR's appendix on the process used to select tools acknowledged that "if an exotic pest becomes permanently established in California, control measures will be needed forever."

That defining the objective as eradication was "artificially narrow" is further apparent from the fact that the last-minute change did not describe the revised program objective as simply "to control" LBAMs. Rather, CDFA's own Findings in approving the control program acknowledge "As revised, the objectives [italics added] of the LBAM Program [include]: [¶] protect the nation's food supply, and protect and promote California's agricultural economy and the environment by keeping LBAM from attaining damaging levels; [¶] ... [¶] protect California's natural resources ... from environmental damage by LBAM; [¶] [and] protect the larger environment by preventing the expansion of LBAM to other states and areas...."

A control program should have been studied as an alternative to eradication, particularly given the EIR's acknowledgement that eradication could become unattainable at any time.

It is possible a control program may not be a viable alternative on the ground its unending nature would be more harmful to the environment than an eradication program. On the other hand, it could be that an eradication program would require more intensive application of tools than a control program, leveling out the environmental *576 impacts. We do not know, because CDFA skirted the issue. Had CDFA evaluated a "control" program as an alternative to an "eradication" program, CDFA may have rejected a control program on the ground the unending nature of it would be more harmful to the environment. At least that is the inference to be drawn from the EIR's drumbeat distinguishing the two. But that just goes to show the prejudice of CDFA's last-minute shift from eradication to an unstudied control program.

We conclude the EIR violated CEQA by failing to study control as an alternative to eradication. That error infected the entire EIR insofar as it *670 dismissed out of hand anything that would not achieve complete eradication of LBAMs. The last-minute change to the control program made the CEQA violation unequivocally prejudicial requiring reversal, as we next discuss.

C. Prejudice

Section 21005 provides: "(a) The Legislature finds and declares that it is the policy of the state that noncompliance with the information disclosure provisions of this division which precludes relevant information from being presented to the public agency, or noncompliance with substantive requirements of this division, may constitute a prejudicial abuse of discretion within the meaning of Sections 21168 and 21168.5, regardless of whether a different outcome would have resulted if the public agency had complied with those provisions. [¶] (b) It is the intent of the Legislature that, in undertaking judicial review pursuant to Sections 21168 and 21168.5, courts shall continue to follow the established principle that there is no presumption that error is prejudicial. [¶] (c) It is further the intent of the Legislature that any court, which finds, or, in the process of reviewing a previous court finding, finds, that a public agency has taken an action without compliance with this division, shall specifically address each of the alleged grounds for noncompliance."

In Neighbors for Smart Rail v. Exposition Metro Line Construction Authority (2013) 57 Cal.4th 439 , 160 Cal.Rptr.3d 1 , 304 P.3d 499 ( Neighbors for Smart Rail ), a plurality of the court discussed prejudice in the CEQA context, that an omission in an EIR (in that case, an omission in significant impacts analysis) "is deemed prejudicial if it deprived the public and decision makers of substantial relevant information about the project's likely adverse impacts. Although an agency's failure to disclose information called for by CEQA may be prejudicial 'regardless of whether a different outcome would have resulted if the public agency had complied' with the law (§ 21005, subd. (a)), under CEQA 'there is no presumption that error is prejudicial' (§ 21005, subd. (b)). Insubstantial or merely technical omissions are not grounds for relief. [Citation.] 'A prejudicial abuse of discretion occurs if the failure to include relevant information precludes informed decisionmaking and informed public participation, thereby thwarting the statutory goals of the EIR process.' [Citation.]" ( Neighbors for Smart Rail, supra, 57 Cal.4th at p. 463 , 160 Cal.Rptr.3d 1 , 304 P.3d 499 [EIR's use exclusively of future conditions baseline to analyze traffic congestion impacts was not prejudicial, where EIR demonstrated lack of grounds to suppose analysis of existing traffic conditions would have produced any substantially different information].)

The burden is on the agency to establish lack of prejudice. ( *671 Environmental Protection Information Center v. California Dept. of Forestry and Fire Protection (2008) 44 Cal.4th 459 , 488, 80 Cal.Rptr.3d 28 , 187 P.3d 888 ; *577 Laurel Heights I , supra, 47 Cal.3d at p. 388 , 253 Cal.Rptr. 426 , 764 P.2d 278 .)

Here, the prejudice in the EIR's failure to address control as an alternative to eradication is apparent, because the EIR dismissively rejected anything that would not achieve eradication. As we discuss post, this prejudice was compounded because the EIR's cumulative impacts discussion failed to address the reasonably foreseeable need for continued control activities. Accordingly, CDFA's ultimate selection of an alternative (control program) not analyzed in the EIR left the EIR inadequate as failing to include relevant information and precluding informed decisionmaking and informed public participation.

The last-minute change from an eradication program to a control program did not cure the prejudice. Chaparral Greens v. City of Chula Vista (1996) 50 Cal.App.4th 1134 , 58 Cal.Rptr.2d 152 ( Chaparral Greens ) held the failure to revise an EIR based on new information did not violate CEQA, because substantial evidence supported the agency's decision that the new information was insignificant. ( Id. at pp. 1147-1148, 58 Cal.Rptr.2d 152 .) Here, the new information was clearly significant, since it caused CDFA to change the program. Although CDFA claims the change reduced the scope of the program, appellants make the plausible argument that the change expanded the scope of the program into one of indefinite duration. The manner in which CDFA conducted the environmental review process-ignoring anything that would not achieve eradication-leaves the record without substantial evidence to support CDFA's claim.

Though not cited by the parties, we discussed the situation where a public agency selected an alternative not analyzed in the EIR based on new information, in Western Placer Citizens for an Agricultural and Rural Environment v. County of Placer (2006) 144 Cal.App.4th 890 , 50 Cal.Rptr.3d 799 ( Western Placer CARE ).

In Western Placer CARE, supra, 144 Cal.App.4th 890 , 50 Cal.Rptr.3d 799 , we held a county's EIR analyzing a proposed aggregate mine did not violate CEQA by failing to include and analyze a revised project description submitted by the developer after the final EIR was prepared. A "mitigated design alternative" had been selected as the project description because of its ability to reduce and/or eliminate significant effects. ( Id. at p. 896, 50 Cal.Rptr.3d 799 .) But as the environmental review process progressed, the developer proposed to implement the mitigated design alternative but with change in phasing to avoid mining on lands affected by conservation laws. ( Id. at p. 894, 50 Cal.Rptr.3d 799 .) The final EIR said the project could avoid those conflicts by delaying mining on those lands *672 affected by conservation contracts. The final EIR did not, however, include a revised description reflecting the developer's decision to proceed with that suggestion and change its phasing to avoid the affected lands, nor did the EIR analyze whether the change in phasing created additional impacts. ( Ibid. ) The developer submitted a revised project application to implement the mitigated design alternative with the changes, which became known as the "revised mitigated design alternative." ( Id. at pp. 894-895, 50 Cal.Rptr.3d 799 .) The county then certified the EIR and approved the project, finding the revisions did not result in any additional impacts. ( Id. at pp. 895, 896, 50 Cal.Rptr.3d 799 .) The trial court determined the EIR violated CEQA by not identifying and addressing the new phasing. ( Id. at pp. 895-896, 50 Cal.Rptr.3d 799 .) We disagreed and held CEQA does not per se require a revised project description *578 be included in the final EIR itself, and substantial evidence in the record demonstrated the changed phasing was not significant new information requiring additional analysis in, or recirculation of, the final EIR. ( Ibid. )

Here, in contrast, the record does not contain evidence demonstrating the change from eradication to control was insignificant.

CDFA cites case law finding there was no error where the approved project was narrower than the EIR's project description. (E.g., Dusek v. Redevelopment Agency of City of Anaheim (1985) 173 Cal.App.3d 1029 , 1040, 219 Cal.Rptr. 346 [EIR described acquisition and demolition of multiple buildings but agency approved only the demolition of one building].)

However, we cannot tell whether the approved control program was narrower than the EIR's described eradication program, because the EIR expressly declined to evaluate a control program as an alternative to an eradication program. CDFA's findings stated the difference between eradication and control was in the intensity, in that control is less intensive than eradication. CDFA claims the change reduced the scope of the project: "The difference between control and eradication does not lie in the tools used but in the intensity of their use. Control targets a portion of the pest population, while eradication targets the entire pest population. Thus, the PEIR analyzed a more intense Program than CDFA is now proposing to implement. The environmental impacts of the reduced Program will be no greater than-and will almost certainly be less than-the environmental impacts of the more aggressive Program analyzed in the PEIR." (Orig. italics.) This was supposition on CDFA's part. CDFA does not direct our attention to anything in the record that would constitute substantial evidence supporting this supposition.

Instead, the record supports an opposing inference that a control program may be more harmful to the environment than an eradication, because the EIR acknowledged a control program would need to go on "forever."

*673 We conclude the EIR was fatally defective in failing to study a range of reasonable alternatives.

We continue on with appellants' other contentions as needed to comply with section 21005, subdivision (c), which requires us to address other contentions despite finding reversible error as to one contention.

V

The "No Program Alternative" (NPA )

A. Substantial Evidence

Appellants argue no substantial evidence supports two assumptions CDFA made about potential impacts of the "No-Program Alternative" (NPA) of doing nothing about LBAMs. NCRA claims CDFA merely assumed based on "flawed studies and unsubstantiated claims" that doing nothing would cause (1) a dramatic increase in private pesticide use and (2) crop damage and reduced agricultural revenues. We see no reason to reverse the judgment on this ground.

Under CEQA, "substantial evidence" is "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).) Thus, appellants' references to evidence they think support different conclusions is unavailing.

Substantial evidence includes facts, reasonable assumptions predicated upon facts, *579 and expert opinion supported by facts. (§ 21082.2, subd. (c); Guidelines, § 15384, subd. (b).) It does not include "[a]rgument, speculation, unsubstantiated opinion or narrative, evidence which is clearly inaccurate or erroneous, or evidence of social or economic impacts which do not contribute to, or are not caused by, physical impacts on the environment ...." (§ 21082.2, subd. (c); accord, § 21080, subd. (e)(2).)

1. Crop Damage

Appellants claim there was only one incident of crop damage, which "ultimately proved to be from a different insect." The DEIR acknowledged it used evidence from outside California due to the lack of any published studies of crop damage caused by LBAMs in California.

In light of our conclusion that the program cannot continue without further environmental review, we need not address sufficiency of evidence regarding crop damage, because there is doubtless new information on this point since the EIR was certified.

*674 2. Private Use

As to anticipated private pesticide use in the absence of CDFA action, the EIR separately addressed urban and agricultural uses. For private urban use, the EIR relied on studies performed in 2001-2003 by University of California scientists Cheryl Wilen and Mary Louise Flint, as analyzed by LBAM Program Director Robert Dowell in his April 2008 "Calculation of the Potential Increase in Urban Homeowner Pesticide Use Caused by a Generalized [LBAM] Infestation in Coastal California." For private agricultural use, the EIR relied on a University of California report (Johnson et al. 2007) and literature from the USDA, Australia, and New Zealand.

Appellants do not always make clear whether they are arguing about urban or agricultural use. They complain the studies showed that most private urban pesticide use targeted ants, and only three to seven percent targeted plant-pests. Appellants claims the existing pesticide use against plant-pests should have been considered part of the environmental baseline, but CDFA instead extrapolated it to project increased pesticide use. However, existing use was considered a baseline. Moreover, that increased pests will mean increased pesticide use is a reasonable assumption.

OCEF maintains no additional spraying would be needed for individuals who already spray for backyard pests, because any current pesticide would likely kill LBAM as well. However, OCEF does not direct our attention to anything in the multiple cited pages that supports this assertion.

OCEF's claims "[t]he point is that CDFA's assumed increase in private pesticide use would also occur under the Program because LBAM will continue to exist even under the Program. During the eradication program, increased private pesticide use would have occurred until LBAM eradication. Now that the goal is control, such increased private pesticide use will potentially occur indefinitely because LBAM will never be eradicated. Yet CDFA did not consider any increased private pesticide use when analyzing Program impacts, thereby artificially inflating pesticide impacts under the No Program scenario and artificially minimizing impacts under the Program scenario." (Orig. emphasis.) However, it is OCEF which is making an unsupported assumption that, because LBAM will continue to exist despite the Program, then the projected increase in private use will occur with or without the Program. OCEF cites no supporting *580 evidence. The projection of increased private use was the expectation in the absence of a program -i.e., if the government does not take action, private citizens will. *675 OCEF argues there was no evidence private pesticide use would increase, because a UC Study concluded, "[a]lthough LBAM attacks many types of plants, it is not likely to cause serious damage to them in backyard situations. In many cases, treatment would not be needed in backyards."

OCEF argues the Dowell Report "speculated" that private individuals would use the "highly toxic pesticides permethrin or chlorpyrifos," whereas it was not known what people were then using. However, the EIR merely listed chlorpyrifos among the registered products that consumers "may" use and stated it was using permethrin as a "representative material." Though OCEF cites evidence that use of permethrin had decreased from 2002 to 2007, the record contains reports of the U.S. Environmental Protection Agency that individuals used about two million pounds of permethrin annually, over 70 percent of which was used in non-agricultural settings, and 41 percent of which was applied by homeowners in residential areas. CDFA selected permethrin because it is a broad-spectrum pesticide that kills LBAM and is readily available to and commonly used by consumers, and CDFA planned to use it. It was reasonable for CDFA to use permethrin as representative in light of the evidence.

OCEF argues there was no evidence that farmers would increase pesticide use in the absence of a program, because there was no evidence the LBAM had caused any crop damage, and the only cited incident of damage was ultimately shown to be from a different insect. OCEF's appellate brief argues LBAM "is primarily located in two counties," so there is no evidence supporting a conclusion of increased pesticide use by farmers. The reference to two counties is perplexing, since LBAM had already spread to 13 counties during the review process.

In any event, we conclude appellants' claims of insufficiency of the evidence do not constitute a separate ground for reversal of the judgment.

B. Asserted Effect of Quarantines on No-Program Alternative (NPA)

In 2007, CDFA and USDA established quarantines that required "trapping, inspection, and certification of all nursery stock and host commodities" in eight counties. The quarantines later expanded to 13 counties and were to continue throughout the seven-year program.

OCEF argues the NPA was defective in failing to consider how the quarantines would affect LBAM spread under the no-program alternative. OCEF claims, *676 "Because CDFA and USDA's quarantines will prevent this spread, even if only to some extent, CDFA's conclusions about private pesticide use are wrong, and its No Program analysis is flawed." OCEF cites Planning and Conservation League v. Dept. of Water Resources (2000) 83 Cal.App.4th 892 , 100 Cal.Rptr.2d 173 , as holding an EIR proposing new water shortage protocols was deficient for failing to analyze the impact of existing water shortage protocols under the "no-project alternative." ( Id. at pp. 915-916, 100 Cal.Rptr.2d 173 .)

However, in this case the record shows LBAMs continued to spread throughout California during the environmental review process, despite the quarantines, increasing from 10,000 to more than 71,000 in two years.

*581 We conclude appellants fail to show reversible error regarding the No-Program Alternative.

VI

Asserted Failure to Analyze Impacts

NCRA argues the EIR failed adequately to analyze impacts on several points, while OCEF makes this argument as to one point (site-specific impacts). We see no need to reverse the judgments on these grounds, though we do not foreclose the possibility that updated evidence may alter the analysis in the event CDFA proceeds with further environmental review in order to continue its control program.

A. Twist-Tie Tool (MD-1 )

CDFA found using plastic twist-ties attached to aluminum wire and infused with LBAM pheromone on trees and shrubs would not result in significant impacts, because they would be used only to address "small isolated infestations," and they do not kill LBAMs but disrupt mating behavior thus diminishing the likelihood of successful reproduction. The pheromone degrades over time. Twist-ties are placed by hand and retrieved after three to six months. A search can be conducted if any are missing. A potential exists for ingestion by birds, but the twist-ties lack flavor and were not likely to be swallowed. The twist-ties are placed six to eight feet off the ground and therefore not likely to be ingested by animals. Toxicity studies of twist-ties under federal law (Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), 7 U.S.C. § 136a, subd. (c)(5) ) and under the California Department of Fish and Game Pesticide Investigations Unit indicated they were not harmful except at high doses, which were not expected in this case due to CDFA's commitment to remove the twist-ties.

*677 NCRA argues retrieving 250 twist-ties per acre in densely vegetated or populated areas is easier said than done, and new ones will be placed as old ones are retrieved, and the plastic and metal parts of the twist-ties will not degrade, and the twist-ties have a sweet odor, and the EIR does not explain why birds but not mammals might consume them. However, the EIR did explain mammals were unlikely to ingest twist-ties because they are not flavored and are placed off the ground. Appellants complain the EIR did not evaluate the twist-ties' inert ingredients, which were trade secrets unknown to the parties. However, CDFA notes the entire formulation of the product was tested in the pesticide product registration process. Moreover, none of appellants' arguments render the EIR defective. We do not pass upon the correctness of the EIR's environmental conclusions, but only upon its sufficiency as an informative document. ( Chaparral Greens, supra, 50 Cal.App.4th at p. 1145 , 58 Cal.Rptr.2d 152 .) CEQA requires an EIR to reflect a good faith effort at full disclosure; it does not mandate perfection, nor does it require an analysis to be exhaustive. ( Ibid. )

B. Ground Application Tool (MD-2)

NCRA argues the EIR failed adequately to address impacts of MD-2, application of chemical pheromones (Hercon Bio-Flake and SPLAT) to trees and shrubs in residential yards and telephone poles and trees on public property along roads.

NCRA complains the EIR said potential ingestion of significant quantities of Hercon Bio-Flake was extraordinarily small because a "very low number of flakes" would be applied per acre, "approximately 1 cup." NCRA complains the volume of these tiny flakes is a "meaningless statistic" that reveals nothing about the actual *582 number of flakes. NCRA fails to persuade us why this should matter.

NCRA argues the evidence does not support the conclusions about toxicity of Hercon Bio-Flake and SPLAT. However, we are hard pressed to find support for NCRA's position in the few pages of the record cited by NCRA, and the record contains much evidence, including toxicity assessments, not cited by NCRA. NCRA has thus forfeited its substantial evidence claim. ( State Water Resources Control Board Cases (2006) 136 Cal.App.4th 674 , 749-750, 39 Cal.Rptr.3d 189 [CEQA does not exempt appellants from ordinary rule that substantial evidence argument is forfeited by failing to cite all material evidence on the subject].) "[T]he burden to provide a fair summary of the evidence 'grows with the complexity of the record.' "

*678 Boeken v. Philip Morris, Inc. (2005) 127 Cal.App.4th 1640 , 1658, 26 Cal.Rptr.3d 638 .) That certainly applies here, where the administrative record exceeds 70,000 pages.

C. Aerial Spraying Tool (MD-3)

NCRA argues the EIR inadequately analyzed the MD-3 tool of aerial spraying of Hercon Bio-Flake or SPLAT. However, CDFA Findings stated it would not use this tool. We therefore need not consider the contentions on appeal. We disregard the argument in NCRA's reply brief, unsupported by any evidence or authority whatsoever, that maybe CDFA (having found the tool was not feasible "at this time") will change its mind and use the trial court's ruling upholding adequacy of the EIR's discussion of MD-3 as a bar to any future challenge.

D. Organically approved Insecticide Tools (Btk and S)

NCRA argues the EIR inadequately analyzed the impacts of insecticides Btk and S. on non-target species such as butterflies, bees, and birds. The EIR stated that impacts were considered less than significant due, in part, to the "short-term length of the Program." NCRA argues the last-minute change from eradication to control renders this conclusion defective now that CDFA has approved a program "of indefinite duration." However, CDFA approved only a seven-year program.

NCRA also argues no evidence supports the conclusion that "highly localized" spraying will avoid significant impacts to non-target species, and there was no evidence of how far the spray will drift. However, as CDFA points out, NCRA fails to cite evidence studying drift and has therefore forfeited the contention.

E. Sterile Insect Technique Tool (SIT)

SIT would produce and release millions of sterile moths. NCRA argues the EIR failed to analyze impacts of the reasonably foreseeable accidental release of fertile moths, development of asexually-reproducing LBAM, and resistance of LBAM to the radiation rendering them sterile. NCRA cites its own comment letter asserting a sterilization accident occurred in 2003, resulting in the release of fertile flies around a Mexico facility. However, the EIR explained quality control procedures in California were more effective than in Mexico.

*679 In its reply brief, NCRA says its comment letter also cited a report of an accidental release in California. However, NCRA forfeited this point by failing to include it in the opening brief. ( Garcia v. McCutchen (1997) 16 Cal.4th 469 , 482, fn. 10, 66 Cal.Rptr.2d 319 , 940 P.2d 906 [appellant may not raise new argument in reply brief].)

NCRA complains CDFA did not disclose the contents of the "emergency plan" it *583 stated it has for dealing with the remote possibility of accidental release during transport. The final EIR referred to "Chapter 6-Hazardous Materials." CDFA submitted the emergency plan to the trial court with a declaration stating it was inadvertently omitted from the administrative record. To the extent NCRA suggests it was not made available to the public, our reversal renders the point moot.

F. Site-Specific Impacts

Appellants argue the EIR is defective because it does not discuss site-specific impacts. We conclude appellants fail to show reversible error on this basis.

The EIR stated site-specific evaluation was beyond the scope of this programmatic evaluation. The EIR identified a number of likely areas that would be subject to program activities, stated it was impossible to determine in advance where all LBAM populations will occur, and promised that prior to deploying treatments CDFA would follow notification procedures to local elected officials, the agricultural commissioner, and affected residents and would hold informational open houses for residents. The final EIR noted the DEIR evaluated the significance of impacts of each tool in various types of land use and topography.

CEQA does not require identification of every possible treatment site. (Guidelines, § 15151 ["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 level of specificity is determined by the nature of the project and the rule of reason, not by any semantic label. ( Laurel Heights I, supra, 47 Cal.3d at p. 407 , 253 Cal.Rptr. 426 , 764 P.2d 278 .)

After briefing was completed in this case, we recently held in Center for Biological Diversity, supra, 234 Cal.App.4th 214 , 183 Cal.Rptr.3d 736 , that a fish hatchery and stocking enterprise of the Department of Fish and Wildlife was an appropriate project for a program EIR. The Center argued the EIR was inadequate because *680 it did not contain site-specific analysis for each water body to be stocked, and it deferred that analysis until the Department performed for each site an "evaluation protocol" described in the EIR. ( Id. at p. 237, 183 Cal.Rptr.3d 736 .) We found the EIR adequate. We distinguished our earlier opinion in Friends of Mammoth v. Town of Mammoth Lakes Redevelopment Agency (2000) 82 Cal.App.4th 511 , 98 Cal.Rptr.2d 334 ( Friends of Mammoth ), where we held inadequate a program EIR for a redevelopment plan containing detailed descriptions of 72 different projects, where the EIR did not review potential impacts from any of the 72 projects to the extent information was known or could reasonably be known. ( Id. at p. 535, 98 Cal.Rptr.2d 334 .) Unlike Friends of Mammoth , the EIR in Center for Biological Diversity analyzed every impact that reasonably could occur by stocking fish in any water body in the state based on the information currently known. ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 237, 183 Cal.Rptr.3d 736 .) Due to the nature of the project, the EIR explained what the impacts would likely be to "decision species" (threatened amphibian or other species that may be harmed by stocking the water body with fish) at any site where the Department operated hatcheries and stocked fish. ( Id. at pp. 227, 237, 183 Cal.Rptr.3d 736 .) "Site-specific analysis will likely not reveal any unanticipated impacts; instead, it will reveal whether the impacts discussed in the EIR are occurring at that site." ( Id. at p. 237, 183 Cal.Rptr.3d 736 .) *584 The Center complained that if site-specific analysis were not done during the EIR process, it may never be done or done in a way allowing public input. ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 238, 183 Cal.Rptr.3d 736 .) We said CEQA does not require additional site-specific environmental review if the agency determines the site-specific impacts were sufficiently addressed in the program EIR, nor does it require that determination to be made in a public process. ( Ibid. ) Only if the agency discovers new impacts will they be addressed in a public process. ( Ibid. ) "When a program EIR is used to avoid preparing subsequent EIR's, such as here, the public agency must examine site-specific program activities 'in the light of the program EIR to determine whether an additional environmental document must be prepared.' (CEQA Guidelines, § 15168, subd. (c).) If the site-specific activity will not create effects or require mitigation measures that were not discussed in the program EIR, the public agency is not required to prepare any other site-specific environmental document. (CEQA Guidelines, § 15168, subd. (c)(2).)" ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 238, 183 Cal.Rptr.3d 736 .)

"In effect, after a sufficiently comprehensive and specific program EIR has been certified, CEQA allows much of the initial site-specific review to occur outside a formal CEQA process and beyond public view. CEQA does not require the Department to engage in a public process when it determines *681 whether the impacts from a site-specific project were addressed and adequately mitigated in the program EIR. And if the Department finds the impacts were addressed, it need not prepare a new environmental document at all." ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 239, 183 Cal.Rptr.3d 736 .) To require a project-specific EIR for all sites proposed after certification of the program EIR, even where the subsequent activity is within the scope of the project described in the program EIR, would contravene one of the essential purposes of program EIRs, i.e., " 'to streamline environmental review of projects within the scope of a previously completed program EIR. We conclude that a program EIR may serve as the EIR for a subsequently proposed project to the extent it contemplates and adequately analyzes the potential environmental impacts of the project....' [Citation.]" ( Ibid. )

Before the Department would stock a water body, it would utilize the "evaluation protocol"-a checklist to document site-specific impacts and determine whether they were sufficiently analyzed in the program EIR-to determine if any "decision species" were present. If they were present, the agency would determine whether stocking would have a substantial environmental effect, a review that "will by necessity include application of the impacts analysis contained in the EIR, as well as a determination of any other impacts that may not have been addressed in the EIR. This is exactly the type of process CEQA requires an agency to utilize outside of public review when it intends to approve a site-specific project that is part of a program previously reviewed in a program EIR. If the Department upon using the evaluation protocol discovers an impact that was not sufficiently addressed in the EIR, it will then be obligated to begin a CEQA process, but only if the Department intends to approve the activity." ( Center for Biological Diversity, supra, 234 Cal.App.4th at p. 239, 183 Cal.Rptr.3d 736 .)

We conclude appellants fail to show reversible error in the EIR's impacts analysis.

*585 VII

Cumulative Impacts

NCRA argues the EIR's cumulative impacts analysis violated CEQA by failing to address (1) the synergistic effects of the project, and (2) individually minor but collectively significant cumulative impacts. OCEF argues the cumulative impacts analysis violated CEQA by (1) failing to use one of two methods (the list method or the summary-of-projections method) in Guidelines, § 15130, subd. (b), and (2) failing to discuss ongoing programs targeting other pests.

*682 We need not address these contentions.

As indicated, section 21005, subdivision (c), provides that any court that finds a CEQA violation "shall specifically address each of the alleged grounds for noncompliance." However, section 21005 does not require us to address additional alleged defects that may be addressed in a completely different and more comprehensive manner upon further CEQA review following remand. ( Communities for a Better Environment, supra, 184 Cal.App.4th at pp. 101-102, 108 Cal.Rptr.3d 478 .)

Our finding of reversible error (in the EIR's failure to analyze a control program as an alternative to an eradication program) necessarily requires a new cumulative impacts discussion in the event CDFA undertakes the further environmental review needed to proceed with the program.

Thus, the cumulative impacts discussion in an EIR must discuss incremental impacts of the project when added to other, closely related past, present, and reasonably foreseeable probable future projects. (§ 21083, subd. (b)(2).) "An EIR shall discuss cumulative impacts of a project when the project's incremental effect is cumulatively considerable, as defined in section 15065(a)(3) [cumulatively considerable means 'the incremental effects of an individual project are significant when viewed in connection with the effects of past projects, the effects of other current projects, and the effects of probable future projects ']. Where a lead agency is examining a project with an incremental effect that is not 'cumulatively considerable,' a lead agency need not consider that effect significant, but shall briefly describe its basis for concluding that the incremental effect is not cumulatively considerable...." (Guidelines, § 15130, subd. (a), italics added.) " 'Cumulative impacts' refer to two or more individual effects which, when considered together, are considerable or which compound or increase other environmental impacts. [¶] (a) The individual effects may be changes resulting from a single project or a number of separate projects. [¶] (b) The cumulative impact from several projects is the change in the environment which results from the incremental impact of the project when added to other closely related past, present, and reasonably foreseeable probable future projects. Cumulative impacts can result from individually minor but collectively significant projects taking place over a period of time." (Guidelines, § 15355.)

Here, any revised cumulative impacts discussion will have to include the reasonably foreseeable need to continue anti-LBAM activities after expiration of the seven-year period and into the foreseeable future, as acknowledged in *683 the EIR ("if an exotic pest becomes permanently established in California, control measures will be needed forever").

Moreover, CDFA's own argument in effect acknowledges the current cumulative impacts discussion for a control program will be different from the discussion for the eradication program, because the EIR and CDFA's findings stated a control program and an eradication differ in "intensity."

*586 While CDFA claims a control program is less intense, appellants disagree, and the manner in which CDFA has conducted the environmental review leaves the record bereft of evidence to resolve the dispute.

On appeal, CDFA does not try to argue that the program could continue without incremental impacts. Instead, CDFA argues it is speculative whether the program will have to continue after the seven-year period. This argument is disingenuous, given the EIR's statements that an eradication program was chosen specifically to avoid the unending nature of a control program that would have to go on forever. CDFA also argues the exact details of a continued program would be too speculative to study. However, the exact details need not be known with certainty; what matters is what is reasonably foreseeable.

VIII

Summary

We conclude the EIR violated CEQA by failing to analyze a control program as an alternative to an eradication program, with the consequence that the EIR dismissively rejected anything that would not achieve full eradication. The error was prejudicial, requiring reversal of the judgments. CDFA's last-minute change from an eradication program to a control program did not cure the prejudice, because the EIR dismissively rejected control feature that would not achieve eradication, and the EIR's cumulative impacts discussion did not address the reasonably foreseeable need to continue pest control efforts after expiration of the seven-year period.

DISPOSITIONS

The judgment against the NCRA appellants is reversed and the matter remanded to the superior court with directions to enter a new judgment granting the writ petition. Appellants shall recover costs on appeal. (Cal. Rules of Court, rule 8.278.)

*684 The judgment against the OCEF appellants is reversed and the matter remanded to the superior court with directions to enter a new judgment granting the writ petition. Appellants shall recover costs on appeal. (Cal. Rules of Court, rule 8.278.)

We concur:

RAYE, P.J.

BLEASE, J.