ORDER
KARLTON, Senior District Judge.Pending before the court are cross-motions for summary judgment filed by plaintiffs and state defendants.1 The suit involves two claims made by plaintiffs effectively brought against several state agencies, under the Clean Air Act (“The Act” or “CAA”), 42 U.S.C. §§ 7401, et seq.2 *916In their first cause of action, plaintiffs, unincorporated associations and non-profit organizations representing members who live and work in affected areas,3 allege that defendants failed to adopt and implement regulations by June 15, 1997 as required by the California State Implementation Plan (“SIP”). In their second claim, plaintiffs allege that defendants improperly calculated the 1990 baseline emission inventory in violation of the SIP.4 Plaintiffs and state defendants move for summary judgment on both claims.
I resolve the matter based on the pleadings,. the parties’ papers, and after oral argument. Remarkably, I conclude that although plaintiffs’ motion for summary judgment as to the second cause of action must be denied, by virtue of the fact that the defendants calculated the baseline in a manner inconsistent with the SIP, the plaintiffs’ motion for summary judgment on the first cause of action must be granted.5
I.
THE STATUTE AND REGULATORY PROCESS
In its previous order, the court outlined the statutory and regulatory framework of the CAA. Nonetheless, the framework is indispensable for resolution of the instant motion, and thus the court will repeat itself.
The CAA requires the U.S. Environmental Protection Agency (EPA) to promulgate health-based standards for certain pollutants, including hydrocarbons and nitrogen oxides which produce ground level ozone. These standards are called the National Ambient Air Quality Standards (NAAQS). 42 U.S.C. § 7409(a), (b). Each state is required under the CAA to adopt a State Implementation Plan (SIP) to satisfy the NAAQS requirements. 42 U.S.C. § 7410(a)(1). Specifically, each state is mandated under § 110(a) of the Act, 42 U.S.C. § 7410(a), to adopt a “plan which provides for implementation, maintenance, and enforcement” of the National Ambient *917Air Quality Standards and to submit its SIP to the EPA for approval.
A SIP must be approved by the EPA to insure that it meets the criteria specified in § 7410. A SIP, “once adopted by a state and approved by the EPA, becomes controlling and must be carried out by the state.” See Bayview Hunters Point Community Advocates v. Metropolitan Transportation Commission, 366 F.3d 692, 695 (9th Cir.2004) (citing Friends of the Earth v. Carey, 535 F.2d 165, 169 (2nd Cir.1976), cert. denied, 434 U.S. 902, 98 S.Ct. 296, 54 L.Ed.2d 188 (1977)). Approved SIPs are enforceable by either the State, the EPA, or via citizen suits brought under Section 304(a) of the CAA. Id. (citing Baughman v. Bradford Coal Co., 592 F.2d 215, 217 n. 1 (3d Cir.1979)); 42 U.S.C. § 7604(a).
The EPA designates areas of states as “attainment” or “nonattainment” based on whether they meet the NAAQS for a particular pollutant, such as ozone, or whether each areas “attains” the NAAQS. 42 U.S.C. § 7407(d). Ozone nonattainment areas are further classified as Marginal, Moderate, Serious, Severe or Extreme, depending on the severity of the ozone pollution problem.6 42 U.S.C. § 7511(a).
A. ESTABLISHING THE 1990 BASE YEAR INVENTORY AND TRACKING
One requirement for all nonattainment areas is the establishment of an emissions inventory. The CAA requires that each state submit a comprehensive inventory of all sources of the relevant pollutant in the nonattainment area. 42 U.S.C. § 7502(c)(3). The inventory sets out the classes and categories of emission sources so control strategies can be used to address emissions from those sources.
In the case of ozone, which is the product of a chemical reaction, the inventory focuses on those sources that emit the ozone-forming pollutants. Nitrogen oxides is such a precursor. VOCs are other ozone precursors. VOCs are found in many products, including pesticides. Because ozone is not a directly-emitted pollutant, the inventory for ozone requires the cataloguing of the total amount of actual VOC and NOx emissions. In the case of the pesticide element of the Ozone SIP, because volatile organic compounds (VOCs) react with oxides of nitrogen (Nox) in the presence of heat and sunlight to form ozone, the state must monitor the VOCs emitted by each pesticidal product to determine the inventory, or the catalog of emission sources.
1. Enforceable Control Measures
Each SIP must include enforceable emission limitations and other control measures necessary to attain the NAAQS, as well as timetables for compliance. 42 U.S.C. § 7410(a)(2)(A) (a SIP must include “enforceable emission limitations, and such other control measures, means or techniques ... as may be necessary or appropriate to meet the applicable requirements of [the Act]”); 42 U.S.C. § 7502(c)(6) (a SIP [for nonattainment areas] shall include enforceable emission limitations and control measures “as may be necessary or appropriate to provide for attainment” of the NAAQS by the applicable attainment date.). These control measures are “strategies” that the court may enforce. Bayview Hunters Point Community Advo*918cates v. Metro. Transportation Comm’n, 366 F.3d 692, 701 (9th Cir.2004).
II.
FACTS7
Pesticides, i.e., chemical products used to treat agricultural land, agricultural crops, and structures for protection from insects, fungus, rodents, and other pests emit VOC, which reacts with oxides of nitrogen to form ozone. Pis.’ SUF 2. Ozone pollution damages lung tissue, resulting, inter alia, in reduced lung capacity. It also damages crops and vegetation, exacerbates asthma, increases respiratory and cardiovascular hospital admissions, and increases school and work absenteeism. Pis.’SUF 3.
The Sacramento Metropolitan, San Joaquin Valley, South Coast, Southeast Desert, and Ventura areas are designated non-attainment under the federal one-hour and eight-hour ozone standards. Pis.’ SUF 4.
A. DEVELOPMENT OF THE PESTICIDE ELEMENT OF THE 1994 OZONE SIP
On May 5,1994, EPA proposed a Federal Implementation Plan (“FIP”) containing measures, including pesticide controls, intended to bring the South Coast, Sacramento and Ventura areas into compliance with the one-hour ozone NAAQs standard. Pis.’ SUF 5. Under U.S. EPA Region 9 Air division Director David Howekamp’s (“Howekamp”) direction, and parallel with the development of the FIP, EPA worked with California regulatory agencies to develop a legally sufficient ozone SIP.8 Pis.’ SUF 6.
EPA’s goal was to approve a SIP that would withstand legal challenge and allow California state and local regulatory agencies to replace the FIP and regain control of ozone regulation.9 Pis.’ SUF 7. On November 15, 1994, the California Air Resources Board (“ARB” or “CARB”) adopted and submitted the 1994 Ozone State Implementation Plan (“SIP”). The SIP contains the Pesticide Element, as adopted by the Department of Pesticide Regulation (DPR). Pis.’ SUF 8; Defs.’ SUF 1.
B. THE PESTICIDE ELEMENT
The Pesticide Element, as submitted to EPA on November 15, 1994, stated that *919the “plan is designed to reduce volatile organic compound (VOC) emissions from agricultural and commercial structural pesticide applications by a maximum of 20 percent from the 1990 baseline emission inventory to the year 2005.” Pis.’ SUF 9. The Pesticide Element declared that “[a] decision whether additional regulatory measures to ensure that reductions in pes-ticidal VOC emissions are achieved will .be made by 1997.” Pis.’ SUF 10 (citing 1994 Pesticide Plan (“Plan”) attached as Exhibit 1 to Newell Dec. at 10).10
The Pesticide Element relies on an inventory of pesticide VOC emissions to “provide the basis to determine what additional regulatory measures are needed to achieve targeted pesticidal VOC reductions.” Pis.’ SUF 11. The initial inventory—the so-called “1990 baseline inventory”—and all subsequent inventories were generated by multiplying the amount of pesticides used by the VOC content of those pesticides. Pis.’ SUF 12. Pesticide Use Report (“PUR”) data provides the amount of pesticides used while VOC emission factor (or “emission potential”) estimates pesticides’ VOC content. Dr. Randy Segawa, Senior Environmental Research Scientist at DPR and the supervisor for DPR’s VOC emissions program, explained how DPR calculates the inventory:
The calculation of our VOC inventory is based on two pieces of information of which pesticide use report data is one. The pesticide use reports provide the information regarding the amount of pesticides applied, the number of acres treated, the crop treated, the location and things regarding the application of a particular pesticide product. We take the amount reported in the pesticide use reports and multiply it by the emission potential to calculate the VOC emission for each pesticide application reported in the pesticide use reports.
Pis.’ SUF 14.
To create the VOC emission factor, DPR either solicited emission potential data from pesticide registrants (manufacturers) or, in the absence of that data, applied the highest emission potential as a default.11 Pis.’ SUF 15. If the registrant did not submit the emission potential data, the product was to “be given the highest VOC emission factor for the pesticide product.” Pis.’ SUF 17.
The Pesticide Element also states that “[i]n cooperation with DPR, the California Air Resources Board (ARB) will develop a baseline inventory of estimated 1990 pesti-cidal VOC emissions based on 1991 pesticide use data, adjusted to represent the 1990 base year, and on the VOC emission potential data.”12 Pis.’ SUF 19. It expli*920cates that “[t]he base year inventory will be created from the 1991 Pesticide Use Report and then adjusted by a factor to represent the 1990 base year.”13 The Pesticide Element also explained that “[f]ull pesticide use reporting began in 1990” and that “[i]t is believed that the 1991 pesticide use report would be a more accurate source to determine the 1990 pesticidal VOC emissions.” Pis.’ SUF 20.
DPR Deputy Director Paul Gosselin (“Gosselin”) testified as to the difference between using 1990 and 1991 Pesticide Use Reports:
Before we had put this plan together we had conducted a number of workshops on different aspects of the methodology knowing that the beginning part of this process had to begin with 1990. It just so happened that that coincided with the beginning of the Full Use Report Program which began in 1990.
There was a lot of concern from all parties that the first year of the collection of pesticide use data was probably not the most accurate in terms of compliance, in terms of accuracy for any sort of huge program getting undertaken. So what was decided was there was a lot more comfort from all parties that the 1991 pesticide use report was far more accurate, far more complete than 1990, and that was probably the better year to start this program with and use that as sort of a starting point to determine looking back as to what the starting point in 1990 should be.
Pis.’ SUF 21.
B. HOWEKAMP’S DECLARATION
As "Director of the Air Division at EPA Region 9, Howekamp oversaw the implementation of the Clean Air Act in California, including the approval of the 1994 Ozone SIP. Pis.’ SUF 22. In order for EPA to approve the SIP and relieve itself of the obligation to supervise the FIP, EPA had to ensure that, as the CAA required, all measures in the SIP were legally enforceable (including enforcement by citizen law suit) and that all needed emission reductions would be achieved by the attainment deadlines established in the Act.
On March 20, 1995, after reviewing the Pesticide Element as submitted to EPA, Howekamp requested a “technical clarification” from DPR Director James Wells (“Wells”) on (1) the calculation of the 1990 baseline inventory for pesticide-related VOC emissions; (2) the percentage reduction targets (as specified in a table in the letter); (3) the deadline to adopt necessary regulations; and (4) why VOC emissions from adjuvants were not regulated by the pesticide component. Pis.’ SUF 24.
On March 31, 1995, Wells responded to Howekamp’s March 20 letter. Pis.’ SUF 25. Even after this clarification, EPA could not approve the pesticide component because it was inconsistent with the CAA. Pis.’ SUF 26. In order to ensure compliance with the CAA, Howekamp, in a letter dated April 21, 1995, requested that Wells explicitly state that California commits to adopt and submit regulations by June 15, 1997 (or earlier). Those regulations, he explained, must limit VOC emissions from pesticides to specific percentages of the 1990 base year emissions by specific years and in specific nonattainment areas (described in 40 C.F.R. Pt. 81.305) as listed in the following table:
*921Reductions from 1990 Baseline
Ozone Nonattainment Area 1996 1999 2002 2005
Sacramento Metro_8% 12% 16% 20%
San Joaquin Valley_8% 12% 16% 20%
South Coast_8% 12% 16% 20%
Southeast Desert_8% 12% 16% 20%
Ventura_8% 12% 16% 20%
Exhibit 3 to Howekamp Dec. at 0223.
Howekamp specifically noted that “[i]n order to be consistent with Sections 110(k)(2) and 110(k)(4) of the CAA, such commitments should not extend beyond June 1997.” Accordingly, he wrote that “[t]he primary issue regarding the pesticide component is the date by which California has committed to adopt and submit regulations should they be needed to reduce VOC emissions.” Defs.’ SUF 2. Put directly, EPA’s goal to replace the FIP with a legally sufficient regulation was dependent on such an enforceable commitment to ensure that the needed emission reductions would be achieved on or before the attainment dates prescribed by the Act.
1. Clarifícation of the SIP
CARB Executive Officer James Boyd submitted DPR’s clarification of the Pesticide Element in a letter dated May 11, 1995. His letter reads in part:
I am writing to transmit a clarification by the California Department of Pesticide (DPR) of the pesticide element of the California State Implementation Plan (SIP) for Ozone, submitted to the U.S. Environmental Protection Agency (U.S.EPA) on November 15, 1994. The enclosed letter from DPR clarifies that in the SIP, California has committed to adopt and submit to U.S. EPA by June 15, 1997 any regulations necessary to achieve the emission reductions from pesticides specified in the letter.
Pis.’ SUF 31.
The “enclosed letter from DPR” adverted to by Boyd was actually a May 9, 1995 memorandum from Wells of DPR to Boyd (“Wells memo”). The Wells memo provided that:
The Department of Pesticide Regulation commits to adopt and submit to U.S. EPA by June 15, 1997, any regulations necessary to reduce volatile organic compound emissions from agricultural and commercial structural pesticides by specific percentages of the 1990 base year emissions, by specific years, and in specific nonattainment areas (described in 40 C.F.R. 81.305), as listed in the following table:
Reductions from 1990 Baseline
Ozone Nonattainment Area 1996 1999 2002 2005
Sacramento Metro 8% 12% 16% 20%
San Joaquin Valley 8% 12% 16% 20%
South Coast 8% 12% 16% 20%
Southeast Desert 8% 12% 16% 20%
Ventura 8% 12% 16% 20%
Pis.’ SUF 32.
The Wells memo satisfied Howekamp that the pesticide component of the 1994 Ozone SIP now included an enforceable commitment to adopt and submit any regulations needed to reduce VOC emissions by the applicable attainment dates in the Act. According to Howekamp, the May 11, 1995 letter from Boyd and the May 1995 memo were considered by EPA to be part of California’s SIP when EPA proposed approval of the measure on March 18, 1996.14 Pis.’ SUF 33. The Notice of Proposed Rulemaking also noted that “California’s pesticide commitment is described in *922a letter from DPR to CARB [James Well (DPR) to James Boyd (CARB), dated November 15,1994].” Defs.’ SUF 7.
C.LETTER FROM JAMES BOYD (CARB) TO DAVID HOWEKAMP (U.S.EPA) (“The Howekamp Letter”)
On June 13, 1996, Boyd of CARB submitted a letter to Howekamp of the U.S. EPA. It requested that EPA delete the table labeled “Reductions from 1990 Pesticide Emissions Baselines” and explained that the pesticide “commitment is for a 20% reduction from 1990 levels by 2005 in each SIP area, except [San Diego].” Pis.’ SUF 35. The letter contained an Attachment A which was asserted to be a detailed table of corrections and an Attachment C, a summary spreadsheet identifying the reductions that the State committed to achieve. Defs.’ SUF 10.
As Howekamp understood the June 13, 1996 letter from CARB, California requested that the informational tables included in the notice of proposed rule-making be revised to show that emission reductions from the pesticides measure are used in the SIP only for the overall demonstration of attainment, not for reasonable further progress. Pis.’ SUF 81. Thus, in his declaration, Howekamp avers that although, in compliance with the State’s request, the Final Rule omitted the tables that set forth specific percentage reductions, “the Final Rule made no change” to the treatment of the May 11 letter and May'9, 1995 memorandum as “part of California’s SIP when EPA proposed approval of the measure.” U.S. EPA approved the pesticide measure on January 8, 1997.
D. CALIFORNIA’S FOIA REQUEST FROM U.S. EPA
On June 13, 2005, the California Attorney General’s office asked U.S. EPA Region 9 to provide it with “copies of all documents held in EPA Region 9’s records pursuant to part 52.02 as incorporated by reference that create, clarify, revise or amend the California Department of Pesticide Regulation’s federally enforceable SIP obligations.” Defs.’ SUF 12. On July 19, 2005, the state defendants agreed to convert the request to a Freedom of Information Act request. Defs.’ SUF 13. On July 25, 205, U.S. EPA Region 9 responded to the FOIA request by providing copies of the Pesticide Plan Exhibit A and the 1996 Boyd-Howekamp letter. Defs.’ SUF 14.15
E. 1997 DECISION THAT REGULATIONS WERE NOT NECESSARY
On February 18, 1998, Wells and John Dunlap responded to the Environmental Defense Center’s Notice of Intent to File Suit Under Section 304 of the Clean Air Act. (“Chytilo letter”). The response asserted that the agencies had decided in April 1997 that, based on an “April 1997 analysis of data through 1995,” regulations were not necessary. The letter explained that “[o]ur April 1997 analysis of data through 1995 showed that all nonatttainment areas (NAAs) were meeting annual interim goals, and that four out of five NAAs were already meeting the final 2005 year goals.” Pis.’ SUF 48. A graph of the San Joaquin Valley inventory 1990-1995 (Bates stamped 0084) in the 1998 workshop package “reflects] the data known to the Department at the time of the analysis referred to in the Chytilo letter.” Pis.’ SUF 49.
*923According to Dr. Randall T. Segawa, a Senior Environmental Research Scientist with DPR, the first estimates of pesticide VOC emissions for the years 1990 through 1995, published in February 1998, showed that the South Coast, Ventura, Sacramento Metro and San Joaquin Valley nonattainment areas had achieved the 20% pesticide VOC emission reductions specified in the SIP element approved by the U.S. EPA. Defs.’ SUF 19.16 It is undisputed that DPR did not submit any regulations to the U.S. EPA by June 15,1997.
1. 1990 Baseline Calculation
Contained in the record is a memorandum dated February 13,1997 addressed to Paul Gosselin from Mark Pepple which explained that there is another option for calculating the 1990 base year.17 Pis.’ SUF 37. The Pepple memo states:
In the SIP, we state that the 1990 figure will be based on a backcast of 1991 data. It is uncertain when the ARB will provide the backcast figure. In the meantime, there is another option (called the PUR option) for calculating the 1990 baseyear. This option involves multiplying the emission factor for each pesticide product by the use of that product. We considered this option because the agricultural use figures determined by the ARB Methodology appeared to be low compared to successive years (which were all based on PUR data) and the commercial structural use figures seemed to be too high.
Pis.’ SUF 39.
The Pepple Memo mentions “a draft workshop package that includes an inventory for the years 1990-1995,” with the figure for the 1990 baseline inventory estimated by CARB “using their current methodology.” Pis.’ SUF 40. The Pepple Memo stated that “[o]verall, the PUR option increased the 1990 base inventory by 31%” [as compared to the baseline calculated by CARB] and “appear to confirm our suspicions about the ARB methodology.” Pis.’ SUF 41. The Pepple Memo included inventory graphs with a baseline calculated from the “PUR option.” Pis.’ SUF 42.
Pepple also attached graphs of the lower baseline calculated by CARB for comparison. Pis.’ SUF 43. On February 3, 1998, DPR released a public workshop package to consider pesticide VOC regulatory options (“1998 workshop package”). Pis.’ SUF 44. The 1998 workshop package states that the 1990 baseline inventory was “calculated by multiplying the VOC Emission Factor value for each product by the use of that product in 1990.” Pis.’ SUF 45. The 1998 workshop package also states that:
Table 1 lists estimate pesticide VOC emissions from agricultural and commercial structural use pesticides for the years 1990-1995. The VOC emissions for the 1990 based year in this table were calculated using 1990 Pesticide Use Report data.
Pis.’ SUF 46.
Defendants admit.that DPR used 1990 Pesticide Use Report data instead of back-casting 1991 PUR data to calculate the 1990 baseline inventory. Pis.’ SUF 47.
2. Revision of Default Emission Potentials
“Beginning in 2001, DPR made a change in its methodology for calculating VOC *924emissions that had a significant impact on the total inventory calculated.” Pis.’ SUF 51. DPR staff formally explained the inventory revision in a 2002 memorandum. Pis.’ SUF 52. DPR revised the “default” values for pesticide products so that when a pesticide registrant does not provide VOC emission data, then DPR would assign the median emission potential as a default instead of the maximum emission potential as the default value. Pis.’ SUF 53. The default emission potential revision caused DPR’s initial 1990 baseline inventory to be higher than the 1990 baseline inventory after DPR revised the default values. Pis.’ SUF 54.
The change affected the inventory for every year, but affected the 1990 baseline inventory the most. “While the maximum-value defaults caused upward bias in all years of the inventory, the effect is particularly severe for the early inventory years—and especially, the base year 1990—when the fraction of applied products with maximum-value defaults EPs was the greatest.” Pis.’ SUF 55 (citing Memorandum from Frank Spurlock to Randy Segawa regarding Analysis of the Historical and Revised Base Year 1990 Volatile Organic Compound Emission Inventories, dated December 16, 2002 at Bates stamp 100374).
3. Pesticide VOC Emission Inventories for 2000-2003
By using baseline and subsequent emissions inventories calculated with the methodology used by DPR beginning in 2001 to calculate the 1990-2000 inventory, defendants (a) admit that 8%, 12%, and 16% reductions from pesticide VOC reductions in 1996, 1999, and 2002, respectively, were not attained in the San Joaquin Valley, Southeast Desert, and Ventura NAAs; and (b) with respect to the Sacramento NAA, admit that the 8% and 12% reductions were not attained in 1996 and 1999 respectively, but assert that by 2002, pesticide VOC emissions had met the 2005 attainment year target. Pis.’ SUF 56.
The 2000 Inventory was the first inventory to use the hew default emission potential data. Pis.’ SUF 57. The 2000 inventory describes the San Joaquin Valley’s overall reduction goal as 16.6 million pounds by 2005. Pis.’ SUF 58. After reviewing the 2000 inventory, DPR decided that “no regulatory measures were needed at this time.” Pis.’ SUF 62. The 2001 inventory describes the San Joaquin Valley’s overall reduction goal as 16.6 million pounds by 2005. Pis.’ SUF 63. In 2001, DPR made a major change in the way it calculated the 1990 baseline emissions inventory and each year’s inventory. Defs.’ SUF 22. After DPR changed the way it calculated pesticide VOC emissions in 2001, the inventories showed that the Ventura and Southeast Desert nonattainment areas were not meeting their attainment year goals and the San Joaquin Valley nonattainment area was not meeting its 1999 attainment year goal. Defs.’ SUF 23. After reviewing the 2001 inventory, DPR decided that “we did not need to put regulations in place.” Pis.’ SUF 66. Plaintiffs filed this citizen suit on May 4, 2004. Pis.’ SUF 67.
The 2002 inventory, dated May 17, 2004, states that the San Joaquin Valley has a 12% reduction goal by 1999, the Valley’s attainment year when EPA approved the SIP, while the other four nonattainment areas have a 20% reduction goal in their respective attainment years (either 2005, 2007, or 2010). Pis.’ SUF 68. The 2002 inventory shows that the San Joaquin Valley Nonattainment Area increased VOC emissions by 34% over the 2001 level to approximately the 1990 baseline, while the Southeast Desert Nonattainment Area posted emissions at approximately the *925same level as the 1990 baseline. Pis.’ SUF 69. The 2002 inventory shows that the San Joaquin Valley Nonattainment Area increased VOC emissions by 34% over the 2001 level to approximately the 1990 baseline, while the Southeast Desert Nonattainment Area posted emissions at approximately the same level as the 1990 baseline. Pis.’ SUF 70.
The 2003 inventory—the most recent dated April 5, 2005—states that the San Joaquin Valley has a 12% reduction goal by 1999, the Valley’s attainment year when EPA approved the SIP, while the other four nonattainment areas have a 20% reduction goal in their respective attainment years (either 2005, 2007, or 2010). Pis.’ SUF 72. The 2003 inventory shows that the San Joaquin Valley, Southeast Desert, and Ventura all exceed the 1990 baseline, while the Sacramento and South Coast nonattainment areas appeared to meet the reduction goals. Pis.’ SUF 73. DPR at some point started to pursue the option of reformulation of non-fumigant products to reduce VOC content.18 Pis.’ SUF 74.
The 2003 inventory shows that the Ven-tura Nonattainment Area had VOC levels nearly double the 1990 baseline, approximately 93% of which are from fumigants. Pis.’ SUF 76. The 2003 inventory shows that the San Joaquin Valley and Southeast Desert nonattainment areas’ VOC levels exceeded the 1990 baseline. Pis.’ SUF 77.
The exhaustively detailed exposition noted above was necessary to explain the unexpected result adverted to earlier. I turn first to disposition of the second cause of action since, despite the order of plaintiffs’ complaint, its disposition is necessary to resolve the first claim.
III.
ANALYSIS
Plaintiffs bring suit to enforce two asserted commitments contained in the Pesticide Element of the 1994 Ozone SIP. The first cause of action alleges a commitment to adopt regulations as necessary to ensure that the Pesticide Element would meet the 20% reduction by 2005 goal, as well as interim reduction goals in 1996, 1999, and 2002. The second cause of action asserts a commitment to calculate the 1990 baseline inventory of pesticide VOC emissions.
As explained below, the court must grant plaintiffs’ motion for summary judgment as to the first cause of action and grant summary judgment to defendants as to the second cause of action.
*926A. PLAINTIFFS’ SECOND CLAIM: ALLEGED VIOLATION OF THE SIP BY IMPROPERLY CALCULATING THE 1990 BASELINE INVENTORY
Plaintiffs argue that defendants have violated the SIP by improperly calculating the 1990 baseline inventory.19 They argue that the SIP requires CARB and DPR to use 1991 Pesticide Use Report data (PUR) to calculate the 1990 baseline inventory. Defendants, on the other hand, argue that they have not violated the SIP, and in any event, a baseline inventory does not constitute an enforceable control strategy under the CAA.
It is undisputed that DPR used 1990 Pesticide Use Report data instead of back-casting 1991 PUR data to calculate the 1990 baseline inventory. Pis.’ SUF 47.20 One of the many questions is whether this use violates an enforceable commitment.
1. SIP Language
A “regulation should be construed to give effect to the natural and plain meaning of its words,” Bayview Hunters Point Community Advocates, 366 F.3d at 695. While, as I have pointed out in the earlier order, it is sometimes difficult to know what is in the regulation, much less identify its plain words, the instant issue does not pose that problem. The Pesticide Element provides that:
In cooperation with DPR, the California Air Resources Board (ARB) will develop a baseline inventory of estimated 1990 pesticidal VOC emissions based on 1991 pesticide use data, adjusted to represent the 1990 base year, and on the VOC emission potential data.
Pis.’ SUF 19 (citing SIP at 4).
It also states that “[t]his baseline inventory may be adjusted if empirical data are developed to determine the impact of temperature, treated substrate (foliate, soil, water, etc.), application technique, and other conditions on VOC emissions.” Pis.’ SUF 19-20.
Elsewhere, the SIP explains:
“[t]he base year inventory will be created from the 1991 Pesticide Use Report and then adjusted by a factor to represent the 1990 base year.” The Pesticide element also explained that “[f]ull pesticide use reporting began in 1990” and that “[i]t is believed that the 1991 pesticide use report would be a more accurate source to determine the 1990 pesti-cidal VOC emissions.”
Pis.’ SUF 20.
In sum, the SIP commits defendants’ use of a baseline inventory of estimated 1990 pesticidal VOC emissions based on 1991 pesticide use data. The SIP utilizes mandatory language, ARB “will develop a baseline inventory of estimated 1990 pesticide VOC emissions based on 1991 pesticide use data, adjusted to represent the 1990 base year, and on the VOC emission potential data”; “[t]he base year inventory will be created from the 1991 Pesticide Use Report [“PUR”] and then adjusted by a factor to represent the 1990 base year” (emphasis supplied).
The SIP also provides, however, that “[t]his baseline inventory may be adjusted if empirical data are developed to determine the impact of temperature, treated *927substrate (foliate, soil, water, etc.), application technique, and other conditions on VOC emissions.” Pis.’ SUF 19-20. Thus, the SIP leaves room for modification or “adjustment]” of the baseline inventory if it is impacted by specified conditions.
Defendants explain that when the SIP was written, it was assumed that the “1991 PUR would provide a more accurate basis to establish the 1990 emissions because staff anticipated that there would be low compliance in 1990.” Defs.’ Mot. at 32. Defendants elaborate that they expected 1991 would allow for more “complete reporting of PUR data.” Later, though, “staff reviewed the data” and “determined that the 1990 reporting was comparable to the reporting in 1991 and subsequent years,” and was “not underreported compared to the 1991 data.” Id. at 33.
Although the SIP allows for adjustments which may affect VOC emissions, it does not allow for adjustments relating to accuracy. Of course, that does not mean that a determination that a more accurate means of establishing the baseline must be ignored. If the state defendants concluded that the 1990 PUR data was more accurate, they could have submitted a revision to U.S. EPA for approval. They had no right, however, to unilaterally decide to use the 1990 data. It is established that a SIP, “once adopted by a state and approved by the EPA, becomes controlling and must be carried out by the state.” See Bayview Hunters Point Community Advocates, 366 F.3d at 692.
Below I conclude, however, that despite the fact that the State could not unilaterally alter the means of establishing the baseline, plaintiffs’ second claim fails because a “baseline inventory” is not an “emission standard or limitation” subject to challenge under the Clean Mr Act.
2. Section 304 of the Clean Air Act and “Emission Standard or Limitation” Requirement
Plaintiffs contend that the baseline inventory is a key element of the pesticide element’s strategy, and that “manipulation of the starting point skews the relative percentage comparison with subsequent years’ inventories.”21 Pis.’ Mot. at 11. They insist that the calculation of the 1990 baseline inventory qualifies as “an emission standard or limitation that Warmerdam22 has violated.” Pis.’ Opp’n at 17. Plaintiffs, however, face a significant barrier to their claim. The CAA’s creation of a private cause of action to enforce the Act is not all encompassing. Plaintiffs must limit their suit to those actions permitted under the Act. As I now explain, it appears to the court that the second claim is not so authorized.
Under section 304(a)(1) of the Act, plaintiffs may enforce an “emission standard” or “limitations” of a SIP. The Act defines “emission standard or limitation” as “a schedule or timetable of compliance, emission limitation, standard of performance or emission standard.” 42 U.S.C. § 7604(f)(l)-(4). Elsewhere, the CAA defines the terms “emission limitation” and “emission standard” to mean:
a requirement established by the State or the Administrator which limits the quantity, rate, or concentration of emissions of air pollutants on a continuous basis, including any requirement relating to the operation or maintenance of a source to assure continuous emission re*928duction, and any design, equipment, work practice or operational standard promulgated under this chapter.
42 U.S.C. § 7602(k).
Plaintiffs rely on the well-established principle that because the Clean Air Act is a remedial statute, the court should interpret the Act broadly to effectuate its purpose. Bayview Hunters Point Community Advocates, 366 F.3d at 692; Citizens For a Better Environment v. Deukmejian, 731 F.Supp. 1448, 1454 (N.D.Cal.1990). The problem with plaintiffs’ argument lies not with the principle they rely on. The problem is they confound construction and rewriting.
Neither Citizens for a Better Environment nor Bayview Hunters Point support plaintiffs’ attempt to rewrite the statute. In both cases, the measures at issue were designed to reduce emissions.23.
The case at bar, unlike the two cases relied on by plaintiffs, involves a baseline inventory which, as defendants note, is not a schedule or timetable of compliance, emission limitation, or standard as defined by the statute. Indeed, the baseline does not “limit[ ] the quantity, rate, or concentration of emissions of air pollutants on a continuous basis” and does not “relat[e] to the operation or maintenance of a source to assure continuous emission reduction.” By its own terms, the baseline identifies emission sources and then quantifies the amount of emissions attributed to those sources. As defendants argue, once the sources of air pollution are identified, control strategies can then be formulated to control emissions entering the air from those sources. From all the above, I must conclude that the baseline is not an emission “standard” or “limitation” within the meaning of 42 U.S.C. § 7604(f)(l)-(4).
The court’s conclusion above, while compelled by the language of the statute and the function of a baseline, is nonetheless troubling. Plaintiffs make a compelling argument that “manipulation of the starting point skews the relative percentage comparison with subsequent years’ inventories.” Pis.’ Mot. at 11. Thus, if defendants are able to arbitrarily alter the baseline they can “skew” the numbers in order to meet the NAAQS in subsequent years, defeating the purpose of the Clean Air Act’s mandate to reduce emissions and improve air quality.
Although plaintiffs urge that the citizen suit provision contemplates a broader reading of what can be defined as an emissions “limitation” or “standard,” and such a reading makes sense, it is unfortunately not supported by the statute. Plaintiffs’ motion for summary judgment as to the second cause of action must be DENIED.
C. PLAINTIFFS’ FIRST CLAIM-FAILURE TO ADOPT REGULATIONS IN VIOLATION OF THE SIP
Plaintiffs assert that defendants violated the SIP by failing to adopt regulations necessary to ensure interim and final reduction goals. Pis.’ Mot. at 15. They contend that the obligation to adopt regulations became an enforceable SIP strategy upon EPA approval of the Pesticide Element, as clarified by a May 9, 1995 memorandum submitted to EPA and included in the SIP. Id.
Defendants argue that the May 9, 1995 Memorandum cannot be the source of a federally-enforceable SIP strategy because *929it was not incorporated by reference into the Code of Federal Regulations. Defs.’ Mot. at 19. Alternatively, defendants argue that even if the Wells Memo had been approved as part of the SIP, the commitment to adopt regulations was a one-time obligation permitting the state to make a discretionary determination of necessity. They maintain that obligation was discharged when a decision was made not to adopt regulations. Defs.’ Mot. at 22. As the court explains below, plaintiffs’ motion for summary judgment as to the first cause of action must be granted because defendants failed to utilize 1991 PUR data to calculate the baseline.
1. Was there a Commitment in the SIP to Adopt Regulations?
“An obligation cannot be imposed based upon a SIP if that obligation was not actually undertaken in the SIP.” Bayview Hunters Point, 366 F.3d at 698.
Plaintiffs contend here, as they did in the motion for judgment on the pleadings, that the State’s commitment or “enforceable control measure,” which is to adopt regulations, derives from the May 1995 clarification letter signed by DPR Director James Wells which CARB director, James Boyd, forwarded to the EPA. To repeat, the clarification letter reads, in pertinent part:
The Department of Pesticide Regulation commits to adopt and submit to U.S. EPA by June 15, 1997, any regulations necessary to reduce volatile organic compound emissions from agricultural and commercial structural pesticides by specific percentages of the 1990 base year emissions, by specific years, and in specific nonattainment areas (described in 40 C.F.R. 81.305), as listed in the following table:
Reductions from 1990 Baseline_
Ozone Nonattainment Area 1996 1999 2002 2005
Sacramento Metro_8% 12% 16% 20%
San Joaquin Valley_8% 12% 16% 20%
South Coast _8% 12% 16% 20%
Southeast Desert_8% 12% 16% 20%
Ventura_8% 12% 16% 20%
Pis.’ SUF ¶ 32, Ex. 4 to Newell Dec. at 017-018.
The clarification letter further states that “[t]he Department of Pesticide Regulation commits to adopt and submit to U.S. EPA by June 15, 1997, any regulations necessary to reduce volatile organic compound emissions.... ”
As the court noted in its previous order, SIPs are composed of a compilation of numerous documents, including new and previously submitted plans, district rules, state regulations, and federal controls. See Order at 14, n. 12. The court also noted that it is apparently not unusual for a revision or letter of clarification to be written by an agency official and forwarded to EPA to be come part of the SIP. These letters often clarify the agencies’ intent with regard to SIPs. Id.24 It is this method of proceeding which is a major source of the confusion in the matter at bar.
The issue would be vastly simpler if the parties agreed that this clarification letter was or was not part of the SIP. Defendants, however, maintain here, as they did in connection with the motion for judgment on the pleadings, that, contrary to the plaintiffs’ contention, the Wells memorandum was not incorporated by reference into the Code of Federal Regulations.
*930As I noted in disposition of the previous motion, neither the statute, the regulatory-process, nor the record provide a clear resolution of the problem. On the one hand, the language contained in the final rule is similar to that in the CARB clarification letter. It states:
As described in the SIP, California has committed to adopt and submit to U.S. EPA by June 15, 1997 any regulations necessary to reduce VOC emissions from agricultural and commercial structural pesticides by 20 percent of the 1990 base yéar emissions in the attainment years for Sacramento, Ventura, Southeast Desert, and the South Coast, and by 12 percent in 1999 for San Joaquin Valley.
62 FedReg. 1150,1170.
It would seem that, based on this language, the court could conclude that California committed to submit regulations if 20 percent of the 1990 base year emissions for the above-mentioned basins were not met.
As the court discussed in the previous order, however, the final rule does not set out the relevant table titled “Reductions from 1990 Baseline” which was contained in the May 1995 memo and subsequently published in the proposed regulations. This deletion may signal the commitment was not included in the final rule.
The defendants also direct the court to the preamble of the final regulations, which they argue may serve “as a source of evidence concerning contemporaneous agency intent,” citing Entergy Services, Inc. v. F.E.R.C., 375 F.3d 1204, 1209 (D.C.Cir.2004). In the preamble, the EPA, while mentioning that it was approving other revisions to the California SIP, did not mention the May 1995 memorandum at issue.25 Again, this is a possible signal that the commitment was not included in the final rule.26
In the disposition of the motion for judgment on the pleadings, the court relied on the presumption of administrative regularity to conclude the commitment had been made, since such a commitment appeared to the court to be necessary for the EPA to approve the SIP. Plaintiffs have now provided persuasive evidence that the court’s disposition accorded with the understanding of the EPA administrator tasked with approval of the California SIP.27
*931Plaintiffs tender the declaration of David Howekamp, who served as the Director of the Air Division at EPA Region 9 during the period in question. The declaration appears dispositive. It confirms, inter alia, that EPA understood that California had made the commitment to submit and adopt regulations if CARB and DPR found it necessary for the meeting of attainment dates.28
According to his declaration, the letter Howekamp sent to DPR Director Wells requested various “technical clarifications” on several issues, including the deadline to adopt necessary regulations. Pis.’ SUF 24. Howekamp explained that even after Wells’ response to his letter, he could not approve the pesticide component because it was inconsistent with the Clean Air Act. Pis.’ SUF 26. Thereafter, in a letter to DPR dated April 21, 1995, Howekamp identified “approvability issues,” and reiterated that the State must modify the SIP to be explicit about the dates of rule adoption and the emission reduction goals. According to Howekamp, the modification was requested because the pesticides measure in the SIP “did not contain an adequate enforceable commitment to adopt and submit regulations by no later than June 15, 1997.” Pis.’ SUF 29 (citing How-ekamp Dec. at 14).
By virtue of the declaration, it is apparent that it was Howekamp’s interactions with the state agencies which prompted DPR to write to CARB clarifying its commitment, which in turn CARB submitted to EPA in the May 11, 1995 cover letter. It is significant that it was Howekamp’s position that the final rule made no change to the clarification in the May 1995 memorandum, regardless of the final rule’s omission of the table. Pis.’ SUF 36. In sum, Howekamp’s account of the rulemaking process demonstrates that the May 1995 memorandum was “an essential component of SIP approval.” Mott at 22.
Thus, despite the uncertainties noted above, based on the Howekamp declaration and the language contained in the final regulations under the “emissions reduction” section,29 the court concludes that EPA approved and promulgated the strategy to implement the regulations committed to by DPR in the 1995 memorandum and submitted by CARB.
2. The Scope of the Commitment to Implement Regulations
Having held that the Wells memo contained a commitment, which was incorporated into the final SIP, the court must now decide exactly what that commitment was. The language contained in the Wells memorandum explains:
As described in the SIP, California has committed to adopt and submit to U.S. EPA by June 15, 1997 any regulations necessary to reduce VOC emissions from agricultural and commercial structural pesticides by 20 percent of the 1990 base year emissions in the attainment years for Sacramento, Ventura, *932Southeast Desert, and the South Coast, and by 12 percent in 1999 for San Joaquin Valley.
62 Fed.Reg. 1150, 1170.
Plaintiffs contend that defendants have violated the SIP and continue to violate the SIP by “failing to adopt regulations by June 15, 1997, and subsequent to June 15, 1997.” That is, plaintiffs contend that any failure to meet interim goals would require defendants to adopt regulations and that defendants’ obligation was ongoing and mandatory. They cite various reports on emissions from years subsequent to June 1997 to support their contention that the State did not have discretion in determining whether regulations would be adopted. I cannot agree.
The language of the 1995 Wells Memo is relatively clear that defendants contemplated a one-time obligation.30 As defendants argue, the Wells memorandum could have stated that regulations must be adopted when and if the interim targets are missed or that the commitment was ongoing. Instead, the regulation identifies only a single deadline for adopting such regulations. Reading the language of the regulation as plaintiffs suggest—so that the commitment is ongoing—would strain the plain meaning. Clearly, the preposition “by” is used to mean “not later than.” In this regard, the pre-revision language of the SIP is of some help. The language from the SIP proposed by DPR provided:
A decision whether additional regulatory measures to ensure that reductions in pesticidal VOC emissions are achieved will be made by 1997. The determination that additional regulatory measures will be made if it is found that the existing program elements are not sufficient in achieving established VOC reductions and that additional VOC reductions from the use of pesticides are feasible.
SIP at 10.
Elsewhere, the SIP contains a “summary of the plan,” which is essentially a timeline. The timeline indicated that by December 1998 “[i]mplementation of additional regulatory measures, if necessary will take place to ensure that targeted pesticidal VOC reductions will occur.”31 SIP at 10-11.
Defendants also point out that in April 1995, when Dr. Howekamp wrote to DPR requesting further information about the obligation to adopt regulations, he noted that “[i]n order to be consistent with Sections 110(k)(2) and 110(k)(4) of the CAA, such commitments should not extend beyond June 1997.”32 Indeed, as defendants contend, it appears from this correspondence that “[t]he commitment to adopt regulations, if necessary,” was in essence a commitment to submit another SIP revi*933sion and promulgate new regulations not later than June 15, 1997, if CARB and DPR determined they were needed.33 The court concludes that based on the record before it, the SIP called for a commitment on the part of DPR and CARB to look at the emissions data before it and to promulgate regulations it thought would allow it to meet attainment dates.
3. Did Defendants Fulfill the OneTime Obligation to Adopt Regulations if necessary by June IS, 1997? And does this Comply with the mandate of the Clean Air Act?
I concluded above that defendants committed to make an assessment as to emissions and to pass regulations before June 15,1997 if they thought it was necessary to meet the NAAQS for future years.
It is undisputed that DPR did not submit any regulations to the U.S. EPA by June 15, 1997. There is some evidence that defendants did indeed consider whether regulations were necessary prior to the June 15, 1997 deadline, and decided that regulations were not necessary. A letter from Wells and John Dunlap of DPR to Marc Chytilo of the Environmental Defense Center (“Chytilo letter”) explained that “[o]ur April 1997 analysis of data through 1995 showed that all nonattainment areas (NAAs) were meeting annual interim goals, and that four out of five NAAs were already meeting the final 2005 year goals.” Pis.’ SUF 48.34 Defendants also submitted a graph of the inventory for 1990-1995 which “reflect[s] the data known to the Department at the time of the analysis referred to in the Chytilo letter.” Pis.’ SUF 49.35
As I now explain, however, even assuming that an analysis was undertaken and a decision reached, plaintiffs’ motion must be granted because, under defendants’ contention, the decision could be based on wishful thinking rather than an accurate forecast of future emission trends.
Defendants contend that they complied with the CAA by merely committing to pass regulations before June 15, 1997 only if they thought that regulations were necessary to meet the NAAQS for future years. If the court adopted that contention, it would be required to conclude that the EPA had violated its duty in accepting the Pesticide Element of the SIP. Under the statute, EPA must ensure that SIPs contain “enforceable emission limitations, and such other control measures, means or techniques ... as may be *934necessary or appropriate to meet the applicable requirements of [the Act].” 42 U.S.C. § 7502(6). Moreover, the statute requires that a SIP for nonattainment areas include enforceable emission limitations and control measures “as may be necessary or appropriate to provide for attainment of the NAAQS by the applicable attainment date.” Id. at § 7506(4). Thus, for a SIP to be valid, enforceable standards rather than a discretionary decision must be included. In the absence of a state commitment to promulgate regulations based upon meaningful data, rather than an option to pass regulations, the statutory requirements would not be met.
The court, however, need not hold that the SIP is invalid. Even if there was a formal decision premised upon an evaluation of the data demonstrating no objective need for regulations, the process utilized to reach that decision would clearly not be in compliance with the SIP.
It is undisputed that defendants were utilizing 1990 PUR data in 1997 when they wrote the letter to Mr. Chytilo explaining why they believed regulations were unnecessary. This violated the SIP which required the use of 1991 PUR data and backcasting to 1990. Defendants maintain that they used the 1990 PUR data because they believed the 1990 data was more accurate than they had previously expected it to be. Defendants argue that it would be “sheer absurdity” to “preclud[e] the use of best data available in [implementing the Clean Air Act].” Defs.’ Mot. at 36.
Of course, using the best data makes sense. Defendants’ problem is that the SIP did not permit use of 1990 data. If, in fact, defendants concluded that the 1991 PUR data was not the best data to use, they could have, and more importantly, should have, submitted a revision to the SIP. As plaintiffs argue, defendants failed to avail themselves of the opportunity to amend the SIP to allow for a change in the baseline calculation commitment. As I noted above, a “SIP once adopted by a state becomes controlling and must be carried out by the state.” Bayview Hunters Point, 366 F.3d at 695. Defendants simply are not permitted to violate federal law, whether they had a better thought or not. Under the CAA, California may submit a revision to change the baseline calculation if it will not interfere with any applicable requirement concerning attainment and reasonable further progress. 42 U.S.C. §§ 7410(a), 7410(1). Where they have failed to do so, states are “relegated to a lone option: compliance,” Friends of the Earth, 535 F.2d at 178. Because DPR did not exercise the procedural option to revise the SIP, it had to comply with the SIP and utilize the 1991 PUR.
Defendants failed to carry.out the commitment as promulgated in the regulations. Accordingly, the court must grant summary judgment for plaintiffs as to the first cause of action.
For the foregoing reasons the court hereby ORDERS that:
1. Defendants’ motion for summary judgment as to plaintiffs’ second cause of action is GRANTED;
2. Plaintiffs’ motion for summary judgment as to plaintiffs’ first cause of action is GRANTED;
3. In all other respects, the motions are DENIED; and
4. The parties are directed to file cross-motions as to remedies not later than twenty (20) days from the date of this order. The court will then either take the matter of remedies under submission or set the matter for oral argument.
IT IS SO ORDERED.
. On December 16, 2004, this court granted Air Coalition Team's motion to intervene pursuant to Fed.R.Civ.P. 24(a). On April 25, 2005, the court denied defendants' motion for judgment on the pleadings.
. Plaintiffs bring suit against the following defendants in their official capacities: Paul Helliker, Director, Department of Pesticide Regulation; Terry Tamminen, Secretary, California Environmental Protection Agency; *916Catherine Witherspoon, Executive Officer, Air Resources board; Allan Loyd, Chairman, Air Resources Board; and William Burke, Joseph Calhoun, Dorene D'Adama, Mark Desaulnier, C. Hugh Friedman, William F. Friedman, Matthew McKinnon, Barbara Patrick, Barbara Riordan, Ron Roberts, members, Air Resources Board.
In their original complaint, plaintiffs brought only one claim under the Act. On July 14, 2005, plaintiffs amended their complaint to add a second cause of action.
. Plaintiffs include El Comité para El Bienes-tar de Earlimart, Association of Irritated Residents, Community and Children’s Advocates Against Pesticide Poisoning, Wishtoyo Foundation, and Ventura Coastkeeper. Pis.' Compl. at 1. Defendants do not challenge plaintiffs’ standing, nor could they reasonably do so.
. Plaintiffs request the following relief;
a.An order declaring that defendants are in violation of the Act and the 1994 Ozone SIP by failing to adopt and implement regulations
b. A preliminary and permanent injunction directing defendants to adopt, implement, and submit to the U.S. EPA regulations necessary to reduce pesticide-related VOC emissions by 20% from the 1990 baseline for the Sacramento, San Joaquin Valley, Ventura, and Southeast Desert nonattainment areas no later than January 1, 2005.
c. That the court retain jurisdiction over this matter until defendants have complied with their duties as described above
d. Costs of litigation, including reasonable attorney and expert witness fees pursuant to the CAA.
.To say the least, the court finds itself bemused by the result. Nonetheless, given the statute, the means of adopting SIPS, and the record in the matter at bar, the result, while ironic, is not inconsistent.
. Plaintiffs allege that all five air basins or "ozone nonattainment areas” at issue in this case have been designated "severe” or "extreme” for the one-hour and eight-hour ozone NAAQS, except the Southeast Desert and Ventura nonattainment areas which have been designated “moderate” for the eight-hour ozone NAAQS. Compl. at 6.
. Facts are undisputed unless otherwise noted.
. Dr. Howekamp explained that his primary responsibility at U.S. EPA was to "oversee implementation of the Clean Air Act and radiation programs in California, Arizona, Nevada and Hawaii.” Among his duties, Dr. How-ekamp’ "interpreted applicable state and local laws and regulations and integrated them with rFederal Laws.” His employment required him to work "closely with state and local regulatory agency officials and elected officials in state legislatures and on local air district boards to adopt needed laws, regulations, and measures.” Howekamp Deck ¶¶ 3, 6.
. California developed the 1994 Ozone Sip under threat of an EPA takeover of California's ozone air quality planning. On July 1, 1992, the Ninth Circuit held that EPA had a duty to promulgate a FIP for the South Coast Nonattainment area because EPA had disapproved of the area's ozone and carbon monoxide state implementation plans in 1998. See Coalition for Clean Air v. U.S. EPA, 971 F.2d 219, 228-229 (9th Cir.1992). Section 110(c)(1)(A) requires EPA to promulgate a FIP two years after EPA disapproves of a state implementation plan, unless the state corrects such deficiencies and EPA approves the plan. 42 U.S.C. § 7410(c)(1)(A). On May 5, 1994, EPA proposed a FIP containing measures, including pesticide controls, for bringing the South Coast, Sacramento and Ventura nonattainment areas into attainment within the one hour ozone NAAQS. 59 Fed.Reg. 23264, 23220-23323 (May 5, 1994).
. The Plan also stated that a “determination that additional regulatory measures will be made if it is found that the existing program measures are not sufficient in achieving established VOC reductions and that VOC reductions from the use of pesticides are feasible.” Plan at 10.
. Dr. Segawa explained how DPR used default emission potentials:
The department requested emission potential data from several hundred registrants in which they would provide information regarding the VOC content of the pesticide products included in the reevaluation. Once the department received that information, then DPR staff reviewed the information and if appropriated approved it and entered the data into an electronic database ... In the case where we did not receive the emission potential data from registrants, then DPR had other methods to estimate or make assumptions as to what the emission potential data for those remaining products was.
Pis’ SUF 16.
.The pesticide element also states that ”[t]his baseline inventory may be adjusted if empirical data are developed to determine the *920impact of temperature, treated substrate (foliage, soil, water, etc.), application technique, and other conditions on VOC emissions.”
. The element notes that ”[u]nder the Clean Air Act Amendments of 1990, the baseline inventory must be based on 1990 emissions.”
. The proposed rule provided that "[t]his clarification is considered part of California’s SIP.”
. The FOIA Response, Exhibit S, states that the attached documents are identified and codified at 40 C.F.R. §§ 52.220(c)(204)(i)(A)(6) and 52.220(c)(236)(i)(A)(l). Defs.’ SUF 15.
. Dr. Segawa also explained that DPR revises its calculations of the baseline and each year’s inventory every year as more information became available. Defs.’ SUF 21.
. Gosselin Dep. at 15:12-13. Mark Pepple is a Senior Research Scientist at DPR. Pepple Dec. at 1:18-20.
. Dr. Segawa explains the reformulation option during his deposition as follows:
Q. Did you participate in decisions whether or not to adopt regulations based on the data in this document, Exhibit 15?
A. Yes.
Q. Can you describe those decisions?
A. We decided to pursue another reevaluation in place of regulations.
Q. What is the reevaluation?
A. In this case I’m referring to a reevaluation to request pesticide registrants reformulate a number of different products to lower the VOC content.
Q. What other options were discussed?
A. As I recall, we also discussed the possibility of reducing emissions from fumigants.
Q. What possibilities were those?
A. To adopt changes in applications methods, limit—or limit the amount of fumigants that could be used.
Q. Has a decision regarding those possibilities been made?
A. Yes.
Q. What is that decision?
A. We decided-—at that time we decided not to pursue regulatory measures for fumigants.
Q. Who is "we”?
A. Paul Gosselin and Maiy-Ann Warmer-dam made that decision.
Pis.’ SUF 75.
. As defendants explain, "[a]ny plan to reduce VOC emissions must establish a 'baseline' against which VOC emission reductions can be measured.” Defs.’ Mot. at 31.
. The record reflects that beginning in 1997, DPR began to rethink how it should calculate the baseline, both in an internal memorandum and in material produced for a DPR workshop. Pis.’ SUF 37.
. The initial inventory in 1990, and all subsequent inventories are generated by multiplying the amount of pesticides used by the VOC content of those pesticides. Pis.’ SUF 19.
. Maty Ann Warmerdam is the current director of the DPR and replaced Helliker during the course of this litigation.
. In Citizens for a Better Environment, at issue were "twenty-three [stationary source] measures,” which would "be implemented as new regulations by the BAAQMD” to reduce emissions. 731 F.Supp. at 1455. In Bayview, it was Transportation Control Measures ("TCM”) designed to reduce air pollutants by increasing public transit. 366 F.3d at 694.
. See, e.g., 40 C.F.R. Pt. 52.200 (2004) (letter of clarification was submitted by the Arkansas governor regarding Arkansas' pollution control); 40 C.F.R. Pt. 52.780 (2004) (Attorney General of the State of Indiana sent a letter to U.S. EPA clarifying Indiana's interpretation of the definition of "federally enforceable” under the Act.).
. The final regulation lists the following revisions as being submitted to the SIP: November 14, 1994, November 15, 1994, December 28, 1994, December 29, 1994, February 7, 1995, March 30, 1995, January 22, 1996, April 4, 1996, May 17, 1996, June 13, 1996, July 10, 1996, and July 12, 1996. See 62 Fed.Reg. 1150. The regulation does not list the May 1995 date, the date of the Wells memorandum.
Defendants also explain that, based on the FOIA response noted above they received from defendants, they have shown that the memorandum was not part of the SIP.
. That the court is, in essence, being urged to read tea leaves, strongly suggests that there is something profoundly wrong with the process.
.The court notes that the CAA is governed by the judicial review provisions of the Administrative Procedure Act. 5 U.S.C. §§ 701-706. Public Citizen v. Dept. of Transp., 316 F.3d 1002, 1021 (9th Cir.2003). Although an argument could be made that the -issue must be resolved on the basis of the administrative record, a number of reasons support consideration of the declaration. First, defendants have not objected to its receipt, and thus it is in evidence. Second, as the court has repeatedly noted, the administrative record is ambiguous, and limiting consideration to the record frustrates the ability to resolve the issue. Third, an exception to the rule exists where resort to extrinsic material is required to “determine whether the agency has considered all relevant factors and has explained its decision,” or when it is "necessary to explain technical terms or complex subject matter.” *931Southwest Center for Biological Diversity v. United States Forest Svc., 100 F.3d 1443, 1450 (9th Cir.1996).
. An assertion by defendants that they did not intend to so commit would be unavailing. Defendants’ unexpressed intentions are irrelevant. Moreover, once the SIP is approved by the EPA, it becomes federal law, and this court need only look to see whether state defendants complied with the governing law. As I explained in disposing of the motion for judgment on the pleadings, a commitment was necessary to comply with the CAA. Indeed, as the Howekamp declaration demonstrates, that was also EPA’s understanding and the SIP would not have been approved without such a commitment.
. The language is derived verbatim from the Wells memorandum.
. Again, the pertinent language is "California has committed to adopt and submit to U.S. EPA by June 15, 1997, any regulations necessary to reduce VOC emissions ... by 20 percent ... in the attainment years
. It seems evident that it was the 1998 date which, inter alia, required Howekamp’s disapproval of the proposed SIP.
. Apparently, these two sections have to do with deadlines that Howekamp as an administrator was facing under the Clean Air Act. Section 110(k)(2) states that within twelve months of determination by the Administrator that a state has submitted a plan, the administrator has to act on the submission. Section 110(k)(4) has to do with conditional approval. It states that the Administrator may approve a plan revision based on a commitment of the State to adopt specific enforceable measures, but not later than one year after the date of the approval of the plan revision. Thus, when the State wanted to move the obligation date to 1998, Howekamp was obligated to have them move the date up to 1997 so he could comply with his own timeline under the statute.
.Like everything else in the rule making process, this correspondence is not free of ambiguity. On the one hand, Howekamp writes in the April 21, 1995 letter to Wells that "[t]he primary issue [he was concerned with] is the date by which California has committed to adopt and submit regulations should they be needed to reduce VOC emissions.” This conditional language supports defendants’ reading that their obligation was only one-time and that obligation only required defendants to make an assessment in 1997 whether regulations were necessary to meet reductions by the attainment dates. On the other hand, the Howekamp letters also suggest that EPA understood that the obligation to adopt regulations was not discretionary. See, e.g., Letter of March 20, 1995 to Mr. Wells ("We understand that California has committed to limiting VOC emissions from pesticides to specific percentages of the 1990 base year emissions by specific years in specific nonattainment areas ... regardless of future growth in emissions that might otherwise occur.”).
. The letter was in response to Mr. Chytilo's Notice of Intent to File Suit under the Clean Air Act.
. It is somewhat disconcerting that the defendants have failed to provide any documentation that an analysis had in fact been undertaken and a decision actually reached prior to the letter to Mr. Chytilo, or at any time for that matter.