# Approval and Promulgation of Implementation Plans; Designation of Areas for Air Quality Planning Purposes; California; San Joaquin Valley Moderate Area Plan and Reclassification as Serious Nonattainment for the 2006 PM2.5 NAAQS

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A2015-00270

## Record

- **Collection:** Federal Register
- **Document type:** Proposed Rule
- **Published:** January 13, 2015
- **Citation:** 80 FR 1816

## Text

ENVIRONMENTAL PROTECTION AGENCY
40 CFR Parts 52 and Part 81
[EPA-R09-OAR-2014-0636; FRL-9921-48-Region 9]

Approval and Promulgation of Implementation Plans; Designation of Areas for Air Quality Planning Purposes; California; San Joaquin Valley Moderate Area Plan and Reclassification as Serious Nonattainment for the 2006 PM
2.5
NAAQS

AGENCY:

U.S. Environmental Protection Agency.

ACTION:

Proposed rule.

SUMMARY:

The Environmental Protection Agency (EPA) is proposing to approve state implementation plan (SIP) revisions submitted by California to address Clean Air Act (CAA or Act) requirements for the 2006 24-hour fine particulate matter (PM
2.5
) national ambient air quality standard (NAAQS) in the San Joaquin Valley (SJV) Moderate PM
2.5
nonattainment area. These SIP revisions are the 2012 PM
2.5
Plan, submitted March 4, 2013, and the Supplement, submitted November 6, 2014. The EPA is also proposing to disapprove interpollutant trading ratios identified in these SIP submittals for nonattainment new source review permitting purposes. Finally, the EPA is proposing to reclassify the SJV area, including Indian country within it, as a Serious nonattainment area for the 2006 PM
2.5
NAAQS based on EPA's determination that the area cannot practicably attain this standard by the applicable Moderate area attainment date of December 31, 2015. Upon final reclassification as a Serious area, California will be required to submit a Serious area plan including a demonstration that the plan provides for attainment of the 2006 PM
2.5
NAAQS in the SJV area by the applicable Serious area attainment date, which is no later than December 31, 2019, or by the most expeditious alternative date practicable, in accordance with the requirements of part D of title I of the CAA.

DATES:

Any comments must arrive by February 27, 2015.

ADDRESSES:

Submit comments, identified by docket number EPA-R09-OAR-2014-0636, by one of the following methods:

• Federal eRulemaking Portal:
www.regulations.gov.
Follow the on-line instructions.

• Email:
tax.wienke@epa.gov.

• Mail or deliver: Wienke Tax, Office of Air Planning (AIR-2), U.S. Environmental Protection Agency Region 9, 75 Hawthorne Street, San Francisco, CA 94105.

Instructions:
All comments will be included in the public docket without change and may be made available online at
www.regulations.gov,
including any personal information provided, unless the comment includes Confidential Business Information (CBI) or other information whose disclosure is restricted by statute. Information that you consider CBI or otherwise protected should be clearly identified as such and should not be submitted through
www.regulations.gov
or email. The
www.regulations.gov
Web site is an “anonymous access” system, and the EPA will not know your identity or contact information unless you provide it in the body of your comment. If you send email directly to the EPA, your email address will be automatically captured and included as part of the public comment. If the EPA cannot read your comments due to technical difficulties and cannot contact you for clarification, the EPA may not be able to consider your comment.

Docket:
The index to the docket (docket number EPA-R09-OAR-2014-0636) for this proposed rule is available electronically on the
www.regulations.gov
Web site and in hard copy at EPA Region 9, 75 Hawthorne Street, San Francisco, California, 94105. While all documents in the docket are listed in the index, some information may be publicly available only at the hard copy location (
e.g.,
copyrighted material), and some may not be publicly available at either location (
e.g.,
CBI). To inspect the hard copy materials, please schedule an appointment during normal business hours with the contact listed in the
FOR FURTHER INFORMATION CONTACT
section below.

FOR FURTHER INFORMATION CONTACT:

Wienke Tax, Air Planning Office (AIR-2), U.S. Environmental Protection Agency, Region 9, (415) 947-4192,
tax.wienke@epa.gov
.

SUPPLEMENTARY INFORMATION:

Throughout this document, “we,” “us” and “our” refer to the EPA.

Table of Contents

I. Background for Proposed Actions

II. Clean Air Act Requirements for Moderate PM
2.5
Nonattainment Area Plans

III. Clean Air Act Procedural Requirements for SIP Submittals

IV. Review of the San Joaquin Valley 2012 PM
2.5
Plan and Supplement

A. Emissions Inventory

B. Air Quality Modeling

C. PM
2.5
Precursors

D. Reasonably Available Control Measures/Reasonably Available Control Technology

E. Major Stationary Source Control Requirements Under CAA Section 189(e)

F. Adopted Control Strategy

G. Demonstration That Attainment by the Moderate Area Attainment Date Is Impracticable

H. Reasonable Further Progress and Quantitative Milestones

I. Contingency Measures

J. Interpollutant Trading Ratios for Nonattainment New Source Review Permits

K. Motor Vehicle Emission Budgets

V. Reclassification as Serious Nonattainment and Serious Area SIP Requirements

A. Reclassification as Serious and Applicable Attainment Date

B. Clean Air Act Requirements for Serious PM
2.5
Nonattainment Area Plans

C. Statutory Deadline for Submittal of the Serious Area Plan

VI. Reclassification of Indian Country

VII. Summary of Proposed Actions and Request for Public Comment

VIII. Statutory and Executive Order Reviews

I. Background for Proposed Actions

On October 17, 2006, the EPA revised the 24-hour national ambient air quality standard (NAAQS or standard) for PM
2.5
, particulate matter with a diameter of 2.5 microns or less, to provide increased protection of public health by lowering its level from 65 micrograms per cubic meter (μg/m
3
) to 35 μg/m
3
(40 CFR 50.13).
1

Epidemiological studies have shown statistically significant correlations between elevated PM
2.5
levels and premature mortality. Other important health effects associated with PM
2.5
exposure include aggravation of respiratory and cardiovascular disease (as indicated by increased hospital admissions, emergency room visits, absences from school or work, and restricted activity days), changes in lung function and increased respiratory symptoms. Individuals particularly sensitive to PM
2.5
exposure include older adults, people with heart and lung disease, and children (78 FR 3086 at 3088, January 15, 2013). PM
2.5
can be emitted directly into the atmosphere as a solid or liquid particle (“primary PM
2.5
” or “direct PM
2.5
”) or can be formed in the atmosphere as a result of

various chemical reactions among precursor pollutants such as nitrogen oxides, sulfur oxides, volatile organic compounds, and ammonia (“secondary PM
2.5
”).
2

1

See
71 FR 61224 (October 17, 2006). The EPA set the first NAAQS for PM
2.5
on July 18, 1997 (62 FR 36852), including annual standards of 15.0 μg/m
3
based on a 3-year average of annual mean PM
2.5
concentrations and 24-hour (daily) standards of 65 μg/m
3
based on a 3-year average of 98th percentile 24-hour concentrations (40 CFR 50.7). In 2012, the EPA revised the annual standard to lower its level to 12 μg/m
3
(78 FR 3086, January 15, 2013, codified at 40 CFR 50.18). Unless otherwise noted, all references to the PM
2.5
standard in this notice are to the 2006 24-hour standard of 35 μg/m
3
codified at 40 CFR 50.13.

2

See
EPA, Regulatory Impact Analysis for the Final Revisions to the National Ambient Air Quality Standards for Particulate Matter (EPA-452/R-12-005, December 2012), p. 2-1.

Following promulgation of a new or revised NAAQS, the EPA is required by CAA section 107(d) to designate areas throughout the nation as attaining or not attaining the NAAQS. On November 13, 2009, the EPA designated the SJV as nonattainment for the 2006 PM
2.5
standard of 35 μg/m
3
(74 FR 58688, November 13, 2009). This designation became effective on December 14, 2009 (40 CFR 81.305). The SJV area is also designated nonattainment for the 1997 annual and 24-hour PM
2.5
standards.
3

3

See
70 FR 944 (January 5, 2005) and 40 CFR 81.305. In June 2008, California submitted the 2008 PM
2.5
Plan to provide for attainment of the 1997 PM
2.5
standards in the SJV. In November 2011, the EPA approved all but the contingency measures in the 2008 PM
2.5
Plan (76 FR 69896, November 9, 2011). In July 2013, the State submitted a revised contingency measure plan, which the EPA approved in May 2014 (79 FR 29327, May 22, 2014).

The SJV PM
2.5
nonattainment area is home to 4 million people and is the nation's leading agricultural region. Stretching over 250 miles from north to south, it is partially enclosed by the Coast Mountain range to the west, the Tehachapi Mountains to the south, and the Sierra Nevada range to the east. It encompasses over 23,000 square miles and includes all or part of eight counties: San Joaquin, Stanislaus, Merced, Madera, Fresno, Tulare, Kings, and the valley portion of Kern. For a precise description of the geographic boundaries of the San Joaquin Valley PM
2.5
nonattainment area, see 40 CFR 81.305.

The local air district with primary responsibility for developing plans to attain the NAAQS in the area is the San Joaquin Valley Air Pollution Control District (SJVAPCD or District). The District works cooperatively with the California Air Resources Board (CARB) in preparing these plans. Authority for regulating sources under state jurisdiction in the SJV is split between the District, which has responsibility for regulating stationary and most area sources, and CARB, which has responsibility for regulating most mobile sources.

Ambient 24-hour PM
2.5
NAAQS design value levels in the San Joaquin Valley are among the highest recorded in the United States for the 2011-2013 period.
4

Exceedances of the 24-hour PM
2.5
standard occur almost exclusively during the late fall and winter months from October to March, when ambient PM
2.5
is dominated by ammonium nitrate (a secondary particulate formed from nitrogen oxides (NO
X
) and ammonia emissions) and directly-emitted particulates, such as wood smoke. During the winter, the SJV experiences extended periods of stagnant weather with cold foggy conditions which encourage wood burning and are conducive to the formation of ammonium nitrate (2012 PM
2.5
, Appendix G, pp. 7 to 9).

4

See
U.S. EPA, 2013 Design Value Reports, PM
2.5
Detailed Information Updated 8/24/14, available at
http://www.epa.gov/airtrends/values.html
(“PM
2.5
_DesignValues_20112013_FINAL_08_28_14”) (hereafter “2013 PM
2.5
Design Value Reports”). “Design values” are the 3-year average NAAQS metrics that are compared to the NAAQS levels to determine when a monitoring site meets or does not meet the NAAQS.
See
40 CFR part 50 Appendix N, Section 1.0(c).

II. Clean Air Act Requirements for PM
2.5
Moderate Nonattainment Area Plans

In April 2007, the EPA issued the Clean Air Fine Particle Implementation Rule (“2007 PM
2.5
Implementation Rule”) to assist states with the development of SIPs to meet the Act's attainment planning requirements for the 1997 PM
2.5
standards (72 FR 20583, April 25, 2007, codified at 40 CFR part 51, subpart Z). This rule was premised on the EPA's prior interpretation of the Act as allowing for implementation of the PM
2.5
NAAQS solely pursuant to the general nonattainment area provisions in subpart 1 of part D, title I of the CAA (“subpart 1”) and not the more specific provisions for particulate matter nonattainment areas in subpart 4 of part D, title I of the Act (“subpart 4”). Among other things, the 2007 PM
2.5
Implementation Rule included nationally applicable presumptions regarding the need to evaluate and potentially control emissions of certain PM
2.5
precursors.
5

5
Specifically, in 40 CFR 51.1002(c), the EPA provided, among other things, that a state was “not required to address VOC [and ammonia] as . . . PM
2.5
attainment plan precursor[s] and to evaluate sources of VOC [and ammonia] emissions in the State for control measures,” unless the State or the EPA provided an appropriate technical demonstration showing that emissions from sources of these pollutants “significantly contribute” to PM
2.5
concentrations in the nonattainment area (40 CFR 51.1002(c)(3), (4) and 72 FR 20586 at 20589-97 (April 25, 2007)).

In March 2012, the EPA issued a guidance document to aid states in preparing SIPs to meet the Act's attainment planning requirements for the 2006 24-hour PM
2.5
standard.
6

The 2012 guidance was based, in large part, on the requirements in the 2007 PM
2.5
Implementation Rule, which the EPA based solely upon the statutory requirements of subpart 1.

6

See
Memorandum, dated March 2, 2012 (withdrawn June 6, 2013), from Stephen D. Page, Director, Office of Air Quality Planning and Standards, to EPA Regional Air Directors, Regions I-X re: “Implementation Guidance for the 2006 24-Hour Fine Particle (PM
2.5
) National Ambient Air Quality Standards (NAAQS)” (“2012 Guidance”) Available at:
http://epa.gov/ttn/naaqs/pm/pm25_guide.html.

California had three years from the effective date of SJV's designation as nonattainment for the 2006 PM
2.5
standard to submit a SIP for the SJV that addressed the applicable requirements of the Act.
7

On December 20, 2012, the District adopted the 2012 PM
2.5
Plan to provide for attainment of the 2006 PM
2.5
standard. On January 24, 2013, CARB adopted the Plan as an element of the California SIP and submitted it to the EPA on March 4, 2013.

7

See
CAA section 172(b) and 40 CFR 51.1002(a).

On January 4, 2013, several weeks after the District's adoption of the Plan, the U.S. Court of Appeals for the DC Circuit issued its decision in a challenge to the EPA's 2007 mPM
2.5
Implementation Rule (
NRDC
v.
EPA,
706 F.3d 428 (D.C. Cir. 2013)). In
NRDC,
the court held that the EPA erred in implementing the 1997 PM
2.5
standards solely pursuant to the general implementation requirements of subpart 1, without also considering the requirements specific to particulate matter nonattainment areas in subpart 4.
8

The court reasoned that the plain meaning of the CAA requires implementation of the 1997 PM
2.5
standards under subpart 4 because PM
2.5
particles fall within the statutory definition of PM
10
and are thus subject to the same statutory requirements as PM
10
. The court remanded the 2007 PM
2.5
Implementation Rule in its entirety, including the presumptions concerning VOC and ammonia in 40 CFR 51.1002, and instructed the EPA “to repromulgate these rules pursuant to Subpart 4 consistent with this opinion.”

8
The
NRDC
decision also remanded the EPA's 2008 final rule to implement the nonattainment New Source Review (NNSR) permitting requirements for PM
2.5
(73 FR 28231, May 16, 2008) which, like the 2007 PM
2.5
Implementation Rule, was premised on the requirements of subpart 1. Today's proposal does not address requirements for NNSR programs other than the requirements concerning PM
2.5
precursors in CAA section 189(e), which we discuss in Section IV.C below, and PM
2.5
interpollutant trading ratios, which we discuss in Section IV.J below.

Consistent with the
NRDC
decision, on June 2, 2014 (79 FR 31566), the EPA published a final rule classifying all areas currently designated nonattainment for the 1997 and/or 2006 PM
2.5
standards as “Moderate” under subpart 4 and establishing a deadline of December 31, 2014 for states to submit

any attainment-related and nonattainment new source review (NNSR) SIP elements required for these areas pursuant to subpart 4. The EPA provided its rationale for these actions in both the proposed and final classification/deadline rule.
9

9

See
79 FR 69806, 69809 (November 21, 2013) and 79 FR 31566, 31568 (June 2, 2014).

On September 18, 2014, the District adopted the “Supplemental Document, Clean Air Act Subpart 4: The 2012 PM
2.5
Plan for the 2006 PM
2.5
Standard and District Rule 2201 (New and Modified Stationary Source Review)” (“Supplement”) as a revision to the 2012 PM
2.5
Plan. The District adopted the Supplement to address subpart 4 requirements for the 2006 PM
2.5
standard to the extent that these requirements were not adequately addressed in the 2012 PM
2.5
Plan. CARB submitted the Supplement to the EPA on November 6, 2014. The Supplement includes information on the implementation of reasonably available controls for ammonia sources in the SJV and the District's demonstration that attainment by the Moderate area attainment date of December 31, 2015 is impracticable (“impracticability demonstration”). As a consequence of the
NRDC
decision, we are reviewing the 2012 PM
2.5
Plan and Supplement for compliance with the applicable requirements of both subpart 1 and subpart 4.

The EPA has longstanding guidance interpreting the subpart 4 requirements for particulate matter nonattainment areas (
see
“State Implementation Plans; General Preamble for the Implementation of Title I of the Clean Air Act Amendments of 1990” (57 FR 13498, April 16, 1992) (“General Preamble”) and “State Implementation Plans for Serious PM-10 Nonattainment Areas, and Attainment Date Waivers for PM-10 Nonattainment Areas Generally; Addendum to the General Preamble for the Implementation of Title I of the Clean Air Act Amendments of 1990” (59 FR 41998, August 16, 1994) (“Addendum”)). The General Preamble at 13538 discusses the relationship of subpart 1 and subpart 4 SIP requirements, and notes that SIPs for moderate nonattainment areas must meet the general provisions in subpart 1 to the extent that these provisions are not otherwise “subsumed by, or integrally related to, the more specific [subpart 4] requirements.” Some subpart 1 provisions have no subpart 4 equivalent (
e.g.,
the emission inventories (CAA section 172(c)(3)) and contingency measures (CAA section 172(c)(9)) and for these provisions, subpart 1 continues to govern. Other provisions of subpart 1 are subsumed or superseded by more specific requirements in subpart 4 (
e.g.,
certain provisions concerning attainment dates).

Because the 2012 PM
2.5
Plan was initially developed and submitted to meet the requirements of subpart 1 and the 2007 PM
2.5
Implementation Rule, certain elements of the Plan address provisions of subpart 1 rather than the applicable provisions of subpart 4. Specifically, these elements are the State's request for an attainment date extension from 2014 to 2019 under CAA section 172(a)(2)(A); the demonstration of attainment by 2019; those portions of the reasonably available control measure (RACM) demonstration that show there are no section 172(c)(1) RACM that would expedite attainment from 2019 to 2018; the transportation conformity motor vehicle emission budgets for 2019; and the contingency measures for failure to attain. We are not proposing any action on these specific SIP elements at this time.

As part of this proposal, the EPA is proposing to determine that the SJV cannot practicably attain the 2006 PM
2.5
standard by the Moderate area attainment date of December 31, 2015 and to reclassify the area from Moderate to Serious nonattainment under subpart 4. Should the EPA finalize this proposal to reclassify the SJV area as a Serious area, the State will be required to adopt and submit a new plan addressing the Serious area requirements in subpart 4. We discuss these Serious area requirements in more detail in section V. below.

III. Clean Air Act Procedural Requirements for SIP Submittals

We are proposing action on two California SIP submittals. The first is the “2012 PM
2.5
Plan,” which the State submitted to EPA on March 4, 2013 (hereafter “2012 PM
2.5
Plan” or “the Plan”)
10

and the second is the “Supplemental Document, Clean Air Act Subpart 4: The 2012 PM
2.5
Plan for the 2006 PM
2.5
Standard and District Rule 2201 (New and Modified Stationary Sources),” which the State submitted to EPA on November 6, 2014 (hereafter “the Supplement”).
11

10

See
Letter dated March 4, 2013, from James N. Goldstene, Executive Officer, California Air Resources Board, to Jared Blumenfeld, Regional Administrator, EPA Region 9, with enclosures.

11

See
Letter dated November 6, 2014, from James N. Goldstene, Executive Officer, California Air Resources Board, to Jared Blumenfeld, Regional Administrator, EPA Region 9, with enclosures.

CAA sections 110(a)(1) and (2) and 110(l) require each state to provide reasonable public notice and opportunity for public hearing prior to the adoption and submittal of a SIP or SIP revision. To meet this requirement, every SIP submittal should include evidence that adequate public notice was given and an opportunity for a public hearing was provided consistent with the EPA's implementing regulations in 40 CFR 51.102.

Both the District and CARB satisfied applicable statutory and regulatory requirements for reasonable public notice and hearing prior to adoption and submittal of the 2012 PM
2.5
Plan. The District conducted public workshops, provided public comment periods, and held a public hearing prior to the adoption of the Plan on December 20, 2012.
12

CARB provided the required public notice and opportunity for public comment prior to its January 24, 2013 public hearing on the Plan.
13

The SIP submittal includes proof of publication of notices for these public hearings. We find, therefore, that the 2012 PM
2.5
Plan meets the procedural requirements for public notice and hearing in CAA sections 110(a) and 110(l).

12

See
2012 PM
2.5
Plan, p. ES-6 and SJVAPCD Governing Board Resolution 2012-12-19, “In the Matter of Adopting the San Joaquin Valley Unified Air Pollution Control District 2012 PM
2.5
Plan,” December 20, 2012.

13

See
CARB Resolution 13-2, “San Joaquin Valley PM
2.5
State Implementation Plan,” January 24, 2013.

The District adopted the Supplement after reasonable public notice and hearing.
14

CARB adopted the Supplement for submittal as a SIP revision at its October 24, 2014 Board meeting after reasonable public notice.
15

14

See
Notice of Public Hearing to Adopt Supplemental Document to the
2012 PM
2.5

Plan for the 2006 PM
2.5

Standard
and New Source Review Rule to address Clean Air Act Subpart 4 requirements, available at
http://www.valleyair.org/Air_Quality_Plans/docs/NPH_SD_PM25Plan_09_2014.pdf.

15

See
CARB, Notice of Public Meeting to Consider the Supplemental Document for the San Joaquin Valley 24-Hour PM
2.5
SIP, September 23, 2014, and CARB Board Resolution 14-37, October 24, 2014.

CAA section 110(k)(1)(B) requires the EPA to determine whether a SIP submittal is complete within 60 days of receipt. This section also provides that any plan that the EPA has not affirmatively determined to be complete or incomplete will become complete by operation of law six months after the date of submittal. The EPA's SIP completeness criteria are found in 40 CFR part 51, Appendix V.

The March 4, 2013 submittal of the 2012 PM
2.5
Plan became complete by operation of law on September 4, 2014. We find that the Supplement satisfies the completeness criteria in 40 CFR part

51, appendix V (
see
our Technical Support Document at section I.B).

IV. Review of the San Joaquin Valley 2012 PM
2.5
Plan and Supplement

We summarize our evaluation of the 2012 PM
2.5
Plan and Supplement below. Our detailed evaluation can be found in the Technical Support Document (TSD) for this proposal which is available online at
www.regulations.gov
in docket number EPA-R09-OAR-2014-0636, on EPA Region 9's Web site at
www.epa.gov/region9/air/sjv-pm25/index.html,
or from the EPA contact listed at the beginning of this notice.

A. Emissions Inventory

1. Requirements for Emissions Inventories

CAA section 172(c)(3) requires that each SIP include a “comprehensive, accurate, current inventory of actual emissions from all sources of the relevant pollutant or pollutants in [the] area. . . .” By requiring an accounting of actual emissions from all sources of the relevant pollutants in the area, this section provides for the base year inventory to include all emissions that contribute to the formation of a particular NAAQS pollutant. For the 2006 24-hour PM
2.5
standard, this includes direct PM
2.5
as well as the main chemical precursors to the formation of secondary PM
2.5
: NO
X
, SO
2
, VOC, and ammonia (NH
3
). Primary PM
2.5
includes condensable and filterable particulate matter.

A state should include in its SIP submittal documentation explaining how the emissions data were calculated. In estimating mobile source emissions, a state should use the latest emissions models and planning assumptions available at the time the SIP is developed. California is required to use EMFAC2011 to estimate tailpipe and brake and tire wear emissions of PM
2.5
, NO
X
, SO
2
, and VOC from on-road mobile sources (78 FR 14533, March 6, 2013). States are required to use the EPA's AP-42 road dust method for calculating re-entrained road dust emissions from paved roads (76 FR 6328, February 4, 2011).

In addition to the base year inventory submitted to meet the requirements of CAA section 172(c)(3), the state must also submit future “baseline inventories” for the projected attainment year and each reasonable further progress (RFP) milestone year, and any other year of significance for meeting applicable CAA requirements. By “baseline inventories” (also referred to as “projected baseline inventories”), we mean projected emissions inventories for future years that account for, among other things, the ongoing effects of economic growth and adopted emissions control requirements. The SIP should include documentation explaining how the emissions projections were calculated.

2. Emissions Inventories in the 2012 PM
2.5
Plan

The planning inventories for direct PM
2.5
and all PM
2.5
precursors (NO
X
, SO
2
, VOC, and ammonia) for the SJV PM
2.5
nonattainment area together with documentation for the inventories are found in Appendix B of the 2012 PM
2.5
Plan. Winter daily average inventories, representing conditions in the period November through April, are provided for the base year 2007 and the baseline year of 2012 and each baseline year from 2014 to 2019. A winter inventory is used because exceedances of the 35 μg/m
3
PM
2.5
standard in the SJV occur mostly during the winter months (p. 3-4 and Appendix G, p. G-6). Baseline inventories reflect all control measures adopted prior to January 2012. Growth factors used to project these baseline inventories are derived from data obtained from a number of sources such as the California Energy Commission and Department of Finance as well as studies commissioned by the SJV's metropolitan planning organizations (pp. B-21 to B-30).

Each inventory includes emissions from point, area, on-road, and non-road sources. The inventories use EMFAC2011 for estimating on-road motor vehicle emissions (p. B-26). After EMFAC2011 was released in 2011, new information on statewide diesel fuel usage and economic forecasts became available to the State. For the 2012 PM
2.5
Plan, CARB adjusted EMFAC2011 emissions estimates for heavy-duty trucks to reflect this new information (p. B-26). The EPA allowed the use of these adjustment factors in transportation conformity determinations in the SJV.
16

Re-entrained paved road dust emissions were calculated using the EPA's AP-42 road dust methodology (Appendix B, p. B-25).

16
Letter dated January 14, 2014, Matthew Lakin, Chief, Air Planning Office, EPA-Region 9 to John Taylor, Branch Chief, Transportation Planning Branch, CARB; Subject: Use of San Joaquin Valley Heavy Duty Diesel Vehicle Recession Adjustment Methodology.

Table 1 provides a summary of the winter daily average inventories of direct PM
2.5
and PM
2.5
precursors for the base year of 2007. These inventories provide the basis for the control measure analysis and the RFP and attainment demonstrations in the 2012 PM
2.5
Plan.

Table 1—San Joaquin Valley Emissions Inventory Summary for Direct PM2
.
5 and PM2
.
5 Precursors for the 2007 Base Year

[Winter daily average in tons]

Direct PM2
.
5

NO
X

SO2
VOC
Ammonia

Stationary Sources
9.4
45.6
10.4
96.2
19.8

Area Sources
62.5
19.0
0.8
213.2
342.2

On-Road Mobile Sources
9.1
296.5
0.6
67.3
5.3

Off-Road Mobile Sources
6.1
103.9
1.0
38.0
0.0

Total
87.1
465.1
12.8
414.8
367.3

Source: 2012 PM2
.
5 Plan, Appendix B, Tables B-1 to B-5.

3. Evaluation and Proposed Action

The inventories in the 2012 PM
2.5
Plan are based on the most current and accurate information available to the State and District at the time the Plan and its inventories were being developed in 2011 and 2012, including the latest EPA-approved version of California's mobile source emissions model, EMFAC2011. The inventories comprehensively address all source categories in the SJV and were developed consistent with the EPA's inventory guidance. For these reasons, we are proposing to approve the 2007 base year emissions inventory in the 2012 PM
2.5
Plan as meeting the requirements of CAA section 172(c)(3).

We are also proposing to find that the baseline inventories in the Plan provide an adequate basis for the RACM, RFP, and impracticability demonstrations.

B. Air Quality Modeling

1. Requirements for Air Quality Modeling

CAA section 189(a)(1)(B) requires each state in which a Moderate area is located to submit a plan that includes a demonstration either (i) that the plan will provide for attainment by the applicable attainment date, or (ii) that attainment by that date is impracticable. The 2012 PM
2.5
Plan and Supplement include a demonstration that attainment by the Moderate area attainment date is impracticable.

Air quality modeling is used to establish emissions attainment targets, the combination of emissions of PM
2.5
and PM
2.5
precursors that the area can accommodate and still attain the standard, and to assess whether the proposed control strategy will result in attainment of the standard. Air quality modeling is performed for a base year and compared to air quality monitoring data collected during that year in order to determine model performance. Once the model performance is determined to be acceptable, future year changes to the emissions inventory are simulated with the model to determine the relationship between emissions reductions and changes in ambient air quality. To project future design values, the model response to emission reductions, in the form of Relative Response Factors (RRFs), is applied to monitored design values from the base year.

For demonstrating attainment, the EPA's recommendations for model input preparation, model performance evaluation, use of the model output for the attainment demonstration, and modeling documentation are described in
Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM
2.5,

and Regional Haze
, EPA-454/B-07-002, April 2007 (“Modeling Guidance”), as amended by “Update to the 24 Hour PM
2.5
NAAQS Modeled Attainment Test,” Memorandum dated June 28, 2011, from Tyler Fox, Air Quality Modeling Group, OAQPS, EPA to Regional Air Program Managers, EPA (“Modeling Guidance Update”).
17

The EPA has not issued modeling guidance specific to impracticability demonstrations but believes that a state seeking to make such a demonstration generally should provide air quality modeling similar to that required for an attainment demonstration. The main difference is that for an impracticability demonstration, the model's projected design value on the required attainment date would be above the NAAQS, despite full implementation of the SIP control strategy including all reasonably available control measures (RACM). Alternatively, a model projection could show that the implementation of the SIP control strategy (including RACM) results in attainment of the standard, but that this is achieved only after the applicable attainment date. We are using the latter alternative in evaluating the 2012 PM
2.5
Plan, since its modeling focuses on an attainment year of 2019, instead of 2015, which is the Moderate area attainment year for this area under subpart 4 (CAA section188(c)(1)).
18

17
The EPA Modeling Guidance and Modeling Guidance Update are available on EPA's SCRAM Web site, Web page:
http://www.epa.gov/scram001/guidance_sip.htm;
direct links:
http://www.epa.gov/scram001/guidance/guide/final-03-pm-rh-guidance.pdf
and
http://www.epa.gov/ttn/scram/guidance/guide/Update_to_the_24-hour_PM25_Modeled_Attainment_Test.pdf.

18
The District developed the Plan to address the requirements of subpart 1 as interpreted in the 2007 p.m.2.5 Implementation Rule (prior to the D.C. Circuit's remand of this rule in
NRDC
) which authorized the EPA to extend the attainment date as appropriate for a period no greater than 10 years from the date of designation, considering the severity of nonattainment and the availability and feasibility of pollution control measures (
see
CAA section 172(a)(2)(A) and 40 CFR 51.1004). Because the SJV areas was designated nonattainment effective December 14, 2009 (74 FR 58688 (November 13, 2009), the date “10 years from the date of designation” would be December 14, 2019.

The EPA recommends that states prepare modeling protocols as part of their modeled attainment demonstrations (Guidance, p. 133). The Guidance (at pp. 133-134) describes the topics to be addressed in this modeling protocol. A modeling protocol should detail and formalize the procedures for conducting all phases of the modeling analysis, such as describing the background and objectives, creating a schedule and organizational structure, developing the input data, conducting model performance evaluations, interpreting modeling results, describing procedures for using the model to demonstrate whether proposed strategies are sufficient to attain the applicable standard, and producing documentation to be submitted for EPA Regional Office review and approval prior to actual modeling.

In addition to a modeled attainment demonstration, which focuses on locations with an air quality monitor, EPA's Guidance describes an Unmonitored Area Analysis (UAA). This analysis is intended to ensure that a control strategy leads to reductions in PM
2.5
at other locations that have no monitor but that might have base year and future baseline (projection year) ambient PM
2.5
levels exceeding the standard.

Finally, as discussed below, the Modeling Guidance recommends supplemental air quality analyses. These may be used as part of a Weight of Evidence analysis (WOEA), which assesses attainment by considering evidence other than the main air quality modeling attainment test. While supplemental analyses can increase confidence in the reliability of the modeling, they are less important for evaluating the impracticability demonstration
per se.
That is, the level of rigor in the modeling analyses supporting the Plan's conclusion that attainment will occur by 2019 is less important when the object is to demonstrate that attainment is not practicable by 2015. Supplemental analyses to support a demonstration of attainment by the end of 2019 will be necessary in a new Serious area plan.

2. Air Quality Modeling in the 2012 PM
2.5
Plan

A brief description of the modeling in 2012 PM
2.5
Plan and our evaluation of it follows. More detailed information about the modeling is available in section II.B. of the TSD.

CARB and the District jointly performed the air quality modeling for the 2012 PM
2.5
Plan. The modeling analysis uses the Community Multiscale Air Quality (CMAQ) photochemical model, developed by the EPA.
19

It incorporates routinely available meteorological and air quality data collected during 2007. The MM5 (Mesoscale Model version 5) was used to prepare meteorological input for CMAQ. Air Quality modeling was performed only for the first and fourth quarters (Q1 and Q4) of 2007 which is sufficient for modeling the 2006 24-hour PM
2.5
standard in the SJV because the high 24-hour PM
2.5
concentrations occur only during the colder part of the year. Only the top 10 percent of modeled days is required for projecting the 98th percentile-based design values into the future. (Modeling Guidance Update, p. B-1) The 2012 PM
2.5
Plan's modeling protocol is contained in Appendix F and includes descriptions of the photochemical modeling. Additional

description of the photochemical modeling is also covered in the Weight of Evidence Analysis (”WOEA”).
20

The protocol was reviewed by the EPA and a number of academic experts, and covers all of the topics recommended in the Guidance, including thorough discussions of past modeling results and emission inventory preparation procedures.

19
The 2012 PM
2.5
Plan also included receptor modeling source apportionment analyses, using both the Chemical Mass Balance (CMB) model and the Positive Matrix Factorization (PMF) model. These models distinguish the ambient PM
2.5
contributions of several broad emissions source categories based on how they match the chemical species components of PM
2.5
measurements. These results generally corroborated results from the photochemical modeling, but were not themselves part of the attainment demonstration.

20
The WOEA is Appendix B to the “[CARB] Staff Report, Proposed Revision to the PM
2.5
State Implementation Plan (SIP) for the San Joaquin Valley,” Release Date: January 11, 2013 (“Staff Report”), which can be found in the docket for this proposed rule. Appendix G to the 2012 PM
2.5
Plan also contains a weight of evidence analysis which is identical to the one in the CARB Staff Report except for the two additional appendices 5 and 6.

The air quality modeling and results are summarized in Chapter 4 of the Plan (section 4.5, p. 4-22) and in the WOEA (section 10, p. 62). The Plan's meteorological model and air quality model performance statistics and graphics are available from the CARB Web site (“Meteorology and Air Quality Modeling for the 2012 24-Hour PM
2.5
Plan for the San Joaquin Valley”,
http://arb.ca.gov/eos/SIP_Modeling_PM25/24hr_PM25_ModelingPage.htm
). The air quality model performance appears to be quite good, with bias within the criteria for acceptance, and usually within the original performance goals; performance is very good for total PM
2.5
and for nitrate, the largest component of PM
2.5
; however, time series plots show that some high PM
2.5
periods were underestimated.

The 2012 PM
2.5
Plan included a UAA in Appendix 6 of the Weight of Evidence Analysis in Appendix G of the Plan (p. G-175). (ARB Staff Report Appendix B is identical to Plan Appendix G, except that it does not include the latter's Appendices 5 and 6).

3. Evaluation of the Air Quality Modeling in the 2012 PM
2.5
Plan

The modeling showed that existing State and District control measures are not sufficient to attain the 2006 PM
2.5
standard by 2019 at all monitoring sites in the SJV. Modeling of the additional measures in the Plan (additional direct PM
2.5
reductions from residential wood burning and from commercial charbroiling) showed attainment at all sites by 2019.
Id.

Given the extensive discussion of modeling procedures, tests, and performance analyses called for in the Modeling Protocol and the good model performance, the EPA finds that the modeling is adequate for purposes of supporting the RACM demonstration, the RFP demonstration, and the demonstration of impracticability in the 2012 PM
2.5
Plan and Supplement. To satisfy the statutory requirements for a serious area attainment demonstration, however, the State will need to address documentation gaps outlined in the TSD (section II.B. of the TSD).

While the State included a UAA in the Plan, it makes no difference for the impracticability demonstration we are concerned with here. Any unmonitored peaks with concentrations higher than at the monitors would merely strengthen the case for attainment being impracticable by the required date. A demonstration that attainment is impracticable at monitor locations is sufficient to meet the requirements of section 189(a)(1)(B)(ii). The EPA finds that the supplemental analyses presented in the WOEA are useful in a weight of evidence analysis, and support the demonstration of the impracticability of attainment by 2015.

We note finally that existing ambient air quality monitoring data also support the modeled demonstration that attainment of the 2006 PM
2.5
standard by December 31, 2015 is impracticable. Compliance with the 2006 24-hour PM
2.5
standard is assessed using the three-year average of the yearly 98th percentile concentrations. The most recent monitored PM
2.5
concentrations show that compliance with the 2006 standard is not possible by the end of 2015.
See
discussion in section II.F.3 of the TSD and section V.A. of this proposal.

C. PM
2.5
Precursors

1. Requirements for the Control of PM
2.5
Precursors

The composition of PM
2.5
is complex and highly variable due in part to the large contribution of secondary PM
2.5
to total fine particle mass in most locations, and to the complexity of secondary particle formation processes. A large number of possible chemical reactions, often non-linear in nature, can convert gaseous SO
2
, NO
X
, VOC and ammonia to PM
2.5
, making them precursors to PM
2.5
.
21

Formation of secondary PM
2.5
may also depend on atmospheric conditions, including solar radiation, temperature, and relative humidity, and the interactions of precursors with preexisting particles and with cloud or fog droplets.
22

21
EPA, Air Quality Criteria for Particulate Matter (EPA/600/P-99/002aF, October 2004), Chapter 3.

22
EPA, Regulatory Impact Analysis for the Final Revisions to the National Ambient Air Quality Standards for Particulate Matter (EPA-452/R-12-005, December 2012), p. 2-1.

The 2007 PM
2.5
Implementation Rule contained rebuttable presumptions concerning the four PM
2.5
precursors applicable to attainment plans and control measures related to those plans.
See
40 CFR 51.1002(c). Although the rule included presumptions that states should address SO
2
and NOx emissions in their attainment plans, it also included presumptions that regulation of VOCs and ammonia was not necessary. Specifically, in 40 CFR 51.1002(c), the EPA provided, among other things, that a state was “not required to address VOC [and ammonia] as . . . PM
2.5
attainment plan precursor[s] and to evaluate sources of VOC [and ammonia] emissions in the state for control measures,” unless the state or the EPA provided an appropriate technical demonstration showing that emissions from sources of these pollutants “significantly contribute” to PM
2.5
concentrations in the nonattainment area. 40 CFR 51.1002(c)(3), (4); see also 2007 p.m.2.5 Implementation Rule, 72 FR 20586 at 20589-97 (April 25, 2007).

In
NRDC,
however, the D.C. Circuit remanded the EPA's 2007 PM
2.5
Implementation Rule in its entirety, including the presumptions concerning VOC and ammonia in 40 CFR 51.1002.
See NRDC
v.
EPA,
706 F.3d 428 (D.C. Cir. 2013). Although the court expressly declined to decide the specific challenge to these presumptions (
see
706 F.3d at 437, n. 10 (D.C. Cir. 2013)), the court cited CAA section 189(e)
23

to support its observation that “[a]mmonia is a precursor to fine particulate matter, making it a precursor to both PM
2.5
and PM
10
” and that “[f]or a PM
10
nonattainment area governed by subpart 4, a precursor is presumptively regulated.” 706 F.3d at 436, n. 7 (
citing
CAA section 189(e)). Consistent with the
NRDC
decision, EPA now interprets the Act to require that under subpart 4, a state must evaluate all PM
2.5
precursors for regulation unless it provides a demonstration adequate to rebut the presumption for a particular precursor in a particular nonattainment area.

23
Section 189(e) of the CAA states that “[t]he control requirements applicable under plans in effect under this part for major stationary sources of PM
10
shall also apply to major stationary sources of PM
10
precursors, except where the Administrator determines that such sources do not contribute significantly to PM
10
levels which exceed the standard in the area.”

The provisions of subpart 4 do not define the term “precursor” for purposes of PM
2.5
, nor do they explicitly require the control of any specifically identified particulate matter (PM) precursor. The statutory definition of “air pollutant,” however, provides that the term “includes any precursors to the formation of any air pollutant, to the

extent the Administrator has identified such precursor or precursors for the particular purpose for which the term `air pollutant' is used.” CAA section 302(g). The EPA has identified SO
2
, NO
X
, VOC, and ammonia as precursors to the formation of PM
2.5
. Accordingly, the attainment plan requirements of subpart 4 presumptively apply to emissions of all four precursor pollutants and direct PM
2.5
from all types of stationary, area, and mobile sources, except as otherwise provided in the Act (
e.g.
CAA section 189(e)).

Section 189(e) of the Act requires that the control requirements for major stationary sources of direct PM
10
also apply to major stationary sources of PM
10
precursors, except where the Administrator determines that such sources do not contribute significantly to PM
10
levels that exceed the standard in the area. Section 189(e) contains the only express exception to the control requirements under subpart 4 (
e.g.,
requirements for RACM and RACT, best available control measures (BACM) and best available control technology (BACT), most stringent measures, and NSR) for sources of direct PM
2.5
and PM
2.5
precursor emissions. Although section 189(e) explicitly addresses only major stationary sources, the EPA interprets the Act as authorizing it also to determine, under appropriate circumstances, that regulation of specific PM
2.5
precursors from other source categories in a given nonattainment area is not necessary. For example, under the EPA's longstanding interpretation of the control requirements that apply to stationary, area, and mobile sources of PM
10
precursors area-wide under CAA section 172(c)(1) and subpart 4 (
see
General Preamble, 57 FR 13498 at 13539-42), a state may demonstrate in a SIP submittal that control of a certain precursor pollutant is not necessary in light of its insignificant contribution to ambient PM
10
levels in the nonattainment area.
24

24
Courts have upheld this approach to the requirements of subpart 4 for PM
10
.
See, e.g., Assoc. of Irritated Residents
v.
EPA, et al.,
423 F.3d 989 (9th Cir. 2005).

We are evaluating the SJV PM
2.5
Plan in accordance with the presumption embodied within subpart 4 that all PM
2.5
precursors must be addressed in the state's evaluation of potential control measures, unless the state adequately demonstrates that emissions of a particular precursor do not contribute significantly to ambient PM
2.5
levels that exceed the PM
2.5
NAAQs in the nonattainment area. In reviewing any determination by the state to exclude a PM
2.5
precursor from the required evaluation of potential control measures, we consider both the magnitude of the precursor's contribution to ambient PM
2.5
concentrations in the nonattainment area and the sensitivity of ambient PM
2.5
concentrations in the area to reductions in emissions of that precursor.

2. Evaluation of Precursors in SJV PM
2.5
Plan and Supplement

In the 2012 PM
2.5
Plan, the State and District identify NO
X
, and SO
X
as the precursors that it must control in order to attain the 2006 PM
2.5
standard in the San Joaquin Valley within 10 years of the area's designation as nonattainment for the PM
2.5
NAAQS (
i.e.,
by 2019).
25

Although no technical demonstration is necessary to support a conclusion consistent with the regulatory presumptions under subpart 4, the Plan nevertheless provides supporting evidence describing the need for NO
X
and SO
X
controls. The Plan states that further reductions in VOC and ammonia emissions would not contribute to attainment of the PM
2.5
standard in the area and provides analyses to support this position. In the following, we discuss the technical basis that the District provided in the Plan to support its positions with respect to SO
2,
NO
X
, VOC, and ammonia.

25
This identification is made in the 2012 PM
2.5
Plan, Appendix G, “PM
2.5
Weight of Evidence Analysis” (“WOEA”) at pp. iv and 66, and in the CARB, “Staff Report, Proposed Revision to the PM
2.5
State Implementation Plan (SIP) for the San Joaquin Valley,” Release Date: January 11, 2013 (“Staff Report”) at p. 9, which can be found in the docket for this proposed rule. The WOEA is also included as Appendix B to the Staff Report.

a. SO
2

The 2012 PM
2.5
Plan recognizes that emissions of SO
2
contribute significantly to ambient PM
2.5
levels in the San Joaquin Valley, and that ambient PM
2.5
concentrations are sensitive to reductions in SO
2
.

The 2012 PM
2.5
Plan shows the measured contribution of SO
2
emissions to ambient PM
2.5
concentrations in pie charts portraying the contribution of various pollutant species. Depending on time period and location, the ammonium sulfate contribution is 6 to 9 percent,
26

and the corresponding contribution of just the sulfate part of the ammonium sulfate molecules is 4 to 7 percent of ambient PM
2.5
. The Plan's “Weight of Evidence Analysis” (“WOEA”) also gives the ammonium sulfate portion of the 2007 design value concentration, 4.7 μg/m
3
at Bakersfield.
27

The corresponding sulfate-only portion is 3.4 μg/m
3
. These contribution levels are substantial, although smaller than the contributions of some other components.

26
2009-2010 peak day average in 2012 PM
2.5
Plan Appendix A, Figures A-29 and A-30, p. A-52; 2009-2011 peak day average in WOEA Figure 7, p. 10.

27
WOEA at Table 5, p. 63.

Ambient PM
2.5
sensitivity to reductions of SO
2
emissions is also presented in the Plan in the form of modeling results. The results from the sensitivity modeling is cited and discussed below in the NO
X
subsection. The SO
2
-specific results are that a 25% reduction in Valley-wide SO
2
emissions would result in a 0.18 μg/m
3
decrease in ambient PM
2.5
concentrations at the design value monitoring site, Bakersfield-California. It can also be inferred from the modeling that there is an ambient PM
2.5
decrease of 0.08 μg/m
3
per ton of SO
2
reduction (WOEA, Tables 6 and 7, p. 65). The 0.18 μg/m
3
PM
2.5
decrease for a 25% SO
2
reduction is considerably lower than the 3.75 μg/m
3
decrease that would result from a 25% NO
X
reduction, but the 0.08 μg/m
3
PM
2.5
decrease per ton of emissions reduction is the same for SO
2
as it is for NO
X
. The reason the 25% NO
X
reduction provides a larger reduction in ambient PM
2.5
levels than a 25% SO
2
reduction is simply that the NO
X
emission inventory for the area is much larger than the SO
2
inventory. The 2007 winter planning inventory for SO
2
is just 12.8 tpd, whereas for NO
X
it is 465.1 tpd, more than 35 times larger (
see
2012 PM
2.5
Plan at Appendix B, Emission Inventory, grand totals in tables B-3 for SO
2
and B-2 for NO
X
.) Even though the relatively small SO
2
contribution to ambient PM
2.5
concentrations leaves little scope for reductions, the sensitivity of ambient PM
2.5
to SO
2
emission reductions indicates that SO
2
emissions contribute significantly to PM
2.5
levels above the standards.

Based on the technical analyses provided in the Plan, the EPA agrees with the State's and District's conclusion that SO
2
controls must be included in the evaluation of potential control measures for the 2006 PM
2.5
standard in the SJV, consistent with the requirements of subpart 4.

b. NO
X

The 2012 PM
2.5
Plan recognizes that emissions of NO
X
contribute significantly to ambient PM
2.5
levels in the San Joaquin Valley, and that ambient PM
2.5
concentrations are sensitive to reductions in NO
X
. The Plan discusses NO
X
in conjunction with ammonia, because these precursors react together to create ammonium

nitrate, the largest component of ambient PM
2.5
particles by species in the SJV.
28

The chemical products of ammonia and NO
X
(ammonium and nitrate) combine in a 1:1 molecular ratio, but as discussed below, this ratio does not mean that emissions controls for the two precursor pollutants would be equally effective at reducing ambient PM
2.5
. The Plan provides several lines of evidence to indicate that reductions in NO
X
emissions are effective in reducing PM
2.5
concentrations exceeding the standard, and also that they are more effective than reductions in ammonia emissions. The evidence includes ambient contributions, model simulations of NO
X
emission reductions, historical trends, and the relative amounts of NO
X
and ammonia.

28
2012 PM
2.5
Plan Appendix A, Figures A-30 and A-29, p. A-52.

The Plan indicates that the ambient contribution of NO
X
to PM
2.5
levels in the SJV is substantial. Ammonium nitrate is the largest chemical component of ambient PM
2.5
in the SJV, comprising 65 percent of the 2009-2011 average peak 24-hour PM
2.5
concentration at Bakersfield (WOEA, p. 10). Using the 2009-2011 24-hour PM
2.5
design value of 61.8 μg/m
3
,
29

the ammonium nitrate concentration on peak PM
2.5
days is approximately 40.2 μg/m
3
. If only nitrate itself is considered (
i.e.,
the nitrate part of the ammonium nitrate molecules), the contribution of NO
X
represents approximately 50.3 percent of the 2009-2011 average peak 24-hour PM
2.5
concentration at Bakersfield, which is an ambient contribution of 31.1 μg/m
3
.
30

Whether considered as ammonium nitrate or simply as nitrate, NO
X
is clearly a significant contributor to ambient PM
2.5
levels above the standard in the SJV.

29

See
2012 PM
2.5
Plan, p. A-10. The design value for Bakersfield-California for 2009-2011 is given as a rounded value of 62 μg/m
3
in Table A-5 in Appendix A of the Plan. For greater precision in estimating species contributions, we have used the unrounded value of 61.8 μg/m
3
calculated as the average of the 98th percentiles values for each year (66.7, 53.3, and 65.5) as listed in Table A-4.

30
The nitrate fraction of ammonia nitrate is calculated as molecular weight of nitrate (62) divided by the molecular weight of ammonium nitrate (80) and equals 77.5 percent.

In addition to this evidence on the contribution of NO
X
to PM
2.5
concentrations, the Plan provides evidence that ambient PM
2.5
concentrations are sensitive to NO
X
reductions (
i.e.,
nitrate PM
2.5
concentrations go down when NO
X
emissions are reduced). The evidence is from modeling, historical trends, and relative proportions of NO
X
and ammonia. The 2012 PM
2.5
Plan provides evidence from past and current photochemical modeling simulations that ambient ammonium nitrate is sensitive to NO
X
reductions. The Plan describes past modeling studies that were documented in academic journals.
31

In the various studies, when NO
X
emissions were reduced by 50 percent, ambient ammonium nitrate decreased by 25 to 50 percent, depending on the episode modeled and the geographic location.
32

Modeling for the 2012 PM
2.5
Plan also shows substantial sensitivity of ambient PM
2.5
concentrations to reductions in NO
X
emissions. In the Plan, the State modeled a 25 percent reduction in NO
X
emissions and showed a decrease in Bakersfield PM
2.5
concentrations of 3.75 μg/m
3
, a 6 percent decrease in the 2009-2011 design value of 61.8 μg/m
3
and similar levels of ambient decreases at other monitors (WOEA, Table 6, p. 65).

31
The academic journal papers and are described in Appendix F, section 2.7 (p. 28), and in WOEA, section 5.c (p. 64).

32
Kleeman, M.J., Ying, Q., and Kaduwela, A., Control strategies for the reduction of airborne particulate nitrate in California's San Joaquin Valley, Atmospheric Environment, 2005, 39, 5325-5341. Liang, J., Gürer, K., Allen, P.D., Zhang, K.M., Ying, Q., Kleeman, M., Wexler, A., and Kaduwela, A., 2006, A photochemical model investigation of an extended winter PM episode observed in Central California: Model Performance Evaluation, Proceedings of the 5th Annual CMAQ Models-3 User's Conference, Chapel Hill, NC. Pun, B.K., Balmori R.T.F, and Seigneur, C., 2009, Modeling wintertime particulate matter formation in Central California, Atmospheric Environment, 43, 402-409. Different models and emission inventories in these studies conducted over the years also contribute to the variation in results.

The 2012 PM
2.5
Plan provides additional (non-modeling) evidence on the effectiveness of NO
X
reductions. The historical downward trends of NO
X
emissions and of ambient nitrate concentrations are discussed in Chapter 4 and the WOEA of the Plan.
33

Daily NO
X
emissions levels and winter nitrate concentrations appear correlated over time on an annual basis. Both have decreased by about a third during the period 2004 to 2011. This is evidence that existing NO
X
controls are effective at reducing ammonium nitrate. The evidence is strengthened by the fact that this reduction in ambient nitrate occurred despite an increase in emissions of ammonia, the other precursor to ammonium nitrate, during the same period (Plan p. 4-8).

33
2012 PM
2.5
Plan, Chapter 4, p. 4-8 and Figure 4-7, p. 4-10; more detail in WOEA section 9, p. 58, Figures 49-52.

The Plan further describes the effectiveness of NO
X
controls by characterizing it as the “limiting precursor” in ammonium nitrate formation, based on the relative amounts of NO
X
and ammonia. Based on monitored concentrations and the emissions inventory, the Plan concludes that NO
X
is the limiting precursor. The limiting precursor concept is illustrated briefly in Chapter 4 and described more fully in the WOEA.
34

One molecule of each of NO
X
and ammonia is required to form each molecule of ammonium nitrate. If NO
X
is in short supply relative to ammonia, then NO
X
is the limiting factor in ammonium nitrate formation.
35

34
Chapter 4, Figures 4-4 and 4-5 (Plan, p. 4-9); WOEA, section 5b, p. 16. See also CARB Staff Report, p. E-3.

35
As noted below in the ammonia subsection, the “limiting precursor” concept is not absolute, and must be used with caution. However, for NO
X
it does support evidence from the modeling results that NO
X
significantly contributes to PM
2.5
exceedances.

The WOEA includes plots (Figures 16 and 17, p. 19) of ammonia and nitric acid (which contains nitrate) concentrations at two monitoring sites in the SJV (Angiola and Fresno) that were measured during the winter 2000-2001 CRPAQS
36

study and reported in Lurmann
et al.
(2006).
37

The Plan notes that ammonia concentrations are at least an order of magnitude larger than those of nitrate and notes Lurmann
et al.'
s conclusion that NO
X
is the limiting precursor.

36
CRPAQS is the California Regional Particulate Air Quality Study. More information is available about CRPAQS at
http://www.arb.ca.gov/airways/ccaqs.htm.

37
Lurmann, F.W., Brown, S.G., McCarthy, M.C., and Roberts, P.T., December 2006, Processes Influencing Secondary Aerosol Formation in the San Joaquin Valley during Winter, Journal of Air and Waste Management Association, 56, 1679-1693.

The WOEA also considers emissions inventories to support the argument that NO
X
is the limiting precursor. The WOEA normalized NO
X
emissions using the relative molecular weights of NO
X
and ammonia, in order to reflect the number of molecules of each available to react with each other (p. 18, Table 1). In 2000, the amount of NO
X
available was only about two-thirds the amount of ammonia; in 2011 NO
X
was only one-third of ammonia. This shows the scarcity of NO
X
relative to ammonia and implies that NO
X
is the limiting precursor in the formation of ammonium nitrate.

Based on the range of technical analyses provided in the Plan and other information available to EPA, the EPA agrees with the State's and District's conclusion that NO
X
controls must be included in the evaluation of potential control measures for the 2006 PM
2.5
standard in the SJV, consistent with the requirements of subpart 4.

c. Ammonia

The 2012 PM
2.5
Plan and Supplement state that “[b]ecause of the regional

surplus in ammonia, even substantial ammonia emissions reductions yield a relatively small reduction in nitrate” (Plan p. 4-8) and that “ammonia reductions would not significantly contribute to the Valley's attainment of the 2006 PM
2.5
standard” (Plan p. 4-11). To support this finding, the Plan and Supplement discuss the ambient contribution of ammonia to measured PM
2.5
levels in the SJV, and the sensitivity of ambient PM
2.5
to ammonia reductions. The latter includes discussion of the relative abundance of NO
X
and ammonia, and of modeled simulations of further reductions in ammonia emissions.

The Plan indicates that ammonia contributes to ambient concentrations of PM
2.5
, in the form of ammonium nitrate and ammonium sulfate. As noted above in the NO
X
discussion, ammonium nitrate contributes 65 percent of the 2009-2011 average peak PM
2.5
ambient levels at Bakersfield. Ammonium sulfate contributes an additional 7 percent (p. G-10; WOEA, p. 10). Thus, ammonium nitrate and ammonium sulfate together account for a total of 72 percent of the peak 24-hour PM
2.5
concentrations, or 44.5 μg/m
3
, and ammonia emissions are essential to the formation of both of these components of the ambient particulate matter. If only the ammonium portion of these molecules is considered, the corresponding figures are 16.6 percent of peak PM
2.5
ambient levels, or 10.3 μg/m
3
.
38

This level of contribution is a substantial fraction of the SJV's 2009-2011 design value of 61.8 μg/m
3
, and indicates that emissions of ammonia contribute significantly to ambient PM
2.5
concentrations that exceed the 24-hour PM
2.5
standard in the SJV.

38
The ammonium fraction of ammonia nitrate is calculated as molecular weight of ammonium (18) divided by the molecular weight of ammonium nitrate (80) and equals 23.5 percent. The ammonium fraction of ammonia sulfate is calculated as molecular weight of ammonium portion (36) divided by the molecular weight of ammonium sulfate (132) and equals 27.3 percent.

Next we examine information in the Plan regarding the sensitivity of ambient PM
2.5
levels in the SJV to ammonia emission control. On this issue there is conflicting evidence. Based on evidence that ammonia is not the limiting precursor and that modeled ammonia reductions are relatively ineffective, the Plan concludes that controls for ammonia are not warranted. However, a detailed evaluation of the modeling shows that ammonia controls can be effective at reducing ambient PM
2.5
in some locations.

The Plan's evidence discussed above to support the argument that NO
X
is the limiting precursor for ammonia nitrate formation is also presented as evidence that ammonia is not the limiting precursor, and so ambient PM
2.5
would not be sensitive to ammonia reductions (WOEA, p. 16-20). The Plan notes that there is both an abundance of ambient ammonia relative to ambient nitrate, and an abundance of ammonia emissions relative to NO
X
emissions. The Plan also indicates that there is an abundance of gaseous ammonia relative to particulate ammonium at multiple locations during the 2000-2001 winter episode in the CRPAQS study (WOEA, p. 20 and Figure 18). This abundance suggests that even under conditions favorable to ammonium nitrate formation, a substantial amount of unreacted ammonia remains. Based on these multiple pieces of evidence on the abundance of ammonia, the Plan concludes that ammonia is not the limiting factor for ammonium nitrate formation and, thus, that reducing ammonia emissions would not reduce ambient PM
2.5
in the SJV.

The 2012 PM
2.5
Plan also considered air quality modeling analyses to evaluate the effectiveness of reducing ammonia as compared to other precursors, and to PM
2.5
decreases needed for attainment. Modeling for the 2012 PM
2.5
Plan indicated that a 25 percent reduction in ammonia emissions resulted in a 0.55 μg/m
3
decrease in ambient PM
2.5
(WOEA, p. 65, Table 6). This benefit is roughly one-seventh the corresponding benefit for a 25 percent reduction in NO
X
.
Id.
Restating the inventory reduction percentages in terms of tons per day reductions, the Plan notes that reducing ammonia emissions by one ton per day is only about 10 percent as effective as reducing one ton per day of NO
X
(Plan p. 4-11). Thus, based on this air quality modeling, the Plan concludes that additional ammonia control is considerably less effective than NO
X
control.

The Plan also notes that, assuming the same rate of improvement in ambient PM
2.5
concentration per ton of ammonia reduced, it would take a 34 percent reduction in ammonia emissions to decrease ambient PM
2.5
by 1 μg/m
3
, the amount that would have been needed to advance projected attainment by one year from 2019 to 2018. The Plan considers this to be “an infeasible amount, since there are no control strategies that exist or have been identified which could achieve such large reductions” (Plan, p. 4-11).

The Plan assumes that additional ammonia control, as modeled, would provide limited benefit for attainment planning purposes. The Plan concludes, based upon the various information and analyses described above, that “ammonia reductions would not significantly contribute to the Valley's attainment of the 2006 PM
2.5
standard” (Plan p. 4-11), and therefore additional control measures should not be evaluated.

After reviewing the information discussed above, EPA believes that the information provided by the State and District in the Plan and Supplement shows that ammonia contributes to a large fraction of measured PM
2.5
concentrations in the SJV area, in the form of ammonium nitrate and ammonium sulfate. Additionally, modeling analyses submitted by the State and studies available to EPA indicate that although ammonia control is generally less effective at reducing PM
2.5
concentrations when compared to NO
X
control, it remains true that reducing ammonia emissions in the SJV would reduce PM
2.5
by varying amounts throughout the nonattainment area. Moreover, reductions in ammonia in conjunction with reductions of SO
2
and NO
X
would help to provide for attainment of the NAAQS in the SJV.

While EPA finds the modeling and other analyses presented in the 2012 PM
2.5
plan to be credible, the fact remains that the modeling analyses show that additional reductions in ammonia may reduce ambient PM
2.5
levels to varying degrees. In the various studies, when ammonia emissions were reduced by up to 50 percent, ambient ammonium nitrate decreased by 5 to 25 percent, depending on the episode modeled and the geographic location evaluated. (WOEA, p. 64) These percentages for ammonia benefits are generally smaller than those for NO
X
reductions, but these modeling results show that reductions in ammonia emissions under certain circumstances can effectively reduce ambient PM
2.5
. The fact that all the modeling studies, including the modeling done for the current Plan, find at least some benefit from ammonia control shows that the concept of a “limiting precursor” discussed above is not absolute. In addition, the test for determining whether emission reduction measures for a particular precursor must be evaluated for purposes of timely attainment should not be exclusively based on the control effectiveness of the precursor relative to other precursors, but should also consider whether emissions of the precursor “contribute significantly” to ambient PM
2.5
levels which exceed the PM
2.5
standard in the nonattainment area.

Regarding the Plan's statement that it would take a 34 percent reduction in

ammonia emissions to decrease ambient PM
2.5
by 1 μg/m
3
, the amount needed to advance attainment by one year from 2019 to 2018, EPA notes that the test for advancing the attainment date is based not on an evaluation of control measures for a single pollutant but rather on an evaluation of potential control measures for direct PM
2.5
and all PM
2.5
precursors from all types of sources in the nonattainment area. We also note that the appropriate inquiry in this context is whether reasonably available control measures would advance attainment by one year from 2015 to 2014 (not from 2019 to 2018), given under subpart 4 the applicable attainment date for the SJV area for the 2006 PM
2.5
NAAQS is December 31, 2015.

In summary, the information provided by the State and District in the Plan and Supplement shows that ammonia contributes to a large fraction of measured PM
2.5
concentrations in the SJV area, in the form of ammonium nitrate and, to a lesser extent, ammonium sulfate. Additionally, modeled evidence submitted by the State and studies available to EPA indicate that although ammonia control is less effective at reducing PM
2.5
concentrations compared to NO
X
control, reducing ammonia emissions in the SJV would reduce PM
2.5
by some amount in parts of the Valley. Given the severity of the PM
2.5
nonattainment problem in the SJV, the demonstration that attainment in SJV by 2015 is impracticable, and the documentation in the 2012 PM
2.5
Plan showing that ammonia emissions are responsible for more than 10 μg/m
3
of ambient PM
2.5
concentrations in the SJV area, EPA does not agree with the statement in the Plan that additional ammonia controls need not be evaluated to satisfy CAA requirements for the 2006 PM
2.5
standard in the SJV.

While stating that additional ammonia reductions are not necessary, the State nevertheless provided in the Supplement an evaluation of ammonia control measures currently implemented in the SJV and other potential ammonia control measures. We discuss the State's ammonia control evaluation in section D, below (“Reasonably Available Control Measures/Reasonably Available Control Technology”).

d. VOC

The 2012 PM
2.5
Plan states that further reductions in VOC emissions would not contribute to PM
2.5
attainment in the San Joaquin Valley. To support this finding, the Plan discusses the ambient contribution of VOC to measured PM
2.5
levels in the Valley, the indirect role of VOC in ammonium nitrate formation, and modeled simulations of further reductions in VOC emissions.

There are two routes by which VOC can contribute to ambient PM
2.5
. The first is through various chemical reactions leading to the formation of Secondary Organic Aerosols (SOA). The second is through photochemical reactions that create oxidants such as ozone and the hydroxyl radical (OH), which in turn oxidize NO
X
emissions to nitrate, leading to the formation of particulate ammonium nitrate. The Plan's Chapter 4 (p. 4-12) discussed both routes in section 4.4.2. The WOEA discusses SOA in section 6 (WOEA, p. 32) and the oxidant route in section 5d (WOEA, p. 24).

For the direct contribution of VOC to PM
2.5
, SOA, the 2012 PM
2.5
Plan states that modeling for annual average PM
2.5
for the 2008 PM
2.5
Plan found that anthropogenic SOA were about 3 to 5 percent of total organic aerosol, and that SOA were mainly formed during the summer from nonanthropogenic sources. The winter anthropogenic contribution that is of interest for the 24-hour PM
2.5
standard would necessarily be lower, since SOA formation is less at winter temperatures, which are lower than the annual average. The Plan also cites a study by Chen
et al.

39

for the winter 2000-2001 CRPAQS episode. This study found that the SOA portion of total organic aerosol had a maximum value of 4.26 μg/m
3
with concentrations at Fresno and Bakersfield of 2.46 and 2.28 μg/m
3
, respectively, which represent 6 percent and 4 percent of the total organic aerosol at those locations. Applying this roughly 5 percent SOA proportion to the organic carbon portion of the measured 2009-2011 winter PM
2.5
composition shows that SOA is around 0.9 percent of total ambient PM
2.5
at Bakersfield, and 1.5 percent of ambient PM
2.5
at Fresno.
40

Because anthropogenic SOA is only a portion of the total, the portion due to controllable anthropogenic sources would be even less. These modeling studies show that SOA is not a substantial component of ambient PM
2.5
in the SJV and that the potential for reducing ambient PM
2.5
through VOC emission reductions is very limited.

39
Chen, J., Ying, Q., and Kleeman, M.J., 2010, Source apportionment of wintertime secondary organic aerosol during the California regional PM
10
/PM
2.5
air quality study, Atmospheric Environment, 44(10), 1331-1340.

40
The contribution of Organic Matter to 2009-2011 peak day PM
2.5
levels was 17 percent at Bakersfield and 29 percent at Fresno (
see
Staff Report, Appendix B, p. 10 [pdf.52], Figure 6. Five percent of these gives 0.85 percent SOA at Bakersfield and 1.45 percent at Fresno. As a fraction of the 2007 design values of 66 μg/m
3
at Bakersfield and 63 μg/m
3
at Fresno, these give SOA contributions of 0.56 and 0.91 μg/m
3
at Bakersfield and Fresno, respectively.

For the indirect contribution of VOC to PM
2.5
, nitrate formation via daytime photochemistry, it appears that this route is also not a substantial contributor, based on modeled sensitivity to VOC reductions. For one such study there were relatively low modeled concentrations of ozone, which did not appear consistent with nitrate formation via daytime oxidant (ozone) photochemistry, which would be expected to have elevated ozone levels.
41

The Plan reviews the same studies that the State relied on in the 2008 PM
2.5
Plan for attainment of the 1997 PM
2.5
standards (Plan Modeling Protocol, p. F-31). The EPA's review of these studies and of the 2008 PM
2.5
Plan's examination of them is covered in the TSD for the EPA's final action on the 2008 PM
2.5
Plan.
42

The findings from those reviews remain the same for the current Plan: past modeling studies vary on whether controlling VOC reduces PM
2.5
, but the most reliable ones show VOC control has little benefit, or even a disbenefit. As detailed in that previous TSD and in the Plan's Modeling Protocol, the studies for which VOC control showed a benefit at some times and places are less reliable because they used unrealistic emissions levels, unrealistic control scenarios, or the effect occurred at PM
2.5
concentrations no longer reached in the SJV. The Protocol also suggested that there is sufficient background ozone flowing in from outside the SJV area to perform the oxidizing role, even in the absence of oxidants generated from VOC emissions within the SJV,
43

implying that VOC reductions would have little effect on ambient PM
2.5
levels exceeding the standard in the SJV. The overall conclusion is that the effect of reducing VOC emissions is somewhat uncertain, but in general produces little benefit or

even a disbenefit in PM
2.5
concentrations.

41
Pun, B.K., Balmori R.T.F, and Seigneur, C., 2009, Modeling Wintertime Particulate Matter Formation in Central California, Atmospheric Environment, 43: 402-409. doi: 10.1016/j.atmosenv.2008.08.040.

42
EPA-Region 9, Technical Support Document and Responses to Comments Final Rule on the San Joaquin Valley 2008 PM
2.5
State Implementation Plan,” September 30, 2011, section II.C.

43
Kleeman, M.K., Ying, Q., and Kaduwela, A., 2005, Control strategies for the reduction of airborne particulate nitrate in California's San Joaquin Valley, Atmospheric Environment, 39: 5325-5341 September 2005. doi:10.1016/j.atmosenv.2005.05.044; cited in Plan Modeling Protocol. p. F-36).

Added to this evidence from past studies is the 2012 Plan's current modeling, which indicates a disbenefit from controlling VOC at important geographic locations. This is shown by negative PM
2.5
sensitivities (that is, decreased VOC emissions result in increased PM
2.5
levels) for multiple locations.
44

In addition, diagrams of model PM
2.5
response at Bakersfield to various combinations of NO
X
and VOC reductions show graphically that VOC reductions increase PM
2.5
, for any given level of NO
X
.
45

For all the monitoring sites, these NO
X
vs. VOC diagrams show either no decrease or an actual increase in PM
2.5
as VOC emissions are reduced. The current modeling provides strong evidence that additional VOC reductions would not decrease ambient PM
2.5
concentrations in the SJV.

44
See VOC columns of WOEA Tables 6 and 7 (p. 65) for Bakersfield, Visalia, Corcoran, and Merced.

45
WOEA Figure 54 (p. 67) shows the model PM
2.5
response at Bakersfield to reductions in various combinations of precursors. Subfigure “b)” shows NO
X
reductions plotted against VOC reductions. For a given level of NO
X
, in decreasing VOC by moving leftward along a horizontal line (representing constant NO
X
), one crosses the lines of constant PM
2.5
(isopleths) into regions of increased PM
2.5.
The Plan presents similar diagrams in Chapter 4, Figures 4-15 through 4-23 (pp. 4-31
ff
) for the various monitoring sites.

In sum, the information provided by the State and District in the Plan and Supplement shows that: (a) wintertime levels of secondary organic aerosol measured in the SJV are low and therefore the direct products of VOC emissions do not directly contribute significantly to PM
2.5
levels above the standard in the SJV; and (b) wintertime reductions in VOC emissions in the SJV, when PM
2.5
concentrations are high, would not reduce ambient PM
2.5
levels, and therefore the indirect products of VOC emissions also do not directly contribute significantly to PM
2.5
levels above the standard in the SJV. Based on this information, we propose to determine that at this time VOC emissions do not contribute significantly to ambient PM
2.5
levels that exceed the 2006 PM
2.5
standard in the SJV nonattainment area.

3. Proposed Action

Based on a review of the information provided by the District and other information available to EPA, we propose to determine that at this time VOC emissions do not contribute significantly to ambient PM
2.5
levels which exceed the 2006 PM
2.5
standard in the SJV and, therefore, that VOCs may be excluded from the State's evaluation of potential control measures for purposes of this standard in this area. Consistent with the regulatory presumptions under subpart 4, all other PM
2.5
precursors (NO
X
, SO
2
, and ammonia) must be included in the State's evaluation of potential control measures for the 2006 PM
2.5
standard in the SJV area, including NNSR provisions to implement the requirements of subpart 4.
46

We discuss the State's evaluation of potential control measures for NO
X
, SO
2
, and ammonia, as well as direct PM
2.5
, in section D (“Reasonably Available Control Measures/Reasonably Available Control Technology”).

46
Absent a demonstration to EPA's satisfaction that major stationary sources of ammonia emissions do not contribute significantly to ambient PM
2.5
levels that exceed the NAAQS in the SJV area, under CAA section 189(e) major stationary sources of ammonia are subject to the control requirements that apply to major stationary sources of direct PM
2.5
, including NNSR requirements. We intend to evaluate the adequacy of SJV's NNSR program for PM
2.5
purposes in a separate rulemaking.

D. Reasonably Available Control Measures/Reasonably Available Control Technology

1. Requirements for RACM/RACT

CAA section 172(c) under subpart 1 describes the general attainment plan requirement for RACM and RACT, requiring that attainment plan submissions “provide for the implementation of all reasonably available control measures as expeditiously as practicable (including such reductions in emissions from existing sources in the area as may be obtained through the adoption, at a minimum, of reasonably available control technology) and shall provide for attainment” of the NAAQS. The attainment planning requirements specific to PM
2.5
under subpart 4 likewise impose upon states an obligation to develop attainment plans that require RACM on sources of direct PM
2.5
and those PM
2.5
precursors determined to be subject to the RACM/RACT requirement. CAA section 189(a)(1)(C) requires that Moderate area PM
2.5
SIPs contain provisions to assure that RACM are implemented by no later than 4 years after designation of the area. The EPA reads CAA sections 172(c)(1) and 189(a)(1)(C) together to require that attainment plans for Moderate nonattainment areas must provide for the implementation of RACM and RACT for existing sources of PM
2.5
and PM
2.5
precursors in the nonattainment area as expeditiously as practicable but no later than 4 years after designation.
47

As part of the RACM/RACT analysis, all available controls should be evaluated, and reasonable controls should be adopted.

47
This interpretation is consistent with guidance provided in the General Preamble at 13540.

The terms RACM and RACT are not specifically defined in the Act, nor do the provisions of subpart 4 specify how states are to meet the RACM and RACT requirements. In longstanding guidance, however, the EPA has interpreted the RACM requirement to include any potential control measure for a point, area, on-road and non-road emission source that is technologically and economically feasible (General Preamble at 13540) and is not “absurd, unenforceable, or impracticable” (General Preamble at 13560). The EPA has historically defined RACT as the lowest emission limitation that a particular stationary source is capable of meeting by the application of control technology (
e.g.,
devices, systems, process modifications, or other apparatus or techniques that reduce air pollution) that is reasonably available considering technological and economic feasibility.
See
General Preamble at 13541 and 57 FR 18070, 18073-74 (April 28, 1992).

The EPA recommends that states use the following process to identify RACM/RACT:

1. Develop a comprehensive list of available control measures for all source categories in the nonattainment area. In developing this list, the state should identify existing control measures currently being implemented in the area and also include any control measures suggested in public comments.

2. Remove from the list any measure that is unreasonable because emissions from the affected source or source category are insignificant (
i.e.,
de minimis). The state should appropriately document any determination that a source or source category is insignificant.

3. Evaluate each remaining available control measure for its “reasonableness,” considering technological and economic feasibility and any potentially adverse impacts. The state should provide a reasoned justification if it rejects a listed control measure as unreasonable.

4. Include in its submitted Moderate area attainment plan, in enforceable form, each control measure found to be reasonable unless the measure is already either federally promulgated, part of the state's SIP, or otherwise creditable in SIPs. In areas that demonstrate attainment by the Moderate area attainment date, the state may justify rejection of an otherwise reasonable measure by demonstrating that the measure would not, individually or collectively with other reasonable measures being rejected, advance attainment by one year or more. For areas that demonstrate that attainment by the Moderate area attainment date is impracticable, the state must provide for the implementation of all available control measures that are technologically and economically feasible.

See
General Preamble at 13540-13544.

An evaluation of technological feasibility should include consideration of factors such as a source's process and operating conditions, raw materials, physical plant layout, and non-air quality and energy impacts (
e.g.,
increased water pollution, waste disposal, and energy requirements) (57 FR 18070, 18073).

An evaluation of economic feasibility should include consideration of factors such as cost per ton of pollution reduced (cost-effectiveness), capital costs, and annualized cost (57 FR 18070, 18074). Absent other indications, the EPA presumes that it is reasonable for similar sources to bear similar costs of emissions reductions. Economic feasibility of RACM and RACT is thus largely informed by evidence that other sources in a source category have in fact applied the control technology, process change, or measure in question in similar circumstances.
Id.

2. RACM/RACT Analysis in the 2012 PM
2.5
Plan

The 2012 PM
2.5
Plan's RACM/RACT evaluation for direct PM
2.5
, NO
X
and SO
2
sources is summarized in section 9.2 (p. 9-3) and detailed in Appendices C (“Mobile Source Control Strategies”) and D (“Stationary and Area Source Control Strategy Evaluation”). RACM for ammonia sources is discussed in Attachment A of the Supplement. The Plan includes a short discussion of the District's current VOC control measures (p. 5-17 and Supplement, p. 13) but includes no detailed evaluation of them. The treatment of VOCs is thus consistent with the District's determination that additional VOC controls are not necessary in the SJV area for purposes of attaining the PM
2.5
NAAQS.

The evaluation of potential controls in the 2012 PM
2.5
Plan is presented by source category. For stationary and area source categories, the evaluation is broken down by the current District rule or rules that apply to that source category. The evaluation also addresses a number of source categories that are not currently subject to emission controls (
e.g.,
fireworks).
See
2012 Plan, Appendix D, p. D-163.

The following information is provided in Appendix D for each stationary or area source category or District rule or, for ammonia sources, in Attachment A to the Supplement:

• A description of the sources within the category or sources subject to the rule;

• base year (2007), current year (2012), and projected baseline year emissions (for every year from 2014 to 2019) in the source category or affected by the rule;

• a discussion of the current rule requirements and/or listing and discussion of existing rules, regulations, or other control efforts that address the source category; and

• identification and discussion of potential new controls, including in many cases, a discussion of the technological and economic feasibility of the new controls. Rules adopted by other agencies (including the EPA, South Coast AQMD, and Bay Area AQMD) are discussed and compared to existing SJV rules. Measures proposed by the public for the source category/rule are also identified and discussed. In addition, non-regulatory approaches to reducing emissions in each stationary and area source category are discussed, including the use of incentives, opportunities for technology advancement programs, policy initiatives, and education/outreach programs.

The following information is provided in Appendix C for each major on- and off-road mobile source category:

• A description of the type of sources in the category;

• base year (2007), current year (2012), and projected baseline year emissions (for every year from 2014 to 2019) in the source category;

• a discussion of the current rule requirements and/or listing and discussion of existing rules, regulations, or other control efforts that address the source category; and

• identification and discussion of potential new controls. Measures proposed by the public for the source category/rule are also identified and discussed with the majority of this discussion presented in the responses to comments in Appendix I, pp. I-10 to I-13.

Through this evaluation process, the District identified 5 new control measures for adoption. The 2012 PM
2.5
plan includes enforceable commitments to adopt these measures.
See
2012 PM
2.5
Plan, section 5.3.1 (p. 5-21) and Table 2 below.

Table 2—SJVAPCD 2012 PM
2.5
Plan Specific Rule Amendment Commitments

District rule No.
Measure number and description
Amend date
Comply date
Rule amendment commitment

4308
Boilers, Steam Generators, and Process Heaters 0.075 to <2 MMBtu/hr
2013
2015

Lower the NO
X
emission limit for instantaneous water heaters in the size range of 0.075-0.4 MMBtu/hr from the current level of 55 ppmv to 20 ppmv.

4692
Commercial Charbroiling
2016
2017
Add requirements for under-fired charbroilers.

1. Lower the threshold level for calling wood-burning curtailments from 30 μg/m
3
to ≥20 μg/m
3
.

2. Review the meteorological conditions that lead to elevated PM
2.5
, to prevent the buildup of PM
2.5
that may lead to a potential exceedance day.

4901
Wood Burning Fireplaces and Wood Burning Heaters
2016
2016/2017
3. Consider expanding the wood burning season to include October and/or March.

4. Analyze the feasibility of allowing the use of the cleanest certified wood burning devices at specified curtailment levels.

4905
Natural Gas-Fired, Fan Type Residential Central Furnaces
2014
2015

Lower the NO
X
emission limits for residential furnaces and examine the possibility of incorporating NO
X
limits for natural gas-fired, fan-type, commercial central furnaces into the rule.

9610
SIP Creditability of Incentives
2013
2013
Establish appropriate mechanisms for the District to take SIP credit for eligible incentive programs.

Source: 2012 PM
2.5
Plan, p. 5-21, Table 5-3, and section 5.3.1.

The District also identified a number of source categories for which existing information is inadequate to determine the feasibility of additional controls. These categories and the additional controls to be studied are discussed in section 5.3.3. (p. 5-24). The schedule for these studies is given in Table 5-4 (p. 5-24).

The Plan also includes descriptions of the District's incentive programs

(Chapter 6); its technology advancement program (Chapter 7), and its public outreach program (Chapter 8).

The efforts by the SJV's eight metropolitan planning organizations (MPO)
48

to implement cost-effect transportation control measures (TCM) are described in Appendix C, section C-11-4 (p. C-33). While no additional TCMs were identified by the MPOs, the Plan includes a discussion of the on-going implementation of a broad range of TCMs in the Valley. There is also a discussion of the MPOs' Congestion Management and Air Quality (CMAQ) funding policy which is a standardized process across the Valley for distributing 20 percent of the CMAQ funds to projects that meet a minimum cost-effectiveness. During the comment period for the 2012 PM
2.5
Plan, a number of TCMs were suggested by the public for consideration.
See
Appendix I, pp. I-10 to I-13. The feasibility of these measures is discussed in the District's responses to comments.
Id.

48
These eight MPOs represent the eight counties in the San Joaquin Valley air basin: the San Joaquin Council of Governments, the Stanislaus Council of Governments, the Merced County Association of Governments, the Madera County Transportation Commission, the Council of Fresno County Governments, Kings County Association of Governments, the Tulare County Association of Governments and the Kern Council of Governments.

The Supplement identifies three categories of ammonia (NH
3
) emission sources in the SJV, which are estimated to account for 96% of ammonia air emissions in the Valley—farming operations, including confined animal facilities (239.2 tpd), evaporation from agricultural fertilizers (66.1 tpd), and composting solid waste operations (20.5 tpd) (
see
Supplement at 11). Information presented in the Supplement then compares District rules for confined animal facility (CAF) and composting operations to analogous requirements in other areas and discusses water quality measures for agricultural fertilizer application that may also reduce air emissions (
see generally
Supplement at Attachment A).

Farming operations:
The Supplement states that the control measures in SIP-approved Rule 4570 (Confined Animal Facilities, adopted 2010)
49

have a secondary benefit of limiting NH
3
emissions and have reduced ammonia emissions by more than 100 tpd (
see
Supplement at A-2, citing Staff Report for June 2009 re-adoption of Rule 4570 (June 2009) at Appendix F). Rule 4570 is a work practice rule that requires farmers to implement management practices (
e.g.,
feed according to NRCS guidelines, remove manure from corrals at least four times per year) for different components of the CAF operation, such as feeding, milking parlors, housing/bedding, manure management and land application. Rule 4570 applies to livestock operations, including dairy, beef, poultry and swine, above certain size thresholds. The District assumes in its RACM analysis that, for most control measures, the ammonia control efficiencies are the same as the VOC control efficiencies. The District compares Rule 4570 to livestock rules in seven other California air districts, including the South Coast Air Quality Management District (SCAQMD), Bay Area Air Quality Management District (BAAQMD), and the Sacramento Metropolitan Air Quality Management District (SMAQMD). Based on a review of the number of work practices required by the other California rules, the District concludes that Rule 4570 is at least as stringent as the air quality rules for livestock operations in these other areas. In cases where the work practice standards in other rules may be more stringent than the requirements in Rule 4570, the District considers these measures beyond RACT or the emissions differences not significant (
see, e.g.,
Supplement at A-7, noting frequency of mitigation requirements in South Coast Rule 223, adopted June 2, 2006).

49

See
77 FR 2228 (January 17, 2012) (final rule approving Rule 4570 into the California SIP).

The Supplement also compares Rule 4570 to the Idaho Department of Environmental Quality's (IDEQ) dairy ammonia permit by rule, adopted in 2009 (Idaho rule).
50

While Rule 4570 is based on implementing a certain number of BMPs in specific categories, the Idaho rule is based on a “points” system, where each BMP is assigned a certain number of points based on control effectiveness. The District states that Rule 4570 is more stringent than the Idaho rule based on Rule 4570's lower applicability threshold (
e.g.,
500 milking cows v. 1,638 cows (@1,400 lbs) for free stall/flush dairies), more stringent requirements and greater number of required mitigation measures (
see
Supplement at A-25 and A-29). The District also states that Rule 4570 is more stringent regarding temporary suspension and substitution of mitigation measures, and contains more stringent testing, monitoring and recordkeeping requirements. It also states that Rule 4570 is more stringent because it provides specific mitigation measures for various operations at dairies, while the Idaho rule is based on what the District characterizes as an “arbitrary” point system that does not guarantee a specific degree of control (
see
Supplement at A-24 to A-29).
51

50
The Idaho rule requires dairy farms above specified threshold numbers of cows or animal units to register with Idaho DEQ and to implement industry best management practices (BMPs) to control ammonia emissions. A list of BMPs is contained in the rule. Each BMP is assigned a point value based on relative effectiveness in reducing ammonia. Dairy farms must employ BMPs totaling 27 points.
See
Idaho Administrative Procedure Act (IDAPA) 58.01.01 Sections 760-764: Rules for the Control of Ammonia from Dairy Farms (adopted March 30, 2007; amended May 8, 2009).

51
Citing IDEQ's technical documentation for the Idaho rule, the Supplement states that the Idaho rule's point system is “arbitrary” and that overall ammonia emission reductions from the rule may not be substantial given the flexibility in the rule and the absence of a direct correlation between the points required and the amount of emission reductions achieved (
see
Supplement at A-25).

Fertilizer application:
The Supplement identifies fertilizer application to crops as the second largest source of ammonia emissions in the Valley and suggests that research to identify maximum efficiency of nitrogen use for various crop types could lead to a reduction of excess nitrogen in the soil that is susceptible to leaching and volatilization. The Supplement also refers to a “Regional Board General Order” issued by the Central Valley Regional Water Quality Control Board as a control measure that has “significantly reduced losses of nitrogen compounds to the environment, including leaching of nitrogen compounds to groundwater and air emissions such as ammonia and nitrous oxide” (Supplement at A-25 to A-27).
52

The Supplement does not, however, identify any state or local air pollution control measure that limits ammonia emissions from fertilizer application operations.

52

See
Reissued Waste Discharge Requirements General Order for Existing Milk Cow Dairies, California Regional Water Quality Control Board Central Valley Region Order R5-2013-0122, October 3, 2013, available at
http://www.waterboards.ca.gov/centralvalley/board_decisions/adopted_orders/general_orders/r5-2013-0122.pdf

Composting:
The District compares its two SIP-approved composting rules, Rule 4566 (Organic Water Materials, adopted in 2011)
53

and Rule 4565 (Biosolids, Animal Manure, and Poultry Litter Operations, adopted in 2007),
54

to analogous rules adopted by the SCAQMD, Rule 1133.3 (Emission Reductions from Greenwaste Composting Operations, adopted in 2011) and Rule 1133.2 (Emission Reductions from Co-Composting Operations, adopted in 2003). For this analysis, the District assumes that ammonia control efficiencies achieved by these rules are the same as the VOC

control efficiencies. In its comparison of Rule 4566 with SCAQMD Rule 1133.3, the District acknowledges that Rule 1133.3 has a much lower size threshold for implementing most stringent controls (80% control efficiency) but notes that neither SCAQMD nor the District currently has any facilities that trigger the most stringent controls. Therefore, the District states that, in practice, the controls are equivalent. In its comparison of Rule 4565 with SCAQMD Rule 1133.2, the District finds that the controls in Rule 4565 are generally more stringent than Rule 1133.2 for small facilities but less stringent than Rule 1133.2 for large facilities. The Supplement states that the most stringent measures in Rule 1133.2 are not cost-effective for facilities in the Valley.

53

See
77 FR 71129 (November 29, 2012) (final rule approving Rule 4566 into California SIP).

54

See
77 FR 2228 (January 17, 2012) (final rule approving Rule 4565 into California SIP).

3. Evaluation and Proposed Action

The process followed by the District in the 2012 PM
2.5
Plan and Supplement to identify RACM/RACT is generally consistent with the EPA's recommendations in the General Preamble. The process included compiling a comprehensive list of potential controls measures for sources of direct PM
2.5
, NO
X
, SO
2
, and ammonia in the SJV. This list included measures suggested in public comments on the Plan.
See
2012 PM
2.5
Plan, Appendices C, D, and I. No source categories were identified as insignificant except by implication for a few source categories of ammonia which collectively contributed less than 5 percent to the base year ammonia inventory (Supplement, p. A-1.) As part of this process, the District evaluated potential controls for all relevant source categories for economic and technological feasibility and provided justifications for the rejection of certain identified measures.
Id.
After completing this evaluation, the District included enforceable commitments to adopt and expeditiously implement those measures it found to be reasonable.
See
2012 PM
2.5
Plan, Table 5-3, p. 21 and Table D-1 above. Since submittal of the 2012 PM
2.5
Plan in March 2013, the District has adopted all but two of these measures. One measure, amendments to Rule 4905 (Natural Gas-Fired, Fan Type Residential Central Furnaces) is scheduled for adoption this winter. The second measure, amendments to Rule 4692 (Commercial Charbroiling), is not scheduled until 2016. Research and development of cost-effective control technology are currently underway for certain measures in Rule 4692, the addition of PM
2.5
emission limits for under-fired charbroilers (Plan, p. 5-22 and Supplement, p. 8).

We have reviewed the District's determination in the 2012 PM
2.5
Plan that its stationary and area source control measures represent RACM/RACT for direct PM
2.5
, NO
X
, and SO
2
. As discussed in Section II.C. of the TSD, the EPA is proposing to determine that at this time, VOCs do not contribute significantly to ambient PM
2.5
levels that exceed the 2006 PM
2.5
standard in the SJV and that a VOC control evaluation therefore is not necessary for purposes of this standard in this area.
55

In our review, we relied mainly on our previous evaluations of the District's rules in connection with our approval of the SJV RACT SIP for the 1997 8-hour ozone standard, our comments on the 2012 PM
2.5
Plan, and our comments on the District's RACT SIP for the 2008 8-hour ozone standard.
56

We also reviewed measures suggested by the public in comments on the 2012 PM
2.5
Plan. Based on this review, we believe that the District's rules provide for the implementation of RACM/RACT for stationary and area sources of direct PM
2.5
, NO
X
, and SO
2
.
57

55
VOC sources in the San Joaquin Valley are currently subject to a broad range of controls measures adopted by the District and CARB as part of their strategy to attain the ozone standards in the area. The SJV is currently designated nonattainment and classified as extreme for the 1-hour ozone standard and for both the 1997 and 2008 8-hour ozone standards.
See
40 CFR 81.305. Extreme ozone nonattainment areas are required by CAA section 172(c)(1) to implement RACM for VOC sources and by section 182(b)(2) to the implement RACT for all major sources of VOC and certain other sources of VOC. A major source of VOC in an extreme ozone nonattainment area is a source that emits or has the potential to emit 10 tons per year or more of VOC (CAA section 182(e)) which is much lower than the 100 tpy threshold for major stationary sources in Moderate PM
2.5
nonattainment areas (CAA section 302(j)). In 2012, the EPA approved the RACM demonstration in SJV's SIP for attainment of the 1997 8-hour ozone standard (77 FR 12652, 12670 (March 1, 2012)) and found that the State had met the section 182(b)(2) RACT control requirement for the 1997 8-hour ozone standards (limited approval, 77 FR 1417 (January 10, 2012); final corrective rule approval, 77 FR 71129 (November 29, 2012)).

As noted in the General Preamble, Congress recognized that PM precursors may be otherwise controlled and expressly recommended that the EPA consider other provisions of the CAA in addressing precursors (General Preamble at 13542, n. 22). Accordingly, the General Preamble provides that a state may demonstrate in a SIP submittal that control of VOC under other CAA requirements relieves it from the need to adopt additional controls for VOC as a precursor under section 189(e).
Id.
at 13542.

56

See
77 FR 1417 (January 10, 2012); EPA Region 9, Technical Support Document for EPA's Notice of Proposed Rulemaking for the California State Implementation Plan—EPA's Evaluation of San Joaquin Valley Unified Air Pollution Control District's Reasonably Available Control Technology (RACT) Demonstration for Ozone State Implementation Plan (SIP), Adopted April 16, 2009 (dated August 29, 2011); letter dated October 19, 2012, from Kerry Drake, Associate Director, Air Division EPA-Region 9 to Samir Sheikh, SJVAPCD; and letter dated June 4, 2014, from Andrew Steckel, Chief, Rules Office, EPA Region 9 to Errol Villegas, Planning Manager, SJVAPCD.

57
A full list of the District's rules, including cites to our most recent final or proposed rulemaking on each can be found in Appendix B to the TSD.

We have also reviewed the District's analysis of current and potentially available ammonia controls for the three largest sources of ammonia emissions in the SJV: CAF operations, agricultural fertilizer application, and composting operations. Collectively these sources account for 96 percent of total base year (2007) ammonia emissions.
58

See
2012 PM
2.5
Plan, Appendix B, Table B-5.

58
The balance of the ammonia inventory is spread among a number of source categories from electric utilities to gasoline-powered on-road vehicles. The largest of these sources, at 6.3 tpd in 2007 (1.7 percent of the total ammonia inventory), is the area source category “Other (Miscellaneous Processes).”
See
2012 PM
2.5
Plan, Appendix B, Table B-5. Ammonia emissions in this category are from domestic sources including pets, untreated human waste (
e.g.,
diapers), and perspiration.
See
ENVIRON International Corporation, “Final Work Plan California Regional PM
10
/PM
2.5
Air Quality Study—Ammonia Emissions Improvement Projects in Support of CRPAQS Aerosol Modeling and Data Analyses: Draft Ammonia Inventory Development,” April 13, 2001, p. 2-16, found at
http://www.arb.ca.gov/ei/areasrc/nh3environworkplan.pdf.

We note that two of the remaining source categories (open burning and residential wood burning at a combined 2.9 tpd in 2007) are covered by SIP-approved rules that prohibit and/or curtail burning and therefore also reduce ammonia emissions from these sources.

Farming Operations and Composting:
The District adopted Rule 4565, Rule 4566 and Rule 4570 primarily to control VOC emissions, and EPA has determined that the control requirements in each of these rules represent RACT-level controls for VOCs.
See
77 FR 2228 (January 17, 2012) (final rule approving Rule 4565 and Rule 4570 into California SIP) and 77 FR 71129 (November 29, 2012) (final rule approving Rule 4566 into California SIP). Although the Supplement does not provide a detailed analysis of the extent to which these rules also reduce ammonia emissions, the District's general conclusion that the work practice standards in these rules reduce ammonia emissions appears to be factually correct. The District's evaluation of Rule 4565, Rule 4566 and Rule 4570 generally supports a conclusion that these SIP-approved rules for CAFs and composting operations are as stringent as analogous rules implemented in other California districts.

With respect to the Idaho rule, because it is based on a point system and Rule 4570 is based on numbers of BMPs for different components of the CAF operation, it is difficult to compare the requirements in these two rules

directly. Both rules contain options for controlling ammonia emissions in different parts of the CAF operation, but Rule 4570 may be more prescriptive in requiring a certain number of BMPs for each component, while the Idaho rule does not set a specific number of BMPs and instead requires that the points associated with the selected BMPs total to at least 27. It appears that in some respects Rule 4570 has more stringent provisions than the Idaho rule (
e.g.,
rule applicability threshold, and provisions for temporary suspension/substitution, testing and records retention). On the other hand, the Idaho rule contains options (
e.g.,
chemical amendments, lagoon nitrification/de-nitrification systems, lagoon covers, tunnel ventilation with biofilter, incorporation of manure within 24 hours of land application) that do not appear to be included in Rule 4570. It is not clear whether these control options are commonly implemented in Idaho or reasonably available or appropriate for use in the SJV.

In the absence of specific information regarding more stringent ammonia air emission control measures that may be technologically and economically feasible for implementation in the SJV area, we find the District's evaluation of Rule 4565, Rule 4566 and Rule 4570 in comparison to analogous rules adopted elsewhere is adequate to demonstrate that the District is implementing all available control measures for ammonia emissions from CAFs and composting operations that are technologically and economically feasible for implementation in the SJV at this time. We note, however, that the District's analyses of these rules appear to rest on certain assumptions concerning ammonia emissions reductions that are not supported by specific documentation or analyses (
e.g.,
assumptions concerning ammonia control efficiencies based on VOC control efficiencies). The point system in the Idaho rule appears to be based upon detailed scientific studies on the ammonia control efficiencies of the identified BMPs,
59

some of which may be available for implementation in the SJV. We note also that the timing of the
NRDC
decision in early 2013 may have constrained the State's and District's ability to fully evaluate additional ammonia control measures as part of a RACM/RACT control strategy ahead of the applicable Moderate area attainment date (December 31, 2015). Taking these unique circumstances into account, we find the District's ammonia control evaluation adequate for RACM/RACT purposes at this time but recommend that the State and District conduct a more thorough evaluation of all available ammonia control measures for farming and composting operations as part of its development of a Serious area plan. Specifically, we encourage the District to revisit its control evaluation for CAFs and composting operations at its earliest opportunity, both to specifically consider the ammonia control efficiencies and cost effectiveness of Rule 4565, Rule 4566 and Rule 4570, and to address any additional control options that may be technologically and economically feasible for implementation in the SJV,
e.g.,
the BMPs identified in the Idaho rule and other control options identified by EPA or by citizens during the District's rulemaking processes.

59
IDEQ's technical documentation indicates that under the Idaho rule, BMP points are awarded based on a detailed assessment of each BMP's relative effectiveness in controlling ammonia emissions.
See
IDEQ, “Scientific Basis for the Control of Ammonia from Dairy Farms Best Management Practices,” July 18, 2006,
available at http://www.deq.idaho.gov/media/635665-58_0101_0502_scientific_basis_final.pdf
.

Fertilizer application:
Although certain water pollution control and other requirements contained in the “Regional Board General Order” cited in the Supplement may indirectly reduce ammonia emissions to the air from fertilizer application operations, these requirements have not been approved into the SIP for purposes of attainment of the NAAQS and therefore cannot provide a basis for approval of a RACM demonstration under the CAA. We are not aware, however, of ammonia air emission control measures that have been adopted and implemented for fertilizer application operations elsewhere. In the absence of information regarding air emission control measures for agricultural fertilizer application that may be technologically and economically feasible for implementation in the SJV area, we find the District's analysis in the Supplement acceptable for RACM purposes at this time. We note also that the timing of the
NRDC
decision in early 2013 may have constrained the State's and District's ability to fully evaluate additional ammonia control measures as part of a RACM/RACT control strategy ahead of the applicable Moderate area attainment date (December 31, 2015). Taking these unique circumstances into account, we find the District's ammonia control evaluation adequate for RACM/RACT purposes at this time, but we encourage the District to revisit its control evaluation for fertilizer application at its earliest opportunity to specifically evaluate the technical and economic feasibility of potential air pollution control measures that may reduce ammonia emissions from fertilizer application in the SJV.

Mobile Sources:
Finally, we have reviewed the analysis of current and potentially available controls for both on and off-road mobile sources in Chapter 5 and Appendix C. As we have noted in previous actions on SJV plans,
60

California is a leader in the development and implementation of stringent control measures for on-road and off-road mobile sources. Its current program addresses the full range of mobile sources in the SJV through regulatory programs for both new and in use vehicles and through incentive grant programs.
See
2012 PM
2.5
Plan, Table 5-2 and Appendix A of the TSD. The District has also adopted measures to reduce emissions from mobile sources including its indirect source review rule (Rule 9510) and employer trip reduction rule (Rule 9410) and has a well-funded incentive grants program focused on mobile sources.
See
Chapter 6. The MPOs also have a program to fund cost-effective TCMs.
See
Appendix C, p. C-33. Overall, we believe that the State, District, and MPO programs provide for the implementation of RACM for PM
2.5
and its precursors from mobile sources in the SJV.

60
The proposed approvals of the SJV 2007 [8-hour] Ozone Plan at 76 FR 57846, 57852 (September 16, 2011) and the 2008 PM
2.5
Plan at 76 FR 41338, 41345 (July 13, 2011) include discussions of California's control programs for mobile sources.

For the foregoing reasons, we propose to find that the 2012 PM
2.5
Plan and Supplement provide for the implementation of all RACM/RACT that can be implemented prior to the applicable Moderate area attainment date as required by CAA sections 189(a)(1)(C) and 172(c)(1) and to approve the RACM/RACT demonstration in the Plan.
61

61
The 2012 PM
2.5
Plan is the latest in a series of air quality plans and control strategies that the District and CARB have adopted to provide for attainment of the NAAQS in the SJV. These plans include the 2003 PM
10
Plan (approved 69 FR 30005 (May 26, 2004)); the 2004 Extreme [1-hour] Ozone Attainment Plan (approved 75 FR 10420 (March 8, 2010), approval withdrawn 77 FR 70376 (November 26, 2012)); the 2007 [8-hour] Ozone Plan (approved 77 FR 12652 (March 1, 2012)); the 2008 PM
2.5
SIP (approved 76 FR 69896 (November 9, 2011)); and the 2009 RACT SIP (limited approval January 10, 2012 (77 FR 1417); last corrective rule approval November 29, 2012 (77 FR 71129)).

E. Major Stationary Source Control Requirements Under CAA Section 189(e)

CAA section 189(e) specifically requires that the control requirements applicable to major stationary sources of

direct PM
2.5
62

also apply to major stationary sources of PM
2.5
precursors, except where the Administrator determines that such sources do not contribute significantly to PM
2.5
levels that exceed the standard in the area (General Preamble at 13539 and 13541 to 42). The Supplement contains a discussion of the District's Rule 2201 (New and Modified Stationary Source Review Rule) (Supplement at 17-20) and three potential major stationary sources of ammonia operating in the SJV (Supplement at A-39 to A-41). The EPA is not proposing to act on these components of the Plan at this time and will do so in a separate rulemaking to address the control requirements specific to major stationary sources of direct PM
2.5
and PM
2.5
precursors under section 189(e) in the SJV.

62
The control requirements applicable to major stationary sources of direct PM
2.5
in a Moderate area include, at minimum, the requirements of a nonattainment new source review (NNSR) permit program meeting the requirements of CAA section 172(c)(5) (
see
CAA 189(a)(1)(A)).

F. Adopted Control Strategy

1. Requirements for Control Strategies and Enforceable Commitments

CAA section 110(a)(2)(A) provides that each SIP “shall include enforceable emission limitations and other control measures, means or techniques . . . as well as schedules and timetables for compliance, as may be necessary or appropriate to meet the applicable requirement of the Act.” Section 172(c)(6) of the Act, which applies to nonattainment SIPs, is virtually identical to section 110(a)(2)(A).
63

Measures necessary to meet RACM/RACT requirements should also be adopted by the State in an enforceable form (General Preamble at 13541).

63
The language in sections 110(a)(2)(A) and 172(c)(6) is quite broad, allowing a SIP to contain any enforceable “means or techniques” that EPA determines are “necessary or appropriate” to meet CAA requirements, such that the area will attain as expeditiously as practicable, but no later than the designated date. Furthermore, the express allowance for “schedules and timetables” demonstrates that Congress understood that all required controls might not have to be in place before a SIP could be fully approved.

Generally, the EPA will approve a State plan that takes emissions reduction credit for a control measure only where the EPA has approved the measure as part of the SIP, or in the case of certain on-road and nonroad measures, where the EPA has issued the related waiver of preemption or authorization under CAA section 209(b) or section 209(e). The EPA has, however, approved enforceable commitments that are limited in scope where circumstances exist that warrant the use of such commitments in place of adopted measures. Commitments approved by the EPA under CAA section 110(k)(3) are enforceable by the EPA and citizens under CAA sections 113 and 304, respectively. In the past, the EPA has approved enforceable commitments and courts have enforced actions against states that failed to comply with them.
64

Additionally, if a state fails to meet its commitments, the EPA could make a finding of failure to implement the SIP under CAA section 179(a)(4), which starts an 18-month period for the state to correct the non-implementation before mandatory sanctions are imposed.

64

E.g., American Lung Ass'n of N.J.
v.
Kean,
670 F. Supp. 1285 (D.N.J. 1987), aff'd, 871 F.2d 319 (3rd Cir. 1989);
NRDC, Inc.
v.
N.Y. State Dept. of Env. Cons.,
668 F. Supp. 848 (S.D.N.Y. 1987);
Citizens for a Better Env't
v.
Deukmejian,
731 F. Supp. 1448, recon. granted in par, 746 F. Supp. 976 (N.D. Cal. 1990);
Coalition for Clean Air
v.
South Coast Air Quality Mgt. Dist.,
No. CV 97-6916-HLH, (C.D. Cal. Aug. 27, 1999).

Once the EPA determines that circumstances warrant use of an enforceable commitment, the EPA considers three factors in determining whether to approve the use of an enforceable commitment to meet a CAA requirement: (a) Does the commitment address a limited portion of the CAA-required program; (b) is the state capable of fulfilling its commitment; and (c) is the commitment for a reasonable and appropriate period of time.
65

65
The U.S. Court of Appeals for the Fifth Circuit upheld the EPA's interpretation of CAA section 110(a)(2)(A) and 172(c)(6) and the Agency's use and application of the three factor test in approving enforceable commitments in the Houston-Galveston ozone SIP in
BCCA Appeal Group et al.
v.
EPA et al.,
355 F.3d 817 (5th Cir. 2003).

2. Control Strategy in the 2012 PM
2.5
Plan

For purposes of evaluating the 2012 PM
2.5
Plan, we have divided the measures relied on to satisfy the applicable control requirements into two categories: baseline measures and control strategy measures.

As the term is used here, baseline measures are federal, State, and District rules and regulations adopted prior to January 2012 (
i.e.,
prior to the development of 2012 PM
2.5
Plan) that continue to achieve emissions reductions through the current attainment year of 2015 and beyond.
66

The Plan describes many of these measures in Chapter 5, Section 5.1.1 and 5.1.2 and in Appendices C and D. Reductions from these measures are incorporated into the baseline inventory and, for the most part, not individually quantified. According to the Plan, these measures provide all the emissions reductions projected to occur between the base year of 2007 and the current attainment year of 2015 and, based on the model-predicted level of emission reductions needed to meet the 2006 PM
2.5
standard in the SJV in 2019, most of the direct PM
2.5
, NO
X
, SO
2
, and ammonia reductions needed to meet the PM
2.5
standard.
See
Table F-1 in the TSD and Appendices A and B in the TSD.

66
These measures are typically rules that have compliance dates that occur after the adoption date of a plan and mobile source measures that achieve reductions as older engines are replaced through attrition (
e.g.,
through fleet turnover).

Control strategy measures are the new rules, rule r

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2015-00270. Public record. Not legal advice.