On August 2, 2010, EPA proposed the long- awaited replacement for the Clean Air Interstate Rule (CAIR), which was rejected in 2008 by the D.C. Circuit Court of Appeals. Now called the Transport Rule (TR) by EPA, the new rule revises the CAIR approach to conform with the court ruling. In this first version (Round 1) of TR, overall emission budgets for NOX are similar to the initial period in CAIR; overall SO2 budgets are approximately 50% of the initial CAIR budget; and only utilities are regulated (see attached table). However, EPA has stated its intention in Round 2 to revise TR at minimum to consider additional NOX reductions to assist at least two (likely three) metropolitan areas in meeting the 1997 ozone NAAQS of 0.08 ppm, for which an impact of 84 ppb was acceptable.
Unlike Round 1, Round 2 is expected to require many industrial NOX emitters to reduce emissions. Further, EPA has clearly designed TR to be readily adaptable to future NAAQS revisions, and is defining the process for future TR versions in this initial rulemaking. As proposed, TR only seeks to remove upwind pollution that “significantly contributes to or interferes with maintenance of” the 1997 ozone NAAQS, the 1997 annual PM2.5 NAAQS, and the 2006 24-hr PM2.5 NAAQS, each of which is currently under review. A new ozone standard is expected in Fall 2010 and new PM2.5 standards are expected in 2011, which may lead to TR Round 3. EPA expects to propose revised TR requirements approximately one year after new NAAQS designations, with a final rule one year later.
Why Address Transport?
… any source … within the state from emitting any air pollutant in amounts which will contribute significantly to nonattainment in, or interfere with maintenance by, any other state with respect to any [NAAQS]….
Congress has long recognized the potential for transported emissions to impact air quality, and so called “good neighbor” provisions date to the 1970 Clean Air Act Amendments (CAAA). The original requirement was amended and is now contained in §110(a)(2)(D), which requires each state, in its state implementation plan (SIP), to prohibit: Prior to CAIR, EPA regulated transport through the NOx SIP Call, which addressed emissions contributing to ozone generally east of the Mississippi River. CAIR built upon the approach used in the NOx SIP Call with both an expanded group of states and expanded NAAQS (both ozone and PM). Despite its grounding in the prior NOx SIP Call, in a unexpected 2008 ruling, the court found CAIR so fundamentally flawed that it vacated the rule in its entirety. As a result of petitions regarding the lack of any requirement for additional emissions reductions while EPA developed a new rule, the court amended its vacatur to simply a remand, which the current TR seeks to address.
Development of Required Emission Reductions
For TR, EPA first determined via air dispersion modeling which states (considering all emissions statewide) had a significant impact on an out-of-state nonattainment area (or maintenance area), where EPA defined significant as 1% of the respective NAAQS. For example, Virginia was linked as follows for the annual PM2.5 and ozone standards (listed by predominant cities in area). Comparatively more areas were linked for the PM2.5 24-hr standard (20 vs. 8).
Annual PM2.5 – nonattainment
- Lancaster, PA
- York, PA
- Huntington, WV/Ashland, KY
- Charleston, WV
Annual PM2.5 – maintenance
- New York, NY
- Reading, PA
- Hagerstown, MD/Martinsburg, WV
- Fairmont, WV
Ozone – nonattainment & maintenance
- New York, NY
- Philadelphia, PA
Once linkages were defined, EPA proceeded to determine the quantity of the “significant contribution,” which is the amount to be eliminated via TR. For the NOx SIP Call and CAIR, the determination was performed regionally; in TR, state-specific factors are used in response to the court ruling. For TR, EPA used a 4-step approach.
- Identify reductions available at various costs
- Use simplified air quality assessment tool to quantify impact of reductions at different $/ton levels
- Identify “breakpoints” based on cost and air quality information
- Quantify reductions available in each linked state at chosen cost threshold; verify using refined air model.
EPA’s approach assumes the same cost per ton in upwind and downwind states, though EPA recognizes that unique local conditions can necessitate non-equal costs. EPA’s proposed rule notes Liberty-Clairton PA as one area where there is a disproportionate local contribution due to coke ovens and other local sources.
In Step 1, EPA identified what controls were available and when. For instance, EPA believes that scrubbers and SCRs could not be installed until 2014; low NOX burners (LNB) and fuel switching are deemed available in 2012. Of note, EPA does not include any considerations for air permitting in Its timeframe, although some units have previously triggered major New Source Review for LNB (and potentially other) projects. After determining available controls, EPA then models those impacts using a screening model in Step 2.
Step 3 comprises a review for each state of the appropriate cost breakpoint, and identifies a specific $/ton threshold for quantifying the amount of significant contribution. For PM2.5 in 2014, after considering costs for SO2 from $100/ton to $2,400/ton, EPA narrowed consideration to either $2,000/ton or $2,400/ton. Since the air quality model predicted no change in areas impacted by reducing from $2,000/ton to $2,4000/ton, the proposed rule is founded on $2,000/ton. However, changes costing $2,000/ton (scrubbers) could not be installed until 2014, and EPA considered what changes could be in place by 2012 (primarily operating scrubbers that are installed but not required to be operated under an existing regulatory program). EPA modeled this smaller set of emission reductions and found that these reductions (at less than $2,000/ton) eliminated the contribution from many states. Based on that finding, EPA separated the states into two groups for SO2. Group 2 states are able to fully remove their significant contribution by 2012 at less than $2,000/ton and have only a single SO2 budget. In contrast, Group 1 states have a 2-step approach, with one budget for 2012-13 and a more stringent budget for 2014 onward.
For NOx (still with respect to PM2.5 attainment) EPA determined $500/ton was appropriate since SO2 contributes more to PM2.5 than NOx and to obtain additional reductions beyond those available at $500/ton required an increase to at least $2,400/ton.
Since the SO2 budgets will require substantial SO2 reductions, and the allocations are no longer linked to the Acid Rain program, SO2 credits under Acid Rain will be greatly devalued, resulting in SO2 “leakage” to non-covered states. EPA identified five states likely to see SO2 increase by at least 5,000 tpy due to this leakage: Arkansas, Mississippi, North and South Dakota, and Texas. Additional modeling of these states’ post-leakage budgets bumped Texas into a significant PM2.5 contribution, and EPA seeks comment on whether Texas should be included in Group 2.
For ozone, EPA similarly considered available controls in 2012 and 2014 in Step 3. At $500/ton, EPA believes that utilities essentially are the only sector to be considered, while at higher costs other sectors should be considered. However, EPA did not have adequate time to consider higher thresholds for this rule-making, and plans a future rulemaking to consider whether reductions at a higher cost per ton are appropriate for utilities and other source categories. Based on $500/ton, all areas except Houston, TX and Baton Rouge, LA will be in attainment with the ozone NAAQS, while New York, NY flirts with nonattainment based on historical year-to-year ozone levels. For areas that contribute to Houston or Baton Rouge (and potentially areas that contribute to New York), EPA plans a review of additional reductions that could be achieved at more than $500/ton and at least up to $3,200/ton. At the highest cost thresholds, EPA expects to consider NOX reductions from (1) industrial boilers, (2) reciprocating internal combustion engines (RICE), (3) portland cement manufacturing, (4) petroleum refining, (5) glass manufacturing, (6) pulp and paper production and, (7) iron and steel production.
State Emissions Budgets
Using the approach outlined previously, EPA ran its Integrated Planning Model with state-specific inputs to determine unit-by-unit expected emissions for Group 1 SO2 states in 2014. For 2012 (Group 2 SO2 and ozone), EPA used a mix of actual performance modified by pending controls. The details of EPA’s approach are complex and unit-by-unit data should be closely reviewed by those subject to the rule; there are several exceptions to the general approach, both on a state and unit basis.The unit-by-unit values were summed to determine state budgets. The state budgets represent the emissions that would remain after significant contributions and interference with maintenance have been addressed, in an average year. For example, in Virginia, IPM predicts that the following changes by 2014.
- Clinch River – three 230 MW units each add scrubbers/SCR
- Bremo Bluff No. 4 – 150 MW unit adds a scrubber
- Chesterfield No. 3 – 100 MW unit adds a scrubber
Compared to CAIR, TR budgets may be lower or higher. When comparing TR to the first-phase CAIR budgets (TR/CAIR), the ratio ranges from 30% to 190%. In the aggregate, NOX emissions (ozone season and annual) are similar in TR Round 1 and CAIR Phase 1, while TR SO2 emissions are approximately 50% of CAIR Phase 1. The low SO2 budget, and dissociation from the Acid Rain Program (ARP) credits, results in essentially flooding the national ARP market, and projected increases in SO2 emissions in states not covered by TR.
To address variability, EPA identified both 1-year and 3-year variability limits, which are values allowed above the state budgets, subject to conditions. The 1-year variability is either 10% of the annual budget or 5,000 tons (SO2)/1,700 tons (NOX), whichever is greater. The 3-year variability is based on dividing 1-year varability by the square root of three, while ozone season is the higher of 10% or 2,100 tons. The variability limits would not apply until 2014 (1-year) and 2016 (3-year). Several alternatives are also considered in the proposal.
EPA proposed three approaches to implement the state budgets. The preferred approach allows unlimited intrastate trading and limited interstate trading. Alternative A allows solely intrastate trading, and Alternative B is a direct control option, likely with lb/MMBtu limits for individual sources.
The reason interstate trading is limited is due to what EPA terms “assurance provisions.” If, in any 1-year or 3-year period, a state exceeds its variability limits, sources in that state are penalized by a requirement to submit “assurance” allowances. In essence, across a state, allowances submitted in excess of variability provisions are discounted by 50%, requiring submittal of additional allowances. EPA would calculate the quantities and apportion assurance allowances to individual units. The assurance allowances are above and beyond allowances already submitted by each unit for actual emissions. EPA believes the likelihood of triggering these provisions is low.
In concert with the proposed trading approach, EPA has proposed unit-specific emissions allocations based on the analysis used to develop the state emissions budgets. These allocations would be one-time and permanent (at least until EPA issues Round 2 TR). As with CAIR, EPA proposes a new unit set-aside, based on 3% of the state budget.
Under EPA’s preferred option, most of the remaining trading requirements are similar to CAIR. For example, under TR, states have the option to develop their own SIPs to implement state-specific budgets.
Applicability to TR is generally the same as for CAIR, namely limited to fossil fuel-fired units serving a generator greater than 25 MW producing electricity for sale. The exemptions for co-generation units and solid waste incinerators are generally the same as CAIR with minor technical corrections. Notably, EPA did not provide a broad exemption for trivial usage of fossil fuel, such as for startup only; any fossil fuel usage at any time makes a unit fossil fuel-fired forever. For example, biomass boilers that combust 100% biomass but use natural gas or diesel to startup would be classified as fossil fuel-fired.
EPA does propose to modify the definition of “fossil” solely for the solid waste incineration unit definition. The definition of solid waste incineration unit is tied to §129(g)(1) of the Clean Air Act, and thus is linked to the currently pending rulemaking to define solid waste (75 FR 31844, June 4, 2010). Further review of this exemption is warrantedafter finalization of the solid waste rule.
While EPA received multiple requests to extend the comment period on this complex rule, the comment period ends on October 1, 2010 for the core rule and on October 15, 2010 for additional IPM data made available on September 1. On September 17, in supporting its denial of requests for more review time, EPA announced that it believes sufficient time was provided to review the rule.
For industrials and those in states marked as “review pending” for ozone, expect to see additional proposed requirements to address NOxemissions, at least for the summer ozone season, with a proposal likely in 2012. Lastly, both industrials and utilities should monitor future TR revisions (Round 3) that will follow pending NAAQS revisions.