Daily Screening for Wildfire Impacts on Ozone using a Photochemical Model A Proposal to the Texas Near-Nonattainment Areas Greg Yarwood (gyarwood@ramboll.com) Sue Kemball-Cook (skemballcook@ramboll.com) Ramboll Environ March 20, 2017

The 2015 Ozone NAAQS and the Exceptional Event Rule In 2015, the U.S. EPA promulgated a more stringent ozone NAAQS of 70 ppb It is now more likely that background ozone due to natural sources can cause U.S. ozone monitors to exceed the NAAQS even when local emissions from human activities are minimal Background ozone is not controllable The Exceptional Events Rule1 addresses monitoring data influenced by natural and/or uncontrollable events State must prepare a demonstration to EPA showing Clear, causal relationship between the event and the monitored exceedance The event was not reasonably controllable or preventable The event was a human activity that is unlikely to recur at a particular location or was a natural event In 2005, Congress provided the statutory authority for the exclusion of data influenced by “exceptional events” meeting specific criteria by adding section 319(b) to the Clean Air Act (CAA). To implement this 2005 CAA amendment, the EPA promulgated the 2007 Exceptional Events Rule (72 FR 13560, March 22, 2007). 1https://www.epa.gov/air-quality-analysis/exceptional-events-rule-and-guidance

The Exceptional Events Rule and Design Values An exceedance or violation due to an exceptional event may be excluded from the calculation of the monitor’s ozone design value* Can lower the design value and even change nonattainment status to attainment Example: Exceptional Event 4 Highest Values of Daily Maximum 8-hour Ozone Ozone design values are the metric calculated for comparison with the NAAQS and are defined as the consecutive three year average of annual 4th highest maximum daily 8-hour average (MDA8) ozone concentration. *Design value:  3-Year Average of 4th high daily max 8-hour ozone not to exceed 70 ppb per 2015 Ozone NAAQS

Wildfires and Ozone Wildfires create plumes with enhanced PM, NOx and ozone Plumes affect ozone locally and regionally via transport Wildfires can be classified as exceptional events Wildfires are the most likely exceptional event to affect East Texas ozone Photochemical models can quantify wildfire influences on ozone Can provide prompt alerts of fire impacts on ozone at a monitor when used as a screening tool that is run every day throughout the ozone season (AP Photo/Gregory Bull) http://www1.ncdc.noaa.gov/pub/data/cmb/images/fire/2011/09/TX_fires_20110906.jpg

Photochemical Model Exceptional Event Screening System Ramboll Environ proposes to deploy a photochemical model exceptional event screening system for East Texas from March 1, 2018 through October 31, 2018 and March 1, 2019 through October 31, 2019 Two CAMx photochemical model runs will be made for each day during the ozone season. The two runs will be identical except: One CAMx run with fire emissions One CAMx run without fire emissions The difference in ozone between the two runs, (CAMx ozone with fire emissions – CAMx ozone without fire emissions), is the ozone impact of the fire emissions The system will run automatically on each day and provide modeling products to the NNAs, including a daily assessment of fire impacts on ozone at NNA regulatory monitors

Exceptional Event Screening System Deliverables The screening system output/results will be made available via a password-protected modeling system website hosted by Ramboll Environ Example model fire impact results shown in slides 10-12 Ramboll Environ will alert NNA staff and the NNA’s TCEQ technical liaison when a fire event affects a monitor in their area For each participating NNA, Ramboll Environ will prepare a report documenting the results of the exceptional event screening modeling at the Area’s monitors All days with fire influence at NNA regulatory CAMS will be identified and the magnitude of the fire influence on the daily maximum 8-hour average ozone will be reported A list of days warranting further analysis as a potential exceptional event will be developed

Exceptional Event Screening System Cost Cost to deploy and operate the screening system is $150,000 per year Total cost for FY18-19 biennium is $300,000 Costs can be shared among the NNAs Cost to each NNA depends on number of participating areas

Examples of Photochemical Model Wildfire Analysis Results Animations in slides 10-11 show only the ozone produced by fires Ozone is produced by fires in California, New Mexico and Arizona. The ozone produced by the fires is blown east by the winds and affects the CAMS 12 monitor in El Paso during midday on June 21 Ozone produced by fires appears as green, yellow and red. Areas where fire emissions consume ozone are blue. Ozone forms during the day and dissipates at night

How to Run the Animation on the Next Slide Move your cursor to the bottom of the figure A control bar will appear under the contour color bar Start/stop the animation by clicking on the arrow at the left of the control bar The smaller arrows on the right of the control bar can be used to step through the animation frame by frame Regions colored green and red show where fire emissions increased ground level ozone, while blue regions show where fire emissions decreased ground level ozone There is a time stamp at the top of each animation frame: 20150617_03 YearMonthDay__Hour

Impact of Fire Emissions on Ground Level Ozone: June 17-22 Animation of CAMx Ozone Modeling Results Impact of Fire Emissions on Ground Level Ozone: June 17-22 CAMx model results showing: Ozone impact of fire emissions = CAMx ozone run with fire emissions – CAMx ozone run without fire emissions CAMS 12 4 km Domain

Animation of CAMx Regional Ozone Modeling Results Impact of Fire Emissions on Ground Level Ozone: June 17-22 CAMx model results showing: Ozone impact of fire emissions = CAMx ozone run with fire emissions – CAMx ozone run without fire emissions CAMS 12 12 km Domain

Example of Fire Impact: El Paso CAMS 12 on June 21, 2015 Ozone in the CAMx runs with (blue) and without (green) fire emissions differs substantially on June 21, the day when fires affected CAMS 12 Magnitude of fire impact is shown in red and is plotted on the right axis The CAMx run with fires had 7 ppb more ozone at 11 am (red line) than the CAMx run without fires Model results show a clear, causal relationship between fire emissions and ozone at CAMS 12