Ozone enhancement in Western U.S. wildfire plumes at the Mt. Bachelor Observatory: The Role of NOx
Author
Baylon, Leo Miguel Paolo
Metadata
Show full item recordAbstract
We looked at 19 wildfire events that were observed in the summers of 2012 (July 22 - September 20) and 2013 (July 12 - August 30) at the Mt. Bachelor Observatory (MBO, 2.7 km a.s.l.), a mountaintop site located in central Oregon. Fire events were identified based on the following criteria: 5-minute ambient aerosol scattering > 20 Mm-1 for at least two hours; 5-minute CO > 150 ppbv for at least two hours; strong correlation (R2 > 0.70) between aerosol scattering and CO; and backward trajectories consistently indicating transport over fire spots. We identified wildfire plumes using enhancement ratios (deltaY/deltaX), which we calculated for each plume by taking the Reduced Major Axis linear regression slope of various species. We reported the calculated enhancement ratios and explored their variation with source distance. We measured NOx at MBO during Summer 2012 and 2013 using a two-channel chemiluminescence instrument manufactured by Air Quality Design, Inc. The instrument has been used at MBO previously. During 2012, average summertime NO sensitivities in the NO and NOx channels were 2.7 and 2.2 counts per second (cps)/pptv, respectively. Average summertime NO2 sensitivity in the NOx channel was 0.7 cps/pptv. These values correspond to NO and NO2 detection limits of 15 pptv and 35 pptv, respectively. During 2013, a different ozone generator was used. Average summertime NO sensitivities in the NO and NOx channels were 4.0 and 3.5 cps/pptv, respectively. Average NO2 sensitivity in the NOx channel was 0.9 cps/pptv. These correspond to detection limits of 7 pptv for NO and 30 pptv for NO2. We observed a negative correlation between deltaO3/deltaCO and deltaNOx/deltaNOy (r = -0.72). This showed that the degree of NOx oxidation is a key predictor of ozone production. The highest deltaNOx/deltaNOy (0.57 pptv/pptv) was associated with ozone loss (ozone titration). Low deltaNOx/deltaNOy values (ranging from 0.049 to 0.15 pptv/pptv) are generally associated with ozone enhancement. We also found that even if deltaO3/deltaCO is low, deltaO3 may still be significant if CO enhancements are large. We then explored events that are not associated with any O3 enhancement/loss. Out of 19 fire events, 3 belong to this category. We discovered that these events are either BL-influenced (O3 deposition), associated with low deltaNOy/deltaCO, and/or associated with minimal photochemistry (due to nighttime transport). Absolute ozone enhancements ranged from 3.8 to 32 ppbv, while ozone production efficiencies (OPEs) ranged from 2.1 to 17. However, because PAN comprises most of the reactive nitrogen in fire plumes, the calculated OPEs underestimated the true ozone mixing ratios. OPEs may therefore be misleading indicators of ozone production in wildfires. Finally, we segregated the data into plume/non-plume time periods. From this we found that the average O3 mixing ratio was significantly higher in fire plumes compared to non-plume time periods, and the noontime NO/NO2 ratios were also higher. This later result gives insight into the photochemical environment in the fire plumes. We looked at several example events in detail. Event 5 was associated with ozone loss that could be explained by high NOx conditions (which lead to NOx titration) and nighttime chemistry. The Pole Creek Fire led to three large peaks (events 9, 10 and 11) in CO, aerosol scattering and O3 observed at MBO. We then uses a dispersion model with continuous particle release run and looked at the three peaks in aerosol scattering that were observed at MBO on September 18, 19 and 20, 2012. The dispersion results are consistent with the timing of the three peaks observed at MBO. We have NOx data for only the first (event 9) of these peaks. Calculated OPE for event 9 suggests that 2.1 ppbv of ozone is produced per 1 ppbv of NOx lost. Finally, we looked at event 6, which had a deltaNOx/deltaNOy of 0.11. This low value indicates a high degree of chemical processing. Because event 6 has all the necessary requirements for ozone production, it is associated with a relatively high deltaO3/deltaCO of 0.11 ppbv/ppbv. To our knowledge, this is the first work that looks at case studies of different O3 enhancement scenarios (i.e., O3 enhancement, O3 loss, no O3 enhancement/loss), and therefore furthers our knowledge about O3 production, which until now has been poorly understood. We believe this work represents a significant advance in our understanding of wildfire plume chemistry and O3 production.
Collections
- Atmospheric sciences [315]