Ozone Production and Its Sensitivity to NO<sub><i>x</i></sub> and VOCs: Results from the DISCOVER-AQ Field Experiment, Houston 2013

Abstract

<p><strong>Abstract.</strong> An observation-constrained box model based on the Carbon Bond mechanism, Version 5 (CB05), was used to study photochemical processes along the NASA P-3B flight track and spirals over eight surface sites during the September 2013 Houston, Texas deployment of the NASA DISCOVER-AQ campaign. Data from this campaign provided an opportunity to examine and improve our understanding of atmospheric photochemical oxidation processes related to the formation of secondary air pollutants such as ozone (O<sub>3</sub>). O<sub>3</sub> production and its sensitivity to NO<sub><i>x</i></sub> and VOCs were calculated at different locations and times of day. Ozone production efficiency (OPE), defined as the ratio of the ozone production rate to the NO<sub><i>x</i></sub> oxidation rate, was calculated using the observations and the simulation results of the box and Community Multiscale Air Quality (CMAQ) models. Correlations of these results with other parameters, such as radical sources and NO<sub><i>x</i></sub> mixing ratio, were also evaluated. It was generally found that O<sub>3</sub> production tends to be more VOC sensitive in the morning along with high ozone production rates, suggesting that control of VOCs may be an effective way to control O<sub>3</sub> in Houston. In the afternoon, O<sub>3</sub> production was found to be mainly NO<sub><i>x</i></sub> sensitive with some exceptions. O<sub>3</sub> production at near major emissions sources such as Deer Park was mostly VOC sensitive for the entire day, other urban areas near Moody Tower and Channelview were VOC sensitive or in the transition regime, and areas farther from downtown Houston such as Smith Point and Conroe were mostly NO<sub><i>x</i></sub> sensitive for the entire day. It was also found that the control of NO<sub><i>x</i></sub> emissions has reduced O<sub>3</sub> concentrations over Houston, but led to larger OPE values. The results from this work strengthen our understanding of O<sub>3</sub> production; they indicate that controlling NO<sub><i>x</i></sub> emissions will provide air quality benefits over the greater Houston metropolitan area in the long run, but in selected areas controlling VOC emissions will also be beneficial.</p>

New insights into the column CH2O/NO2 ratio as an indicator of near‐surface ozone sensitivity

“…Thus, even though we are unable to pigeonhole the time‐dependent O 3 sensitivity at Moody Tower as being radical limited or NO x limited, one could still deduce that reducing NO x emissions, especially later in the day, would be effective for reducing local O 3 . It should be noted, however, that our conclusions about P(O 3 ) in Houston do not always agree with the results presented in a similar study, particularly in the morning [ Mazzuca et al ., ]. At Deer Park, for example, Mazzuca et al reported a morning median P(O 3 ) value of 30 ppbv/h, while we report a morning median value of 17.2 ppbv/h.…”

“…This discrepancy is also present at every other site (with the exception of Galveston and Conroe), where our modeled P(O 3 ) in the morning tends to be 5–10 ppbv/h lower than the reported values in Mazzuca et al . []. This discrepancy can be attributed to differences in model chemistry: In a previous ground‐based study conducted at Moody Tower, the Carbon Bond mechanism, version 5 (CB05, also used in Mazzuca et al .…”

“…This discrepancy can be attributed to differences in model chemistry: In a previous ground‐based study conducted at Moody Tower, the Carbon Bond mechanism, version 5 (CB05, also used in Mazzuca et al . []) was found to predict higher radical abundances than other mechanisms, including the LaRC box model [ Ren et al ., ]. This may be attributed to the simplified alkene chemistry used in CB05, which has been shown to lead to overpredictions of O 3 production in the morning in locations where alkenes dominate VOC reactivity (such as Deer Park) [ Heo et al ., ].…”

“…Additional contributions to the discrepancy between our work and Mazzuca et al . [] include (1) Mazzuca et al performed additional filtering of their data beyond what is stated in their text, which pushed their statistics toward more polluted air masses and (2) many of the VOCs used as model inputs in Mazzuca et al . [] were derived from Community Multiscale Air Quality Model (CMAQ) outputs (including some highly reactive species like ethene and propene), while our VOC inputs were entirely derived from airborne measurements (X. Ren, private communication, 2017).…”

Ozone Production Efficiencies at Rural New York State Locations: Relationship to Oxides of Nitrogen Concentrations

“…Estimating the OPE as ∆O 3 /∆NO z is known to be an upper limit for describing how efficiently O 3 can be produced for multiple reasons. First, if there is considerable loss of NO z due to deposition of HNO 3 , the OPE is biased high (Mazzuca et al, ; Neuman et al, ; Trainer et al, , ). Second, approximating the OPE as ∆O 3 /∆NO z does not account for mixing and fresh emissions (Zaveri et al, ).…”