Understanding ozone episodes during the TRACER-AQ campaign in Houston, Texas: The role of transport and ozone production sensitivity to precursors

This study investigated transport pathways and photochemical formation responsible for ozone exceedances during the September 2021 deployment of the Tracking Aerosol Convection Interactions ExpeRiment/Air Quality (TRACER-AQ) campaign in Houston, Texas. We focused on two ozone episodes, September 6th-September 11th ("Episode 1") and September 23rd-September 26th ("Episode 2"), when the maximum daily eight-hour average (MDA8) ozone at surface monitors exceeded 70 ppbv. Long-range transport patterns of air masses during these episodes were from the central/northern US. High-resolution (4 km resolution) trajectory analysis with FLEXible PARTicle (FLEXPART) dispersion model revealed local recirculation of air masses and the accumulation of pollutants across Houston contribute to the ozone exceedances. Comprehensive Air Quality Model with extensions (CAMx) driven by 1.33-km resolution meteorology from the Weather Research and Forecast (WRF) tool simulated elevated ozone production rates during ozone episodes across the Houston metropolitan area, with ozone production hotspots mostly over Houston city and industrial districts of the Houston Ship Channel (HSC). The regional increase in ozone production rates was due to the transport of VOC-rich air masses (via northerly flows) that brought ozone precursors to the region, which ultimately caused a transition in the ozone formation tendency from generally VOC-limited to NOx-limited conditions. However, the city of Houston and the HSC remained in a VOC-limited regime because of local NOx emissions that, to some extent, preponderated the impact of transported VOCs. While approximately 37 % of the elevated ozone production was attributed to local photochemistry, the remaining ∼63 % increase in ozone production was due to the transported ozone to the region during episodes, bringing ozone to the Houston region and contributing to ozone exceedances. The outcomes of this study illustrated the synergy between transport and ozone production, both long-range and local scale, which resulted in ozone exceedances in Houston.

Effect of photochemical losses of ambient volatile organic compounds on their source apportionment

Photochemical losses of ambient volatile organic compounds (VOCs) substantially affect source apportionment analysis. Hourly speciated VOC data measured from April to August 2020 in Tianjin, China were used to analyze the photochemical losses of VOC species and assess the impacts of photochemical losses on source apportionment by comparing the positive matrix factorization (PMF) results based on observed and initial concentration data (OC-PMF and IC-PMF). The initial concentrations of the VOC species were estimated using a photochemical age-based parameterization method. The results suggest that the average photochemical loss of total VOCs (TVOCs) during the ozone pollution period was 2.4 times higher than that during the non-ozone pollution period. The photochemical loss of alkenes was more significant than that of the other VOC species. Temperature has an important effect on photochemical losses, and different VOC species have different sensitivities to temperature; high photochemical losses mainly occurred at temperatures between 25 °C and 35 °C. Photochemical losses reduced the concentrations of highly reactive species in the OC-PMF factor profile. Compared with the IC-PMF results, the OC-PMF contributions of biogenic emissions and polymer production-related industrial sources were underestimated by 73 % and 50 %, respectively, likely due to the oxidation of isoprene and propene, respectively. The contribution of diesel and gasoline evaporation was underestimated by 39 %, which was likely due to the loss of m,p-xylene. Additionally, the contributions of liquefied petroleum gas, vehicle emissions, natural gas, and oil refinery were underestimated by 31 %, 29 %, 23 %, and 13 %, respectively. When the O₃ concentrations were higher than 140 μg m^-3 or the temperatures were higher than 30 °C, the photochemical losses from most sources increased substantially. Additionally, solar radiation produced different photochemical losses for different source types.

Contributions of meteorology to ozone variations: Application of deep learning and the Kolmogorov-Zurbenko filter

From hourly ozone observations obtained from three regions⸻Houston, Dallas, and West Texas⸻we investigated the contributions of meteorology to changes in surface daily maximum 8-h average (MDA8) ozone from 2000 to 2019. We applied a deep convolutional neural network and Shapely additive explanation (SHAP) to examine the complex underlying nonlinearity between variations of surface ozone and meteorological factors. Results of the models showed that between 2000 and 2019, specific humidity (38% and 27%) and temperature (28% and 37%) contributed the most to ozone formation over the Houston and Dallas metropolitan areas, respectively. On the other hand, the results show that solar radiation (50%) strongly impacted ozone variation over West Texas during this time. Using a combination of the Kolmogorov-Zurbenko (KZ) filter and multiple linear regression, we also evaluated the influence of meteorology on ozone and quantified the contributions of meteorological parameters to trends in surface ozone formation. Our findings showed that in Houston and Dallas, meteorology influenced ozone variations to a large extent. The impacts of meteorology on West Texas, however, showed meteorological factors had fewer influences on ozone variabilities from 2000 to 2019. This study showed that SHAP analysis and the KZ approach can investigate the contributions of the meteorological factors on ozone concentrations and help policymakers enact effective ozone mitigation policies.

Traffic, transport, and vegetation drive VOC concentrations in a major urban area in Texas

The population of Texas has increased rapidly in the past decade. The San Antonio Field Study (SAFS) was designed to investigate ozone (O₃) production and precursors in this rapidly changing, sprawling metropolitan area. There are still many questions regarding the sources and chemistry of volatile organic compounds (VOCs) in urban areas like San Antonio which are affected by a complex mixture of industry, traffic, biogenic sources and transported pollutants. The goal of the SAFS campaign in May 2017 was to measure inorganic trace gases, VOCs, methane (CH₄), and ethane (C₂H₆). The SAFS field design included two sites to better assess air quality across the metro area: an urban site (Traveler's World; TW) and a downwind/suburban site (University of Texas at San Antonio; UTSA). The results indicated that acetone (2.52 ± 1.17 and 2.39 ± 1.27 ppbv), acetaldehyde (1.45 ± 1.02 and 0.93 ± 0.45 ppbv) and isoprene (0.64 ± 0.49 and 1.21 ± 0.85 ppbv; TW and UTSA, respectively) were the VOCs with the highest concentrations. Additionally, positive matrix factorization showed three dominant factors of VOC emissions: biogenic, aged urban mixed source, and acetone. Methyl vinyl ketone and methacrolein (MVK + MACR) exhibited contributions from both secondary photooxidation of isoprene and direct emissions from traffic. The C₂H₆:CH₄ demonstrated potential influence of oil and gas activities in San Antonio. Moreover, the high O₃ days during the campaign were in the NO_x-limited O₃ formation regime and were preceded by evening peaks in select VOCs, NO_x and CO. Overall, quantification of the concentration and trends of VOCs and trace gases in a major city in Texas offers vital information for general air quality management and supports strategies for reducing O₃ pollution. The SAFS campaign VOC results will also add to the growing body of literature on urban sources and concentrations of VOCs in major urban areas.

Ground-based lidar for atmospheric boundary layer ozone measurements

Ground-based lidars are suitable for long-term ozone monitoring as a complement to satellite and ozonesonde measurements. However, current ground-based lidars are unable to consistently measure ozone below 500 m above ground level (AGL) due to both engineering issues and high retrieval sensitivity to various measurement errors. In this paper, we present our instrument design, retrieval techniques, and preliminary results that focus on the high-temporal profiling of ozone within the atmospheric boundary layer (ABL) achieved by the addition of an inexpensive and compact mini-receiver to the previous system. For the first time, to the best of our knowledge, the lowest, consistently achievable observation height has been extended down to 125 m AGL for a ground-based ozone lidar system. Both the analysis and preliminary measurements demonstrate that this lidar measures ozone with a precision generally better than ±10% at a temporal resolution of 10 min and a vertical resolution from 150 m at the bottom of the ABL to 550 m at the top. A measurement example from summertime shows that inhomogeneous ozone aloft was affected by both surface emissions and the evolution of ABL structures.

Cluster Analysis of Surface Winds in Houston, Texas, and the Impact of Wind Patterns on Ozone

The city of Houston, Texas, is near a complex coastline and numerous petrochemical plants, the combination of which plays a large role in Houston’s air pollution events. It has long been known that the thermally driven afternoon onshore flow (sea breeze or gulf breeze) transports ozone-rich air inland. As a way of quantifying the role of the gulf breeze in Houston’s high-ozone events, cluster analysis of hourly averaged surface winds from a regional network of meteorological sensors was performed for 27 summer days of 2000, with the dates coinciding with the Texas Air Quality Study 2000 (TexAQS 2000). Hourly averaged winds were partitioned into 16 independent clusters, or wind patterns, while simultaneously keeping track of the maximum ozone in the network for each hour. Clusters emerged that represented various wind patterns, including thermally driven flows, stagnant winds, and a thunderstorm outflow. All clusters were used to assess which wind patterns were most likely to be coincident with the maximum ozone of the day. High ozone was most likely to occur with clusters representing the gulf breeze. Clusters occurring before the ozone maximum of the day were analyzed to determine which sequences of wind patterns were most likely to precede high ozone. A transition from offshore flow to onshore flow, with at least 1 h of stagnant winds in between, routinely occurred in the 6 h preceding ozone measurements reaching ≥ 120 parts per billion by volume (ppbv). On nontransition days with high ozone, ozone maxima ≥ 120 ppbv often occurred the hour after a wind direction shift of greater than about 45°.