Houston-Galveston-Brazoria Research Articles

Characterizing Overwater High Ozone Events in the Houston–Galveston–Brazoria Region during the 2021 GO3 and TRACER-AQ Campaigns

Abstract Photochemical modeling outputs showing high ozone concentrations over the Gulf of Mexico and Galveston Bay during ozone episodes in the Houston–Galveston–Brazoria (HGB) region have not been previously verified using in situ observations. Such data were collected systematically, for the first time, from July to October 2021 from three boats deployed for the Galveston Offshore Ozone Observations (GO3) and Tracking Aerosol Convection Interactions Experiment—Air Quality (TRACER-AQ) field campaigns. A pontoon boat and a commercial vessel operated in Galveston Bay, while another commercial vessel operated in the Gulf of Mexico offshore of Galveston. All three boats had continuously operating sampling systems that included ozone analyzers and weather stations, and the two boats operating in Galveston Bay had a ceilometer. The sampling systems operated autonomously on the two commercial boats as they traveled their daily routes. Thirty-seven ozonesondes were launched over water on forecast high ozone days in Galveston Bay and the Gulf of Mexico. During the campaigns, multiple periods of ozone exceeding 100 ppbv were observed over water in Galveston Bay and the Gulf of Mexico. These events included previously identified conditions for high ozone events in the HGB region, such as the bay/sea-breeze recirculation and postfrontal environments, as well as a localized coastal high ozone event after the passing of a tropical system (Hurricane Nicholas) that was not well forecast.

Long-term trend in surface ozone in Houston-Galveston-Brazoria: Sectoral contributions based on changes in volatile organic compounds

This study investigated the long-term variations in ambient levels of surface ozone, volatile organic compounds (VOCs), and nitrogen oxides (NOx) within the Houston-Galveston-Brazoria (HGB) region. Analysis of ozone levels revealed an overall reduction in the maximum daily 8-h average ozone (MDA8 O3) from 2000 to 2019 (April–October) with an average rate of ∼ −0.48 ppb/yr across HGB. With a few exceptions, the MDA8 O3 reduction rates were more pronounced for the monitoring sites closer to the Houston Ship Channel (HSC). Meanwhile, ambient levels of NOx and most VOC species (across the three representative sites as Houston Bayland Park, Haden Road, and Lynchburg Ferry) decreased significantly within the same investigation period, reflecting the impact of emission reductions. The positive matrix factorization (PMF) model applied to the mentioned sites identified regional background ozone, petrochemical emissions, engine combustion, natural gas/fuel evaporation, and solvent/painting/rubber industries as the major sources of MDA8 O3. The regional background ozone was the predominant source, accounting for 59–70% of MDA8 O3 across the three sites. Regarding the local anthropogenic emissions, natural gas/fuel evaporation was the largest contributor (19.5 ± 6.1%) to MDA8 O3 at Houston Bayland Park, whereas petrochemical facilities (10.9 ± 4.9%) and solvent/painting/rubber industries (18.1 ± 9.5%) were the largest factor at Haden Road and Lynchburg Ferry, respectively. Notable reductions were found in the contributions of petrochemical emissions, engine combustion, and natural gas/fuel evaporation to MDA8 O3 within 2000–2019, but an increasing trend was revealed in the role of solvent/painting/rubber industries on MDA8 O3 most probably due to the enhanced demand for their products. Results of this study corroborated the success of emission control policies in limiting ozone precursors and provided useful details for prioritizing emission reduction policies to further reduce ozone pollution in the HGB.

Assessing the impact of episodic flare emissions on ozone formation in the Houston-Galveston-Brazoria area of Texas

Ozone formation due to episodic industrial emissions was modeled for the Houston-Galveston-Brazoria (HGB) region in Texas. A total of more than 1000 synthetic emission scenarios were modeled, accounting for different sources, emission magnitudes, emission compositions, duration and timing of the emission events, meteorology at the time of the emissions, and location of the emission source. Episodic emissions consistent with the characteristics of flaring resulted in the greatest amount of ozone formation. An index, based on incremental reactivity, for characterizing the ozone formation potential of flaring emissions was developed using the synthetic scenarios and was applied to the ~4500 flares that report emissions in the state of Texas. The flares with the highest ranking index in the HGB were identified and additional modeling was performed on the ozone formation potential of these flares. If these flares are modeled with their annual average emission rates, the maximum increase in maximum daily 8-h average ozone (MDA8) formation associated with flaring emissions is <1 ppb. If, however, it is assumed that emissions from flaring are episodic, rather than continuous, and temporal patterns consistent with previous data collection on flare emission variability in the HGB are applied, MDA8 enhancements associated with flaring emissions are >10 ppb for multiple scenarios. This analysis suggests that, for flares emissions that have high ozone formation potential, the temporal patterns of flare emissions should be accounted for in developing ozone mitigation plans.

Anthropogenic Source Contributions to Ozone Formation in the Greater Houston Area

The Houston-Galveston-Brazoria (HGB) area of Texas has historically experienced severe air pollution events with high concentrations of ozone (O3) during the summer season. This study evaluates the contribution of different anthropogenic sources to ozone formation in the HGB area. The Emission Processing System (EPS3) is used to process emission files in four different scenarios (Base case as including All emission sources (BC), All sources— Area sources (AMA), All sources—Point sources (AMP), and All sources— Mobile sources (AMM). These files are used as input in photochemical modeling with the Comprehensive Air Quality Model with Extensions (CAMx) to simulate ozone formation. The data is analyzed for daily maximum ozone concentrations and contribution of source categories at three air quality monitoring locations (La Porte Sylvan beach-C556, Houston Texas avenue-C411, and Texas city in Galveston-C683) for a study period of June 1-June 30, 2012. The contribution of the point sources to ozone formation is dominated at all three locations, followed by mobile sources and area sources on high ozone days. The relative contributions of point sources are 27.51% ± 3.53%, 21.45% ± 7.36%, and 30.30% ± 9.36%; and mobile sources are 18.27% ± 2.22%, 20.60% ± 6.89%, and 18.61% ± 7.43%; and area sources were 4.2% ± 1.65%, 5.21% ± 1.59%, and 3.72% ± 1.52% at C556, C411, and C683, respectively. These results demonstrate the importance of regulatory focus on controlling point and mobile source emissions for NAAQS attainment in the study region.

Prediction of High-ozone Events Using GAM, SMOTE, and Tail Dependence Approaches in Texas (2005–2019)

We test three methods for ozone prediction in the El Paso (ELP) and Houston-Galveston-Brazoria (HGB) regions of Texas from 2005–2019: (1) a Generalized Additive Model (GAMs) approach; (2) a GAM approach with the addition of the Synthetic Minority Over-sampling TEchnique (SMOTE) and (3) a tail dependence modeling approach based in extreme value theory (EVT). We also compare the feature selection capabilities of the tail dependence approach to other feature selection methods. We find that the GAM+SMOTE model outperformed the GAM-only model when predicting ozone values for the root mean square error metric, particularly with regard to the above-threshold ozone values, which may be of particularly useful for extreme ozone event prediction. In addition, we find that the improvement of above-threshold MDA8 O3 prediction for the GAM+SMOTE method tends to come at the cost of below-threshold prediction, which is particularly important if MDA8 O3 trends are of interest. We also find that the tail dependence approach is capable of predicting extreme ozone events, but algorithmic stability and configuration complexity can make this approach difficult to operationalize on a broad scale and that the selection of the threshold needs to be carefully considered. Finally, the feature selection via the tail dependence method performs comparably to other forms of machine learning-based feature selection and we find that there are multiple parameter sets that can predict MDA8 O3 with equal success.

The COVID-19 Pandemic Vulnerability Index (PVI) Dashboard: Monitoring County-Level Vulnerability Using Visualization, Statistical Modeling, and Machine Learning.

The COVID-19 Pandemic Vulnerability Index (PVI) Dashboard: Monitoring County-Level Vulnerability Using Visualization, Statistical Modeling, and Machine Learning.

HGBEnviroScreen: Enabling Community Action through Data Integration in the Houston-Galveston-Brazoria Region.

The Houston–Galveston–Brazoria (HGB) region faces numerous environmental and public health challenges from both natural disasters and industrial activity, but the historically disadvantaged communities most often impacted by such risks have limited ability to access and utilize big data for advocacy efforts. We developed HGBEnviroScreen to identify and prioritize regions of heightened vulnerability, in part to assist communities in understanding risk factors and developing environmental justice action plans. While similar in objectives to existing environmental justice tools, HGBEnviroScreen is unique in its ability to integrate and visualize national and local data to address regional concerns. For the 1090 census tracts in the HGB region, we accrued data into five domains: (i) social vulnerability, (ii) baseline health, (iii) environmental exposures and risks, (iv) environmental sources, and (v) flooding. We then integrated and visualized these data using the Toxicological Prioritization Index (ToxPi). We found that the highest vulnerability census tracts have multifactorial risk factors, with common drivers being flooding, social vulnerability, and proximity to environmental sources. Thus, HGBEnviroScreen is not only helping identify communities of greatest overall vulnerability but is also providing insights into which domains would most benefit from improved planning, policy, and action in order to reduce future vulnerability.

Meteorological detrending of long-term (2003-2017) ozone and precursor concentrations at three sites in the Houston Ship Channel Region

ABSTRACTAmbient ozone is influenced by meteorology in addition to concentrations of precursor compounds (oxides of nitrogen (NOx) and volatile organic compounds (VOC)). The efficacy of regulatory measures in nonattainment areas, such as the Houston-Galveston-Brazoria (HGB) area of Texas, can be efficiently evaluated by separating the meteorologically induced variability from ozone data. This study applies the Kolmogorov-Zurbenko (KZ) filter for obtaining a temporal resolution of ozone, NOx and VOC data into short-term, seasonal and long-term components, at three stations located near the Houston Ship Channel. Air quality and meteorological data for Clinton (AQS Site ID: 48-201-1035), Deer Park (48-201-1039) and Lynchburg Ferry (48-201-1015) stations were analyzed for the period between 2003 and 2017. A combination of KZ filter and multiple linear regression, with predictor variables (solar radiation, temperature, dew point, and wind speed) is employed to develop meteorologically independent ozone, NOx and VOC trends. This study indicates that variability from meteorology accounts for 51%, 35% and 41% in baseline MDA8 ozone at Clinton, Deer Park, and Lynchburg stations, respectively. For the 15-year study period, long-term MDA8 ozone trends for Deer Park and Lynchburg stations were decreasing at a linear rate of 0.689 ± 0.016, and 0.573 ± 0.019 ppb/yr, respectively. At the Deer Park and Lynchburg stations, a high degree of correlation for meteorologically detrended MDA8 ozone with NOx (ρ: 0.899, 0.678) and VOC (ρ: 0.912) concentrations was observed. For the Clinton station, decreases in NOx and VOC levels d at the rate of 2.068 ± 0.032 ppb/yr and 14.637 ± 0.412 ppb C/yr, were not reflected in MDA8 ozone, which showed no discernable decrease over the 15 years. The regional transport of ozone plumes from the east and south-east directions of the Clinton station were identified as the likely factors for this pattern.Implications: The efficacy of emission control policies in the Houston-Galveston-Brazoria area can be evaluated by isolating the meteorological forcing from air quality time series data and developing long-term trends for ozone and precursor compounds. This paper applies the Kolmogorov–Zurbenko filter technique in combination with Multiple Linear Regression analysis to MDA8/MDA1 ozone, NOx, and VOC data between 2003-2017 at three air monitoring stations near the Houston Ship Channel. Estimates for trends of air quality are calculated and underlying causes are investigated to provide a guidance for further investigation into air quality management of the Greater Houston Area.

Standard and alternative procedures for projecting future ozone in the Houston area using relative reduction factors

Abstract Air quality model results are used in a relative sense as part of the process to demonstrate attainment of the U.S. ozone standard. Ozone model predictions for a future year are divided by corresponding model results for a base year to obtain a Relative Reduction Factor (RRF), and the base-year measured ozone design value (DV) is multiplied by the RRF to estimate the future-year design value (DVF). We developed alternative RRFs and DVFs that focus on the increment of local ozone production above regional background (RB). An advantage of the alternative DVFs is that measurements can be used to determine RB ozone, thereby increasing the use of measurements in the DVF estimates. We simulated ozone formation in the Houston area for May–September 2012 and 2028 using the Comprehensive Air Quality Model with Extensions and compared the standard and alternative RRFs and DVFs. If model results are used for all the RB ozone concentrations in our new equation to calculate DVFs, the alternative and standard DVFs are very similar for Houston-Galveston-Brazoria (HGB). If measurements are used for the RB ozone concentrations we combine with the base-year DVs, the alternative DVFs are noticeably larger than the standard DVFs for about half the sites, with the increase being 7 ppb at the site (Deer Park) with the largest standard and alternative DVFs. The differences arise primarily from including different days in the averaging process for the RRFs versus the base-year DVs. We evaluated model performance for the days included in the RRFs and found that the model over-predicted RB ozone (normalized mean bias (NMB) = 26%) and under-predicted the anthropogenic ozone increment (NMB = -22%) from HGB emissions and therefore can be expected to under-estimate ozone response to HGB emission controls.

Influence of Cold Fronts on Variability of Daily Surface O3 over the Houston-Galveston-Brazoria Area in Texas USA during 2003–2016

We investigated the impacts of cold fronts on area-wide peak O3 and regional backgroundO3 mixing ratios on a daily scale over the Houston-Galveston-Brazoria (HGB) area of southeasternTexas during the O3 seasons (April–October) of 2003–2016. Back trajectories showed that an 18h timelag existed between arrival of cold fronts in the HGB area and onset of a predominately northerlyflow. Cold fronts showed increasing effects on both peak and background O3 over the HGB area.Compared to no front days, average peak O3 mixing ratios during the cold front 1st days, coldfront 2+ days, and post frontal days increased 0.7, 5.9, and 9.0 ppbv, respectively while averagebackground O3 increased 2.9, 6.8, and 8.6 ppbv, respectively. The change in wind direction fromsoutherly to northerly was the most important factor causing increasing O3 levels. Wind directionshifts caused variation of other meteorological factors (i.e., wind speed, precipitation, temperature,cloud cover, and relative humidity) and tended to overshadow their effects on O3 over the HGB area.On a long-term and large-scale view, cold fronts over the HGB area could be regarded as interruptionsin the cleansing effects of predominantly marine southerly flow from the Gulf of Mexico.

Investigation of Primary Factors Affecting the Variation of Modeled Oak Pollen Concentrations: A Case Study for Southeast Texas in 2010

Oak pollen concentrations over the Houston-Galveston-Brazoria (HGB) area in southeastern Texas were modeled and evaluated against in-situ data. We modified the Community Multi-scale Air Quality (CMAQ) model to include oak pollen emission, dispersion, and deposition. The Oak Pollen Emission Model (OPEM) calculated gridded oak pollen emissions, which are based on a parameterized equation considering a plant-specific factor (C e ), surface characteristics, and meteorology. The simulation period was chosen to be February 21 to April 30 in the spring of 2010, when the observed monthly mean oak pollen concentrations were the highest in six years (2009-2014). The results indicated C e and meteorology played an important role in the calculation of oak pollen emissions. While C e was critical in determining the magnitude of oak pollen emissions, meteorology determined their variability. In particular, the contribution of the meteorology to the variation in oak pollen emissions increased with the oak pollen emission rate. The evaluation results using in-situ surface data revealed that the model underestimated pollen concentrations and was unable to accurately reproduce the peak pollen episodes. The model error was likely due to uncertainty in climatology-based C e used for the estimation of oak pollen emissions and inaccuracy in the wind fields from the Weather Research and Forecast (WRF) model.

Air Quality Impact of Biomass Co-Firing with Coal at a Power Plant in the Greater Houston Area

The Houston-Galveston-Brazoria (HGB) area of Texas is a moderate nonattainment region for ozone, and has a history of severe summer ozone episodes. W. A. Parish power plant (WAP) located in the greater Houston area is the largest coal and natural gas based electricity generating unit (EGU) in Texas. Forest residue is an abundant renewable resource, and can be used to offset coal usage at EGUs. This study evaluates the impact of co-firing 5%, 10%, and 15% (energy-basis) of forest residue at WAP on the air quality of the HGB area. Photochemical modeling with Comprehensive Air Quality Model with Extensions (CAMx) was conducted to investigate the air quality at three air quality monitoring sites (C696, C53, C556) in the HGB area, under two source scenarios (all-sources, point + biogenic sources). Significant reduction of SO2 and O3 was observed for 10% and 15% co-firing ratios at monitoring station (C696) close to WAP. The maximum reduction of ozone observed for 15% co-firing is 4.7% and 6.3% for all-sources and point + biogenic sources scenarios respectively. The reduction in other criteria air pollutants is not significant at all locations. The overall results from this study indicate that biomass co-firing at WAP would not lead to a significant reduction in ozone concentrations in the region during periods of peak ozone.

Relative Contribution of Different Source Categories to Ozone Exceedances in the Houston-Galveston-Brazoria Area

The goal of this study is to analyze the relative contribution of different emission source categories to ozone in the Houston-Galveston-Brazoria (HGB) area of Texas. Emission Processing System (EPS3) is used to prepare the emission files for five different source combination cases (Base case, Biogenic, Area + Biogenic, Mobile + Biogenic, Low-level Point + Biogenic). These emission files are used to perform photochemical modeling with Comprehensive Air Quality Model with Extensions (CAMx), and the results are analyzed with Visual Environment for Rich Data Interpretation (VERDI) tool. The daily maximum ozone concentrations and individual contributions of the source categories were analyzed over a 15-day study period between June 1-15, 2012, at three locations (University of Houston-Sugarland, Bayland Park and Conroe). Biogenic sources contributed an average of 49.7% ± 12.8%, 43.1% ± 12.0%, and 39.9% ± 9.28% at Sugarland, Bayland Park and Conroe sites respectively, indicating the significance of isoprene emissions from the vegetation in northeast Houston. On peak ozone days, contribution of Mobile + Biogenic source category averages about 80.1% ± 12.6%, 79.9% ± 6.50%, and 75.9% ± 10.9% at Sugarland, Bayland Park and Conroe sites respectively, indicating the dominance of mobile source NOX emissions and the necessity for regulatory focus on mobile source emissions control.

A data mining framework for environmental and geo-spatial data analysis

Mining geo-spatial data is an important task in many application domains, such as environmental science, geographic information science, and social networks. In this paper, we introduce a data mining framework, which includes pre-processing of environmental and geo-spatial data, geo-spatial data mining techniques, and visual analysis of environmental and geo-spatial data. In particular, we propose new density-based clustering algorithms to identify interesting distribution patterns from geo-spatial data, a change pattern discovery technique to detect dynamic change patterns within spatial clusters, and a post-processing technique to extract interesting patterns and useful knowledge from geo-spatial data. Our density-based clustering algorithms are based on the well-established density-based shared nearest neighbor clustering algorithm, which can find clusters of different shape, size, and densities in high-dimensional data. The post-processing analysis technique allows automatic screening of interesting spatial clusters. The change pattern discovery algorithm is able to detect and analyze dynamic patterns of changes within spatial clusters. This paper focuses on developing a framework integrating a sequence of data mining process including clustering algorithm, analysis technique and pattern changing discovery algorithm. In contrast to previous works in this area, our approaches can cluster and analyze dynamically evolved complex objects, i.e., polygons. We evaluate the effectiveness of our techniques through a challenging real case study involving ozone pollution events in the Houston–Galveston–Brazoria area. The experimental results show that our approaches can discover interesting patterns and useful information from geo-spatial air-quality data.

Regional background O&amp;lt;sub&amp;gt;3&amp;lt;/sub&amp;gt; and NO&amp;lt;sub&amp;gt;&amp;lt;i&amp;gt;x&amp;lt;/i&amp;gt;&amp;lt;/sub&amp;gt; in the Houston–Galveston–Brazoria (TX) region: a decadal-scale perspective

Abstract. Ozone (O3) in the lower troposphere is harmful to people and plants, particularly during summer, when photochemistry is most active and higher temperatures favor local chemistry. Local precursor emissions, such as those of volatile organic compounds (VOCs) and nitrogen oxides (NOx), together with their chemistry contribute to the O3 and NOx mixing ratios in the Houston–Galveston–Brazoria (HGB) region. In addition to local emissions, chemistry and transport, larger-scale factors also contribute to local O3 and NOx. These additional contributions (often referred to as regional background) are not well quantified within the HGB region, impeding more efficient controls on precursor emissions to achieve compliance with the National Ambient Air Quality Standards for O3. In this study, we estimate ground-level regional background O3 and NOx in the HGB region and quantify their decadal-scale trends.We use four different approaches based on principal component analysis (PCA) to quantify background O3 and NOx. Three of these approaches consist of independent PCA on both O3 and NOx for both 1 and 8 h levels to compare our results with previous studies and to highlight the effect of both temporal and spatial scales. In the fourth approach, we co-varied O3, NOx and meteorology.Our results show that the estimation of regional background O3 has less inherent uncertainty when it was constrained by NOx and meteorology, yielding a statistically significant temporal trend of −0.68 ± 0.27 ppb yr−1. Likewise, the estimation of regional background NOx trend constrained by O3 and meteorology was −0.04 ± 0.02 ppb yr−1 (upper bound) and −0.03 ± 0.01 ppb yr−1 (lower bound). Our best estimates of the 17-year average of season-scale background O3 and NOx were 46.72 ± 2.08 ppb and 6.80 ± 0.13 ppb (upper bound) or 4.45 ± 0.08 ppb (lower bound), respectively. Average background O3 is consistent with previous studies and between the approaches used in this study, although the approaches based on 8 h averages likely overestimate background O3 compared to the hourly median approach by 7–9 ppb. Similarly, the upper bound of average background NOx is consistent between approaches in this study (A–C) but overestimated compared to the hourly approach by 1 ppb, on average. We likely overestimate the upper-bound background NOx due to instrument overdetection of NOx and the 8 h averaging of NOx and meteorology coinciding with MDA8 O3.Regional background O3 and NOx in the HGB region both have declined over the past 2 decades. This decline became steadier after 2007, overlapping with the effects of controlling precursor emissions and a prevailing southeasterly–southerly flow.

Influence of the Bermuda High on interannual variability of summertime ozone in the Houston–Galveston–Brazoria region

Abstract. The Bermuda High (BH) quasi-permanent pressure system is the key large-scale circulation pattern influencing summertime weather over the eastern and southern US. Here we developed a multiple linear regression (MLR) model to characterize the effect of the BH on year-to-year changes in monthly-mean maximum daily 8 h average (MDA8) ozone in the Houston–Galveston–Brazoria (HGB) metropolitan region during June, July, and August (JJA). The BH indicators include the longitude of the BH western edge (BH-Lon) and the BH intensity index (BHI) defined as the pressure gradient along its western edge. Both BH-Lon and BHI are selected by MLR as significant predictors (p &lt; 0.05) of the interannual (1990–2015) variability of the HGB-mean ozone throughout JJA, while local-scale meridional wind speed is selected as an additional predictor for August only. Local-scale temperature and zonal wind speed are not identified as important factors for any summer month. The best-fit MLR model can explain 61–72 % of the interannual variability of the HGB-mean summertime ozone over 1990–2015 and shows good performance in cross-validation (R2 higher than 0.48). The BH-Lon is the most important factor, which alone explains 38–48 % of such variability. The location and strength of the Bermuda High appears to control whether or not low-ozone maritime air from the Gulf of Mexico can enter southeastern Texas and affect air quality. This mechanism also applies to other coastal urban regions along the Gulf Coast (e.g., New Orleans, LA, Mobile, AL, and Pensacola, FL), suggesting that the BH circulation pattern can affect surface ozone variability through a large portion of the Gulf Coast.

Use of high-order sensitivity analysis and reduced-form modeling to quantify uncertainty in particulate matter simulations in the presence of uncertain emissions rates: A case study in Houston

Abstract Regional air quality models are widely used to evaluate control strategy effectiveness. As such, it is important to understand the accuracy of model simulations to establish confidence in model performance and to guide further model development. Particulate matter with aerodynamic diameter less than 2.5 μm (PM2.5) is regulated as one of the criteria pollutants by the National Ambient Air Quality Standards (NAAQS), and PM2.5 concentrations have a complex dependence on the emissions of a number of precursors, including SO2, NOx, NH3, VOCs, and primary particulate matter (PM). This study quantifies how the emission-associated uncertainties affect modeled PM2.5 concentrations and sensitivities using a reduced-form approach. This approach is computationally efficient compared to the traditional Monte Carlo simulation. The reduced-form model represents the concentration-emission response and is constructed using first- and second-order sensitivities obtained from a single CMAQ/HDDM-PM simulation. A case study is conducted in the Houston-Galveston-Brazoria (HGB) area. The uncertainty of modeled, daily average PM2.5 concentrations due to uncertain emissions is estimated to fall between 42% and 52% for different simulated concentration levels, and the uncertainty is evenly distributed in the modeling domain. Emission-associated uncertainty can account for much of the difference between simulation and ground measurements as 60% of observed PM2.5 concentrations fall within the range of one standard deviation of corresponding simulated PM2.5 concentrations. Uncertainties in meteorological fields as well as the model representation of secondary organic aerosol formation are the other two key contributors to the uncertainty of modeled PM2.5. This study also investigates the uncertainties of the simulated first-order sensitivities, and found that the larger the first-order sensitivity, the lower its uncertainty associated with emissions. Sensitivity of PM2.5 to primary PM has the lowest uncertainty while sensitivity of PM2.5 to VOC has the highest uncertainty associated with emission inputs.

Photochemical model simulations of air quality for Houston–Galveston–Brazoria area and analysis of ozone–NO x –hydrocarbon sensitivity

As part of the State Implementation Plan for the Houston–Galveston–Brazoria (HGB) area which has been declared as a non-attainment area, it was required to assess the impacts of pollutant emissions, meteorological conditions, and initial and boundary conditions on air quality. In this study, photochemical model simulations using the Comprehensive Air Quality Model with extensions (CAMx) were conducted for three ozone episodes between May and October of 2006. The CAMx simulations compared fairly well with the hourly average concentrations of ozone measured at the monitoring sites, and the index of agreement (IOA) values for all the three ozone episodes were found greater than 0.65, which proved to be good for CAMx model performance. Both simulated and observed peak ozone concentrations were observed to occur between 12:00 and 15:00 h. The CAMx simulation results showed that there were 68 days where the maximum 8-h mean ozone concentrations exceeded the NAAQS of 75 ppb out of a total of 72 days simulated in the three ozone episodes. For all three episodes, ozone concentrations were highest during the third quarter of the day. The photochemical indicator ratios of hydrogen peroxide (H2O2) to nitric acid (HNO3) ([H2O2]/[HNO3]) showed the HGB area to be a nitrogen oxide (NO x )-sensitive regime for the episodes studied. This indicated that volatile organic compounds (VOCs) are plentiful, and NO x is the limiting precursor for ozone formation in the HGB area. The plentiful VOCs in this region are possibly produced from industries and refineries in the HGB area.

Prediction of surface ozone episodes using clusters based generalized linear mixed effects models in Houston–Galveston–Brazoria area, Texas

A two–stage strategy is proposed to predict regional peak ozone episodes in the Houston–Galveston–Brazoria (HGB) area of Texas, USA. With the forecasted meteorological information, ozone episodes can be predicted one day in advance. Three generalized linear mixed effects models (GLMMs) are built with air quality and meteorological data monitored at CAMS35, CAMS403 and CAMS1015; wind field data from 8 monitoring sites in HGB area are used to generate clusters which represent distinct weather patterns. Air quality and meteorological data during ozone seasons (Apr. 1st – Oct. 31st) from 2003 to 2005 are used to build site–specific prediction models. Data of ozone season from 2006 to 2007 are used to test these models. Compared to linear regression models (LM), generalized linear models (GLMs), multilayer perceptron (MLP) and support vector machine (SVM), GLMM which considers differences in ozone formation and diffusion in distinct weather patterns has the smallest fitting and prediction error on ozone exceedances and can detect the most number of exceedance days correctly.

Influence of Reactive Volatile Organic Compounds on Ozone Production in Houston-Galveston-Brazoria Area

Secondary pollutant ozone (O3) formation in a particular area is often influenced by various factors. Source of emissions is one of the factors. In south east Texas, Houston-Galveston-Brazoria (HGB) is a marginal non-attainment area for ozone (O3). A summer episode of May 28 to July 2, 2006 is simulated using Comprehensive Air Quality Model with extensions (CAMx). During this period O3 concentration in HGB often exceeds the National Ambient Air Quality Standards (NAAQS) 0.075 ppm of average 8 hour O3 concentration. HGB area has numerous point sources. Various studies found that some specific volatile organic compounds are very reactive in atmosphere. The objective of this study is to analyze the influence of volatile organic compounds present in point source emissions on the air quality of HGB area. For this purpose ozone sensitivity for HGB area is analyzed by the ratio of hydrogen peroxides (H2O2) to nitric acid (HNO3). HGB area is found NOx limited but reactive VOCs are found to be influential too. From (1-4 June, 2006) maximum O3 concentration was found on weekend, June 3. VOCs such as Acetaldehyde (ALD2), Formaldehyde (FORM) and Alkane (ETHA) showed good correlation with O3 concentrations on that day. In addition, Peroxyacetyl nitrate (PAN) formation was found correlated to higher ozone production. Criteria pollutant Sulfur dioxide (SO2) was found to influence the ALD2 and ETHA concentrations, and thus indirectly influenced O3 production.