Predicted Ozone Concentrations Research Articles

Identifying the geospatial relationship of surface ozone pollution in China: Implications for key pollution control regions

Surface ozone pollution, as a pressing environmental concern, has garnered widespread attention across China. Due to air mass transport, effective control of ozone pollution is highly dependent on collaborative efforts across neighboring regions. However, specific regions with strong internal interactions of ozone pollution are not yet well identified. Here, we introduced the Geospatial SHapley Additive exPlanation (GeoSHAP) approach, which primarily involves machine learning and geostatistical algorithms. Based on extensive atmospheric environmental monitoring data from 2017 to 2021, machine learning models were employed to train and predict ozone concentrations at the target location. The R2 values on the test sets of different scale regions all reached 0.98 in the overall condition, indicating that the core model has good accuracy and generalization ability. The results highlight key regions with high ozone geospatial relationship (OGR) index, predominantly located in the Northern District (ND), spanning the Fen-Wei Plain, the Loess Plateau, and the North China Plain, as well as within portions of the Yangtze River Delta (YRD) and the Pearl River Delta (PRD). Further investigation indicated that high geospatial relationships stem from a synergy between anthropogenic and natural factors, with anthropogenic factors serving as a pivotal element. This study revealed key regions with the most urgent need for joint control of anthropogenic sources to mitigate ozone pollution.

Enhancing ozone nowcasting over East Asia using a data-to-data translation approach with observations from a geostationary environment monitoring spectrometer

The degradation of air quality caused by excessive anthropogenic ozone (O3) concentrations negatively affects ecosystems and the atmosphere. To monitor this issue, the Geostationary Environment Monitoring Spectrometer (GEMS) aboard the Geostationary Korea Multi-Purpose Satellite (GK)-2B provides ozone products over Asia-centered circular regions based on the ozone algorithm derived from representative polar-orbit atmospheric environmental satellites. In this study, we developed a few-hour GEMS O3 nowcasting model using data-to-data (D2D) translation with a conditional generative adversarial network. This model is based on hourly GEMS O3 time-series products and can be used to predict ozone concentrations. The D2D model underwent training and testing employing paired input and output datasets of GEMS O3, with data collected from March 22, 2020, to June 21, 2020, and from March 22, 2021, to June 18, 2021, respectively. The resulting D2D ozone nowcasting model was used to determine ozone concentrations in time zones where GEMS O3 products were unavailable. Test results of the D2D model demonstrated excellent statistical scores, including a bias of 2.162 Dobson Units (DU, where 1 DU corresponds to 2.687 × 1016 molecules/cm2), root-mean-square error (RMSE) of 5.606 DU, a correlation coefficient (CC) of 0.994 for 1-h prediction, a bias of 1.421 DU, RMSE of 5.903 DU, and CC of 0.992 for 2-h prediction, and a bias of 1.169 DU, RMSE of 6.797 DU, and CC of 0.988 for 3-h prediction. Despite the dataset pairing and number limitations, the D2D prediction model accurately forecasted GEMS O3 within 3 h in East Asia.

Prediction of tropospheric ozone concentration with Bagging-MLP method

Prediction of tropospheric ozone concentration with Bagging-MLP method

Spatiotemporal Analysis of Ozone Concentration Using Semi-Functional Partial Linear Models

As weather warms up in China, ozone pollution rises to the top of the list of air pollutants. In this research, we examine the spatiotemporal variability of particulate matter components using contemporary functional data analysis techniques. The technique models the yearly pollutant profiles to describe their dynamic behavior over time and location. These cutting-edge methods offer dimension reduction for better data display and permit us to forecast annual profiles for locations and years for which data are lacking. In order to accurately estimate hourly ozone concentrations for 12 stations in China over two years (2015-2016), this study set out to showcase the best prediction models currently available. To accurately predict Ozone concentration, several methods are used, including Kernel Functional Classical Estimation (KFCE), Kernel Functional Quantile estimation (KFQE), Semi-Partial Linear Functional Classical Estimation (SPLFCE), Semi-Partial Linear Functional Quantile Estimation (SPLFQE), and Semi-Partial Linear Functional Expectile Estimation (SPLFEE). These functional models were chosen based on their ability to establish a forecast region with a given level of confidence. In terms of prediction accuracy, we may conclude that the Semi-Partial linear models outperform conventional models.

Land-Use Regression Analysis of Summer Tropospheric Ozone Concentrations in Ireland

Tropospheric ozone is a powerful oxidant that can damage living organisms; it is widely monitored, as air concentrations have more than doubled since the Industrial Revolution. However, in general air quality monitoring stations are limited spatially to large urban centres; accordingly, accurate prediction of concentrations outside of cities is important for protecting human and plant health. Land-use regression has been successfully used for modelling air pollutant concentrations by establishing a relationship between observed concentrations and landscape features representing sources and sinks. In this study, we developed a land-use regression model that explained 68% of the variance of summer average ozone concentrations in the Republic of Ireland. Ozone was measured at 14 active and 20 passive monitoring sites; air concentrations varied spatially, with the highest ozone measured in rural upland (64.5 µg/m3) and Atlantic coastal (50.2–60.5 µg/m3) sites and the lowest generally in urban centres (38.9–45.7 µg/m3). A total of 74 land-use predictor variables were tested, and their inclusion in the model was based on their impact on the coefficient of determination (R2). The final model included variables linked primarily to deposition processes and included “forest woodland and scrub area” and “distance to coast”. The meteorological variable “rain” and an indicator for NOx emissions “distance to EPA Integrated Pollution Control facilities” were also included in the final model. Our results demonstrate the potential effectiveness of land-use regression modelling in predicting ozone concentrations, at a scale relevant for ecosystem protection.

Variation trend prediction of ground-level ozone concentrations with high-resolution using landscape pattern data.

Scientifically configuring landscape patterns based on their relationship with ground-level ozone concentrations (GOCs) is an effective way to prevent and control ground-level ozone pollution. In this paper, a GOC variation trend prediction model (hybrid model) combining a generalized linear model (GLM) and a logistic regression model (LRM) was established to analyze the spatiotemporal variation patterns in GOCs as well as their responses to landscape patterns. The model exhibited satisfactory performance, with percent of samples correctly predicted (PCP) value of 82.33% and area under receiver operating characteristics curve (AUC) value of 0.70. Using the hybrid model, the per-pixel rise probability of annual average GOCs at a spatial resolution of 1 km in Shenzhen were generated. The results showed that (1) annual average GOCs were increasing in Shenzhen from 2015 to 2020, and had obvious spatial differences, with a higher value in the west and a lower value in the east; (2) variation trend in GOCs was significant positively correlated with landscape heterogeneity (HET), while significant negatively correlated with dominance (DMG) and contagion (CON); (3) GOCs in Shenzhen has a great risk of rising, especially in GuangMing, PingShan, LongGang, LuoHu and BaoAn. The results provide not only a preliminary index for estimating the GOC variation trend in the absence of air quality monitoring data but also guidance for landscape optimizing design from the perspective of controlling ground-level ozone pollution.

Short-Term Forecasting of Ozone Concentration in Metropolitan Lima Using Hybrid Combinations of Time Series Models

In the modern era, air pollution is one of the most harmful environmental issues on the local, regional, and global stages. Its negative impacts go far beyond ecosystems and the economy, harming human health and environmental sustainability. Given these facts, efficient and accurate modeling and forecasting for the concentration of ozone are vital. Thus, this study explores an in-depth analysis of forecasting the concentration of ozone by comparing many hybrid combinations of time series models. To this end, in the first phase, the hourly ozone time series is decomposed into three new sub-series, including the long-term trend, the seasonal trend, and the stochastic series, by applying the seasonal trend decomposition method. In the second phase, we forecast every sub-series with three popular time series models and all their combinations In the final phase, the results of each sub-series forecast are combined to achieve the results of the final forecast. The proposed hybrid time series forecasting models were applied to four Metropolitan Lima monitoring stations—ATE, Campo de Marte, San Borja, and Santa Anita—for the years 2017, 2018, and 2019 in the winter season. Thus, the combinations of the considered time series models generated 27 combinations for each sampling station. They demonstrated significant forecasts of the sample based on highly accurate and efficient descriptive, statistical, and graphic analysis tests, as a lower mean error occurred in the optimized forecast models compared to baseline models. The most effective hybrid models for the ATE, Campo de Marte, San Borja, and Santa Anita stations were identified based on their superior out-of-sample forecast results, as measured by RMSE (4.611, 3.637, 1.495, and 1.969), RMSPE (4.464, 11.846, 1.864, and 15.924), MAE (1.711, 2.356, 1.078, and 1.462), and MAPE (14.862, 20.441, 7.668, and 76.261) errors. These models significantly outperformed other models due to their lower error values. In addition, the best models are statistically significant (p < 0.05) and superior to the rest of the combination models. Furthermore, the final proposed models show significant performance with the least mean error, which is comparatively better than the considered baseline models. Finally, the authors also recommend using the proposed hybrid time series combination forecasting models to predict ozone concentrations in other districts of Lima and other parts of Peru.

A New Hybrid Forecasting Model Based on Dual Series Decomposition with Long-Term Short-Term Memory

In recent years, ozone (O3) has gradually become the primary pollutant plaguing urban air quality. Accurate and efficient ozone prediction is of great significance to the prevention and control of ozone pollution. The air quality monitoring network provides multisource pollutant concentration monitoring data for ozone prediction, but ozone prediction based on multisource monitoring data still faces the challenges of each station’s series of data. Aiming at the problems of low prediction accuracy and low computational efficiency in traditional atmospheric ozone concentration prediction, ozone concentration prediction using dual series decomposition was proposed by variational mode decomposition (VMD), ensemble empirical mode decomposition (EEMD), and long short-term memory (LSTM). First, the historical data series of Nanjing air quality monitoring stations is decomposed by VMD, and then the EEMD algorithm is applied to the residual of VMD to obtain several characteristic intrinsic mode function (IMF) components; each characteristic IMF component is trained by LSTM to obtain the prediction result of each component, and then the final result can be obtained by linear superposition. The proposed method achieved the best results with R2 = 99%, MSE = 5.38, MAE = 4.54, and MAPE = 3.12. Because LSTM has strong adaptive learning ability and good memory function, it has the learning advantage of long-term memory for long-term data, and the prediction results are more accurate. According to the data, the proposed method is superior to the baseline models in terms of statistical metrics. As a result, the proposed hybrid method can serve as a reliable model for ozone forecasting.

A comparison of machine learning methods for ozone pollution prediction

Precise and efficient ozone (hbox {O}_{3}) concentration prediction is crucial for weather monitoring and environmental policymaking due to the harmful effects of high hbox {O}_{3} pollution levels on human health and ecosystems. However, the complexity of hbox {O}_{3} formation mechanisms in the troposphere presents a significant challenge in modeling hbox {O}_{3} accurately and quickly, especially in the absence of a process model. Data-driven machine-learning techniques have demonstrated promising performance in modeling air pollution, mainly when a process model is unavailable. This study evaluates the predictive performance of nineteen machine learning models for ozone pollution prediction. Specifically, we assess how incorporating features using Random Forest affects hbox {O}_{3} concentration prediction and investigate using time-lagged measurements to improve prediction accuracy. Air pollution and meteorological data collected at King Abdullah University of Science and Technology are used. Results show that dynamic models using time-lagged data outperform static and reduced machine learning models. Incorporating time-lagged data improves the accuracy of machine learning models by 300% and 200%, respectively, compared to static and reduced models, under RMSE metrics. And importantly, the best dynamic model with time-lagged information only requires 0.01 s, indicating its practical use. The Diebold-Mariano Test, a statistical test used to compare the forecasting accuracy of models, is also conducted.

WITHDRAWN: Machine Learning with Spatial Interpolation Techniques for Constructing 2-Dimensional Ozone Concentrations in Southern California during the COVID-19 Shutdown

In this study, we combine machine learning and geospatial interpolations to create a two-dimensional high-resolution ozone concentration fields over the South Coast Air Basin for the entire year of 2020. Three spatial interpolation methods (bicubic, IDW, and ordinary kriging) were employed. The predicted ozone concentration fields were constructed using 15 building sites, and random forest regression was employed to test predictability of 2020 data based on input data from past years. Spatially interpolated ozone concentrations were evaluated at twelve sites that were independent of the actual spatial interpolations to find the most suitable method for SoCAB. Ordinary kriging interpolation had the best performance overall for 2020: concentrations were overestimated for Anaheim, Compton, LA North Main Street, LAX, Rubidoux, and San Gabriel sites and underestimated for Banning, Glendora, Lake Elsinore, and Mira Loma sites. The model performance improved from the West to the East, exhibiting better predictions for inland sites. The model is best at interpolating ozone concentrations inside the sampling region (bounded by the building sites), with R2 ranging from 0.56 to 0.85 for those sites, as prediction deficiencies occurred at the periphery of the sampling region, with the lowest R2 of 0.39 for Winchester. All the interpolation methods poorly predicted and underestimated ozone concentrations in Crestline during summer (up to 19 ppb). Poor performance for Crestline indicates that the site has a distribution air pollution levels independent from all other sites. Therefore, historical data from coastal and inland sites should not be used to predict ozone in Crestline using data-driven spatial interpolation approaches. The study demonstrates the utility of machine learning and geospatial techniques for evaluating air pollution levels during anomalous periods.

Graph convolutional network – Long short term memory neural network- multi layer perceptron- Gaussian progress regression model: A new deep learning model for predicting ozone concertation

Ozone is one of the most important air pollutants. The high ozone concertation (OZC) affects the environment and public health. Since OZC depends on the number of different variables, it is difficult and complex to predict ozone concertation. A multilayer perceptron model is robust for predicting ozone concertation, although it has some limitations. The multi-layer perceptron (MLP) model may be unable to extract spatial and temporal features from time series data. Additionally, modelers need to developed models for handling unstructured and structured data. The purpose of this study is to combine deep learning models with MLP models to extract spatial and temporal features and improve the accuracy of predictions. We coupled deep learning models with the MLP model to overcome the limitations of the MLP model. We propose two new hybrid deep models for predicting ozone concertation. Graph convolutional network (GCN)-Long short term memory neural network (LSTM)-multi layer perceptron model (MLP)- Gaussian progress models (GPR) and convolutional neural network (CNN)-LSTM-MLP-GPR are introduced to predict hourly ozone concentration. Novelties of the papers include the development of a new model for spatial feature extraction, the introduction of a new model for temporal feature extraction, and the quantification of uncertainty. Spatial features can be extracted using CNN and GCN models. The LSTM model is a robust model for extracting temporal features. The GPR model quantifies uncertainty values. We use the models to predict ozone concentrations in three stations of Malaysia and one basin in Iran (Sefidrood basin). For the first case study (Malaysian stations), air pollution data and meteorological data were used. GCN-LSTM-MLP-GPR decreased the training MAE of CNN-LSTM-MLP-GPR, CNN-MLP-GPR, GNC-LSTM, CNN-LSTM, GCN-MLP, CNN-MLP, GCN-GPR, CNN-GPR, LSTM-GPR, MLP-GPR, GCNN, CNN, LSTM, MLP, and GPR by 1.7%, 9.2%, 14%, 15%, 18%, 21%, 22%, 27%, 30%, 32%, 34%, 35%, 37%, 42%, and 50% respectively at station 1. At station 2, the PBIAS of the CNN-LSTM-MLP-GPR and GCN-LSTM-LSTM-GPR was 2 and 4%, respectively. The testing MAE of the GCN-LSTM, CNN-LSTM, GCN-MLP, and CNN-MLP was 0.130, 0.131, 0.133, and 0.141 at station 3. In the Sefidrrod basin, the GCN-LSTM-MLP-GPR decreased the MAE of the GCN, CNN, LSTM, MLP, and GPR models by 53%, 55%, 56%, 57%, and 59%, respectively. The uncertainty analysis indicated that the uncertainty of the GCN-LSTM-MLP-GPR model was lower than that of the CNN-LSTM-MLP-GPR, CNN-MLP-GPR, GNC-LSTM, CNN-LSTM, GCN-MLP, CNN-MLP, GCN-GPR, CNN-GPR, LSTM-GPR, MLP-GPR, GCNN, CNN, LSTM, MLP, and GPR model. The GCN-LSTM-MLP-GPR model was a reliable tool for extracting data and predicting ozone concentration. The CGN-LSTM-MLP-GPR is a robust model for spatial and temporal predictions. It can be used as an early warning system to monitor air quality.

Assessing uncertainty and heterogeneity in machine learning-based spatiotemporal ozone prediction in Beijing-Tianjin- Hebei region in China

Accurate prediction of spatiotemporal ozone concentration is of great significance to effectively establish advanced early warning systems and regulate air pollution control. However, the comprehensive assessment of uncertainty and heterogeneity in spatiotemporal ozone prediction remains unknown. Here, we systematically analyze the hourly and daily spatiotemporal predictive performances using convolutional long short term memory (ConvLSTM) and deep convolutional generative adversarial network (DCGAN) models over the Beijing-Tianjin-Hebei region in China from 2013 to 2018. In extensive scenarios, our results show that the machine learning-based (ML-based) models achieve better spatiotemporal ozone concentration prediction performance with multiple meteorological conditions. A further comparison to the air pollution model-Nested Air Quality Prediction Modelling System (NAQPMS) and monitoring observations, the ConvLSTM model demonstrates the practical feasibility of identifying high ozone concentration distribution and capturing spatiotemporal ozone variation patterns at a high spatial resolution (here 15 km × 15 km).

Representing chemical history in ozone time-series predictions – a model experiment study building on the MLAir (v1.5) deep learning framework

Abstract. Tropospheric ozone is a secondary air pollutant that is harmful to living beings and crops. Predicting ozone concentrations at specific locations is thus important to initiate protection measures, i.e. emission reductions or warnings to the population. Ozone levels at specific locations result from emission and sink processes, mixing and chemical transformation along an air parcel's trajectory. Current ozone forecasting systems generally rely on computationally expensive chemistry transport models (CTMs). However, recently several studies have demonstrated the potential of deep learning for this task. While a few of these studies were trained on gridded model data, most efforts focus on forecasting time series from individual measurement locations. In this study, we present a hybrid approach which is based on time-series forecasting (up to 4 d) but uses spatially aggregated meteorological and chemical data from upstream wind sectors to represent some aspects of the chemical history of air parcels arriving at the measurement location. To demonstrate the value of this additional information, we extracted pseudo-observation data for Germany from a CTM to avoid extra complications with irregularly spaced and missing data. However, our method can be extended so that it can be applied to observational time series. Using one upstream sector alone improves the forecasts by 10 % during all 4 d, while the use of three sectors improves the mean squared error (MSE) skill score by 14 % during the first 2 d of the prediction but depends on the upstream wind direction. Our method shows its best performance in the northern half of Germany for the first 2 prediction days. Based on the data's seasonality and simulation period, we shed some light on our models' open challenges with (i) spatial structures in terms of decreasing skill scores from the northern German plain to the mountainous south and (ii) concept drifts related to an unusually cold winter season. Here we expect that the inclusion of explainable artificial intelligence methods could reveal additional insights in future versions of our model.

Spatio-Temporal Prediction of Ground-Level Ozone Concentration Based on Bayesian Maximum Entropy by Combining Monitoring and Satellite Data

Ozone (O3) pollution is one of the predominant environmental problems, and exposure to high O3 concentrations has a significant negative influence on both human health and ecosystems. Therefore, it is essential to analyze spatio-temporal characteristics of O3 distribution and to evaluate O3 exposure levels. In this study, O3 monitoring and satellite data were used to estimate O3 daily, seasonal and one-year exposure levels based on the Bayesian maximum entropy (BME) model with a spatial resolution of 1 km × 1 km in the Beijing-Tianjin-Hebei (BTH) region, China. Leave-one-out cross-validation (LOOCV) results showed that R2 for daily and one-year exposure levels were 0.81 and 0.69, respectively, and the corresponding values for RMSE were 19.58 μg/m3 and 4.40 μg/m3, respectively. The simulation results showed that the heavily polluted areas included Tianjin, Cangzhou, Hengshui, Xingtai, and Handan, while the clean areas were mainly located in Chengde, Qinhuangdao, Baoding, and Zhangjiakou. O3 pollution in summer was the most severe with an average concentration of 134.5 μg/m3. In summer, O3 concentrations in 87.7% of the grids were more than 100 μg/m3. In contrast, winter was the cleanest season in the BTH region, with an average concentration of 51.1 μg/m3.

Clustering-based spatial transfer learning for short-term ozone forecasting

Ground-level ozone is a critical atmospheric pollutant, and high concentrations of ozone can damage human health, affect plant growth and cause ecological harm. Traditional chemical transport models and popular machine learning models have difficulty in predicting ozone concentrations, especially in times with high concentrations. We proposes a clustering-based spatial transfer learning Multilayer Perceptron (SPTL-MLP) to predict ozone concentration at the target observation station for the next three days. We use k-means clustering algorithm to find similar stations and train them together to get a base model for spatial transfer learning. For practical applications, a weighted loss function has been designed with an extra emphasis on reducing prediction errors of high ozone concentrations. Evaluation using historical data of stations in Germany shows that our SPTL-MLP model has a smaller error (reduced by 9.13%) and higher prediction accuracies of ozone exceedances (improved by 8.21% and 16.9%) compared to MLP (without spatial transfer). The results demonstrate the effectiveness of the SPTL-MLP in the short-term ozone forecast. It can be used for timely warning of ozone exceedances and help governments to detect air quality.

Research Trends, Hotspots and Frontiers of Ozone Pollution from 1996 to 2021: A Review Based on a Bibliometric Visualization Analysis

A total of 2932 publications in the field of ozone pollution were obtained from the Web of Science and briefly reviewed using a bibliometric analysis and WOS-based citation reports. CiteSpace 5.7.R3 (64 bit) was used to perform a visualization of knowledge mapping by keywords co-words, burst analysis, co-cited reference analysis, and cooperation network analysis. Research topics in this field have mainly focused on three aspects: risk assessment of ozone pollution for both humans and plants under short-term and long-term exposure; ozone pollution characterization and modeling of ozone transport on different scales; and elucidating the mechanism of ozone formation and source apportionment. By clustering the co-cited references using the data from 2016 to 2021, the frontiers are found to be: (1) VOCs’ precursors and ozone transformation mechanism; (2) modeling of source apportionment and source-oriented chemical transport considering meteorological influence to predict ozone concentration at different spatial and temporal scales; and (3) premature mortality and health burden with relation to ozone exposure. It should be mentioned that an emerging research hotspot is the utilization of artificial intelligence (AI) tools (e.g., machine learning, deep learning, etc.) to facilitate the modeling of big data at different scales.

Correction to: A hybrid CNN-Transformer model for ozone concentration prediction

Correction to: A hybrid CNN-Transformer model for ozone concentration prediction

Feature selection for global tropospheric ozone prediction based on the BO-XGBoost-RFE algorithm

Ozone is one of the most important air pollutants, with significant impacts on human health, regional air quality and ecosystems. In this study, we use geographic information and environmental information of the monitoring site of 5577 regions in the world from 2010 to 2014 as feature input to predict the long-term average ozone concentration of the site. A Bayesian optimization-based XGBoost-RFE feature selection model BO-XGBoost-RFE is proposed, and a variety of machine learning algorithms are used to predict ozone concentration based on the optimal feature subset. Since the selection of the underlying model hyperparameters is involved in the recursive feature selection process, different hyperparameter combinations will lead to differences in the feature subsets selected by the model, so that the feature subsets obtained by the model may not be optimal solutions. We combine the Bayesian optimization algorithm to adjust the parameters of recursive feature elimination based on XGBoost to obtain the optimal parameter combination and the optimal feature subset under the parameter combination. Experiments on long-term ozone concentration prediction on a global scale show that the prediction accuracy of the model after Bayesian optimized XGBoost-RFE feature selection is higher than that based on all features and on feature selection with Pearson correlation. Among the four prediction models, random forest obtained the highest prediction accuracy. The XGBoost prediction model achieved the greatest improvement in accuracy.

Prediction of Ozone Hourly Concentrations Based on Machine Learning Technology

To optimize the accuracy of ozone (O3) concentration prediction, this paper proposes a combined prediction model of O3 hourly concentration, FC-LsOA-KELM, which integrates multiple machine learning methods. The model has three parts. The first part is the feature construction (FC), which is based on correlation analysis and incorporates time-delay effect analysis to provide a valuable feature set. The second part is the kernel extreme learning machine (KELM), which can establish a complex mapping relationship between feature set and prediction object. The third part is the lioness optimization algorithm (LsOA), which is purposed to find the optimal parameter combination of KELM. Then, we use air pollution data from 11 cities on Fenwei Plain in China from 2 January 2015 to 30 December 2019 to test the validity of FC-LsOA-KELM and compare it with other prediction methods. The experimental results show that FC-LsOA-KELM can obtain better prediction results and has a better performance.

Forecasting ground-level ozone concentration levels using machine learning

Ground-level ozone (GLO) has been widely recognized as a critical air pollutant that has the potential to induce various adverse environmental and health effects. To eliminate its hazardous impacts, the development of an accurate and effective approach to forecast the upcoming pollution concentration levels is in urgent need. Recent studies show that machine learning algorithms have excellent abilities in ground-level ozone concentration forecasting, however many of their forecasting models do not consider the contributing effects on meteorological measurements and traffic situations which were also identified as potential influencing factors to ground-level ozone concentration by some previous research. In the meantime, most of the existing models target short-term forecasting with rough temporal resolutions, such as daily and weekly scales. This paper aims to propose a methodology that can provide long-term GLO concentration forecasting with a high temporal resolution. To achieve this, a frequent pattern mining approach is utilized to analyze the local intercorrelation between GLO concentration and contributing factors such as meteorological parameters and transportation situations. Then, a series of machine learning algorithms were identified to forecast the ground-level ozone concentration levels using traffic and meteorological measurements data. A case study was conducted in the Houston region with 10 years of historical measurements, each of the historical ground-level ozone concentration records is associated with a series of meteorology and traffic situation parameters. Data from 2010 to 2019 was used to select and train the machine learning models, and data from 2020 was used to perform the final validation and evaluation. Results show that the extreme gradient boost (XGBoost) machine learning algorithm provides the most accurate prediction of the hourly ground-level ozone concentration on a yearly scale, which shows the annually forecasting ability and the robustness of the model built. The proposed approach could be applied to other similar regions and other critical air pollutants that are also influenced by transportation and meteorological factors.