Pollutant Concentration Data Research Articles

The dynamic evolution characteristics of PM2.5 concentrations and health risk assessment during typical forest fires in China

In the context of climate change, increasingly frequent wildfire events are exacerbating the air quality crisis in China and have become a significant source of atmospheric PM2.5. This study selected 12 major forest wildfire events in China from 2018 to 2023. Using atmospheric pollutant and component concentration data, as well as meteorological data, the study employed MK trend analysis, forward trajectory simulation, potential source contribution function (PSCF), and health risk indices to investigate the evolution patterns of PM2.5 concentrations before and after each wildfire event. The study explored the driving role of meteorological factors in this evolution process and quantitatively analyzed the spatial distribution characteristics of health risks in surrounding cities post-wildfire. The results indicated that: 1) PM2.5 levels changed significantly before and after each wildfire event, although the proportion of components remained relatively stable. 2) PM2.5 concentrations exhibited varying characteristics within different buffer zones of the wildfire areas. In the forest wildfire events in Foshan, Guangdong Province, and Jinzhong, Shanxi Province, the average PM2.5 concentrations within the 50 km buffer zones reached approximately 90 μg/m3, about 4–5 times the multi-year average for these areas. 3) Wildfire emissions were mainly influenced by meteorological and topographical factors. Wind field dependence diagrams showed that PM2.5 tended to accumulate in all wind directions, particularly at lower wind speeds of 0–5 m/s 4) Potential source analysis revealed that PM2.5 emissions from wildfires posed significant risks to surrounding areas. Multiple events had high-risk areas (WPSCF ≥0.8), and among the cities in these high-risk areas, the health risk value (ΔM) for Ya'an City reached an extremely high 85.1. This study provides a theoretical basis for policymakers to develop locally tailored wildfire management policies, ensuring the protection of the public's right to breathe clean air.

Ground visibility prediction using tree-based and random-forest machine learning algorithm: Comparative study based on atmospheric pollution and atmospheric boundary layer data

To mitigate haze impacts, three visibility simulation schemes were designed using decision tree and random forest algorithms, leveraging atmospheric boundary layer meteorological data, pollutant concentrations, and ground observations. The optimal approach was identified to investigate the boundary layer's effect on simulations. The results showed that the simulation effect of the random forest algorithm for two haze processes was better than that of the decision tree algorithm. In the first haze process, the random forest algorithm had a more significant reduction in root mean square error than the decision tree algorithm in the same visibility range (Scheme 3, visibility<200 m, mean absolute error reduced by 5.9%, root mean square error reduced by 19.1%). Simulation models significantly improved the accuracy of the models by adding atmospheric boundary layer observation data to the two fog-hazes process visibility. However, the addition of atmospheric boundary layer meteorological data in the first haze process had a better improvement effect (random forest: visibility<200 m, mean absolute errors of 25.0 (relative error<12.5%) and 25.5 m (relative error<12.8%) in Scheme 2 and 3, respectively). The addition of atmospheric boundary-layer pollutant concentrations data was more effective in the second haze process (random forest: visibility<200 m, scheme 2 and scheme 3 had mean absolute errors of 25.6 (relative error<12.8%) and 11.1 m (relative error<5.6%), respectively). The influence of atmospheric boundary layer meteorological data and pollutant data on the two fog processes is affected by the cause of the fog process.

Environmental Insights: Democratizing access to ambient air pollution data and predictive analytics with an open-source Python package

Ambient air pollution is a pervasive issue with wide-ranging effects on human health, ecosystem vitality, and economic structures. Utilizing data on ambient air pollution concentrations, researchers can perform comprehensive analyses to uncover the multifaceted impacts of air pollution across society. To this end, we introduce Environmental Insights, an open-source Python package designed to democratize access to air pollution concentration data. This tool enables users to easily retrieve historical air pollution data and employ a Machine Learning model for forecasting potential future conditions. Moreover, Environmental Insights includes a suite of tools aimed at facilitating the dissemination of analytical findings and enhancing user engagement through dynamic visualizations. This comprehensive approach ensures that the package caters to the diverse needs of individuals looking to explore and understand air pollution trends and their implications. Code repository clickable linkEnvironmental Insights Github Home Page.

Pollution concentration monitoring using a new Birnbaum–Saunders control chart

AbstractAir pollution monitoring is an important issue in environmental science. The Birnbaum–Saunders (BS) distribution, originally applied to describe product failure time distribution to fatigue failures and general random wear failures, is also well to describe the pollutant concentration data due to accumulations of various pollutants in the air over time. Sulfur dioxide (SO2) is a critical factor in air pollution. Hence, it is important to monitor its concentration variation for air pollution prevention. Due to the complexity of its distribution form, there is no reliable and easy‐to‐use control chart for monitoring pollutant concentrations based on the BS distribution. We found that the SO2 concentration data follows the BS distribution. In this study, we propose a new median control chart based on the exact sampling distribution of the monitoring statistic to detect shifts in the median of BS distribution. Thus, given the false alarm rate, the control limits for such control charts can be obtained precisely satisfying a preset in‐control average run length using Monte Carlo simulations. The out‐of‐control average run lengths are calculated by simulation to evaluate the detection performance of the proposed chart when the median shifts occur. We further compare the detection performance of the proposed chart and those of the existing control charts based on asymptotic sampling distributions. In order to improve the detection ability of the proposed chart for small median shifts, an exponentially weighted moving average (EWMA) control chart is constructed. The results of numerical analyses demonstrated that the proposed EWMA chart performs much better than all existing control charts for monitoring the median of BS distribution. Finally, the proposed control charts are applied to monitor the median of SO2 concentrations for air pollution control, showing that both charts can effectively detect a shift in the median of SO2 concentrations. The proposed EWMA control chart even detects out a small shift in the median of SO2 concentrations. The results provide a continuous monitoring solution for air pollution prevention.

Unit Maxwell-Boltzmann Distribution and Its Application to Concentrations Pollutant Data

In the vast statistical literature, there are numerous probability distribution models that can model data from real-world phenomena. New probability models, nevertheless, are still required in order to represent data with various spread behaviors. It is a known fact that there is a great need for new models with limited support. In this study, a flexible probability model called the unit Maxwell-Boltzmann distribution, which can model data values in the unit interval, is derived by selecting the Maxwell-Boltzmann distribution as a base-line model. The important characteristics of the derived distribution in terms of statistics and mathematics are investigated in detail in this study. Furthermore, the inference problem for the mentioned distribution is addressed from the perspectives of maximum likelihood, method of moments, least squares, and maximum product space, and different estimators are obtained for the unknown parameter of the distribution. The derived distribution outperforms competitive models according to different fit tests and information criteria in the applications performed on four actual air pollutant concentration data sets, indicating that it is an effective model for modeling air pollutant concentration data.

Evaluation of the Effectiveness of Feature Selection Methods Combined with Regression Algorithms to Predict Particulate Matter (PM10) in Gandhinagar, Gujarat, India

Feature selection is one of the important data pre-processing techniques that are used to increase the performance of machine learning models, to build faster and more cost-effective algorithms, and to make it easier to interpret the predictions made by the models. The main objective of this research work is to investigate the influence features to predict particulate matter (PM10). This research uses 24-hour average pollutant concentration data of 36 air quality monitoring stations provided by Gandhinagar Smart City Development Limited (GSCDL), Gandhinagar, Gujarat. Important features were identified using five feature selection techniques (correlation, forward selection, backward elimination, Exhaustive Feature Selection (EFS), and feature importance derived using Random Forest Regressor). With selected features six regression algorithms (Multiple Linear Regression, Random Forest, Decision Tree, K-nearest Neighbour, XGBoost, and Support Vector Regressor) were trained to predict PM10. Further, the models were compared based on the Root Mean Square Error (RMSE) and Coefficient of determination (R2) parameters to identify the model with good performance. This proposed model can be utilized as an early warning system, providing air quality information to local authorities to develop air-quality improvement initiatives.

Rapid weather changes are associated with daily hospital visitors for atrial fibrillation accompanied by abnormal ECG repolarization: a case-crossover study

BackgroundAtrial fibrillation (AF) is highly prevalent in the population, yet the factors contributing to AF events in susceptible individuals remain partially understood. The potential relationship between meteorological factors and AF, particularly with abnormal electrocardiograph (ECG) repolarization, has not been adequately studied. This case-crossover study aims to investigate the association between meteorological factors and daily hospital visits for AF with abnormal ECG repolarization in Shanghai, China.MethodsThe study cohort comprised 10,325 patients with ECG-confirmed AF who sought treatment at Shanghai Sixth People's Hospital between 2015 and 2018. Meteorological and air pollutant concentration data were matched with the patient records. Using a case-crossover design, we analyzed the association between meteorological factors and the daily count of hospital visitors for AF with abnormal ECG repolarization at our AF center. Lag analysis models were applied to examine the temporal relationship between meteorological factors and AF events.ResultsThe analysis revealed statistically significant associations between AF occurrence and specific meteorological factors. AF events were significantly associated with average atmospheric pressure (lag 0 day, OR 0.9901, 95% CI 0.9825–0.9977, P < 0.05), average temperature (lag 1 day, OR 0.9890, 95% CI 0.9789–0.9992, P < 0.05), daily pressure range (lag 7 days, OR 1.0195, 95% CI 1.0079–1.0312, P < 0.01), and daily temperature range (lag 5 days, OR 1.0208, 95% CI 1.0087–1.0331, P < 0.01). Moreover, a significant correlation was observed between daily pressure range and daily temperature range with AF patients, particularly those with abnormal ECG repolarization, as evident in the case-crossover analysis.ConclusionThis study highlights a significant correlation between meteorological factors and daily hospital visits for AF accompanied by abnormal ECG repolarization in Shanghai, China. In addition, AF patients with abnormal ECG repolarization were found to be more vulnerable to rapid daily changes in pressure and temperature compared to AF patients without such repolarization abnormalities.

Unveiling the air pollution tapestry in China: A comprehensive assessment of spatiotemporal variations through geographically and temporally weighted regression

Study of air pollution informs environmentally-conscious policies and urban planning by government and businesses, leading to more scientific decision-making. This paper comprehensively analyzes the spatiotemporal correlation of six air pollutants (PM2.5, PM10, SO2, NO2, CO, and O3) at the national level of China from 2015 to 2021. The temporal changes in environmental air pollutants are examined through statistical data analysis. Secondly, the annual and quarterly spatial distribution characteristics of air pollutants are analyzed by utilizing global and local spatial autocorrelation analysis and visualizing pollutant concentration data. To quantify the spatiotemporal correlation information obtained from the data, we proposed a novel framework based on the Geographically and Temporally Weighted Regression (GTWR) model to provide a local approach for detecting the spatiotemporal correlation between urban air pollutants on a large scale. The framework's experimental results showed that the significant correlation between urban air pollutants in time and space ranged from 0.45 to 0.9, which is better than traditional correlation algorithm, and the degree of local spatiotemporal correlations can be explained by spatiotemporal coefficients. We find noticeable differences in the correlations between different pollutants and geographical boundaries in the spatial dimension. Furthermore, in terms of spatiotemporal location, PM2.5, PM10, NO2, SO2, and CO exhibit positive correlations with each other, while O3 shows both positive and negative correlations with other pollutants. This study offers an important reference for air quality monitoring and prediction, contributing to the improvement of accuracy and timeliness of air quality warnings.

Estimation of pollution sources and hydraulic conductivity field in a coastal aquifer under tidal effects

In the past decades, most of the efforts have often been invested in the joint identification of contamination and hydraulic conductivity field in inland subsurface aquifers. However, the booming economy and dense population around the coast have brought a new challenge to the forefront: the serious pollution of coastal aquifers. Coastal aquifers are highly vulnerable ecological areas, which are affected by inland pollutants carried by groundwater during its entry into the sea and are also influenced by seawater intrusion. This article addresses the inverse problem of coastal aquifer pollution, which is subject to tidal effects, using the Iterative Local Updating Ensemble Smoother (ILUES) algorithm. The adopted ILUES algorithm is leveraged to efficiently extract pollutant concentration and salinity data for the identification of the source release history and characterization of the hydraulic conductivity field of a site with multiple pollutant sources. This study is based on a variant of the traditional benchmark problem, analyzing the influence of algorithm parameters, tidal effects, and heterogeneity on data assimilation. The article further demonstrates the great potential of ILUES to address new challenges. In addition, the significance of multi-level monitoring wells for the estimation of hydraulic conductivity in coastal multi-layered heterogeneous aquifers is explained.

An integrated framework for predicting air quality index using pollutant concentration and meteorological data

An integrated framework for predicting air quality index using pollutant concentration and meteorological data

Seasonal Characteristics of Long-Range Transport and Potential Associated Sources of Particulate Matter (Pm&lt;sub&gt;10&lt;/sub&gt;) Pollution at the Station Elk, Poland, on 2021-2022 Data

The current study aimed to determine the potential sources of distant emissions of PM10 particles that significantly affect PM10 levels at a given site in southeastern Baltic. The EEA Air Quality Monitoring Station in Elk City, northeastern Poland, was selected for this study. This station is located approximately 50 km from the border of the Russian exclave (Kaliningrad Region). In this study, the NOAA HYSPLIT_4 trajectory model, potential source contribution function (PSCF), and concentration-weight trajectory (CWT) were employed to investigate the origin of the measured PM10 mass at a receptor site. PSCF and CWT utilize back-trajectory analysis and Lagrangian particle dispersion simulations to reconstruct the advection pathways of air masses arriving at the site. These reconstructed retroplumes provide detailed information regarding the geographic locations traversed by polluted air masses on their way to the receptor. By integrating trajectory information with concurrent pollutant concentration data, the PSCF and CWT enable the identification of potential source regions and quantification of their impact on the observed atmospheric levels. From January 1, 2021, to December 31, 2022, at 200 m the 72h backward trajectories of air masses entering the receptor point were calculated and categorized by clustering them into 5-4-4-5 clusters. Subsequently, the PM10 levels at the Elk site associated with each air mass cluster were examined during the observation period. The seasonal variation in PM10 was generally characterized by a peak in winter and minimum values in summer. PM10 was lower during warmer periods, particularly during summer, and significantly, higher concentrations were observed during colder periods. Cluster analyses showed that airflow followed a seasonal pattern, with different results obtained in different seasons. According to the PSCF and CWT results, in winter and spring, the receptor site was influenced more by long-range PM10 pollution, particularly from heavily industrialized areas in Central-Eastern Europe. In contrast, in summer and autumn, the receptor site was less influenced by long-range pollution. The findings demonstrate that the seasonal distributions of PM10 source areas obtained using these two methods generally share similar characteristics, suggesting the credibility and accuracy of the analytical results.

Air Quality Monitoring Using Machine Learning Techniques

Abstract-Air pollution (AP) is a major global environmental issue, drawing attention from researchers due to its impact on human health. Predicting air quality is crucial to inform people about health risks and protect against the effects of air pollution, particularly in metropolitan cities facing severe environmental challenges. Real-time monitoring of pollution data enables local authorities to analyze traffic conditions and make informed decisions. In this study, we developed a machine learning model to forecast the air quality index of India, which is a standard measure of pollutant levels (e.g., SO2, NO2) over a specific period. Our model is based on historical data from previous years and uses Gradient Descent-boosted multivariable regression to predict the air quality index for an upcoming year. To improve model efficiency, we applied cost estimation to the predictive problem. The proposed model demonstrates high accuracy, achieving 96% on the current available dataset for predicting India's air quality index. Moreover, we utilized XG Boost and Light GBM algorithms to determine the order of preference based on similarity to the ideal solution, further enhancing the model's performance. Our model has the capability to predict the air quality index for an entire county, state, or any bounded region, given historical pollutant concentration data. By implementing parameter-reducing formulations, we outperformed standard regression models, making our approach a valuable tool for air quality prediction and environmental decision-making to protect human health. Keywords—Machine Learning, Air Quality, Air quality prediction, Monitoring System, Intelligent System.

Ventilation Strategies for Highly Occupied Public Environments: A Review

In urban public transportation and highly diversified air environments, air pollutant exposure is becoming an increasing concern in terms of public health and personal safety. Herein, the scientific literature on air quality and virus transmission in densely crowded environments is reviewed to determine effective control methods. The research results are classified on the basis of different crowded environments. Much research has been conducted on pollutants in subways and buses. High particulate matter concentrations in public transportation are still a serious problem, but few studies on the spread of viruses exist. With existing types of ventilation systems, increasing local exhaust may be an efficient way to remove pollutants. Air quality sensors should be distributed in densely crowded spaces to achieve real-time display of pollutant concentration data. When pollution levels exceed the safe values, scientifically designed ventilation and filtration schemes should be implemented to reduce the pollution levels. Occupant activities are among the important factors that make pollutant transmission more complex. The analysis results herein contribute to the assessment of indoor pollutant concentrations and the protection of occupants from cross-infection.

Spatial and temporal characteristics analysis and prediction model of PM2.5 concentration based on SpatioTemporal-Informer model.

The primary cause of hazy weather is PM2.5, and forecasting PM2.5 concentrations can aid in managing and preventing hazy weather. This paper proposes a novel spatiotemporal prediction model called SpatioTemporal-Informer (ST-Informer) in response to the shortcomings of spatiotemporal prediction models commonly used in studies for long-input series prediction. The ST-Informer model implements parallel computation of long correlations and adds an independent spatiotemporal embedding layer to the original Informer model. The spatiotemporal embedding layer captures the complex dynamic spatiotemporal correlations among the input information. In addition, the ProbSpare Self-Attention mechanism in this model can focus on extracting important contextual information of spatiotemporal data. The ST-Informer model uses weather and air pollutant concentration data from numerous stations as its input data. The outcomes of the trials indicate that (1) The ST-Informer model can sharply capture the peaks and sudden changes in PM2.5 concentrations. (2) Compared to the current models, the ST-Informer model shows better prediction performance while maintaining high-efficiency prediction [Formula: see text]. (3) The ST-Informer model has universal applicability, and the model was applied to the concentration of other pollutants prediction with good results.

Association between ambient air pollutants and upper respiratory tract infection and pneumonia disease burden in Thailand from 2000 to 2022: a high frequency ecological analysis

BackgroundA pertinent risk factor of upper respiratory tract infections (URTIs) and pneumonia is the exposure to major ambient air pollutants, with short term exposures to different air pollutants being shown to exacerbate several respiratory conditions.MethodsHere, using disease surveillance data comprising of reported disease case counts at the province level, high frequency ambient air pollutant and climate data in Thailand, we delineated the association between ambient air pollution and URTI/Pneumonia burden in Thailand from 2000 – 2022. We developed mixed-data sampling methods and estimation strategies to account for the high frequency nature of ambient air pollutant concentration data. This was used to evaluate the effects past concentrations of fine particulate matter (PM2.5), sulphur dioxide (SO2), and carbon monoxide (CO) and the number of disease case count, after controlling for the confounding meteorological and disease factors.ResultsAcross provinces, we found that past increases in CO, SO2, and PM2.5 concentration were associated to changes in URTI and pneumonia case counts, but the direction of their association mixed. The contributive burden of past ambient air pollutants on contemporaneous disease burden was also found to be larger than meteorological factors, and comparable to that of disease related factors.ConclusionsBy developing a novel statistical methodology, we prevented subjective variable selection and discretization bias to detect associations, and provided a robust estimate on the effect of ambient air pollutants on URTI and pneumonia burden over a large spatial scale.

Significant variation in air quality in South Indian cities during COVID-19 lockdown and unlock phases.

With the spread of COVID-19 pandemic worldwide, the Government of India had imposed lockdown in the month of March 2020 to curb the spread of the virus furthermore. This shutdown led to closure of various institutions, organizations, and industries, and restriction on public movement was also inflicted which paved way to better air quality due to reduction in various industrial and vehicular emissions. To brace this, the present study was carried out to statistically analyze the changes in air quality from pre-lockdown period to unlock 6.0 in South Indian cities, namely, Bangalore, Chennai, Coimbatore, and Hyderabad, by assessing the variation in concentration of PM2.5, PM10, NO2, and SO2 during pre-lockdown, lockdown, and unlock phases. Pollutant concentration data was obtained for the selected timeframe (01 March 2020-30 November 2020) from CPCB, and line graph was plotted which had shown visible variation in the concentration of pollutants in cities taken into consideration. Analysis of variance (ANOVA) was applied to determine the mean differences in the concentration of pollutants during eleven timeframes, and the results indicated a significant difference (F (10,264) = 3.389, p < 0.001). A significant decrease in the levels of PM2.5, PM10, NO2, and SO2 during the lockdown phases was asserted by Tukey HSD results in Bangalore, Coimbatore, and Hyderabad stations, whereas PM10 and NO2 significantly increased during lockdown period in Chennai station. In order to understand the cause of variation in the concentration of pollutants and to find the association of pollutants with meteorological parameters, the Pearson correlation coefficient was used to study the relationship between PM2.5, PM10, NO2, and SO2 concentrations, temperature, rainfall, and wind speed for a span of 15months, i.e., from January 2020 to March 2021. At a significant level of 99.9%, 99%, and 95%, a significant correlation among the pollutants, rainfall had a major impact on the pollutant concentration in Bangalore, Coimbatore, Hyderabad, and Chennai followed by wind speed and temperature. No significant influence of temperature on the concentration of pollutants was observed in Bangalore station.

Spatial and Temporal Variation of Aerosol Optical Depth in Huaihai Economic Zone from 1982 to 2021

Aerosol Optical Depth (AOD), quantifying the amount of aerosol in the atmosphere, is widely regarded as a crucial indicator for research on atmospheric physics and regional air quality. At present, the inversion of AOD from observation of satellite remote sensing sensors has become the main technology for large-scale monitoring of aerosol load. The Huaihai Economic Zone is the connecting belt of two key areas of atmospheric governance (the Yangtze River Delta and the Beijing-Tianjin-Hebei region, China), and it has been suffering from air pollution for many years and few studies of AOD focus on this region. Therefore, the spatial and temporal characteristics of the AOD are explored using MODIS AOD data and AVHRR AOD data in this region during the period from 1982 to 2021 in this study. The correlation coefficients between the AOD of satellite observation and actual air pollution were analyzed by combining PM2.5 pollutant concentration and air quality index (AQI) data. The results showed that the AOD is higher in the northwest than in the southeast, and it is different from season to season. The annual variation of AOD in the Huaihai Economic Zone is a W-shaped trend from 1982 to 2011, while the trend of annual AOD is decreasing after 2011. In terms of seasons, the whole differences in AOD are evident, exhibiting AOD values in summer &gt; those in spring &gt; those in autumn &gt; and those in winter. Furthermore, it indicated that the quarterly and monthly variation of the AOD tends to be flat in recent years. Since 2015, the concentration of PM2.5 has continued to decline, the same as that of AQI. Meanwhile, the quarterly and monthly differences in PM2.5 are still obvious, with higher PM2.5 in winter and lower PM2.5 in summer. However, it also represented that PM2.5 is significantly higher in spring than in autumn from 2015 to 2018, which is the opposite for 2019 to 2021. Lastly, the correlation between AOD and PM2.5/AQI is also given; i.e., the correlation coefficients of AOD with PM2.5/AQI are 0.84/0.82, with the highest correlation coefficient in autumn (R = 0.86/0.91) and the lowest in winter (R = 0.46/0.48).

A spatial correlation prediction model of urban [formula omitted] concentration based on deconvolution and LSTM

Precise prediction of air pollutants can effectively reducre the occurrence of heavy pollution incidents. With the current surge of massive data, deep learning appears to be a promising technique to achieve dynamic prediction of air pollutant concentration from both the spatial and temporal dimensions. This paper presents Dev-LSTM, a prediction model building on deconvolution and LSTM. The novelty of Dev-LSTM lies in its capability to fully extract the spatial feature correlation of air pollutant concentration data, preventing the excessive loss of information caused by traditional convolution. At the same time, the feature associations in the time dimension are mined to produce accurate prediction results. Experimental results show that Dev-LSTM outperforms traditional prediction models on a variety of indicators.

An ensemble interval prediction model with change point detection and interval perturbation-based adjustment strategy: A case study of air quality

Point prediction has been used to predict air pollutant concentrations in recent years. However, it is still a challenge to characterize the time series data of pollutant concentrations in the presence of high volatility and uncertainty. Since interval prediction can quantity this uncertainty and provide more information than point prediction, we propose an improved interval prediction model based on lower upper bound estimation (LUBE) to construct the prediction intervals (PIs) of PM2.5 concentrations. First, we decompose the original time series into a trend term and a fluctuation term. Then, the trend term with regularity is analyzed by point prediction, while the fluctuation term with uncertainty is studied by LUBE interval prediction. To improve the efficiency and stabilize the randomness in the existing LUBE, we propose a new method based on change point detection and interval perturbation-based adjustment strategy (IPAS). IPAS is used to replace the optimization algorithm in LUBE for improving efficiency. Meanwhile, change point detection is introduced to optimize the initialized parameters in LUBE. In addition, since PM2.5 concentration is influenced by various factors, partial autocorrelation function and maximal information coefficient are applied to select the optimal input features from meteorological and other air pollution factors associated with PM2.5 concentrations. Moreover, we ensemble the prediction results of the trend term and fluctuation term to obtain the ultimate PIs. To evaluate the effectiveness and efficiency of our proposed model, the daily PM2.5 concentration data in Wuhan, China, are analyzed in the empirical study. Comparison results and ablation study clearly indicate that our proposed model has better comprehensive performance than classical models and benchmark models. The proposed new method can provide high-quality PIs and achieve better stability.

COVID-19 Pandemic: Impacts on Air Quality during Partial Lockdown in the Metropolitan Area of São Paulo

Air pollution has become one of the factors that most affect the quality of life, human health, and the environment. Gaseous pollutants from motor vehicles have a significantly harmful effect on air quality in the Metropolitan Area of São Paulo (MASP)—Brazil. Motor vehicles emit large amounts of particulate matter (PM), carbon monoxide (CO), nitrogen oxides (NOx), and volatile organic compounds (VOCs), the last three acting as the main tropospheric ozone (O3) precursors. In this study, we evaluated the effects of these pollutants on air quality in the MASP during the partial lockdown that was imposed to ensure the social distancing necessitated by the COVID-19 pandemic. We compared the monthly data for nitrogen dioxide (NO2) from the Ozone Monitoring Instrument (OMI) and CO, SO2, and BC from MERRA-2 for the period between April and May 2020 (during the pandemic) with the average for the same period for the (pre-pandemic) years 2017 to 2019 in the southeast region of Brazil. The meteorological and pollutant concentration data from the CETESB air quality monitoring stations for the MASP were compared with the diurnal cycle of three previous years, with regard to the monthly averages of April and May (2017, 2018, and 2019) and the same period in 2020, when the partial lockdown was first imposed in southeast Brazil. Our findings showed that there was a decrease in NO2 concentrations ranging from 10% to more than 60% in the MASP and the Metropolitan Area of Rio de Janeiro (MARJ), whereas in the Metropolitan Area of Belo Horizonte and Vitoria (MABH and MAV, respectively), there was a reduction of around 10%. In the case of the concentrations of CO and BC from MERRA-2, there was a considerable decrease (approx. 10%) during the period of partial lockdown caused by COVID-19 throughout almost the entire state of São Paulo, particularly in the region bordering the state of Rio de Janeiro. The concentration of SO2 from MERRA-2 was 5 to 10% lower in the MASP and MARJ and the west of the MABH, and there was a decrease of 30 to 50% on the border between the states of São Paulo and Rio de Janeiro, while in the MAV region, there was an increase in pollutant levels, as this region was not significantly affected by the COVID-19 pandemic. Sharp reductions in the average hourly concentrations of CO (38.8%), NO (44.9%), NO2 (38.7%), and PM2.5 (6%) were noted at the CETESB air quality monitoring stations in the MASP during the partial lockdown in 2020 compared with the hourly average rate in the pre-pandemic period. In contrast, there was an increase of approximately 16.0% in O3 concentrations in urban areas that are seriously affected by vehicular emissions, which is probably related to a decrease in NOx.