A Review on Predicting Ground PM2.5 Concentration Using Satellite Aerosol Optical Depth

Interpreting hourly mass concentrations of PM2.5 chemical components with an optimal deep-learning model

PM2.5 constitutes a complex and diverse mixture that significantly impacts the environment, human health, and climate change. However, existing observation and numerical simulation techniques have limitations, such as a lack of data, high acquisition costs, and multiple uncertainties. These limitations hinder the acquisition of comprehensive information on PM2.5 chemical composition and effectively implement refined air pollution protection and control strategies. In this study, we developed an optimal deep learning model to acquire hourly mass concentrations of key PM2.5 chemical components without complex chemical analysis. The model was trained using a randomly partitioned multivariate dataset arranged in chronological order, including atmospheric state indicators, which previous studies did not consider. Our results showed that the correlation coefficients of key chemical components were no less than 0.96, and the root mean square errors ranged from 0.20 to 2.11 µg/m3 for the entire process (training and testing combined). The model accurately captured the temporal characteristics of key chemical components, outperforming typical machine-learning models, previous studies, and global reanalysis datasets (such as Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) and Copernicus Atmosphere Monitoring Service ReAnalysis (CAMSRA)). We also quantified the feature importance using the random forest model, which showed that PM2.5, PM1, visibility, and temperature were the most influential variables for key chemical components. In conclusion, this study presents a practical approach to accurately obtain chemical composition information that can contribute to filling missing data, improved air pollution monitoring and source identification. This approach has the potential to enhance air pollution control strategies and promote public health and environmental sustainability.

Forecasting O3 and NO2 concentrations with spatiotemporally continuous coverage in southeastern China using a Machine learning approach

Ozone (O3) is a significant contributor to air pollution and the main constituent ofphotochemical smog that plagues China. Nitrogen dioxide (NO2) is a significant air pollutant and a critical trace gas in the Earth's atmosphere. The presence of O3 and NO2 has detrimental effects on human health, the ecosystem, and agricultural production. Forecasting accurate ambient O3 and NO2 concentrations with full spatiotemporal coverage is pivotal for decision-makers to develop effective mitigation strategies and prevent harmful public exposure. Existing methods, including chemical transport models (CTMs) and time series at air monitoring sites, forecast O3 and NO2 concentrations either with nontrivial uncertainty or without spatiotemporally continuous coverage. In this research, we adopted a forecasting model that integrates the random forest algorithm with NASA's Goddard Earth Observing System "Composing Forecasting" (GEOS-CF) product. This approach offers spatiotemporally continuous forecasts of O3 and NO2 concentrations across southeastern China for up to five days in advance. Both overall validation and spatial cross-validation revealed that our forecast framework significantly surpassed the initial GEOS-CF model for all validation metrics, substantially reducing the errors in the GEOS-CF forecast data. Our model could provide accurate near-real-time O3 and NO2 forecasts with continuous spatiotemporal coverage.

Spatio-temporal analysis of bicyclists' PM2.5 exposure levels in a medium sized urban agglomeration

Many cities have promoted decarbonized transportation modes to mitigate climate change, reduce air pollution and promote healthy behaviors. However, cyclists may be particularly exposed to higher concentrations of on-road air pollutants than other commuters due to their proximity to traffic, higher breathing rates, and prolonged commutes. In addition, there is scarce information analyzing the geographic exposure levels of cyclists in medium-sized urban agglomerations of Latin American cities. We aimed to assess cyclists' exposure to PM2.5 at the spatio-temporal level using low-cost sensors in the Mendoza Metropolitan Area, Argentina. We calculated PM2.5 inhalation doses (IDs) for different routes with distinctive characteristics, considering different age ranges and gender of cyclists. The dose was represented as temporally and spatially disaggregated exposure maps, one of the first to represent it in Latin America using this method. All link types analyzed exhibited significant differences in PM2.5 concentrations, although the most frequent concentrations were less than 5 µg m-3, with secondary peaks of 6.5 and 9 µg m-3. As expected, ID increases with age and is greater in males than in females. Our findings further reinforce the fact that route choice (busy vs. quiet) and time of day (peak vs. off-peak) appreciably affect the pollutant exposure of cyclists. All these results could be helpful in the selection of alternative cycling routes with lower PM2.5 at different hours of day but also enable us to investigate further implications of exposure to PM2.5 for the health of urban bicycle commuters.

Unmasking the sky: high-resolution PM2.5 prediction in Texas using machine learning techniques

BACKGROUND

Although PM2.5 (fine particulate matter with an aerodynamic diameter less than 2.5 µm) is an air pollutant of great concern in Texas, limited regulatory monitors pose a significant challenge for decision-making and environmental studies.

OBJECTIVE

This study aimed to predict PM2.5 concentrations at a fine spatial scale on a daily basis by using novel machine learning approaches and incorporating satellite-derived Aerosol Optical Depth (AOD) and a variety of weather and land use variables.

METHODS

We compiled a comprehensive dataset in Texas from 2013 to 2017, including ground-level PM2.5 concentrations from regulatory monitors; AOD values at 1-km resolution based on images retrieved from the MODIS satellite; and weather, land-use, population density, among others. We built predictive models for each year separately to estimate PM2.5 concentrations using two machine learning approaches called gradient boosted trees and random forest. We evaluated the model prediction performance using in-sample and out-of-sample validations.

RESULTS

Our predictive models demonstrate excellent in-sample model performance, as indicated by high R2 values generated from the gradient boosting models (0.94-0.97) and random forest models (0.81-0.90). However, the out-of-sample R2 values fall within a range of 0.52-0.75 for gradient boosting models and 0.44-0.69 for random forest models. Model performance varies slightly across years. A generally decreasing trend in predicted PM2.5 concentrations over time is observed in Eastern Texas.

IMPACT STATEMENT

We utilized machine learning approaches to predict PM2.5 levels in Texas. Both gradient boosting and random forest models perform well. Gradient boosting models perform slightly better than random forest models. Our models showed excellent in-sample prediction performance (R2 > 0.9).

Meteorological data source comparison-a case study in geospatial modeling of potential environmental exposure to abandoned uranium mine sites in the Navajo Nation

Meteorological (MET) data is a crucial input for environmental exposure models. While modeling exposure potential using geospatial technology is a common practice, existing studies infrequently evaluate the impact of input MET data on the level of uncertainty on output results. The objective of this study is to determine the effect of various MET data sources on the potential exposure susceptibility predictions. Three sources of wind data are compared: The North American Regional Reanalysis (NARR) database, meteorological aerodrome reports (METARs) from regional airports, and data from local MET weather stations. These data sources are used as inputs into a machine learning (ML) driven GIS Multi-Criteria Decision Analysis (GIS-MCDA) geospatial model to predict potential exposure to abandoned uranium mine sites in the Navajo Nation. Results indicate significant variations in results derived from different wind data sources. After validating the results from each source using the National Uranium Resource Evaluation (NURE) database in a geographically weighted regression (GWR), METARs data combined with the local MET weather station data showed the highest accuracy, with an average R2 of 0.74. We conclude that local direct measurement-based data (METARs and MET data) produce a more accurate prediction than the other sources evaluated in the study. This study has the potential to inform future data collection methods, leading to more accurate predictions and better-informed policy decisions surrounding environmental exposure susceptibility and risk assessment.

Influence of land-sea breeze on PM[Formula: see text] prediction in central and southern Taiwan using composite neural network

PM[Formula: see text] prediction plays an important role for governments in establishing policies to control the emission of excessive atmospheric pollutants to protect the health of citizens. However, traditional machine learning methods that use data collected from ground-level monitoring stations have reached their limit with poor model generalization and insufficient data. We propose a composite neural network trained with aerosol optical depth (AOD) and weather data collected from satellites, as well as interpolated ocean wind features. We investigate the model outputs of different components of the composite neural network, concluding that the proposed composite neural network architecture yields significant improvements in overall performance compared to each component and the ensemble model benchmarks. The monthly analysis also demonstrates the superiority of the proposed architecture for stations where land-sea breezes frequently occur in the southern and central Taiwan in the months when land-sea breeze dominates the accumulation of PM[Formula: see text].

Current topics and challenges in geoAI

Taken literally, geoAI is the use of Artificial Intelligence methods and techniques in solving geo-spatial problems. Similar to AI more generally, geoAI has seen an influx of new (big) data sources and advanced machine learning techniques, but also a shift in the kind of problems under investigation. In this article, we highlight some of these changes and identify current topics and challenges in geoAI.

Exploring the detailed spatiotemporal characteristics of PM2.5: Generating a full-coverage and hourly PM2.5 dataset in the Sichuan Basin, China

Fine particulate matter (PM_2.5) has received worldwide attention due to its threat to public health. In the Sichuan Basin (SCB), PM_2.5 is causing heavy health burdens due to its high concentrations and population density. Compared with other heavily polluted areas, less effort has been made to generate a full-coverage PM_2.5 dataset of the SCB, in which the detailed PM_2.5 spatiotemporal characteristics remain unclear. Considering commonly existing spatiotemporal autocorrelations, the top-of-atmosphere reflectance (TOAR) with a high coverage rate and other auxiliary data were employed to build commonly used random forest (RF) models to generate accurate hourly PM_2.5 concentration predictions with a 0.05° × 0.05° spatial resolution in the SCB in 2016. Specifically, with historical concentrations predicted from a spatial RF (S-RF) and observed at stations, an alternative spatiotemporal RF (AST-RF) and spatiotemporal RF (ST-RF) were built in grids with stations (type 1). The predictions from the AST-RF in grids without stations (type 2) and observations in type 1 formed the PM_2.5 dataset. The LOOCV R², RMSE and MAE were 0.94/0.94, 8.71/8.62 μg∕m³ and 5.58/5.57 μg∕m³ in the AST-RF/ST-RF, respectively. Using the produced dataset, spatiotemporal analysis was conducted for a detailed understanding of the spatiotemporal characteristics of PM_2.5 in the SCB. The PM_2.5 concentrations gradually increased from the edge to the center of the SCB in spatial distribution. Two high-concentration areas centered on Chengdu and Zigong were observed throughout the year, while another high-concentration area centered on Dazhou was only observed in winter. The diurnal variation had double peaks and double valleys in the SCB. The concentrations were high at night and low in daytime, which suggests that characterizing the relationship between PM_2.5 and adverse health outcomes by daily means might be inaccurate with most human activities conducted in daytime.

Nowcasting Applications of Geostationary Satellite Hourly Surface PM2.5 Data

The mass concentration of fine particulate matter (PM2.5; diameters less than 2.5 μm) estimated from geostationary satellite aerosol optical depth (AOD) data can supplement the network of ground monitors with high temporal (hourly) resolution. Estimates of PM2.5 over the United States (US) were derived from NOAA's operational geostationary satellites Advanced Baseline Imager (ABI) AOD data using a geographically weighted regression with hourly and daily temporal resolution. Validation versus ground observations shows a mean bias of -21.4% and -15.3% for hourly and daily PM2.5 estimates, respectively, for concentrations ranging from 0 to 1000 μg/m3. Because satellites only observe AOD in the daytime, the relation between observed daytime PM2.5 and daily mean PM2.5 was evaluated using ground measurements; PM2.5 estimated from ABI AODs were also examined to study this relationship. The ground measurements show that daytime mean PM2.5 has good correlation (r > 0.8) with daily mean PM2.5 in most areas of the US, but with pronounced differences in the western US due to temporal variations caused by wildfire smoke; the relation between the daytime and daily PM2.5 estimated from the ABI AODs has a similar pattern. While daily or daytime estimated PM2.5 provides exposure information in the context of the PM2.5 standard (> 35 μg/m3), the hourly estimates of PM2.5 used in Nowcasting show promise for alerts and warnings of harmful air quality. The geostationary satellite based PM2.5 estimates inform the public of harmful air quality ten times more than standard ground observations (1.8 vs. 0.17 million people per hour).

Prediction of daily mean and one-hour maximum PM2.5 concentrations and applications in Central Mexico using satellite-based machine-learning models

BACKGROUND

Machine-learning algorithms are becoming popular techniques to predict ambient air PM_2.5 concentrations at high spatial resolutions (1 × 1 km) using satellite-based aerosol optical depth (AOD). Most machine-learning models have aimed to predict 24 h-averaged PM_2.5 concentrations (mean PM_2.5) in high-income regions. Over Mexico, none have been developed to predict subdaily peak levels, such as the maximum daily 1-h concentration (max PM_2.5).

OBJECTIVE

Our goal was to develop a machine-learning model to predict mean PM_2.5 and max PM_2.5 concentrations in the Mexico City Metropolitan Area from 2004 through 2019.

METHODS

We present a new modeling approach based on extreme gradient boosting (XGBoost) and inverse-distance weighting that uses AOD, meteorology, and land-use variables. We also investigated applications of our mean PM_2.5 predictions that can aid local authorities in air-quality management and public-health surveillance, such as the co-occurrence of high PM_2.5 and heat, compliance with local air-quality standards, and the relationship of PM_2.5 exposure with social marginalization.

RESULTS

Our models for mean and max PM_2.5 exhibited good performance, with overall cross-validated mean absolute errors (MAE) of 3.68 and 9.20 μg/m³, respectively, compared to mean absolute deviations from the median (MAD) of 8.55 and 15.64 μg/m³. In 2010, everybody in the study region was exposed to unhealthy levels of PM_2.5. Hotter days had greater PM_2.5 concentrations. Finally, we found similar exposure to PM_2.5 across levels of social marginalization.

SIGNIFICANCE

Machine learning algorithms can be used to predict highly spatiotemporally resolved PM_2.5 concentrations even in regions with sparse monitoring.

IMPACT

Our PM_2.5 predictions can aid local authorities in air-quality management and public-health surveillance, and they can advance epidemiological research in Central Mexico with state-of-the-art exposure assessment methods.

Urban Particulate Matter Hazard Mapping and Monitoring Site Selection in Nablus, Palestine

Few air pollution studies have been applied in the State of Palestine and all showed an increase in particulate matter concentrations above WHO guidelines. However, there is no clear methodology for selecting monitoring locations. In this study, a methodology based on GIS and locally calibrated low-cost sensors was tested. A GIS-based weighted overlay summation process for the potential sources of air pollution (factories, quarries, and traffic), taking into account the influence of altitude and climate, was used to obtain an air pollution hazard map for Nablus, Palestine. To test the methodology, eight locally calibrated PM sensors (AirUs) were deployed to measure PM2.5 concentrations for 55 days from 7 January to 2 March 2022. The results of the hazard map showed that 82% of Nablus is exposed to a high and medium risk of PM pollution. Sensors’ readings showed a good match between the hazard intensity and PM concentrations. It also shows an elevated PM2.5 concentrations above WHO guidelines in all areas. In summary, the overall average for PM2.5 in the Nablus was 48 µg/m3. This may indicate the effectiveness of mapping methodology and the use of low-cost, locally calibrated sensors in characterizing air quality status to identify the potential remediation options.

A Suitable Model for Spatiotemporal Particulate Matter Concentration Prediction in Rural and Urban Landscapes, Thailand

Spatiotemporal particulate matter (PM) concentration prediction using MODIS AOD with significant PM factors in rural and urban landscapes in Thailand is necessary for public health and has been complicated by the limitations of PM monitoring stations. The research objectives were (1) to identify significant factors affecting PM10 concentrations in rural landscapes and PM2.5 in urban landscapes; (2) to predict spatiotemporal PM10 and PM2.5 concentrations using geographically weighted regression (GWR) and mixed-effect model (MEM), and (3) to evaluate a suitable spatiotemporal model for PM10 and PM2.5 concentration prediction and validation. The research methodology consisted of four stages: data collection and preparation, the identification of significant spatiotemporal factors affecting PM concentrations, the prediction of spatiotemporal PM concentrations, and a suitable spatiotemporal model for PM concentration prediction and validation. As a result, the predicted PM10 concentrations using the GWR model varied from 50.53 to 85.79 µg/m3 and from 36.92 to 51.32 µg/m3 in winter and summer, while the predicted PM10 concentrations using the MEM model varied from 50.68 to 84.59 µg/m3 and from 37.08 to 50.81 µg/m3 in both seasons. Likewise, the PM2.5 concentration prediction using the GWR model varied from 25.33 to 44.37 µg/m3 and from 16.69 to 24.04 µg/m3 in winter and summer, and the PM2.5 concentration prediction using the MEM model varied from 25.45 to 44.36 µg/m3 and from 16.68 and 23.75 µg/m3 during the two seasons. Meanwhile, according to Thailand and U.S. EPA standards, the monthly air quality index (AQI) classifications of the GWR and MEM were similar. Nevertheless, the derived average corrected Akaike Information Criterion (AICc) values of the GWR model for PM10 and PM2.5 predictions during both seasons were lower than that of the MEM model. Therefore, the GWR model was chosen as a suitable model for spatiotemporal PM10 and PM2.5 concentration predictions. Furthermore, the result of spatial correlation analysis for GWR model validation based on a new dataset provided average correlation coefficient values for PM10 and PM2.5 concentration predictions with a higher than the expected value of 0.5. Subsequently, the GWR model with significant monthly and seasonal factors could predict spatiotemporal PM 10 and PM2.5 concentrations in rural and urban landscapes in Thailand.

A New Coupling Method for PM2.5 Concentration Estimation by the Satellite-Based Semiempirical Model and Numerical Model

Aerosol optical and chemical properties play a major role in the retrieval of PM2.5 concentrations based on aerosol optical depth (AOD) data from satellites in the conventional semiempirical model (SEM). However, limited observation information hinders the high-resolution estimation of PM2.5. Therefore, a new method for evaluating near-surface PM2.5 at high spatial resolution is developed by coupling the SEM and the chemical transport model (CTM)-based numerical (CSEN) model. The numerical model can provide large-scale information for aerosol properties with high spatial resolution at a large scale based on emissions and meteorology, though it can still be biased in simulating absolute PM2.5 concentrations. Therefore, the two crucial aerosol characteristic parameters, including the coefficient integrated humidity effect (γ′) and the comprehensive reference value of aerosol properties (K) in SEM, have been redefined using the WRF-Chem numerical model. Improved model performance was observed for these results compared with the original SEM results. The monthly averaged correlation coefficients (R) by CSEN were 0.92, 0.82, 0.84, and 0.83 in January, April, July, and October, respectively, whereas those of the SEM were 0.80, 0.77, 0.72, and 0.72, respectively. All the statistical metrics of the model validation showed significant improvements in all seasons. The reduced biases of estimated PM2.5 by CSEN indicated the effect of hygroscopic growth and aerosol properties affected by the meteorology on the relationship between AOD and estimated PM2.5 concentrations, especially in winter and summer. The better performance of the CSEN model provides insight for air quality monitoring at different scales, which supplies important information for air pollution control policies and health impact analysis.

Spring 2020 Atmospheric Aerosol Contamination over Kyiv City

Extraordinarily high aerosol contamination was observed in the atmosphere over the city of Kyiv, Ukraine, during the March–April 2020 period. The source of contamination was the large grass and forest fires in the northern part of Ukraine and the Kyiv region. The level of PM2.5 load was investigated using newly established AirVisual sensor mini-networks in five areas of the city. The aerosol data from the Kyiv AERONET sun-photometer site were analyzed for that period. Aerosol optical depth, Ångström exponent, and the aerosol particles properties (particle size distribution, single-scattering albedo, and complex refractive index) were analyzed using AERONET sun-photometer observations. The smoke particles observed at Kyiv site during the fires in general correspond to aerosol with optical properties of biomass burning aerosol. The variability of the optical properties and chemical composition indicates that the aerosol particles in the smoke plumes over Kyiv city were produced by different burning materials and phases of vegetation fires at different times. The case of enormous PM2.5 aerosol contamination in the Kyiv city reveals the need to implement strong measures for forest fire control and prevention in the Kyiv region, especially in its northwest part, where radioactive contamination from the Chernobyl disaster is still significant.

Mutual Information Variational Autoencoders and Its Application to Feature Extraction of Multivariate Time Series

The application of deep learning in time-series prediction has developed gradually. In this paper, we propose a deep generative network model for feature extraction of multivariate time series, namely, mutual information variational autoencoders (MI-VAE). In the architecture of the proposed model, we use the latent space of VAE for feature learning, which can extract the essential features of multivariate time-series data effectively. The latent space employed directly as a feature extractor can avoid poor interpretability of model. In addition, we introduce a mutual information term into the loss function, which improves the expression capability and accuracy of model. The proposed model, combining the merits of VAE and mutual information, extracts features for multivariate time-series data from a new perspective. The Lorenz system and Beijing air quality time series are used to test performance of the proposed model and comparative models. Results show that the proposed model is superior to other similar models in terms of accuracy and expression capability of latent space.

A novel multi-pollutant space-time learning network for air pollution inference

Detailed information about air pollution in space and time is essential to manage risks to public health. In this paper we propose a multi-pollutant space-time learning network (Multi-AP learning network), which estimates pixel-wise (grid-level) concentrations of multiple air pollutant species based on fixed-station measurements and multi-source urban features, including land use information, traffic data, and meteorological conditions. We infer concentrations of multiple pollutants within one integrated learning network, which is applied to and evaluated on a case study in Chengdu (4900 km², 26 April - 12 June 2019), where air pollutant (PM_2.5, PM₁₀ and O₃) measurements from 40 monitoring sites are used to train the network to estimate pollutant concentrations in 4900 grid-cells (1 km²). The Multi-AP learning network allows us to estimate highly-resolved (1 km × 1 km, hourly) air pollution maps based on pollutant measurements which cover less than 1% of the grid-cells with better accuracy compared to other approaches, and with significant computational efficiency improvements. The time-cost is 1/3 of the time-cost of modelling each pollutant individually. Furthermore, we evaluate the relative importance of features and find that the meteorological feature set is the most important followed the land use features. The proposed Multi-AP method could be used to estimate air pollution exposure across a city using a limited set of air pollution monitoring sites.

Combining Machine Learning and Numerical Simulation for High-Resolution PM2.5 Concentration Forecast

Forecasting ambient PM_2.5 concentrations with spatiotemporal coverage is key to alerting decision makers of pollution episodes and preventing detrimental public exposure, especially in regions with limited ground air monitoring stations. The existing methods rely on either chemical transport models (CTMs) to forecast spatial distribution of PM_2.5 with nontrivial uncertainty or statistical algorithms to forecast PM_2.5 concentration time series at air monitoring locations without continuous spatial coverage. In this study, we developed a PM_2.5 forecast framework by combining the robust Random Forest algorithm with a publicly accessible global CTM forecast product, NASA's Goddard Earth Observing System "Composition Forecasting" (GEOS-CF), providing spatiotemporally continuous PM_2.5 concentration forecasts for the next 5 days at a 1 km spatial resolution. Our forecast experiment was conducted for a region in Central China including the populous and polluted Fenwei Plain. The forecast for the next 2 days had an overall validation R² of 0.76 and 0.64, respectively; the R² was around 0.5 for the following 3 forecast days. Spatial cross-validation showed similar validation metrics. Our forecast model, with a validation normalized mean bias close to 0, substantially reduced the large biases in GEOS-CF. The proposed framework requires minimal computational resources compared to running CTMs at urban scales, enabling near-real-time PM_2.5 forecast in resource-restricted environments.

Evaluating the Utility of High-Resolution Spatiotemporal Air Pollution Data in Estimating Local PM2.5 Exposures in California from 2015–2018

Air quality management is increasingly focused not only on across-the-board reductions in ambient pollution concentrations but also on identifying and remediating elevated exposures that often occur in traditionally disadvantaged communities. Remote sensing of ambient air pollution using data derived from satellites has the potential to better inform management decisions that address environmental disparities by providing increased spatial coverage, at high-spatial resolutions, compared to air pollution exposure estimates based on ground-based monitors alone. Daily PM2.5 estimates for 2015–2018 were estimated at a 1 km2 resolution, derived from NASA’s Moderate Resolution Imaging Spectroradiometer (MODIS) satellite instrument and the Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm in order to assess the utility of highly refined spatiotemporal air pollution data in 92 California cities and in the 13 communities included in the California Community Air Protection Program. The identification of pollution hot-spots within a city is typically not possible relying solely on the regulatory monitoring networks; however, day-to-day temporal variability was shown to be generally well represented by nearby ground-based monitoring data even in communities with strong spatial gradients in pollutant concentrations. An assessment of within-ZIP Code variability in pollution estimates indicates that high-resolution pollution estimates (i.e., 1 km2) are not always needed to identify spatial differences in exposure but become increasingly important for larger geographic areas (approximately 50 km2). Taken together, these findings can help inform strategies for use of remote sensing data for air quality management including the screening of locations with air pollution exposures that are not well represented by existing ground-based air pollution monitors.

Comparison of PM2.5 in Seoul, Korea Estimated from the Various Ground-Based and Satellite AOD

Based on multiple linear regression (MLR) models, we estimated the PM2.5 at Seoul using a number of aerosol optical depth (AOD) values obtained from ground-based and satellite remote sensing observations. To construct the MLR model, we consider various parameters related to the ambient meteorology and air quality. In general, all AOD values resulted in the high quality of PM2.5 estimation through the MLR method: mostly correlation coefficients >~0.8. Among various polar-orbit satellite AODs, AOD values from the MODIS measurement contribute to better PM2.5 estimation. We also found that the quality of estimated PM2.5 shows some seasonal variation; the estimated PM2.5 values consistently have the highest correlation with in situ PM2.5 in autumn, but are not well established in winter, probably due to the difficulty of AOD retrieval in the winter condition. MLR modeling using spectral AOD values from the ground-based measurements revealed that the accuracy of PM2.5 estimation does not depend on the selected wavelength. Although all AOD values used in this study resulted in a reasonable accuracy range of PM2.5 estimation, our analyses of the difference in estimated PM2.5 reveal the importance of utilizing the proper AOD for the best quality of PM2.5 estimation.

Full-coverage spatiotemporal mapping of ambient PM2.5 and PM10 over China from Sentinel-5P and assimilated datasets: Considering the precursors and chemical compositions

Ambient concentrations of particulate matters (PM_2.5 and PM₁₀) are significant indicators for monitoring the air quality relevant to living conditions. At present, most remote sensing based approaches for the estimation of PM_2.5 and PM₁₀ employed Aerosol Optical Depth (AOD) products as the main variate. Nevertheless, the coverage of missing data is generally large in AOD products, which can cause deviations in practical applications of estimated PM_2.5 and PM₁₀ (e.g., air quality monitoring and exposure evaluation). To efficiently address this issue, our study explores a novel approach using the datasets of the precursors & chemical compositions for PM_2.5 and PM₁₀ instead of AOD products. Specifically, the daily full-coverage ambient concentrations of PM_2.5 and PM₁₀ are estimated at 5-km (0.05°) spatial girds across China based on Sentinel-5P and assimilated datasets (GEOS-FP). The estimation models are acquired via an advanced ensemble learning method named Light Gradient Boosting Machine in this paper. For comparison, the Deep Blue AOD product from VIIRS is adopted in a similar framework as a baseline (AOD-based). Validation results show that the ambient concentrations are well estimated through the proposed approach, with the space-based Cross-Validation R²s and RMSEs of 0.88 (0.83) and 11.549 (22.9) μg/m³ for PM_2.5 (PM₁₀), respectively. Meanwhile, the proposed approach achieves better performance than the AOD-based in different cases (e.g., overall and seasonal). Compared to the related previous works over China, the estimation accuracy of our method is also satisfactory. Regarding the mapping, the estimated results through the proposed approach display consecutive spatial distribution and can exactly express the seasonal variations of PM_2.5 and PM₁₀. The proposed approach could efficiently present daily full-coverage results at 5-km spatial grids. It has a large potential to be extended for providing global accurate ambient concentrations of PM_2.5 and PM₁₀ at multiple temporal scales (e.g., daily and annual).

Multivariate Spatial Prediction of Air Pollutant Concentrations with INLA

Estimates of daily air pollution concentrations with complete spatial and temporal coverage are important for supporting epidemiologic studies and health impact assessments. While numerous approaches have been developed for modeling air pollution, they typically only consider each pollutant separately. We describe a spatial multipollutant data fusion model that combines monitoring measurements and chemical transport model simulations that leverages dependence between pollutants to improve spatial prediction. For the contiguous United States, we created a data product of daily concentration for 12 pollutants (CO, NOx, NO₂, SO₂, O₃, PM₁₀, and PM_2.5 species EC, OC, NO₃, NH₄, SO₄) during the period 2005 to 2014. Out-of-sample prediction showed good performance, particularly for daily PM_2.5 species EC (R² = 0.64), OC (R² = 0.75), NH₄ (R² = 0.84), NO₃ (R2 = 0.73), and SO₄ (R² = 0.80). By employing the integrated nested Laplace approximation (INLA) for Bayesian inference, our approach also provides model-based prediction error estimates. The daily data product at 12km spatial resolution will be publicly available immediately upon publication. To our knowledge this is the first publicly available data product for major PM_2.5 species and several gases at this spatial and temporal resolution.

A National-Scale 1-km Resolution PM2.5 Estimation Model over Japan Using MAIAC AOD and a Two-Stage Random Forest Model

Satellite-based models for estimating concentrations of particulate matter with an aerodynamic diameter less than 2.5 μm (PM2.5) have seldom been developed in islands with complex topography over the monsoon area, where the transport of PM2.5 is influenced by both the synoptic-scale winds and local-scale circulations compared with the continental regions. We validated Multi-Angle Implementation of Atmospheric Correction (MAIAC) aerosol optical depth (AOD) with ground observations in Japan and developed a 1-km-resolution national-scale model between 2011 and 2016 to estimate daily PM2.5 concentrations. A two-stage random forest model integrating MAIAC AOD with meteorological variables and land use data was applied to develop the model. The first-stage random forest model was used to impute the missing AOD values. The second-stage random forest model was then utilised to estimate ground PM2.5 concentrations. Ten-fold cross-validation was performed to evaluate the model performance. There was good consistency between MAIAC AOD and ground truth in Japan (correlation coefficient = 0.82 and 74.62% of data falling within the expected error). For model training, the model showed a training coefficient of determination (R2) of 0.98 and a root mean square error (RMSE) of 1.22 μg/m3. For the 10-fold cross-validation, the cross-validation R2 and RMSE of the model were 0.86 and 3.02 μg/m3, respectively. A subsite validation was used to validate the model at the grids overlapping with the AERONET sites, and the model performance was excellent at these sites with a validation R2 (RMSE) of 0.94 (1.78 μg/m3). Additionally, the model performance increased as increased AOD coverage. The top-ten important predictors for estimating ground PM2.5 concentrations were day of the year, temperature, AOD, relative humidity, 10-m-height zonal wind, 10-m-height meridional wind, boundary layer height, precipitation, surface pressure, and population density. MAIAC AOD showed high retrieval accuracy in Japan. The performance of the satellite-based model was excellent, which showed that PM2.5 estimates derived from the model were reliable and accurate. These estimates can be used to assess both the short-term and long-term effects of PM2.5 on health outcomes in epidemiological studies.

Evaluation of MODIS Aerosol Optical Depth and Surface Data Using an Ensemble Modeling Approach to Assess PM2.5 Temporal and Spatial Distributions

The use of statistical models and machine-learning techniques along satellite-derived aerosol optical depth (AOD) is a promising method to estimate ground-level particulate matter with an aerodynamic diameter of ≤2.5 μm (PM2.5), mainly in urban areas with low air quality monitor density. Nevertheless, the relationship between AOD and ground-level PM2.5 varies spatiotemporally and differences related to spatial domains, temporal schemes, and seasonal variations must be assessed. Here, an ensemble multiple linear regression (EMLR) model and an ensemble neural network (ENN) model were developed to estimate PM2.5 levels in the Monterrey Metropolitan Area (MMA), the second largest urban center in Mexico. Four AOD-SDSs (Scientific Datasets) from MODIS Collection 6 were tested using three spatial domains and two temporal schemes. The best model performance was obtained using AOD at 0.55 µm from MODIS-Aqua at a spatial resolution of 3 km, along meteorological parameters and daily scheme. EMLR yielded a correlation coefficient (R) of ~0.57 and a root mean square error (RMSE) of ~7.00 μg m−3. ENN performed better than EMLR, with an R of ~0.78 and RMSE of ~5.43 μg m−3. Satellite-derived AOD in combination with meteorology data allowed for the estimation of PM2.5 distributions in an urban area with low air quality monitor density.

Assessment and valuation of health impacts of fine particulate matter during COVID-19 lockdown: a comprehensive study of tropical and sub tropical countries

A novel coronavirus disease (COVID-19) continues to challenge the whole world. The disease has claimed many fatalities as it has transcended from one country to another since it was first discovered in China in late 2019. To prevent further morbidity and mortality associated with COVID-19, most of the countries initiated a countrywide lockdown. While physical distancing and lockdowns helped in curbing the spread of this novel coronavirus, it led to massive economic losses for the nations. Positive impacts have been observed due to lockdown in terms of improved air quality of the nations. In the current research, ten tropical and subtropical countries have been analysed from multiple angles, including air pollution, assessment and valuation of health impacts and economic loss of countries during COVID-19 lockdown. Countries include Brazil, India, Iran, Kenya, Malaysia, Mexico, Pakistan, Peru, Sri Lanka, and Thailand. Validated Simplified Aerosol Retrieval Algorithm (SARA) binning model is used on data collated from moderate resolution imaging spectroradiometer (MODIS) for particulate matters with a diameter of less than 2.5 μm (PM2.5) for all the countries for the month of January to May 2019 and 2020. The concentration results of PM2.5 show that air pollution has drastically reduced in 2020 post lockdown for all countries. The highest average concentration obtained by converting aerosol optical depth (AOD) for 2020 is observed for Thailand as 121.9 μg/m3 and the lowest for Mexico as 36.27 μg/m3. As air pollution is found to decrease in the April and May months of 2020 for nearly all countries, they are compared with respective previous year values for the same duration to calculate the reduced health burden due to lockdown. The present study estimates that cumulative about 100.9 Billion US$ are saved due to reduced air pollution externalities, which are about 25% of the cumulative economic loss of 435.9 Billion US$.

A multi-analysis approach for estimating regional health impacts from the 2017 Northern California wildfires

Smoke impacts from large wildfires are mounting, and the projection is for more such events in the future as the one experienced October 2017 in Northern California, and subsequently in 2018 and 2020. Further, the evidence is growing about the health impacts from these events which are also difficult to simulate. Therefore, we simulated air quality conditions using a suite of remotely-sensed data, surface observational data, chemical transport modeling with WRF-CMAQ, one data fusion, and three machine learning methods to arrive at datasets useful to air quality and health impact analyses. To demonstrate these analyses, we estimated the health impacts from smoke impacts during wildfires in October 8-20, 2017, in Northern California, when over 7 million people were exposed to Unhealthy to Very Unhealthy air quality conditions. We investigated using the 5-min available GOES-16 fire detection data to simulate timing of fire activity to allocate emissions hourly for the WRF-CMAQ system. Interestingly, this approach did not necessarily improve overall results, however it was key to simulating the initial 12-hr explosive fire activity and smoke impacts. To improve these results, we applied one data fusion and three machine learning algorithms. We also had a unique opportunity to evaluate results with temporary monitors deployed specifically for wildfires, and performance was markedly different. For example, at the permanent monitoring locations, the WRF-CMAQ simulations had a Pearson correlation of 0.65, and the data fusion approach improved this (Pearson correlation = 0.95), while at the temporary monitor locations across all cases, the best Pearson correlation was 0.5. Overall, WRF-CMAQ simulations were biased high and the geostatistical methods were biased low. Finally, we applied the optimized PM2.5 exposure estimate in an exposure-response function. Estimated mortality attributable to PM2.5 exposure during the smoke episode was 83 (95% CI: 0, 196) with 47% attributable to wildland fire smoke.Implications: Large wildfires in the United States and in particular California are becoming increasingly common. Associated with these large wildfires are air quality and health impact to millions of people from the smoke. We simulated air quality conditions using a suite of remotely-sensed data, surface observational data, chemical transport modeling, one data fusion, and three machine learning methods to arrive at datasets useful to air quality and health impact analyses from the October 2017 Northern California wildfires. Temporary monitors deployed for the wildfires provided an important model evaluation dataset. Total estimated regional mortality attributable to PM2.5 exposure during the smoke episode was 83 (95% confidence interval: 0, 196) with 47% of these deaths attributable to the wildland fire smoke. This illustrates the profound effect that even a 12-day exposure to wildland fire smoke can have on human health.

Environmental impact of COVID-19 led lockdown: A satellite data-based assessment of air quality in Indian megacities

The strategies to contain the spread of COVID-19 pandemic, including restricted human movement and economic activities, have shown positive impacts on the environment. Present research analysed the effects of COVID-19 led lockdown on air quality with special reference to major pollutants, namely nitrogen dioxide (NO₂), carbon monoxide (CO), sulphur dioxide (SO₂) and aerosol optical depth (AOD). The assessment has been conducted for megacities of India (Delhi, Mumbai, Bengaluru, Chennai and Kolkata) for four months, that is, March and April in 2019 and 2020 using Sentinel 5P and MCD19A2 data. A decrease in concentrations of air pollutants, specifically NO₂ and SO_2, has been observed during the lockdown period in all the cities; whereas CO and AOD have exhibited discrete pattern of spatio-temporal variation. Four megacities except Kolkata have revealed a positive correlation between NO₂ concentration and population density. The results conclude overall improvement in air quality during COVID-19 led lockdown. The current situation provides a unique opportunity to implement a structural economic change that could help us move towards a city with low emission economy. Realizing the achievable improvement of air quality, the study suggests further in-depth research on source attribution of individual pollutants to assess the prospect of emission reduction actions.

Hourly Ground-Level PM2.5 Estimation Using Geostationary Satellite and Reanalysis Data via Deep Learning

This study proposes an improved approach for monitoring the spatial concentrations of hourly particulate matter less than 2.5 μm in diameter (PM2.5) via a deep neural network (DNN) using geostationary ocean color imager (GOCI) images and unified model (UM) reanalysis data over the Korean Peninsula. The DNN performance was optimized to determine the appropriate training model structures, incorporating hyperparameter tuning, regularization, early stopping, and input and output variable normalization to prevent training dataset overfitting. Near-surface atmospheric information from the UM was also used as an input variable to spatially generalize the DNN model. The retrieved PM2.5 from the DNN was compared with estimates from random forest, multiple linear regression, and the Community Multiscale Air Quality model. The DNN demonstrated the highest accuracy compared to that of the conventional methods for the hold-out validation (root mean square error (RMSE) = 7.042 μg/m3, mean bias error (MBE) = −0.340 μg/m3, and coefficient of determination (R2) = 0.698) and the cross-validation (RMSE = 9.166 μg/m3, MBE = 0.293 μg/m3, and R2 = 0.49). Although the R2 was low due to underestimated high PM2.5 concentration patterns, the RMSE and MBE demonstrated reliable accuracy values (<10 μg/m3 and 1 μg/m3, respectively) for the hold-out validation and cross-validation.

Imputing Satellite-Derived Aerosol Optical Depth Using a Multi-Resolution Spatial Model and Random Forest for PM2.5 Prediction

A task for environmental health research is to produce complete pollution exposure maps despite limited monitoring data. Satellite-derived aerosol optical depth (AOD) is frequently used as a predictor in various models to improve PM2.5 estimation, despite significant gaps in coverage. We analyze PM2.5 and AOD from July 2011 in the contiguous United States. We examine two methods to aid in gap-filling AOD: (1) lattice kriging, a spatial statistical method adapted to handle large amounts data, and (2) random forest, a tree-based machine learning method. First, we evaluate each model’s performance in the spatial prediction of AOD, and we additionally consider ensemble methods for combining the predictors. In order to accurately assess the predictive performance of these methods, we construct spatially clustered holdouts to mimic the observed patterns of missing data. Finally, we assess whether gap-filling AOD through one of the proposed ensemble methods can improve prediction of PM2.5 in a random forest model. Our results suggest that ensemble methods of combining lattice kriging and random forest can improve AOD gap-filling. Based on summary metrics of performance, PM2.5 predictions based on random forest models were largely similar regardless of the inclusion of gap-filled AOD, but there was some variability in daily model predictions.

Ensemble-based deep learning for estimating PM2.5 over California with multisource big data including wildfire smoke

INTRODUCTION

Estimating PM2.5 concentrations and their prediction uncertainties at a high spatiotemporal resolution is important for air pollution health effect studies. This is particularly challenging for California, which has high variability in natural (e.g, wildfires, dust) and anthropogenic emissions, meteorology, topography (e.g. desert surfaces, mountains, snow cover) and land use.

METHODS

Using ensemble-based deep learning with big data fused from multiple sources we developed a PM2.5 prediction model with uncertainty estimates at a high spatial (1 km × 1 km) and temporal (weekly) resolution for a 10-year time span (2008-2017). We leveraged autoencoder-based full residual deep networks to model complex nonlinear interrelationships among PM2.5 emission, transport and dispersion factors and other influential features. These included remote sensing data (MAIAC aerosol optical depth (AOD), normalized difference vegetation index, impervious surface), MERRA-2 GMI Replay Simulation (M2GMI) output, wildfire smoke plume dispersion, meteorology, land cover, traffic, elevation, and spatiotemporal trends (geo-coordinates, temporal basis functions, time index). As one of the primary predictors of interest with substantial missing data in California related to bright surfaces, cloud cover and other known interferences, missing MAIAC AOD observations were imputed and adjusted for relative humidity and vertical distribution. Wildfire smoke contribution to PM2.5 was also calculated through HYSPLIT dispersion modeling of smoke emissions derived from MODIS fire radiative power using the Fire Energetics and Emissions Research version 1.0 model.

RESULTS

Ensemble deep learning to predict PM2.5 achieved an overall mean training RMSE of 1.54 μg/m3 (R2: 0.94) and test RMSE of 2.29 μg/m3 (R2: 0.87). The top predictors included M2GMI carbon monoxide mixing ratio in the bottom layer, temporal basis functions, spatial location, air temperature, MAIAC AOD, and PM2.5 sea salt mass concentration. In an independent test using three long-term AQS sites and one short-term non-AQS site, our model achieved a high correlation (>0.8) and a low RMSE (<3 μg/m3). Statewide predictions indicated that our model can capture the spatial distribution and temporal peaks in wildfire-related PM2.5. The coefficient of variation indicated highest uncertainty over deciduous and mixed forests and open water land covers.

CONCLUSION

Our method can be generalized to other regions, including those having a mix of major urban areas, deserts, intensive smoke events, snow cover and complex terrains, where PM2.5 has previously been challenging to predict. Prediction uncertainty estimates can also inform further model development and measurement error evaluations in exposure and health studies.

Estimation of PM2.5 Concentrations in New York State: Understanding the Influence of Vertical Mixing on Surface PM2.5 Using Machine Learning

In New York State (NYS), episodic high fine particulate matter (PM2.5) concentrations associated with aerosols originated from the Midwest, Mid-Atlantic, and Pacific Northwest states have been reported. In this study, machine learning techniques, including multiple linear regression (MLR) and artificial neural network (ANN), were used to estimate surface PM2.5 mass concentrations at air quality monitoring sites in NYS during the summers of 2016–2019. Various predictors were considered, including meteorological, aerosol, and geographic predictors. Vertical predictors, designed as the indicators of vertical mixing and aloft aerosols, were also applied. Overall, the ANN models performed better than the MLR models, and the application of vertical predictors generally improved the accuracy of PM2.5 estimation of the ANN models. The leave-one-out cross-validation results showed significant cross-site variations and were able to present the different predictor-PM2.5 correlations at the sites with different PM2.5 characteristics. In addition, a joint analysis of regression coefficients from the MLR model and variable importance from the ANN model provided insights into the contributions of selected predictors to PM2.5 concentrations. The improvements in model performance due to aloft aerosols were relatively minor, probably due to the limited cases of aloft aerosols in current datasets.

Estimating Wildfire Smoke Concentrations during the October 2017 California Fires through BME Space/Time Data Fusion of Observed, Modeled, and Satellite-Derived PM2.5

Exposure to wildfire smoke causes adverse health outcomes, suggesting the importance of accurately estimating smoke concentrations. Geostatistical methods can combine observed, modeled, and satellite-derived concentrations to produce accurate estimates. Here, we estimate daily average ground-level PM_2.5 concentrations at a 1 km resolution during the October 2017 California wildfires, using the Constant Air Quality Model Performance (CAMP) and Bayesian Maximum Entropy (BME) methods to bias-correct and fuse three concentration datasets: permanent and temporary monitoring stations, a chemical transport model (CTM), and satellite-derived estimates. Four BME space/time kriging and data fusion methods were evaluated. All BME methods produce more accurate estimates than the standalone CTM and satellite products. Adding temporary station data increases the R² by 36%. The data fusion of observations with the CAMP-corrected CTM and satellite-derived concentrations provides the best estimate (R² = 0.713) in fire-impacted regions, emphasizing the importance of combining multiple datasets. We estimate that approximately 65,000 people were exposed to very unhealthy air (daily average PM_2.5 ≥ 150.5 μg/m³).

Spatial and Temporal Distribution of PM2.5 Pollution over Northeastern Mexico: Application of MERRA-2 Reanalysis Datasets

Aerosol and meteorological remote sensing data could be used to assess the distribution of urban and regional fine particulate matter (PM2.5), especially in locations where there are few or no ground-based observations, such as Latin America. The objective of this study is to evaluate the ability of Modern-Era Retrospective Analysis for Research and Application, version 2 (MERRA-2) aerosol components to represent PM2.5 ground concentrations and to develop and validate an ensemble neural network (ENN) model that uses MERRA-2 aerosol and meteorology products to estimate the monthly average of PM2.5 ground concentrations in the Monterrey Metropolitan Area (MMA), which is the main urban area in Northeastern Mexico (NEM). The project involves the application of the ENN model to a regional domain that includes not only the MMA but also other municipalities in NEM in the period from January 2010 to December 2014. Aerosol optical depth (AOD), temperature, relative humidity, dust PM2.5, sea salt PM2.5, black carbon (BC), organic carbon (OC), and sulfate (SO42−) reanalysis data were identified as factors that significantly influenced PM2.5 concentrations. The ENN estimated a PM2.5 monthly mean of 25.62 μg m−3 during the entire period. The results of the comparison between the ENN and ground measurements were as follows: correlation coefficient R ~ 0.90; root mean square error = 1.81 μg m−3; mean absolute error = 1.31 μg m−3. Overall, the PM2.5 levels were higher in winter and spring. The highest PM2.5 levels were located in the MMA, which is the major source of air pollution throughout this area. The estimated data indicated that PM2.5 was not distributed uniformly throughout the region but varied both spatially and temporally. These results led to the conclusion that the magnitude of air pollution varies among seasons and regions, and it is correlated with meteorological factors. The methodology developed in this study could be used to identify new monitoring sites and address information gaps.

Atmospheric Aerosol Distribution in 2016–2017 over the Eastern European Region Based on the GEOS-Chem Model

The spatial and temporal distributions of atmospheric aerosols have been simulated using the GEOS-Chem model over the sparsely investigated Eastern European region. The spatial distribution of the particulate matter (PM2.5) concentration, mineral dust, black carbon, organic aerosols, sea salt, as well as nitrate, sulfate, and ammonium aerosols during 2016–2017 were considered. The aerosols’ concentration, seasonality and spatial features were determined for the region. Particulate matter (PM2.5) contamination prevails in Poland in late autumn and winter. The monthly mean PM2.5 concentration reached 55 µg m−3 over the Moscow region in the early spring of both years. The mineral dust concentration varied significantly, reaching 40 µg m−3 over the southwestern part of Eastern Europe in March 2016. The areas most polluted by black carbon aerosols were the central and southern parts of Poland in the winter. The organic aerosols’ concentration was the largest in March and April, reaching 10 µg m−3 over East Belarus. The sea salt aerosol concentration increased in the coastal regions in winter due to the wind strength. Mineral dust aerosols in Eastern Europe are mainly composed of dust, partially transported from the Ukrainian steppe and partially from the Saharan Desert.

Global Distribution of Column Satellite Aerosol Optical Depth to Surface PM2.5 Relationships

Using a combined Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) mid-visible aerosol optical depth (AOD) product at 0.1 × 0.1-degree spatial resolution and collocated surface PM2.5 (particulate matter with aerodynamic diameter smaller than 2.5 μm) monitors, we provide a global five-year (2015–2019) assessment of the spatial and seasonal AOD–PM2.5 relationships of slope, intercepts, and correlation coefficients. Only data from ground monitors accessible through an open air-quality portal that are available to the worldwide community for air quality research and decision making are used in this study. These statistics that are reported 1 × 1-degree resolution are important since satellite AOD is often used in conjunction with spatially limited surface PM2.5 monitors to estimate global distributions of surface particulate matter concentrations. Results indicate that more than 3000 ground monitors are now available for PM2.5 studies. While there is a large spread in correlation coefficients between AOD and PM2.5, globally, averaged over all seasons, the correlation coefficient is 0.55 with a unit AOD producing 54 μgm−3 of PM2.5 (Slope) with an intercept of 8 μgm−3. While the number of surface PM2.5 measurements has increased by a factor of 10 over the last decade, a concerted effort is still needed to continue to increase these monitors in areas that have no surface monitors, especially in large population centers that will further leverage the strengths of satellite data.

Influence of meteorological conditions on PM2.5 concentrations across China: A review of methodology and mechanism

Air pollution over China has attracted wide interest from public and academic community. PM_2.5 is the primary air pollutant across China. Quantifying interactions between meteorological conditions and PM_2.5 concentrations are essential to understand the variability of PM_2.5 and seek methods to control PM_2.5. Since 2013, the measurement of PM_2.5 has been widely made at 1436 stations across the country and more than 300 papers focusing on PM_2.5-meteorology interactions have been published. This article is a comprehensive review on the meteorological impact on PM_2.5 concentrations_. We start with an introduction of general meteorological conditions and PM_2.5 concentrations across China, and then seasonal and spatial variations of meteorological influences on PM_2.5 concentrations. Next, major methods used to quantify meteorological influences on PM_2.5 concentrations are checked and compared. We find that causality analysis methods are more suitable for extracting the influence of individual meteorological factors whilst statistical models are good at quantifying the overall effect of multiple meteorological factors on PM_2.5 concentrations. Chemical Transport Models (CTMs) have the potential to provide dynamic estimation of PM_2.5 concentrations by considering anthropogenic emissions and the transport and evolution of pollutants. We then comprehensively examine the mechanisms how major meteorological factors may impact the PM_2.5 concentrations, including the dispersion, growth, chemical production, photolysis, and deposition of PM_2.5. The feedback effects of PM_2.5 concentrations on meteorological factors are also carefully examined. Based on this review, suggestions on future research and major meteorological approaches for mitigating PM_2.5 pollution are made finally.

Contribution of Satellite-Derived Aerosol Optical Depth PM2.5 Bayesian Concentration Surfaces to Respiratory-Cardiovascular Chronic Disease Hospitalizations in Baltimore, Maryland

The fine particulate matter baseline (PMB), which includes PM_2.5 monitor readings fused with Community Multiscale Air Quality (CMAQ) model predictions, using the Hierarchical Bayesian Model (HBM), is less accurate in rural areas without monitors. To address this issue, an upgraded HBM was used to form four experimental aerosol optical depth (AOD)-PM_2.5 concentration surfaces. A case-crossover design and conditional logistic regression evaluated the contribution of the AOD-PM_2.5 surfaces and PMB to four respiratory-cardiovascular hospital events in all 99 12 km² CMAQ grids, and in grids with and without ambient air monitors. For all four health outcomes, only two AOD-PM_2.5 surfaces, one not kriged (PMC) and the other kriged (PMCK), had significantly higher Odds Ratios (ORs) on lag days 0, 1, and 01 than PMB in all grids, and in grids without monitors. In grids with monitors, emergency department (ED) asthma PMCK on lag days 0, 1 and 01 and inpatient (IP) heart failure (HF) PMCK ORs on lag days 01 were significantly higher than PMB ORs. Warm season ORs were significantly higher than cold season ORs. Independent confirmation of these results should include AOD-PM_2.5 concentration surfaces with greater temporal-spatial resolution, now easily available from geostationary satellites, such as GOES-16 and GOES-17.

Optimal Inversion of Conversion Parameters from Satellite AOD to Ground Aerosol Extinction Coefficient Using Automatic Differentiation

Satellite aerosol optical depth (AOD) plays an important role for high spatiotemporal-resolution estimation of fine particulate matter with diameters ≤2.5 μm (PM2.5). However, the MODIS sensors aboard the Terra and Aqua satellites mainly measure column (integrated) AOD using the aerosol (extinction) coefficient integrated over all altitudes in the atmosphere, and column AOD is less related to PM2.5 than low-level or ground-based aerosol (extinction) coefficient (GAC). With recent development of automatic differentiation (AD) that has been widely applied in deep learning, a method using AD to find optimal solution of conversion parameters from column AOD to the simulated GAC is presented. Based on the computational graph, AD has considerably improved the efficiency in applying gradient descent to find the optimal solution for complex problems involving multiple parameters and spatiotemporal factors. In a case study of the Jing-Jin-Ji region of China for the estimation of PM2.5 in 2015 using the Multiangle Implementation of Atmospheric Correction AOD, the optimal solution of the conversion parameters was obtained using AD and the loss function of mean square error. This solution fairly modestly improved the Pearson’s correlation between simulated GAC and PM2.5 up to 0.58 (test R2: 0.33), in comparison with three existing methods. In the downstream validation, the simulated GACs were used to reliably estimate PM2.5, considerably improving test R2 up to 0.90 and achieving consistent match for GAC and PM2.5 in their spatial distribution and seasonal variations. With the availability of the AD tool, the proposed method can be generalized to the inversion of other similar conversion parameters in remote sensing.

Children's acute respiratory symptoms associated with PM2.5 estimates in two sequential representative surveys from the Mexico City Metropolitan Area

BACKGROUND

Respiratory diseases are a major component of morbidity in children and their symptoms may be spatially and temporally exacerbated by exposure gradients of fine particulate matter (PM_2.5) in large polluted urban areas, like the Mexico City Metropolitan Area (MCMA).

OBJECTIVES

To analyze the association between satellite-derived and interpolated PM_2.5 estimates with children's (≤9 years old) acute respiratory symptoms (ARS) in two probabilistic samples representing the MCMA.

METHODS

We obtained ARS data from the 2006 and 2012 National Surveys for Health and Nutrition (ENSaNut). Two week average exposure to PM_2.5 was assessed for each household with spatial estimates from a hybrid model with satellite measurements of aerosol optical depth (AOD-PM_2.5) and also with interpolated PM_2.5 measurements from ground stations, from the Mexico City monitoring network (MNW-PM_2.5). We used survey-adjusted logistic regressions to analyze the association between PM_2.5 estimates and ARS reported on children.

RESULTS

A total of 1,005 and 1,233 children were surveyed in 2006 and 2012 representing 3.1 and 3.5 million children, respectively. For the same years and over the periods of study, the estimated prevalence of ARS decreased from 49.4% (95% CI: 44.9,53.9%) to 37.8% (95% CI: 34,41.7%). AOD-PM_2.5 and MNW-PM_2.5 estimates were associated with significantly higher reports of ARS in children 0-4 years old [OR₂₀₀₆ = 1.29 (95% (CI): 0.99,1.68) and OR₂₀₀₆ = 1.24 (95% CI: 1.08,1.42), respectively]. We observed positive non-significant associations in 2012 in both age groups and in 2006 for children 5-9 years old. No statistically significant differences in health effect estimates of PM_2.5 were found comparing AOD-PM_2.5 or MNW-PM_2.5 for exposure assessment.

CONCLUSIONS

Our findings suggest that PM_2.5 is a risk factor for the prevalence of ARS in children and expand the growing evidence of the utility of new satellite AOD-based methods for estimating health effects from acute exposure to PM_2.5.

Contribution of low-cost sensor measurements to the prediction of PM2.5 levels: A case study in Imperial County, California, USA

Regulatory monitoring networks are often too sparse to support community-scale PM2.5 exposure assessment while emerging low-cost sensors have the potential to fill in the gaps. To date, limited studies, if any, have been conducted to utilize low-cost sensor measurements to improve PM2.5 prediction with high spatiotemporal resolutions based on statistical models. Imperial County in California is an exemplary region with sparse Air Quality System (AQS) monitors and a community-operated low-cost network entitled Identifying Violations Affecting Neighborhoods (IVAN). This study aims to evaluate the contribution of IVAN measurements to the quality of PM2.5 prediction. We adopted the Random Forest algorithm to estimate daily PM2.5 concentrations at a 1-km spatial resolution using three different PM2.5 datasets (AQS-only, IVAN-only, and AQS/IVAN combined). The results show that the integration of low-cost sensor measurements is an effective way to significantly improve the quality of PM2.5 prediction with an increase of cross-validation (CV) R2 by ~0.2. The IVAN measurements also contributed to the increased importance of emission source-related covariates and more reasonable spatial patterns of PM2.5. The remaining uncertainty in the calibrated IVAN measurements could still cause apparent outliers in the prediction model, highlighting the need for more effective calibration or integration methods to relieve its negative impact.

Stacking machine learning model for estimating hourly PM2.5 in China based on Himawari 8 aerosol optical depth data

Aerosol optical depth (AOD) from polar orbit satellites and meteorological factors have been widely used to estimate concentrations of surface particulate matter with an aerodynamic diameter <2.5 μm (PM_2.5). However, estimations with high temporal resolution remain lacking because of the limitations of satellite observations. Here, we used AOD data with a temporal resolution of 1 h provided by a geostationary satellite called Himawari 8 to overcome this problem. We developed a stacking model, which contained three submodels of machine learning, namely, AdaBoost, XGBoost and random forest, stacked through a multiple linear regression model. Then, we estimated the hourly concentrations of PM_2.5 in Central and Eastern China. The accuracy evaluation showed that the proposed stacking model performed better than the single models when applied to the test set, with an average coefficient of determination (R²) of 0.85 and a root-mean-square error (RMSE) of 17.3 μg/m³. Model precision reached its peak at 14:00 (local time), with an R² (RMSE) of 0.92 (12.9 μg/m³). In addition, the spatial and temporal distributions of PM_2.5 in Central and Eastern China were plotted in this study. The North China Plain was determined to be the most polluted area in China, with an annual mean PM_2.5 concentration of 58 μg/m³ during daytime. Moreover, the pollution level of PM_2.5 was the highest in winter, with an average concentration of 73 μg/m³.

Methods, availability, and applications of PM2.5 exposure estimates derived from ground measurements, satellite, and atmospheric models

Fine particulate matter (PM2.5) is a well-established risk factor for public health. To support both health risk assessment and epidemiological studies, data are needed on spatial and temporal patterns of PM2.5 exposures. This review article surveys publicly available exposure datasets for surface PM2.5 mass concentrations over the contiguous U.S., summarizes their applications and limitations, and provides suggestions on future research needs. The complex landscape of satellite instruments, model capabilities, monitor networks, and data synthesis methods offers opportunities for research development, but would benefit from guidance for new users. Guidance is provided to access publicly available PM2.5 datasets, to explain and compare different approaches for dataset generation, and to identify sources of uncertainties associated with various types of datasets. Three main sources used to create PM2.5 exposure data are ground-based measurements (especially regulatory monitoring), satellite retrievals (especially aerosol optical depth, AOD), and atmospheric chemistry models. We find inconsistencies among several publicly available PM2.5 estimates, highlighting uncertainties in the exposure datasets that are often overlooked in health effects analyses. Major differences among PM2.5 estimates emerge from the choice of data (ground-based, satellite, and/or model), the spatiotemporal resolutions, and the algorithms used to fuse data sources.Implications: Fine particulate matter (PM2.5) has large impacts on human morbidity and mortality. Even though the methods for generating the PM2.5 exposure estimates have been significantly improved in recent years, there is a lack of review articles that document PM2.5 exposure datasets that are publicly available and easily accessible by the health and air quality communities. In this article, we discuss the main methods that generate PM2.5 data, compare several publicly available datasets, and show the applications of various data fusion approaches. Guidance to access and critique these datasets are provided for stakeholders in public health sectors.

Modeling Wildland Fire-Specific PM2.5 Concentrations for Uncertainty-Aware Health Impact Assessments

Wildland fire is a major emission source of fine particulate matter (PM_2.5), which has serious adverse health effects. Most fire-related health studies have estimated human exposures to PM_2.5 using ground observations, which have limited spatial/temporal coverage and could not separate PM_2.5 emanating from wildland fires from other sources. The Community Multiscale Air Quality (CMAQ) model has the potential to fill the gaps left by ground observations and estimate wildland fire-specific PM_2.5 concentrations, although the issues around systematic bias in CMAQ models remain to be resolved. To address these problems, we developed a two-step calibration strategy under the consideration of prediction uncertainties. In a case study of the eastern U.S. in 2014, we evaluated the calibration performance using three cross-validation methods, which consistently indicated that the prediction accuracy was improved with an R² of 0.47-0.64. In a health impact study based on the wildland fire-specific PM_2.5 predictions, we identified regions with excess respiratory hospital admissions due to wildland fire events and quantified the estimation uncertainty propagated from multiple components in health impact function. We concluded that the proposed calibration strategy could provide reliable wildland fire-specific PM_2.5 predictions and health burden estimates to support policy development for reducing fire-related risks.

Predicting ground-level PM2.5 concentrations in the Beijing-Tianjin-Hebei region: A hybrid remote sensing and machine learning approach

An accurate estimation of PM_2.5 (fine particulate matters with diameters ≤ 2.5 μm) concentration is critical for health risk assessment and generating air pollution control strategies. In this study, a hybrid remote sensing and machine learning approach, named RSRF model is proposed to estimate daily ground-level PM_2.5 concentrations, which integrates Random Forest (RF), one of machine learning (ML) models, and aerosol optical depth (AOD), one of remote sensing (RS) products. The proposed RSRF model provides an opportunity for an adequate characterization of real-time spatiotemporal PM_2.5 distributions at uninhabited places and complex surfaces. It also offers advantages in handling complicated non-linear relationships among a large number of meteorological, environmental and air pollutant factors, as well as ever-increasing environmental data sets. The applicability of the proposed RSRF model is tested in the Beijing-Tianjin-Hebei region (BTH region) during 2015-2017. Deep Blue (DB) AOD from Aqua-retrieved Collection 6.1 (C_61) aerosol products of Moderate Resolution Imaging Spectroradiometer (MODIS) is validated with Aerosol Robotic Network. The validation results indicate C_61 DB AOD has a high correlation with ground based AOD in the BTH region. The proposed RSRF model performed well in characterizing spatiotemporal variations of annual and seasonal PM_2.5 concentrations. It not only is useful to quantify the relationships between PM_2.5 and relevant factors such as DB AOD, meteorological and air pollutant variables, but also can provide decision support for air pollution control at a regional environment during haze periods.

PM2.5 mapping using integrated geographically temporally weighted regression (GTWR) and random sample consensus (RANSAC) models

An uncertainty in the relationship between aerosol optical depth (AOD) and fine particulate matter (PM_2.5) comes from the uncertainty of AOD by aerosol models and the estimated surface reflectance, a mismatch in spatiotemporal resolution, integration of AOD and PM_2.5 data, and data modeling. In this study, an integrated geographically temporally weighted regression (GTWR) and RANdom SAmple Consensus (RANSAC) models, which provide fine goodness-of-fit between observed PM_2.5 and AOD data, were used for mapping of PM_2.5 over Taiwan for the year 2014. For this, dark target (DT) AOD observations at 3-km resolution (DT_3K) only for high-quality assurance flag (QA = 3) were obtained from the scientific data set (SDS) "Optical_Depth_Land_And_Ocean". AOD observations were also obtained from the merged DT and DB (deep blue) product (DTB_3K) which was generated using the simplified merge scheme (SMS), i.e., using an average of the DT and DB highest quality AOD retrievals or the available one. The GTWR model integrated with RANSAC can use the effective sampling and fitting to overcome the estimation problem of AOD-PM_2.5 with the uncertainty and outliers of observation data. Results showed that the model dealing with spatiotemporal heterogeneity and uncertainty is a powerful tool to infer patterns of PM_2.5 from a RANSAC subset samples. Moreover, spatial variability and hotspot analysis were applied after PM_2.5 mapping. The hotspot and spatial variability of PM_2.5 maps can give us a summary of the spatiotemporal patterns of PM_2.5 variations.

A nonparametric approach to filling gaps in satellite-retrieved aerosol optical depth for estimating ambient PM2.5 levels

Satellite-retrieved aerosol optical depth (AOD) is commonly used to estimate ambient levels of fine particulate matter (PM_2.5), though it is important to mitigate the estimation bias of PM_2.5 due to gaps in satellite-retrieved AOD. A nonparametric approach with two random-forest submodels is proposed to estimate PM_2.5 levels by filling gaps in satellite-retrieved AOD. This novel approach was employed to estimate the spatiotemporal distribution of daily PM_2.5 levels during 2013-2015 in the Sichuan Basin of Southwest China, where the coverage rate of composite AOD retrieved by the Terra and Aqua satellites was only 11.7%. Based on the retrieved AOD and various covariates (including meteorological conditions and land use types), the first random-forest submodel (named AOD-submodel) was trained to fill the gaps in the AOD dataset, giving a cross-validation R² of 0.95. Subsequently, the second random-forest submodel (named PM_2.5-submodel) was trained to estimate the PM_2.5 levels for unmonitored areas/days based on the gap-filled AOD, ground-monitored PM_2.5 levels, and the covariates, and achieved a cross-validation R² of 0.86. By comparing the complete and incomplete (i.e., without the days when AOD data were missing) estimates, we found that the monthly PM_2.5 levels could be overestimated by 34.6% if the PM_2.5 values coincident with AOD gaps were not considered. The newly developed approach is valuable for deriving the complete spatiotemporal distribution of daily PM_2.5 from incomplete remote-sensing data, which is essential for air quality management and human exposure assessment.

Bayesian geostatistical modelling of PM10 and PM2.5 surface level concentrations in Europe using high-resolution satellite-derived products

Air quality monitoring across Europe is mainly based on in situ ground stations, which are too sparse to accurately assess the exposure effects of air pollution for the entire continent. The demand for precise predictive models that estimate gridded geophysical parameters of ambient air at high spatial resolution has rapidly grown. Here, we investigate the potential of satellite-derived products to improve particulate matter (PM) estimates. Bayesian geostatistical models addressing confounding between the spatial distribution of pollutants and remotely sensed predictors were developed to estimate yearly averages of both, fine (PM2.5) and coarse (PM10) surface PM concentrations, at 1 km2 spatial resolution over 46 European countries. Model outcomes were compared to geostatistical, geographically weighted and land-use regression formulations. Rigorous model selection identified the Earth observation data which contribute most to pollutants' estimation. Geostatistical models outperformed the predictive ability of the frequently employed land-use regression. The resulting estimates of PM10 and PM2.5, which represent the main air quality indicators for the urban Sustainable Development Goal, indicate that in 2016, 66.2% of the European population was breathing air above the WHO air quality guidelines thresholds. Our estimates are readily available to policy makers and scientists assessing the effects of long-term exposure to pollution on human and ecosystem health.

Evaluation of machine learning techniques with multiple remote sensing datasets in estimating monthly concentrations of ground-level PM2.5

Fine particulate matter (PM_2.5) has been recognized as a key air pollutant that can influence population health risk, especially during extreme cases such as wildfires. Previous studies have applied geospatial techniques such as land use regression to map the ground-level PM_2.5, while some recent studies have found that Aerosol Optical Depth (AOD) derived from satellite images and machine learning techniques may be two elements that can improve spatiotemporal prediction. However, there has been a lack of studies evaluating use of different machine learning techniques with AOD datasets for mapping PM_2.5, especially in areas with high spatiotemporal variability of PM_2.5. In this study, we compared the performance of eight predictive algorithms with the use of multiple remote sensing datasets, including satellite-derived AOD data, for the prediction of ground-level PM2.5 concentration. Based on the results, Cubist, random forest and eXtreme Gradient Boosting were the algorithms with better performance, while Cubist was the best (CV-RMSE = 2.64 μg/m3, CV-R² = 0.48). Variable importance analysis indicated that the predictors with the highest contributions in modelling were monthly AOD and elevation. In conclusion, appropriate selection of machine learning algorithms can improve ground-level PM2.5 estimation, especially for areas with nonlinear relationships between PM2.5 and predictors caused by complex terrain. Satellite-derived data such as AOD and land surface temperature (LST) can also be substitutes for traditional datasets retrieved from weather stations, especially for areas with sparse and uneven distribution of stations.

Modeling Wildfire Smoke Pollution by Integrating Land Use Regression and Remote Sensing Data: Regional Multi-Temporal Estimates for Public Health and Exposure Models

To understand the health effects of wildfire smoke, it is important to accurately assess smoke exposure over space and time. Particulate matter (PM) is a predominant pollutant in wildfire smoke. In this study, we develop land-use regression (LUR) models to investigate the impact that a cluster of wildfires in the northwest USA had on the level of PM in southern Alberta (Canada), in the summer of 2015. Univariate aerosol optical depth (AOD) and multivariate AOD-LUR models were used to estimate the level of PM2.5 in urban and rural areas. For epidemiological studies, it is also important to distinguish between wildfire-related PM2.5 and PM2.5 originating from other sources. We therefore subdivided the study period into three sub-periods: (1) Pre-fire, (2) during-fire, and (3) post-fire. We then developed separate models for each sub-period. With this approach, we were able to identify different predictors significantly associated with smoke-related PM2.5 verses PM2.5 of different origin. Leave-one-out cross-validation (LOOCV) was used to evaluate the models’ performance. Our results indicate that model predictors and model performance are highly related to the level of PM2.5, and the pollution source. The predictive ability of both uni- and multi-variate models were higher in the during-fire period than in the pre- and post-fire periods.

Incorporating long-term satellite-based aerosol optical depth, localized land use data, and meteorological variables to estimate ground-level PM2.5 concentrations in Taiwan from 2005 to 2015

Satellite-based aerosol optical depth (AOD) is now comprehensively applied to estimate ground-level concentrations of fine particulate matter (PM_2.5). This study aimed to construct the AOD-PM_2.5 estimation models over Taiwan. The AOD-PM_2.5 modeling in Taiwan island is challenging owing to heterogeneous land use, complex topography, and humid tropical to subtropical climate conditions with frequent cloud cover and prolonged rainy season. The AOD retrievals from the MODerate resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites were combined with the meteorological variables from reanalysis data and high resolution localized land use variables to estimate PM_2.5 over Taiwan island from 2005 to 2015. Ten-fold cross validation was carried out and the residuals of the estimation model at various locations and seasons are assessed. The cross validation (CV) R² based on monitoring stations were 0.66 and 0.66, with CV root mean square errors of 14.0 μg/m³ (34%) and 12.9 μg/m³ (33%), respectively, for models based on Terra and Aqua AOD. The results provided PM_2.5 estimations at locations without surface stations. The estimation revealed PM_2.5 concentration hotspots in the central and southern part of the western plain areas, particularly in winter and spring. The annual average of estimated PM_2.5 concentrations over Taiwan consistently declined during 2005-2015. The AOD-PM_2.5 model is a reliable and validated method for estimating PM_2.5 concentrations at locations without monitoring stations in Taiwan, which is crucial for epidemiological study and for the assessment of air quality control policy.