Incorporating local land use regression and satellite aerosol optical depth in a hybrid model of spatiotemporal PM2.5 exposures in the Mid-Atlantic states

Capabilities of satellite Geostationary Environment Monitoring Spectrometer (GEMS) NO2 data for hourly ambient NO2 exposure modeling

The Geostationary Environment Monitoring Spectrometer (GEMS) is the world's first geostationary instrument that monitors hourly gaseous air pollutant levels, including nitrogen dioxide (NO2). Using the first-of-its-kind capabilities of GEMS NO2 data, we examined how well GEMS NO2 levels can explain the spatiotemporal variabilities in hourly NO2 concentrations in the Republic of Korea for the year 2022. A correlation analysis between hourly GEMS NO2 levels and ground NO2 concentrations showed a higher spatial correlation [Pearson r = 0.56 (SD = 0.20)] than a temporal one [Pearson r = 0.42 (SD = 0.14)], on average. To take advantage of the enhanced spatial predictability of GEMS NO2 data, we employed a mixed effects model to allow hour-specific relationships between GEMS NO2 and NO2 concentrations on a given day in each region and subsequently estimated hourly NO2 concentrations in all urban and rural areas. The 10-fold cross validation demonstrated R2 = 0.72, mean absolute error (MAE) = 3.7 ppb, and root mean squared error (RMSE) = 5.5 ppb. The hourly variations of the relationships were attributed particularly to those of wind speed among meteorological parameters considered in this study. The spatial distributions of hourly estimated NO2 concentrations were highly correlated between hours [average r = 0.91 (SD = 0.06)]. Nonetheless, they represented the diurnal patterns of urban versus rural NO2 contrasts during the day [urban/rural NO2 ratios from 1.22 (5 p.m.) to 1.37 (12 p.m.)]. The newly retrieved GEMS NO2 data enable temporally as well as spatially resolved NO2 exposure assessment. In combination with the time-activity patterns of individual subjects, the GEMS NO2 data can generate 'sub-population' exposure estimates and therefore enhance health effect studies.

A review of geospatial exposure models and approaches for health data integration

BACKGROUND

Geospatial methods are common in environmental exposure assessments and increasingly integrated with health data to generate comprehensive models of environmental impacts on public health.

OBJECTIVE

Our objective is to review geospatial exposure models and approaches for health data integration in environmental health applications.

METHODS

We conduct a literature review and synthesis.

RESULTS

First, we discuss key concepts and terminology for geospatial exposure data and models. Second, we provide an overview of workflows in geospatial exposure model development and health data integration. Third, we review modeling approaches, including proximity-based, statistical, and mechanistic approaches, across diverse exposure types, such as air quality, water quality, climate, and socioeconomic factors. For each model type, we provide descriptions, general equations, and example applications for environmental exposure assessment. Fourth, we discuss the approaches used to integrate geospatial exposure data and health data, such as methods to link data sources with disparate spatial and temporal scales. Fifth, we describe the landscape of open-source tools supporting these workflows.

Global economic structure transition boosts PM2.5-related human health impact in Belt and Road Initiative

The Belt and Road Initiative (BRI) is an open platform for international cooperation proposed by China to promote common global development and prosperity. The BRI can promote the optimal allocation of resources and promote in-depth cooperation in international trade. Meanwhile, it can establish a green supply chain cooperation network to help BRI countries achieve green transformation. BRI has made a notable contribution to the rapid growth of cross-border trade. However, it has also brought environmental impacts. Given that little attention has been paid to the trade-embodied particulate matter 2.5 related human health impacts (PM2.5-HHI) throughout the BRI, this study accounts for and traces the embodied PM2.5-HHI flows between the BRI countries and non-Belt and Road Initiative (non-BRI) countries. Moreover, this study also uncovers the critical socioeconomic drivers of PM2.5-HHI changes in BRI countries during 1990-2015, based on the multi-regional input-output based structural decomposition analysis (MRIO-SDA). Results show that, firstly, BRI countries had significantly increased their economic added value by exporting products to the non-BRI countries. They also have brought PM2.5-HHI to themselves. Secondly, the final demand of BRI countries was the largest potential driving force of PM2.5-HHI of BRI countries. Thirdly, the emission intensity change of BRI is the key socioeconomic factor for reducing PM2.5-HHI. While per capita final demand level change of BRI and production structure change of non-BRI are the key socioeconomic factors for increasing PM2.5-HHI. The study's findings on the one hand can help reduce the PM2.5-HHI and impacts of environmental pollution of BRI countries from a global perspective by providing scientific support. On the other hand, they can help provide relevant policy recommendations for the green transformation of BRI and the construction of green BRI.

Time-sensitive prediction of NO2 concentration in China using an ensemble machine learning model from multi-source data

Nitrogen dioxide (NO2) poses a critical potential risk to environmental quality and public health. A reliable machine learning (ML) forecasting framework will be useful to provide valuable information to support government decision-making. Based on the data from 1609 air quality monitors across China from 2014-2020, this study designed an ensemble ML model by integrating multiple types of spatial-temporal variables and three sub-models for time-sensitive prediction over a wide range. The ensemble ML model incorporates a residual connection to the gated recurrent unit (GRU) network and adopts the advantage of Transformer, extreme gradient boosting (XGBoost) and GRU with residual connection network, resulting in a 4.1%±1.0% lower root mean square error over XGBoost for the test results. The ensemble model shows great prediction performance, with coefficient of determination of 0.91, 0.86, and 0.77 for 1-hr, 3-hr, and 24-hr averages for the test results, respectively. In particular, this model has achieved excellent performance with low spatial uncertainty in Central, East, and North China, the major site-dense zones. Through the interpretability analysis based on the Shapley value for different temporal resolutions, we found that the contribution of atmospheric chemical processes is more important for hourly predictions compared with the daily scale predictions, while the impact of meteorological conditions would be ever-prominent for the latter. Compared with existing models for different spatiotemporal scales, the present model can be implemented at any air quality monitoring station across China to facilitate achieving rapid and dependable forecast of NO2, which will help developing effective control policies.

Estimating daily air temperature and pollution in Catalonia: A comprehensive spatiotemporal modelling of multiple exposures

Environmental epidemiology studies require models of multiple exposures to adjust for co-exposure and explore interactions. We estimated spatiotemporal exposure to surface air temperature and pollution (PM2.5, PM10, NO2, O3) at high spatiotemporal resolution (daily, 250 m) for 2018-2020 in Catalonia. Innovations include the use of TROPOMI products, a data split for remote sensing gap-filling evaluation, estimation of prediction uncertainty, and use of explainable machine learning. We compiled meteorological and air quality station measurements, climate and atmospheric composition reanalyses, remote sensing products, and other spatiotemporal data. We performed gap-filling of remotely-sensed products using Random Forest (RF) models and validated them using Out-Of-Bag (OOB) samples and a structured data split. The exposure modelling workflow consisted of: 1) PM2.5 station imputation with PM10 data; 2) quantile RF (QRF) model fitting; and 3) geostatistical residual spatial interpolation. Prediction uncertainty was estimated using QRF. SHAP values were used to examine variable importance and the fitted relationships. Model performance was assessed via nested CV at the station level. Evaluation of the gap-filling models using the structured split showed error underestimation when using OOB. Temperature models had the best performance (R2 =0.98) followed by the gaseous air pollutants (R2 =0.81 for NO2 and 0.86 for O3), while the performance of the PM2.5 and PM10 models was lower (R2 =0.57 and 0.63 respectively). Predicted exposure patterns captured urban heat island effects, dust advection events, and NO2 hotspots. SHAP values estimated a high importance of TROPOMI tropospheric NO2 columns in PM and NO2 models, and confirmed that the fitted associations conformed to prior knowledge. Our work highlights the importance of correctly validating gap-filling models and the potential of TROPOMI measurements. Moderate performance in PM models can be partly explained by the poor station coverage. Our exposure estimates can be used in epidemiological studies potentially accounting for exposure uncertainty.

Machine learning-based country-level annual air pollutants exploration using Sentinel-5P and Google Earth Engine

Climatic condition is triggering human health emergencies and earth's surface changes. Anthropogenic activities, such as built-up expansion, transportation development, industrial works, and some extreme phases, are the main reason for climate change and global warming. Air pollutants are increased gradually due to anthropogenic activities and triggering the earth's health. Nitrogen Dioxide (NO₂), Carbon Monoxide (CO), and Aerosol Optical Depth (AOD) are truthfully important for air quality measurement because those air pollutants are more harmful to the environment and human's health. Earth observational Sentinel-5P is applied for monitoring the air pollutant and chemical conditions in the atmosphere from 2018 to 2021. The cloud computing-based Google Earth Engine (GEE) platform is applied for monitoring those air pollutants and chemical components in the atmosphere. The NO₂ variation indicates high during the time because of the anthropogenic activities. Carbon Monoxide (CO) is also located high between two 1-month different maps. The 2020 and 2021 results indicate AQI change is high where 2018 and 2019 indicates low AQI throughout the year. The Kolkata have seven AQI monitoring station where high nitrogen dioxide recorded 102 (2018), 48 (2019), 26 (2020) and 98 (2021), where Delhi AQI stations recorded 99 (2018), 49 (2019), 37 (2020), and 107 (2021). Delhi, Kolkata, Mumbai, Pune, and Chennai recorded huge fluctuations of air pollutants during the study periods, where ~ 50-60% NO₂ was recorded as high in the recent time. The AOD was noticed high in Uttar Pradesh in 2020. These results indicate that air pollutant investigation is much necessary for future planning and management otherwise; our planet earth is mostly affected by the anthropogenic and climatic conditions where maybe life does not exist.

A daily and complete PM2.5 dataset derived from space observations for Vietnam from 2012 to 2020

PM_2.5 pollution is a serious problem in Vietnam and around the world, having bad impacts on human health, animals and environment. Regular monitoring at a large scale is important to assess the status of air pollution, develop solutions and evaluate the effectiveness of policy implementation. However, air quality monitoring stations in Vietnam are limited. In this article, we propose an approach to estimate daily PM_2.5 concentration from 2012 to 2020 over the Vietnamese territory, which is strongly affected by cloudy conditions, using a modern statistical model named Mixed Effect Model (MEM) on a dataset consisting of ground PM_2.5 measurements, integrated satellite Aerosol Optical Depth (AOD), meteorological and land use maps. The result of this approach is the first long-term, full coverage and high quality PM_2.5 dataset of Vietnam. The daily mean PM_2.5 maps have high validation results in comparison with ground PM_2.5 measurement (Pearson r of 0.87, R² of 0.75, RMSE of 11.76 μg/m³, and MRE of 36.57 % on a total of 13,886 data samples). The aggregated monthly and annual average maps from 2012 to 2020 in Vietnam have outstanding quality when compared with another global PM_2.5 product. The PM_2.5 concentration maps has shown spatial distribution and seasonal variations of PM_2.5 concentration in Vietnam in a long period from 2012 to 2020 and has been used in other studies and applications in the environment and public health at the national scale, which has not been possible before because of the lack of monitoring stations and an appropriate PM_2.5 modeling approach.

Spatiotemporal PM2.5 estimations in China from 2015 to 2020 using an improved gradient boosting decision tree

Fine particulate matter (PM_2.5) with spatiotemporal continuity can provide important basis for the assessment of adverse effects on human health. In recent years, researchers have done a lot of work on the surface PM_2.5 simulation. However, due to the limitations of data and models, it is difficult to accurately evaluate the spatial and temporal PM_2.5 variations on a fine scale. In this study, we adopted the multi-angle implementation of atmospheric correction (MAIAC) aerosol products, and proposed a spatiotemporal model based on the gradient boosting decision tree (GBDT) algorithm to retrieve PM_2.5 concentration across China from 2015 to 2020 at 1-km resolution. Our model achieved excellent performance, with overall CV-R² of 0.92, and annual CV-R² of 0.90-0.93. In addition, the model can also be used for evaluation on different time scales. Compared with previous studies, the model developed in our study performed better and more stable, which showed the highest accuracies in PM_2.5 estimation works at 1-km resolution. During the study period, the overall national PM_2.5 pollution showed a downward trend, with the annual mean concentration dropping from 42.42 μg/m³ to 27.91 μg/m³. The largest decrease occurred in Beijing-Tianjin-Hebei (BTH), with a trend of -5.17 μg/m³/yr, while it remains the most polluted region. The area meeting the secondary national air quality standard (<35 μg/m³) increased from ∼34% to ∼79%. These results indicate that the atmospheric environment has improved significantly. Moreover, different regions have different time nodes for the start of the continuous standard-met day during the year, and the duration is different as well. Overall, this study can provide reliable large-scale PM_2.5 estimations.

Identifying Spatiotemporal Heterogeneity of PM2.5 Concentrations and the Key Influencing Factors in the Middle and Lower Reaches of the Yellow River

Fine particulate matter (PM2.5) is a harmful air pollutant that seriously affects public health and sustainable urban development. Previous studies analyzed the spatial pattern and driving factors of PM2.5 concentrations in different regions. However, the spatiotemporal heterogeneity of various influencing factors on PM2.5 was ignored. This study applies the geographically and temporally weighted regression (GTWR) model and geographic information system (GIS) analysis methods to investigate the spatiotemporal heterogeneity of PM2.5 concentrations and the influencing factors in the middle and lower reaches of the Yellow River from 2000 to 2017. The findings indicate that: (1) the annual average of PM2.5 concentrations in the middle and lower reaches of the Yellow River show an overall trend of first rising and then decreasing from 2000 to 2017. In addition, there are significant differences in inter-province PM2.5 pollution in the study area, the PM2.5 concentrations of Tianjin City, Shandong Province, and Henan Province were far higher than the overall mean value of the study area. (2) PM2.5 concentrations in western cities showed a declining trend, while it had a gradually rising trend in the middle and eastern cities of the study area. Meanwhile, the PM2.5 pollution showed the characteristics of path dependence and region locking. (3) the PM2.5 concentrations had significant spatial agglomeration characteristics from 2000 to 2017. The “High-High (H-H)” clusters were mainly concentrated in the southern Hebei Province and the northern Henan Province, and the “Low-Low (L-L)” clusters were concentrated in northwest marginal cities in the study area. (4) The influencing factors of PM2.5 have significant spatiotemporal non-stationary characteristics, and there are obvious differences in the direction and intensity of socio-economic and natural factors. Overall, the variable of temperature is one of the most important natural conditions to play a positive impact on PM2.5, while elevation makes a strong negative impact on PM2.5. Car ownership and population density are the main socio-economic influencing factors which make a positive effect on PM2.5, while the variable of foreign direct investment (FDI) plays a strong negative effect on PM2.5. The results of this study are useful for understanding the spatiotemporal distribution characteristics of PM2.5 concentrations and formulating policies to alleviate haze pollution by policymakers in the Yellow River Basin.

Individual and joint effects of prenatal green spaces, PM2.5 and PM1 exposure on BMI Z-score of children aged two years: A birth cohort study

BACKGROUND

Few studies examined the association of prenatal exposure to green spaces with children's body mass index (BMI) Z-score, and no study evaluated the joint effect of prenatal green spaces and PM_2.5 or PM₁ exposure on children's BMI Z-score. We aimed to assess the individual and joint effects of prenatal green spaces, PM_2.5, and PM₁ exposure on BMI Z-score of children aged two years.

METHODS

The study was based on a birth cohort in Beijing, China, in which 13,253 mothers (LMP from 2014 to 2017) and their children were included. We estimated prenatal green spaces exposure by calculating average normalized difference vegetation index with 500 m buffers (NDVI-500), prenatal PM_2.5 and PM₁ exposure based on maternal residential addresses. Weight and height of children were measured at 2 years old. We calculated children's BMI Z-score based on the WHO Standards. Generalized linear regression was used to examine the individual and joint effects of prenatal NDVI-500, PM_2.5 and PM₁ exposure on children's BMI Z-score.

RESULTS

A 0.1 increase in prenatal NDVI-500 exposure, a 10 μg/m³ decrease in PM_2.5, a 10 μg/m³ decrease in PM₁ were associated with 0.185 [95% confidence interval (95%CI): 0.155, 0.216], 0.034 (95%CI: 0.015, 0.052) and 0.041 (95%CI: 0.020, 0.061) increase of children's BMI Z-score, respectively. Compared with those exposed to low-level NDVI-500 (not greater than median) and high-level PM_2.5 (greater than median), the BMI Z-score was higher in children whose mother exposed to high-level of NDVI-500 and low-level PM_2.5 [β:0.172 (95%CI: 0.131, 0.214), P_interaction = 0.003]. Compared with those exposed to low-level NDVI-500 and high-level PM₁, the BMI Z-score was higher in children whose mother exposed to high-level of NDVI-500 and low-level PM₁ [β:0.169 (95%CI: 0.127, 0.210), P_interaction<0.001]. In the trimester-specific analysis, NDVI-500 and PM exposure during the second trimester have a consistent individual effect, together with a joint effect, on child growth.

CONCLUSION

The study suggested the beneficial effect of prenatal exposure to green spaces on child growth and its interaction with PM_2.5 and PM₁, especially in the second trimester. The findings call for developing public health policy to improve green infrastructure and control PM_2.5 and PM₁ concentrations, in order to promote child growth.

A Deep-Neural-Network-Based Aerosol Optical Depth (AOD) Retrieval from Landsat-8 Top of Atmosphere Data

The 30 m resolution Landsat data have been used for high resolution aerosol optical depth (AOD) retrieval based on radiative transfer models. In this paper, a Landsat-8 AOD retrieval algorithm is proposed based on the deep neural network (DNN). A total of 6390 samples were obtained for model training and validation by collocating 8 years of Landsat-8 top of atmosphere (TOA) data and aerosol robotic network (AERONET) AOD data acquired from 329 AERONET stations over 30°W–160°E and 60°N–60°S. The Google Earth Engine (GEE) cloud-computing platform is used for the collocation to avoid a large download volume of Landsat data. Seventeen predictor variables were used to estimate AOD at 500 nm, including the seven bands TOA reflectance, two bands TOA brightness (BT), solar and viewing zenith and azimuth angles, scattering angle, digital elevation model (DEM), and the meteorological reanalysis total columnar water vapor and ozone concentration. The leave-one-station-out cross-validation showed that the estimated AOD agreed well with AERONET AOD with a correlation coefficient of 0.83, root-mean-square error of 0.15, and approximately 61% AOD retrievals within 0.05 + 20% of the AERONET AOD. Theoretical comparisons with the physical-based methods and the adaptation of the developed DNN method to Sentinel-2 TOA data with a different spectral band configuration are discussed.

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.

Development of high-resolution spatio-temporal models for ambient air pollution in a metropolitan area of China from 2013 to 2019

Modeling high-resolution air pollution concentrations is essential to accurately assess exposure for population studies. The aim of this study is to establish an advanced exposure model to predict spatiotemporal changes in fine particulate matter (PM_2.5), nitrogen dioxides (NO₂), and ozone (O₃) concentrations in Shanghai, China. The model is constructed on a geo-statistical modeling framework that incorporates a dimension reduction regression approach and a spatial smoothing function to deal with fine-scale exposure variations. We used a dataset with comprehensive observational and predictor variables that included monitoring data from both national and local agencies from 2013 to 2019, a high-resolution geographical dataset of predictor variables, and a full-coverage weekly satellite data of the aerosol optical depth at a 1 × 1 km² resolution. Our model performed well in terms of the spatial and temporal prediction ability assessed by cross-validation (CV) for PM_2.5 (spatial R² = 0.89, temporal R² = 0.91), NO₂ (R² = 0.49, 0.78), and O₃ (R² = 0.67, 0.81) at the national monitors over seven years according to the leave-one-out CV. For the predictions at the local agency monitoring stations, the overall CV R² was between 0.77 and 0.89 across the air pollutants. We visualized the long-term and seasonal averaged predictions of the PM_2.5, NO_2, and O₃ exposure on maps with a spatial resolution of 100 × 100 m². Our study provides a useful tool to accurately estimate air pollution exposure with high spatial and temporal resolution at the urban scale. These model predictions will be useful to assess both short-term and long-term air pollution exposure for health studies.

Early life exposure to air pollution and cell-mediated immune responses in preschoolers

BACKGROUND

Exposure to air pollution has been linked with altered immune function in adults, but little is known about its effects on early life. This study aimed to investigate the effects of exposure to air pollution during prenatal and postnatal windows on cell-mediated immune function in preschoolers.

METHODS

Pre-school aged children (2.9 ± 0.5 y old, n = 391) were recruited from a mother-child cohort study in Wuhan, China. We used a spatial-temporal land use regression (LUR) model to estimate exposures of particulate matter with aerodynamic diameters ≤2.5 μm (PM_2.5) and ≤10 μm (PM₁₀), and nitrogen dioxide (NO₂) during the specific trimesters of pregnancy and the first two postnatal years. We measured peripheral blood T lymphocyte subsets and plasma cytokines as indicators of cellular immune function. We used multiple informant models to examine the associations of prenatal and postnatal exposures to air pollution with cell-mediated immune function.

RESULTS

Prenatal exposures to PM_2.5, PM₁₀, and NO₂ during early pregnancy were negatively associated with %CD3⁺ and %CD3⁺CD8⁺ cells, and during late pregnancy were positively associated with %CD3⁺ cells. Postnatal exposures to these air pollutants during 1-y or 2-y childhood were positively associated with IL-4, IL-5, IL-6, and TNF-α. We also observed that the associations of prenatal or postnatal air pollution exposures with cellular immune responses varied by child's sex.

CONCLUSIONS

Our results suggest that exposure to air pollution during different critical windows of early life may differentially alter cellular immune responses, and these effects appear to be sex-specific.

High-Resolution PM2.5 Estimation Based on the Distributed Perception Deep Neural Network Model

The accurate measurement of the PM2.5 individual exposure level is a key issue in the study of health effects. However, the lack of historical data and the minute coverage of ground monitoring points are obstacles to the study of such issues. Based on the aerosol optical depth provided by NASA, combined with ground monitoring data and meteorological data, it is an effective method to estimate the near-ground concentration of PM2.5. With the deepening of related research, the models used have developed from univariate and multivariate linear models to nonlinear models such as support vector machine, random forest model, and deep learning neural network model. Among them, the depth neural network model has better performance. However, in the existing research, the variables used are input into the same neural network together, that is, the complex relationship caused by the lag effect of features and the correlation and partial correlation between features have not been considered. The above neural network framework can not be well applied to the complex situation of atmospheric systems and the estimation accuracy of the model needs to be improved. This is the first problem that we need to be overcome. Secondly, in the missing data value processing, the existing studies mostly use single interpolation methods such as linear fitting and Kriging interpolation. However, because the time and place of data missing are complex and changeable, a single method is difficult to deal with a large area of strip and block missing data. Moreover, the linear fitting method is easy to smooth out the special data in bad weather. This is the second problem that we need to overcome. Therefore, we construct a distributed perception deep neural network model (DP-DNN) and spatiotemporal multiview interpolation module to overcome problems 1 and 2. In empirical research, based on the Beijing–Tianjin–Hebei–Shandong region in 2018, we introduce 50 features such as meteorology, NDVI, spatial-temporal feature to analyze the relationship between AOD and PM2.5, and test the performance of DP-DNN and spatiotemporal multiview interpolation module. The results show that after applying the spatiotemporal multiview interpolation module, the average proportion of missing data decreases from 52.1% to 4.84%, and the relative error of the results is 27.5%. Compared with the single interpolation method, this module has obvious advantages in accuracy and level of completion. The mean absolute error, relative error, mean square error, and root mean square error of DP-DNN in time prediction are 17.7 μg/m3, 46.8%, 766.2 g2/m6, and 26.9 μg/m3, respectively, and in space prediction, they are 16.6 μg/m3, 41.8%, 691.5 μg2/m6, and 26.6 μg/m3. DP-DNN has higher accuracy and generalization ability. At the same time, the estimation method in this paper can estimate the PM2.5 of the selected longitude and latitude, which can effectively solve the problem of insufficient coverage of China’s meteorological environmental quality monitoring stations.

Temporal and Spatial Autocorrelation as Determinants of Regional AOD-PM2.5 Model Performance in the Middle East

Exposure to fine particulate matter (PM2.5) air pollution has been shown in numerous studies to be associated with detrimental health effects. However, the ability to conduct epidemiological assessments can be limited due to challenges in generating reliable PM2.5 estimates, particularly in parts of the world such as the Middle East where measurements are scarce and extreme meteorological events such as sandstorms are frequent. In order to supplement exposure modeling efforts under such conditions, satellite-retrieved aerosol optical depth (AOD) has proven to be useful due to its global coverage. By using AODs from the Multiangle Implementation of Atmospheric Correction (MAIAC) of the MODerate Resolution Imaging Spectroradiometer (MODIS) and the Multiangle Imaging Spectroradiometer (MISR) combined with meteorological and assimilated aerosol information from the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2), we constructed machine learning models to predict PM2.5 in the area surrounding the Persian Gulf, including Kuwait, Bahrain, and the United Arab Emirates (U.A.E). Our models showed regional differences in predictive performance, with better results in the U.A.E. (median test R2 = 0.66) than Kuwait (median test R2 = 0.51). Variable importance also differed by region, where satellite-retrieved AOD variables were more important for predicting PM2.5 in Kuwait than in the U.A.E. Divergent trends in the temporal and spatial autocorrelations of PM2.5 and AOD in the two regions offered possible explanations for differences in predictive performance and variable importance. In a test of model transferability, we found that models trained in one region and applied to another did not predict PM2.5 well, even if the transferred model had better performance. Overall the results of our study suggest that models developed over large geographic areas could generate PM2.5 estimates with greater uncertainty than could be obtained by taking a regional modeling approach. Furthermore, development of methods to better incorporate spatial and temporal autocorrelations in machine learning models warrants further examination.

Encoder-Decoder Full Residual Deep Networks for Robust Regression and Spatiotemporal Estimation

Although increasing hidden layers can improve the ability of a neural network in modeling complex nonlinear relationships, deep layers may result in degradation of accuracy due to the problem of vanishing gradient. Accuracy degradation limits the applications of deep neural networks to predict continuous variables with a small sample size and/or weak or little invariance to translations. Inspired by residual convolutional neural network in computer vision, we developed an encoder-decoder full residual deep network to robustly regress and predict complex spatiotemporal variables. We embedded full shortcuts from each encoding layer to its corresponding decoding layer in a systematic encoder-decoder architecture for efficient residual mapping and error signal propagation. We demonstrated, theoretically and experimentally, that the proposed network structure with full residual connections can successfully boost the backpropagation of signals and improve learning outcomes. This novel method has been extensively evaluated and compared with four commonly used methods (i.e., plain neural network, cascaded residual autoencoder, generalized additive model, and XGBoost) across different testing cases for continuous variable predictions. For model evaluation, we focused on spatiotemporal imputation of satellite aerosol optical depth with massive nonrandomness missingness and spatiotemporal estimation of atmospheric fine particulate matter [Formula: see text] (PM2.5). Compared with the other approaches, our method achieved the state-of-the-art accuracy, had less bias in predicting extreme values, and generated more realistic spatial surfaces. This encoder-decoder full residual deep network can be an efficient and powerful tool in a variety of applications that involve complex nonlinear relationships of continuous variables, varying sample sizes, and spatiotemporal data with weak or little invariance to translation.

Estimating PM2.5 concentrations via random forest method using satellite, auxiliary, and ground-level station dataset at multiple temporal scales across China in 2017

Fine particulate matter with aerodynamic diameters less than 2.5 μm (PM_2.5) poses adverse impacts on public health and the environment. It is still a great challenge to estimate high-resolution PM_2.5 concentrations at moderate scales. The current study calibrated PM_2.5 concentrations at a 1 km resolution scale using ground-level monitoring data, Aerosol Optical Depth (AOD), meteorological data, and auxiliary data via Random Forest (RF) model across China in 2017. The three ten-folded cross-validations (CV) methods including sample-based, time-based, and spatial-based validation combined with Coefficient Square (R²), Root-Mean-Square Error (RMSE), and Mean Predictive Error (MPE) have been used for validation at different temporal scales in terms of daily, monthly, heating seasonal, and non-heating seasonal. Finally, the distribution map of PM_2.5 concentrations was illustrated based on the RF model. Some findings were achieved. The RF model performed well, with a relatively high sample-based cross-validation R² of 0.74, a low RMSE of 16.29 μg × m^-3, and a small MPE of -0.282 μg × m^-3. Meanwhile, the performance of the RF model in inferring the PM_2.5 concentrations was well at urban scales except for Chengyu (CY). North China, the CY urban agglomeration, and the northwest of China exhibited relatively high PM_2.5 pollution features, especially in the heating season. The robustness of the RF model in the present study outperformed most statistical regression models for calibrating PM_2.5 concentrations. The outcomes can supply an up-to-date scientific dataset for epidemiological and air pollutants exposure risk studies across China.

Estimation of Ultrahigh Resolution PM2.5 Mass Concentrations Based on Mie Scattering Theory by Using Landsat8 OLI Images over Pearl River Delta

The aerosol optical depth (AOD), retrieved by satellites, has been widely used to estimate ground-level PM2.5 mass concentrations, due to its advantage of large-scale spatial continuity. However, it is difficult to obtain urban-scale pollution patterns from the coarse resolution retrieval results (e.g., 1 km, 3 km, or 10 km) at present, and little research has been conducted on PM2.5 mass concentration retrieval from high resolution remote sensing data. In this study, a physical model is proposed based on Mie scattering theory to evaluate the PM2.5 mass concentrations by using Landsat8 Operational Land Imager (OLI) images. First, the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) model (which can simulate the transmission process of solar radiation in the Earth-atmosphere system and calculate the radiance at the top of the atmosphere) is used to build a lookup table to retrieve the AOD of the coast and blue bands based on the improved deep blue (DB) method. Then, the Angstrom formula is used to obtain the AOD of the green and red bands. Second, the dry near-surface AOD of four bands (coast, blue, green, red) is obtained through vertical correction and humidity correction. Third, aerosol particles are divided into four types based on the standard radiation atmosphere (SRA) model, and the optical properties of different aerosol types are analyzed to derive the volume distribution of aerosol particles. Finally, the relationship between the dry near-surface AOD of each band and the volume distribution of four aerosol particles is correlated, based on Mie scattering theory, and a physical model is established between the AOD and PM2.5 mass concentrations. Then, the distribution of PM2.5 mass concentrations is obtained. The retrieval results show that the distribution of AOD and PM2.5 at the urban scale in detail. The AOD results show that a reasonable relationship with a correlation coefficient (R2) of 0.66 and root mean square error (RMSE) of 0.1037 between Landsat8 OLI AOD and MODO4 DB AOD at 550 nm. The PM2.5 retrieval results are compared with the PM2.5 values measured by ground monitoring stations. The RMSEs for a certain day in different years, including 2017, 2018, 2019, and 2020, are 11.9470 μg/m³, 11.9787 μg/m³, 7.4217 μg/m³, and 5.4723 μg/m³, respectively. The total RMSE is 10.0224 μg/m³. The ultrahigh resolution PM2.5 results can provide pollution details at the urban scale and support better decisions on urban atmospheric environmental governance.

Estimation of ambient PM2.5 in Iraq and Kuwait from 2001 to 2018 using machine learning and remote sensing

Iraq and Kuwait are in a region of the world known to be impacted by high levels of fine particulate matter (PM2.5) attributable to sources that include desert dust and ambient pollution, but historically have had limited pollution monitoring networks. The inability to assess PM2.5 concentrations have limited the assessment of the health impact of these exposures, both in the native populations and previously deployed military personnel. As part of a Department of Veterans Affairs Cooperative Studies Program health study of land-based U.S. military personnel who were previously deployed to these countries, we developed a novel approach to estimate spatially and temporarily resolved daily PM2.5 exposures 2001-2018. Since visibility is proportional to ground-level particulate matter concentrations, we were able to take advantage of extensive airport visibility data that became available as a result of regional military operations over this time period. First, we combined a random forest machine learning and a generalized additive mixed model to estimate daily high resolution (1 km × 1 km) visibility over the region using satellite-based aerosol optical depth (AOD) and airport visibility data. The spatially and temporarily resolved visibility data were then used to estimate PM2.5 concentrations from 2001 to 2018 by converting visibility to PM2.5 using empirical relationships derived from available regional PM2.5 monitoring stations. We adjusted for spatially resolved meteorological parameters, land use variables, including the Normalized Difference Vegetation Index, and satellite-derived estimates of surface dust as a measure of sandstorm activity. Cross validation indicated good model predictive ability (R2 = 0.71), and there were considerable spatial and temporal differences in PM2.5 across the region. Annual average PM2.5 predictions for Iraq and Kuwait were 37 and 41 μg/m3, respectively, which are greater than current U.S. and WHO standards. PM2.5 concentrations in many U.S. bases and large cities (e.g. Bagdad, Balad, Kuwait city, Karbala, Najaf, and Diwaniya) had annual average PM2.5 concentrations above 45 μg/m3 with weekly averages as high as 150 μg/m3 depending on calendar year. The highest annual PM2.5 concentration for both Kuwait and Iraq were observed in 2008, followed by 2009, which was associated with extreme drought in these years. The lowest PM2.5 values were observed in 2014. On average, July had the highest concentrations, and November had the lowest values, consistent with seasonal patterns of air pollution in this region. This is the first study that estimates long-term PM2.5 exposures in Iraq and Kuwait at a high resolution based on measurements data that will allow the study of health effects and contribute to the development of regional environmental policies. The novel approach demonstrated may be used in other parts of the world with limited monitoring networks.

Estimating ground-level PM2.5 concentrations by developing and optimizing machine learning and statistical models using 3 km MODIS AODs: case study of Tehran, Iran

PURPOSE

In this study we aimed to develop an optimized prediction model to estimate a fine-resolution grid of ground-level PM_2.5 levels over Tehran. Using remote sensing data to obtain fine-resolution grids of particulate levels in highly polluted environments in areas such as Middle East with the abundance of brightly reflecting deserts is challenging.

METHODS

Different prediction models implementing 3 km AOD products from the MODIS collection 6 and various effective parameters were used to obtain a reliable model to estimate ground-level PM_2.5 concentrations. In this regards, the linear mixed effect model (LME), multi-variable linear regression model (MLR), Gaussian process model (GPM), artificial neural network (ANN) and support vector regression (SVR) were developed and their performance were compared. Since the LME and GPM outperformed other models, they were further optimized based on meteorological and topographical variables. These models were used to estimate PM_2.5 values over the highly polluted megacity, Tehran, Iran. Moreover, the influence of site effect term on the performance of different shapes of LME models was evaluated by considering the random intercept for sites.

RESULTS

Results showed LME models without the site effect term were able to explain ground-level variabilities of PM_2.5 concentrations in ranges of 60-66% (RMSE = 9.6 to 10.3 μg/m³) and 35-41% (RMSE = 12.7 to 13.3 μg/m³) during the model-fitting and cross-validation, respectively. By considering the site effect term, the performance of LME models during calibrations and validations improved by 20% and 50% on average, respectively (18.5% and 17% decrease in the RSME) as compared to LME models without the site effect term. The optimized shape of LME models had a good agreement during both model-fitting (R² of 0.76) and cross-validation (R² of 0.6). Site-specific and seasonal performances of all types of models revealed that LME models had highest R² values over all monitoring stations and all seasons during the cross-validation. LME models had the best performance in May and March compared to other months during the model-fitting and cross-validation. However, LME models had a significant weakness in predicting extreme values of PM_2.5 during the cross-validation. Among all other types of models, GPM with the R² value of 0.59 and the RMSE of 10.2 μg/m³ had the best performance during the cross-validation.

CONCLUSIONS

While the best shape of LME and GPM had similar and reliable performances in predicting ground-level PM_2.5 values during the cross-validation, GPM was able to predict extreme values of ground-level PM_2.5 concentrations, which was the weakness of LME models and was an important issue in urban polluted environments. In this respect, GPM could be a good alternative for LME models for high levels of PM_2.5 concentrations. The spatial distribution of estimated PM_2.5 values represented that central parts of Tehran were the most polluted area over the studied region which was consistent with the ground-level recording PM_2.5 data over monitoring stations.

The association of prenatal exposure to particulate matter with infant growth: A birth cohort study in Beijing, China

Limited studies examined the associations of prenatal exposure to particulate matter (PM) and children's growth with inconsistent results, and no study focused on PM₁. We matched a birth cohort (10,547 children) with daily PM₁ and PM_2.5 concentrations by maternal home addresses. Air pollution concentrations were predicted by satellite remote sensing data, meteorological factors, and land use information. The weight and length of children in the birth cohort were measured at approximately one year old. We calculated the Z-score of weight for length (WFL) and body mass index (BMI) and then defined overweight and obesity (OWOB) based on WHO Standards. Generalized linear regression and modified Poisson regression were used to identify the association of prenatal exposure to PM₁ or PM_2.5 with anthropometric measurements and risk of OWOB. We also determined the mediation effect of preterm birth on the associations. Results showed that a 10 μg/m³ increase in prenatal exposure to PM₁ and PM_2.5 was significantly associated with a 0.105 [95% confidence interval (CI): 0.067, 0.144] and 0.063 (95% CI: 0.029, 0.097) increase in WFL Z-score for one-year-old children. Similar associations were found for BMI Z-score. A 10 μg/m³ increase in prenatal PM₁ and PM_2.5 exposure was significantly associated with 1.012 (95%CI: 1.003, 1.021) and 1.010 (95%CI: 1.002, 1.018) times higher risk of OWOB. . Preterm birth mediated 7.5% [direct effect (DE) = 0.106, P < 0.001; indirect effect (IE) = 0.009, P < 0.001)] and 9.9% (DE = 0.064, P < 0.001; IE = 0.007, P < 0.001) of the association between prenatal PM₁ and PM_2.5 exposure and WFL Z-score of the children. The association of prenatal PM₁ and PM_2.5 exposure with BMI Z-score of children was also mediated by preterm birth by 6.6% (DE = 0.111, P < 0.001; IE = 0.008, P < 0.001) and 9.1% (DE = 0.064, P < 0.001; IE = 0.006, P < 0.001). These results remained robust in the sensitivity analyses. In conclusion, prenatal exposure to PM₁ and PM_2.5 increased WFL, BMI Z-scores and higher risk of OWOB for one-year-old children. The associations were partially mediated by preterm birth. These findings call for the urgent action on air pollution regulation to protect early-life health among children.

Examining the status of improved air quality in world cities due to COVID-19 led temporary reduction in anthropogenic emissions

Clean air is a fundamental necessity for human health and well-being. Anthropogenic emissions that are harmful to human health have been reduced substantially under COVID-19 lockdown. Satellite remote sensing for air pollution assessments can be highly effective in public health research because of the possibility of estimating air pollution levels over large scales. In this study, we utilized both satellite and surface measurements to estimate air pollution levels in 20 cities across the world. Google Earth Engine (GEE) and Sentinel-5 Precursor TROPOspheric Monitoring Instrument (TROPOMI) application were used for both spatial and time-series assessment of tropospheric Nitrogen Dioxide (NO₂) and Carbon Monoxide (CO) statuses during the study period (1 February to May 11, 2019 and the corresponding period in 2020). We also measured Population-Weighted Average Concentration (PWAC) of particulate matter (PM_2.5 and PM₁₀) and NO₂ using gridded population data and in-situ air pollution estimates. We estimated the economic benefit of reduced anthropogenic emissions using two valuation approaches: (1) the median externality value coefficient approach, applied for satellite data, and (2) the public health burden approach, applied for in-situ data. Satellite data have shown that ~28 tons (sum of 20 cities) of NO₂ and ~184 tons (sum of 20 cities) of CO have been reduced during the study period. PM_2.5, PM₁₀, and NO₂ are reduced by ~37 (μg/m³), 62 (μg/m³), and 145 (μg/m³), respectively. A total of ~1310, ~401, and ~430 premature cause-specific deaths were estimated to be avoided with the reduction of NO₂, PM_2.5, and PM₁₀. The total economic benefits (Billion US$) (sum of 20 cities) of the avoided mortality are measured as ~10, ~3.1, and ~3.3 for NO₂, PM_2.5, and PM₁₀, respectively. In many cases, ground monitored data was found inadequate for detailed spatial assessment. This problem can be better addressed by incorporating satellite data into the evaluation if proper quality assurance is achieved, and the data processing burden can be alleviated or even removed. Both satellite and ground-based estimates suggest the positive effect of the limited human interference on the natural environments. Further research in this direction is needed to explore this synergistic association more explicitly.

Satellite-based ground PM2.5 estimation using a gradient boosting decision tree

Fine particulate matter with an aerodynamic diameter less than 2.5 μm (PM_2.5) is one of the major air pollutants risks to human health worldwide. Satellite-based aerosol optical depth (AOD) products are an effective metric for acquiring PM_2.5 information, featuring broad coverage and high resolution, which compensate for the sparse and uneven distribution of existing monitoring stations. In this study, a gradient boosting decision tree (GBDT) model for estimating ground PM_2.5 concentration directly from AOD products across China in 2017, integrating human activities and various natural variables was proposed. The GBDT model performed well in estimating temporal variability and spatial contrasts in daily PM_2.5 concentrations, with relatively high fitted model (10-fold cross-validation) coefficients of determination of 0.98 (0.81), low root mean square errors of 3.82 (11.57) μg/m³, and mean absolute error of 1.44 (7.45) μg/m³. Seasonal examinations revealed that summer had the cleanest air with the highest estimation accuracies, whereas winter had the most polluted air with the lowest estimation accuracies. The model successfully captured the PM_2.5 distribution pattern across China in 2017, showing high levels in southwest Xinjiang, the North China Plain, and the Sichuan Basin, especially in winter. Compared with other models, the GBDT model showed the highest performance in the estimation of PM_2.5 with a 3-km resolution. This algorithm can be adopted to improve the accuracy of PM_2.5 estimation with higher spatial resolution, especially in summer. In general, this study provided a potential method of improving the accuracy of satellite-based ground PM_2.5 estimation.

The comparison of AOD-based and non-AOD prediction models for daily PM2.5 estimation in Guangdong province, China with poor AOD coverage

The large amount of missing values has challenged the application of satellite-retrieved aerosol optical depth (AOD) in mapping surface PM_2.5 concentrations. In this study, we developed a non-AOD random forest model to estimate daily concentrations of PM_2.5 in Guangdong Province, China, where more than 80% of AOD data were missing. The predictive ability of the non-AOD model was compared with that of a AOD-based model. Daily ground-based measurements of PM_2.5 were obtained from 148 ground-monitoring sites in Guangdong with a 60 km rectangle buffer from January 2016 to December 2018. Daily MODIS MAIAC AOD were downloaded from NASA at a resolution of approximately 1 km. Two random forest models were developed to predict ground-level PM_2.5 with one included AOD as a predictor and the other one without AOD. The two random forest models were developed using the same dataset and their predictive abilities were compared. The results of 10-fold cross validation (CV) showed that the non-AOD and AOD-based random forest models yielded similar performance. The CV R² (RMSE) for the non-AOD model in 2016-2018 were 0.82 (8.4 μg/m³), 0.81 (9.5 μg/m³) and 0.78 (9.4 μg/m³), and those for AOD-based model were 0.83 (8.2 μg/m³), 0.82 (9.2 μg/m³) and 0.80 (9.0 μg/m³), respectively. Higher consistency of estimated PM_2.5 were observed in areas close to monitoring sites than those far away from sites, and in southern coastal than northern areas. As the non-AOD random forest model is not affected by AOD missingness, it can be used for epidemiological studies to estimate individual-level exposure to PM_2.5 at a high resolution without spatial or temporal gaps.

Impact of long-term exposure to ambient air pollution on the incidence of chronic obstructive pulmonary disease: A systematic review and meta-analysis

BACKGROUND

It is well known that air pollution causes respiratory morbidity and mortality by inducing airway inflammation. However, whether long-term exposure to air pollution is associated with increased incidence of chronic obstructive pulmonary disease (COPD) is controversial.

METHODS

We conducted a systematic review and meta-analysis with a random-effects model to calculate the pooled risk estimates of COPD development per 10 μg/m³ increase in individual air pollutants. PubMed, Embase, and Cochrane Library were searched from the date of their inception to August 2019 to identify long-term (at least three years of observation) prospective longitudinal studies that reported the risk of COPD development due to exposure to air pollutants. The air pollutants studied included particulate matter (PM_2.5 and PM₁₀) and nitrogen dioxide (NO₂).

RESULTS

Of the 436 studies identified, seven met our eligibility criteria. Among the seven studies, six, three, and five had data on PM_2.5, PM₁₀, and NO₂, respectively. The meta-analysis results showed that a 10 μg/m³ increase in PM_2.5 is associated with increased incidence of COPD (pooled HR 1.18, 95% CI 1.13-1.23). We also noted that a 10 μg/m³ increase in NO₂ is marginally associated with increased incidence of COPD (pooled HR 1.07, 95% CI 1.00-1.16). PM₁₀ seems to have no significant impact on the incidence of COPD (pooled HR 0.95, 95% CI 0.83-1.08), although the number of studies was too small. Meta-regression analysis found no significant effect modifiers.

CONCLUSIONS

Long-term exposure to PM_2.5 and NO₂ can be associated with increased incidence of COPD.

Estimating PM2.5 concentrations in Northeastern China with full spatiotemporal coverage, 2005-2016

Predicting long-term spatiotemporal characteristics of fine particulate matter (PM_2.5) is important in China to understand historical levels of PM_2.5, to support health effects research of both long-term and short-term exposures to PM_2.5, and to evaluate the efficacy of air pollution control policies. Satellite-retrieved aerosol optical depth (AOD) provides a unique opportunity to characterize the long-term trends of ground-level PM_2.5 at high spatial resolution. However, the missing rate of AOD in Northeastern China (NEC) is very high, especially in winter, and challenges the accuracy of long-term predictions of PM_2.5 if left unresolved. Using random forest algorithms, this study developed a gap-filling approach combing satellite AOD, meteorological data, land use parameters, population and visibility in the NEC during 2005-2016. The model, including all predictors, combined with a model without AOD was able to fill the gap of PM_2.5 predictions caused by missing AOD at 1-km resolution. The R² (RMSE) of the full-coverage predictions was 0.81 (18.5 μg/m³) at the daily level. Gap-filled PM_2.5 predictions on days with missing AOD reduced the relative prediction error from 28% to 2.5% in winter. The leave-one-year-out-cross-validation R² (RMSE) of the full-coverage predictions was 0.65 (16.3 μg/m³) at the monthly level, indicating relatively high accuracy of predicted historical PM_2.5 concentrations. Our results suggested that AOD helped increase the reliability of historical PM_2.5 prediction when ground PM_2.5 measurements were unavailable, even though predictions from the AOD model only accounted for approximate 37% of the whole dataset. Predicted PM_2.5 level in NEC have increased since 2005, reached its peak during 2013-2015, then saw a major decline in 2016. Our high-resolution predictions also showed a south to north gradient and many pollution hot spots in the city clusters surrounding provincial capitals, as well as within large cities. Overall, by combining predictions from the AOD model with higher accuracy and predictions from the non-AOD model to achieve full coverage, our modeling approach could produce long-term, full-coverage historical PM_2.5 levels in high-latitude areas in China, despite the widespread and persistent AOD missingness.

Identifying the spatiotemporal dynamic of PM2.5 concentrations at multiple scales using geographically and temporally weighted regression model across China during 2015-2018

Fine particulate matter (PM_2.5) is closely related to the air quality and public health. Numerous models have been introduced to simulate the PM_2.5 concentrations at large scale based on remote sensing and auxiliary data. However, the data precision provided by these models are inadequate for epidemiology and pollutant exposure studies at medium or small scale. The present study aims to calibrate PM_2.5 concentrations at 1 km resolution scale across China during 2015-2018 based on monitoring station data and auxiliary data using a novel geographically and temporally weighted regression model (GTWR). The cross-validation (CV) method and the geographically weighted regression (GWR) model are conducted for validation and cross-comparison. Additionally, the spatial autocorrelation and slope analysis methods are implemented to detect the spatiotemporal dynamic of PM_2.5 concentrations. A sample-based CV of the GTWR model demonstrates an acceptable precision with a coefficient of determination equal to 0.67, a root-mean-square error of 10.32 μg/m³, and a mean prediction error of-6.56 μg/m³. This result proves that the GTWR model can simulate PM_2.5 concentrations at a higher spatial resolution and accuracy across China than some previous models. Besides, the heterogeneity and spatiotemporal dynamic of PM_2.5 concentrations are obvious, that is, the High-High (H-H) agglomeration areas with strong haze pollution were mainly concentrated in Beijing-Tianjin-Hebei (BTH), Yangtze River Delta (YRD), Chengdu-Chongqing (CY), and Guanzhong Plain (GZP). In addition, the PM_2.5 concentrations are undergoing a decreasing trend in most of the study area, and the decrease in the BTH is dramatic. The results of the present study are helpful for calibrating and detecting the spatiotemporal dynamic of PM_2.5 concentrations and useful for the government to make decisions about decreasing haze pollution in urban agglomeration scale.

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.

Himawari-8 Aerosol Optical Depth (AOD) Retrieval Using a Deep Neural Network Trained Using AERONET Observations

Spectral aerosol optical depth (AOD) estimation from satellite-measured top of atmosphere (TOA) reflectances is challenging because of the complicated TOA-AOD relationship and a nexus of land surface and atmospheric state variations. This task is usually undertaken using a physical model to provide a first estimate of the TOA reflectances which are then optimized by comparison with the satellite data. Recently developed deep neural network (DNN) models provide a powerful tool to represent the complicated relationship statistically. This study presents a methodology based on DNN to estimate AOD using Himawari-8 Advanced Himawari Imager (AHI) TOA observations. A year (2017) of AHI TOA observations over the Himawari-8 full disk collocated in space and time with Aerosol Robotic Network (AERONET) AOD data were used to derive a total of 14,154 training and validation samples. The TOA reflectance in all six AHI solar bands, three TOA reflectance ratios derived based on the dark-target assumptions, sun-sensor geometry, and auxiliary data are used as predictors to estimate AOD at 500 nm. The DNN AOD is validated by separating training and validation samples using random k-fold cross-validation and using AERONET site-specific leave-one-station-out validation, and is compared with a random forest regression estimator and Japan Meteorological Agency (JMA) AOD. The DNN AOD shows high accuracy: (1) RMSE = 0.094, R2 = 0.915 for k-fold cross-validation, and (2) RMSE = 0.172, R2 = 0.730 for leave-one-station-out validation. The k-fold cross-validation overestimates the DNN accuracy as the training and validation samples may come from the same AHI pixel location. The leave-one-station-out validation reflects the accuracy for large-area applications where there are no training samples for the pixel location to be estimated. The DNN AOD has better accuracy than the random forest AOD and JMA AOD. In addition, the contribution of the dark-target derived TOA ratio predictors is examined and confirmed, and the sensitivity to the DNN structure is discussed.

Air Pollution Measurements and Land-Use Regression in Urban Sub-Saharan Africa Using Low-Cost Sensors—Possibilities and Pitfalls

Air pollution is recognized as the most important environmental factor that adversely affects human and societal wellbeing. Due to rapid urbanization, air pollution levels are increasing in the Sub-Saharan region, but there is a shortage of air pollution monitoring. Hence, exposure data to use as a base for exposure modelling and health effect assessments is also lacking. In this study, low-cost sensors were used to assess PM2.5 (particulate matter) levels in the city of Adama, Ethiopia. The measurements were conducted during two separate 1-week periods. The measurements were used to develop a land-use regression (LUR) model. The developed LUR model explained 33.4% of the variance in the concentrations of PM2.5. Two predictor variables were included in the final model, of which both were related to emissions from traffic sources. Some concern regarding influential observations remained in the final model. Long-term PM2.5 and wind direction data were obtained from the city’s meteorological station, which should be used to validate the representativeness of our sensor measurements. The PM2.5 long-term data were however not reliable. Means of obtaining good reference data combined with longer sensor measurements would be a good way forward to develop a stronger LUR model which, together with improved knowledge, can be applied towards improving the quality of health. A health impact assessment, based on the mean level of PM2.5 (23 µg/m3), presented the attributable burden of disease and showed the importance of addressing causes of these high ambient levels in the area.

Estimating PM2.5 concentrations in Yangtze River Delta region of China using random forest model and the Top-of-Atmosphere reflectance

Previous studies that have used remote sensing data to estimate the PM_2.5 concentrations mainly focused on the retrieval of aerosol optical depth (AOD) with moderate-to-low spatial resolution. However, the complex process of retrieving AOD from satellite Top-of-Atmosphere (TOA) reflectance always generates the missingness of AOD values due to the limitation of AOD retrieval algorithms. This study validated the possibility of using satellite TOA reflectance for estimating PM_2.5 concentrations, rather than using conventional AOD products retrieved from remote sensing imageries. Given that the TOA-PM_2.5 relationship cannot be accurately expressed by simple linear correlation, we developed a random forest model that integrated satellite TOA reflectance from Moderate Resolution Imaging Spectroradiometer (MODIS) Level 1B product, meteorological fields and land-use variables to estimate the ground-level PM_2.5 concentrations. The highly-polluted Yangtze River Delta (YRD) region of eastern China was employed as our study case. The results showed that our model performed well with a site-based and a time-based CV R² of 0.92 and 0.88, respectively. The derived annual and seasonal distributions of PM_2.5 concentrations exhibited high PM_2.5 values in northern YRD region (i.e., Jiangsu province) and relatively low values in southern region (i.e., Zhejiang province), which shared a similar distribution trend with the observed PM_2.5 concentrations. This study demonstrated the outstanding performance of random forest model using satellite TOA reflectance, and also provided an effective method for remotely sensed PM_2.5 estimation in regions where AOD retrievals are unavailable.

Spatiotemporal variability in long-term population exposure to PM2.5 and lung cancer mortality attributable to PM2.5 across the Yangtze River Delta (YRD) region over 2010-2016: A multistage approach

The Yangtze River Delta region (YRD) is one of the most densely populated regions in the world, and is frequently influenced by fine particulate matter (PM_2.5). Specifically, lung cancer mortality has been recognized as a major health burden associated with PM_2.5. Therefore, this study developed a multistage approach 1) to first create dasymetric population data with moderate resolution (1 km) by using a random forest algorithm, brightness reflectance of nighttime light (NTL) images, a digital elevation model (DEM), and a MODIS-derived normalized difference vegetation index (NDVI), and 2) to apply the improved population dataset with a MODIS-derived PM_2.5 dataset to estimate the association between spatiotemporal variability of long-term population exposure to PM_2.5 and lung cancer mortality attributable to PM_2.5 across YRD during 2010-2016 for microscale planning. The created dasymetric population data derived from a coarse census unit (administrative unit) were fairly matched with census data at a fine spatial scale (street block), with R² and RMSE of 0.64 and 27,874.5 persons, respectively. Furthermore, a significant urban-rural difference of population exposure was found. Additionally, population exposure in Shanghai was 2.9-8 times higher than the other major cities (7-year average: 192,000 μg·people/m³·km²). More importantly, the relative risks of lung cancer mortality in high-risk areas were 28%-33% higher than in low-risk areas. There were 12,574-14,504 total lung cancer deaths attributable to PM_2.5, and lung cancer deaths in each square kilometer of urban areas were 7-13 times higher than for rural areas. These results indicate that moderate-resolution information can help us understand the spatiotemporal variability of population exposure and related health risk in a high-density environment.

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.

Polycyclic aromatic hydrocarbons in the soils of the Yangtze River Delta Urban Agglomeration, China: Influence of land cover types and urbanization

With the development of urbanization, urban areas have become the main sources and sinks of polycyclic aromatic hydrocarbons (PAHs). The effects of human activities on the behaviors of PAHs in urban agglomerations have attracted significant attention. We collected soil samples (n = 330) to investigate the distribution, composition, and sources of 16 PAHs in the Yangtze River Delta Urban Agglomeration using the land resolution of 24 km × 24 km. The concentrations of Σ₁₆PAHs ranged from 21 to 2034 ng/g, with a median value of 124 ± 338 ng/g. The concentrations of PAHs were highest in impervious surfaces (350 ± 352 ng/g), followed by grassland (259 ± 322 ng/g), cropland (254 ± 341 ng/g), forest (190 ± 303 ng/g), and water (68 ± 34 ng/g). PAHs were dominated by medium-molecular-weight components (4 rings PAHs), followed by PAHs with high-molecular-weight (5-6 rings PAHs) and low-molecular-weight (2-3 rings PAHs) components. Fluoranthene, benzo[a]anthracene and chrysene are three major pollutants in YRDUA. A positive matrix factorization model indicated that fossil fuel combustion, coal combustion and volatilization, vehicle emission, and biomass burning were the main sources of PAHs, contributing 36%, 29%, 22%, and 12% of PAH sources, respectively. Urbanization parameters were positively correlated with PAH concentrations. A land use regression (LUR) model integrated with urbanization parameters showed evidence of the strong relationship between measured PAHs and predicted PAHs. These findings together highlighted that land cover types and human activities intensively influenced the PAHs pollution in the highly urbanized zones.

Spatiotemporal trends of PM2.5 concentrations in central China from 2003 to 2018 based on MAIAC-derived high-resolution data

Long-term PM_2.5 levels with high precision at fine spatiotemporal resolution are essential for quantitatively understanding the health risk of exposure to ambient fine particulate matter (PM_2.5) and making effective air pollution control policies. The emerging statistically derived PM_2.5 estimations from satellite remote sensing observations of aerosol optical depth (AOD) data are an effective alternative to reconstruct global, long-term, high spatiotemporal resolution PM_2.5 information. However, studies on PM_2.5 estimation and its application to exposure and health-related studies are limited in China due to the lack of historical in-situ measurements before 2013. In this study, we explored the long-term trends of PM_2.5 exposure in central China, a hotspot that has recently been experiencing severe particulate pollution, at the local scale. We first developed a spatiotemporal model incorporating periodical characteristics within the data to estimate daily concentrations of historical PM_2.5 at a fine scale of 1 km for 2003-2018. The linear effects of predictors including AOD, meteorological and land-use parameters and the non-linear interaction between AOD and meteorological parameters were considered in the modeling process. The most recently released high-resolution satellite aerosol product, Multi-Angle Implementation of Atmospheric Correction (MAIAC) was used to help to represent the fine-scale particle gradients. Our daily estimates correlated well with in-situ observations (cross-validation R² = 0.59), achieving precision comparable to previous statistical models. Through linking with gridded demographic data, the population-weighted PM_2.5 average during 2003 to 2018 was found to be high (62.23 μg/m³ for the whole domain) with obvious spatial variations and seasonality. An inverse U pattern was seen in the time series, with two inflection points around 2008 and 2015. Our model provides reliable particulate information with high spatial resolution and long-term temporal coverage, which can inform local-scale PM_2.5-related epidemiological studies and health-risk assessments for central China.

Satellite-Derived PM2.5 Composition and Its Differential Effect on Children’s Lung Function

Studies of the association between air pollution and children’s health typically rely on fixed-site monitors to determine exposures, which have spatial and temporal limitations. Satellite observations of aerosols provide the coverage that fixed-site monitors lack, enabling more refined exposure assessments. Using aerosol optical depth (AOD) data from the Multiangle Imaging SpectroRadiometer (MISR) instrument, we predicted fine particulate matter, PM 2.5 , and PM 2.5 speciation concentrations and linked them to the residential locations of 1206 children enrolled in the Southern California Children’s Health Study. We fitted mixed-effects models to examine the relationship between the MISR-derived exposure estimates and lung function, measured as forced expiratory volume in 1 second (FEV 1 ) and forced vital capacity (FVC), adjusting for study community and biological factors. Gradient Boosting and Support Vector Machines showed excellent predictive performance for PM 2.5 (test R 2 = 0.68 ) and its chemical components (test R 2 = –0.71). In single-pollutant models, FEV 1 decreased by 131 mL (95% CI: − 232 , − 35 ) per 10.7-µg/m 3 increase in PM 2.5 , by 158 mL (95% CI: − 273 , − 43 ) per 1.2-µg/m 3 in sulfates (SO 4 2 − ), and by 177 mL (95% CI: − 306 , − 56 ) per 1.6-µg/m 3 increase in dust; FVC decreased by 175 mL (95% CI: − 310 , − 29 ) per 1.2-µg/m 3 increase in SO 4 2 − and by 212 mL (95% CI: − 391 , − 28 ) per 2.5-µg/m 3 increase in nitrates (NO 3 − ). These results demonstrate that satellite observations can strengthen epidemiological studies investigating air pollution health effects by providing spatially and temporally resolved exposure estimates.

Predicting Fine Particulate Matter (PM2.5) in the Greater London Area: An Ensemble Approach using Machine Learning Methods

Estimating air pollution exposure has long been a challenge for environmental health researchers. Technological advances and novel machine learning methods have allowed us to increase the geographic range and accuracy of exposure models, making them a valuable tool in conducting health studies and identifying hotspots of pollution. Here, we have created a prediction model for daily PM2.5 levels in the Greater London area from 1st January 2005 to 31st December 2013 using an ensemble machine learning approach incorporating satellite aerosol optical depth (AOD), land use, and meteorological data. The predictions were made on a 1 km × 1 km scale over 3960 grid cells. The ensemble included predictions from three different machine learners: a random forest (RF), a gradient boosting machine (GBM), and a k-nearest neighbor (KNN) approach. Our ensemble model performed very well, with a ten-fold cross-validated R2 of 0.828. Of the three machine learners, the random forest outperformed the GBM and KNN. Our model was particularly adept at predicting day-to-day changes in PM2.5 levels with an out-of-sample temporal R2 of 0.882. However, its ability to predict spatial variability was weaker, with a R2 of 0.396. We believe this to be due to the smaller spatial variation in pollutant levels in this area.

Hourly PM2.5 Estimation over Central and Eastern China Based on Himawari-8 Data

In this study, an improved geographically and temporally weighted regression (IGTWR) model for the estimation of hourly PM2.5 concentration data was applied over central and eastern China in 2017, based on Himawari-8 Advanced Himawari Imager (AHI) data. A generalized distance based on the longitude, latitude, day, hour, and land use type was constructed. AHI aerosol optical depth, surface relative humidity, and boundary layer height (BLH) data were used as independent variables to retrieve the hourly PM2.5 concentrations at 1:00, 2:00, 3:00, 4:00, 5:00, 6:00, 7:00, and 8:00 UTC (Coordinated Universal Time). The model fitting and cross-validation performance were satisfactory. For the model fitting set, the correlation coefficient of determination (R2) between the measured and predicted PM2.5 concentrations was 0.886, and the root-mean-square error (RMSE) of 437,642 samples was only 12.18 µg/m3. The tenfold cross-validation results of the regression model were also acceptable; the correlation coefficient R2 of the measured and predicted results was 0.784, and the RMSE was 20.104 µg/m3, which is only 8 µg/m3 higher than that of the model fitting set. The spatial and temporal characteristics of the hourly PM2.5 concentration in 2017 were revealed. The model also achieved stable performance under haze and dust conditions.

Spatiotemporal Imputation of MAIAC AOD Using Deep Learning with Downscaling

Aerosols have adverse health effects and play a significant role in the climate as well. The Multiangle Implementation of Atmospheric Correction (MAIAC) provides Aerosol Optical Depth (AOD) at high temporal (daily) and spatial (1 km) resolution, making it particularly useful to infer and characterize spatiotemporal variability of aerosols at a fine spatial scale for exposure assessment and health studies. However, clouds and conditions of high surface reflectance result in a significant proportion of missing MAIAC AOD. To fill these gaps, we present an imputation approach using deep learning with downscaling. Using a baseline autoencoder, we leverage residual connections in deep neural networks to boost learning and parameter sharing to reduce overfitting, and conduct bagging to reduce error variance in the imputations. Downscaled through a similar auto-encoder based deep residual network, Modern-Era Retrospective analysis for Research and Applications Version 2 (MERRA-2) GMI Replay Simulation (M2GMI) data were introduced to the network as an important gap-filling feature that varies in space to be used for missingness imputations. Imputing weekly MAIAC AOD from 2000 to 2016 over California, a state with considerable geographic heterogeneity, our full (non-full) residual network achieved mean R2 = 0.94 (0.86) [RMSE = 0.007 (0.01)] in an independent test, showing considerably better performance than a regular neural network or non-linear generalized additive model (mean R2 = 0.78-0.81; mean RMSE = 0.013-0.015). The adjusted imputed as well as combined imputed and observed MAIAC AOD showed strong correlation with Aerosol Robotic Network (AERONET) AOD (R = 0.83; R2 = 0.69, RMSE = 0.04). Our results show that we can generate reliable imputations of missing AOD through a deep learning approach, having important downstream air quality modeling applications.

Novel Approaches to Air Pollution Exposure and Clinical Outcomes Assessment in Environmental Health Studies

An accurate assessment of pollutants’ exposure and precise evaluation of the clinical outcomes pose two major challenges to the contemporary environmental health research. The common methods for exposure assessment are based on residential addresses and are prone to many biases. Pollution levels are defined based on monitoring stations that are sparsely distributed and frequently distanced far from residential addresses. In addition, the degree of an association between outdoor and indoor air pollution levels is not fully elucidated, making the exposure assessment all the more inaccurate. Clinical outcomes’ assessment, on the other hand, mostly relies on the access to medical records from hospital admissions and outpatients’ visits in clinics. This method differentiates by health care seeking behavior and is therefore, problematic in evaluation of an onset, duration, and severity of an outcome. In the current paper, we review a number of novel solutions aimed to mitigate the aforementioned biases. First, a hybrid satellite-based modeling approach provides daily continuous spatiotemporal estimations with improved spatial resolution of 1 × 1 km2 and 200 × 200 m2 grid, and thus allows a more accurate exposure assessment. Utilizing low-cost air pollution sensors allowing a direct measurement of indoor air pollution levels can further validate these models. Furthermore, the real temporal-spatial activity can be assessed by GPS tracking devices within the individuals’ smartphones. A widespread use of smart devices can help with obtaining objective measurements of some of the clinical outcomes such as vital signs and glucose levels. Finally, human biomonitoring can be efficiently done at a population level, providing accurate estimates of in-vivo absorbed pollutants and allowing for the evaluation of body responses, by biomarkers examination. We suggest that the adoption of these novel methods will change the research paradigm heavily relying on ecological methodology and support development of the new clinical practices preventing adverse environmental effects on human health.

Satellite-based estimation of hourly PM2.5 levels during heavy winter pollution episodes in the Yangtze River Delta, China

In the developing countries such as China, most well-developed areas have suffered severe haze pollution, which was associated with increased premature morbidity and mortality and attracted widespread public concerns. Since ground-based PM_2.5 monitoring has limited temporal and spatial coverage, satellite aerosol remote sensing data has been increasingly applied to map large-scale PM_2.5 characteristics through advanced spatial statistical models. Although most existing research has taken advantage of the polar orbiting satellite instruments, a major limitation of the polar orbiting platform is its limited sampling frequency (e.g., 1-2 times/day), which is insufficient for capturing the PM_2.5 variability during short but intense heavy haze episodes. As the first attempt, we quantitatively investigated the feasibility of using the aerosol optical depth (AOD) data retrieved by the Geostationary Ocean Color Imager (GOCI) to estimate hourly PM_2.5 concentrations during winter haze episodes in the Yangtze River Delta (YRD). We developed a three-stage spatial statistical model, using GOCI AOD and fine mode fraction, as well as corresponding monitoring PM_2.5 concentrations, meteorological and land use data on a 6-km modeling grid with complete coverage in time and space. The 10-fold cross-validation R² was 0.72 with a regression slope of 1.01 between observed and predicted hourly PM_2.5 concentrations. After gap filling, the R² value for the three-stage model was 0.68. We further analyzed two representative large regional episodes, i.e., a "multi-process diffusion episode" during December 21-26, 2015 and a "Chinese New Year episode" during February 7-8, 2016. We concluded that AOD retrieved by geostationary satellites could serve as a new valuable data source for analyzing the heavy air pollution episodes.

Assessing Adverse Health Effects of Long-Term Exposure to Low Levels of Ambient Air Pollution: Phase 1

INTRODUCTION

This report provides a summary of major findings and key conclusions supported by a Health Effects Institute grant aimed at "Assessing Adverse Health Effects of Long-Term Exposure to Low Levels of Ambient Pollution." Our study was designed to advance four critical areas of inquiry and methods development.

METHODS

First, our work focused on predicting short- and long-term exposures to ambient PM2.5 mass (particulate matter ≤ 2.5μm in aerodynamic diameter) and ozone (O3) at high spatial resolution (1 km × 1 km) for the continental United States during the period 2000-2012 and linking these predictions to health data. Second, we developed new causal inference methods for exposure-response (ER) that account for exposure error and adjust for measured confounders. We applied these methods to data from the New England region. Third, we applied standard regression methods using Medicare claims data to estimate health effects that are associated with short- and long-term exposure to low levels of ambient air pollution. We conducted sensitivity analyses to assess potential confounding bias due to lack of extensive information on behavioral risk factors in the Medicare population using the Medicare Current Beneficiary Survey (MCBS) (nationally representative sample of approximately 15,000 Medicare enrollees per year), which includes abundant data on individual-level risk factors including smoking. Finally, we have begun developing tools for reproducible research - including approaches for data sharing, record linkage, and statistical software.

RESULTS

Our HEI-funded work has supported an extensive portfolio of analysis and the development of statistical methods that can be used to robustly understand the health effects of long- and short-term exposure to low levels of ambient air pollution. This report provides a high-level overview of statistical methods, data analysis, and key findings, as grouped into the following four areas: (1) Exposure assessment and data access; (2) Epidemiological studies of ambient exposures to air pollution at low levels; (3) Methodological contributions in causal inference; and (4) Open science research data platform.

CONCLUSION

Our body of work, advanced by HEI, lends extensive evidence that short- and long-term exposure to PM2.5 and O3 is harmful to human health, increasing the risks of hospitalization and death, even at levels that are well below the National Ambient Air Quality Standards (NAAQS).

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.

Cluster-based bagging of constrained mixed-effects models for high spatiotemporal resolution nitrogen oxides prediction over large regions

BACKGROUND

Accurate estimation of nitrogen dioxide (NO2) and nitrogen oxide (NOx) concentrations at high spatiotemporal resolutions is crucial for improving evaluation of their health effects, particularly with respect to short-term exposures and acute health outcomes. For estimation over large regions like California, high spatial density field campaign measurements can be combined with more sparse routine monitoring network measurements to capture spatiotemporal variability of NO2 and NOx concentrations. However, monitors in spatially dense field sampling are often highly clustered and their uneven distribution creates a challenge for such combined use. Furthermore, heterogeneities due to seasonal patterns of meteorology and source mixtures between sub-regions (e.g. southern vs. northern California) need to be addressed.

OBJECTIVES

In this study, we aim to develop highly accurate and adaptive machine learning models to predict high-resolution NO2 and NOx concentrations over large geographic regions using measurements from different sources that contain samples with heterogeneous spatiotemporal distributions and clustering patterns.

METHODS

We used a comprehensive Kruskal-K-means method to cluster the measurement samples from multiple heterogeneous sources. Spatiotemporal cluster-based bootstrap aggregating (bagging) of the base mixed-effects models was then applied, leveraging the clusters to obtain balanced and less correlated training samples for less bias and improvement in generalization. Further, we used the machine learning technique of grid search to find the optimal interaction of temporal basis functions and the scale of spatial effects, which, together with spatiotemporal covariates, adequately captured spatiotemporal variability in NO2 and NOx at the state and local levels.

RESULTS

We found an optimal combination of four temporal basis functions and 200 m scale spatial effects for the base mixed-effects models. With the cluster-based bagging of the base models, we obtained robust predictions with an ensemble cross validation R2 of 0.88 for both NO2 and NOx [RMSE (RMSEIQR): 3.62 ppb (0.28) and 9.63 ppb (0.37) respectively]. In independent tests of random sampling, our models achieved similarly strong performance (R2 of 0.87-0.90; RMSE of 3.97-9.69 ppb; RMSEIQR of 0.21-0.27), illustrating minimal over-fitting.

CONCLUSIONS

Our approach has important implications for fusing data from highly clustered and heterogeneous measurement samples from multiple data sources to produce highly accurate concentration estimates of air pollutants such as NO2 and NOx at high resolution over a large region.

Prenatal exposure to fine particulate matter, maternal hemoglobin concentration, and fetal growth during early pregnancy: associations and mediation effects analysis

BACKGROUND

Fetal essential organ development is completed during early pregnancy which is important for fetal and postnatal health. However, the effect of exposure to PM_2.5 on fetal growth during early pregnancy is less studied and the related mechanisms are largely unknown.

METHODS

We conducted a birth cohort study of 1945 pregnant women with measurement of the fetal crown to rump length (CRL) by ultrasound between the gestational age of 11 and 14 weeks. We estimated residential exposures of PM_2.5 from the date of LMP to the date of ultrasound examination using a spatial-temporal land use regression model. Maternal hemoglobin concentration was examined by maternal blood samples during the same gestational period or ±1 week of the ultrasound examination. The associations of exposure to PM_2.5 with maternal hemoglobin concentration, and exposure to PM_2.5 with fetal CRL during early pregnancy were estimated by multiple linear regression models. The mediation effect of maternal hemoglobin concentration on the association between exposure to PM_2.5 and fetal CRL was explored by a casual mediation analysis.

RESULTS

One IQR increment of prenatal exposure to PM_2.5 was associated with a 0.929 g/L (95% CI: 0.068, 1.789) increase in maternal hemoglobin concentration, and associated with a -0.082 cm (95% CI: 0.139, -0.025) decrease in fetal CRL. One g/L increment of maternal hemoglobin concentration was associated a -0.011 cm (95% CI: 0.014, -0.008) decrease in fetal CRL. The mediation analysis indicated that 12.1% of the total effect of prenatal exposure to PM_2.5 on reducing fetal CRL was mediated by increased maternal hemoglobin concentration.

CONCLUSION

Exposure to PM_2.5 was associated with reduced fetal growth during early pregnancy and elevated maternal hemoglobin concentration mediated this association.

Assessing exposure to outdoor air pollution for epidemiological studies: Model-based and personal sampling strategies

Epidemiologic studies have found air pollution to be causally linked to respiratory health including the exacerbation and development of childhood asthma. Accurately characterizing exposure is paramount in these studies to ensure valid estimates of health effects. Here, we provide a brief overview of the evolution of air pollution exposure assessment ranging from the use of ground-based, single-site air monitoring stations for population-level estimates to recent advances in spatiotemporal models, which use advanced machine learning algorithms and satellite-based data to accurately estimate individual-level daily exposures at high spatial resolutions. In addition, we review recent advances in sensor technology that enable the use of personal monitoring in epidemiologic studies, long-considered the "holy grail" of air pollution exposure assessment. Finally, we highlight key advantages and uses of each approach including the generalizability and public health relevance of air pollution models and the accuracy of personal monitors that are useful to guide personalized prevention strategies. Investigators and clinicians interested in the effects of air pollution on allergic disease and asthma should carefully consider the pros and cons of each approach to guide their application in research and practice.

Satellite-Based Estimation of Daily Ground-Level PM2.5 Concentrations over Urban Agglomeration of Chengdu Plain

Monitoring particulate matter with aerodynamic diameters of less than 2.5 μm (PM2.5) is of great importance to assess its adverse effects on human health, especially densely populated regions. In this paper, an improved linear mixed effect model (LMEM) was developed. The model introduced meteorological variable, column water vapor (CWV), which has as the same resolution as satellite-derived aerosol optical thickness (AOT), to enhance PM2.5 estimation accuracy by considering spatiotemporal consistency of CWV and AOT. The model was implemented to urban agglomeration of Chengdu Plain during 2015. The results show that model accuracy has been improved significantly compared to linear regression model (R2 = 0.49), with R2 of 0.81 and root mean squared prediction error (RMSPE) of 15.47 μg/m3, mean prediction error (MPE) of 11.09 μg/m3, and effectively revealed the characteristics of spatiotemporal variations PM2.5 level across the study area: The PM2.5 level is higher in the central and southern areas with dense population, while it is lower in the northwest and southwest mountain areas; and the PM2.5 level is higher during autumn and winter, while it is lower during spring and summer. The product data in this paper are valuable for local government pollution monitoring, public health research, and urban air quality control.

Estimation of daily PM10 and PM2.5 concentrations in Italy, 2013-2015, using a spatiotemporal land-use random-forest model

Particulate matter (PM) air pollution is one of the major causes of death worldwide, with demonstrated adverse effects from both short-term and long-term exposure. Most of the epidemiological studies have been conducted in cities because of the lack of reliable spatiotemporal estimates of particles exposure in nonurban settings. The objective of this study is to estimate daily PM₁₀ (PM < 10 μm), fine (PM < 2.5 μm, PM_2.5) and coarse particles (PM between 2.5 and 10 μm, PM_2.5-10) at 1-km² grid for 2013-2015 using a machine learning approach, the Random Forest (RF). Separate RF models were defined to: predict PM_2.5 and PM_2.5-10 concentrations in monitors where only PM₁₀ data were available (stage 1); impute missing satellite Aerosol Optical Depth (AOD) data using estimates from atmospheric ensemble models (stage 2); establish a relationship between measured PM and satellite, land use and meteorological parameters (stage 3); predict stage 3 model over each 1-km² grid cell of Italy (stage 4); and improve stage 3 predictions by using small-scale predictors computed at the monitor locations or within a small buffer (stage 5). Our models were able to capture most of PM variability, with mean cross-validation (CV) R² of 0.75 and 0.80 (stage 3) and 0.84 and 0.86 (stage 5) for PM₁₀ and PM_2.5, respectively. Model fitting was less optimal for PM_2.5-10, in summer months and in southern Italy. Finally, predictions were equally good in capturing annual and daily PM variability, therefore they can be used as reliable exposure estimates for investigating long-term and short-term health effects.

Quantifying population exposure to air pollution using individual mobility patterns inferred from mobile phone data

A critical question in environmental epidemiology is whether air pollution exposures of large populations can be refined using individual mobile-device-based mobility patterns. Cellular network data has become an essential tool for understanding the movements of human populations. As such, through inferring the daily home and work locations of 407,435 mobile phone users whose positions are determined, we assess exposure to PM_2.5. Spatiotemporal PM_2.5 concentrations are predicted using an Aerosol Optical Depth- and Land Use Regression-combined model. Air pollution exposures of subjects are assigned considering modeled PM_2.5 levels at both their home and work locations. These exposures are then compared to residence-only exposure metric, which does not consider daily mobility. In our study, we demonstrate that individual air pollution exposures can be quantified using mobile device data, for populations of unprecedented size. In examining mean annual PM_2.5 exposures determined, bias for the residence-based exposures was 0.91, relative to the exposure metric considering the work location. Thus, we find that ignoring daily mobility potentially contributes to misclassification in health effect estimates. Our framework for understanding population exposure to environmental pollution could play a key role in prospective environmental epidemiological studies.