Search

Short-term mobile monitoring campaigns are increasingly used to assess long-term air pollution exposure in epidemiology. Little is known about how monitoring network design features, including the number of stops and sampling temporality, impacts exposure assessment models. We address this gap by leveraging an extensive mobile monitoring campaign conducted in the greater Seattle area over the course of a year during all days of the week and most hours. The campaign measured total particle number concentration (PNC; sheds light on ultrafine particulate (UFP) number concentration), black carbon (BC), nitrogen dioxide (NO

Based on human and animal experimental studies, exposure to ambient carbon monoxide (CO) may be associated with cardiovascular disease outcomes, but epidemiological evidence of this link is limited. The number and distribution of ground-level regulatory agency monitors are insufficient to characterize fine-scale variations in CO concentrations.To develop a daily, high-resolution ambient CO exposure prediction model at the city scale.We developed a CO prediction model in Baltimore, Maryland, based on a spatiotemporal statistical algorithm with regulatory agency monitoring data and measurements from calibrated low-cost gas monitors. We also evaluated the contribution of three novel parameters to model performance: high-resolution meteorological data, satellite remote sensing data, and copollutant (The CO model had spatial cross-validation (CV)The three novel parameters did not substantially improve model performance, suggesting that, on its own, our spatiotemporal modeling framework based on geographic features was reliable and robust. As low-cost air monitors become increasingly available, this approach to CO concentration modeling can be generalized to resource-restricted environments to facilitate comprehensive epidemiological research. https://doi.org/10.1289/EHP10889.

Background Air pollution health studies have been increasingly using prediction models for exposure assessment even in areas without monitoring stations. To date, most studies have assumed that a single exposure model is correct, but estimated effects may be sensitive to the choice of exposure model. Methods We obtained county-level daily cardiovascular (CVD) admissions from the New York (NY) Statewide Planning and Resources Cooperative System (SPARCS) and four sets of fine particulate matter (PM2.5) spatio-temporal predictions (2002–2012). We employed overdispersed Poisson models to investigate the relationship between daily PM2.5 and CVD, adjusting for potential confounders, separately for each state-wide PM2.5 dataset. Results For all PM2.5 datasets, we observed positive associations between PM2.5 and CVD. Across the modeled exposure estimates, effect estimates ranged from 0.23% (95%CI: -0.06, 0.53%) to 0.88% (95%CI: 0.68, 1.08%) per 10 µg/m3 increase in daily PM2.5. We observed the highest estimates using monitored concentrations 0.96% (95%CI: 0.62, 1.30%) for the subset of counties where these data were available. Conclusions Effect estimates varied by a factor of almost four across methods to model exposures, likely due to varying degrees of exposure measurement error. Nonetheless, we observed a consistently harmful association between PM2.5 and CVD admissions, regardless of model choice.

Current Instance Transfer Learning (ITL) methodologies use domain adaptation and sub-space transformation to achieve successful transfer learning. However, these methodologies, in their processes, sometimes overfit on the target dataset or suffer from negative transfer if the test dataset has a high variance. Boosting methodologies have been shown to reduce the risk of overfitting by iteratively re-weighing instances with high-residual. However, this balance is usually achieved with parameter optimization, as well as reducing the skewness in weights produced due to the size of the source dataset. While the former can be achieved, the latter is more challenging and can lead to negative transfer. We introduce a simpler and more robust fix to this problem by building upon the popular boosting ITL regression methodology, two-stage TrAdaBoost.R2. Our methodology, S-TrAdaBoost.R2, is a boosting and random-forest based ensemble methodology that utilizes importance sampling to reduce the skewness due to the source dataset. We show that S-TrAdaBoost.R2 performs better than competitive transfer learning methodologies 63% of the time. It also displays consistency in its performance over diverse datasets with varying complexities, as opposed to the sporadic results observed for other transfer learning methodologies.

Low-cost air quality sensors are promising supplements to regulatory monitors for PM2.5 exposure assessment. However, little has been done to incorporate the low-cost sensor measurements in large-scale PM2.5 exposure modeling. We conducted spatially varying calibration and developed a downweighting strategy to optimize the use of low-cost sensor data in PM2.5 estimation. In California, PurpleAir low-cost sensors were paired with air quality system (AQS) regulatory stations, and calibration of the sensors was performed by geographically weighted regression. The calibrated PurpleAir measurements were then given lower weights according to their residual errors and fused with AQS measurements into a random forest model to generate 1 km daily PM2.5 estimates. The calibration reduced PurpleAir's systematic bias to ∼0 μg/m3 and residual errors by 36%. Increased sensor bias was found to be associated with higher temperature and humidity, as well as longer operating time. The weighted prediction model outperformed the AQS-based prediction model with an improved random cross-validation (CV) R2 of 0.86, an improved spatial CV R2 of 0.81, and a lower prediction error. The temporal CV R2 did not improve due to the temporal discontinuity of PurpleAir. The inclusion of PurpleAir data allowed the predictions to better reflect PM2.5 spatial details and hotspots.

Although short-term ozone (The present study aimed to estimate the spatiotemporal distribution of monthly mean daily maximum 8-h averageWe developed a machine learning model with a satellite-derived boundary-layerThe random, spatial, and temporal cross-validationThe present model had enhanced performance in predicting ground-level

High-resolution, high-quality exposure modeling is critical for assessing the health effects of ambient PM2.5 in epidemiological studies. Using sparse regulatory PM2.5 measurements as principal model inputs may result in two issues in exposure prediction: (1) they may affect the models’ accuracy in predicting PM2.5 spatial distribution; (2) the internal validation based on these measurements may not reliably reflect the model performance at locations of interest (e.g., a cohort’s residential locations). In this study, we used the PM2.5 measurements from a publicly available commercial low-cost PM2.5 network, PurpleAir, with an external validation dataset at the residential locations of a representative sample of participants from the Adult Changes in Thought - Air Pollution (ACT-AP) study, to improve the accuracy of exposure prediction at the cohort participant locations. We also proposed a metric based on principal component analysis (PCA) - the PCA distance - to assess the similarity between monitor and cohort locations to guide monitor deployment and data selection. The analysis was based on a spatiotemporal modeling framework with 51 “gold-standard” monitors and 58 PurpleAir monitors for model development, as well as 105 home monitors at the cohort locations for model validation, in the Puget Sound region of Washington State from June 2017 to March 2019. After including calibrated PurpleAir measurements as part of the dependent variable, the external spatiotemporal validation R2 and root-mean-square error, RMSE, for two-week concentration averages improved from 0.84 and 2.22 μg/m3 to 0.92 and 1.63 μg/m3, respectively. The external spatial validation R2 and RMSE for long-term averages over the modeling period improved from 0.72 and 1.01 μg/m3 to 0.79 and 0.88 μg/m3, respectively. The exposure predictions incorporating PurpleAir measurements demonstrated sharper urban-suburban concentration gradients. The PurpleAir monitors with shorter PCA distances improved the model’s prediction accuracy more substantially than the monitors with longer PCA distances, supporting the use of this similarity metric.

Exposure to fine particulate matter (PM(2.5)) has been linked to a substantial disease burden globally, yet little has been done to estimate the population health risks of PM(2.5) in South Africa due to the lack of high-resolution PM(2.5) exposure estimates. We developed a random forest model to estimate daily PM(2.5) concentrations at 1 km(2) resolution in and around industrialized Gauteng Province, South Africa, by combining satellite aerosol optical depth (AOD), meteorology, land use, and socioeconomic data. We then compared PM(2.5) concentrations in the study domain before and after the implementation of the new national air quality standards. We aimed to test whether machine learning models are suitable for regions with sparse ground observations such as South Africa and which predictors played important roles in PM(2.5) modeling. The cross-validation R(2) and Root Mean Square Error of our model was 0.80 and 9.40 μg/m(3), respectively. Satellite AOD, seasonal indicator, total precipitation, and population were among the most important predictors. Model-estimated PM(2.5) levels successfully captured the temporal pattern recorded by ground observations. Spatially, the highest annual PM(2.5) concentration appeared in central and northern Gauteng, including northern Johannesburg and the city of Tshwane. Since the 2016 changes in national PM(2.5) standards, PM(2.5) concentrations have decreased in most of our study region, although levels in Johannesburg and its surrounding areas have remained relatively constant. This is anadvanced PM(2.5) model for South Africa with high prediction accuracy at the daily level and at a relatively high spatial resolution. Our study provided a reference for predictor selection, and our results can be used for a variety of purposes, including epidemiological research, burden of disease assessments, and policy evaluation.

BACKGROUND AND AIM: Air quality management is increasingly focused not only on across-the-board reductions in ambient pollution concentrations, but also on identifying and remediating elevated expo...

BACKGROUND AND AIM: High-resolution, high-quality exposure modeling is critical for assessing the health effects of PM2.5 in epidemiological cohorts. Sparse ground-level PM2.5 measurements, as key ...

High-resolution, high-quality exposure modeling is critical for assessing the health effects of ambient PM(2.5) in epidemiological studies. Using sparse regulatory PM(2.5) measurements as principal model inputs may result in two issues in exposure prediction: (1) they may affect the models’ accuracy in predicting PM(2.5) spatial distribution; (2) the internal validation based on these measurements may not reliably reflect the model performance at locations of interest (e.g., a cohort’s residential locations). In this study, we used the PM(2.5) measurements from a publicly available commercial low-cost PM(2.5) network, PurpleAir, with an external validation dataset at the residential locations of a representative sample of participants from the Adult Changes in Thought - Air Pollution (ACT-AP) study, to improve the accuracy of exposure prediction at the cohort participant locations. We also proposed a metric based on principal component analysis (PCA) - the PCA distance - to assess the similarity between monitor and cohort locations to guide monitor deployment and data selection. The analysis was based on a spatiotemporal modeling framework with 51 “gold-standard” monitors and 58 PurpleAir monitors for model development, as well as 105 home monitors at the cohort locations for model validation, in the Puget Sound region of Washington State from June 2017 to March 2019. After including calibrated PurpleAir measurements as part of the dependent variable, the external spatiotemporal validation R(2) and root-mean-square error, RMSE, for two-week concentration averages improved from 0.84 and 2.22 μg/m(3) to 0.92 and 1.63 μg/m(3), respectively. The external spatial validation R(2) and RMSE for long-term averages over the modeling period improved from 0.72 and 1.01 μg/m(3) to 0.79 and 0.88 μg/m(3), respectively. The exposure predictions incorporating PurpleAir measurements demonstrated sharper urban-suburban concentration gradients. The PurpleAir monitors with shorter PCA distances improved the model’s prediction accuracy more substantially than the monitors with longer PCA distances, supporting the use of this similarity metric.

Estimating ground-level ozone concentrations is crucial to study the adverse health effects of ozone exposure and better understand the impacts of ground-level ozone on biodiversity and vegetation. However, few studies have attempted to use satellite retrieved ozone as an indicator given their low sensitivity in the boundary layer. Using the Troposphere Monitoring Instrument (TROPOMI)'s total ozone column together with the ozone profile information retrieved by the Ozone Monitoring Instrument (OMI), as TROPOMI ozone profile product has not been released, we developed a machine learning model to estimate daily maximum 8-hour average ground-level ozone concentration at 10 km spatial resolution in California. In addition to satellite parameters, we included meteorological fields from the High-Resolution Rapid Refresh (HRRR) system at 3 km resolution and land-use information as predictors. Our model achieved an overall 10-fold cross-validation (CV) R 2 of 0.84 with root mean square error (RMSE) of 0.0059 ppm, indicating a good agreement between model predictions and observations. Model predictions showed that the suburb of Los Angeles Metropolitan area had the highest ozone levels, while the Bay Area and the Pacific coast had the lowest. High ozone levels are also seen in Southern California and along the east side of the Central Valley. TROPOMI data improved the estimate of extreme values when compared to a similar model without it. Our study demonstrates the feasibility and value of using TROPOMI data in the spatiotemporal characterization of ground-level ozone concentration.

Supplemental Digital Content is available in the text., Background: Toxicological evidence has shown that fine particulate matter (PM2.5) may affect distant organs, including kidneys, over the short term. However, epidemiological evidence is limited. Objectives: We investigated associations between short-term exposure to PM2.5, major PM2.5 components [elemental carbon (EC), organic carbon (OC), sulfate, and nitrate], and gaseous co-pollutants (O3, CO, SO2, NO2, and NOx) and emergency department (ED) visits for kidney diseases during 2002–2008 in Atlanta, Georgia. Methods: Log-linear time-series models were fitted to estimate the acute effects of air pollution, with single-day and unconstrained distributed lags, on rates of ED visits for kidney diseases [all renal diseases and acute renal failure (ARF)], controlling for meteorology (maximum air and dew-point temperatures) and time (season, day of week, holidays, and long-term time trend). Results: For all renal diseases, we observed positive associations for most air pollutants, particularly 8-day cumulative exposure to OC [rate ratio (RR) = 1.018, (95% confidence interval [CI]: 1.003, 1.034)] and EC [1.016 (1.000, 1.031)] per interquartile range increase exposure. For ARF, we observed positive associations particularly for 8-day exposure to OC [1.034 (1.005, 1.064)], EC [1.032 (1.002, 1.063)], nitrate [1.032 (0.996, 1.069)], and PM2.5 [1.026 (0.997, 1.057)] per interquartile range increase exposure. We also observed positive associations for most criteria gases. The RR estimates were generally higher for ARF than all renal diseases. Conclusions: We observed positive associations between short-term exposure to fine particulate air pollution and kidney disease outcomes. This study adds to the growing epidemiological evidence that fine particles may impact distant organs (e.g., kidneys) over the short term.

Background Air pollution health studies have been increasingly using prediction models for exposure assessment even in areas without monitoring stations. To date, most studies have assumed that a single exposure model is correct, but estimated effects may be sensitive to the choice of exposure model. Methods We obtained county-level daily cardiovascular (CVD) admissions from the New York (NY) Statewide Planning and Resources Cooperative System (SPARCS) and four sets of fine particulate matter (PM2.5) spatio-temporal predictions (2002–2012). We employed overdispersed Poisson models to investigate the relationship between daily PM2.5 and CVD, adjusting for potential confounders, separately for each state-wide PM2.5 dataset. Results For all PM2.5 datasets, we observed positive associations between PM2.5 and CVD. Across the modeled exposure estimates, effect estimates ranged from 0.23% (95%CI: -0.06, 0.53%) to 0.88% (95%CI: 0.68, 1.08%) per 10 µg/m3 increase in daily PM2.5. We observed the highest estimates using monitored concentrations 0.96% (95%CI: 0.62, 1.30%) for the subset of counties where these data were available. Conclusions Effect estimates varied by a factor of almost four across methods to model exposures, likely due to varying degrees of exposure measurement error. Nonetheless, we observed a consistently harmful association between PM2.5 and CVD admissions, regardless of model choice. Supplementary Information The online version contains supplementary material available at 10.1186/s12940-021-00782-3.

Epidemiological research on the adverse health outcomes due to PM

Quantification of PM

BACKGROUND: Emissions control programs targeting certain air pollution sources may alter PM(2.5) composition, as well as the risk of adverse health outcomes associated with PM(2.5). OBJECTIVES: We examined temporal changes in the risk of emergency department (ED) visits for cardiovascular diseases (CVDs) and asthma associated with short-term increases in ambient PM(2.5) concentrations in Los Angeles, California. METHODS: Poisson log-linear models with unconstrained distributed exposure lags were used to estimate the risk of CVD and asthma ED visits associated with short-term increases in daily PM(2.5) concentrations, controlling for temporal and meteorological confounders. The models were run separately for three predefined time periods, which were selected based on the implementation of multiple emissions control programs (EARLY: 2005-2008; MIDDLE: 2009-2012; LATE: 2013-2016). Two-pollutant models with individual PM(2.5) components and the remaining PM(2.5) mass were also considered to assess the influence of changes in PM(2.5) composition on changes in the risk of CVD and asthma ED visits associated with PM(2.5) over time. RESULTS: The relative risk of CVD ED visits associated with a 10 μg/m(3) increase in 4-day PM(2.5) concentration (lag 0-3) was higher in the LATE period (rate ratio = 1.020, 95% confidence interval = [1.010, 1.030]) compared to the EARLY period (1.003, [0.996, 1.010]). In contrast, for asthma, relative risk estimates were largest in the EARLY period (1.018, [1.006, 1.029]), but smaller in the following periods. Similar temporal differences in relative risk estimates for CVD and asthma were observed among different age and season groups. No single component was identified as an obvious contributor to the changing risk estimates over time, and some components exhibited different temporal patterns in risk estimates from PM(2.5) total mass, such as a decreased risk of CVD ED visits associated with sulfate over time. CONCLUSIONS: Temporal changes in the risk of CVD and asthma ED visits associated with short-term increases in ambient PM(2.5) concentrations were observed. These changes could be related to changes in PM(2.5) composition (e.g., an increasing fraction of organic carbon and a decreasing fraction of sulfate in PM(2.5)). Other factors such as improvements in healthcare and differential exposure misclassification might also contribute to the changes.

Low-cost air quality sensors are promising supplements to regulatory monitors for PM

Exposure to fine particulate matter (PM2.5) has been linked to a substantial disease burden globally, yet little has been done to estimate the population health risks of PM2.5 in South Africa due to the lack of high-resolution PM2.5 exposure estimates. We developed a random forest model to estimate daily PM2.5 concentrations at 1 km2 resolution in and around industrialized Gauteng Province, South Africa, by combining satellite aerosol optical depth (AOD), meteorology, land use, and socioeconomic data. We then compared PM2.5 concentrations in the study domain before and after the implementation of the new national air quality standards. We aimed to test whether machine learning models are suitable for regions with sparse ground observations such as South Africa and which predictors played important roles in PM2.5 modeling. The cross-validation R2 and Root Mean Square Error of our model was 0.80 and 9.40 μg/m3, respectively. Satellite AOD, seasonal indicator, total precipitation, and population were among the most important predictors. Model-estimated PM2.5 levels successfully captured the temporal pattern recorded by ground observations. Spatially, the highest annual PM2.5 concentration appeared in central and northern Gauteng, including northern Johannesburg and the city of Tshwane. Since the 2016 changes in national PM2.5 standards, PM2.5 concentrations have decreased in most of our study region, although levels in Johannesburg and its surrounding areas have remained relatively constant. This is anadvanced PM2.5 model for South Africa with high prediction accuracy at the daily level and at a relatively high spatial resolution. Our study provided a reference for predictor selection, and our results can be used for a variety of purposes, including epidemiological research, burden of disease assessments, and policy evaluation.

Emissions in many sources are estimated in municipal district totals and spatially disaggregated onto grid cells using empirically selected spatial proxies such as population density, which might introduce biases, especially in fine spatial scale. Efforts have been made to improve the spatial representation of emission inventory, by incorporating comprehensive point source database (e.g. power plants, industrial facilities) in emission estimates. Satellite-based observations from the TROPOspheric Monitoring Instrument (TROPOMI) with unprecedented pixel sizes (3.5 × 7 km2) and signal-to-noise ratios offer the opportunity to evaluate the spatial accuracy of such highly resolved emissions from space. Here, we compare the city-level NO x emissions from a proxy-based emission inventory named the Multi-resolution Emission Inventory for China (MEIC) with a highly resolved emission inventory named the Multi-resolution Emission Inventory for China - High Resolution (MEIC-HR) that has nearly 100 000 industrial facilities, and evaluate them through NO x emissions derived from the TROPOMI NO2 tropospheric vertical column densities (TVCDs). We find that the discrepancies in city-level NO x emissions between MEIC and MEIC-HR are influenced by the proportions of emissions from point sources and NO x emissions per industrial gross domestic product (IGDP). The use of IGDP as a spatial proxy to disaggregate industrial emissions tends to overestimate NO x emissions in cities with lower industrial emission intensities or less industrial facilities in the MEIC. The NO x emissions of 70 cities are derived from one year TROPOMI NO2 TVCDs using the exponentially modified Gaussian function. Compared to the satellite-derived emissions, the cities with higher industrial point source emission proportions in MEIC-HR agree better with space-constrained results, indicating that integrating more point sources in the inventory would improve the spatial accuracy of emissions on city scale. In the future, we should devote more efforts to incorporating accurate locations of emitting facilities to reduce uncertainties in fine-scale emission estimates and guide future policies.

PM2.5 hourly average measurements from 33 outdoor PurpleAir particle monitors were compared with hourly measurements from 27 nearby US EPA Air Quality System (AQS) stations employing Federal Equivalent Method (FEM) monitors in California over an 18-month (77-week) period. A transparent and reproducible alternative method (ALT) of calculating PM2.5 from the particle numbers in three size categories was used in place of the estimates provided by Plantower, the manufacturer of the sensors used in PurpleAir monitors. The ALT method was superior in several ways (better precision, lower limit of detection, improved size distribution) compared to Plantower's CF1 or ATM data series. PurpleAir monitors were strongly correlated with the nearby US EPA Air Quality System AQS stations. A calibration factor (CF) ranging between 2.9 and 3.1 was empirically derived for the PurpleAir estimates using the ALT method. This value was based on comparing the average value of 177,329 PurpleAir measurements to the value calculated from the FEM stations. The monitoring period included about 13 weeks showing very high outdoor values due to several major fires covering several hundred thousand acres. The CF during these 13 weeks averaged 2.39, whereas the CF for the remaining 64 weeks averaged about 3.21, suggesting a different response to the smoke from wildfires compared with normal ambient fine particulate matter (PM2.5). The standard Plantower CF1 data series overestimated the FEM values by about 40%, in agreement with several other studies.

Regulatory monitoring networks are often too sparse to support community-scale PM(2.5) exposure assessment while emerging low-cost sensors have the potential to fill in the gaps. To date, limited studies, if any, have been conducted to utilize low-cost sensor measurements to improve PM(2.5) prediction with high spatiotemporal resolutions based on statistical models. Imperial County in California is an exemplary region with sparse Air Quality System (AQS) monitors and a community-operated low-cost network entitled Identifying Violations Affecting Neighborhoods (IVAN). This study aims to evaluate the contribution of IVAN measurements to the quality of PM(2.5) prediction. We adopted the Random Forest algorithm to estimate daily PM(2.5) concentrations at 1-km spatial resolution using three different PM(2.5) datasets (AQS-only, IVAN-only, and AQS/IVAN combined). The results showed that the integration of low-cost sensor measurements is an effective way to significantly improve the quality of PM(2.5) prediction with an increase of cross-validation (CV) R(2) by ~0.2. The IVAN measurements also contributed to the increased importance of emission source-related covariates and more reasonable spatial patterns of PM(2.5). The remaining uncertainty in the calibrated IVAN measurements could still cause apparent outliers in the prediction model, highlighting the need for more effective calibration or integration methods to relieve its negative impact.

Previous PM

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

Epidemiological research on the adverse health outcomes due to PM2.5 exposure frequently relies on measurements from regulatory air quality monitors to provide ambient exposure estimates, whereas personal PM2.5 exposure may deviate from ambient concentrations due to outdoor infiltration and contributions from indoor sources. Research in quantifying infiltration factors (Finf), the fraction of outdoor PM2.5 that infiltrates indoors, has been historically limited in space and time due to the high costs of monitor deployment and maintenance. Recently, the growth of openly accessible, citizen-based PM2.5 measurements provides an unprecedented opportunity to characterize Finf at large spatiotemporal scales. In this analysis, 91 consumer-grade PurpleAir indoor/outdoor monitor pairs were identified in California (41 residential houses and 50 public/commercial buildings) during a 20-month period with around 650000 h of paired PM2.5 measurements. An empirical method was developed based on local polynomial regression to estimate site-specific Finf. The estimated site-specific Finf had a mean of 0.26 (25th, 75th percentiles: [0.15, 0.34]) with a mean bootstrap standard deviation of 0.04. The Finf estimates were toward the lower end of those reported previously. A threshold of ambient PM2.5 concentration, approximately 30 μg/m3, below which indoor sources contributed substantially to personal exposures, was also identified. The quantified relationship between indoor source contributions and ambient PM2.5 concentrations could serve as a metric of exposure errors when using outdoor monitors as an exposure proxy (without considering indoor-generated PM2.5), which may be of interest to epidemiological research. The proposed method can be generalized to larger geographical areas to better quantify PM2.5 outdoor infiltration and personal exposure.

Evaluating the efficacy of air pollution control policies is an essential part of the decision-making process to develop new policies and improve existing measures. Since 2005, Fujian Province of Mainland China and Taiwan across the Taiwan Strait have both implemented aggressive air pollution control policies designed based on different principles, but a comprehensive evaluation of these control policies on PM2.5pollution levels is still lacking. In the current study, we assessed the effects of these policies in the Taiwan Strait Region from 2005 to 2018 using full-coverage, high-resolution PM2.5generated by a satellite-driven machine learning model. A ten-fold cross-validation for our prediction model showed an R2value of 0.89, demonstrating that these predictions can be used for policy evaluation. During the 14-year period, PM2.5levels in all areas of Fujian and Taiwan underwent a significant decrease. Separate regression models for policy evaluation in Taiwan and Fujian showed that all considered policies have mitigated PM2.5pollution to various degrees. The Clean Air Action Plans (CAAP) is the most effective control policy in Taiwan, while the Action Plan of Air Pollution Prevention and Control (APPC-AP) and Three-year Action Plan for Blue Skies (3YAP-BS) as well as their provincial implementation plans are the most successful in Fujian. The effectiveness of control policies, however, varies by land-use types especially for Taiwan.

Background Emissions control programs targeting certain air pollution sources may alter PM2.5 composition, as well as the risk of adverse health outcomes associated with PM2.5. Objectives We examined temporal changes in the risk of emergency department (ED) visits for cardiovascular diseases (CVDs) and asthma associated with short-term increases in ambient PM2.5 concentrations in Los Angeles, California. Methods Poisson log-linear models with unconstrained distributed exposure lags were used to estimate the risk of CVD and asthma ED visits associated with short-term increases in daily PM2.5 concentrations, controlling for temporal and meteorological confounders. The models were run separately for three predefined time periods, which were selected based on the implementation of multiple emissions control programs (EARLY: 2005-2008; MIDDLE: 2009-2012; LATE: 2013-2016). Two-pollutant models with individual PM2.5 components and the remaining PM2.5 mass were also considered to assess the influence of changes in PM2.5 composition on changes in the risk of CVD and asthma ED visits associated with PM2.5 over time. Results The relative risk of CVD ED visits associated with a 10 μg/m3 increase in 4-day PM2.5 concentration (lag 0-3) was higher in the LATE period (rate ratio = 1.020, 95% confidence interval = [1.010, 1.030]) compared to the EARLY period (1.003, [0.996, 1.010]). In contrast, for asthma, relative risk estimates were largest in the EARLY period (1.018, [1.006, 1.029]), but smaller in the following periods. Similar temporal differences in relative risk estimates for CVD and asthma were observed among different age and season groups. No single component was identified as an obvious contributor to the changing risk estimates over time, and some components exhibited different temporal patterns in risk estimates from PM2.5 total mass, such as a decreased risk of CVD ED visits associated with sulfate over time. Conclusions Temporal changes in the risk of CVD and asthma ED visits associated with short-term increases in ambient PM2.5 concentrations were observed. These changes could be related to changes in PM2.5 composition (e.g., an increasing fraction of organic carbon and a decreasing fraction of sulfate in PM2.5). Other factors such as improvements in healthcare and differential exposure misclassification might also contribute to the changes.

Background: Studies of PM2.5 health effects are influenced by the spatiotemporal coverage and accuracy of exposure estimates. The use of satellite remote sensing data such as aerosol optical depth (AOD) in PM2.5 exposure modeling has increased recently in the US and elsewhere in the world. However, few studies have addressed this issue in southern California due to challenges with reflective surfaces and complex terrain. Methods: We examined the factors affecting the associations with satellite AOD using a two-stage spatial statistical model. The first stage estimated the temporal PM2.5/AOD relationships using a linear mixed effects model at 1 km resolution. The second stage accounted for spatial variation using geographically weighted regression. Goodness of fit for the final model was evaluated by comparing the daily PM2.5 concentrations generated by cross-validation (CV) with observations. These methods were applied to a region of southern California spanning from Los Angeles to San Diego. Results: Mean predicted PM2.5 concentration for the study domain was 8.84 µg m−3. Linear regression between CV predicted PM2.5 concentrations and observations had an R 2 of 0.80 and RMSE 2.25 µg m−3. The ratio of PM2.5 to PM10 proved an important variable in modifying the AOD/PM2.5 relationship (β = 14.79, p ≤ 0.001). Including this ratio improved model performance significantly (a 0.10 increase in CV R 2 and a 0.56 µg m−3 decrease in CV RMSE). Discussion: Utilizing the high-resolution MAIAC AOD, fine-resolution PM2.5 concentrations can be estimated where measurements are sparse. This study adds to the current literature using remote sensing data to achieve better exposure data in the understudied region of Southern California. Overall, we demonstrate the usefulness of MAIAC AOD and the importance of considering coarser particles in dust prone areas.

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

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 PM2.5 monitoring has limited temporal and spatial coverage, satellite aerosol remote sensing data has been increasingly applied to map large-scale PM2.5 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 PM2.5 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 PM2.5 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 PM2.5 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 PM2.5 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.