Spatiotemporal models to estimate ambient exposures at high spatiotemporal resolutions are crucial in large-scale air pollution epidemiological studies that follow participants over extended periods. Previous models typically rely on central-site monitoring data and/or covered short periods, limiting their applications to long-term cohort studies. Here we developed a spatiotemporal model that can reliably predict nitrogen oxide concentrations with a high spatiotemporal resolution over a long time span (>20 years). Leveraging the spatially extensive highly clustered exposure data from short-term measurement campaigns across 1-2 years and long-term central site monitoring in 1992-2013, we developed an integrated mixed-effect model with uncertainty estimates. Our statistical model incorporated nonlinear and spatial effects to reduce bias. Identified important predictors included temporal basis predictors, traffic indicators, population density, and subcounty-level mean pollutant concentrations. Substantial spatial autocorrelation (11-13%) was observed between neighboring communities. Ensemble learning and constrained optimization were used to enhance reliability of estimation over a large metropolitan area and a long period. The ensemble predictions of biweekly concentrations resulted in an R² of 0.85 (RMSE: 4.7 ppb) for NO₂ and 0.86 (RMSE: 13.4 ppb) for NO_x. Ensemble learning and constrained optimization generated stable time series, which notably improved the results compared with those from initial mixed-effects models.

Background

Increasingly ensemble learning-based spatiotemporal models are being used to estimate residential air pollution exposures in epidemiological studies. While these machine learning models typically have improved performance, they suffer from exposure measurement error that is inherent in all models. Our objective is to develop a framework to formally assess shared, multiplicative measurement error (SMME) in our previously published three-stage, ensemble learning-based nitrogen oxides (NO_x) model to identify its spatial and temporal patterns and predictors.

Methods

By treating the ensembles as an external dosimetry system, we quantified shared and unshared, multiplicative and additive (SUMA) measurement error components in our exposure model. We used generalized additive models (GAMs) with a smooth term for location to identify geographic locations with significantly elevated SMME and explain their spatial and temporal determinants.

Results

We found evidence of significant shared and unshared multiplicative error (p < 0.0001) in our ensemble-learning based spatiotemporal NO_x model predictions. Unshared multiplicative error was 26 times larger than SMME. We observed significant geographic (p < 0.0001) and temporal variation in SMME with the majority (43%) of predictions with elevated SMME occurring in the earliest time-period (1992-2000). Densely populated urban prediction regions with complex air pollution sources generally exhibited highest odds of elevated SMME.

Conclusions

We developed a novel statistical framework to formally evaluate the magnitude and drivers of SMME in ensemble learning-based exposure models. Our framework can be used to inform building future improved exposure models.

Background

Accurate estimation of nitrogen dioxide (NO₂) and nitrogen oxide (NO_x) 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 NO_x 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 NO_x 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 NO_x 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 NO_x [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 NO_x at high resolution over a large region.