You are here

Spatial statistical learning methods for estimating ambient air pollution

Principal Investigator: 

Utrecht University, Netherlands

Hoek and colleagues will prepare maps of modeled annual average air pollution across the Netherlands, validate the maps using new measurements from over 100 sites, and evaluate the performance of several exposure models. The investigators will conduct cross-comparisons to evaluate how different exposure assessment methods compare in their ability to predict long-term pollutant concentrations, with a particular focus on spatial variability of pollutants.

Funded under

Poster abstract for HEI Annual Conference 2022

Comparison of Long-term Air Pollution Exposure Assessment Based on Mobile Monitoring, Low-cost Sensors, Dispersion Modelling and Routine Monitoring-based Models

Gerard Hoek1, Femke Bouma1, Kees Meliefste1, Ulrike Gehring1, Roel Vermeulen1, Kees de Hoogh2, Sjoerd van Ratingen3, Wouter Hendrickx3 Nicole Janssen3, Joost Wesseling3

z1Institute for Risk Assessment Sciences, Utrecht University, the Netherlands.

2Swiss Tropical and Public Health Institute, Switzerland

3National Institute for Public Health and the Environment, the Netherlands

Background. Assessment of long-term exposure to traffic-related outdoor air pollution remains a major challenge for epidemiological studies. One challenge is the characterization of the spatial variation of the ambient concentrations of key traffic-related air pollutants including ultrafine particles (UFP), Black carbon (BC) and NO2 as these pollutants vary on a fine spatial scale. Recently, epidemiological studies have used different approaches, including land use regression (LUR) models based upon mobile monitoring (UFP, BC), models based upon low-cost sensors (PM2.5, NO2) or routine monitoring data and increasingly hybrid models incorporating routine surface monitoring, satellite, chemical transport and land use data. Very little information is available about the relative performance of these different approaches to assess long-term exposure to traffic-related air pollution. Differences in performance may affect conclusions from epidemiological studies applying different exposure assessment approaches regarding the relevance of these pollutants for health.

Methods. The project will generate and evaluate annual average air pollution maps using eight different exposure assessment methods, which differ in modeling approach (empirical LUR, deterministic dispersion models and hybrid models) and monitoring data used (low-cost sensors, mobile monitoring). For all empirical models we will test three model development algorithms covering major families of modeling approaches: supervised linear regression, random forest and LASSO.

The predictions from and performance of the models will be compared at 20,000 addresses across the Netherlands and tested on newly collected and existing external validation data. Epidemiological analyses in three cohort studies will be conducted to compare health risk estimates for/between the different exposure assessment methods. The studies include a national administrative cohort, an adult cohort study and a mature birth cohort in which we will assess mortality, cardiovascular disease incidence, lung function and asthma, respectively. 

Results. Data collection for the low-cost sensor network and the new external validation started in September 2021 and is currently ongoing. Exposure models based on pre-existing data are currently being developed. Epidemiological analyses of the association between long-term UFP exposure and allergic sensitization in the mature birth cohort showed no significant association and no significant differences between UFP exposure models based on different algorithms. Significant associations were found between NO2 exposure and sensitization to food allergens. An analysis of UFP-related mortality in the national administrative cohort is ongoing.

Conclusions. As data collection and exposure modelling have not yet finished, no conclusions about the comparison on exposure models can be drawn currently. We found no evidence that UFP exposure was associated with allergic sensitization in children.