Empirical spatial air pollution models have been applied extensively to assess exposure in epidemiological studies with increasingly sophisticated and complex statistical algorithms beyond ordinary linear regression. However, different algorithms have rarely been compared in terms of their predictive ability. This study compared 16 algorithms to predict annual average fine particle (PM2.5) and nitrogen dioxide (NO2) concentrations across Europe. The evaluated algorithms included linear stepwise regression, regularization techniques and machine learning methods. Air pollution models were developed based on the 2010 routine monitoring data from the AIRBASE dataset maintained by the European Environmental Agency (543 sites for PM2.5 and 2399 sites for NO2), using satellite observations, dispersion model estimates and land use variables as predictors. We compared the models by performing five-fold cross-validation (CV) and by external validation (EV) using annual average concentrations measured at 416 (PM2.5) and 1396 sites (NO2) from the ESCAPE study. We further assessed the correlations between predictions by each pair of algorithms at the ESCAPE sites. For PM2.5, the models performed similarly across algorithms with a mean CV R2 of 0.59 and a mean EV R2 of 0.53. Generalized boosted machine, random forest and bagging performed best (CV R2~0.63; EV R2 0.58-0.61), while backward stepwise linear regression, support vector regression and artificial neural network performed less well (CV R2 0.48-0.57; EV R2 0.39-0.46). Most of the PM2.5 model predictions at ESCAPE sites were highly correlated (R2 > 0.85, with the exception of predictions from the artificial neural network). For NO2, the models performed even more similarly across different algorithms, with CV R2s ranging from 0.57 to 0.62, and EV R2s ranging from 0.49 to 0.51. The predicted concentrations from all algorithms at ESCAPE sites were highly correlated (R2 > 0.9). For both pollutants, biases were low for all models except the artificial neural network. Dispersion model estimates and satellite observations were two of the most important predictors for PM2.5 models whilst dispersion model estimates and traffic variables were most important for NO2 models in all algorithms that allow assessment of the importance of variables. Different statistical algorithms performed similarly when modelling spatial variation in annual average air pollution concentrations using a large number of training sites.

The aim of the present study is to establish exposure-response relationships reflecting the percentage highly annoyed (%HA) as functions of road traffic, railway, and aircraft noise exposure, measured as day-evening-night level (Lden), as well as to elucidate the degree to which the acoustic indicator Intermittency Ratio (IR), which reflects the "eventfulness" of a noise situation, predicts noise annoyance. We conducted a mixed-mode representative population survey in a stratified random sample of 5592 residents exposed to transportation noise all over Switzerland. Source-specific noise exposure was calculated for each floor and each façade based on comprehensive traffic data. Noise annoyance was measured using the ICBEN 11-point scale. The survey was carried out in 4 waves at different times of the year. We hypothesized that in addition to Lden, the effects of noise on annoyance can be better explained when also considering the intensity of short-term variations of noise level over time. We therefore incorporated the acoustic indicator IR in the statistical models. For all noise sources, results revealed significant associations between Lden and %HA after controlling for confounders and independent predictors such as IR (measured over 24 h), exposure to other transportation noise sources, sex and age, language, home ownership, education level, living duration, temperature, and access to a quiet side of the dwelling. Aircraft noise annoyance scored markedly higher than annoyance to railway and road traffic noise at the same Lden level. Railway noise elicited higher percentages of highly annoyed persons than road traffic noise. Results furthermore suggest that for road traffic noise, IR has an additional effect on %HA and can explain shifts of the exposure-response curve of up to about 6 dB between low IR and high IR exposure situations, possibly due to the effect of different durations of noise-free intervals between events. For railway and aircraft noise annoyance, the predictive value of IR was limited.