Spatiotemporal Modelling of Airborne Birch and Grass Pollen Concentration across Switzerland: a comparison of statistical, machine learning and ensemble methods.

BACKGROUND: Statistical and machine learning models are commonly used to estimate spatial and temporal variability in exposure to environmental stressors, supporting epidemiological studies. We aimed to compare the performances, strengths and limitations of six different algorithms in the retrospective spatiotemporal modeling of daily birch and grass pollen concentrations at a spatial resolution of 1 km across Switzerland. METHODS: Daily birch and grass pollen concentrations were available from 14 measurement sites in Switzerland for 2000-2019. To develop the spatiotemporal models, we considered spatial-temporal, spatial and temporal predictors including meteorological factors, land-use, elevation, species distribution and Normalized Difference Vegetation Index (NDVI). We used six statistical and machine learning algorithms: LASSO, Ridge, Elastic net, Random forest, XGBoost and ANNs. We optimized model structures through feature selection and grid search techniques to obtain the best predictive performance. We used train-test split and cross-validation to avoid overfitting and overoptimistic performance indicators. We then combined these six models through multiple linear regression to develop an ensemble hybrid model. RESULTS: The 5th-95th percentiles of birch and grass pollen concentrations were 0-151 and 0-105 grains/m3, respectively. The hybrid ensemble model achieved the best RMSE on the test dataset for both birch and grass pollen with 94.4 and 19.7 grains/m3, respectively. Nonlinear models (Random forest, XGBoost and ANNs) achieved lower test RMSE's than linear models (LASSO, Ridge, Elastic net) for both pollen types, with RMSE's ranging from 105.9 to 140.5 grains/m3 for birch and from 20 to 25.4 grains/m3 for grass pollen. The Random forest algorithm yielded the best spatial and temporal performance among the six evaluated modelling methods. The ensemble hybrid model outperformed the six linear and nonlinear algorithms. Country-wide pollen concentration, land use, weather, and NDVI were important predictors. CONCLUSION: Nonlinear algorithms outperformed linear models and accurately explained complex, nonlinear relationships between environmental factors and measured concentrations.

Hourly land-use regression modeling for NO2 and PM2.5 in the Netherlands.

Annual average land-use regression (LUR) models have been widely used to assess spatial patterns of air pollution exposures. However, they fail to capture diurnal variability in air pollution and consequently might result in biased dynamic exposure assessments. In this study we aimed to model average hourly concentrations for two major pollutants, NO2 and PM2.5, for the Netherlands using the LUR algorithm. We modelled the spatial variation of average hourly concentrations for the years 2016-2019 combined, for two seasons, and for two weekday types. Two modelling approaches were used, supervised linear regression (SLR) and random forest (RF). The potential predictors included population, road, land use, satellite retrievals, and chemical transport model pollution estimates variables with different buffer sizes. We also temporally adjusted hourly concentrations from a 2019 annual model using the hourly monitoring data, to compare its performance with the hourly modelling approach. The results showed that hourly NO2 models performed overall well (5-fold cross validation R2 = 0.50-0.78), while the PM2.5 performed moderately (5-fold cross validation R2 = 0.24-0.62). Both for NO2 and PM2.5 the warm season models performed worse than the cold season ones, and the weekends' worse than weekdays'. The performance of the RF and SLR models was similar for both pollutants. For both SLR and RF, variables with larger buffer sizes representing variation in background concentrations, were selected more often in the weekend models compared to the weekdays, and in the warm season compared to the cold one. Temporal adjustment of annual average models performed overall worse than both modelling approaches (NO2 hourly R2 = 0.35-0.70; PM2.5 hourly R2 = 0.01-0.15). The difference in model performance and selection of variables across hours, seasons, and weekday types documents the benefit to develop independent hourly models when matching it to hourly time activity data.

Ambient air pollution and primary liver cancer incidence in four European cohorts within the ESCAPE project.

BACKGROUND: Tobacco smoke exposure increases the risk of cancer in the liver, but little is known about the possible risk associated with exposure to ambient air pollution. OBJECTIVES: We evaluated the association between residential exposure to air pollution and primary liver cancer incidence. METHODS: We obtained data from four cohorts with enrolment during 1985-2005 in Denmark, Austria and Italy. Exposure to nitrogen oxides (NO2 and NOX), particulate matter (PM) with diameter of less than 10µm (PM10), less than 2.5µm (PM2.5), between 2.5 and 10µm (PM2.5-10) and PM2.5 absorbance (soot) at baseline home addresses were estimated using land-use regression models from the ESCAPE project. We also investigated traffic density on the nearest road. We used Cox proportional-hazards models with adjustment for potential confounders for cohort-specific analyses and random-effects meta-analyses to estimate summary hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: Out of 174,770 included participants, 279 liver cancer cases were diagnosed during a mean follow-up of 17 years. In each cohort, HRs above one were observed for all exposures with exception of PM2.5 absorbance and traffic density. In the meta-analysis, all exposures were associated with elevated HRs, but none of the associations reached statistical significance. The summary HR associated with a 10-µg/m3 increase in NO2 was 1.10 (95% confidence interval (CI): 0.93, 1.30) and 1.34 (95% CI: 0.76, 2.35) for a 5-µg/m3 increase in PM2.5. CONCLUSIONS: The results provide suggestive evidence that ambient air pollution may increase the risk of liver cancer. Confidence intervals for associations with NO2 and NOX were narrower than for the other exposures.