Compute Daily Particulate Matter (PM2.5) Predictions for Europe Using Machine Learning
A space-time extremely randomised trees model was used to estimate PM2.5 concentrations over Europe. Satellite remote sensing data, meteorological data, and land variables were used as the independent variables, PM2.5 ground-observations were used as the dependent variable to create our model. Used data is shown in Table 1.
|Name of the variable||Unit|
|Aerosol optical depth||–|
|Total Column Water Vapour||Kg/m2|
The model achieved good results with out of sample cross validated R2 of 0.69, RMSE of 5 μg/m3 and MAE of 3.3 μg/m3. Other validation methods were applied as shown in Table 2.
The outcome of this study is a daily full coverage PM2.5 dataset with 1 km spatial resolution for a three-year period 2018–2020.
We selected November 2020 for downloading demonstration data, the same principle applies to the remaining months by changing the month and the year in the links.