Application of machine learning (individual vs stacking) models on MERRA-2 data to predict surface PM_2.5...

The spatial coverage of PM_2.5 monitoring is non-uniform across India due to the limited number of ground monitoring stations. Alternatively, Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2), is an atmospheric reanalysis data used for estimating PM_2.5. MERRA-2 does not explicitly measure PM_2.5 but rather follows an empirical model. MERRA-2 data were spatiotemporally collocated with ground observation for validation across India. Significant underestimation in MERRA-2 prediction of PM_2.5 was observed over many monitoring stations ranging from -20 to 60 μg m^-3. The utility of Machine Learning (ML) models to overcome this challenge was assessed. MERRA-2 aerosol and meteorological parameters were the input features used to train and test the individual ML models and compare them with the stacking technique. Initially, with 10% of randomly selected data, individual model performance was assessed to identify the best model. XGBoost (XGB) was the best model (r² = 0.73) compared to Random Forest (RF) and LightGBM (LGBM). Stacking was then applied by keeping XGB as a meta-regressor. Stacked model results (r² = 0.77) outperformed the best standalone estimate of XGB. Stacking technique was used to predict hourly and daily PM_2.5 in different regions across India and each monitoring station. The eastern region exhibited the best hourly prediction (r² = 0.80) and substantial reduction in Mean Bias (MB = -0.03 μg m^-3), followed by the northern region (r² = 0.63 and MB = -0.10 μg m^-3), which showed better output due to the frequent observation of PM_2.5 >100 μg m^-3. Due to sparse data availability to train the ML models, the lowest performance was for the central region (r² = 0.46 and MB = -0.60 μg m^-3). Overall, India's PM_2.5 prediction was good on an hourly basis compared to a daily basis using the ML stacking technique.