Multi-Horizon Forecasting of Average Daily $\text{PM}_{2.5}$ Concentration: A Case Study of Air Pollution in Singapore

Understanding the spatial and temporal dynamics of surface concentrations of particulate matter (
$\mathbf{PM}_{2.5}$
and PM
10
) is essential in air quality modelling and climate research. Forecasting of air pollutants is necessary to understand the variation of pollutants and to plan and implement air pollution control measures. This work proposes a methodology for forecasting
$\mathbf{PM}_{2.5}$
concentration using various meteorological parameters over multiple time horizons. The proposed deep learning-based models forecast the daily average values concentration for
$\mathbf{PM}_{2.5}$
in Singapore. The performance metrics indicate the efficacy of the proposed model in forecasting
$\mathbf{PM}_{2.5}$
concentrations over different horizons. The analysis shows that
$\mathbf{PM}_{2.5}$
concentrations are best forecasted for the 3-days ahead scenario, with RMSE and MAPE being 3.767
$\mu \mathrm{g}/\mathrm{m}3$
and 6.82%, respectively.