What question did this study set out to answer?

The central aim is to forecast air quality by effectively modeling extreme pollutant levels using a novel approach.

February 14, 2026Journal of the Royal Statistical Society Series A (Statistics in Society)3 citationsOpen Access

E-STGCN: extreme spatio-temporal graph convolutional networks for air quality forecasting

MPMadhurima Panja TCTanujit Chakraborty ABAnimesh Biswas

Puntos clave

The central aim is to forecast air quality by effectively modeling extreme pollutant levels using a novel approach.
Development of an E-STGCN model combining graph convolutional networks and extreme value theory.
Incorporation of long short-term memory units for learning temporal sequences.
Application of the model to air pollution data from 37 monitoring stations in Delhi.
Empirical evaluation of forecasting performance compared to benchmark models.
Utilization of conformal prediction to generate probabilistic prediction intervals.
E-STGCN demonstrated consistent forecasting performance across all seasons.
The model effectively captured extreme behavior of air pollutants.
Forecasts were superior compared to other temporal and spatio-temporal benchmarks.
Probabilistic intervals were generated, enhancing the reliability of predictions.

Resumen

Abstract Modelling and forecasting air quality is crucial for effective air pollution management and protecting public health. Air quality data, characterized by nonlinearity, nonstationarity, and spatio-temporal correlations, often include extreme pollutant levels in severely polluted cities (e.g. Delhi, the capital of India). This is ignored by various geometric deep learning models, such as spatio-temporal graph convolutional networks (STGCNs), which are otherwise effective for spatio-temporal forecasting. This study develops an extreme value theory (EVT) guided modified STGCN model (E-STGCN) for air pollution data to incorporate extreme behaviour across pollutant concentrations. E-STGCN combines graph convolutional networks for spatial modelling and EVT-guided long short-term memory units for temporal sequence learning. Along with spatial and temporal components, it incorporates a generalized Pareto distribution to capture the extreme behaviour of different air pollutants and embed this information into the learning process. The proposal is then applied to analyse air pollution data of 37 monitoring stations across Delhi, India. The forecasting performance for different test horizons is compared to benchmark forecasters (both temporal and spatio-temporal). It is found that E-STGCN has consistent performance across all seasons. The robustness of our results has also been evaluated empirically. Moreover, combined with conformal prediction, E-STGCN can produce probabilistic prediction intervals.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo