Spatio-Temporal Forecasting of PM2.5 via Spatial-Diffusion guided Encoder-Decoder Architecture

TL;DR

This paper introduces a novel spatio-temporal graph neural network architecture that effectively captures diffusion and correlations to forecast PM2.5 levels across large regions, outperforming existing models.

Contribution

The paper proposes a new encoder-decoder GNN model with GRU and TransformerConv layers specifically designed for spatio-temporal diffusion in air quality forecasting.

Findings

Model accurately predicts PM2.5 levels across diverse regions.

Outperforms existing models on real-world datasets.

Effectively captures spatial diffusion and temporal dependencies.

Abstract

In many problem settings that require spatio-temporal forecasting, the values in the time-series not only exhibit spatio-temporal correlations but are also influenced by spatial diffusion across locations. One such example is forecasting the concentration of fine particulate matter (PM2.5) in the atmosphere which is influenced by many complex factors, the most important ones being diffusion due to meteorological factors as well as transport across vast distances over a period of time. We present a novel Spatio-Temporal Graph Neural Network architecture, that specifically captures these dependencies to forecast the PM2.5 concentration. Our model is based on an encoder-decoder architecture where the encoder and decoder parts leverage gated recurrent units (GRU) augmented with a graph neural network (TransformerConv) to account for spatial diffusion. Our model can also be seen as a generalization of various existing models for time-series or spatio-temporal forecasting. We demonstrate the model's effectiveness on two real-world PM2.5 datasets: (1) data collected by us using a recently deployed network of low-cost PM$_{2.5}$ sensors from 511 locations spanning the entirety of the Indian state of Bihar over a period of one year, and (2) another publicly available dataset that covers severely polluted regions from China for a period of 4 years. Our experimental results show our model's impressive ability to account for both spatial as well as temporal dependencies precisely.