Spatiotemporal Causal Decoupling Model for Air Quality Forecasting

2025-05-26

AI Summary

To address insufficient modeling of complex spatiotemporal causal relationships between air quality index (AQI) and meteorological factors, poor interpretability, and low robustness in multi-step forecasting, this paper proposes AirCade—a novel causal disentanglement framework. AirCade explicitly separates synchronous causal effects in AQI dynamics via a first-of-its-kind causal disentanglement module. It integrates causal graph priors, meteorological knowledge embedding, graph neural networks, and temporal attention mechanisms, while incorporating counterfactual intervention for uncertainty-aware robust modeling. Experiments on open-source datasets demonstrate that AirCade reduces average relative error by over 20% compared to state-of-the-art methods. It significantly improves forecasting accuracy and generalization stability for horizons ranging from 1 to 72 hours. Moreover, AirCade enables interpretable causal pathway analysis, facilitating transparent attribution of AQI variations to underlying meteorological drivers.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Due to the profound impact of air pollution on human health, livelihoods, and economic development, air quality forecasting is of paramount significance. Initially, we employ the causal graph method to scrutinize the constraints of existing research in comprehensively modeling the causal relationships between the air quality index (AQI) and meteorological features. In order to enhance prediction accuracy, we introduce a novel air quality forecasting model, AirCade, which incorporates a causal decoupling approach. AirCade leverages a spatiotemporal module in conjunction with knowledge embedding techniques to capture the internal dynamics of AQI. Subsequently, a causal decoupling module is proposed to disentangle synchronous causality from past AQI and meteorological features, followed by the dissemination of acquired knowledge to future time steps to enhance performance. Additionally, we introduce a causal intervention mechanism to explicitly represent the uncertainty of future meteorological features, thereby bolstering the model's robustness. Our evaluation of AirCade on an open-source air quality dataset demonstrates over 20% relative improvement over state-of-the-art models.

Problem

• Modeling causal relationships between AQI and meteorological features
• Improving air quality prediction accuracy via causal decoupling
• Enhancing robustness with causal intervention for meteorological uncertainty

Innovation

• Causal decoupling for AQI and meteorology
• Spatiotemporal module with knowledge embedding
• Causal intervention for future uncertainty