Spatio-temporal Causal Learning for Streamflow Forecasting

TL;DR

This paper introduces a novel causal learning approach using spatio-temporal graph neural networks and river flow graphs to improve streamflow forecasting accuracy and efficiency in hydrologic modeling.

Contribution

It presents the Causal Streamflow Forecasting (CSF) model that integrates domain-specific river flow graphs with STGNNs to learn causal relationships and enhance prediction performance.

Findings

Outperforms standard STGNNs in accuracy.

Achieves higher computational efficiency than traditional methods.

Validated on Brazos River basin data.

Abstract

Streamflow plays an essential role in the sustainable planning and management of national water resources. Traditional hydrologic modeling approaches simulate streamflow by establishing connections across multiple physical processes, such as rainfall and runoff. These data, inherently connected both spatially and temporally, possess intrinsic causal relations that can be leveraged for robust and accurate forecasting. Recently, spatio-temporal graph neural networks (STGNNs) have been adopted, excelling in various domains, such as urban traffic management, weather forecasting, and pandemic control, and they also promise advances in streamflow management. However, learning causal relationships directly from vast observational data is theoretically and computationally challenging. In this study, we employ a river flow graph as prior knowledge to facilitate the learning of the causal structure and then use the learned causal graph to predict streamflow at targeted sites. The proposed model, Causal Streamflow Forecasting (CSF) is tested in a real-world study in the Brazos River basin in Texas. Our results demonstrate that our method outperforms regular spatio-temporal graph neural networks and achieves higher computational efficiency compared to traditional simulation methods. By effectively integrating river flow graphs with STGNNs, this research offers a novel approach to streamflow prediction, showcasing the potential of combining advanced neural network techniques with domain-specific knowledge for enhanced performance in hydrologic modeling.