Physics-Guided Recurrent Graph Model for Predicting Flow and Temperature in River Networks [journal]

Journal

Proceedings of the 2021 SIAM International Conference on Data Mining (SDM) - April 29 - May 3, 2021

Authors

Xiaowei Jia (Ph.D. 2020), Jacob Zwart, Jeffrey Sadler, Alison Appling, Samantha Oliver, Steven Markstrom, Jared Willard (Ph.D. student), Shaoming Xu (Ph.D. student), Michael Steinbach (researcher), Jordan Read, Vipin Kumar (professor)

Abstract

This paper proposes a physics-guided machine learning approach that combines machine learning models and physics-based models to improve the prediction of water flow and temperature in river networks. We first build a recurrent graph network model to capture the interactions among multiple segments in the river network. Then we transfer knowledge from physics-based models to guide the learning of the machine learning model. We also propose a new loss function that balances the performance over different river segments. We demonstrate the effectiveness of the proposed method in predicting temperature and streamflow in a subset of the Delaware River Basin. In particular, the proposed method has brought a 33%/14% accuracy improvement over the state-of-the-art physics-based model and 24%/14% over traditional machine learning models (e.g., LSTM) in temperature/streamflow prediction using very sparse (0.1%) training data. The proposed method has also been shown to produce better performance when generalized to different seasons or river segments with different streamflow ranges.

Link to full paper

Physics-Guided Recurrent Graph Model for Predicting Flow and Temperature in River Networks

Keywords

data mining, machine learning, environmental data science