thumbnail

Physics-guided graph meta learning for predicting water temperature and streamflow in stream networks

By: , , and 
https://doi.org/10.1145/3534678.3539115

Links

Abstract

This paper proposes a graph-based meta learning approach to separately predict water quantity and quality variables for river segments in stream networks. Given the heterogeneous water dynamic patterns in large-scale basins, we introduce an additional meta-learning condition based on physical characteristics of stream segments, which allows learning different sets of initial parameters for different stream segments. Specifically, we develop a representation learning method that leverages physical simulations to embed the physical characteristics of each segment. The obtained embeddings are then used to cluster river segments and add the condition for the meta-learning process. We have tested the performance of the proposed method for predicting daily water temperature and streamflow for the Delaware River Basin (DRB) over a 14 year period. The results confirm the effectiveness of our method in predicting target variables even using sparse training samples. We also show that our method can achieve robust performance with different numbers of clusterings.

Publication type Conference Paper
Publication Subtype Conference Paper
Title Physics-guided graph meta learning for predicting water temperature and streamflow in stream networks
DOI 10.1145/3534678.3539115
Year Published 2022
Language English
Publisher ACM Digital Library
Contributing office(s) WMA - Integrated Information Dissemination Division
Description 10 p.
Larger Work Type Book
Larger Work Subtype Conference publication
Larger Work Title Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining
First page 2752
Last page 2761
Conference Title 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
Conference Location Washington DC
Conference Date August 14-18, 2022
Google Analytic Metrics Metrics page
Additional publication details