Geo-Aware Models for Stream Temperature Prediction across Different Spatial Regions and Scales

2025-10-10

AI Summary

Stream water temperature prediction faces significant challenges in cross-basin and multi-scale generalization, primarily due to inherent spatial heterogeneity of ecological data and sparse observational sampling. To address this, we propose the Geography-aware Gated Spatio-Temporal Graph Neural Network (Geo-GSTGNN), which innovatively integrates geographic embedding with a gated graph neural network to explicitly model both basin-shared physical principles and localized hydrological dynamics. Evaluated on a real-world long-term dataset spanning 37 years across 32 basins in the eastern United States, Geo-GSTGNN substantially outperforms existing physics-based models and purely data-driven approaches. It demonstrates strong generalization capability on unseen basins and across multiple temporal scales (daily, weekly, and monthly). This work establishes a transferable paradigm for environmental intelligent modeling under data scarcity and supports high-resolution decision-making in water resources management.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Understanding environmental ecosystems is vital for the sustainable management of our planet. However,existing physics-based and data-driven models often fail to generalize to varying spatial regions and scales due to the inherent data heterogeneity presented in real environmental ecosystems. This generalization issue is further exacerbated by the limited observation samples available for model training. To address these issues, we propose Geo-STARS, a geo-aware spatio-temporal modeling framework for predicting stream water temperature across different watersheds and spatial scales. The major innovation of Geo-STARS is the introduction of geo-aware embedding, which leverages geographic information to explicitly capture shared principles and patterns across spatial regions and scales. We further integrate the geo-aware embedding into a gated spatio-temporal graph neural network. This design enables the model to learn complex spatial and temporal patterns guided by geographic and hydrological context, even with sparse or no observational data. We evaluate Geo-STARS's efficacy in predicting stream water temperature, which is a master factor for water quality. Using real-world datasets spanning 37 years across multiple watersheds along the eastern coast of the United States, Geo-STARS demonstrates its superior generalization performance across both regions and scales, outperforming state-of-the-art baselines. These results highlight the promise of Geo-STARS for scalable, data-efficient environmental monitoring and decision-making.

Problem

• Predicting stream temperature across diverse watersheds and scales
• Addressing generalization issues in environmental models due to data heterogeneity
• Overcoming limited observational data for spatio-temporal temperature prediction

Innovation

• Geo-aware embedding captures patterns across regions
• Gated spatio-temporal graph neural network integration
• Learns complex patterns with sparse observational data