Distinct hydrologic response patterns and trends worldwide revealed by physics-embedded learning

2025-04-14

AI Summary

Global water models (GWMs) suffer from static hydrological response representations and limited data-driven learning capacity, hindering accurate characterization of dynamic features—such as baseflow ratio and green–blue water partitioning—and thus impeding flood risk identification, water supply stress assessment, and estuarine freshwater input prediction. To address these limitations, we propose a physics-embedded deep learning framework. Our approach integrates water-balance priors, multi-source remote sensing and in-situ observations, and spatiotemporally decoupled feature modeling. We first reveal that global baseflow ratios exhibit pronounced temporal variability—up to ~20% over two decades—and enable dynamic representation of both baseflow ratio and green–blue water allocation. The framework reliably simulates nonlinear runoff elasticity and river “flicker”—a metric quantifying rapid flow fluctuations—at monthly and daily scales. Validation demonstrates substantial improvements in flood detection across mid-to-high latitudes, water supply stress assessment in South Asia’s tropics, and estuarine freshwater flux prediction in Europe, delivering high-resolution, interpretable decision support for global water management.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Computer Vision: Low Level & Physics-based Vision

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: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

To track rapid changes within our water sector, Global Water Models (GWMs) need to realistically represent hydrologic systems' response patterns - such as baseflow fraction - but are hindered by their limited ability to learn from data. Here we introduce a high-resolution physics-embedded big-data-trained model as a breakthrough in reliably capturing characteristic hydrologic response patterns ('signatures') and their shifts. By realistically representing the long-term water balance, the model revealed widespread shifts - up to ~20% over 20 years - in fundamental green-blue-water partitioning and baseflow ratios worldwide. Shifts in these response patterns, previously considered static, contributed to increasing flood risks in northern mid-latitudes, heightening water supply stresses in southern subtropical regions, and declining freshwater inputs to many European estuaries, all with ecological implications. With more accurate simulations at monthly and daily scales than current operational systems, this next-generation model resolves large, nonlinear seasonal runoff responses to rainfall ('elasticity') and streamflow flashiness in semi-arid and arid regions. These metrics highlight regions with management challenges due to large water supply variability and high climate sensitivity, but also provide tools to forecast seasonal water availability. This capability newly enables global-scale models to deliver reliable and locally relevant insights for water management.

Problem

• Improving Global Water Models to capture hydrologic response patterns
• Analyzing shifts in green-blue-water partitioning and baseflow ratios
• Enhancing seasonal water availability forecasting for management challenges

Innovation

• Physics-embedded big-data-trained model
• High-resolution hydrologic response capture
• Accurate seasonal runoff simulations