HydroChronos: Forecasting Decades of Surface Water Change

2025-06-17

AI Summary

Existing surface water dynamic forecasting lacks a standardized benchmark that is cross-continental, long-term (multi-decadal), and multimodal. To address this, we introduce the first cross-continental surface water prediction benchmark dataset spanning 30 years, integrating Landsat/Sentinel-2 imagery, climate variables, and digital elevation model (DEM) data. We formalize three core tasks: change detection, directional classification, and magnitude regression. We propose AquaClimaTempo UNet—a novel spatiotemporal U-Net architecture featuring an embedded climate branch—establishing the first climate-aware remote sensing modeling paradigm for time-series water dynamics. Our framework incorporates multi-source temporal alignment, joint task learning, and SHAP-based interpretability analysis to identify dominant climatic drivers and informative spectral channels underlying water body changes. Experiments demonstrate significant improvements over the persistence baseline: +14% F1-score for change detection, +11% for directional classification, and −0.1 MAE for magnitude regression—validating the substantial predictive gain from integrating climate information into long-term hydrological forecasting.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Time-Series/Data Streams
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Forecasting surface water dynamics is crucial for water resource management and climate change adaptation. However, the field lacks comprehensive datasets and standardized benchmarks. In this paper, we introduce HydroChronos, a large-scale, multi-modal spatiotemporal dataset for surface water dynamics forecasting designed to address this gap. We couple the dataset with three forecasting tasks. The dataset includes over three decades of aligned Landsat 5 and Sentinel-2 imagery, climate data, and Digital Elevation Models for diverse lakes and rivers across Europe, North America, and South America. We also propose AquaClimaTempo UNet, a novel spatiotemporal architecture with a dedicated climate data branch, as a strong benchmark baseline. Our model significantly outperforms a Persistence baseline for forecasting future water dynamics by +14% and +11% F1 across change detection and direction of change classification tasks, and by +0.1 MAE on the magnitude of change regression. Finally, we conduct an Explainable AI analysis to identify the key climate variables and input channels that influence surface water change, providing insights to inform and guide future modeling efforts.

Problem

• Lack of comprehensive datasets for surface water dynamics forecasting
• Absence of standardized benchmarks in water change prediction
• Need for accurate climate-influenced water dynamics modeling

Innovation

• Large-scale multi-modal spatiotemporal dataset HydroChronos
• Novel AquaClimaTempo UNet architecture with climate branch
• Explainable AI analysis identifies key climate variables