HydroChronos: Forecasting Decades of Surface Water Change

TL;DR

HydroChronos introduces a comprehensive multi-modal dataset and a novel deep learning model for long-term surface water change forecasting, significantly advancing climate and water resource management research.

Contribution

The paper presents HydroChronos, a large-scale dataset and AquaClimaTempo UNet, a new architecture that improves surface water change prediction accuracy and interpretability.

Findings

Model outperforms persistence baseline by +14% F1 in change detection.

Model achieves +11% F1 in change direction classification.

Model reduces MAE by 0.1 in change magnitude regression.

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.