Globally Scalable Glacier Mapping by Deep Learning Matches Expert Delineation Accuracy

TL;DR

This paper introduces GlaViTU, a deep learning model that achieves expert-level accuracy in global glacier mapping using satellite imagery, with strategies for multitemporal and cross-sensor generalization, and provides a new benchmark dataset.

Contribution

The paper presents a novel deep learning model and strategies for scalable, accurate, and automated global glacier mapping, including a benchmark dataset for the community.

Findings

Achieves intersection over union >0.85 on unseen images in most cases

Performance approaches or matches expert delineation uncertainties

Adding SAR data improves accuracy across all regions

Abstract

Accurate global glacier mapping is critical for understanding climate change impacts. Despite its importance, automated glacier mapping at a global scale remains largely unexplored. Here we address this gap and propose Glacier-VisionTransformer-U-Net (GlaViTU), a convolutional-transformer deep learning model, and five strategies for multitemporal global-scale glacier mapping using open satellite imagery. Assessing the spatial, temporal and cross-sensor generalisation shows that our best strategy achieves intersection over union >0.85 on previously unobserved images in most cases, which drops to >0.75 for debris-rich areas such as High-Mountain Asia and increases to >0.90 for regions dominated by clean ice. A comparative validation against human expert uncertainties in terms of area and distance deviations underscores GlaViTU performance, approaching or matching expert-level delineation. Adding synthetic aperture radar data, namely, backscatter and interferometric coherence, increases the accuracy in all regions where available. The calibrated confidence for glacier extents is reported making the predictions more reliable and interpretable. We also release a benchmark dataset that covers 9% of glaciers worldwide. Our results support efforts towards automated multitemporal and global glacier mapping.