Targeted Semantic Segmentation of Himalayan Glacial Lakes Using Time-Series SAR: Towards Automated GLOF Early Warning

TL;DR

This paper develops an automated deep learning pipeline using time-series SAR data for targeted monitoring of Himalayan glacial lakes to enable early warning of GLOFs, overcoming optical imagery limitations.

Contribution

It introduces a novel 'temporal-first' training strategy and an operational Dockerised architecture for dynamic, automated GLOF early warning systems using Sentinel-1 SAR data.

Findings

Achieved IoU of 0.9130 validating the model's accuracy.

Demonstrated the effectiveness of the 'temporal-first' training approach.

Provided a scalable system architecture for real-time GLOF monitoring.

Abstract

Glacial Lake Outburst Floods (GLOFs) are one of the most devastating climate change induced hazards. Existing remote monitoring approaches often prioritise maximising spatial coverage to train generalistic models or rely on optical imagery hampered by persistent cloud coverage. This paper presents an end-to-end, automated deep learning pipeline for the targeted monitoring of high-risk Himalayan glacial lakes using time-series Sentinel-1 SAR. We introduce a "temporal-first" training strategy, utilising a U-Net with an EfficientNet-B3 backbone trained on a curated dataset of a cohort of 4 lakes (Tsho Rolpa, Chamlang Tsho, Tilicho and Gokyo Lake). The model achieves an IoU of 0.9130 validating the success and efficacy of the "temporal-first" strategy required for transitioning to Early Warning Systems. Beyond the model, we propose an operational engineering architecture: a Dockerised pipeline that automates data ingestion via the ASF Search API and exposes inference results via a RESTful endpoint. This system shifts the paradigm from static mapping to dynamic and automated early warning, providing a scalable architectural foundation for future development in Early Warning Systems.