Strandings from Space

Problem

Cetacean stranding monitoring is difficult in remote, complex or under-resourced coastlines. Very-high-resolution satellite imagery can help extend monitoring, but the workflow needs to be accessible, standardised and reproducible for different users and study regions.

Approach

• Pre-process VHR optical and SAR satellite imagery using open-source Python and QGIS workflows.
• Support structured image annotation with standardised attributes for stranded cetacean observations.
• Compare annotations from multiple observers using semi-automated clustering and count-congruence summaries.
• Keep the public repository useful while respecting satellite-imagery licensing and data-sharing boundaries.

What it demonstrates

This project adds the environmental-monitoring side of my portfolio. It shows collaborative research software, remote-sensing workflows, reproducible notebooks and practical tooling for ecological and conservation applications.

The canonical multi-author repository is maintained with the wider research team; my public fork is included here to show the connection to my portfolio and related code contributions.