Biodiversity & Land-Cover Change Detection

Remote Sensing Science

Used multi-temporal satellite imagery to classify land cover and detect habitat change across a portfolio of sites, producing indicators and a scientific report to inform biodiversity assessments.

40+

Sites Assessed

8 yrs

Time Span

Land Classes

Gallery

01 / 03

Cloud-Free Composites

Annual median composites built from multi-temporal satellite imagery.

Tip: use ← → keys for keyboard navigation.

The story

Scroll through problem, approach, and outcome.

Problem

Assessing biodiversity and land-cover change across many sites by hand was inconsistent and could not scale to a growing portfolio.

Approach

Built a Google Earth Engine workflow to assemble cloud-free composites, ran supervised land-cover classification, and computed change metrics and spectral indices; analysed and validated results in Python and R, with maps prepared in QGIS.

Outcome

Produced consistent, reproducible change indicators and a scientific report that supported biodiversity and earth-systems assessments across the site portfolio.

Tech Stack

Google Earth EnginePythonRGeoPandasQGIS