Impact/context

Deforestation and forest degradation in Brazil’s Amazon and neighbouring biomes threaten biodiversity, indigenous and forest-dependent communities, and global climate stability. Although satellite-based monitoring systems already support deforestation mapping and enforcement actions, early detection remains challenging due to persistent cloud cover, subtle degradation signatures, and the sheer scale and complexity of national monitoring across diverse landscapes and seasonal conditions.

How quantum could help

This use case explores whether quantum machine learning can improve temporal anomaly detection in satellite image time series by learning compact yet expressive representations of forest dynamics under limited labelled data. By focusing quantum resources on detecting subtle, early-stage disturbances in the compressed feature space, the project aims to complement classical monitoring pipelines and assess whether quantum-enhanced temporal modelling can reduce false alarms and improve sensitivity in operational forest monitoring.