Unveiling new horizons in forest floor remote sensing

Modern forestry is increasingly data-driven, yet many ecological and cultural features remain difficult to detect and act on beneath dense canopy. As a result, operational decisions often rely on maps and proxies with uneven reliability, creating blind spots for biodiversity, streams and wetlands, and small cultural heritage objects.

This project addresses these challenges by combining ecological and archaeological knowledge with high-resolution remote sensing and artificial intelligence (AI). The research develops methods to improve the detection of important features on the forest floor, while also assessing how reliable these detections are in different situations.

A key focus is to translate model performance into practical decision support. This means determining when existing airborne data is sufficient, when additional data collection using UAVs (drones) is needed, and when field verification is required.

The project is carried out as part of an industrial PhD within the WIFORCE Research School, at the interface between research and operational forestry. The long-term goal is to develop transparent and reliable decision support that can strengthen forest planning, improve inventories and compliance, and reduce the risk of damage to soils and cultural heritage over time.

Research team

PhD student: Oscar Andersson

Main supervisor: Lars Östlund

Assistant supervisors:

William Lidberg

Chelsea Elissa Budd

Collaborating partner: Holmen AB

Project period: October 2025 – October 2030

This project is part of WIFORCE – the Wallenberg Initiative in Forest Research, funded by the Knut and Alice Wallenberg Foundation.