A multi-source remote sensing dataset for large-scale forest monitoring

Area of Science:

• Earth and Environmental Sciences
• Forestry
• Remote Sensing

Background:

• Accurate forest modelling and monitoring require comprehensive, harmonized auxiliary data.
• Existing datasets may lack consistency across regions or integration of multiple data sources.
• Satellite-based data offers a scalable solution for large-scale forest assessments.

Purpose of the Study:

• To present a novel, multi-source satellite-based auxiliary dataset for forest modelling and monitoring.
• To provide harmonized data across seven European countries for consistent research.
• To support researchers in developing and validating forest-related models.

Main Methods:

• Integration of annual medoid composites from Sentinel-1, Sentinel-2, and Landsat imagery.
• Inclusion of spectral indices, Landsat-based 3I3D change metrics, forest masks, and forest types.
• Derivation of terrain variables using Copernicus GLO-30 Digital Elevation Model (DEM).

Main Results:

• A comprehensive dataset with forest cover, spectral behavior, and change metrics is generated.
• Harmonized predictors are available across seven European countries.
• A curated subset is accessible via Zenodo, with links to the full multi-terabyte archive.

Conclusions:

• The dataset provides essential tools for forest biodiversity conservation, carbon monitoring, and biomass modelling.
• It facilitates climate-change impact assessments by offering consistent forest dynamics data.
• The data enhances the scalability and ease of use for forest research across Europe.