Overview
Developed within the MAMBO project by UvA, this result provides harmonised geospatial data products of vegetation structure derived from airborne LiDAR point clouds, generated at national and site scales across multiple countries and time periods. At its core is Laserfarm, a free, open-source Python workflow for high-throughput, scalable, and reproducible processing of multi-terabyte LiDAR surveys.
Our result
This result delivers harmonised geospatial data products of vegetation structure derived from airborne LiDAR (ALS) point clouds across multiple countries, spatial scales, and time periods.
At its core is Laserfarm, an open-source Python-based workflow enabling high-throughput, scalable, and reproducible processing of multi-terabyte LiDAR datasets at regional to national scales. The workflow transforms raw point clouds into standardised raster products describing key habitat structure attributes at user-defined spatial resolutions. It supports end-to-end processing based on open standards and widely adopted libraries such as PDAL, GDAL, and Laserchicken.
In MAMBO, Laserfarm has been applied to seven demonstration sites across five European countries. Additionally, a full multi-temporal national-scale dataset for the Netherlands was generated at 10 m resolution, demonstrating capability for consistent, wall-to-wall habitat structure mapping.
All datasets are openly accessible via Zenodo and registered in the LifeWatch ERIC Metadata Catalogue and MAMBO community on the EU Open Research Repository. Full processing workflows and reproducible code are available via GitHub and WorkflowHub.
Needs and target audiences
We seek new project funding and strategic support to extend our airborne laser scanning workflows towards pan-European, wall-to-wall vegetation structure datasets for habitat condition monitoring. We aim to collaborate with research infrastructures and e-infrastructures to enable (1) scalable, operational production of harmonised vegetation structure products across Europe, and (2) their long-term integration, maintenance, and uptake in biodiversity monitoring frameworks. Engagement with new partners will be essential to support these goals.
Current stage
Laserfarm has reached TRL 7, with prototype workflows demonstrated in operational environments for scalable processing of airborne laser scanning data at different scales. Applied across seven sites in five countries and used to generate national-scale vegetation structure products for the Netherlands, it has been deployed on HPC systems, cloud infrastructures, and virtual research environments.
Unique value proposition
This result integrates a scalable, high-performance LiDAR processing workflow with the delivery of harmonised, analysis-ready vegetation structure metrics across sites, countries, and time periods. By combining open workflows with distributed computing, it overcomes bottlenecks in handling multi-terabyte datasets, enabling consistent and reproducible habitat structure mapping at different scales.
Keywords
- deep learning
- biodiversity monitoring
- species
- habitat condition
- threatened species
- citizen science
- remote sensing
- automated monitoring
Research project acronyms
- MAMBO
Research projects
- Modern Approaches to the Monitoring of BiOdiversity
Project identifiers
- 101060639
Contributors
- UNIVERSITEIT VAN AMSTERDAM
Source: Horizon Results Platform