3D-FOREST - Novel in-situ 3D forest structure and biomass estimates to validate air/spaceborne products

You are here

Context and objectives

Forests are the ecosystem that contains most biomass. However, estimating biomass without cutting the trees is not easy. Traditional methods of estimating biomass are based on relationships between destructive estimates of volume and diameter and/or height, which can be measured more easily in the field. However, harvesting trees is expensive and often impractical. Terrestrial laser scanning (TLS) has the potential to go beyond simple allometric relationships by providing a detailed three dimensional characterization of forest structure. We will develop a framework to assess the quality of TLS volume estimates and use these estimates to develop new equations for  estimating biomass. We will also acquire ground-based and UAV (unmanned aerial vehicle or drone) laser scanning data at the same time. We will investigate how the fusion of data from ground-based and UAV LiDAR improves our biomass estimates and how we can upscale plot-based measurements to landscape level. Finally, we will use TLS input instead of traditional inventory data to improve forest growth models. The detailed TLS data is expected to produce more reliable models to predict biomass in forests. This project will focus on temperate (SONIA site) and tropical forests. 

Project outcome

Expected scientific results

 

The main innovations within this project are:

  1. A novel quality assessment framework of LiDAR derived volumes;
  2. Improved allometric above-ground biomass equations;
  3. A unique dataset of co-incident TLS and UAV LiDAR covering a wide range of forest ecosystems;
  4. New insights into the upscaling and spatial variation of forest structure;
  5. A proof of concept of improved forest growth modelling using LiDAR input data.

Expected products and services

  • Shared data/code: New datasets and code related to TLS will be made available via TLS RCN community and related data sharing platforms. Improvements and parameterizations for the ED2 model (WP 3) will be made available for the ED modelling community via the ED github system.
  • Teaching: New methods and findings will be used as teaching material for Bachelor and Master students in the two partner universities.
  • Training activities in organized by the TLSRCN community
  • Social media the visual character of TLS data allows to communicate easily through social media (twitter, facebook,…) via pictures, animations, …
  • Also classical press releases will be targeted in case of major breakthroughs or relevant events (field campaigns, …)

Expected outcomes will be of relevance for other scientists and students. We will collaborate closely with the Queensland Government and CSIRO in Australia to communicate our improved knowledge on monitoring biomass. We also aim to disseminate knowledge to the general public through (social) media. We will also communicate our knowledge (and data) to potential stakeholders such as natural resource managers, policy makers, TLS research coordination network, ESA, NASA and FAO.