Geoscientific Model Development
Geoscientific Model Development
GMD
Articles | Volume 18, issue 2
Articles | Volume 18, issue 2
https://doi.org/10.5194/gmd-18-337-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-18-337-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 18, issue 2
Methods for assessment of models
 | 
22 Jan 2025
Methods for assessment of models |  | 22 Jan 2025

Remote-sensing-based forest canopy height mapping: some models are useful, but might they provide us with even more insights when combined?

Nikola Besic, Nicolas Picard, Cédric Vega, Jean-Daniel Bontemps, Lionel Hertzog, Jean-Pierre Renaud, Fajwel Fogel, Martin Schwartz, Agnès Pellissier-Tanon, Gabriel Destouet, Frédéric Mortier, Milena Planells-Rodriguez, and Philippe Ciais

Data sets

M1: Global canopy height map for the year 2020 derived from Sentinel-2 and GEDI Nico Lang et al. https://doi.org/10.3929/ethz-b-000609802

M2: Canopy height and biomass map for Europe Siyu Liu et al. https://doi.org/10.5281/zenodo.8154445

M3: Estimation of forest height and biomass from open-access multi-sensor satellite imagery and GEDI Lidar data: high-resolution maps of metropolitan France David Morin et al. https://doi.org/10.5281/zenodo.8071004

M5: FORMS: Forest Multiple Source height, wood volume, and biomass maps in France at 10 to 30 m resolution based on Sentinel-1, Sentinel-2, and GEDI data with a deep learning approach M. Schwartz et al. https://doi.org/10.5281/zenodo.7840108

Global Forest Canopy Height, 2019 Global Land Analysis \& Discovery https://glad.umd.edu/dataset/gedi/

Model code and software

Code associated to the manuscript "Remote sensing-based forest canopy height mapping: some models are useful, but might they provide us with even more insights when combined?" Nikola Besic https://doi.org/10.5281/zenodo.13909201

Download
Short summary
The creation of advanced mapping models for forest attributes, utilizing remote sensing data and incorporating machine or deep learning methods, has become a key area of interest in the domain of forest observation and monitoring. This paper introduces a method where we blend and collectively interpret five models dedicated to estimating forest canopy height. We achieve this through Bayesian model averaging, offering a comprehensive analysis of these remote-sensing-based products.
Read more