Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning

Abstract Accurately measuring vegetation height is essential for understanding ecosystem structure, carbon storage, and biodiversity, yet global height models have overwhelmingly focused on forests, excluding ecosystems with shorter herbaceous vegetation or shrubs. To address this gap in vegetation...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria O. Hunter, Leandro Parente, Yu-feng Ho, Carmelo Bonannella, Laerte Guimarães Ferreira, Douglas Morton, Davide Consoli, Lindsey Sloat
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-05739-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849226649268125696
author Maria O. Hunter
Leandro Parente
Yu-feng Ho
Carmelo Bonannella
Laerte Guimarães Ferreira
Douglas Morton
Davide Consoli
Lindsey Sloat
author_facet Maria O. Hunter
Leandro Parente
Yu-feng Ho
Carmelo Bonannella
Laerte Guimarães Ferreira
Douglas Morton
Davide Consoli
Lindsey Sloat
author_sort Maria O. Hunter
collection DOAJ
description Abstract Accurately measuring vegetation height is essential for understanding ecosystem structure, carbon storage, and biodiversity, yet global height models have overwhelmingly focused on forests, excluding ecosystems with shorter herbaceous vegetation or shrubs. To address this gap in vegetation structure data, we developed the first global estimate of median vegetation height annually from 2000–2022 at 30 m resolution, using ICESat-2 satellite Lidar, Landsat cloud free composites, and other Earth Observation raster data. Thirty two (32) million ICESat-2 20 m segments were used within 10 independent draws to build ensemble Gradient Boosted Tree (GBT) models and estimate 90% prediction intervals. Our model achieves a root mean square error (RMSE) of 2.35 m, R2 values of 0.515 and a D2 regression score of 0.62 estimated on the testing set. Comparisons with existing global height products show that our approach increases detail and heterogeneity of height in short vegetation ecosystems. Output maps are publicly available together with reference samples and trained models under CC-BY license.
format Article
id doaj-art-68ebe1dcc8a044b8bf58895c755b2269
institution Kabale University
issn 2052-4463
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Data
spelling doaj-art-68ebe1dcc8a044b8bf58895c755b22692025-08-24T11:07:22ZengNature PortfolioScientific Data2052-44632025-08-0112111710.1038/s41597-025-05739-6Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine LearningMaria O. Hunter0Leandro Parente1Yu-feng Ho2Carmelo Bonannella3Laerte Guimarães Ferreira4Douglas Morton5Davide Consoli6Lindsey Sloat7Remote Sensing and GIS Laboratory (LAPIG/UFG)OpenGeoHub FoundationOpenGeoHub FoundationOpenGeoHub FoundationRemote Sensing and GIS Laboratory (LAPIG/UFG)NASA Goddard Space Flight CenterOpenGeoHub FoundationLand & Carbon Lab, World Resources InstituteAbstract Accurately measuring vegetation height is essential for understanding ecosystem structure, carbon storage, and biodiversity, yet global height models have overwhelmingly focused on forests, excluding ecosystems with shorter herbaceous vegetation or shrubs. To address this gap in vegetation structure data, we developed the first global estimate of median vegetation height annually from 2000–2022 at 30 m resolution, using ICESat-2 satellite Lidar, Landsat cloud free composites, and other Earth Observation raster data. Thirty two (32) million ICESat-2 20 m segments were used within 10 independent draws to build ensemble Gradient Boosted Tree (GBT) models and estimate 90% prediction intervals. Our model achieves a root mean square error (RMSE) of 2.35 m, R2 values of 0.515 and a D2 regression score of 0.62 estimated on the testing set. Comparisons with existing global height products show that our approach increases detail and heterogeneity of height in short vegetation ecosystems. Output maps are publicly available together with reference samples and trained models under CC-BY license.https://doi.org/10.1038/s41597-025-05739-6
spellingShingle Maria O. Hunter
Leandro Parente
Yu-feng Ho
Carmelo Bonannella
Laerte Guimarães Ferreira
Douglas Morton
Davide Consoli
Lindsey Sloat
Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
Scientific Data
title Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
title_full Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
title_fullStr Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
title_full_unstemmed Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
title_short Global 30-m annual median vegetation height maps (2000–2022) based on ICESat-2 data and Machine Learning
title_sort global 30 m annual median vegetation height maps 2000 2022 based on icesat 2 data and machine learning
url https://doi.org/10.1038/s41597-025-05739-6
work_keys_str_mv AT mariaohunter global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT leandroparente global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT yufengho global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT carmelobonannella global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT laerteguimaraesferreira global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT douglasmorton global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT davideconsoli global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning
AT lindseysloat global30mannualmedianvegetationheightmaps20002022basedonicesat2dataandmachinelearning