Forest biomass retrieval approaches from earth observation in different biomes

Yrjö Rauste, Pedro Rodríguez-Veiga (Corresponding Author), Shaun Quegan, Joao Carreiras, Henrik Persson, Johan Fransson, Agata Hoscilo, Dariusz Zielkowski, Krzysztof Sterenczak, Sandra Lohenberger, Matthias Stängel, Anna Berninger, Florian Siegert, Valerio Avitabile, Martin Herold, Stéphane Mermoz, Alexandre Bouvet, Thuy Le Toan, Nuno Carvalhais, Maurizio Santoro & 9 others Oliver Cartus, Renaud Mathieu, Gregory Asner, Christian Thiel, Carsten Pathe, Chris Schmullius, Frank Martin Seifert, Kevin Tansey, Heiko Balzter

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)

Abstract

The amount and spatial distribution of forest aboveground biomass (AGB) were estimated using a range of regionally developed methods using Earth Observation data for Poland, Sweden and regions in Indonesia (Kalimantan), Mexico (Central Mexico and Yucatan peninsula), and South Africa (Eastern provinces) for the year 2010. These regions are representative of numerous forest biomes and biomass levels globally, from South African woodlands and savannas to the humid tropical forest of Kalimantan. AGB retrieval in each region relied on different sources of reference data, including forest inventory plot data and airborne LiDAR observations, and used a range of retrieval algorithms. This is the widest inter-comparison of regional-to-national AGB maps to date in terms of area, forest types, input datasets, and retrieval methods. The accuracy assessment of all regional maps using independent field data or LiDAR AGB maps resulted in an overall root mean square error (RMSE) ranging from 10 t ha−1 to 55 t ha−1 (37% to 67% relative RMSE), and an overall bias ranging from −1 t ha−1 to +5 t ha−1 at pixel level. The regional maps showed better agreement with field data than previously developed and widely used pan-tropical or northern hemisphere datasets. The comparison of accuracy assessments showed commonalities in error structures despite the variety of methods, input data, and forest biomes. All regional retrievals resulted in overestimation (up to 63 t ha−1) in the lower AGB classes, and underestimation (up to 85 t ha−1) in the higher AGB classes. Parametric model-based algorithms present advantages due to their low demand on in situ data compared to non-parametric algorithms, but there is a need for datasets and retrieval methods that can overcome the biases at both ends of the AGB range. The outcomes of this study should be considered when developing algorithms to estimate forest biomass at continental to global scale level.
Original languageEnglish
Pages (from-to)53-68
Number of pages16
JournalInternational Journal of Applied Earth Observation and Geoinformation
Volume77
DOIs
Publication statusPublished - 3 Jan 2019
MoE publication typeNot Eligible

Fingerprint

aboveground biomass
biome
Biomass
Earth (planet)
biomass
accuracy assessment
Mean square error
forest inventory
savanna
tropical forest
woodland
Northern Hemisphere
pixel
Spatial distribution
spatial distribution
Pixels
method

Keywords

  • Remote sensing
  • biomass

Cite this

Rauste, Yrjö ; Rodríguez-Veiga, Pedro ; Quegan, Shaun ; Carreiras, Joao ; Persson, Henrik ; Fransson, Johan ; Hoscilo, Agata ; Zielkowski, Dariusz ; Sterenczak, Krzysztof ; Lohenberger, Sandra ; Stängel, Matthias ; Berninger, Anna ; Siegert, Florian ; Avitabile, Valerio ; Herold, Martin ; Mermoz, Stéphane ; Bouvet, Alexandre ; Le Toan, Thuy ; Carvalhais, Nuno ; Santoro, Maurizio ; Cartus, Oliver ; Mathieu, Renaud ; Asner, Gregory ; Thiel, Christian ; Pathe, Carsten ; Schmullius, Chris ; Seifert, Frank Martin ; Tansey, Kevin ; Balzter, Heiko. / Forest biomass retrieval approaches from earth observation in different biomes. In: International Journal of Applied Earth Observation and Geoinformation. 2019 ; Vol. 77. pp. 53-68.
@article{b651dade06b54d09a0595101c5f53273,
title = "Forest biomass retrieval approaches from earth observation in different biomes",
abstract = "The amount and spatial distribution of forest aboveground biomass (AGB) were estimated using a range of regionally developed methods using Earth Observation data for Poland, Sweden and regions in Indonesia (Kalimantan), Mexico (Central Mexico and Yucatan peninsula), and South Africa (Eastern provinces) for the year 2010. These regions are representative of numerous forest biomes and biomass levels globally, from South African woodlands and savannas to the humid tropical forest of Kalimantan. AGB retrieval in each region relied on different sources of reference data, including forest inventory plot data and airborne LiDAR observations, and used a range of retrieval algorithms. This is the widest inter-comparison of regional-to-national AGB maps to date in terms of area, forest types, input datasets, and retrieval methods. The accuracy assessment of all regional maps using independent field data or LiDAR AGB maps resulted in an overall root mean square error (RMSE) ranging from 10 t ha−1 to 55 t ha−1 (37{\%} to 67{\%} relative RMSE), and an overall bias ranging from −1 t ha−1 to +5 t ha−1 at pixel level. The regional maps showed better agreement with field data than previously developed and widely used pan-tropical or northern hemisphere datasets. The comparison of accuracy assessments showed commonalities in error structures despite the variety of methods, input data, and forest biomes. All regional retrievals resulted in overestimation (up to 63 t ha−1) in the lower AGB classes, and underestimation (up to 85 t ha−1) in the higher AGB classes. Parametric model-based algorithms present advantages due to their low demand on in situ data compared to non-parametric algorithms, but there is a need for datasets and retrieval methods that can overcome the biases at both ends of the AGB range. The outcomes of this study should be considered when developing algorithms to estimate forest biomass at continental to global scale level.",
keywords = "Remote sensing, biomass",
author = "Yrj{\"o} Rauste and Pedro Rodr{\'i}guez-Veiga and Shaun Quegan and Joao Carreiras and Henrik Persson and Johan Fransson and Agata Hoscilo and Dariusz Zielkowski and Krzysztof Sterenczak and Sandra Lohenberger and Matthias St{\"a}ngel and Anna Berninger and Florian Siegert and Valerio Avitabile and Martin Herold and St{\'e}phane Mermoz and Alexandre Bouvet and {Le Toan}, Thuy and Nuno Carvalhais and Maurizio Santoro and Oliver Cartus and Renaud Mathieu and Gregory Asner and Christian Thiel and Carsten Pathe and Chris Schmullius and Seifert, {Frank Martin} and Kevin Tansey and Heiko Balzter",
year = "2019",
month = "1",
day = "3",
doi = "10.1016/j.jag.2018.12.008",
language = "English",
volume = "77",
pages = "53--68",
journal = "International Journal of Applied Earth Observation and Geoinformation",
issn = "1569-8432",
publisher = "Elsevier",

}

Rauste, Y, Rodríguez-Veiga, P, Quegan, S, Carreiras, J, Persson, H, Fransson, J, Hoscilo, A, Zielkowski, D, Sterenczak, K, Lohenberger, S, Stängel, M, Berninger, A, Siegert, F, Avitabile, V, Herold, M, Mermoz, S, Bouvet, A, Le Toan, T, Carvalhais, N, Santoro, M, Cartus, O, Mathieu, R, Asner, G, Thiel, C, Pathe, C, Schmullius, C, Seifert, FM, Tansey, K & Balzter, H 2019, 'Forest biomass retrieval approaches from earth observation in different biomes' International Journal of Applied Earth Observation and Geoinformation, vol. 77, pp. 53-68. https://doi.org/10.1016/j.jag.2018.12.008

Forest biomass retrieval approaches from earth observation in different biomes. / Rauste, Yrjö; Rodríguez-Veiga, Pedro (Corresponding Author); Quegan, Shaun; Carreiras, Joao; Persson, Henrik; Fransson, Johan; Hoscilo, Agata; Zielkowski, Dariusz; Sterenczak, Krzysztof; Lohenberger, Sandra; Stängel, Matthias; Berninger, Anna; Siegert, Florian; Avitabile, Valerio; Herold, Martin; Mermoz, Stéphane; Bouvet, Alexandre; Le Toan, Thuy; Carvalhais, Nuno; Santoro, Maurizio; Cartus, Oliver; Mathieu, Renaud; Asner, Gregory; Thiel, Christian; Pathe, Carsten; Schmullius, Chris; Seifert, Frank Martin; Tansey, Kevin; Balzter, Heiko.

In: International Journal of Applied Earth Observation and Geoinformation, Vol. 77, 03.01.2019, p. 53-68.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Forest biomass retrieval approaches from earth observation in different biomes

AU - Rauste, Yrjö

AU - Rodríguez-Veiga, Pedro

AU - Quegan, Shaun

AU - Carreiras, Joao

AU - Persson, Henrik

AU - Fransson, Johan

AU - Hoscilo, Agata

AU - Zielkowski, Dariusz

AU - Sterenczak, Krzysztof

AU - Lohenberger, Sandra

AU - Stängel, Matthias

AU - Berninger, Anna

AU - Siegert, Florian

AU - Avitabile, Valerio

AU - Herold, Martin

AU - Mermoz, Stéphane

AU - Bouvet, Alexandre

AU - Le Toan, Thuy

AU - Carvalhais, Nuno

AU - Santoro, Maurizio

AU - Cartus, Oliver

AU - Mathieu, Renaud

AU - Asner, Gregory

AU - Thiel, Christian

AU - Pathe, Carsten

AU - Schmullius, Chris

AU - Seifert, Frank Martin

AU - Tansey, Kevin

AU - Balzter, Heiko

PY - 2019/1/3

Y1 - 2019/1/3

N2 - The amount and spatial distribution of forest aboveground biomass (AGB) were estimated using a range of regionally developed methods using Earth Observation data for Poland, Sweden and regions in Indonesia (Kalimantan), Mexico (Central Mexico and Yucatan peninsula), and South Africa (Eastern provinces) for the year 2010. These regions are representative of numerous forest biomes and biomass levels globally, from South African woodlands and savannas to the humid tropical forest of Kalimantan. AGB retrieval in each region relied on different sources of reference data, including forest inventory plot data and airborne LiDAR observations, and used a range of retrieval algorithms. This is the widest inter-comparison of regional-to-national AGB maps to date in terms of area, forest types, input datasets, and retrieval methods. The accuracy assessment of all regional maps using independent field data or LiDAR AGB maps resulted in an overall root mean square error (RMSE) ranging from 10 t ha−1 to 55 t ha−1 (37% to 67% relative RMSE), and an overall bias ranging from −1 t ha−1 to +5 t ha−1 at pixel level. The regional maps showed better agreement with field data than previously developed and widely used pan-tropical or northern hemisphere datasets. The comparison of accuracy assessments showed commonalities in error structures despite the variety of methods, input data, and forest biomes. All regional retrievals resulted in overestimation (up to 63 t ha−1) in the lower AGB classes, and underestimation (up to 85 t ha−1) in the higher AGB classes. Parametric model-based algorithms present advantages due to their low demand on in situ data compared to non-parametric algorithms, but there is a need for datasets and retrieval methods that can overcome the biases at both ends of the AGB range. The outcomes of this study should be considered when developing algorithms to estimate forest biomass at continental to global scale level.

AB - The amount and spatial distribution of forest aboveground biomass (AGB) were estimated using a range of regionally developed methods using Earth Observation data for Poland, Sweden and regions in Indonesia (Kalimantan), Mexico (Central Mexico and Yucatan peninsula), and South Africa (Eastern provinces) for the year 2010. These regions are representative of numerous forest biomes and biomass levels globally, from South African woodlands and savannas to the humid tropical forest of Kalimantan. AGB retrieval in each region relied on different sources of reference data, including forest inventory plot data and airborne LiDAR observations, and used a range of retrieval algorithms. This is the widest inter-comparison of regional-to-national AGB maps to date in terms of area, forest types, input datasets, and retrieval methods. The accuracy assessment of all regional maps using independent field data or LiDAR AGB maps resulted in an overall root mean square error (RMSE) ranging from 10 t ha−1 to 55 t ha−1 (37% to 67% relative RMSE), and an overall bias ranging from −1 t ha−1 to +5 t ha−1 at pixel level. The regional maps showed better agreement with field data than previously developed and widely used pan-tropical or northern hemisphere datasets. The comparison of accuracy assessments showed commonalities in error structures despite the variety of methods, input data, and forest biomes. All regional retrievals resulted in overestimation (up to 63 t ha−1) in the lower AGB classes, and underestimation (up to 85 t ha−1) in the higher AGB classes. Parametric model-based algorithms present advantages due to their low demand on in situ data compared to non-parametric algorithms, but there is a need for datasets and retrieval methods that can overcome the biases at both ends of the AGB range. The outcomes of this study should be considered when developing algorithms to estimate forest biomass at continental to global scale level.

KW - Remote sensing

KW - biomass

UR - http://www.scopus.com/inward/record.url?scp=85062857698&partnerID=8YFLogxK

U2 - 10.1016/j.jag.2018.12.008

DO - 10.1016/j.jag.2018.12.008

M3 - Article

VL - 77

SP - 53

EP - 68

JO - International Journal of Applied Earth Observation and Geoinformation

JF - International Journal of Applied Earth Observation and Geoinformation

SN - 1569-8432

ER -