Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data

Jesús Aguirre-Gutiérrez; Sami Rifai; Alexander Shenkin; Imma Oliveras; Lisa Patrick Bentley; Martin Svátek; Cécile A.J. Girardin; Sabine Both; Terhi Riutta; Erika Berenguer; W. Daniel Kissling; David Bauman; Nicolas Raab; Sam Moore; William Farfan-Rios; Axa Emanuelle Simões Figueiredo; Simone Matias Reis; Josué Edzang Ndong; Fidèle Evouna Ondo; Natacha N'ssi Bengone; Vianet Mihindou; Marina Maria Moraes de Seixas; Stephen Adu-Bredu; Katharine Abernethy; Gregory P. Asner; Jos Barlow; David F.R.P. Burslem; David A. Coomes; Lucas A. Cernusak; Greta C. Dargie; Brian J. Enquist; Robert M. Ewers; Joice Ferreira; Kathryn J. Jeffery; Carlos A. Joly; Simon L. Lewis; Ben Hur Marimon-Junior; Roberta E. Martin; Paulo S. Morandi; Oliver L. Phillips; Carlos A. Quesada; Norma Salinas; Beatriz Schwantes Marimon; Miles Silman; Yit Arn Teh; Lee J.T. White; Yadvinder Malhi

doi:10.1016/j.rse.2020.112122

Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data

Jesús Aguirre-Gutiérrez^* (Corresponding Author), Sami Rifai, Alexander Shenkin, Imma Oliveras, Lisa Patrick Bentley, Martin Svátek, Cécile A.J. Girardin, Sabine Both, Terhi Riutta, Erika Berenguer, W. Daniel Kissling, David Bauman, Nicolas Raab, Sam Moore, William Farfan-Rios, Axa Emanuelle Simões Figueiredo, Simone Matias Reis, Josué Edzang Ndong, Fidèle Evouna Ondo, Natacha N'ssi BengoneVianet Mihindou, Marina Maria Moraes de Seixas, Stephen Adu-Bredu, Katharine Abernethy, Gregory P. Asner, Jos Barlow, David F.R.P. Burslem, David A. Coomes, Lucas A. Cernusak, Greta C. Dargie, Brian J. Enquist, Robert M. Ewers, Joice Ferreira, Kathryn J. Jeffery, Carlos A. Joly, Simon L. Lewis, Ben Hur Marimon-Junior, Roberta E. Martin, Paulo S. Morandi, Oliver L. Phillips, Carlos A. Quesada, Norma Salinas, Beatriz Schwantes Marimon, Miles Silman, Yit Arn Teh, Lee J.T. White, Yadvinder Malhi

^*Corresponding author for this work

Aberdeen Centre For Environmental Sustainability

Research output: Contribution to journal › Article › peer-review

34 Citations (Scopus)

Abstract

Tropical forest ecosystems are undergoing rapid transformation as a result of changing environmental conditions and direct human impacts. However, we cannot adequately understand, monitor or simulate tropical ecosystem responses to environmental changes without capturing the high diversity of plant functional characteristics in the species-rich tropics. Failure to do so can oversimplify our understanding of ecosystems responses to environmental disturbances. Innovative methods and data products are needed to track changes in functional trait composition in tropical forest ecosystems through time and space. This study aimed to track key functional traits by coupling Sentinel-2 derived variables with a unique data set of precisely located in-situ measurements of canopy functional traits collected from 2434 individual trees across the tropics using a standardised methodology. The functional traits and vegetation censuses were collected from 47 field plots in the countries of Australia, Brazil, Peru, Gabon, Ghana, and Malaysia, which span the four tropical continents. The spatial positions of individual trees above 10 cm diameter at breast height (DBH) were mapped and their canopy size and shape recorded. Using geo-located tree canopy size and shape data, community-level trait values were estimated at the same spatial resolution as Sentinel-2 imagery (i.e. 10 m pixels). We then used the Geographic Random Forest (GRF) to model and predict functional traits across our plots. We demonstrate that key plant functional traits can be accurately predicted across the tropicsusing the high spatial and spectral resolution of Sentinel-2 imagery in conjunction with climatic and soil information. Image textural parameters were found to be key components of remote sensing information for predicting functional traits across tropical forests and woody savannas. Leaf thickness (R² = 0.52) obtained the highest prediction accuracy among the morphological and structural traits and leaf carbon content (R² = 0.70) and maximum rates of photosynthesis (R² = 0.67) obtained the highest prediction accuracy for leaf chemistry and photosynthesis related traits, respectively. Overall, the highest prediction accuracy was obtained for leaf chemistry and photosynthetic traits in comparison to morphological and structural traits. Our approach offers new opportunities for mapping, monitoring and understanding biodiversity and ecosystem change in the most species-rich ecosystems on Earth.

Original language	English
Article number	112122
Number of pages	18
Journal	Remote Sensing of Environment
Volume	252
Early online date	14 Oct 2020
DOIs	https://doi.org/10.1016/j.rse.2020.112122
Publication status	Published - Jan 2021

Bibliographical note

Funding Information:
This work is a product of the Global Ecosystems Monitoring (GEM) network (gem.tropicalforests.ox.ac.uk). J.A.G. was funded by the Natural Environment Research Council (NERC; NE/T011084/1 and NE/S011811/1) and the Netherlands Organisation for Scientific Research (NWO) under the Rubicon programme with project number 019.162LW.010. The traits field campaign was funded by a grant to Y.M. from the European Research Council (Advanced Grant GEM-TRAIT: 321131) under the European Union‘s Seventh Framework Programme (FP7/2007-2013), with additional support from NERC Grant NE/D014174/1 and NE/J022616/1 for traits work in Peru, NERC Grant ECOFOR (NE/K016385/1) for traits work in Santarem, NERC Grant BALI (NE/K016369/1) for plot and traits work in Malaysia and ERC Advanced Grant T-FORCES (291585) to Phillips for traits work in Australia. Plot setup in Ghana and Gabon were funded by a NERC Grant NE/I014705/1 and by the Royal Society-Leverhulme Africa Capacity Building Programme. The Malaysia campaign was also funded by NERC GrantNE/K016253/1. Plot inventories in Peru were supported by funding from the US National Science Foundation Long-Term Research in Environmental Biology program (LTREB; DEB 1754647) and the Gordon and Betty Moore Foundation Andes-Amazon Program. Plots inventories in Nova Xavantina (Brazil) were supported by the National Council for Scientific and Technological Development (CNPq), Long Term Ecological Research Program (PELD), Proc. 441244/2016-5, and the Foundation of Research Support of Mato Grosso (FAPEMAT), Project ReFlor, Proc. 589267/2016. During data collection, I.O. was supported by a Marie Curie Fellowship (FP7-PEOPLE-2012-IEF-327990). GEM trait data in Gabon was collected under authorisation to Y.M. and supported by the Gabon National Parks Agency. D.B. was funded by the Fondation Wiener-Anspach. W.D.K. acknowledges support from the Faculty Research Cluster ‘Global Ecology’ of the University of Amsterdam. M.S. was funded by a grant from the Ministry of Education, Youth and Sports of the Czech Republic (INTER-TRANSFER LTT19018). Y.M. is supported by the Jackson Foundation. We thank the two anonymous reviewers and Associate Editor G. Henebry for their insightful comments that helped improved this manuscript.

Data Availability Statement

Supplementary data to this article can be found online.

Keywords

Image texture
Pixel-level predictions
Plant traits
Random Forest
Sentinel-2
Tropical forests

Access to Document

10.1016/j.rse.2020.112122Licence: Unspecified

Aguirre-Gutiérrez_etal_RSE_Pantropical_Modelling_Of_AAM
© 2020 Elsevier Inc. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/(opens in new tab/window)
Accepted author manuscript, 1.54 MBLicence: CC BY-NC-ND

Cite this

Aguirre-Gutiérrez, J., Rifai, S., Shenkin, A., Oliveras, I., Bentley, L. P., Svátek, M., Girardin, C. A. J., Both, S., Riutta, T., Berenguer, E., Kissling, W. D., Bauman, D., Raab, N., Moore, S., Farfan-Rios, W., Figueiredo, A. E. S., Reis, S. M., Ndong, J. E., Ondo, F. E., ... Malhi, Y. (2021). Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data. Remote Sensing of Environment, 252, Article 112122. https://doi.org/10.1016/j.rse.2020.112122

Aguirre-Gutiérrez, J, Rifai, S, Shenkin, A, Oliveras, I, Bentley, LP, Svátek, M, Girardin, CAJ, Both, S, Riutta, T, Berenguer, E, Kissling, WD, Bauman, D, Raab, N, Moore, S, Farfan-Rios, W, Figueiredo, AES, Reis, SM, Ndong, JE, Ondo, FE, N'ssi Bengone, N, Mihindou, V, Moraes de Seixas, MM, Adu-Bredu, S, Abernethy, K, Asner, GP, Barlow, J, Burslem, DFRP, Coomes, DA, Cernusak, LA, Dargie, GC, Enquist, BJ, Ewers, RM, Ferreira, J, Jeffery, KJ, Joly, CA, Lewis, SL, Marimon-Junior, BH, Martin, RE, Morandi, PS, Phillips, OL, Quesada, CA, Salinas, N, Schwantes Marimon, B, Silman, M, Teh, YA, White, LJT & Malhi, Y 2021, 'Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data', Remote Sensing of Environment, vol. 252, 112122. https://doi.org/10.1016/j.rse.2020.112122

@article{47c78d69f2f24034ba6f05da4a39d4b0,

title = "Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data",

abstract = "Tropical forest ecosystems are undergoing rapid transformation as a result of changing environmental conditions and direct human impacts. However, we cannot adequately understand, monitor or simulate tropical ecosystem responses to environmental changes without capturing the high diversity of plant functional characteristics in the species-rich tropics. Failure to do so can oversimplify our understanding of ecosystems responses to environmental disturbances. Innovative methods and data products are needed to track changes in functional trait composition in tropical forest ecosystems through time and space. This study aimed to track key functional traits by coupling Sentinel-2 derived variables with a unique data set of precisely located in-situ measurements of canopy functional traits collected from 2434 individual trees across the tropics using a standardised methodology. The functional traits and vegetation censuses were collected from 47 field plots in the countries of Australia, Brazil, Peru, Gabon, Ghana, and Malaysia, which span the four tropical continents. The spatial positions of individual trees above 10 cm diameter at breast height (DBH) were mapped and their canopy size and shape recorded. Using geo-located tree canopy size and shape data, community-level trait values were estimated at the same spatial resolution as Sentinel-2 imagery (i.e. 10 m pixels). We then used the Geographic Random Forest (GRF) to model and predict functional traits across our plots. We demonstrate that key plant functional traits can be accurately predicted across the tropicsusing the high spatial and spectral resolution of Sentinel-2 imagery in conjunction with climatic and soil information. Image textural parameters were found to be key components of remote sensing information for predicting functional traits across tropical forests and woody savannas. Leaf thickness (R2 = 0.52) obtained the highest prediction accuracy among the morphological and structural traits and leaf carbon content (R2 = 0.70) and maximum rates of photosynthesis (R2 = 0.67) obtained the highest prediction accuracy for leaf chemistry and photosynthesis related traits, respectively. Overall, the highest prediction accuracy was obtained for leaf chemistry and photosynthetic traits in comparison to morphological and structural traits. Our approach offers new opportunities for mapping, monitoring and understanding biodiversity and ecosystem change in the most species-rich ecosystems on Earth.",

keywords = "Image texture, Pixel-level predictions, Plant traits, Random Forest, Sentinel-2, Tropical forests",

author = "Jes{\'u}s Aguirre-Guti{\'e}rrez and Sami Rifai and Alexander Shenkin and Imma Oliveras and Bentley, {Lisa Patrick} and Martin Sv{\'a}tek and Girardin, {C{\'e}cile A.J.} and Sabine Both and Terhi Riutta and Erika Berenguer and Kissling, {W. Daniel} and David Bauman and Nicolas Raab and Sam Moore and William Farfan-Rios and Figueiredo, {Axa Emanuelle Sim{\~o}es} and Reis, {Simone Matias} and Ndong, {Josu{\'e} Edzang} and Ondo, {Fid{\`e}le Evouna} and {N'ssi Bengone}, Natacha and Vianet Mihindou and {Moraes de Seixas}, {Marina Maria} and Stephen Adu-Bredu and Katharine Abernethy and Asner, {Gregory P.} and Jos Barlow and Burslem, {David F.R.P.} and Coomes, {David A.} and Cernusak, {Lucas A.} and Dargie, {Greta C.} and Enquist, {Brian J.} and Ewers, {Robert M.} and Joice Ferreira and Jeffery, {Kathryn J.} and Joly, {Carlos A.} and Lewis, {Simon L.} and Marimon-Junior, {Ben Hur} and Martin, {Roberta E.} and Morandi, {Paulo S.} and Phillips, {Oliver L.} and Quesada, {Carlos A.} and Norma Salinas and {Schwantes Marimon}, Beatriz and Miles Silman and Teh, {Yit Arn} and White, {Lee J.T.} and Yadvinder Malhi",

note = "Funding Information: This work is a product of the Global Ecosystems Monitoring (GEM) network (gem.tropicalforests.ox.ac.uk). J.A.G. was funded by the Natural Environment Research Council (NERC; NE/T011084/1 and NE/S011811/1) and the Netherlands Organisation for Scientific Research (NWO) under the Rubicon programme with project number 019.162LW.010. The traits field campaign was funded by a grant to Y.M. from the European Research Council (Advanced Grant GEM-TRAIT: 321131) under the European Union{\textquoteleft}s Seventh Framework Programme (FP7/2007-2013), with additional support from NERC Grant NE/D014174/1 and NE/J022616/1 for traits work in Peru, NERC Grant ECOFOR (NE/K016385/1) for traits work in Santarem, NERC Grant BALI (NE/K016369/1) for plot and traits work in Malaysia and ERC Advanced Grant T-FORCES (291585) to Phillips for traits work in Australia. Plot setup in Ghana and Gabon were funded by a NERC Grant NE/I014705/1 and by the Royal Society-Leverhulme Africa Capacity Building Programme. The Malaysia campaign was also funded by NERC GrantNE/K016253/1. Plot inventories in Peru were supported by funding from the US National Science Foundation Long-Term Research in Environmental Biology program (LTREB; DEB 1754647) and the Gordon and Betty Moore Foundation Andes-Amazon Program. Plots inventories in Nova Xavantina (Brazil) were supported by the National Council for Scientific and Technological Development (CNPq), Long Term Ecological Research Program (PELD), Proc. 441244/2016-5, and the Foundation of Research Support of Mato Grosso (FAPEMAT), Project ReFlor, Proc. 589267/2016. During data collection, I.O. was supported by a Marie Curie Fellowship (FP7-PEOPLE-2012-IEF-327990). GEM trait data in Gabon was collected under authorisation to Y.M. and supported by the Gabon National Parks Agency. D.B. was funded by the Fondation Wiener-Anspach. W.D.K. acknowledges support from the Faculty Research Cluster {\textquoteleft}Global Ecology{\textquoteright} of the University of Amsterdam. M.S. was funded by a grant from the Ministry of Education, Youth and Sports of the Czech Republic (INTER-TRANSFER LTT19018). Y.M. is supported by the Jackson Foundation. We thank the two anonymous reviewers and Associate Editor G. Henebry for their insightful comments that helped improved this manuscript. ",

year = "2021",

month = jan,

doi = "10.1016/j.rse.2020.112122",

language = "English",

volume = "252",

journal = "Remote Sensing of Environment",

issn = "0034-4257",

publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data

AU - Aguirre-Gutiérrez, Jesús

AU - Rifai, Sami

AU - Shenkin, Alexander

AU - Oliveras, Imma

AU - Bentley, Lisa Patrick

AU - Svátek, Martin

AU - Girardin, Cécile A.J.

AU - Both, Sabine

AU - Riutta, Terhi

AU - Berenguer, Erika

AU - Kissling, W. Daniel

AU - Bauman, David

AU - Raab, Nicolas

AU - Moore, Sam

AU - Farfan-Rios, William

AU - Figueiredo, Axa Emanuelle Simões

AU - Reis, Simone Matias

AU - Ndong, Josué Edzang

AU - Ondo, Fidèle Evouna

AU - N'ssi Bengone, Natacha

AU - Mihindou, Vianet

AU - Moraes de Seixas, Marina Maria

AU - Adu-Bredu, Stephen

AU - Abernethy, Katharine

AU - Asner, Gregory P.

AU - Barlow, Jos

AU - Burslem, David F.R.P.

AU - Coomes, David A.

AU - Cernusak, Lucas A.

AU - Dargie, Greta C.

AU - Enquist, Brian J.

AU - Ewers, Robert M.

AU - Ferreira, Joice

AU - Jeffery, Kathryn J.

AU - Joly, Carlos A.

AU - Lewis, Simon L.

AU - Marimon-Junior, Ben Hur

AU - Martin, Roberta E.

AU - Morandi, Paulo S.

AU - Phillips, Oliver L.

AU - Quesada, Carlos A.

AU - Salinas, Norma

AU - Schwantes Marimon, Beatriz

AU - Silman, Miles

AU - Teh, Yit Arn

AU - White, Lee J.T.

AU - Malhi, Yadvinder

N1 - Funding Information: This work is a product of the Global Ecosystems Monitoring (GEM) network (gem.tropicalforests.ox.ac.uk). J.A.G. was funded by the Natural Environment Research Council (NERC; NE/T011084/1 and NE/S011811/1) and the Netherlands Organisation for Scientific Research (NWO) under the Rubicon programme with project number 019.162LW.010. The traits field campaign was funded by a grant to Y.M. from the European Research Council (Advanced Grant GEM-TRAIT: 321131) under the European Union‘s Seventh Framework Programme (FP7/2007-2013), with additional support from NERC Grant NE/D014174/1 and NE/J022616/1 for traits work in Peru, NERC Grant ECOFOR (NE/K016385/1) for traits work in Santarem, NERC Grant BALI (NE/K016369/1) for plot and traits work in Malaysia and ERC Advanced Grant T-FORCES (291585) to Phillips for traits work in Australia. Plot setup in Ghana and Gabon were funded by a NERC Grant NE/I014705/1 and by the Royal Society-Leverhulme Africa Capacity Building Programme. The Malaysia campaign was also funded by NERC GrantNE/K016253/1. Plot inventories in Peru were supported by funding from the US National Science Foundation Long-Term Research in Environmental Biology program (LTREB; DEB 1754647) and the Gordon and Betty Moore Foundation Andes-Amazon Program. Plots inventories in Nova Xavantina (Brazil) were supported by the National Council for Scientific and Technological Development (CNPq), Long Term Ecological Research Program (PELD), Proc. 441244/2016-5, and the Foundation of Research Support of Mato Grosso (FAPEMAT), Project ReFlor, Proc. 589267/2016. During data collection, I.O. was supported by a Marie Curie Fellowship (FP7-PEOPLE-2012-IEF-327990). GEM trait data in Gabon was collected under authorisation to Y.M. and supported by the Gabon National Parks Agency. D.B. was funded by the Fondation Wiener-Anspach. W.D.K. acknowledges support from the Faculty Research Cluster ‘Global Ecology’ of the University of Amsterdam. M.S. was funded by a grant from the Ministry of Education, Youth and Sports of the Czech Republic (INTER-TRANSFER LTT19018). Y.M. is supported by the Jackson Foundation. We thank the two anonymous reviewers and Associate Editor G. Henebry for their insightful comments that helped improved this manuscript.

PY - 2021/1

Y1 - 2021/1

N2 - Tropical forest ecosystems are undergoing rapid transformation as a result of changing environmental conditions and direct human impacts. However, we cannot adequately understand, monitor or simulate tropical ecosystem responses to environmental changes without capturing the high diversity of plant functional characteristics in the species-rich tropics. Failure to do so can oversimplify our understanding of ecosystems responses to environmental disturbances. Innovative methods and data products are needed to track changes in functional trait composition in tropical forest ecosystems through time and space. This study aimed to track key functional traits by coupling Sentinel-2 derived variables with a unique data set of precisely located in-situ measurements of canopy functional traits collected from 2434 individual trees across the tropics using a standardised methodology. The functional traits and vegetation censuses were collected from 47 field plots in the countries of Australia, Brazil, Peru, Gabon, Ghana, and Malaysia, which span the four tropical continents. The spatial positions of individual trees above 10 cm diameter at breast height (DBH) were mapped and their canopy size and shape recorded. Using geo-located tree canopy size and shape data, community-level trait values were estimated at the same spatial resolution as Sentinel-2 imagery (i.e. 10 m pixels). We then used the Geographic Random Forest (GRF) to model and predict functional traits across our plots. We demonstrate that key plant functional traits can be accurately predicted across the tropicsusing the high spatial and spectral resolution of Sentinel-2 imagery in conjunction with climatic and soil information. Image textural parameters were found to be key components of remote sensing information for predicting functional traits across tropical forests and woody savannas. Leaf thickness (R2 = 0.52) obtained the highest prediction accuracy among the morphological and structural traits and leaf carbon content (R2 = 0.70) and maximum rates of photosynthesis (R2 = 0.67) obtained the highest prediction accuracy for leaf chemistry and photosynthesis related traits, respectively. Overall, the highest prediction accuracy was obtained for leaf chemistry and photosynthetic traits in comparison to morphological and structural traits. Our approach offers new opportunities for mapping, monitoring and understanding biodiversity and ecosystem change in the most species-rich ecosystems on Earth.

AB - Tropical forest ecosystems are undergoing rapid transformation as a result of changing environmental conditions and direct human impacts. However, we cannot adequately understand, monitor or simulate tropical ecosystem responses to environmental changes without capturing the high diversity of plant functional characteristics in the species-rich tropics. Failure to do so can oversimplify our understanding of ecosystems responses to environmental disturbances. Innovative methods and data products are needed to track changes in functional trait composition in tropical forest ecosystems through time and space. This study aimed to track key functional traits by coupling Sentinel-2 derived variables with a unique data set of precisely located in-situ measurements of canopy functional traits collected from 2434 individual trees across the tropics using a standardised methodology. The functional traits and vegetation censuses were collected from 47 field plots in the countries of Australia, Brazil, Peru, Gabon, Ghana, and Malaysia, which span the four tropical continents. The spatial positions of individual trees above 10 cm diameter at breast height (DBH) were mapped and their canopy size and shape recorded. Using geo-located tree canopy size and shape data, community-level trait values were estimated at the same spatial resolution as Sentinel-2 imagery (i.e. 10 m pixels). We then used the Geographic Random Forest (GRF) to model and predict functional traits across our plots. We demonstrate that key plant functional traits can be accurately predicted across the tropicsusing the high spatial and spectral resolution of Sentinel-2 imagery in conjunction with climatic and soil information. Image textural parameters were found to be key components of remote sensing information for predicting functional traits across tropical forests and woody savannas. Leaf thickness (R2 = 0.52) obtained the highest prediction accuracy among the morphological and structural traits and leaf carbon content (R2 = 0.70) and maximum rates of photosynthesis (R2 = 0.67) obtained the highest prediction accuracy for leaf chemistry and photosynthesis related traits, respectively. Overall, the highest prediction accuracy was obtained for leaf chemistry and photosynthetic traits in comparison to morphological and structural traits. Our approach offers new opportunities for mapping, monitoring and understanding biodiversity and ecosystem change in the most species-rich ecosystems on Earth.

KW - Image texture

KW - Pixel-level predictions

KW - Plant traits

KW - Random Forest

KW - Sentinel-2

KW - Tropical forests

UR - http://www.scopus.com/inward/record.url?scp=85092472356&partnerID=8YFLogxK

UR - https://eprints.lancs.ac.uk/id/eprint/149321/

U2 - 10.1016/j.rse.2020.112122

DO - 10.1016/j.rse.2020.112122

M3 - Article

AN - SCOPUS:85092472356

SN - 0034-4257

VL - 252

JO - Remote Sensing of Environment

JF - Remote Sensing of Environment

M1 - 112122

ER -

Pantropical modelling of canopy functional traits using Sentinel-2 remote sensing data

Abstract

Bibliographical note

Data Availability Statement

Keywords

Access to Document

Other files and links

Fingerprint

Cite this