Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFLA |
Texto Completo: | http://repositorio.ufla.br/jspui/handle/1/50009 |
Resumo: | The increasing awareness of global climate change has drawn attention to the role of forests as mitigators of this process as they act as carbon sinks to the atmosphere. Understanding the process of carbon storage in forests and its drivers, as well as presenting consistent models for their estimation, is a current demand. In this sense, the aim of this study was to evaluate the performance of machine learning techniques: support vector machines (SVM) and to propose a new nonlinear model extracted from the training of an artificial neural network (ANN) in the modeling of above ground carbon stock in a secondary semideciduous forest. SVM and ANN construction and training process considered independent variables selected by stepwise: minimum DBH (diameter of breast height - 1.3 m), maximum DBH, mean DBH, total height and number of trees, all by plot. SVM and the model extracted from ANN were applied to the data set intended for validation. Both techniques presented satisfactory performance in modeling carbon stock by plot, with homogeneous distribution and low dispersion of residues and predicted values close to those observed. Analysis criteria indicated superior performance of the model extracted from the artificial neural network, which presented a mean relative error of 6.94 %, while the support vector machine presented 13.52 %, combined with lower bias values and higher correlation between predictions and observations. |
id |
UFLA_15e04a0338701e75603b6f6c7ec840c1 |
---|---|
oai_identifier_str |
oai:localhost:1/50009 |
network_acronym_str |
UFLA |
network_name_str |
Repositório Institucional da UFLA |
repository_id_str |
|
spelling |
Machine learning for carbon stock prediction in a tropical forest in Southeastern BrazilAprendizaje de máquina para la predicción de reservas de carbono en un bosque tropical en el sureste de BrasilArtificial intelligenceArtificial neural networksSupport vector machinesForest biomassCarbon stock predictionInteligência artificialRedes neurais artificiaisMáquinas vetoriais de suporteBiomassa florestalPredição do estoque de carbonoThe increasing awareness of global climate change has drawn attention to the role of forests as mitigators of this process as they act as carbon sinks to the atmosphere. Understanding the process of carbon storage in forests and its drivers, as well as presenting consistent models for their estimation, is a current demand. In this sense, the aim of this study was to evaluate the performance of machine learning techniques: support vector machines (SVM) and to propose a new nonlinear model extracted from the training of an artificial neural network (ANN) in the modeling of above ground carbon stock in a secondary semideciduous forest. SVM and ANN construction and training process considered independent variables selected by stepwise: minimum DBH (diameter of breast height - 1.3 m), maximum DBH, mean DBH, total height and number of trees, all by plot. SVM and the model extracted from ANN were applied to the data set intended for validation. Both techniques presented satisfactory performance in modeling carbon stock by plot, with homogeneous distribution and low dispersion of residues and predicted values close to those observed. Analysis criteria indicated superior performance of the model extracted from the artificial neural network, which presented a mean relative error of 6.94 %, while the support vector machine presented 13.52 %, combined with lower bias values and higher correlation between predictions and observations.Universidad Austral de Chile, Facultad de Ciencias Forestales2022-05-25T18:25:24Z2022-05-25T18:25:24Z2021info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfDANTAS, D. et al. Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil. Bosque, Valdivia, v. 42, n. 1, p. 131-140, 2021. DOI: 10.4067/S0717-92002021000100131.http://repositorio.ufla.br/jspui/handle/1/50009Bosquereponame:Repositório Institucional da UFLAinstname:Universidade Federal de Lavras (UFLA)instacron:UFLAAttribution-NonCommercial 4.0 Internationalhttp://creativecommons.org/licenses/by-nc/4.0/info:eu-repo/semantics/openAccessDantas, DanielTerra, Marcela de Castro Nunes SantosSchorr, Luis Paulo BaldisseraCalegario, Natalinoeng2022-05-25T18:25:24Zoai:localhost:1/50009Repositório InstitucionalPUBhttp://repositorio.ufla.br/oai/requestnivaldo@ufla.br || repositorio.biblioteca@ufla.bropendoar:2022-05-25T18:25:24Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA)false |
dc.title.none.fl_str_mv |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil Aprendizaje de máquina para la predicción de reservas de carbono en un bosque tropical en el sureste de Brasil |
title |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
spellingShingle |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil Dantas, Daniel Artificial intelligence Artificial neural networks Support vector machines Forest biomass Carbon stock prediction Inteligência artificial Redes neurais artificiais Máquinas vetoriais de suporte Biomassa florestal Predição do estoque de carbono |
title_short |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
title_full |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
title_fullStr |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
title_full_unstemmed |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
title_sort |
Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil |
author |
Dantas, Daniel |
author_facet |
Dantas, Daniel Terra, Marcela de Castro Nunes Santos Schorr, Luis Paulo Baldissera Calegario, Natalino |
author_role |
author |
author2 |
Terra, Marcela de Castro Nunes Santos Schorr, Luis Paulo Baldissera Calegario, Natalino |
author2_role |
author author author |
dc.contributor.author.fl_str_mv |
Dantas, Daniel Terra, Marcela de Castro Nunes Santos Schorr, Luis Paulo Baldissera Calegario, Natalino |
dc.subject.por.fl_str_mv |
Artificial intelligence Artificial neural networks Support vector machines Forest biomass Carbon stock prediction Inteligência artificial Redes neurais artificiais Máquinas vetoriais de suporte Biomassa florestal Predição do estoque de carbono |
topic |
Artificial intelligence Artificial neural networks Support vector machines Forest biomass Carbon stock prediction Inteligência artificial Redes neurais artificiais Máquinas vetoriais de suporte Biomassa florestal Predição do estoque de carbono |
description |
The increasing awareness of global climate change has drawn attention to the role of forests as mitigators of this process as they act as carbon sinks to the atmosphere. Understanding the process of carbon storage in forests and its drivers, as well as presenting consistent models for their estimation, is a current demand. In this sense, the aim of this study was to evaluate the performance of machine learning techniques: support vector machines (SVM) and to propose a new nonlinear model extracted from the training of an artificial neural network (ANN) in the modeling of above ground carbon stock in a secondary semideciduous forest. SVM and ANN construction and training process considered independent variables selected by stepwise: minimum DBH (diameter of breast height - 1.3 m), maximum DBH, mean DBH, total height and number of trees, all by plot. SVM and the model extracted from ANN were applied to the data set intended for validation. Both techniques presented satisfactory performance in modeling carbon stock by plot, with homogeneous distribution and low dispersion of residues and predicted values close to those observed. Analysis criteria indicated superior performance of the model extracted from the artificial neural network, which presented a mean relative error of 6.94 %, while the support vector machine presented 13.52 %, combined with lower bias values and higher correlation between predictions and observations. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021 2022-05-25T18:25:24Z 2022-05-25T18:25:24Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
DANTAS, D. et al. Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil. Bosque, Valdivia, v. 42, n. 1, p. 131-140, 2021. DOI: 10.4067/S0717-92002021000100131. http://repositorio.ufla.br/jspui/handle/1/50009 |
identifier_str_mv |
DANTAS, D. et al. Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil. Bosque, Valdivia, v. 42, n. 1, p. 131-140, 2021. DOI: 10.4067/S0717-92002021000100131. |
url |
http://repositorio.ufla.br/jspui/handle/1/50009 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
Attribution-NonCommercial 4.0 International http://creativecommons.org/licenses/by-nc/4.0/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution-NonCommercial 4.0 International http://creativecommons.org/licenses/by-nc/4.0/ |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidad Austral de Chile, Facultad de Ciencias Forestales |
publisher.none.fl_str_mv |
Universidad Austral de Chile, Facultad de Ciencias Forestales |
dc.source.none.fl_str_mv |
Bosque reponame:Repositório Institucional da UFLA instname:Universidade Federal de Lavras (UFLA) instacron:UFLA |
instname_str |
Universidade Federal de Lavras (UFLA) |
instacron_str |
UFLA |
institution |
UFLA |
reponame_str |
Repositório Institucional da UFLA |
collection |
Repositório Institucional da UFLA |
repository.name.fl_str_mv |
Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA) |
repository.mail.fl_str_mv |
nivaldo@ufla.br || repositorio.biblioteca@ufla.br |
_version_ |
1807835093994045440 |