Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State

Detalhes bibliográficos
Autor(a) principal: Santos,Eliana Elizabet dos
Data de Publicação: 2020
Outros Autores: Sena,Nathalie Cruz, Balestrin,Diego, Fernandes Filho,Elpidio Inácio, Costa,Liovando Marciano da, Zeferino,Leiliane Bozzi
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Floresta e Ambiente
Texto Completo: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2179-80872020000300119
Resumo: Abstract Fire behavior prediction models can assist environmental agencies with fire prevention and control. This study aimed to adjust a fire prediction model for the state of Minas Gerais, Brazil. Using the R program and hotspots provided by the National Institute for Space Research (INPE) for 2010, prediction of the probability of fires through the Random Forest algorithm was conducted using the Bootstrapping method. The model generated a prediction map with global kappa value of 0.65. External validation was performed with hotspots in 2015. Results showed that 58% of the hotspots are in areas with ignition probability > 50%, being 24% of them in areas with 25-50% probability, and 17% in areas with < 25% probability. These results were considered satisfactory, demonstrating that the model is suitable for predicting fires.
id UFRJ-3_69d7c65ba7c07dfb81df61ae2da16b64
oai_identifier_str oai:scielo:S2179-80872020000300119
network_acronym_str UFRJ-3
network_name_str Floresta e Ambiente
repository_id_str
spelling Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais Statefiresmodelingenvironmental monitoringAbstract Fire behavior prediction models can assist environmental agencies with fire prevention and control. This study aimed to adjust a fire prediction model for the state of Minas Gerais, Brazil. Using the R program and hotspots provided by the National Institute for Space Research (INPE) for 2010, prediction of the probability of fires through the Random Forest algorithm was conducted using the Bootstrapping method. The model generated a prediction map with global kappa value of 0.65. External validation was performed with hotspots in 2015. Results showed that 58% of the hotspots are in areas with ignition probability > 50%, being 24% of them in areas with 25-50% probability, and 17% in areas with < 25% probability. These results were considered satisfactory, demonstrating that the model is suitable for predicting fires.Instituto de Florestas da Universidade Federal Rural do Rio de Janeiro2020-01-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersiontext/htmlhttp://old.scielo.br/scielo.php?script=sci_arttext&pid=S2179-80872020000300119Floresta e Ambiente v.27 n.3 2020reponame:Floresta e Ambienteinstname:Universidade Federal do Rio de Janeiro (UFRJ)instacron:UFRJ10.1590/2179-8087.011518info:eu-repo/semantics/openAccessSantos,Eliana Elizabet dosSena,Nathalie CruzBalestrin,DiegoFernandes Filho,Elpidio InácioCosta,Liovando Marciano daZeferino,Leiliane Bozzieng2020-08-05T00:00:00Zoai:scielo:S2179-80872020000300119Revistahttps://www.floram.org/PUBhttps://old.scielo.br/oai/scielo-oai.phpfloramjournal@gmail.com||floram@ufrrj.br||2179-80871415-0980opendoar:2020-08-05T00:00Floresta e Ambiente - Universidade Federal do Rio de Janeiro (UFRJ)false
dc.title.none.fl_str_mv Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
title Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
spellingShingle Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
Santos,Eliana Elizabet dos
fires
modeling
environmental monitoring
title_short Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
title_full Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
title_fullStr Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
title_full_unstemmed Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
title_sort Prediction of Burned Areas Using the Random Forest Classifier in the Minas Gerais State
author Santos,Eliana Elizabet dos
author_facet Santos,Eliana Elizabet dos
Sena,Nathalie Cruz
Balestrin,Diego
Fernandes Filho,Elpidio Inácio
Costa,Liovando Marciano da
Zeferino,Leiliane Bozzi
author_role author
author2 Sena,Nathalie Cruz
Balestrin,Diego
Fernandes Filho,Elpidio Inácio
Costa,Liovando Marciano da
Zeferino,Leiliane Bozzi
author2_role author
author
author
author
author
dc.contributor.author.fl_str_mv Santos,Eliana Elizabet dos
Sena,Nathalie Cruz
Balestrin,Diego
Fernandes Filho,Elpidio Inácio
Costa,Liovando Marciano da
Zeferino,Leiliane Bozzi
dc.subject.por.fl_str_mv fires
modeling
environmental monitoring
topic fires
modeling
environmental monitoring
description Abstract Fire behavior prediction models can assist environmental agencies with fire prevention and control. This study aimed to adjust a fire prediction model for the state of Minas Gerais, Brazil. Using the R program and hotspots provided by the National Institute for Space Research (INPE) for 2010, prediction of the probability of fires through the Random Forest algorithm was conducted using the Bootstrapping method. The model generated a prediction map with global kappa value of 0.65. External validation was performed with hotspots in 2015. Results showed that 58% of the hotspots are in areas with ignition probability > 50%, being 24% of them in areas with 25-50% probability, and 17% in areas with < 25% probability. These results were considered satisfactory, demonstrating that the model is suitable for predicting fires.
publishDate 2020
dc.date.none.fl_str_mv 2020-01-01
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2179-80872020000300119
url http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2179-80872020000300119
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv 10.1590/2179-8087.011518
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv text/html
dc.publisher.none.fl_str_mv Instituto de Florestas da Universidade Federal Rural do Rio de Janeiro
publisher.none.fl_str_mv Instituto de Florestas da Universidade Federal Rural do Rio de Janeiro
dc.source.none.fl_str_mv Floresta e Ambiente v.27 n.3 2020
reponame:Floresta e Ambiente
instname:Universidade Federal do Rio de Janeiro (UFRJ)
instacron:UFRJ
instname_str Universidade Federal do Rio de Janeiro (UFRJ)
instacron_str UFRJ
institution UFRJ
reponame_str Floresta e Ambiente
collection Floresta e Ambiente
repository.name.fl_str_mv Floresta e Ambiente - Universidade Federal do Rio de Janeiro (UFRJ)
repository.mail.fl_str_mv floramjournal@gmail.com||floram@ufrrj.br||
_version_ 1750128143654453248

Registros relacionados