Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1016/j.rse.2020.112025 http://hdl.handle.net/11449/199243 |
Resumo: | Fire severity assessment is crucial for predicting ecosystem response and prioritizing post-fire forest management strategies. Although a variety of remote sensing approaches have been developed, more research is still needed to improve the accuracy and effectiveness of fire severity mapping. This study proposes a unitemporal simulation approach based on the generation of synthetic spectral databases from linear spectral mixing. To fully exploit the potential of these training databases, the Random Forest (RF) machine learning algorithm was applied to build a classifier and regression model. The predictive models parameterized with the synthetic datasets were applied in a case study, the Sierra de Luna wildfire in Spain. Single date Landsat-8 and Sentinel-2A imagery of the immediate post-fire environment were used to develop the validation spectral datasets and a Pléiades orthoimage, providing the ground truth data. The four defined severity categories – unburned (UB), partial canopy unburned (PCU), canopy scorched (CS), and canopy consumed (CC) – demonstrated high accuracy in the bootstrapped (about 95%) and real validation sets (about 90%), with a slightly better performance observed when the Sentinel-2A dataset was used. Abundance of four ground covers (green vegetation, non-photosynthetic vegetation, soil, and ash) was also quantified with moderate (~45% for NPV) or high accuracy (higher than 75% for the remaining covers). No specific pattern in the comparison of sensors was observed. Variable importance analysis highlighted the complementary behavior of the spectral bands, although the contrast between the near and shortwave infrared regions stood out above the rest. Comparison of procedures reinforced the usefulness of the approach, as RF image-derived models and the multiple endmember spectral unmixing technique (MESMA) showed lower accuracy. The capabilities for detailed mapping are reflected in the development of different types of cartography (classification maps and fraction cover maps). The approach holds great potential for fire severity assessment, and future research needs to extend the predictive modeling to other burned areas – also in different ecosystems – and analyze its competence and the possible adaptations needed. |
id |
UNSP_d2e9e97f1a53d68e45653bd8e83e08a8 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/199243 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random ForestsFire severityLandsat-8Linear spectral mixingMachine learningPost-fire ground coversSentinel-2Fire severity assessment is crucial for predicting ecosystem response and prioritizing post-fire forest management strategies. Although a variety of remote sensing approaches have been developed, more research is still needed to improve the accuracy and effectiveness of fire severity mapping. This study proposes a unitemporal simulation approach based on the generation of synthetic spectral databases from linear spectral mixing. To fully exploit the potential of these training databases, the Random Forest (RF) machine learning algorithm was applied to build a classifier and regression model. The predictive models parameterized with the synthetic datasets were applied in a case study, the Sierra de Luna wildfire in Spain. Single date Landsat-8 and Sentinel-2A imagery of the immediate post-fire environment were used to develop the validation spectral datasets and a Pléiades orthoimage, providing the ground truth data. The four defined severity categories – unburned (UB), partial canopy unburned (PCU), canopy scorched (CS), and canopy consumed (CC) – demonstrated high accuracy in the bootstrapped (about 95%) and real validation sets (about 90%), with a slightly better performance observed when the Sentinel-2A dataset was used. Abundance of four ground covers (green vegetation, non-photosynthetic vegetation, soil, and ash) was also quantified with moderate (~45% for NPV) or high accuracy (higher than 75% for the remaining covers). No specific pattern in the comparison of sensors was observed. Variable importance analysis highlighted the complementary behavior of the spectral bands, although the contrast between the near and shortwave infrared regions stood out above the rest. Comparison of procedures reinforced the usefulness of the approach, as RF image-derived models and the multiple endmember spectral unmixing technique (MESMA) showed lower accuracy. The capabilities for detailed mapping are reflected in the development of different types of cartography (classification maps and fraction cover maps). The approach holds great potential for fire severity assessment, and future research needs to extend the predictive modeling to other burned areas – also in different ecosystems – and analyze its competence and the possible adaptations needed.Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Department of Geography and Spatial Management University of Zaragoza, C/Pedro Cerbuna 12GEOFOREST-IUCA research group Environmental Sciences Institute (IUCA) University of Zaragoza, C/Pedro Cerbuna 12Lab of Vegetation Ecology Instituto de Biociências Universidade Estadual Paulista (UNESP), Avenida 24-A 1515Centro Universitario de la Defensa de Zaragoza Academia General Militar, Ctra. Huesca s/nLab of Vegetation Ecology Instituto de Biociências Universidade Estadual Paulista (UNESP), Avenida 24-A 1515FAPESP: 2019/07357-8University of ZaragozaUniversidade Estadual Paulista (Unesp)Academia General MilitarMontorio, RaquelPérez-Cabello, FernandoBorini Alves, Daniel [UNESP]García-Martín, Alberto2020-12-12T01:34:32Z2020-12-12T01:34:32Z2020-11-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1016/j.rse.2020.112025Remote Sensing of Environment, v. 249.0034-4257http://hdl.handle.net/11449/19924310.1016/j.rse.2020.1120252-s2.0-85089268347Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengRemote Sensing of Environmentinfo:eu-repo/semantics/openAccess2021-10-22T18:20:45Zoai:repositorio.unesp.br:11449/199243Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462021-10-22T18:20:45Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
title |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
spellingShingle |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests Montorio, Raquel Fire severity Landsat-8 Linear spectral mixing Machine learning Post-fire ground covers Sentinel-2 |
title_short |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
title_full |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
title_fullStr |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
title_full_unstemmed |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
title_sort |
Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests |
author |
Montorio, Raquel |
author_facet |
Montorio, Raquel Pérez-Cabello, Fernando Borini Alves, Daniel [UNESP] García-Martín, Alberto |
author_role |
author |
author2 |
Pérez-Cabello, Fernando Borini Alves, Daniel [UNESP] García-Martín, Alberto |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
University of Zaragoza Universidade Estadual Paulista (Unesp) Academia General Militar |
dc.contributor.author.fl_str_mv |
Montorio, Raquel Pérez-Cabello, Fernando Borini Alves, Daniel [UNESP] García-Martín, Alberto |
dc.subject.por.fl_str_mv |
Fire severity Landsat-8 Linear spectral mixing Machine learning Post-fire ground covers Sentinel-2 |
topic |
Fire severity Landsat-8 Linear spectral mixing Machine learning Post-fire ground covers Sentinel-2 |
description |
Fire severity assessment is crucial for predicting ecosystem response and prioritizing post-fire forest management strategies. Although a variety of remote sensing approaches have been developed, more research is still needed to improve the accuracy and effectiveness of fire severity mapping. This study proposes a unitemporal simulation approach based on the generation of synthetic spectral databases from linear spectral mixing. To fully exploit the potential of these training databases, the Random Forest (RF) machine learning algorithm was applied to build a classifier and regression model. The predictive models parameterized with the synthetic datasets were applied in a case study, the Sierra de Luna wildfire in Spain. Single date Landsat-8 and Sentinel-2A imagery of the immediate post-fire environment were used to develop the validation spectral datasets and a Pléiades orthoimage, providing the ground truth data. The four defined severity categories – unburned (UB), partial canopy unburned (PCU), canopy scorched (CS), and canopy consumed (CC) – demonstrated high accuracy in the bootstrapped (about 95%) and real validation sets (about 90%), with a slightly better performance observed when the Sentinel-2A dataset was used. Abundance of four ground covers (green vegetation, non-photosynthetic vegetation, soil, and ash) was also quantified with moderate (~45% for NPV) or high accuracy (higher than 75% for the remaining covers). No specific pattern in the comparison of sensors was observed. Variable importance analysis highlighted the complementary behavior of the spectral bands, although the contrast between the near and shortwave infrared regions stood out above the rest. Comparison of procedures reinforced the usefulness of the approach, as RF image-derived models and the multiple endmember spectral unmixing technique (MESMA) showed lower accuracy. The capabilities for detailed mapping are reflected in the development of different types of cartography (classification maps and fraction cover maps). The approach holds great potential for fire severity assessment, and future research needs to extend the predictive modeling to other burned areas – also in different ecosystems – and analyze its competence and the possible adaptations needed. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020-12-12T01:34:32Z 2020-12-12T01:34:32Z 2020-11-01 |
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 |
http://dx.doi.org/10.1016/j.rse.2020.112025 Remote Sensing of Environment, v. 249. 0034-4257 http://hdl.handle.net/11449/199243 10.1016/j.rse.2020.112025 2-s2.0-85089268347 |
url |
http://dx.doi.org/10.1016/j.rse.2020.112025 http://hdl.handle.net/11449/199243 |
identifier_str_mv |
Remote Sensing of Environment, v. 249. 0034-4257 10.1016/j.rse.2020.112025 2-s2.0-85089268347 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Remote Sensing of Environment |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799965373256171520 |