Registro Completo
|
Biblioteca(s): |
Embrapa Unidades Centrais. |
|
Data corrente: |
18/11/2019 |
|
Data da última atualização: |
18/11/2019 |
|
Autoria: |
MOURA-BUENO, J. M.; DALMOLIN, R. S. D.; HORT-HEINEN, T. Z.; CANCIAN, L. C.; SCHENATO, R. B.; DOTTO, A. C.; FLORES, C. A. |
|
Afiliação: |
Jean Michel Moura-Bueno, Universidade Federal de Santa Maria - UFSM/Centro de Ciências Rurais, Departamento de Solos; Ricardo Simão Diniz Dalmolin, Universidade Federal de Santa Maria - UFSM/Centro de Ciências Rurais, Departamento de Solos; Taciara Zborowski Horst-Heinen, Universidade Federal de Santa Maria - UFSM/Centro de Ciências Rurais, Departamento de Solos; Luciano Campos Cancian, Universidade Federal de Santa Maria - UFSM/Centro de Ciências Rurais, Departamento de Solos; Ricardo Bergamo Schenato, Universidade Federal de Santa Maria - UFSM/Centro de Ciências Rurais, Departamento de Solos; André Carnieletto Dotto, Universidade de São Paulo - USP/Escola Superior de Agricultura Luiz de Queiroz - ESALQ/Departamento de Ciência do Solo; Carlos Alberto Flores, CPACT. |
|
Título: |
Prediction of soil classes in a complex landscape in Southern Brazil. |
|
Ano de publicação: |
2019 |
|
Fonte/Imprenta: |
Pesquisa Agropecuária Brasileira v. 54, e00420, jan./dez. 2019. |
|
Idioma: |
Inglês |
|
Notas: |
Título em português: Predição de classes de solo em uma paisagem complexa no Sul do Brasil. |
|
Conteúdo: |
The objective of this work was to evaluate the use of covariate selection by expert knowledge on the performance of soil class predictive models in a complex landscape, in order to identify the best predictive model for digital soil mapping in the Southern region of Brazil. A total of 164 points were sampled in the field using the conditioned Latin hypercube, considering the covariates elevation, slope, and aspect. From the digital elevation model, environmental covariates were extracted, composing three sets, made up of: 21 covariates, covariates after the exclusion of the multicollinear ones, and covariates chosen by expert knowledge. Prediction was performed with the following models: decision tree, random forest, multiple logistic regression, and support vector machine. The accuracy of the models was evaluated by the kappa index (K), general accuracy (GA), and class accuracy. The prediction models were sensitive to the disproportionate sampling of soil classes. The best predicted map achieved a GA of 71% and K of 0.59. The use of the covariate set chosen by expert knowledge improves model performance in predicting soil classes in a complex landscape, and random forest is the best model for the spatial prediction of soil classes. |
|
Palavras-Chave: |
Covariável preditora; Digital soil mapping; Mapeamento digital de solo; Modelo preditivo; Pedometria; Pedometry; Predictive covariates; Relação solo paisagem; Soil landscape relationship. |
|
Categoria do assunto: |
-- |
|
URL: |
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/204885/1/Prediction-of-soil-classes.pdf
|
|
Marc: |
LEADER 02268naa a2200313 a 4500
001 2114512
005 2019-11-18
008 2019 bl uuuu u00u1 u #d
100 1 $aMOURA-BUENO, J. M.
245 $aPrediction of soil classes in a complex landscape in Southern Brazil.$h[electronic resource]
260 $c2019
500 $aTítulo em português: Predição de classes de solo em uma paisagem complexa no Sul do Brasil.
520 $aThe objective of this work was to evaluate the use of covariate selection by expert knowledge on the performance of soil class predictive models in a complex landscape, in order to identify the best predictive model for digital soil mapping in the Southern region of Brazil. A total of 164 points were sampled in the field using the conditioned Latin hypercube, considering the covariates elevation, slope, and aspect. From the digital elevation model, environmental covariates were extracted, composing three sets, made up of: 21 covariates, covariates after the exclusion of the multicollinear ones, and covariates chosen by expert knowledge. Prediction was performed with the following models: decision tree, random forest, multiple logistic regression, and support vector machine. The accuracy of the models was evaluated by the kappa index (K), general accuracy (GA), and class accuracy. The prediction models were sensitive to the disproportionate sampling of soil classes. The best predicted map achieved a GA of 71% and K of 0.59. The use of the covariate set chosen by expert knowledge improves model performance in predicting soil classes in a complex landscape, and random forest is the best model for the spatial prediction of soil classes.
653 $aCovariável preditora
653 $aDigital soil mapping
653 $aMapeamento digital de solo
653 $aModelo preditivo
653 $aPedometria
653 $aPedometry
653 $aPredictive covariates
653 $aRelação solo paisagem
653 $aSoil landscape relationship
700 1 $aDALMOLIN, R. S. D.
700 1 $aHORT-HEINEN, T. Z.
700 1 $aCANCIAN, L. C.
700 1 $aSCHENATO, R. B.
700 1 $aDOTTO, A. C.
700 1 $aFLORES, C. A.
773 $tPesquisa Agropecuária Brasileira$gv. 54, e00420, jan./dez. 2019.
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Unidades Centrais (AI-SEDE) |
|
