|
|
Registro Completo |
Biblioteca(s): |
Embrapa Solos. |
Data corrente: |
02/04/2018 |
Data da última atualização: |
11/11/2021 |
Tipo da produção científica: |
Artigo em Periódico Indexado |
Autoria: |
PINHEIRO, H. S. K.; CARVALHO JUNIOR, W. de; CHAGAS, C. da S.; ANJOS, L. H. C. dos; OWENS, P. R. |
Afiliação: |
HELENA SARAIVA KOENOW PINHEIRO, UFRRJ; WALDIR DE CARVALHO JUNIOR, CNPS; CESAR DA SILVA CHAGAS, CNPS; LÚCIA HELENA CUNHA DOS ANJOS, UFRRJ; PHILLIP RAY OWENS, USDA. |
Título: |
Prediction of topsoil texture through regression trees and multiple linear regressions. |
Ano de publicação: |
2018 |
Fonte/Imprenta: |
Revista Brasileira de Ciência do Solo, v. 42, article e0170167, 2018. |
DOI: |
https://doi.org/10.1590/18069657rbcs20170167 |
Idioma: |
Inglês |
Conteúdo: |
Users of soil survey products are mostly interested in understanding how soil properties vary in space and time. The aim of digital soil mapping (DSM) is to represent the spatial variability of soil properties quantitatively to support decision-making. The goal of this study is to evaluate DSM techniques (Regression Trees - RT and Multiple Linear Regressions - MLR) and the ability of these tools to predict mineral fraction content under a wide variability of landscapes. The study site was the entire Guapi-Macacu watershed (1,250.78 km²) in the state of Rio de Janeiro in the Southeast region of Brazil. Terrain attributes and remote sensing data (with 30 m of spatial resolution) were used to represent landscape co-variables selected as an input in predictive models in order to develop the explanatory variables. The selection of sampling sites was based on the Latin Hypercube algorithm. A representative set of one hundred points with feasible field access was chosen. Different input databases were tested for prediction of mineral fraction content (harmonized and original data). The Spline algorithm was used to harmonize data according to the GlobalSoil.Net consortium standards. The results showed better performance from the RT models, using input from an average of six covariates; the simplest MLR model used twice as many input variables, creating more complex models without gaining precision. Furthermore, better R² values were obtained using RT models, irrespective of harmonization of soil data. The harmonized dataset from the 0.00-0.05 and 0.05-0.15 m layers, in general, presented better results for the clay and silt, with R2 values of 0.52 (0.00-0.05 m) and 0.69 (0.05-0.15 m), respectively. Prediction of sand content showed better results when the original depth data was used as an input, although all regression tree models had R2 values greater than 0.52. The RT models provided a better statistical index than MLR for all predicted properties; however, the variance between models suggests similarity of performance. Regarding harmonization of soil data, both input databases (harmonized or not) can be used to predict soil properties, since the variance of model performance was low and generalization of the soil maps showed similar trends. The products obtained from the digital soil mapping approach make it possible to integrate the factor of uncertainties, providing easier interpretation for soil management and land use decisions. MenosUsers of soil survey products are mostly interested in understanding how soil properties vary in space and time. The aim of digital soil mapping (DSM) is to represent the spatial variability of soil properties quantitatively to support decision-making. The goal of this study is to evaluate DSM techniques (Regression Trees - RT and Multiple Linear Regressions - MLR) and the ability of these tools to predict mineral fraction content under a wide variability of landscapes. The study site was the entire Guapi-Macacu watershed (1,250.78 km²) in the state of Rio de Janeiro in the Southeast region of Brazil. Terrain attributes and remote sensing data (with 30 m of spatial resolution) were used to represent landscape co-variables selected as an input in predictive models in order to develop the explanatory variables. The selection of sampling sites was based on the Latin Hypercube algorithm. A representative set of one hundred points with feasible field access was chosen. Different input databases were tested for prediction of mineral fraction content (harmonized and original data). The Spline algorithm was used to harmonize data according to the GlobalSoil.Net consortium standards. The results showed better performance from the RT models, using input from an average of six covariates; the simplest MLR model used twice as many input variables, creating more complex models without gaining precision. Furthermore, better R² values were obtained using RT models, irrespective of harmoniz... Mostrar Tudo |
Palavras-Chave: |
Atributos do terreno; Funções de profundidade do solo; Mapeamento digital do solo; Modelos de regressão. |
Categoria do assunto: |
P Recursos Naturais, Ciências Ambientais e da Terra |
URL: |
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/174786/1/2018-010.pdf
|
Marc: |
LEADER 03254naa a2200229 a 4500 001 2089994 005 2021-11-11 008 2018 bl uuuu u00u1 u #d 024 7 $ahttps://doi.org/10.1590/18069657rbcs20170167$2DOI 100 1 $aPINHEIRO, H. S. K. 245 $aPrediction of topsoil texture through regression trees and multiple linear regressions.$h[electronic resource] 260 $c2018 520 $aUsers of soil survey products are mostly interested in understanding how soil properties vary in space and time. The aim of digital soil mapping (DSM) is to represent the spatial variability of soil properties quantitatively to support decision-making. The goal of this study is to evaluate DSM techniques (Regression Trees - RT and Multiple Linear Regressions - MLR) and the ability of these tools to predict mineral fraction content under a wide variability of landscapes. The study site was the entire Guapi-Macacu watershed (1,250.78 km²) in the state of Rio de Janeiro in the Southeast region of Brazil. Terrain attributes and remote sensing data (with 30 m of spatial resolution) were used to represent landscape co-variables selected as an input in predictive models in order to develop the explanatory variables. The selection of sampling sites was based on the Latin Hypercube algorithm. A representative set of one hundred points with feasible field access was chosen. Different input databases were tested for prediction of mineral fraction content (harmonized and original data). The Spline algorithm was used to harmonize data according to the GlobalSoil.Net consortium standards. The results showed better performance from the RT models, using input from an average of six covariates; the simplest MLR model used twice as many input variables, creating more complex models without gaining precision. Furthermore, better R² values were obtained using RT models, irrespective of harmonization of soil data. The harmonized dataset from the 0.00-0.05 and 0.05-0.15 m layers, in general, presented better results for the clay and silt, with R2 values of 0.52 (0.00-0.05 m) and 0.69 (0.05-0.15 m), respectively. Prediction of sand content showed better results when the original depth data was used as an input, although all regression tree models had R2 values greater than 0.52. The RT models provided a better statistical index than MLR for all predicted properties; however, the variance between models suggests similarity of performance. Regarding harmonization of soil data, both input databases (harmonized or not) can be used to predict soil properties, since the variance of model performance was low and generalization of the soil maps showed similar trends. The products obtained from the digital soil mapping approach make it possible to integrate the factor of uncertainties, providing easier interpretation for soil management and land use decisions. 653 $aAtributos do terreno 653 $aFunções de profundidade do solo 653 $aMapeamento digital do solo 653 $aModelos de regressão 700 1 $aCARVALHO JUNIOR, W. de 700 1 $aCHAGAS, C. da S. 700 1 $aANJOS, L. H. C. dos 700 1 $aOWENS, P. R. 773 $tRevista Brasileira de Ciência do Solo$gv. 42, article e0170167, 2018.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Solos (CNPS) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
Registros recuperados : 283 | |
102. | | LOSS, A.; SANTOS, H. G.; DELARMELINDA-HONORE, E. A.; ANJOS, L. H. C.; WADT, P. G. S. Fracionamento granulométrico e oxidável da matéria orgânica em solos sob pastagens no estado do Acre. In: SILVA, L. M. da; ANJOS, L. H. C. dos; LUMBRERAS, J. F.; PEREIRA, M. G.; WADT, P. G. S. (Ed.).. (Org.). Pesquisas coligadas da IX Reunião Brasileira de Classificação e Correlação de Solos: solos de formações sedimentares em sistemas amazônicos: potencialidades e demandas de pesquisa. Brasília, DF: Embrapa, 2019, v. 1, p. 1-1.Tipo: Capítulo em Livro Técnico-Científico |
Biblioteca(s): Embrapa Rondônia. |
| |
103. | | GOMES, F. W. de F.; EBELING, A. G.; ANJOS, L. H. C. dos; PEREIRA, M. G.; PEREZ, D. V. Fracionamento químico da matéria orgânica de Organossolos no estado do Rio de Janeiro. In: CONGRESSO BRASILEIRO DE CIÊNCIA DO SOLO, 32., 2009, Fortaleza. O solo e a produção de bioenergia: perspectivas e desafios. [Viçosa, MG]: SBCS; Fortaleza: UFC, 2009. 1 CD-ROM.Tipo: Resumo em Anais de Congresso |
Biblioteca(s): Embrapa Solos. |
| |
104. | | LOSS, A.; PEREIRA, M. G.; SCHULTZ, N.; ANJOS, L. H. C. dos; SILVA, E. M. R. da. Frações orgânicas e índice de manejo de carbono em diferentes sistemas de produção orgânico. In: REUNIÃO BRASILEIRA DE MANEJO E CONSERVAÇÃO DO SOLO E DA ÁGUA, 17., 2008, Rio de Janeiro. Manejo e conservação do solo e da água no contexto das mudanças ambientais. Rio de Janeiro: SBCS: Embrapa Solos: Embrapa Agrobiologia, 2008. 4 p. 1 CD-ROM. (Embrapa Solos. Documentos, 101). Parceria: UFRRJ.Tipo: Artigo em Anais de Congresso / Nota Técnica |
Biblioteca(s): Embrapa Agrobiologia. |
| |
109. | | PINHEIRO, H. S. K.; BARBOSA, A. M.; ANJOS, L. H. C. dos; CHAGAS, C. da S.; CARVALHO JUNIOR, W. de. Efeitos da resolução espacial em modelos digitais de elevação oriundos de interpolação de dados no mapeamento digital de solos da bacia do rio Guapi-Macacu, RJ. In: CONGRESSO BRASILEIRO DE CIÊNCIA DO SOLO, 33., 2011, Uberlândia. Solos nos biomas brasileiros: sustentabilidade e mudanças climáticas: anais. [Uberlândia]: SBCS: UFU, ICIAG, 2011. 1 CD-ROM.Tipo: Artigo em Anais de Congresso |
Biblioteca(s): Embrapa Solos. |
| |
112. | | DE BEM, I. O.; ANJOS, L. M. dos; RITSCHEL, P. S.; COELHO, A. G. S.; FERREIRA, M. E. Genetic diversity of apirenic vitis accessions of a germplasm bank. Journal of Basic and Applied Genetics, Buenos Aires, v. 23, p. 271, 2012. Suplemento. Resumo apresentado nos anais no XV Congreso Latinoamericano de Genética, 2012, Rosário, Argentina.Tipo: Resumo em Anais de Congresso |
Biblioteca(s): Embrapa Uva e Vinho. |
| |
115. | | PEREIRA, M. G.; PEREZ, D. V.; VALLADARES, G. S.; SOUZA, J. M. P. F.; ANJOS, L. H. C. Comparação de métodos de extração de cobre, zinco, ferro e manganês em solos do Estado do Rio de Janeiro. Revista Brasileira de Ciência do Solo, Viçosa, MG, v. 25, n. 3, p. 655-660, jul./set. 2001.Biblioteca(s): Embrapa Solos. |
| |
117. | | SILVA, S. N. da; PINHEIRO, É. F. M.; ANJOS, L. H. C. dos; LIMA, E.; ALVES, B. J. R. Emissões de óxido nitroso em Argissolo Amarelo e influência da queima ou manutenção da palha na colheita de cana-de-açúcar. In: JORNADA DE INICIAÇÃO CIENTÍFICA DA UNIVERSIDADE FEDERAL RURAL DO RIO DE JANEIRO, 18., 2008, Seropédica, RJ. Seropédica, RJ, 2008. Parceria: UFRRJ.Tipo: Resumo em Anais de Congresso |
Biblioteca(s): Embrapa Agrobiologia. |
| |
118. | | SILVA, S. N.; PINHEIRO, E. F. M.; CEDDIA, M. B.; ANJOS, L. H. C.; ALVES, B. J. R.; UNTERLEITNER, B. Emissões de óxido nitroso em Solos sob pastagem em função da topografia no Rio de Janeiro. In: CONGRESSO BRASILEIRO DE CIÊNCIA DO SOLO, 32., 2 a 7 de agosto de 2009, Fortaleza. Trabalhos... Fortaleza: Sociedade Brasileira de Ciência do Solo, 2009.Tipo: Artigo em Anais de Congresso / Nota Técnica |
Biblioteca(s): Embrapa Agrobiologia. |
| |
Registros recuperados : 283 | |
|
Nenhum registro encontrado para a expressão de busca informada. |
|
|