Abstract:SUMMARYIs it possible to build predictive models (PMs) of soil particle-size distribution (psd) in a region with complex geology and a young and unstable land-surface? The main objective of this study was to answer this question. A set of 339 soil samples from a small slope catchment in Southern Brazil was used to build PMs of psd in the surface soil layer. Multiple linear regression models were constructed using terrain attributes (elevation, slope, catchment area, convergence index, and topographic wetness i… Show more
“…Henderson et al [21] Nanni and Demattê [22] Fiorio et al [23] Shabou et al [24] ETM+ 0.26-0.68 Chagas et al [25] Diek et al [26] Jena-Optronik (RapidEye) 0.24-0.56 Forkuor et al [27] EO-1 ALI (Hyperion) 0.51-0.83 Zhang et al [28] Castaldi et al [29] Relief-Derived Covariates Elevation, Slope, Convergence Index, and Topographic Wetness Index 0.55 Samuel-Rosa et al [30] Relative Elevation and Slope 0.07 Sumfleth and Duttman [31] Slope, Plan Convexity, and Upslope Distance 0.51 Odeh et al [32] Therefore, literature studies have brought us to important points for evaluation towards the usefulness of soil image analysis, as follows: (a) Is there room to increase the prediction of clay content by multispectral sensors, as they have some limitations compared with hyperspectral ones (distance, spectral, and spatial resolution)? (b) An important paradigm supported by many users is that soils cannot be evaluated using images since they are usually covered by vegetation; (c) Relief parameters have been considered to be the best information sources for mapping soil attributes because they are soil formation factors.…”
Soil mapping demands large-scale surveys that are costly and time consuming. It is necessary to identify strategies with reduced costs to obtain detailed information for soil mapping. We aimed to compare multispectral satellite image and relief parameters for the quantification and mapping of clay and sand contents. The Temporal Synthetic Spectral (TESS) reflectance and Synthetic Soil Image (SYSI) approaches were used to identify and characterize texture spectral signatures at the image level. Soil samples were collected (0–20 cm depth, 919 points) from an area of 14,614 km2 in Brazil for reference and model calibration. We compared different prediction approaches: (a) TESS and SYSI; (b) Relief-Derived Covariates (RDC); and (c) SYSI plus RDC. The TESS method produced highly similar behavior to the laboratory convolved data. The sandy textural class showed a greater increase in average spectral reflectance from Band 1 to 7 compared with the clayey class. The prediction using SYSI produced a better result for clay (R2 = 0.83; RMSE = 65.0 g kg−1) and sand (R2 = 0.86; RMSE = 79.9 g kg−1). Multispectral satellite images were more stable for the identification of soil properties than relief parameters.
“…Henderson et al [21] Nanni and Demattê [22] Fiorio et al [23] Shabou et al [24] ETM+ 0.26-0.68 Chagas et al [25] Diek et al [26] Jena-Optronik (RapidEye) 0.24-0.56 Forkuor et al [27] EO-1 ALI (Hyperion) 0.51-0.83 Zhang et al [28] Castaldi et al [29] Relief-Derived Covariates Elevation, Slope, Convergence Index, and Topographic Wetness Index 0.55 Samuel-Rosa et al [30] Relative Elevation and Slope 0.07 Sumfleth and Duttman [31] Slope, Plan Convexity, and Upslope Distance 0.51 Odeh et al [32] Therefore, literature studies have brought us to important points for evaluation towards the usefulness of soil image analysis, as follows: (a) Is there room to increase the prediction of clay content by multispectral sensors, as they have some limitations compared with hyperspectral ones (distance, spectral, and spatial resolution)? (b) An important paradigm supported by many users is that soils cannot be evaluated using images since they are usually covered by vegetation; (c) Relief parameters have been considered to be the best information sources for mapping soil attributes because they are soil formation factors.…”
Soil mapping demands large-scale surveys that are costly and time consuming. It is necessary to identify strategies with reduced costs to obtain detailed information for soil mapping. We aimed to compare multispectral satellite image and relief parameters for the quantification and mapping of clay and sand contents. The Temporal Synthetic Spectral (TESS) reflectance and Synthetic Soil Image (SYSI) approaches were used to identify and characterize texture spectral signatures at the image level. Soil samples were collected (0–20 cm depth, 919 points) from an area of 14,614 km2 in Brazil for reference and model calibration. We compared different prediction approaches: (a) TESS and SYSI; (b) Relief-Derived Covariates (RDC); and (c) SYSI plus RDC. The TESS method produced highly similar behavior to the laboratory convolved data. The sandy textural class showed a greater increase in average spectral reflectance from Band 1 to 7 compared with the clayey class. The prediction using SYSI produced a better result for clay (R2 = 0.83; RMSE = 65.0 g kg−1) and sand (R2 = 0.86; RMSE = 79.9 g kg−1). Multispectral satellite images were more stable for the identification of soil properties than relief parameters.
“…Nesse estudo, os autores encontraram um modelo com R 2 acima de 0,8 e desviopadrão residual acima de 2,0 utilizando 128 amostras de treinamento. Samuel-Rosa et al (2013), ao testarem modelos para predição da distribuição granulométrica nessa mesma área, obtiveram resultados considerados satisfatórios, concluindo porém que a complexidade geológica da área reduziu a acurácia dos modelos preditivos. Esses mesmos autores relataram que não ocorreu aumento da acurácia quando houve a estratificação da superfície Estatística log (Argila) log (Areia) log (Silte) log (MOS) Quadro 2.…”
Section: Discussionunclassified
“…Na região de estudo, localizada no Rebordo do Planalto, centro do Rio Grande do Sul (RS), a geologia é complexa com relevo suave ondulado a montanhoso com solos de profundidade variável com o predomínio de solos rasos (Pedron et al, 2006). Samuel-Rosa et al (2013) utilizaram essa área para desenvolver modelos preditivos da distribuição do tamanho de partículas usando os atributos de terreno como variáveis; e nos resultados, esses autores demonstraram que as incertezas da predição estão relacionadas à complexidade geológica. Poucos trabalhos exploram áreas como essas, consideradas marginais para exploração agrícola empresarial, mas largamente utilizadas para agricultura familiar, sendo necessários estudos que possam apresentar modelos satisfatórios para predição de determinados atributos do solo.…”
Pedron (4) , Alexandre ten Caten (5) & Luis Fernando Chimelo Ruiz (6) RESUMO A espectroscopia de reflectância difusa (ERD) pode ser utilizada como alternativa para quantificação de atributos como granulometria e matéria orgânica do solo (MOS). Essa técnica pode ser opção para quantificar esses atributos em grande volume de amostras de solos, visto ser rápida, com menor custo e sem a geração de resíduos químicos. O objetivo deste estudo foi desenvolver modelos usando análise de regressão linear múltipla para predizer o teor de argila, areia, silte e MOS, utilizando dados de ERD em uma área de relevo e geologia complexa localizada na região central do Rio Grande do Sul. No estudo, foram utilizadas 303 amostras coletadas na profundidade de 0,00-0,20 m para determinar os teores de argila, areia, silte e MOS por meio da análise laboratorial e de reflectância espectral. O desempenho dos modelos de predição apresentaram bons resultados, com capacidade de explicação da variância de 77 e 72 % para areia e argila, respectivamente. Mesmo com a complexidade geológica e pedológica, os resultados evidenciaram que a técnica é promissora, sendo possível a aplicação dessa técnica para predição da granulometria e teor de MOS.Termos de indexação: pedometria, sensoriamento remoto proximal, radiometria, predição de atributos.(1) Parte da Dissertação de Mestrado do primeiro autor. Estudo financiado pela CAPES. Recebido para publicação em 13 de janeiro de 2014 e aprovado em 30 de julho de 2014.
“…Essas abordagens são encontradas com freqüência na literatura (Giasson et al 2006, Connolly et al 2007, Grimm et al 2008, Hansen et al 2009, Odgers et al 2011a, Odgers et al 2011b, ten Caten et al 2011, Häring et al 2012, Kerry et al 2012, Samuel-Rosa et al 2013, ten Caten et al 2013, Cavazzi et al 2013, Dotto et al 2014, Adhikari et al 2014, Hartemink e Minasmy 2014. No entanto, os resultados dessas diferentes abordagens preditivas são dependentes da qualidade dos dados utilizados, a saber: resolução espacial e incerteza das covariveis ambientais, qualidade e quantidade das amostras de treinamento, densidade espacial das amostras, e a capacidade do modelo preditivo em capturar a variação espacial do solo.…”
Section: Figura 1 Estratégias Para O Desenvolvimento De Modelos Predunclassified
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.