Search

Digital soil mapping studies (MDS) have used legacy maps as the main source of information for calibration of predictor models. However new approaches are needed with techniques that allow the use of information contained in georeferenced soil profiles, allowing the application of MDS in sampled areas that do not provide conventional soil maps. The objective of this study was to evaluate the performance in the prediction of soil occurrence of samples collected in pixels of profiles of georeferenced soils and in pixels collected in buffers with radius of 50, 100, 150, 200 and 250 m of soil profiles in the Lajeado Grande and Santo Cristo Rivers Watersheds. Two areas with availability of soil survey data published at scale 1: 50,000 were used for the study. For prediction of the occurrence of soil classes, ten predictive variables were generated from a digital elevation model with spatial resolution of 30 m. For the prediction, Random Forest algorithm was used. The predicted maps were evaluated for accuracy in relation to soil profiles and the reproducibility of conventional maps. The use of the sampling pixels collected in the buffers did not significantly alter the overall accuracy of the predicted maps in the Lajeado Grande river watershed, but allowed a 15.6% the gain at overall accuracy the Santo Cristo river watershed. The use of georeferenced profiles resulted in predicted maps with overall accuracy greater than 75% and agreement of reproducibility equal to or high than 67% in relation to the conventional map. [ABSTRACT FROM AUTHOR]

Digital soil mapping (DSM) has been shown to be feasible to use in soil survey. Although several methods have been exploited, there is a lack in defining methodologies for doing DSM. This study tests five decision trees algorithms that have been identified as suitable (Simple Chart, Random Tree, REP Tree, BF Tree, and J48) and three digital elevation models (AsterGDEM, SRTM and SRTM V3)for DSM at semidetailed level in situations where the main differentiating factor between soil types is the relief. The use of MDE Aster GDEM and decision three algorithms J48, Simple Tree e BF Tree produced decision tree models capable of produce soil maps with larger accuracy related to reference soil maps. [ABSTRACT FROM AUTHOR]

Soil is increasingly being recognized as having an important role in ecosystems, as well as for food production and global climate regulation. For this reason, the demand for relevant and updated soil information is increasing. Soil science researchers are being demanded to produce information in different spatial resolutions with associated quality in what is being called Digital Soil Mapping (DSM). Due to an increasing number of papers related to the DSM in Brazil, it is necessary to discuss the main characteristics of those studies related to the automated mapping of soil classes, which will enable a broader perspective of the subject and guide future works and demands. The mapping of soil classes using DSM techniques is recent in the country, the first publication in this topic occurred just in 2006. Among the predictive functions the predominant is logistic regression. The soil formation factor relief was used in all studies reviewed. Quality of predictive models was evaluated employing error matrix and kappa which were the most common procedures. The consolidation of this automated approach as an auxiliary tool to the conventional soil mapping will demand training of young soil scientists to use geoinformation technologies and quantitative tools to handle aspects of soil variability. [ABSTRACT FROM AUTHOR]

Available technologies for Earth observation offer a wide range of predictors relevant to Digital Soil Mapping (DSM). However, models with a large number of predictors, as well as, the existence of multicollinearity among the data, may be ineffective in the mapping of classes and soil properties. The aim of this study was to use the Principal Component Analysis (PCA) to reduce the number of predictors in the multinomial logistic regression (MLR) used in soil mapping. Nine environmental covariates, related to the relief factor of soil formation, were derived from a digital elevation model and named the original variables, which were submitted to PCA and transformed into principal components (PC). The MLR were developed using the terrain attributes and the PC as explanatory variables. The soil map generated from three PC (65.6% of the original variance) had a kappa index of 37.3%, lower than the 48.5% achieved by the soil map generated from all nine original variables. [ABSTRACT FROM AUTHOR]

The aim of this study was to apply an automated technique of soil mapping from predictive models developed at a reference area, into nearby areas of the same soil-landscape relationships. Generalized linear models were developed using nine terrain attributes derived from a digital elevation model as covariate predictors, with soil classes, obtained from a conventional soil survey, as dependent variables. The models were able to distinguish the three main forms of the local landscape. Soil classes with pedogenesis intimately tied to the predictive covariates obtained the best results. The soil maps generated, showed a reproducibility of 46.12% and an accuracy of 21.06%. [ABSTRACT FROM AUTHOR]