Autoria: MENDONÇA-SANTOS, M. de L.; DART, R. de O.; BERBARA, R. L. L.

Resumo: The use of the Conditional Latin Hypercube (cLHS) to select soil samples location to be used in the prediction of soil properties as soil organic carbon seems to be an important tool to decrease costs and subjectivity of sampling schemes. The main objective of this work was to test this method in a no sampled area, aiming to evaluate its efficiency in Digital Soil Mapping (DSM) procedures. The results show that this method was able to significantly increase the reliability in the spatial distribution of the sampled points in Parque Estadual da Mata Seca (PEMS). The study area is located in the North of Minas Gerais State, Brazil, in Tropical Dry Forest (TDF) ecosystems mainly. In this study, the cLHS algorithm developed by Minasny & McBratney (2006) was used, in addition to some ancillary data as Land Use/Land Cover and Normalized Difference Vegetation Index (NDVI), in order to select 60 soil sampling location to the study of soil organic carbon and others soil properties. The cLHS will be further analyzed in its performance to select representative points to be sampled and how this method could be helpful for DSM.

Ano de publicação: 2008

Tipo de publicação: Artigo em anais e proceedings

Unidade: Embrapa Solos

Palavras-chave: CLHS, LHS, Mapeamento digital, Mata Seca