Abstract
Digital soil mapping (DSM) has become a valuable tool in order to generate accurate and current soil information. It is recognized that digital soil mapping presents some advantages compared to traditional soil surveys. Despite the growing demand for soil information, application of digital soil mapping in tropical context is still insufficient. In order to evaluate the potential use for DSM in agricultural zones managed by cooperatives in the Western Central Valley of Costa Rica, an exercise was performed to mapping three soil fertility properties (pH, acidity and ECEC) using the scorpan-SSPFe conceptual model proposed by McBratney et al. (2003) as basis. Data from 384 soil fertility tests, a digital elevation model (30 m spatial resolution), satellital imagery and climatic records were the inputs to build 24 covariates layers. There were adjusted linear model to make inferences about soil spatial variability and the uncertainty related to prediction was evaluated. The results are presented by each one of the 3 soil fertility attributes, also a global evaluation of scope and constraints of the proposed methodology as an opportunity to be implemented in other agricultural regions in similar contexts