Resumo

FORTIS-HERNANDEZ, Manuel et al. Validation of four interpolation model for mapping nitrate and ammonium in soil. Terra Latinoam [online]. 2010, vol.28, n.4, pp.371-379. ISSN 2395-8030.

Statistical validation of four models of interpolation in a soil from Lagunera Region was conducted. The objective of this research was to establish the model that better fit the distribution of ammonium and nitrate content in samples of soil fertilized with manure and, based on this, to develop a spatial distribution map. Forty-eight points in a plot were georeferenced to obtain the datum NAD 27 with the system of geographical coordinates. Ammonium and nitrate concentrations were obtained from each point of soil samples (from 0-15 and 15-30 cm depth) at the end of growing season to determine ammonium and nitrate contents. Four geostatistical interpolation methods were used: inverse distance weighting (IDW), radial basis function (RBF), global polynomial interpolation (GPI) and kriging (K). Half of the soil samples (Ml) were used to test interpolations and the rest (M2) were excluded for use in the validation process. The statistical test revealed that the IDW interpolation method had less variability, while the Regularized Completely Linear submethod, from the RBF interpolation, was the best. The variation in concentrations of ammonium and nitrate for each thematic map indicated that distribution was punctual due to the way of applying organic amendments or fertilizers.

Palavras-chave : geostatistics; mapping; pollution.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )