Geostatistics provides tools to model, estimate, map and predict spatial patterns of tree size and growth. Variogram models were used to study spatial dependence of basal diameter in young cork oak plantation. Our objective was to determine the spatial dependence of stand by studying the related spatial stochastic model. Despite the fact that high nugget effects affect the model for whole area, cork oak diameter shows a better spatial dependence in the two different geological classes that divide the area. An isotropic exponential and an isotropic spherical variogram were the models chosen to represent granite and quartzite subareas. In both models, cork oak diameter was spatially autocorrelated over distances no greater than 8 m, a measure of average patch diameter in this forest ecosystem. The results of survey suggest that geology exerts a strong control on cork oak diameter variation, and in the whole area that other factors are also involved, such as elevation, soil depth, catchment area, slope, hillshade and aspect are creating a general random effect. Simulation models and area estimates of tree performance in young or old-growth forests may be improved by including geostatistical components to summarize ecological spatial dependence.