Abstract
More in-situ tests are typically carried out over the same volume of soil in comparison to laboratory tests on undisturbed borehole samples, Hence, geostatistical analysis of in-situ test records should in principle provide a more accurate and representative overview of spatial variation, A natural probabilistic model for correlated spatial data is the random field. While the random field provides a concise description of spatial variation, it poses considerable practical difficulties for statistical inference because of the underlying autocorrelation structure, In this paper, a robust geostatistical procedure that is fully consistent with the stationary random field hypothesis is illustrated using a piezocone record. The feasibility of the proposed geostatistical procedure hinges on the new modified Barlett test, which is capable of rejecting the null hypothesis of weak stationarity for spatially correlated data.