On the stability of the 2D interpolation algorithms with uncertain data

In Earth Sciences, the 2D interpolation methods are used to make maps and vertical sections from a set of point data. Since every measure has an uncertainty, the aim of this paper is to study the sensitivity of the three most often used methods (weighting method, kriging and cokriging). The results are as follows: the three methods are more sensitive to an error of localization in space than to an error of measure. The geostatistical methods prove their worth by showing a more robust behavior as regard to the errors than the analytical ones.