IDENTIFICATION OF FIELD ATTRIBUTE ERROR UNDER DIFFERENT MODELS OF SPATIAL VARIATION

Recent developments in theory and computer software mean that it is no w relatively straightforward to evaluate how attribute errors are prop agated through quantitative spatial models in GIS. A major problem, ho wever, is to estimate the errors associated with the inputs to these s patial models. A first approach is to use the root mean square error, but in many cases it is better to estimate the errors from the degree of spatial variation and the method used for mapping. It is essential to decide at an early stage whether one should use a discrete model of spatial variation (DMSV - homogeneous areas, abrupt boundaries), a co ntinuous model (CMSV - a continuously varying regionalized variable he ld) or a mixture of both (MMSV - mixed model of spatial variation). Ma ps of predictions and prediction error standard deviations are differe nt in all three cases, and it is crucial for error estimation which mo del of spatial variation is used. The choice of model has been insuffi ciently studied in depth, but can be based on prior information about the kinds of spatial processes and patterns that are present, or on va lidation results. When undetermined it is sensible to adopt the MMSV i n order to bypass the rigidity of the DMSV and CMSV. These issues are explored and illustrated using data on the mean highest groundwater le vel in a polder area in the Netherlands.