ABSTRACT
The tools of geostatistics are becoming increasingly important components of GIS (Burrough and McDonnell, 1998). Geostatistics can be used for (1) the characterisation of spatial variation, (2) interpolation, (3) simulation and (4) the design of optimal sampling strategies. In particular, interpolation of elevation values and other properties from samples is a major function of GIS. This chapter uses the geostatistical technique of ordinary kriging (OK) with the aim of making optimal estimates. A principal advantage of kriging-based algorithms over other interpolation techniques is that the spatial dependence (as represented by the variogram, for example) in the property of interest is used to inform the estimates. Also, kriging provides an estimation variance as part of the estimation procedure. In this chapter, a digital terrain data set is analysed using geostatistics.
