Geostatistics in Physical Geography. Part II: Applications

Abstract

Regionalized variable theory has many potential applications in physical geography. The principal one is the estimation of properties distributed over the earth's surface to assess their spatial variation. The paper illustrates this with examples in which the thickness of cover loam and the stone content of the soil are estimated and mapped over small regions. The same procedure is used to map the annual rainfall over England and Wales, the results of which demonstrate the difficulty of doing so with too few data. The theory can also be used to design rational sampling schemes for mapping, and an example of a survey in the Wyre Forest describes an efficient two-stage procedure for this. Applying regionalized variable theory to hill slope profile data provides statistical summaries of their spatial structure. The examples from Dartmoor show how the wavelengths and amplitudes of periodic variograms were used to interpret and compare the slope forms.