| Popis: |
Three interpolation techniques were applied to point data on soil moisture supply capacity in a 2km x 2 km area of cover sand in the eastern part of The Netherlands. The interpolation techniques are: Proximal, Weighted Average and Kriging; survey points used for interpolation were stratified random, with an average density of 1.5 per ha. The Root Mean Squared (RMS) Error of these techniques, as estimated from an independent statistical sample, was 37, 29 and 29 mm, respectively, whereas the moisture supply capacity in the area ranged from 50 to 240 mm. Profile descriptions at the same survey points were used to make a soil map (scale 1 : 10 000), the mapping units of which were delineated in the field. Using the means of these mapping units as estimates for points in the area would e.g. have led to an RMS Error of 32 mm if the means were estimated from an additional random sample of 5 profile descriptions per mapping unit or 0.4 per ha. This lower efficiency, compared with Weighted Average and Kriging, should be weighed against the advantage of the soil map of being a basic document that can be directly used for various purposes. |
