64 3 D soil compaction mapping with a three - coefficient polynomial

Soil property mapping is commonly done in two dimensions, and variation by depth is ignored. The only way to consider the vertical variation is mapping in 3D. In this paper, we present a method based on polynomial depth functions. The objective is determining the effectiveness of a simple polynomial to predict cone-index data by depth over a field. In order to study the performances of the proposed method we compared our results with 3D ordinary kriging using cross validation, excluding a percentage of samples and repredicting their values. Results show that the proposed method performs better that 3D kriging.