Algorithm for the production of contour maps from linearized data

Cole has proposed an algorithm which produces a generalized surface for contouring scattered data1. The algorithm works well when the data points are randomly distributed but improbable results have been obtained when the data points are distributed along lines. There are two reasons for this: one is the extrapolations which occur into regions without data and the other is the generation of unlikely gradients across the lines. Measurements of many geophysical properties, made by moving vehicles, ships and so on, are often distributed along lines which are not necessarily straight. They may cross one another but the distance between adjacent measurements on the lines is usually much less than the distance between the lines. Such a distribution of measurement points may be found in other situations where the two dimensions are not spatial.