Self-Correlation of Uncorrelated Data

Self-correlation arises in scatter plots when the data on each axis are scaled using the same parameter. This has been a known problem for a long time and study of the effect goes back to the days of Karl Pearson. It has long been known that the spread of values of the common parameter, relative to the spread of values for the other parameters, is a major determinant of the degree of self-correlation. Despite the long history of study, there are many surprises that arise from self-correlation. Examples pertinent to the boundary layer community will be presented.