Derivation of Quality Parameters for Terrain Models: Examples from Seafloor and Landscape Mapping

Randommeasurement errors in digital terrain models are important to their quality. The present paper describes an untypical use of variograms to compute this kind of measurement errors. After subtracting a low frequency drift from depth/height measurements, a variogram analysis of the residuals is used to derive the random measurement errors of the observations. Numerical experiments on lasercanning and multibeam echo sounder data show how the semivariogram clearly identifies the amount of noise in the original data. The method can run in real time and does not require additional measurements like ground control patches or overlapping survey strips.