Interpolation with pseudo-primaries

William Curry

bill@sep.stanford.edu

ABSTRACT

Large gaps exist in marine data, particularly at near offsets. I generate pseudo-primaries by cross-correlating a multiple model with the original data. These pseudo-primaries are used as training data for a non-stationary prediction- error filter, which is then used to interpolate the missing near offsets. This method yields good results, and also provides a quality control measure to judge the usefulness of the pseudo-primaries.