Conventional deconvolution algorithms usually assume that the spectrum of reflectivity is white. However, the amplitude spectrum of reflectivity calculated from a well log usually demonstrates obvious dependence on frequency, which is referred to as the color of the reflectivity. Consequently, the white reflectivity assumption can lead to distortion of the deconvolution result. A practical color correction method for Gabor deconvolution has been proposed and tested on a synthetic 1D seismic trace. This article tests the proposed color correction method using a synthetic shot record and real field data. Testing on the synthetic shot record shows that color correction gives a better reflectivity estimation than conventional Gabor deconvolution, and can be successfully applied to shot records. Results of processing field data demonstrate that the seismic data have higher resolution and a better tie to the well log data when color correction is applied.
View full article as PDF (0.56 Mb)