Resolution enhancement in tilted coordinates

Abstract

© 2016 Deconvolution is applied to remove source wavelet effects from seismograms. The results are resolution enhancement that enables detection of thin layers. Following enhancement of resolution, low frequency and high angle reflectors, particularly at great depth, appear as low amplitude and semi-invisible reflectors that are difficult to track and pick. A new approach to enhance resolution is introduced that estimates a derivative using continuous wavelet transform in tilted coordinates. The results are compared with sparse spike deconvolution, curvelet deconvolution and inverse quality filtering in wavelet domain. The positive consequence of the new method is to increase sampling of high dip features by changing the coordinate system from Cartesian to tilted. To compare those methods a complex data set was chosen that includes high angle faults and chaotic mass transport complex. Image enhancement using curvelet deconvolution shows a chaotic system as a non-chaotic one. The results show that sparse spike deconvolution and inverse quality filtering in wavelet domain are able to enhance resolution more than curvelet deconvolution especially at great depth but it is impossible to follow steep dip reflectors after resolution enhancement using these methods, especially when their apparent dips are more than 45°. By estimating derivatives in a continuous wavelet transform from tilted data sets similar resolution enhancement as the other deconvolution methods is achieved but additionally steep dipping reflectors are imaged much better than others. Subtracted results of the enhanced resolution data set using new method and the other introduced methods show that steeply dipping reflectors are highlighted as a particular ability of the new method. The results show that high frequency recovery in Cartesian co-ordinate is accompanied by inability to image steeply dipping reflectors especially at great depths. Conversely recovery of high frequency data and imaging of the data both at shallow and great depth are an ability of resolution enhancement in tilted co-ordinates.