Combination of Minimum-Maximum (m-m) Attribute and Zero-INTENS-Difference (z-i-d) Attribute for Estimating Seismically Thin-Bed Thickness

Purnomo, Eko Widi ( Geoscience Program, Universiti Teknologi PETRONAS, Malaysia)
Harith, Zuhar Zahir Tuan ( Geoscience Program, Universiti Teknologi PETRONAS, Malaysia)

Journal of Engineering and Technological Sciences
Vol 43, No 2 (2011)

This  paper  demonstrates  a  new  alternative  way  in  estimating seismically thin-bed (below-tuning) thickness. Initial thickness is built by bandpass filtering the amplitude display of a zero-phase seismic. The filter removes the  non  minimum  and  or  non  maximum  and  left  the  maximum  and  or  the minimum of seismic amplitude. The unresolved below-tuning thickness is then corrected  by  zero-INTENS-difference  (z-i-d)  attribute.  INTENS  is  integrated energy  spectra,  an  attribute  which  can  be  derived  from  spectral  analysis.  z-i-d attribute is zero difference of INTENS between the seismic and its synthetic. The method  generates  INTENS  difference  profile  by  subtracting  seismic  INTENS and its synthetic INTENS iteratively. The iteration is controlled by dipole space shifting from  distance to closer or  vice  versa.  The true thickness is derived  by locating z-i-d which laid in INTENS different profile. It has found that, for free noise  true  seismic  and  perfect-wavelet  (a  wavelet  which  only  approximately similar  with  wavelet  which  constructing  the  true  seismic)  synthetic  seismic,  in INTENS  different  profile,  the  z-i-d  location  always  corresponds  to  true  dipole space or thickness. The method could resolve all thickness of a wedge-modeled seismic with three different dominant frequencies. When the synthetic seismic is constructed with imperfect wavelet, slightly different analysis is needed to locate z-i-d  attribute  and  the  result  is  not  as  perfect  as  when  perfect  wavelet constructing synthetic seismic. A quiet similar result is got when the method is implemented  for  noisy  wedge-modeled  seismic.  Bad  thickness  estimation  is resulted  for  20%  noise  seismic.  The  method  algorithm  is  extended  for  similar dipole polarity model and multilayer model to bring the method to real seismic data  nearer.  The  extension  is  done  by  estimating  thickness  of  every  layer  of  a stacked-wedge-modeled  seismic. The algorithm then generalized for estimating layers  thickness  with  several  thickness  combinations.  The  method  was  able  to delineate shallow channel of Stratton Field by providing good pseudo-acousticimpedance (pseudo AI) map.