ABSTRACT
The study establishes the thickness and distribution of the ?X? sandstonereservoir in the ASA Field that is located at the southern margin of West NatunaBasin, southwestern South China Sea. The field is located on top of the ?D?horst, which is bordered by the east-northeast (ENE) -west-southwest (WSW)trending basement ridge on the south. By application of discrete Fouriertransform (DFT) on high-resolution 3D seismic data over a short windowcovering the geologic zone of interest, the amplitude spectra of an ?X?sandstone prone channel can help delineate temporal bed thickness variabilityand sandstone distribution.Spectral decomposition is just valid for analysis covered one waveletseismic that will decrease the noise, so that all of frequency range until Nyquistfrequency can be used for analysis. Noise appearance can be used todetermine geological boundaries such as channels and sand bars, but it cannotbe used for the thickness estimation. The maximum value of the first peakfrequency will determine the thinnest layer observable within analysis window.The average tuning thickness ranges is from 30 to 40 feet. The thinnestdetectable layer is about 12 feet that are found at the finite area, this is equalwith 1/12λ, where λ is seismic wavelet wavelength. The tuning thickness of thesandstone reservoir detected by spectral decomposition analysis is thinner thanconventional tuning calculation which is about 35 feet on ¼λ.The paleo-stream flow is interpreted to be from northwest to southeastacross the study area based on spectral decomposition analysis. Faulting wasnot active during ?X? sandstone deposition as evidenced by lack of downthrownthickening and continuity of interpreted sand bars across faulted area.Based on this study, more advance study is recommended to be done tounderstand the optimum window length for spectral decomposition analysisespecially using more than one seismic wavelet to determine sand distributionand its thickness. More advanced method of spectral decomposition analysis isnecessary to detect, within one seismic wavelet, variation in bulk rock propertiesfrom which inferences regarding depositional environment and lithology can bemade.