Seismic Data Processing of a Real Dataset in the Gunsan Basin, Offshore Western Korea using Open-source Software (Madagascar)

We studied on the seismic field data processing using Madagascar (open-source software). Multichannel seismic field data were acquired in Gunsan Basin (Yellow Sea, offshore western Korea) using 480 channels of streamer with a 12.5 m group interval. The source interval is 50 m. Four air-gun arrays (4,578 cu. in.) generated seismic pulses. The length of the marine seismic survey line is about 31 Line-km. To verify the usefulness of the Madagascar, the seismic data processing workflow is composed of data loading, geometry build-up, deconvolution, F-K filter, velocity analysis, normal moveout (NMO) correction, stack, and Stolt's migration. Through the testing, we have confirmed that Madagascar is very effective for the reproducibility and data management. Madagascar also reduces processing time by using automated velocity picking. The F-K filter and predictive deconvolution are useful for the multiple attenuation. We could produce the multiple-suppressed time migration section using open-source software (Madagascar). From the processing results, we have confirmed that Madagascar is effective in attenuating the noise and multiples. However, the multiples for the deeper part of the sub-structure still remain. This is because we don't apply various algorithms for de-multiple. If we adopt various processing modules recently developed by many researchers, the results for the deeper part of the sub-structure will be improved.