Publications

The studied field is located offshore, Indonesia. A number of wells have been drilled on structure, successfully proving the presence of gas in a series of stacked reservoirs within the Miocene interval. Detailed reservoir evaluation has however been hampered by the presence of shallow gas. Simultaneous seismic inversion was used to derive rock properties from the seismic data. The P and Simpedances were ideal for integrating well logs to the seismic data because the impedance values from the inversion can be directly related to the layerbased rock properties. However, the reservoir rocks could not be separated on the basis of P-impedance alone. The biggest challenge presented by the dataset was the attenuation of the seismic response by shallow gas. This resulted in changes in seismic phase, frequency and amplitude. Consequently the existing data was not reliable for conventional inversion using a single wavelet. In order to compensate for the dimmed zone in the seismic data, lateral varying wavelets were applied. Once the lateral varying wavelets were optimized, they were applied to the seismic inversion. Based on well data, reservoir and non reservoir can be separated using a combination of P-impedance and Vp/Vs. These parameters were then applied to the inversion to characterize the reservoir, and identify the pay sand distribution from the inversion results. The result shows that lateral varying wavelets can compensate for the shallow gas effect in the seismic data, and reduce the uncertainty of the seismic inversion modeling for reservoir characterization.Key word: Shallow gas, seismic attenuation, reservoir characterization, simultaneous inversion, lateral varying wavelets