Abstract:

The thin sandstone reservoir in the delta front developed in the upper part of the K₁bs³ formation is an important area for the exploration of structural-lithological oil and gas reservoirs in the XH area. In recent years, Application of P-impedance and pre-stack deterministic inversion could not accurately predict the distribution of thin sandstone reservoirs in the target layer. In this paper, joint impedance and facies inversion（JI-FI） impedance and facies joint inversion technology is used. This technique uses pre-stack seismic gather data, and combines simultaneous inversion and Bayesian classification under the constraints of the P-wave, S-wave velocity and density depth trend of each lithofacies or fluid facies, the maximum expectation algorithm is used to iterate the impedance and facies, and the quantitative prediction of lithofacies/fluid facies and petrophysical properties （such as P-wave, shear wave impedance, density, V_p/V_s, etc.） in the target area is achieved. This method overcomes the uncertainty of inter-well P-impedance interpolation when conventional pre-stack inversion relies on wells to establish low-frequency models, especially under geological conditions with few wells and thin interbeds. The reservoir prediction results are in good agreement with the well, which provides a basis for lithological exploration and well location deployment in this area.

Key words: joint impedance and facies inversion（JI-FI）, lithofacies depth trend, Bayes classification, thin sandstone reservoir, prediction of lithofacies/fluid facies