Yu Qi ^* Kui Wu ^* Qixin Li ^* Xiaowen Zheng ^* Bo Wang ^* Dan Li Wei Tang

• CNOOC Research Institute Ltd., Beijing, China

The development of high-accuracy seismic prediction technology, capable of mitigating the influence of coal events and accurately inverting key reservoir parameters, is critical for enhancing the exploration and development of tight gas reservoirs in coal-bearing strata. Block L, located on the eastern edge of the Ordos Basin, serves as a case study. This area, where tight sandstone gas deposits are widespread in the Upper Paleozoic, is characterized by coal seams that complicate seismic interpretation. The available drilling data shows that the tight gas layers in the Taiyuan and Benxi Formations are well-developed and serve as the primary hydrocarbon source rocks. However, the low-density, low-velocity coal seams create strong wave impedance contrasts, which significantly affect the neighboring tight sandstone gas layers and present substantial challenges for seismic prediction. To address this issue, we employ wavelet decomposition and reconstruction techniques to eliminate coal seam interference and high-resolution waveform simulation technology for predicting thin reservoirs. This approach circumvents the limitations of using a single impedance parameter to differentiate between sand-mudstone and limestone-coal interfaces. Step-by-step inversion and gradual stripping techniques are applied to remove the coal seam and limestone influences, thereby facilitating accurate identification of sandstone distribution. Drilling data confirm the reliability and effectiveness of this reservoir prediction method for coal-bearing strata. Based on successful applications in the eastern Ordos Basin, this workflow demonstrates broad applicability and offers significant guidance for exploring similar reservoirs, given the parallel reservoir characteristics and the foundational technological approach.