A Strategy for Acoustic Impedance Direct Inversion in Depth Domain

Ruiqian Cai,Chengyu Sun,Shizhong Li

DOI: https://doi.org/10.1190/image2022-3751329.1

2022-01-01

Abstract:Due to the strong non-stationarity of the depth-domain seismic data, the traditional convolutional model cannot be used to calculate the synthetic seismogram in depth domain. Therefore, the time-domain inversion methods, which are based on convolutional model, fail to be directly applied to the depth-domain seismic inversion. To carry out seismic inversion in depth domain, we propose a strategy that attains direct inversion. The proposed method integrates the step of depth-time-depth transform and of inversion, and therefore it can directly invert the acoustic impedance from the depth- domain seismic data. Firstly, we derive an analytic formula for synthesizing the depth-domain seismic record directly, which uses the depth-domain model and the time-domain wavelet. Based on this analytic formula, the objective function of direct inversion is derived. Then, we apply our method on the synthetic data and the field dataset. To demonstrate the improvement of direct inversion results, the results of the damped least squares inversion method are used for comparison. It is worth mentioning that the depth- time-depth transform is conducted to realize the inversion of depth-domain seismic data for the damped least squares inversion method. Tests shows that the performance of the proposed method is satisfactory and superior than that of the damped least squares inversion method. Note: This paper was accepted into the Technical Program but was not presented at IMAGE 2022 in Houston, Texas.