ICE 2022

Summary

Seismic data preconditioning is a vital step in optimizing the inputs to pre-stack AVO inversion workflows and generating results more consistent with other measured data. Preconditioning can include denoising, amplitude and alignment corrections, and spectral operations on migrated gathers as well as on angle stacks. As AVO inversion for elastic properties depends on variations of seismic amplitude with offset, these steps are critical to achieve improved accuracy. In this case study, we demonstrate the effects and value of properly preconditioning seismic data prior to inversion, using an unconventional Montney dataset. A focus is placed on noise removal and proper event alignment, in both gather and angle stack domains. Pre-stack, experiments on ordering of RMO correction and radon demultiple operations are tested, with an iterative approach eventually being selected to minimize erroneous corrections. After angle-stacking, seismic warping, or alignment, is applied to correct for any remaining residual moveout by iteratively aligning all anglestacks to a reference stack. At each preconditioning step, thorough QC is performed to justify these processes and their parameters. AVO inversion is conducted on seismic both with and without this rigorous preconditioning applied. The inverted elastic properties from each workflow are compared to each other and to well logs. A relative inversion is used for this step, as the omission of a detailed background model ensures the results and any conclusions from them are driven mainly by the seismic data. Subsequently, a rock-physics inversion for porosity is performed on each dataset using calibrated rock-physics templates. By following this approach and inverting both preconditioned and raw datasets, we can directly assess inversion performance, and thus the uplift of the seismic preconditioning workflow. Small changes in acoustic impedance as a result of the preconditioning workflow can be correlated to non-negligible changes in inverted acoustic impedance, and as a result, on porosity. Additionally, after preconditioning there was found to be an increase in the correlation between inverted subsurface parameters and production.