The process of post-stack inversion is routinely carried out to obtain acoustic impedance (AI) from seismic data and, hence reservoir properties in the target oil field. In this study, we have been describing characterization of a Mid-Cretaceous reservoir using two seismic inversion techniques which have been applied to the main oil bearing complex reservoir in this giant field. These two methods were implemented as an approach to predict the reservoir properties and the reservoir features such as the complex geometries and connectivity patterns, with focusing on benefits and uncertainties between these two techniques.
Analysis of post-stack seismic inversion for various reservoir properties estimation is generally applied for reservoir characterization. The two methods have been interrogated and successfully applied on a heterogeneous carbonate reservoir. Model based inversion has been extensively applied on many case studies and has produced superior results on other techniques. The second methodology, linear programming sparse spike (LPSS) inversion is also much used for reservoir characterization with promising results. In this case study, high resolution 3-D seismic data was used with the data of twenty well logs. First of all, building rock physics model was performed to relate the acoustic properties and the reservoir properties, also to provide a better understating of the main elements that control the seismic amplitude in this heterogeneous carbonate reservoir. Then, both techniques were applied across the 3-D seismic data of the field, using an estimated wavelet and low frequency model.
The rock physics model demonstrated a clear linear relationship between the porosity and acoustic properties with high correlation in the Mishrif interval. The Mishrif reservoir is typically bioclastic-detrital limestone subzones with heterogeneous quality. Consequently, a various reservoir properties observed using these two inversion techniques, also the well data demonstrated a strong correlation at the well locations. Model based and linear programming sparse spike methods have been applied to the Mishrif reservoir over the field. The results of both techniques were compared with computed acoustic impedance logs on the basis of blind well tests, to identify the subzones of the Mishrif reservoir. The qualitatively and quantitatively analysis of inversion results suggest that model based inversion is generally better than those from the LPSS inversion, in terms of reservoir properties and the distribution of carbonate tidal channels.
It can be concluded that the acoustic properties of the Mishrif reservoir, derived from model based inversion, helped to identify the reservoir features such as channel fairways and high quality facies such as reef and clinoforms. This research focuses on how the comparison of inversion methods, used to improve the characterization of the Mishrif reservoir in this giant oil field, and also for similar heterogeneous carbonate reservoirs.