We present a novel sampling algorithm for characterization and uncertainty quantification of heterogeneous multiple facies reservoirs. The method implements a Bayesian inversion framework to estimate physically plausible porosity distributions. This inversion process incorporates data matching at the well locations and constrains the model space by adding
The proposed workflow uses an ensemble-based Markov Chain Monte Carlo approach combined with sampling probability distributions that are physically meaningful. Moreover, the method targets geostatistical modeling to specific zones in the reservoir. Accordingly, it improves fulfilling the inherent stationarity assumption in geostatistical simulation techniques. Parameter sampling and geostatistical simulations are calculated through an inversion process. In other words, the models fit the known porosity field at the well locations and are structurally consistent within main reservoir compartments, zones, and layers obtained from the seismic impedance volume. The new sampling algorithm ensures that the automated history matching algorithm maintains diversity among ensemble members avoiding underestimation of the uncertainty in the posterior probability distribution.
We evaluate the efficiency of the sampling methodology on a synthetic model of a waterflooding field. The predictive capability of the assimilated ensemble is assessed by using production data and dynamic measurements. Also, the qualities of the results are examined by comparing the geological realism of the assimilated ensemble with the reference probability distribution of the model parameters and computing the predicted dynamic data mismatch. Our numerical examples show that incorporating the seismically constrained models as prior information results in an efficient model update scheme and favorable history matching.
Bashir, Yasir (Universiti Teknologi PETRONAS) | Babasafari, Amir Abbas (Universiti Teknologi PETRONAS) | Biswas, Ajay (Universiti Teknologi PETRONAS) | Hamidi, Rositi (Universiti Teknologi PETRONAS) | Moussavi Alashloo, Seyed Yaser (Universiti Teknologi PETRONAS) | Tariq Janjuah, Hammad (American University of Beirut) | Prasad Ghosh, Deva (Universiti Teknologi PETRONAS) | Weng Sum, Chow (Universiti Teknologi PETRONAS)
A majority of remaining proven Oil & Gas reserves is contained by Carbonate reservoir, and much more complicated to explore as imaging of the Carbonate rocks is poor. In case of Carbonate data, seismic diffraction imaging has contributed to an enhancement in the quality of seismic but there is still lack of understanding the lithology and impedance contrast which can be defined by the seismic inversion. In contrast, to the conventional process, an integration of seismic inversion methods are necessary to understand the lithology and include the full band of frequency in our initial model to incorporate and detail study about the basin for prospect evaluation. In this paper, an integrated approch is developed for better deleniation of subsurface structure and lithologies. Seismic post stack inversion technique is applied to the Carbonate field to study Electroficies and lithofacies of subsurface strata for better and detail study of the reservoir.
Acoustic impedance (AI) is often inversely proportional to porosity. This has led to a seismic reservoir characterization technology where the AI from seismic inversion is used to predict porosity away from well locations. The porosity is typically predicted from the absolute seismic AI derived from a model based inversion (MBI). There is an inherent error in AI derived from MBI due to the accuracy of the low frequency model (LFM) provided to the inversion engine, introducing an additional error in the predicted porosity. Here we show an use of relative AI after seismic Coloured Inversion to predict porosity in a target zone. This is possible if the seismic data have good low frequency content and if the target zone is small and appropriate for the field development. Within a zone of interest the sensitivity to the effects of the LFM is minimum and therefore relative AI can be sufficient to predict porosity. An Eagle Ford Shale example from the South Texas is used in this study.
Recently the Mississippi Lime has become one of the most active resource plays. Our study area falls in-between the Fort Worth and Midland Basins. The main production comes from high porosity tripolitic chert. Our objective is to use 3D seismic data to map the areal distribution of discontinuous tripolitic facies.
In the early 1990s several 3D surveys were shot in the study area to image shallower objectives. With the advent of the Mississippi Lime play, four of these surveys were merged and reprocessed using careful statics and velocity analysis. Even after prestack time migration, the target zone is contaminated with the acquisition footprint. The data are low (~15) fold and contaminated by highly aliased, high frequency, high amplitude ground roll. Given the sparsity of the survey, modern f-kx-ky filters were not able to remove ground roll prompting the development of a new ground roll suppression workflow. In workflow, we first window and low-pass filter (f<50 Hz) the data, 3D patch by 3D patch. We then apply linear moveout to approximately flatten the ground roll phases, estimate the dip about this reference moveout, and compute coherence within a 3- channel by 3-shot by 20 ms window for each sample. Using a Kuwahara algorithm, we choose the most coherent window within which we apply a structure-oriented KL filter. At the end we simply modeled the ground roll from the original data. This 3D filter preserves signal amplitude and is flexible enough to model the piece wise continuous ground roll pattern common with irregular topography.