The ultimate goal of reservoir modeling is to obtain a model of the reservoir that is able to predict future flow performance. Achieving this challenging goal requires the model to honor all available static (well log, geological information and 3D seismic) and dynamic (4D seismic and production) data. This paper introduces a general methodology and workflow for reservoir modeling that integrates data from multiple and diverse sources, using a probabilistic approach addressing the possible inconsistency and/or redundancy between various data sources. The workflow is evaluated in the Oseberg Field, located in the Norwegian sector of the North Sea, 140 Km west of Bergen. The focus of this study is the Upper Ness Formation in the Alpha North structural segment of the field. The main contribution of this paper is the inclusion of 4D seismic data, which has previously not been accounted for within this workflow. The reservoir modeling workflow applied to this case study helped modeling the spatial distribution of the channel facies using a multiple-point geostatistical technique while honoring all available data and explicitly accounting for data redundancy. Introduction Creating a model of the reservoir is becoming a common practice during several stages of the reservoir life. From exploration to field abandonment, reservoir modeling pursues the general goal of understanding and predicting important geological, geophysical and engineering components of the reservoir. Reservoir modeling calls for the integration of expertise from different disciplines, as well as the integration of data from various sources. Each type of data provides information about the reservoir heterogeneity on a different scale; therefore, they have different degrees of accuracy and may be redundant towards modeling the reservoir. The reservoir model needs to simultaneously (not hierarchically) honor all available data, both static (well-log, geological information and 3D seismic) and dynamic (4D seismic and production), in order to preserve its predictive capabilities. An approach to integrate static and dynamic data (production and 4D seismic data) for reservoir characterization has been evaluated by Kretz et al (2002) and Mezghani et al (2004), simultaneously integrating different sources of information using an optimization methodology based on the gradual deformation method. In this paper we rely on a probabilistic scheme for data integration of well log, geological information 3D/4D seismic and production data, within the probability perturbation method (Caers, 2003). A fully probabilistic methodology and workflow for reservoir modeling that integrates data from multiple and diverse sources is proposed, using an approach that explicitly addresses the possible inconsistency and/or redundancy between various data sources (Hoffman, 2005). The general goal of the workflow is to model an unknown (facies or petrophysical property) using data from different sources. The information content of each data source is modeled as a spatial probability distribution model; using Journel's tau model (Journel, 2002) these individual spatial probabilities are combined into a simple joint conditional probability from which reservoir models are drawn using sequential simulation.

Small-scale (< 20 m), non-resolvable sand injectites can constitute a large part of the net-to-gross volume and affect fluid flow in the reservoir. However, they may also cause challenges for well placement and reservoir development because they are too small to be reliably constrained by reflection seismic data. It is therefore important to better understand how small-scale injectites influence seismic images and may be recognized and characterized above reservoirs. The Grane Field (North Sea) hosts numerous small-scale sand injectites above the main reservoir unit, causing challenges for well placement, volume estimates and seismic interpretation. Here, we investigate how such small-scale sand injectites influence seismic images and may be characterized by (1) using well-, 3D seismic- and outcrop data to investigate geometries of small-scale sand injectites (0-15 m) and creating conceptual models of injectite geometries, (2) performing seismic convolution modelling to investigate how these would be imaged in seismic data, and (3) compare these synthetic seismic images to actual 3D seismic from the well-investigate Grane Field.Our results show that despite injectites being below seismic resolution, small-scale sand injectites can be detected in seismic data. They are more likely to be detected with high thickness (> 5 m), steep dip (> 30°), densely spaced sand injectites, and homogeneous background stratigraphy. Furthermore, as fraction of sand injectites increases the top reservoir amplitude will decrease. Moreover, comparison of the synthetic seismic images with real seismic data from the Grane Field indicates that the low-amplitude anomalies and irregularities observed above the reservoir may be a result of the overlying sand injectites. Additionally, the comparison strongly suggests that the Grane Field hosts sand injectites that are thicker and located further away from the top reservoir than what is indicated by well observations. These results may be used to improve well planning and develop reservoirs with overlying sand injectites.Supplementary material: A PDF file containing all the seismic modelling results allowing the reader to flip back and forth between the different models is available at https://www.doi.org/10.6084/m9.figshare.14333102 . Well logs from well 25/11-18 T2 are available at https://factpages.npd.no/pbl/wellbore_documents/2358_25_1_18_COMPLETION_REPORT_AND_LOG.pdf

No abstract

No abstract

A work flow to improve the modeling of a fluvial reservoir is presented. Modeling of fluvial reservoirs can be very uncertain when only conditioned to well data. By utilising 3D and 4D seismic inversion data to condition the geological model, the uncertainty in the facies distribution is reduced. Cross plot of 3D elastic inversion data predicts facies better than acoustic impedance data or Vp/Vs data individually. By including 4D elastic inversion data, the correlation between the classified facies from seismic and the facies zonation in the wells, is further improved. A sand probability cube is computed from 3D and 4D elastic inversion data, and used to condition the geological model of the fluvial reservoir. Upscaling and flow simulation are carried out on a set of realisations. This results in a range of simulated production profiles. The dynamic information in the 4D seismic is then used to rank the realisations by comparing the real 4D data with forward modelled elastic parameters from the flow simulation model. The described work flow is a new way of integrating different types of data in the modeling process in order to reduce uncertainties. The method is applied to the fluvial Ness reservoir in the Oseberg field in the North Sea. Introduction Reservoir modeling and flow simulation contribute to reservoir management by predicting the reservoir response in term of production rates and total recovery. Stochastic models and simulations have been used to give a realistic image of the uncertainty in describing the hetereogeneities of the reservoir. A great challenge is to combine and utilize all available data: core and log data, outcrop studies, seismic data and production history, if available. Traditional reservoir modelling utilizes well logs and, in some cases, 3D seismic data. More information can be extracted from the seismic by taking 4D data into account. Cross plot of elastic parameters from a seismic inversion process can, in many cases, be used as input to lithology classification. Several publications have shown that it is possible to do litho-classification by cross plotting Acoustic impedance (AI) and the ratio between the compressional and the shear velocities (Vp/Vs) (Ødegaard and Avseth, 2004, Avseth et. al. 2005, Coléou et. al. 2005). Cross plot of the changes in elastic parameters from 4D seismic can be used to classify 4D effects into saturation related or pressure related changes. The proposed workflow is combining both 3D and 4D elastic inversion data to classify lithology. In the case of the 4D seismic, it is assumed that some of the observed production effects mainly can be related to sand facies only. By utilising this, it is possible to achieve a sand probability cube that has highest probabilities when both 3D and 4D seismic are utilised. The workflow has been developed and tested for the fluvial Upper Ness reservoir unit in the Alfa Nord structure in the Oseberg Field. The Oseberg Field is located on a tilted fault block in the Horda Platform area, in the Norwegian North Sea sector, see Figure 1.