Joint Stochastic Inversion Of Petrophysical Logs And 3D Pre-Stack Seismic Data To Assess The Spatial Continuity Of Fluid Units Away From Wells: Application To A Gulf-Of-Mexico Deepwater Hydrocarbon Reservoir

This paper describes a novel methodology to integrate well logs and 3D pre-stack seismic data. The objective is to assess lateral continuity and spatial extent of lithology and fluid units penetrated by a well. Pre-stack seismic data were used to fill the spatial gap between sparse well locations since they embody the degrees of freedom necesary to uniquely interpret lateral variations of seismic amplitude in terms of variations of lithofacies and petrophysical properties. The proposed approach is based on a stochastic global inversion method that concomitantly honors the well logs and multiple angle stacks of seismic amplitude data. Inversion results consist of 3D spatial distributions of acoustic properties, litho-facies, and petrophysical parameters between wells that exhibit a vertical resolution intermediate between that of well logs and 3D seismic data. Examples of the application of this technique are shown using high-quality 3D seismic data acquired in the deepwater Gulf of Mexico. Reservoir units consist of stacked turbidite sands. Conventional petrophysical interpretation based on well logs and rock-core data was performed for 7 wells. Petrophysical and lithofacies logs were constructed and correlated with elastic parameters inferred from Pand S-wave sonic logs to assess the sensitivity of elastic parameters to variations in porosity and fluid saturation. Both petrophysical logs and elastic-petrophysical correlation cross-plots, together with four angle stacks of pre-stack seismic amplitude data, were entered to the stochastic inversion algorithm to produce 3D distributions of litho-facies, porosity, permeability, and fluid saturation. Results successfully describe the spatial continuity of sand units and of their porosity, permeability, and saturating fluids away from wells, showing the efficiency of the technique for quantitative integration of well logs and pre-stack seismic data. INTRODUCTION Anadarko's Marco Polo deepwater development project is located in Green Canyon Block 608 in the Gulf of Mexico, approximately 175 miles south of New Orleans, in a 4300' water depth environment (Fig. 1). Hydrocarbon production originates from reservoirs consisting of Tertiary deepwater sand deposits. This paper considers a small portion of the Marco Polo Field where hydrocarbon-bearing sand units pertain to the “M” series and are buried at depths between 11500 and 12500 ft (Figs. 2 and 3). The overall “M” series consists of sandy turbidite reservoir deposits interbedded and separated by muddy debris flows. These reservoir intervals are interpreted as stacked, progradational lobes within an overall fan complex. The massive and planar stratified sands exhibit excellent interparticle porosity. Rock-core measurements indicate excellent intrinsic properties: 30%+ porosity, and 100-4000 millidarcies of nominal permeability. The purpose of this research is to assess the lateral continuity and spatial extent of M-series lithology and fluid units penetrated by wells by using pre-stack seismic data. Well logs exhibit a radial length of investigation shorter than 3 m and hence provide limited indication of lateral extent and continuity of reservoir flow units. In the past, post-stack seismic data have been used to fill the spatial gap between sparse well locations. However, post-stack seismic data respond to acoustic impedance (the product of bulk density and P-wave velocity) and, therefore, cannot always uniquely discriminate between spatial variations of porosity, thickness, shale concentration, and fluid saturation. Pre-stack seismic data, on the other hand, are sensitive to S-wave velocity and bulk density in addition to P-wave velocity. This provides additional degrees of freedom to uniquely interpret lateral variations of seismic amplitude in terms of variations of petrophysical properties and flow-unit thickness.