--> Quantification of the Geometry and Compartmentalization of Fluvial Meander-Belt Reservoirs: Empirical Insight From Ancient and Modern Analogs

AAPG Annual Convention and Exhibition

Datapages, Inc.Print this page

Quantification of the Geometry and Compartmentalization of Fluvial Meander-Belt Reservoirs: Empirical Insight From Ancient and Modern Analogs

Abstract

Preserved fluvial meander belts commonly form hydrocarbon reservoirs that comprise mosaics of sand-prone barform elements bordered by genetically related, muddy channel fills. The degree of compartmentalization of meander-belt reservoirs by mud-prone abandoned channel fills is an important control on the effectiveness of primary and enhanced oil recovery. Although this problem is widely recognized, it is typically described and considered at a qualitative level. Through analysis of a large sedimentological database, a quantitative characterization of outcrop, modern, and subsurface systems has been undertaken, the aim being to assess the relative importance of different styles of lithological compartmentalization, and to provide constraints that can be applied to inform predictions of the geometry and connectivity of bar-scale sandbodies in meander-belt reservoirs. These quantitative constraints can be applied either directly, by guiding water-flood programs, or through their incorporation into reservoir models, and consist of: (i) a set of empirical relationships that relate dimensional parameters describing the geometry of point-bar elements, associated channel fills, and resulting potentially un-swept compartments; (ii) probabilistic models that relate well density to both the proportion of compartments intersected by a well array and the maximum volume of untapped barform compartments. The method has several applications: (i) prediction of the likely volume of point-bar reservoir compartments with potential un-swept oil occurring in the upper parts of channel belts; (ii) prediction of the proportion of bars and compartments that are not penetrated by a well array of a certain spacing; (iii) guidance in optimal well placement, by providing a measure of the likely size and spacing of bars; (iv) optimization of infill drilling strategy, by providing a measure of the likely size of bypassed volumes. The database approach can be expanded to the characterization of intra-barform compartmentalization, as determined by the presence and distribution of facies-scale heterogeneities, such as mud drapes or mud-clast conglomerates. Sedimentological and petrophysical data from geologic analogs have been combined and synthesized in numerical models by means of geometric and geostatistical techniques; a detailed case study demonstrates how this approach can be used to investigate dynamic connectivity and its sensitivity to styles of meander-belt facies architecture.