A Multidisciplinary Integrated Approach to Natural Fracture Detection, Characterization and Modeling and Its Application

Abstract

Abstract Understanding anisotropic and heterogeneous characteristics of fractures in the seismic and fluid flow leads to a multidisciplinary approach which can efficiently reduce the uncertainties of the fracture model. This paper presents an integrated workflow in which multi discipline techniques incorporating static and dynamic field data sets deliver an understanding of fracture impacts on well performance and reservoir dynamic behavior at different stages of field development. Imaging logs are used to identify fracture sets and quantify fracture variations of the density and orientation with the reservoir depth. The conductivity of interpreted fractures is calibrated with geomechanical analysis and PLT data. Wave propagation modeling illustrates multiset fractures at different scales can be better detected and characterized with azimuthal dependent seismic data. With defined fracture sets, azimuthal seismic geometry attributes, such as seismic dip, curvature and coherence, are used to detect and image spatial distributions of fracture corridors. 3D fracture corridor network model is generated incorporating fracture characteristics derived from image logs and seismic facies analysis. The diffuse fracture density field, at the seismic scale, can be characterized with azimuthal seismic amplitude and impedance attributes. 3D diffuse fracture density model is constructed with stochastic simulation technique conditioned by fracture density at the log scale and diffuse fracture density field at the seismic scale. The 3D fracture model is built by nesting large scale fracture network model and diffuse fracture density model. Fracture dynamic behavior and flow anisotropy are assessed with mud loss data, gas/water breakthrough, well productivity and PLT data. The assessments provide an understanding of fracture impacts on field production performance and provide insights into locating new wells. This integrating makes fracture model more realistic than purely stochastic fracture models that only honor the statistic field data. In this paper, case studies are used to illustrate applications of proposed approach and its efficiency. Introduction Paleocave reservoirs are usually the product of coalesced collapsed-paleocave systems, which form an important class of carbonate reservoirs. Features and origins of fractures, breccias, and sediment fills associated with coalesced, collapsed paleocave reservoirs have been studied (Kerans, 1988; Loucks and Handford, 1992; Loucks, 1999; Loucks et al 2004; McDonnell et al, 2007). Paleocave reservoirs consist of karst / collapsed related products such as collapsed caves, sinkholes, channels and fracture conduits. The horizontal wells in the paleocave reservoirs are targeting dissolution channels and vertical fractures. Water production within the reservoirs is one of the most interesting challenges of the study. When comparing the water breakthrough dates or the water-cut levels on a well by well basis, it is difficult to give an explicit explanation to the water production behavior. However, it is noted that fracture corridors (conduits) have an impact in term of early water breakthrough at some wells. The objective of this paper is to determine the location of potentially deformed/fractured areas from Ordovician paleocave carbonate reservoirs in the Tarim basin using integrated approach to fracture characterization and modeling. The analyzed results are a fracture model consisting of fracture corridor (conduit) networks and diffuse fracture density filed. Locating different type of fractures and quantifying their distributions are imperative for the success of horizontal wells.