Workstation-Based Fracture Evaluation Using Borehole Images and Wireline Logs

Abstract

Abstract Wireline images and logs can be processed effectively to derive quantitative information relating to fracture orientation, den. sity, aperture and porosity. A methodology based on analyses at the level of individual fractures and fracture sets is presented. A program has been implemented on a workstation presented. A program has been implemented on a workstation allowing a flexible blend of expert-user input and machine processing power. processing power. Introduction In recent years, there have been important advances in the evaluation of fractures using wireline logs. Two of the major developments are :the extensive availability of imaging tools, andtechniques to estimate fracture aperture. These developments have enabled fracture evaluation from logs to progress beyond fracture detection to a quantitative process. progress beyond fracture detection to a quantitative process. Imaging of the borehole wall is achieved using either ultrasonic or electrical scanning methods. Images allow fractures intersecting the borehole wall to be 'directly' observed. Consequently, fracture detection becomes a much less ambiguous process, and in principle, important statistics such as fracture count, orientation and spacing can be obtained. Other information which are evident from images, such as the shape of fractures, their relation to bedding, and the direction of borehole elongation, also contribute towards a better understanding of the fracture system. For the estimation of fracture aperture, several techniques use ing the response of resistivity logs, the reflection of Stoneley waves, and the electrical scanner response itself, have been proposed. These allow, for example, the producibility and proposed. These allow, for example, the producibility and porosity of the fractures to be assessed. porosity of the fractures to be assessed. To fully exploit these developments on a routine basis, however, one must overcome a serious problem of processing and integration which arises from the quantity and diversity of the information contained in the logs, especially the images. In this paper, we demonstrate that this problem can be tackled effectively using a computer workstation. We shall present the methodology and illustrative examples. Mathematical formulae and algorithms are given in the appendices. The problem of extrapolating measurements made on small samples (the volume close to the borehole in the case of wire line measurements) to the fracture system as a whole, is beyond the scope of this paper. The quantitative data obtained from wireline images and logs can however serve as input for methods such as that proposed recently by Howard and Nolen-Hoeksema. Methodology We have adopted a methodology which consists of three main steps: Characterization of individual fractures, Classification of fractures into sets, and, Output of fracture logs. In the first step, the attributes of individual fractures are determined from the log responses. In the second, the statistical properties of the fracture population are ascertained. Both properties of the fracture population are ascertained. Both steps require extensive data manipulation, which is provided by the computer program, and judgements of interpretation, which are input by the user. The information generated by these two steps may be summarized, if desired, in the form of logs giving, for example, the fracture density, porosity, etc. for each fracture set. P. 465