Am I Really Predicting Natural Fractures in the Tight Nordegg Gas Sandstone of West Central Alberta? Part I: Theory and Statistics

Exploration and development of the Nordegg in West Central Alberta is challenging because this prolific zone is deep, structured, and has low permeability. The reservoir is quite thick and is charged with gas. We know from core, samples, well log, and drilling data that the Nordegg zone is fractured to some degree. We expect this fracturing will affect the behavior of the Nordegg to our operations in many ways, from the drilling, to the way that the reservoir behaves under fracture stimulation, and even to the production rates we may achieve from the zone. The fracture density and the production capability of wells drilled into the Nordegg vary materially. In order to address the questions surrounding the effect of fractures on our operations, we must first demonstrate that we can accurately predict the orientation and density of the fracture systems within our area of interest. 3-D surface seismic techniques such as Azimuthal AVO (AVAz), Azimuthal Velocity Analysis (VVAz), Curvature, and Coherence techniques have all been used to predict fractures in a qualitative way in the past. Our approach to the fracture prediction problem will contrast these earlier qualitative attempts, and attempt to cast the problem within the framework of an objective, scientific experiment. Very little information exists in the literature regarding the true accuracy and validity of fracture prediction from surface seismic, so the importance of this endeavor and the need for care in its undertaking is magnified. Our goal is a quantitative analysis of fracture prediction techniques for the Nordegg using objective and scientific validation data. Our validating data includes fracture density data measured from Fullbore Micro Imager Log (FMI) recorded in two horizontal wells as well as microseismic event data recorded over one of the two wells. In part one of this effort, we will describe the experimental set-up and make observations regarding the statistical nature of the FMI data and our seismic attributes. The most crucial result illustrated in part one is that the FMI data clearly illustrates the fractures are almost uniformly vertical and aligned. This satisfies key theoretic requirements of AVAz and VVAz.