Analysis and interpretation of a hydraulic fracture treatment using offset vertical observation wells and a hydraulic fracture simulator
Access full-text files
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Analysis of hydraulic fracture treatments requires incorporating a wide range of data in order to make useful inferences about fracture properties. For example, microseismic monitoring and production decline analysis can be used to obtain the hydraulic fracture half-length, which is an important parameter for field development. The challenge in using these tools is that the methods used for analysis are open to interpretation and can make it difficult to rely on the results. This thesis integrates data from four horizontal wells that were hydraulically fractured in an unconventional shale play and results from a 2-dimensional hydraulic fracture simulator in order to make qualitative observations about fracture properties. The importance of the data set hinges on nine vertical observation wells that recorded pressure vs. time during the hydraulic fracture treatments. The observation wells were located at different distances and depths from the horizontal wells. This is important because it removes some of the ambiguity associated with making interpretations from microseismic data, production decline analysis, or other methods. Results from modeling and the data set indicated the following: (1) the networks of fractures created from these treatments were volumetric and complex, illustrated by the microseismic data and the pressure signals recorded at the observation wells, (2) microseismicity was generally successful in delineating where fluid progressed during pumping, (3) however, flow of fluid into fractures stimulated during previous stages was aseismic, a manifestation of the Kaiser effect, and (4) during long term production, fluid was not produced from the more distant parts of the reservoir that were pressurized and stimulated during the fracturing treatment. To explain these four observations, we hypothesize that proppant was not transported to the regions of the stimulated rock volume that were most distant from the stimulated wells. The stimulated, but unpropped, fractures in this region evidently lost much of their conductivity after closure that they did not contribute significantly to long term production.