John W. Shimeld
 |
 |
 |
|
M. Sc. Thesis
Development of an Algorithm to Detect Subsurface Fractures Using Conventional Well Logs and Fuzzy Inference: Practical Application at the Terra Nova Oil Field, Offshore Newfoundland
(PDF - Mb)
It is implicit in many studies that subsurface fractures are an integral part of the geologic processes that occur in the Earth's crust; fractures are both a control on and a result of fluid/rock interaction. Examples of the importance of fractures are found in diverse areas such as groundwater contaminant transport modelling, studies of hydrothermal ore deposit gensis, prediction of earthquake failure mechanisms, and hydrocarbon charge modelling. Despite their importance, detection of subsurface fractures remains a difficult problem for geologists and geophysicists.
The algorithm developed in this thesis approaches the problem through the use of conventional well logs, which are usually the msot readily available and detailed source of fracture information. The use of conventional well logs is not straightforward, though, since each well logging tool responds to a different set of bulk rock properties from the surrounding mass and no single well log signature is diagnostic of fracturing. Traditionally, fracture detection with conventional well logs is a qualitative and subjective exercise that relies upon the simultaneous interpretation of multiple well logs. The fracture detection algorithm developed in this thesis used fuzzy inference as a tool to manage uncertainty in well log interpretations, and to calculate the relative likelihood of fracturing down the length of a borehole in a uniform, automated, and semi-quantitative manner.
The algorithm is tested using two case studies that illustrate situations where convential well logs are suitable, and where they are not suitable, for fracture detection. Then the Terra Nova oil field, located in the Jeanne d'Arc Basin offshore Eastern Canada, is used as a site for practical application of the fracture detection algorithm. These results are integrated with interpretation of 3D seismic data and fission track analysis to provide a detailed description of fracturing and its influence on fluid distribution at Terra Nova.
Keywords:
Pages:
Supervisor: Mark Williamson