Effects of conductive well casings on electromagnetic surveys: experimental studies and numerical modelling, The

Abstract

Controlled source electromagnetic (CSEM) surveys can be used in petroleum exploration to supplement seismic reflection data because electrical resistivity is highly sensitive to pore space fluid content. Traditionally, CSEM has been used in greenfields exploration, but the method has the potential for application later in the life of a producing field. However, such monitoring would be performed in the presence of electrically conductive steel infrastructure, such as well casings and pipelines. These features introduce considerable artifacts into data gathered for development, production and monitoring purposes. Numerical techniques can be used to model and remove the electromagnetic response from steel infrastructure; one such is the Method of Moments (MoM), which is being actively developed at Mines and will be used throughout this thesis to support experimental data. Although characterizing the effects of steel infrastructure on EM data is of ultimate interest to industry, much can be learned by studying the zero-frequency DC resistivity counterpart of the finite frequency (AC) problem. Throughout this thesis I theoretically examine the relationship between EM data and well casing properties to best design proper experiments at the lab and field scales. In particular, I examine the relationships between the casing response and properties such as length, width, wall thickness and material composition of the casing. I then describe experimental DC resistivity data that I have collected at three scales: lab, mid-, and large-scale and compare results to those predicted by MoM. Modelled responses in zones dominated by the casing response show differences of less than 20% when compared with experimental data, providing confidence in the MoM method. Results show that the most important parameters affecting casing responses are the orientation of the casing with respect to survey geometry, total volume of metal present in the casing, and the casing material - an important finding as industry casings vary extensively in physical properties. Finally, I discuss a series of AC modelling tests conducted at the same three scales to understand the viability of further EM experiments. Results indicate that at the mid-scale and large-scale, such experiments should be possible using transmitting frequencies of 50 kHz and 1 kHz, respectively; however, lab-scale tests will require transmitting frequencies too high for standard EM equipment.