Determining the Efficacy of Magnetic Susceptibility as an Analytical Tool in the Middle Devonian Gas Bearing Shale of Taylor County, West Virginia.

摘要

The magnetic susceptibility of two Middle Devonian shale units, the Mahantango Formation and Marcellus Shale, was recorded in order to determine if magnetic susceptibility could be used to predict (1) transgressive and regressive cycles, (2) brittleness, and (3) total organic content (TOC). A core from Taylor County, West Virginia was selected for this purpose.;Transgressive and regressive cycles were detected through variations of magnetic susceptibility values with maximum flooding surfaces indicated by troughs in the data and maximum regressive surfaces indicated by peaks. A sequence stratigraphic framework based upon variations in gamma ray and density measurements was used to establish a standard to gauge the accuracy of predictions made through magnetic susceptibility. It was found that the accuracy of the magnetic susceptibility method was similar to the gamma-density method in detecting a large 2nd order cycle, when both shale units were evaluated together. When the units were evaluated separately, it was found that both methods detected the same 3rd order cycles. However, within the Mahantango Formation the magnetic susceptibility method was more accurate recording 4th order cycles that the gamma-density method did not. Conversely, within the Marcellus Shale, the gamma-density method was more accurate recording 4th order cycles that the magnetic susceptibility method did not. It was concluded that the increased accuracy of the gamma-density method in the Marcellus shale was due to an increased sensitivity in the gamma ray and density logs as a response to the large amounts of TOC in the formation This increased sensitivity allowed for smaller variations to be more easily detected. The Mahantango Formation does not have large quantities of TOC. This diminished the sensitivity of the gamma and density logs allowing for the magnetic susceptibility method to be more accurate.;It was assumed that variations in brittleness are driven by transgressive and regressive cycles with ductile regions coinciding with maximum flooding surfaces and brittle regions coinciding with maximum regressive surfaces. Since magnetic susceptibility is also linked to transgressive and regressive cycles, it was expected that the sequence stratigraphy established previously could be linked to brittle and ductile couplets. The sequence stratigraphy established with the gamma-density method was also tested. Two methods of calculating a brittleness index were used to establish a baseline to test against. The first was based upon mineralogy and the second was based upon elastic properties. The brittleness index based upon mineralogy compared well to 3rd order transgressive and regressive cycle detected by both methods. Magnetic susceptibility failed to detect 4th order cycles within the Marcellus Shale, but detected 4th order cycles in the Mahantango Formation that mineralogy and elastic properties missed. The brittleness index based upon mineralogy aligns better with the sequence stratigraphy produced by the magnetic susceptibility method than with the gamma-density method. The brittleness index produced by elastic properties did not correlate with either method. Further, the brittleness index produced with elastic properties did not even correlate with the brittleness index produced by mineralogy. This disagreement cast doubt upon the effectiveness of both brittleness indices.;The relationship between magnetic susceptibility and TOC was tested along with the three most common wireline techniques in order to determine which, if any, were superior. All were judged in comparison to TOC derived from the core using Rock-Eval pyrolysis. The methods tested were: Schmoker's 1983 density equation, Schmoker's 1993 density equation, and the DeltalogR method. It was found that both the magnetic susceptibility of the whole rock and the magnetic susceptibility of its isolated kerogen component correlated better with core TOC values than any of the methods tested. The accuracy of the magnetic susceptibility of the whole rock was within the same order of magnitude as the other methods, and the accuracy of the magnetic susceptibility of the isolated kerogen component was an order of magnitude higher. In addition, evidence was found that links the magnetic susceptibility of kerogen within the two units to the composition of the kerogen. Vitrinite reflectance data confirms that variations in the magnetic susceptibility of the kerogen was not caused by variations in maturity. A very strong logarithmic relationship was found between the magnetic susceptibility of kerogen and the weight percent present. Using the hypothesis that variations in the amount of organic material present is linked to episodic algal blooms, it was concluded that the organic material supplied by these blooms significantly lowered the magnetic susceptibility of the organic sediment supplied during the normal habitat of the basin.