Date of Graduation
Eberly College of Arts and Sciences
Geology and Geography
In recent years, more and more attention is paid to shale gas and hydrocarbon liquids exploration and exploitation in unconventional reservoirs. With the development of horizontal drilling and hydraulic fracturing, production from unconventional reservoirs has been greatly increased. However, not all wells, regions and basins harbor highly successful shale gas and liquids producers. In order to improve the production efficiency and reduce the cost of projects, detailed analysis needs to be undertaken to characterize the reservoir. As opposed to conventional reservoirs, extraction of gas in the unconventional reservoir is dependent not only on the reservoir quality, but also on completion quality. The dissertation focuses on three studies that affect reservoir and completion quality of unconventional reservoir units in the Marcellus Shale of the Appalachian basin.
1) Combined with petrophysical analysis from lab and well logging data, nitrogen adsorption is adopted to explore pore characterization of the organic-rich Marcellus Shale and the overlying organic-lean Mahantango Formation. We found that the isotherms of shale can be seen as composite isotherms, which have features of Type I, Type II and Type IV isotherms. The isotherm in the organic-rich Marcellus Shale is more similar to Type H4 of hysteresis loop, suggesting slit-like pores developed in the Marcellus Shale. Fractal analysis shows that the Marcellus Shale with a significantly higher total organic carbon (TOC) content has a more complex pore structure than the Mahantango Formation. Quantitative analysis shows that TOC content has positive relationships with specific surface area, and micropore volume. It indicates that shale samples with high TOC content can store more gas. Therefore, TOC content is a critical parameter to predict gas storage capacity.
2) Hydraulic fracturing is critical to economic production of shale oil and gas from unconventional reservoirs. Success of completion is closely related to gas production. In this study, we found that three factors appear to control hydraulic fracture stimulation. One is the presence of pre-existing natural fractures. Natural fractures were developed in similar patterns with high dip angle in the Marcellus and overlying Mahantango shale units, and are favorable for vertical fracture propagation. The second factor is reservoir geomechanical characterization. The Marcellus Shale has obvious presence of overpressure. According to the study on stress state using a Mohr’s circle, overpressure in the Marcellus Shale increases the possibility of frictional sliding of pre-existing fractures especially at lower value of least principal stress, and will keep some fractures at certain dip angles open. The unconfined compressive strength (UCS) computed from empirical equations for the Marcellus Shale has significantly lower rock strength. The third factor is fracture barrier. The Onondaga Limestone below the Marcellus Shale serves as a critical fracture barrier in our study area. We found that Onondaga fracture barrier with relative high effective minimum horizontal stress and adequate thickness is more effective to prevent the microseismic growing down to other formations.
3) Microseismic monitoring is a useful tool to detect the hydraulic fractures. By observing the microseismic events recorded during the hydraulic fracturing, we found that some basic microseismic characterization such as event number, dominant microseismic azimuth vary stage by stage. To interpret the variations of this characterization, we found that combination of b-value and D-value can be used to interpret the activation of faults and natural fractures. By employing time-distance plot, presence of faults or natural fractures results in higher hydraulic diffusivity. Natural fractures and stress shadow can be used to explain the variation of dominant microseismic azimuth for the stages near the heel of the horizontal well. By applying stress shadow into an unconventional fracture model (UFM), simulated hydraulic fractures are in agreement with distribution of most microseismic events. The hydraulic fractures in the stress shadow grow towards area of next stage, which may be helpful for the production of next stage. However, for the current stage, as fracturing is not sufficient to create more fractures in corresponding area, it may result in the low production in this stage.
Zhu, Yixuan, "Petrophysical and Geomechanical Characterization of the Marcellus Shale in the Appalachian Basin" (2019). Graduate Theses, Dissertations, and Problem Reports. 4113.