The Mississippian Barnett Formation is a prolific shale-gas reservoir that was deposited in the Fort Worth Basin, Texas. Many previous studies of the Barnett Formation have been conducted in the main production area; few studies have been made of the Barnett Formation in the southern part of the basin, which is a less productive area. In the present research, several cores from the Barnett Formation in Hamilton County, southeast Fort Worth Basin, are studied in detail. Two vertical, continuous cores from Hamilton County, Texas, were studied to delineate the depositional setting, lithofacies, pore types, and reservoir quality of the Barnett Formation in the area. Five lithofacies were defined by analysis of the two cores: (1) laminated clay-rich silty and skeletal peloidal siliceous mudstone; 2) laminated skeletal silty peloidal siliceous mudstone; 3) nonlaminated silty peloidal calcareous mudstone; 4) laminated and nonlaminated skeletal calcareous mudstone; and 5) skeletal phosphatic packstone to grainstone. As indicated from this study, the dominant organic matter type is a mixture of Type II (major) and Type III (minor) kerogen having a mean TOC content of approximately 4%. Analysis of Rock Eval data shows that most of the interval is within the oil window; calculated Ro is approximately 0.9%. Organic geochemistry shows that the hydrocarbon generation potential of the abundant oil-prone kerogen was excellent. Mineralogical analysis reveals that the two types of siliceous mudstone, which are similar in composition to the siliceous mudstone in the main producing area in the northern Fort Worth Basin, are good for hydraulic fracturing and production, but they are also limited by their marginal thickness. Organic matter pores, which are the dominant pore types in these two cores, are consistent with pore types found in currently producing wells in the Newark East Field. This research also suggests that the deposition of Barnett Formation was controlled largely by basinal geometry, suspension settling, and slope-originated gravity-flow events. Skeletal deposits and carbonate-silt starved ripples suggest gravity-flow deposits and bottom-current reworking during deposition. Redox-sensitive elements and degree of pyritization both indicate anoxic/euxinic conditions during the deposition of the Barnett Formation. / text

Unconventional natural gas extraction from the Marcellus Shale requires millions of gallons of water to fracture shale and release natural gas from the formation. This process produces water with high levels of total dissolved solids (TDS); and, efforts to recycle these fluids has stimulated microbial growth in produced water. The objective of this study was to analyze the ionic composition of and characterize microorganisms from Marcellus produced water samples. A semi-synthetic culture medium was designed with high TDS to enrich for halophilic microbes, which yielded robust cultures that were able to grow over a wide range of salinities. DNA extracted from aerobic cultures was used for 16s rDNA clone libraries and Automated Ribosomal Intergenic Spacer Analysis (ARISA). ARISA and 16S gene sequencing revealed differences in bacterial composition between Marcellus and freshwater samples. Sequencing of 16S gene indicated the presence of Halomonas, Thalassospira and other genera related to halophilic and petroleum degrading species. / Bayer School of Natural and Environmental Sciences / Environmental Science and Management (ESM) / MS / Thesis

Time-rate analysis and time-rate-pressure analysis methods are available to estimate reserves and study flow performance of wells in unconventional gas reservoirs. However, these tools are often incorrectly used or the analysis can become difficult because of the complex nature of the reservoir system. Conventional methods (e.g., Arps' time-rate relations) are often used incorrectly to estimate reserves from such reservoirs. It was only recently that a serious study was conducted to outline the limitations of these relations and to set guidelines for their correct application. New time-rate relations, particularly the Duong and logistic growth model, were introduced to estimate reserves and forecast production from unconventional reservoirs. These new models are being used with limited understanding of their characteristics and limitations. Moreover, well performance analyses using analytical/semi-analytical solutions (time-rate-pressure) are often complicated from non-uniqueness that arises when estimating well/formation properties. In this work, we present a detailed study of the Duong model and logistic growth model to investigate the behaviors and limitations of these models when analyzing production data from unconventional reservoirs. We consider production data generated from numerical simulation cases and data obtained from unconventional gas reservoirs to study the quality of match to specific flow regimes and compare accuracy of the reserve estimates. We use the power-law exponential model (PLE), which has been shown to model transient, transition and boundary-dominated flow regimes reliably, as a benchmark to study performance of Duong and logistic growth models. Moreover, we use the "continuous EUR" approach to compare these models during reserve estimation. Finally, we develop four new time-rate relations, based on characteristics of the time-rate data on diagnostic plots. Using diagnostic plots we show that the new time-rate relations provide a quality match to the production data across all flow regimes, leading to a reliable reserve estimate. In a preliminary study, we integrated time-rate model parameters with fundamental reservoir properties (i.e., fracture conductivity (Fc) and 30 year EUR (EUR30yr)), by studying 15 numerical simulation cases to yield parametric correlations. We have demonstrated a methodology to integrate time-rate model parameters and reservoir properties. This method avoids the non-uniqueness issues often associated with model-based production data analysis. This study provides theoretical basis for further demonstration of the methodology using field cases.

Fractured reservoirs have been traditionally idealized using dual-porosity models. In these models, all matrix and fractures systems have identical properties. However, it is not uncommon for naturally fractured reservoirs to have orthogonal fractures with different properties. In addition, for hydraulically fractured reservoirs that have preexisting natural fractures such as shale gas reservoirs, it is almost certain that these types of fractures are present. Therefore, a triple-porosity (dual-fracture) model is developed in this work for characterizing fractured reservoirs with different fractures properties. The model consists of three contiguous porous media: the matrix, less permeable micro-fractures and more permeable macro-fractures. Only the macro-fractures produce to the well while they are fed by the micro-fractures only. Consequently, the matrix feeds the micro-fractures only. Therefore, the flow is sequential from one medium to the other. Four sub-models are derived based on the interporosity flow assumption between adjacent media, i.e., pseudosteady state or transient flow assumption. These are fully transient flow model (Model 1), fully pseudosteady state flow model (Model 4) and two mixed flow models (Model 2 and 3). The solutions were mainly derived for linear flow which makes this model the first triple-porosity model for linear reservoirs. In addition, the Laplace domain solutions are also new and have not been presented in the literature before in this form. Model 1 is used to analyze fractured shale gas horizontal wells. Non-linear regression using least absolute value method is used to match field data, mainly gas rate. Once a match is achieved, the well model is completely described. Consequently, original gas in place (OGIP) can be estimated and well future performance can be forecasted.

It was observed that many hydraulically fractured horizontal shale gas wells exhibit transient linear flow behavior. A half-slope on a type curve represents this transient linear flow behavior. Shale gas wells show a significant skin effect which is uncommon in tight gas wells and masks early time linear behavior. Usually 70-85 percent of frac water is lost in the formation after the hydraulic fracturing job. In this research, a shale gas well was studied and simulated post hydraulic fracturing was modeled to relate the effect of frac water to the early significant skin effect observed in shale gas wells. The hydraulically fractured horizontal shale gas well was described in this work by a linear dual porosity model. The reservoir in this study consisted of a bounded rectangular reservoir with slab matrix blocks draining into neighboring hydraulic fractures and then the hydraulic fractures feed into the horizontal well that fully penetrates the entire rectangular reservoir. Numerical and analytical solutions were acquired before building a 3D 19x19x10 simulation model to verify accuracy. Many tests were conducted on the 3D model to match field water production since initial gas production was matching the analytical solutions before building the 3D simulation model. While some of the scenarios tested were artificial, they were conducted in order to reach a better conceptual understanding of the field. Increasing the water saturation in the formation resulted in increasing water production while lowering gas production. Adding a fractured bottom water layer that leaked into the hydraulic fracture allowed the model to have a good match of water and gas production rates. Modeling trapped frac water around the fracture produced approximately the same amount of water produced by field data, but the gas production was lower. Totally surrounding the fracture with frac water blocked all gas production until some of the water was produced and gas was able to pass through. Finally, trapped frac water around the fracture as combined with bottom water showed the best results match. It was shown that frac water could invade the formation surrounding the hydraulic fracture and could cause formation damage by blocking gas flow. It was also demonstrated that frac water could partially block off gas flow from the reservoir to the wellbore and thus lower the efficiency of the hydraulic fracturing job. It was also demonstrated that frac water affects the square root of time plot. It was proven by simulation that the huge skin at early time could be caused by frac water that invades and gets trapped near the hydraulic fractures due to capillary pressure.

With increase of interest in exploiting shale gas/oil reservoirs with multiple stage fractured horizontal wells, complexity of production analysis and reservoir description have also increased. Different methods and models were used throughout the years to analyze these wells, such as using analytical solutions and simulation techniques. The analytical methods are more popular because they are faster and more accurate. The main objective of this paper is to present and demonstrate type curves for production data analysis of shale gas/oil wells using a Dual Porosity model. Production data of horizontally drilled shale gas/oil wells have been matched with developed type curves which vary with effective parameters. Once a good match is obtained, the well dual porosity parameters can be calculated. A computer program was developed for more simplified matching process and more accurate results. As an objective of this thesis, a type curve analytical method was presented with its application to field data. The results show a good match with the synthetic and field cases. The calculated parameters are close to those used on the synthetic models and field cases.

Production forecasting in shale (ultra-low permeability) gas reservoirs is of great interest due to the advent of multi-stage fracturing and horizontal drilling. The well renowned production forecasting model, Arps? Hyperbolic Decline Model, is widely used in industry to forecast shale gas wells. Left unconstrained, the model often overestimates reserves by a great deal. A minimum decline rate is imposed to prevent overestimation of reserves but with less than ten years of production history available to analyze, an accurate minimum decline rate is currently unknown; an educated guess of 5% minimum decline is often imposed. Other decline curve models have been proposed with the theoretical advantage of being able to match linear flow followed by a transition to boundary dominated flow. This thesis investigates the applicability of the Stretched Exponential Production Decline Model (SEPD) and compares it to the industry standard, Arps' with a minimum decline rate. When possible, we investigate an SEPD type curve. Simulated data is analyzed to show advantages of the SEPD model and provide a comparison to Arps' model with an imposed minimum decline rate of 5% where the full production history is known. Long-term production behavior is provided by an analytical solution for a homogenous reservoir with homogenous hydraulic fractures. Various simulations from short-term linear flow (~1 year) to long-term linear flow (~20 years) show the ability of the models to handle onset of boundary dominated flow at various times during production history. SEPD provides more accurate reserves estimates when linear flow ends at 5 years or earlier. Both models provide sufficient reserves estimates for longer-term linear flow scenarios. Barnett Shale production data demonstrates the ability of the models to forecast field data. Denton and Tarrant County wells are analyzed as groups and individually. SEPD type curves generated with 2004 well groups provide forecasts for wells drilled in subsequent years. This study suggests a type curve is most useful when 24 months or less is available to forecast. The SEPD model generally provides more conservative forecasts and EUR estimates than Arps' model with a minimum decline rate of 5%.

This thesis introduces a rock typing method for application in shale gas reservoirs using conventional well logs and core data. Shale gas reservoirs are known to be highly heterogeneous and often require new or modified petrophysical techniques for accurate reservoir evaluation. In the past, petrophysical description of shale gas reservoirs with well logs has been focused to quantifying rock composition and organic-matter concentration. These solutions often require many assumptions and ad-hoc correlations where the interpretation becomes a core matching exercise. Scale effects on measurements are typically neglected in core matching. Rock typing in shale gas provides an alternative description by segmenting the reservoir into petrophysically-similar groups with k-means cluster analysis which can then be used for ranking and detailed analysis of depth zones favorable for production. A synthetic example illustrates the rock typing method for an idealized sequence of beds penetrated by a vertical well. Results and analysis from the synthetic example show that rock types from inverted log properties correctly identify the most organic-rich model types better than rock types detected from well logs in thin beds. Also, estimated kerogen concentration is shown to be most reliable in an under-determined problem. Field cases in the Barnett and Haynesville shale gas plays show the importance of core data for supplementing well logs and identifying correlations for desirable reservoir properties (kerogen/TOC concentration, gas saturation, and porosity). Qualitative rock classes are formed and verified using inverted estimates of kerogen concentration as a rock-quality metric. Inverted log properties identify 40% more of a high-kerogen rock type over well-log based rock types in the Barnett formation. A case in the Haynesville formation suggests the possibility of identifying depositional environments as a result of rock attributes that produce distinct groupings from k-means cluster analysis with well logs. Core data and inversion results indicate homogeneity in the Haynesville formation case. However, the distributions of rock types show a 50% occurrence between two rock types over 90 ft vertical-extent of reservoir. Rock types suggest vertical distributions that exhibit similar rock attributes with characteristic properties (porosity, organic concentration and maturity, and gas saturation). This method does not directly quantify reservoir parameters and would not serve the purpose of quantifying gas-in-place. Rock typing in shale gas with conventional well logs forms qualitative rock classes which can be used to calculate net-to-gross, validate conventional interpretation methods, perform well-to-well correlations, and establish facies distributions for integrated reservoir modeling in hydrocarbon-bearing shale. / text

Horizontal drilling and hydraulic stimulation make hydrocarbon production from organic-rich shales economically viable. Identification of suitable zones to drill a horizontal well and to initiate or contain hydraulic fractures requires detection and quantification of many factors, including elastic mechanical properties. Elastic behavior of rocks is affected by rock composition and fabric, pore pressure, confining stress, and other factors. Rock fabric refers to the arrangement of the rock’s solid and fluid constituents. The objective of this thesis is to quantify rock fabric properties of hydrocarbon-bearing shales affecting elastic properties, including load-bearing matrix, anisotropic cracks, and shape of rock components. Once rock fabric is validated with sonic logs, results can be used to identify suitable zones to drill a horizontal well, initiate hydraulic stimulation, and contain fracture propagation. We develop a method to estimate elastic properties based on rock composition. The differential effective medium (DEM) theory is invoked to model rock elastic properties with a load-bearing component in which remaining minerals and pores are added as spheres or ellipsoids. The method can be combined with the self-consistent approximation (SCA) to construct a load-bearing matrix made of two solid phases. Anisotropic inclusions are added via Hudson’s model. Subsequently, Gassmann’s theory is invoked to saturate the rock with fluids and determine low-frequency elastic properties for comparison to sonic logs. Rock fabric properties remain constant in a vertically homogeneous formation. In vertically heterogeneous strata, the depth interval of interest is divided into rock types, based on rock solid composition, and each rock type is associated with a specific fabric. Quantification of rock fabric properties is a non-unique process, and one should take into account as much petrophysical and geological information as possible to ensure physically viable results. Our simulation and interpretation method is implemented in two wells in both the Haynesville and Barnett shales. Averages of relative errors between estimated velocities and sonic logs are less than 4% in the four wells. Simulations in the Haynesville shale are isotropic, and therefore indicate that rock fabric may not be the main cause of mechanical anisotropy in cases where such behavior is inferred from field data. Rock fabric properties are constant with depth in both wells. Consequently, identification of suitable zones to drill a horizontal well or to contain fracture propagation is not based on rock fabric; it is deduced from Young’s modulus. Simulated Poisson’s ratio is shown to be more sensitive to errors in velocities than Young’s modulus and is therefore not used in the interpretation. Favorable depth intervals for gas production exhibit sizeable volumes of gas and organic content. In the Barnett shale, the two wells exhibit different rock fabrics. Such a behavior indicates that the formation is laterally heterogeneous. Rock physics models should therefore be extrapolated from one well to another with caution. Simulations assume anisotropic elastic behavior and suggest the presence of compliant horizontal pores in one case. Natural vertical fractures are observed on electric image logs in the remaining case and are modeled with Hudson’s theory. This behavior suggests that rock fabric causes mechanical anisotropy in the formation. Heterogeneity of the Barnett shale rock fabric is inferred from the necessary use of rock typing to adequately reproduce sonic logs in both wells. Intervals with large porosity and high gas saturation identify suitable zones to perform hydraulic stimulation. Among such zones, rock types that exhibit stiff load-bearing matrices (comprising mostly calcite, for example) indicate suitable depths to drill horizontal wells or to contain hydraulic fractures. Intervals with dense layering of different rock types are unsuitable for fracture propagation and should be avoided during hydraulic-fracturing operations. / text

The vast shale gas and tight oil reservoirs in North America cannot be economically developed without multi-stage hydraulic fracture treatments. Owing to the disparity in the density of natural fractures in addition to the disparate in-situ stress conditions in these kinds of formations, microseismic fracture mapping has shown that hydraulic fracture treatments develop a range of large-scale fracture networks in the shale plays. In this thesis, an approach is presented, where the fracture networks approximated with microseismic mapping are integrated with a commercial numerical production simulator that discretely models the network structure in both vertical and horizontal wells. A novel approach for reservoir simulation is used, where porosity (instead of permeability) is used as a scaling parameter for the fracture width. Two different fracture geometries have been broadly proposed for a multi stage horizontal well, orthogonal and transverse. The orthogonal pattern represents a complex network with cross cutting fractures orthogonal to each other; whereas transverse pattern maps uninterrupted fractures achieving maximum depth of penetration into the reservoir. The response for a vii single-stage fracture is further investigated by comparing the propagation of the stage to be dendritic versus planar. A dendritic propagation is bifurcation of the hydraulic fracture due to intersection with the natural fracture (failure along the plane of weakness). The impact of fracture spacing to optimize these fracture geometries is studied. A systematic optimization for designing the fracture length and width is also presented. The simulation is motivated by the oil window of Eagle Ford shale formation and the results of this work illustrate how different fracture network geometries impact well performance, which is critical for improving future horizontal well completions and fracturing strategies in low permeability shale and tight oil reservoirs. A rate transient analysis (RTA) technique employing a rate normalized pressure (RNP) vs. superposition time function (STF) plot is used for the linear flow analysis. The parameters that influence linear flow are analytically derived. It is found that picking a straight line on this curve can lead to erroneous results because multiple solutions exist. A new technique for linear flow analysis is used. The ratio of derivative of inverse production and derivative of square root time is plotted against square root time and the constant derivative region is seen to be indicative of linear flow. The analysis is found to be robust because different simulation cases are modeled and permeability and fracture half-length are estimated. / text