A Numerical Modeling Framework for the Optimization and Economic Analysis of Unconventional Gas Production

Abstract

[Abstract] The International Energy Agency describes natural gas as the cleanest burning and fastest growing fossil fuel, contributing almost one-third of the total energy demand growth over the last decade. In 2018 natural gas accounted for nearly half of the increase in global energy demand. The gas demand growth has been concentrated is three key regions that represent paradigmatic disruptions in production technology, in overall economic growth or in the effort towards economic and energetic diversification: the United States, China and the Middle East. This Doctoral Thesis is motivated by the disruptive technological innovations that underpin the recent increase in oil and gas production: the exploitation of unconventional hydrocarbon resources in the USA and elsewhere. Conventional gas refers to methane, and other light hydrocarbons, stored in high-permeability rock formations, which can be recovered economically using century-old production techniques. Unconventional gas refers to natural gas that is difficult to produce, essentially coal-bed methane, tight gas and, the most abundant source, shale gas. Gas shales are tightly packed fine grained sedimentary rocks. Hydrocarbons form in these rocks and remain trapped in their pore space and kerogen inclusions due to their ultra-low permeability. Despite the abundance and relatively homogeneous spatial distribution of gas shale, it was traditionally assumed that these resources could never be developed economically. The sophistication of horizontal drilling and hydraulic fracturing techniques have recently fostered a rapid increase in shale oil and gas production. Objectives. This thesis aims at understanding the physical mechanisms, engineering designs, and financial constraints that control the economic performance of shale gas wells. The general objective is to develop a physically-realistic numerical simulation and optimization framework to evaluate potential shale investments and to improve their economic efficiency. In the first part of the thesis, we use high-fidelity numerical simulations to characterize gas production from hydrofractured shale wells, identifying the relative impact of geology (formation porosity, permeability and kerogen content) and engineering geomechanics (the geometry of the effective propped volume) on gas fluxes. We then study the impact of uncertainty in natural gas prices, due to their stochastic nature, on the financial performance of shale gas wells. In the final part of the thesis, we explore optimal combinations of the geological and design parameters to minimize the risk of shale gas investments. To this end, we propose a simulation and optimization workflow that provides percentiles of various financial metrics for optimal combinations of parameters. Methodology. The basis for the analysis presented in this thesis is a 3D Finite Element model of gas flow towards a horizontal well with multiple hydrofracture stages along its entire length. What sets apart the present simulations from the existing literature is their physical fidelity and the consideration of the particular geometry of the effective propped volumes for each stage. Rather that vertical fracture planes, we consider ellipsoidal volumes enclosing a region of effective stimulated permeability. These volumes account for the generation of a dense and complex network of fractures in the gas-rich mudstone formation upon hydrofracturing. We use this simulation setup to explore the impact of physical parameters and engineering designs of gas production and on the profitability of shale gas investments. To account for the stochastic nature of gas prices, and to quantify the impact of price uncertainty and price evolution on financial performance, we develop a stochastic model of gas price evolution, based on a geometric Brownian process with either constant volatility or with stochastic volatility based on a bootstrapping framework that uses the available historical data of natural gas prices. Finally, the optimal selection of geological sites (sweet spots) and geomechanical engineering processes (fracture geometry) for economic performance is based on minimizing the absolute value of NPV and IRR for a given probability of occurrence, which allows determining the parametric combinations that characterize the sweet spots. Additionally, a new financial indicator called Break Even Time (BET) is developed that allows making speculative investment decisions, in the short term, compared to long-term investments characteristic of shale gas wells. Impact. The economic performance of shale gas wells is highly uncertain, often described as a “sweet spot” business with a few highly productive wells pay for the many underperforming ones in a given play. Understanding the various physical and financial uncertainties is essential for the successful development of the shale gas industry worldwide. The contribution of this thesis is three-fold: through a sensitivity analysis of the geological and fracture design parameters, we show that the geometry of the fractured volumes plays a key role in gas production. We argue that controlling stimulated propped volume, ultimately determined by fracturing fluids and protocols and by geomechanics, is as important as choosing a “sweet spot” in terms of formation permeability and porosity. Studies of the economic performance of shale gas wells have focused so far on the gas production decline curves, with a deterministic gas price to convert from gas flows to cash flows. The approach of this thesis is radically different. We propose that the stochastic nature of natural gas prices must be incorporated in all workflows to evaluate the economic performance of gas wells and to guide shale investments. The resulting analysis is significantly more complex and sophisticated than the existing literature. In exchange, stochastic prices allow us to apply concepts of portfolio analysis and decision making in the presence of uncertainty that are common in the asset management theory. Finally, we present an optimization framework to determine parametric combinations and fracture designs that guarantee profitability and investment recovery periods with a quantified risk. Our results show that profitability is in fact challenging at current price levels. This finding has important practical implications: gas-rich mudrock formations are typically spatially extensive and abundant worldwide, but the limited parametric combinations that guarantee profitable production suggest that only selected regions could accommodate an industry boom comparable to the recent one in the United States.