Early Modeling and Development
Developing new petroleum sites is becoming increasingly more complex as reservoirs are discovered in difficult to access areas. Reservoir modeling has been critical to the industry as a tool to address any arising challenges. The initial stages of modeling and development involve the use of computer modeling to create a digital construction of the shale reservoir. This can be used for the purpose of estimating the available reservices, predicting future production, evaluating reservoir management, and guiding overall decision-making on the development of the field. Aspects such as the environment, geophysics, soil-fluid interactions, and oil recovery techniques are used in reservoir simulations which significantly enhances the practical aspect of petroleum extraction (Hagedorn n.d.).
IGIP
The abbreviation IGIP is deciphered as “initial gas in place” (Kalam et al., 2015). It is often interchangeably used with original gas in place (OGIP). Before unconventional types of reservoirs are developed, it is necessary to conduct an estimation of its reserves and administer an economic analysis in order to determine its feasibility. Calculations of IGIP can be done through various methods, each facing specific challenges. IGIP estimation can use volumetric, simulation, and performance decline methods.
This includes running a decline curve and material balance. The volumetric method is the most common and uses static data, including reservoir bulk volume and in-situ gas content but can be less reliable. Meanwhile, a performance method utilizes dynamic data such as production volumes. IGIP is a tool for reservoir management by helping to evaluate whether the static reservoir model is an accurate representation. IGIP estimation is a critical process that occurs in a continuous cycle and improves as more production data is available (Kalam et al., 2015). Uncertainty and risk are mitigated as the quality and amount of geoscience and production data increases.
IGIP in shale formations is composed of free gas, absorbed gas, and dissolved gas. The latter dissolves into hydrocarbons and is negligible to calculations. Additional information such as porosity, water saturation, and thickness, and volume for the formation is required for the calculation of free gas. The economically recoverable portion of IGIP is also calculated with a recovery factor. The Utica formation was originally identified as highly prospective due to the amounts of IGIP present. Original reports from the early days of surveying the shale indicate an above-average of 93 billion cubic feet per section, with some deeper basin targets exceeding 150 billion cubic feet (Eaton 2010). This implies that IGIP is a crucial indicator of performance in shale development and is crucial to early modeling and estimates.
Core/Pilot Log Data
Reservoir heterogeneity is a vital aspect of the modeling process. Often recovery efficiency is much lower than initially estimated. Heterogeneity can exist at several scales and can result in the presence of non-contracted petroleum regions amongst sediments or oil trapped in small-scale structures. Core analyses consist of observing core desorption at reservoir temperatures. At this point, escaping gas is captured to determine the composition and volume of any given element in the core.
A pilot log analysis is a common practice of making initial records and computer models of a well which includes geologic formations based on information from boreholes. Both these methods are necessary to identify geological facies in cores, including lithology, texture, sedimentary composition, and grain size, amongst others. Integration of core and log information can be vital to represent small-scale heterogeneity in reservoir analysis (Bozorgzadeh et al., 2015).
Modern core and pilot logging tools use innovative technology such as the Schlumberger Platform Express. This allows for a wide range of data to be collected with 100% core recovery, greatly decreasing uncertainty on calibration and accuracy for probabilistic interpretation (Samotorova & Bronnimann 2014). Core and log data are used for reliable reservoir simulation, particularly viable for complex shale or chemical enhanced oil recovery formations. Core flood data and field-scale pilot logs produce critical calibration data, increases the success of modeling, and become applicable to the de-risking process (Bozorgzadeh et al., 2015).
Fracture Modeling
In order to accurately portray realistic fracture networks of a reservoir, new fracture modeling techniques have been developed. Fracture modeling provides stochastic and geomechanical characterizations of a fracture through the use of both stochastic and deterministic methods. Discrete fracture models (DFMs) are particularly viable for unconventional reservoirs where flow modeling remains a challenge due to low matrix permeability, the complexity of the fracture network, and non-linear matrix pressure distribution (Farah 2016).
DFMs explicitly discretize the complex network, involving a non-tractable numerical system and other unknown factors in combination with traditional data of fracture location, anisotropy, and orientation. As a result, the model produces fracture simulations that are homogeneous and can properly portray flow exchange between the matrix and various sorts of fractures (Farah 2016). Fracture models can be distributed and upgraded to 3D grids used for simulation. This is necessary to run complex simulations on advanced permeability and multiple porosities as well as allow integration with various other modeling and simulation technologies or programs.
Fracture modeling must encompass all steps of the hydraulic fracturing method in its essentials that consists of deforming the fracture surface, inserting fluid and tracking its flow within the fracture, and fracture propagation. Linear elasticity is used to calculate deformation, power-law fluid is for fluid flow, and mechanics theory with propagation law is used for the last step. Newer Pseudo-3D models have been developed based on older 2D PKN and KGD models.
The 3D model uses two approaches. First, the fracture is split into cell-based models along a horizontal direction. Another approach is to lump the two half ellipses of equal lateral extent but varies on the vertical parameter. To deal with the challenges of palaeo stress and arbitrary shapes of fractures, PLanar 3D models were developed that simulate micro-cracks, fluid leak-off, and fissure pressure storage using either moving triangular mesh or fixed rectangular mesh techniques (Li et al. 2015). The modeling techniques are vital to the hydraulic fracturing process to mitigate the challenges of complex shale environments and reduce risk.
Transient/Decline Analysis
These types of analyses seek to predict future trends based on historical data. The transient analysis investigates pressure changes over time. Meanwhile, decline analysis focuses on production volumes and capabilities of a well over its performance life. In unconventional reservoirs, production is analyzed through rate-transient techniques, which are based on linear flow models. They use normalized rate pressure that considers reservoir pressure drop divided by the square root of time. Unconventional reservoirs in practical applications include a mixture of well-fluid production, which consists of various types of oil, gas, and water. A transient analysis can be beneficial to account for this multiphase fluid flow in single and dual porosity models of unconventional reservoirs (Uzun et al., 2016).
Meanwhile, decline analysis in unconventional reservoirs is challenging because the Arps standards in petroleum engineering for estimating ultimate recovery are not practically applicable for low permeability fractures. It leads to ambiguous results with invalid assumptions that produce overestimates of reserves, especially if the hyperbolic relation is extrapolated and unconstrained. However, modifying values and hyperbolic relations can help with effectively predicting production data based on diagnostic interpretations (Mangha et al., 2012). An accurate decline analysis is critical to shale development as it produces a long-term forecast for production rates, loss of reservoir pressure, and changing volumes of fluid as part of a function of time.
Different Generational Designs
For unconventional development, wells follow generational designs. A group of wells has similar characteristics such as drilling, casing, perforating, cementing, and hydraulic fracturing design. As more wells are developed, it provides an opportunity to improve the design in order to eliminate deficiencies and improve the effectiveness and performance of the operation. Once the technological leap is made, the next generation of wells follows the new design until further improvements are made (Holloway 2018).
Generation I
First-generation projects are always high-risk due to multi-faceted uncertainties. However, they demonstrate the productive and commercial feasibility of shale locations which paves the way for further development and investment. At generation 1, the Utica and Marcellus wells had shorter lateral strengths, 250’ stage spacing, and Xling gels. Pressure drawdown was 150 psi/week when wet and 200 psi/week dry. Proppant loading ranged between 1600 and 1700 #/ft. Generation 1 initially saw good horizontal results at 2.5-3.3 billion cubic feet equivalent. The primary goal was to increase proppant loading and decrease stage spacing which would lead to improved stimulated reservoir volume (Pickett 2017).
Generation II
The second generational design introduced aspects such as 100-percent slickwater. This demonstrates fluid fracturing efficiency, which is pumped at high rates into the well to generate narrow fractures. Meanwhile, concentrations of proppant decreased in this generation to 1400-1500 #/ft to maximize efficiency. Stage spacing decreased further to the 150-225′ range. Pressure drawdown also decreased to 100-150 psi per week (Pickett 2017). Pressure drawdown is the change between good internal pressure and the flowing wellbore pressure that moves fluids. It has the biggest influence on the production rate of the well. The primary goal for this design generation was to maintain the complete cost structure, which the team was able to achieve within 2% due to operational efficiencies.
Generation III
Generation 3 saw the introduction of super laterals, which could reach depths of upward of 30,00 feet and lateral lengths. This generation offers a high-intensity completion design that operates on 100 percent slickwater and increased proppant loading to 2300-2800 #/ft. Along with reduced stage spacing (150-200’), this generational design serves as a powerful combination at optimizing the production process. The Utica shale formation demonstrated a cumulative performance of 2.4 billion cubic feet of gas equivalent in 185 days of production but demonstrating significantly lower pressure declines than predicted. This is the current generation of the well, and it is expected to outperform previous generations on type well reserve expectations by as much as 50 percent (Pickett, 2017).
Generation 3 optimization is extremely cost-effective due to drilling and engineering solutions such as advanced directional planning, slickwater completions, advanced water logistics, utilization of managed pressure drilling, and casing-conveyed toe guns. Increased sand concentration and the creation of an engineered flow back system further contributed to efficiency. The primary goal for generation 3 was to improve type curve economics across all Utica shale wells, which was able be achieved as the new wells outperformed previous generations on all types of curve assumptions.
Generation III+
The next generation 4 design is currently being developed and tested. It will incorporate many of the practical aspects described above and continue to improve efficiency. A significant breakthrough is predicted to be the optimization of managed pressure drilling that will reduce mud weights, water recycling, reuse of production equipment, and bi-fuel frac fleets (Pickett, 2017). Furthermore, the boundaries of super later well extensions will be pushed to greater than 15,000 feet, longer than Purple Hayes, which would incorporate Utica’s condensate and dry gas areas. Laterals longer than 12,000 feet ultimately change calculations and make it competitive with dry gas. Another direction that future generations of wells will attempt is to re-frac locations and utilize diversion techniques.
General Productivity
As evident, different generational designs improved on various technologies such as longer and stronger laterals, optimization of completion, and decreasing cycle times. These aspects strongly influence productivity as there has been an exponential 150 percent boost in performance since the first generation. This has led to a 35% growth in estimated ultimate recovery, with wells in the Marcellus shale maintain a confident range of 7-20 billion cubic feet of gas equivalent. The potential and growth of Utica and Marcellus shales are estimated to be over 250,000 acres that result in an annual organic growth exceeding 50 percent. By market evaluations, that is worth $4-8 billion (Chesapeake Energy 2014).
References
Bozorgzadeh, M., Stofferis, M.M., Yznaga, R. A., et al. 2015. Strategy for Calibrating a Field-Scale Numerical Simulation Study for the Implementation of a Chemical Enhanced Oil Recovery Pilot Using Core Flooding and Pre-Pilot Experiments. Paper SPE-175666-MS presented at the SPE Reservoir Characterisation and Simulation Conference and Exhibition, Abu Dhabi, UAE, 14–16. Web.
Chesapeake Energy. 2014. Focused on value delivering growth. Web.
Eaton, S. R. 2010. Utica Emerges in Quebec: Shale Play Extends to Canada. AAPG Explorer. Web.
Pickett, A. 2017. Extended lateral lengths, Optimized completions define Eclipse’s best practices. The American Oil and Gas Reporter. Web.
Farah, N. 2016. Flow Modelling in Low Permeability Unconventional Reservoirs. Doctorate thesis, Université Pierre et Marie Curie, Paris, France.
Hagedorn, K. D. Use of Modeling in Petroleum Reservoir Development and Production Enhancement. National Academy of Engineering. Web.
Holloway, M. D. 2018. Fracking: Further Investigations into the Environmental Considerations and Operations of Hydraulic Fracturing. Hoboken, NJ: John Wiley & Sons.
Kalam, S., Khan, R. A., Baig, M. T., et al. 2015. A Review of Recent Developments and Challenges in IGIP Estimation of Coal Bed Methane Reservoirs. Paper SPE-178022-MS presented at the SPE Saudi Arabia Section Annual Technical Symposium and Exhibition, Al-Khobar, Saudi Arabia. Web.
Li, Q., Xing, H., Liu, J. et al. 2015. A Review on Hydraulic Fracturing of Unconventional Reservoir. Petroleum. 1(1): 8-15. Web.
Mangha, V. O., Ilk, D., Blasingame, T. A., et al. 2012. Practical Considerations for Decline Curve Analysis in Unconventional Reservoirs – Application of Recently Developed Rate-Time Relations. Paper SPE-162910-MS presented at the SPE Hydrocarbon Economics and Evaluation Symposium, Calgary, Canada. Web.
Samotorova, G., and Bronnimann, B. 2014. Exploring Petrophysical Uncertainties Thanks to Stochastic Well Log Data Interpretation: Termokarstovoye Field Case. Paper SPE-171211-MS presented at SPE Russian Oil and Gas Exploration & Production Technical Conference and Exhibition, Moscow, Russia. Web.
Uzun, I., Kurtoglu, B., and Kazemi, H. 2015. Multiphase Rate-Transient Analysis in Unconventional Reservoirs: Theory and Application. SPE Reservoir Evaluation & Engineering. 19(4): 8-15. Web.