Low injected fracturing fluid recovery has been an issue during flowback period that is highly impacted by the fracture closure behavior. Although existing flowback models consider fracture closure volumetrically, they do not represent the true situation of non-uniform fracture closure. In this paper, we proposed a coupled geomechanics and fluid flow model for early-time flowback in shale oil reservoirs. The fluid flow model is coupled with an elastic fracture closure model through finite element methods. In this study, three stages are modeled: fracture propagation, well shut-in and flowback. Cohesive Zone Method (CZM) has been used for modeling fracture propagation. The presented model distinguished the propped part from the unpropped part of the fracture. At the beginning of flowback, the proppants may not be completely compacted in early shut-in time. Thus, permeability evolution during closure is tracked using a smooth permeability transition function. The numerical results have shown that fracture closure during the flowback period is often not uniform. While the uniform fracture closure leads to maximum fracturing fluid recovery, an aggressive pressure drawdown strategy may damage fracture connectivity to the wellbore. An integrated flowback model enables modelling nonuniform fracture closure in a complex fracture network. This study highlights that by choke/pressure drawdown management, operators can influence fluid recovery and even maintain high fracture conductivity. Furthermore, the methodology presented in the paper can also be used for inverse analysis on early flowback data.

"Sweet spots" in the unconventional reservoirs such as organic-rich mudrocks are zones with high productivity. However, identifying such regions in unconventional reservoirs depends non-only on their petrophysical and but also on their geomechanical properties. Supervised learning methods can help in integrating numerical simulation and legacy field data in sweet-spot identification workflows and enhance their analysis in complex reservoirs. The objectives of this paper are to: (i) demonstrate the use of supervised learning in parameter selection and evaluation for fracture design and (ii) provide non-linear models for sweet-spot analysis in complex reservoirs. We used fracture simulator that combines with fracture deformation with fluid-flow in discrete fracture networks. We started by selecting different geomechanical rock properties related to its fracability. We then used quasi-random design approach to obtain wide variation in aforementioned properties and performed 200 fracture simulations using the hydraulic fracturing simulator. We used the short-term Stimulated Reservoir Volumes (SRV) obtained at the end of numerical simulations, to quantify the performance of hydraulic fracturing operations. We used supervised learning techniques like support vector machines, decision trees, and random forests to perform parameter ranking and create non-linear regression models that can correlate the SRV to formation geomechanical properties. The inputs for the analysis are: initial aperture, toughness, dilation angle, closure stress, and friction coefficient of initial fractures, stress anisotropy, shear modulus and a ratio of the reservoir rock. We analyzed the results using β-linear and multinomial regression, support vector machines, decision trees, and random forests. The linear models and non-linear models can explain up to 89.1% of output variance. The classification accuracy of support vector machines was at most 35% higher than other algorithms like random forests. Parameter rating using non-linear models showed that stress anisotropy and dilation angle demonstrated the highest effect on SRVs. Shear modulus and fracture toughness show minimal effect on the SRV but these parameters might still be useful they could be correlated to other formation parameters. The outcomes of this paper demonstrated that parameters pertaining to unpropped fracture conductivity play a significant role in determining the success of hydraulic fracturing treatments. We have also compared the performances of supervised machine learning algorithms in assessing the impact of rock properties on fracturing treatments. Such supervised machine learning algorithms can help integrate field legacy data and numerical simulation outputs to develop proxy models that improve sweet-spot analysis and production estimates in unconventional reservoirs.

Microseismic data and post-fracturing production have confirmed the positive role of fracture complexity on production enhancement in fractured wells. While operators are looking for different fluids and pumping schedules to enhance fracture complexity, the mechanisms ruling the process is not fully understood. This paper provides a comprehensive workflow to model the fracture pattern development by accounting for interactions with numerous natural fractures. We present a robust finite element model with adaptive insertion of three-dimensional cohesive elements for fracture propagation through the intact rock as well as the network of intersecting natural fractures. Cohesive elements are coupled with general Darcy's flow to incorporate fluid flow as well as elastic and plastic deformations of rock during initiation, propagation and closure of hydraulic fractures. Hydraulic fracturing treatment has been simulated for different natural fracture patterns. Fluid injection pressure fluctuations are observed while reopening natural fractures. The impact of operation schedules on network complexity such as hesitation time is investigated. The complexity of fracture network is characterized by the ratio of total fracture length to its effective radius from the wellbore. Our analysis has shown that in addition to the differential stress and the fracture intersection angle which are already determined by the nature, pumping injection rate and hesitation time can play a significant role in fracture branching and its diversion to different natural fracture sets. Higher injection rate is found to have a positive effect to overcome the resistance of natural fractures in different directions, and hesitation in the middle of pumping can force the fracture to divert into other directions, both of which help develop a more complex fracture pattern.

Natural fractures are the main producibility factor in the weathered granite reservoirs (basement rock) and volcanic-rock reservoirs. Production practices show that these reservoirs could have high production rate, but the difference of well productivity between single wells is obvious. These reservoirs have complex natural fractures oriented at medium-high angles, which could bring high complexity and heterogeneity to the reservoirs, adding anisotropy to reservoir permeability. It is very hard to effectively simulate complex fractures in naturally fractured reservoirs and study the applicability of different well type and well pattern by using common reservoir simulators. A fast EDFM (Embedded Discrete Fracture Model) method was put forward for production simulation of complex fractures in naturally fractured reservoirs. The EDFM processor combining commercial reservoir simulators (ECLIPSE or CMG) is fully integrated to forecast production performance of the weathered granite reservoir. With a new set of EDFM formulations, the non-neighboring connections (NNCs) in the EDFM are converted into regular connections in traditional reservoir simulators, and the NNCs factors are linked with gridblock permeabilities. So complex dynamic behaviors of natural fractures can be captured, which can maintain the accuracy of DFMs (discrete fracture models) and keep the efficiency offered by structured gridding. In this paper, a 3D model with complex natural fractures was built to model the performance of different well types and well patterns. The results show that wells with higher density of natural fractures produce higher oil production, and horizontal wells with higher density of natural fractures have larger oil production than vertical wells because horizontal wells have a larger contact area than vertical wells. What’s more, heterogeneity and anisotropy have a great effect on well pattern and well type, which need to be studied carefully in the oilfield development.

The natural gas from Marcellus Shale can be produced most efficiently through horizontal wells stimulated by multi-stage hydraulic fracturing. The objective of this study is to investigate the impact of the geomechanical factors and non-uniform formation properties on the gas recovery for the horizontal wells with multiple hydraulic fractures completed in Marcellus Shale. Various information including core analysis, well log interpretations, completion records, stimulation design and field information, and production data from the Marcellus Shale wells in Morgantown, WV at the Marcellus Shale Energy and Environment Laboratory (MSEEL) were collected, compiled, and analyzed. The collected shale petrophysical properties included laboratory measurements that provided the impact of stress on core plug permeability and porosity. The petrophysical data were analyzed to estimate the fissure closure stress. The hydraulic fracture properties (half-length and conductivity) were estimated by analyzing the completion data with the aid of a commercial P3D fracture model. In addition, the information from the published studies on Marcellus Shale cores plugs were utilized to determine the impact of stress on the propped fracture conductivity and fissure permeability. The results of the data collection and analysis were utilized to generate a base reservoir model. Various gas storage mechanisms inherent in shales, i.e., free gas (matrix and fissure porosity), and adsorbed gas were incorporated in the model. Furthermore, the geomechanical effects for matrix permeability, fissure permeability, and hydraulic fracture conductivity were included in the model. A commercial reservoir simulator was then employed to predict the gas production for a horizontal well with multi-stage fracture stimulation using the base model. The production data from two horizontal wells (MIP-4H and MIP-6H), that were drilled in 2011 at the site, were utilized for comparison with the model predictions. The model was then also used to perform a number of parametric studies to investigate the impact of the geomechanical factors and non-uniform formation properties on hydraulic fractures and the gas recovery. The matrix permeability geomechanical effect was determined by an innovative method using the core plug analysis results. The results of the modeling study revealed that the fracture stage contribution has a more significant impact on gas recovery than the fracture half-length. Furthermore, the predicted production by the model was significantly higher than the observed field production when the geomechanical effects were excluded from the model. The inclusion of the geomechanical factors, even though it reduced the differences between the predictions and field results to a large degree, was sufficient to obtain an agreement with field data. This lead to the conclusion that various fracture stages do not have the same contribution to the total production. Based on well trajectory, variation in instantaneous shut-in pressure ISIP along the horizontal length, shale lithofacies variation and natural fracture (fissure) in the reservoir, it is possible to estimate the contribution of different stages to the production for both wells MIP-4H and MIP-6H.

Microseismicity is a physical phenomenon which allows us to estimate the production capability of the well after hydraulic fracturing (HF) in a naturally fractured (NF) reservoir. Some of the microseismic events are reactivations of NFs induced by a direct hit of HF, while others are induced by the fluid leak-off from the previous stages or by elastic waves emitted into the reservoir with hydraulic fracture plane propagation. The former NFs have a chance to be propped there as the latter will not significantly increase their contribution to the production. Identification of such microseismic events helps to reduce uncertainty in the description of fracture network geometry. Based on inferred data from core analysis NF densities and orientations, we generated multiple realizations of the semi-stochastic Discrete Fracture Network (DFN). In order to constrain them, we used time evolution of microseismic cloud in addition to results of core analysis. Fluid and proppant pumping schedule is used to identify such microseismic events because they should be located close to the pressure diffusion front generated by hydraulic fluid. Events outside of proposed region may be triggered by other factors, such as stress-strain relaxation from other stages and correspondent fractures. In most cases, they are not wide enough to take proppant from the main HF. This approach was used to reduce range of production for DFN realizations. This workflow is implanted to a 15-stage hydraulic fracture treatment on a horizontal well placed in a siltstone reservoir with intrinsic fractures. The spatio-temporal dynamics of microseismic events are classified into two groups by the front of nonlinear pressure diffusion caused by 3-dimensional hydraulic fracturing, considered as effective and ineffective events. DFNs with only effective microseismicity and with all the induced events are generated. Then, two types of DFN related uncertainties on production are performed to evaluate the impact of filtration. Results of aleatory uncertainty quantification caused by the randomness of DFN modeling indicate the filtered events can generate a production DFN with a more consistent connected fracture area. Moreover, sensitivity analysis caused by lack of accuracy in natural fracture characterization shows the production area of DFN with filtration process is more insensitive to the variation of fracture parameters. Finally, a history match with production data and pressure data indicates this DFN model properly represents the reservoir and completion. Our methodology characterizes well the conductive fracture network utilizing core data, microseismic data, and pumping schedule. It could restore the true productivity of each fractured stage from a massive microseismic cloud, which helps understand the contribution of fracturing job right after the treatment.

In Teklu et al. (2017a , and 2017b ) matrix and fracture permeability of carbonate rich and clay rich shale cores were measured before and after exposing the core samples to spontaneous dilute acid imbibition. In this study, effect of brine and dilute acid imbibition on long term production of shale gas reservoir is investigated using numerical modeling approach. Experimental observations show that: (1) matrix permeability of clay rich shale samples are usually impaired / damaged by dilute HCl imbibition. This impairment is due to both the clay swelling and related issue of brine (within the dilute HCl solution) and rock-acid reaction. (2) Matrix permeability of calcite rich shales improved with dilute acid imbibition. (3) Effective fracture permeability of un-propped (natural fracture) calcite rich and clay rich shales are reduced by dilute acid imbibition, this is because of "rock softening" and "etching/smoothing" of fracture roughness on the "fracture faces". (4) Propped permeability of clay and carbonate rich shales decrease with brine imbibition, and permeability further decreased after dilute acid imbibition. The further decrease in permeability and porosity after acid imbibition is caused by three phenomena: (a) rock weakening by rock-acid reaction, (b) further proppant embedment during repeated testes vs. stress, and (c) scale precipitation. These observations are similar to results reported in Teklu et al. (2017a and 2017b ). Modeling results show that: (1) For carbonate rich shale reservoirs, despite dilute acid injection/imbibition/fracturing can cause considerable damage to propped fractures and natural fractures (mainly due to rock softening/weakening, fracture roughness damage, and proppant embedment), it can still lead to considerable production improvement This is mainly caused by the matrix permeability improvement which leads to improvement in stimulated reservoir volume. (2) For clay rich shale reservoirs, both brine (slickwater) and dilute acid imbibition/injection/fracturing reduces hydrocarbon production, due to damages caused in to propped fractures, natural fractures, and matrix. Therefore, dilute acid injection or imbibition is recommended pre, post, or during hydraulic fracturing of carbonate rich shale reservoirs (but not for clay rich shale reservoirs) Moreover, slickwater fracturing in clay rich shale reservoirs can create considerable formation damage and decrease in hydrocarbon production, hence, dry gas (such as N 2 , CO 2 , LNG, etc.) might be an alternative solution hydraulic fracturing of in clay rich shale reservoirs.

The production from a hydraulically fractured unconventional well depends on the stimulated permeability and its interaction with the naturally fractured background permeability. Since the propagation of a hydraulic fracture is often asymmetric and depends on geomechanical factors, the ensuing pressure depletion and the EUR depends on this asymmetric behavior. An analytical asymmetric tri-linear model to approximate pressure depletion is presented. The model uses asymmetric frac design results as input and estimates the pressure depletion around a parent well. This new approach represents an acceptable alternative to full reservoir simulation when investigating frac hits problems. This asymmetric tri-linear model was combined with our poro-elastic geomechanical modeling simulator in order to capture the physics created by the depleted pressure sink zone. This physics combines the stimulation operations in the neighboring infill well and their interactions with the complex local and far scale geologic features such as natural fractures and faults. The pressure depletion determined at an Eagle Ford well using the asymmetric tri-linear model was similar to those found with a full reservoir simulator. Hydraulic fracture modeling of a child well located in the vicinity of a parent well with a pressure depleted zone highlighted the potential of developing a frac hit if geological features in the area were creating fluid and pressure conduits. A similar observation is made for a Wolfcamp well where a fault affected the nearby stage causing interference between potential stacked wells. The integration of the asymmetric tri-linear model and our geomechanical simulator presents the necessary completion modeling tool to quickly, yet accurately design hydraulic fracturing while preventing frac hits, especially now with the increasing of number of infill unconventional wells.

Unconventional reservoirs require hydraulically fractured horizontal wells in order to produce reservoir fluids economically. Induced hydraulic fractures interacting with pre-existing natural fractures results in complex fracture networks (CFNs). Even though hydraulic fracture propagation has been investigated extensively, there is indeed a lack of good understanding of characterization approaches for pre-existing natural fractures. This work presents a practical CFN generation approach by incorporating stochastic algorithms, core, and microseismic data. Based on analysis of geological structures and core observation, natural fracture density, length, and strike distributions can be obtained. Fracture length follows a power law distribution constrained by minimum, maximum and cutoff lengths as well as a distribution exponent. Fracture strike follows a Fisher distribution. Then, microseismic event locations are used to constrain fracture centers of stochastically generated natural fractures. Moreover, a fast proxy model is developed for hydraulic fracture propagation, which honors both the total mass volume of the pumped proppants, and the pre-defined reference lengths. Finally, a field case study is used to demonstrate how to apply the proposed fracture generation and simulation workflow to model hydraulically fractured horizontal wells. The proposed workflow implemented stochastic algorithms with the capabilities to incorporate as much information as possible such as core analysis and microseismic information, and to evaluate uncertainties due to pre-existing natural fractures. With the assumption that microseismic event locations are reactivation of pre-existing natural fractures, the perturbed event locations constrained the locations of the natural fracture centers, resulting in a better description of microseismic-derived stimulated reservoir volume. The simplified hydraulic fracture propagation scheme was able to efficiently estimate the resulting complex fracture networks, and to accurately honor the material balance during fracturing treatment. Sensitivity analysis showed that fracture permeability, matrix porosity, and matrix permeability of the CFNs affect well production performance, significantly. This paper discusses how to utilize available data resources to generate representative CFNs for hydraulically fractured horizontal wells. The process of data preparation for each step of the workflow is discussed in details to facilitate engineers to solve practical problems with the developed methodology.

Unconventional shale reservoirs require massive and multistage hydraulically fractured horizontal wells in order to produce economically. Induced hydraulic fractures interacting with in-situ natural fractures results in complex or discrete fracture networks (DFN). Even though well testing characteristics in fractured reservoirs with vertical wells have been investigated extensively, there is indeed a lack of good understanding of well testing behaviors for hydraulically fractured horizontal wells in complex fracture networks. First, three practical approaches are presented regarding how to generate complex fracture networks in the context of developing unconventional shale reservoirs with hydraulically fractured horizontal wells. Complex fracture networks can be generated 1) from stochastic algorithms that input fracture density, length and strike distributions, or 2) from the flowing-producing DFN (FP-DFN) area that is constrained by microseismic information, or 3) from digitization of realistic outcrop maps. Then, new unstructured fracture gridding and discretization techniques specially tailored for complex fracture networks are developed to handle nonuniform fracture apertures, extensively fracture clustering and nonorthogonal fracture intersections. Finally, numerical simulations of pressure build-up are performed in complex fracture networks that are generated from three proposed approaches using both synthetic and field examples. Flow regimes are identified and discussed based on pressure derivative plots. Complex fracture networks show that the most representative characteristics are formation-fracture bilinear flow and formation linear flow regimes. The appearance of the bilinear flow regime during early period might be not clear due to the impact of wellbore storage effect for the fractal fracture generation approach. In addition, the microseismic-based approach reduces uncertainties of fracture characterization by using percentiles of FP-DFN areas. The pressure build-up responses clearly indicate that the higher the percentiles of FP-DFN areas, the lower the pressure difference and derivative curves. The fracture mineralization affects pressure build-up responses significantly. The decrease in nonuniform fracture apertures cause pressure diagnostic plots shift upward. The effect of boundary in the outcrop-based complex fracture network shows an early deviation from the formation linear flow regime. No classic dual porosity behavior is observed in all cases to quantify related parameters. Three practical techniques are proposed to generate complex fracture networks. Pressure transient characteristics are identified and summarized. The open research areas are discussed and highlighted. This work helps us better understand pressure transient behavior of complex fracture networks and after-closure analysis of fracturing calibration test.

To model hydraulic fracture propagation in naturally fractured reservoirs, fracture propagation is generally assumed to be a single planar fracture propagated only in the vertical direction from a horizontal well, regardless of the presence of natural fractures. In this paper, we developed a multi-stage hydraulic fracture propagation model using a twisted multiple planar fracture that is able to describe the propagation of hydraulic fractures more realistically. In propagating hydraulic fractures, we used two criteria: maximum tangential stress, to determine the fracture initiation angle, and whether a hydraulic fracture passes through a natural fracture. The developed model was used with a commercial reservoir simulator through grid mapping in the form of a discrete fracture network using microseismic data. The results of the verification matched the experimental results well for various intersection angles and maximum horizontal stress directions. In the investigation of the direction of maximum horizontal stress, the frictional coefficient of the fracture interface, and fracture orientation, hydraulic fracture propagation modeling results showed that the hydraulic fracture passed through a natural fracture, and thereafter propagated in a manner suitably consistent with the theoretical results, based on a fracture interaction criterion. After confirming the twisted multiple planar fracture model suggested in this work, discrepancies were found in the fracture connectivity and the stimulated reservoir volume. This indicates that the twisted multiple planar fracture approach, which is more realistic in terms of fracture propagation, is extremely important in evaluating the initial gas in place, calculated according to the stimulated reservoir volume.

The most effective method for stimulating unconventional reservoirs is using properly designed and successfully implemented hydraulic fracture treatments. The interaction between pre-existing natural fractures and the engineered propagating hydraulic fracture is a critical factor affecting the complex fracture network. However, many existing numerical simulators use simplified model to either ignore or not fully consider the significant impact of pre-existing fractures on hydraulic fracture propagation. Pursuing development of numerical models that can accurately characterize propagation of hydraulic fractures in naturally fractured formations is important to better understand their behavior and optimize their performance. In this paper, an innovative and efficient modeling approach was developed and implemented which enabled integrated simulation of hydraulic fracture network propagation, interactions between hydraulic fractures and pre-existing natural fractures, fracture fluid leakoff and fluid flow in reservoir. This improves stability and convergence, and increases accuracy, and computational speed. Computing time of one stage treatment with a personal computer is now reduced to 2.2 minutes from 12.5 minutes than using single porosity model. Parametric studies were then conducted to quantify the effect of horizontal differential stress, natural fracture spacing (the density of pre-existing fractures), matrix permeability and fracture fluid viscosity on the geometry of the hydraulic fracture network. Using the knowledge learned from the parametric studies, the fracture-reservoir contact area is investigated and the method to increase this factor is suggested. This new knowledge helps us understand and improve the stimulation of naturally fractured unconventional reservoirs.

It is estimated that only one third of the remaining worldwide oil and gas reserves are conventional, the remainder being in unconventional reservoirs whose evaluation requires appropriate measurements delivered in a cost-effective way. In the case of shales and other tight reservoirs, the defining characteristics are low matrix porosity and low or ultra-low permeability which requires artificial stimulation to encourage production. The optimum stimulation strategy for a particular reservoir is strongly dependent on the distribution of organic material, and on the mechanical and geometrical properties of the rock, and the associated stress field. It is essential to quantify these to an appropriate level of certainty, and well logs are the primary source of such data. Until recently the options for acquiring appropriate logs in high angle and horizontal wells have been constrained either by the limited available sensors or tool conveyance methods. However, the introduction of memory capable small diameter specialized tools and multiple innovative conveyance options has changed the cost-benefit balance for the better. This paper reviews the current status of open hole log measurements with full spectrum conveyance options, and how they impact the evaluation of these challenging reservoirs.

The use of a quintuple porosity system for calculation of original petroleum in place (OPIP) in shales is important as neglecting some of the porosities can result in pessimistic values of OPIP and production rates. Based on the concept of Total Petroleum System (Magoon and Beaumont, 1999) the word ‘Petroleum’ includes (1) thermal and biological hydrocarbon gas, (2) condensates, (3) crude oils and (4) natural bitumen. In the case of natural gas, the gas is trapped and stored in shale in different ways: (1) gas adsorbed in the kerogen material, (2) free gas trapped in nonorganic inter-particle (matrix) porosity, (3) free gas trapped in microfracture and slot porosity, (4) free gas stored in hydraulic fractures created during the stimulation of the shale reservoir, and (5) free gas trapped in a pore network developed within the organic matter or kerogen. An additional storage element is provided by gas dissolved in kerogen. The governing equations that describe the gas mass balance in the quintuple porosity model are presented in detail. The series and parallel gas transport approaches discussed previously in the literature are shown to be special cases of the new general gas transport formulation developed in this study. The effects of permeability stress-dependency are taken into account. Real data from Devonian gas shales are used to illustrate the effect of free gas, adsorbed gas and dissolved gas in a material balance crossplot of P/Z vs. cumulative gas production. Historically, the large contribution of organic porosity, natural fractures and hydraulic fractures that can contribute a significant amount of free petroleum in place has not been taken into account. And many of the laboratory experiments used for determining data utilized in computations of OPIP and production rates have been carried out in crushed samples, which by their very nature do not generally preserve natural fractures, slots and all the porosity present in the organic matter. This leads to pessimistic values of OPIP and rates. This helps to explain the larger than anticipated rates and recoveries of natural gas from some of these formations, for example Devonian shales, which have been producing for several decades. Although the quintuple porosity characterization mentioned above indicates very heterogeneous systems, the production performance is less heterogeneous than that of carbonates, sandstones and naturally fractured tight reservoirs. This surprising result is demonstrated with the use of actual production data from various petroleum reservoirs around the world. The subject matter is significant because of the large volume of petroleum resources in shales throughout the world, which probably are underestimated because of not considering in a single model all types of porosity discussed in this study.

There are many mathematical models for production data analysis of shale wells. Certain models assume a set of parallel hydraulic fractures perpendicular to a horizontal well. Some of these models include natural fractures perpendicular to the hydraulic fractures. In these models, there is uncertainty in whether transient linear flow is dominated by fractures or matrix. Identification of the medium in transient flow usually requires assuming fracture properties. In this paper, we introduce an identification method which does not require explicit knowledge of fracture properties. This work also helps forecast wells which have not reached boundary dominated flow decline. The paper focuses on four fracture distribution models that could describe shale wells. We introduce the identification procedure to determine the true fracture distribution model for a region. Once the model is identified, it is possible to determine whether linear flow is dominated by matrix or fractures. Identification of the true model can also improve forecast accuracy. The model identification procedure is based on the assumption that rock and fluid properties are similar within the region of interest. Shale wells often exhibit long transient linear flow. Forecast projection is uncertain because the end of linear flow cannot be calculated without knowledge of fracture properties. The model identification procedure presented here enables more accurate production forecast and analysis. In addition, model identification might help optimize fracture design for future wells. This work reduces uncertainty and eliminates inapplicable models.

Economic production from shale has been intimately tied to hydraulic fracturing since the first signs of success in Barnet Shale in the late 90s. The introduction of horizontal wells and multi-stage hydraulic fracturing was met by a huge move by operators towards developing shale formations that were mainly ignored in the past. Today using pad drilling, multiple horizontal wells share surface facilities and infrastructure, a development that minimizes the industry's environmental footprint. To understand production from shale reservoirs one must understand the network of natural fractures in the shale and the role of hydraulically induced fractures and their interaction. In this article author proposes a new view of the network of natural fractures in shale that when interfaced with the induced hydraulic fractures, will provide a completely different picture of how stored hydrocarbon is produced. Modeling this new network of natural fractures and its interactions with induced fracture requires fundamental changes in our existing simulation models. Hydraulic fracturing has been around and been studied by engineers for decades. Analytical, numerical and data-driven models have been built to explain their behavior and contribution to flow. Contribution of natural fracture networks to storage and flow in carbonate (and some sandstone) reservoirs had led to the development of techniques to study and model them. Since they are the predominant source of connected porosity and permeability in shale, more attention has been focused on their characteristics in the recent years. Studies of methane production from coal seams in the mid 80s provided insights on sorption as a storage mechanism and desorption and diffusion as a transport phenomenon in reservoirs that came to be known as CBM (Coalbed Methane). Today, production from shale is mainly modeled based on the lessons learned in the past several decades where all the above techniques are integrated to create the modern shale reservoir models. In other words, we use the "Pre-Shale" technology to understand and model hydrocarbon production from shale. This may not be the most efficient path forward 1 . The coupling of hydraulic fractures and natural fracture networks and their integration and interaction with the shale matrix remains the major challenge in reservoir simulation and modeling of shale formations. This article reviews the methods used by the scientists and engineers in recent years to understand the complexities associated with production from shale. This will shed light on the commonly held belief amongst some of the best minds in reservoir engineering (those that have been intimately involved in modeling production from shale) that there is much to be learned about this complex resource and that our best days in understanding and modeling how oil and gas are produced from shale are still ahead of us. Furthermore, an alternative solution to the conventional simulation and moldeing currently used in the industry is proposed. This technology that is used and implemented today can enhance our understanding of production from shale.

Shale reservoirs with multistage hydraulic fractures are commonly characterized by analyzing long-term gas production data, but flowback data is usually not included in the analysis. However, this work shows there can be benefits to including flowback data in well analysis. The flowback period is dominated by water flow. Field data indicate that only 15-30% of the frac water is recovered after the flowback. Past publications have suggested that the lost water is trapped in the natural fracture or imbibed into the rock matrix near the fracture face. In this paper, lost water scenarios are tested and examples are presented for including flowback and production data in the analysis of shale gas wells. A gas-water model was constructed for simulating the flowback and long-term production periods. Various physical assumptions were investigated for the saturations and properties in the fracture/matrix system that exists after hydraulic fracturing. The results of these simulations were compared with data from actual wells. The result of these comparisons led to certain conclusions and procedures that describe possible well/reservoir conditions after hydraulic fracturing and during production. In this work, the challenge of simulating a natural fracture with trapped water without imbibition is solved using a new hybrid relative permeability jail. This concept was tested for the period of flowback, shut-in and production. Natural fracture spacing could be a possible explanation of the lost water. In addition, this paper shows the benefits of combining flowback and long-term water production data in the analysis of shale gas wells. In some cases the time shift on diagnostic plots changes the apparent flow regime identification of the early gas production data. This leads to different models of the fracture/matrix system. The presented work encourages the engineer to collect flowback data in order to include it in the long-term production analysis.

Tight gas reservoirs normally have production problems due to very low matrix permeability and different damage mechanisms during drilling, completion and stimulation. Tight reservoirs need advanced drilling and completion techniques to efficiently connect wellbore to the formation open natural fractures and produce gas at commercial rates. Stress regimes have significant influence on tight gas reservoirs production performance. The stress regimes cause wellbore instability issues while drilling, which can result in large wellbore breakouts. The stress regimes can also control the well long-term production performance, since they affect permeability anisotropy. The preferred horizontal flow direction is expected to be parallel to the maximum in situ horizontal stress. The production and welltest data in non-fractured as well as hydraulically fractured wells in tight reservoirs have indicated the presence of a long-term linear flow regime due to the well and reservoir geometry and also as a result of the permeability anisotropy. The stress anisotropy leads to different permeabilities in different directions, and the natural fractures that are aligned with maximum horizontal stress; they might have larger aperture and greater permeability. Due to the more severe stress anisotropy in tight formations, permeability in maximum stress direction might significantly be larger than permeability in the direction of minimum stress. This study represents evaluation of parameters that might control well productivity and long-term well production performance in tight gas reservoirs. Geomechanical modeling is performed in order to understand the effect of stress anisotropy on aperture evolution of natural fractures in different directions. Furthermore, single well reservoir simulation study is performed in order to generate pressure build-up data for a typical tight gas reservoir, in order to evaluate effect of reservoir geometry and permeability anisotropy on late time linear flow regime, and also assess the well production performance for different well and reservoir conditions.

Abstract This paper provides a look-back review of lessons learned from early exploration to the full-scale development phase of the Marcellus shale in Pennsylvania. Fracture stimulation of over 100 wells has resulted in an in-depth understanding of details needed to achieve optimal frac performance. Much of the necessary learning curve is derived from the empirical testing of theory and what many refer to as "trial and error." The ability to evaluate and capture the best practices and develop them into a standing operating procedure (SOP) is one of the most important aspects in development of a new unconventional play. Lessons learned involving fracturing strategies and technologies to date have greatly narrowed the learning curve enabling more rapid advancement toward full-scale development. Introduction Shale reservoirs are characterized by extremely low permeability rock that has a number of unique attributes, including high organic content, high clay content, extremely fine grain size, plate-like microporosity, little to no macroporosity, and fickian vs. darcy flow through the rock matrix. This combination of traits has led to the evolution of hydraulic-fracture stimulation involving high rates, low-viscosities, and large volumes of proppant. Production from shale is dependent upon many variables including hydrocarbon content, total organic carbon, shale maturity, porosity, permeability, kerogen content, formation pressure, and net thickness. Improvements in drilling and completion techniques have improved gas recovery, namely landing a horizontal borehole strategically and creating a series of multiple staged hydraulic fractures. Even though horizontal drilling and fracturing have become the completion methods most commonly applied, a significant number of successful wells are being completed vertically in the Marcellus Shale. The extremely low permeability of shale requires a complex fracture to create primary induced fractures, reactivate and/or intercept more naturally occurring fractures or parting planes and ultimately expose more surface area. Early increased production is dependent upon the number of natural fractures intercepted and long-term production is dependent upon the amount of surface area exposed in the fracture network. Total improved production is dependent upon complex fracture geometry, which is influenced by many factors: stress contrasts, fluid leakoff, natural fractures, layering, weak planes, brittleness, fracture height growth, differing critical stress, post-fracture retention of connectivity to the created frac network, and mechanical stratigraphy, which controls the frac network creation. Large stimulation volumes of slickwater have been employed to create the extremely complex fracture fairway. High rate is needed to carry the large proppant volumes in a slickwater system and stimulation is achieved by bridging and diverting in induced fractures and natural fractures. The created fracture network is more productive than a dominant single fracture plane in a shale reservoir because more surface area is exposed for gas desorption and long-term natural gas production. Although many lessons have been learned from previous successful shale plays, many new lessons and unique "tricks of the trade" have been developed or tailored specifically for the Marcellus. This is especially true of the fluid systems and geochemical environment that has driven a number of new developments and fluid innovations.

Abstract This paper presents a study on the impact of reservoir characteristics such as matrix porosity, matrix permeability, initial reservoir pressure and pay thickness as well as the length and the orientation of horizontal wells on gas production in New Albany Shale. The study was conducted using a publicly available numerical model, specifically developed to simulate gas production from naturally fractured reservoirs. Reasons for selecting a non-commercial simulator for this study were two folds. First, we wanted to make sure that our results, discussions, and conclusions are accessible and repeatable by all interested operators and individuals that are currently producing or plan to produce from New Albany Shale since the simulator we used is readily available. Secondly, we wanted to demonstrate the utility and ease of use of this publicly available simulation software. The study focuses on several New Albany Shale wells in Western Kentucky. Production from these wells is analyzed and history matched. During the history matching process, natural fracture length, density and orientations as well as fracture bedding of the New Albany Shale are modeled using information found in the literature and outcrops and by performing sensitivity analysis on key reservoir and fracture parameters. Sensitivity analyses are performed to identify the impact of reservoir characteristics and natural fracture aperture, density and length on gas production. Economic analyses are performed to identify and rank the impact of the above parameters on the Net Present Value of investing on gas wells producing from New Albany Shale. Introduction New Albany Shale Gas -The New Albany Shale is predominantly an organic-rich brownish-black and grayish-black shale that is present in the subsurface throughout the Illinois Basin. The total gas content of the New Albany Shale (Devonian and Mississippian) in the Illinois Basin has been estimated to be 86 trillion cubic feet (TCF) (1). Although the New Albany Shale has produced commercial quantities of gas for more than 100 years from many fields in southern Indiana and western Kentucky, only a small fraction of its potential has been realized (2) The Shale is shallow, biogenic and thermogenic that lie at depth of 600–5,000 feet and are 100–200+ feet thick. Natural fractures are believed to provide the effective reservoirs permeability in these zones and gas is stored both as free gas in fractures and as absorbed gas on kerogen and clay surfaces. (3) The lack of dense natural fractures does not eliminate the potential for an economic fracture play in the New Albany. The New Albany Shale has great potential for natural gas reserves. Gas-in-place (GIP) measures from 8 bcfg/square mile to 20 or more bcfg/square mile, depending on locations and depths.