The story of the US shale revolution is well known. Hydraulic fracturing techniques were executed by Mitchell Energy in vertical Barnett Play gas wells in the early 2000's, vertical wells matured into horizontal multi-stage frac wells, and one of the largest land leasing campaigns in history exploded as operators chased high gas prices. As the natural gas market became saturated, the industry started to strip the natural gas liquids (NGLs) out of the gas stream to take advantage of the ever-rising oil pricing. When gas prices tumbled in 2011, and oil prices climbed north of $100/bbl, the industry looked to the liquid rich/oil plays, such as the Williston Basin, the DJ Basin, and the Permian Basin. The turning point came in November 2014 when oil prices fell rapidly. As prices bottomed out at $22/bbl in February 2015, the industry saw a large exodus of operators and capital from the gas rich plays around the US to the liquid rich Permian. The Permian proved to be the haven for oil and gas development with its multiple pay zone targets, high EURs, low break-even costs, friendly regulatory environment, and access to markets. The rush for land, once again ensued, with the hope of an oil price rebound and promise of high returns to capital investors. The rapid ramp up in activity from 2015–2018 did not come without challenges as it put strain on the availability of services and people, access to pipelines and markets, and access to frac sand/water. This drove up costs and resulted in mixed results for many companies. In addition, operators soon saw that with higher-than-expected gas and water production, expenses to manage these by-products sky-rocketed. Water handling and disposal became a huge portion of operating expenses and with gas export facilities at full capacity, companies started to flare gas in large volumes. Associated gas became a waste product, causing operators needed remove the gas and associated liquids from the revenue stream, and in some cases pay a high cost for flaring permits, rather than shutting in wells. By 2019, a shift in the investment community was well underway. The days of growth-focused investment were coming to an end, and investors wanted to see returns on their investments. As prices still hovered around the $55/bbl range, investors were getting anxious to recover their capital invested in the industry, and throughout 2019 operators all talked about the ability to generate free cash flow. This paper analyses the free cash flow for three key unconventional basins across the US and discusses the associated economic impacts in each basin.

The objective of this work is to predict the estimated ultimate recovery (EUR) and rank locations following application of the Digital Analogue Shale (DAS) model to exploration data types (character, geomechanics, quantity, quality, maturity, mineralogy) across shales worldwide. To address this objective, we developed the reduced order DAS model using a machine-learning (ML) approach for rapid determination of EUR with applications to independent data sets including the East Coast Basin, NZ and the Reconcavo Basin, BR. A priori knowledge of the EUR will save money and reduce time by prioritizing economic investments and returns in unconventional shale assets. Development of the DAS model follows a ML workflow - training, feature selection, testing, and uncertainty quantification. The unsupervised ML network involves competitive training and self-organization of publicly-available reservoir data from unconventional shale plays in the USA (e.g. Barnett, Bakken, Eagle Ford, Haynesville, Marcellus, Niobrara, others) that include character (depth and thickness), geomechanics (porosity, permeability, Poisson ratio, and Young’s modulus), quantity (free hydrocarbons, amount of hydrocarbons generated through thermal cracking, and total organic carbon, EUR), quality (hydrogen index), maturity (maximum temperature and vitrinite reflectance), and mineralogy (clay content, carbonate content, and silica content). Minimization of quantization and topographical error vectors provide EUR predictions for testing generalizability and quantifying uncertainty by stochastic cross-validation. The DAS model provides unbiased EUR predictions and their uncertainty estimates when applied to independent shale data. Differences in average and median EUR predictions reveal a nonlinear process underscoring the importance of using the unsupervised ML approach to develop the DAS model. Given the range of estimation uncertainty, the preferred DAS model predictions (closest to observations) are median EUR values. We successfully applied the DAS model to quantify amount and rank locations of EUR across three structural blocks in the Renconcavo basin; and by block, formation and member in the East Coast Basin. The DAS model represents an innovative step-change toward near real-time determination of EUR with quantifiable uncertainty anywhere that (sparse) exploration data are available. By ranking EUR predictions, the DAS model facilitates rapid prospect generation consistent with world-class shale plays. As new reservoir data become available, the DAS model can be refined and redeployed across unconventional shale assets. The DAS model can be applied using only sparse data from exploratory wells and can be applied at an existing field to identify targets for infill drilling. Preliminary tests demonstrate model ability to predict (in addition to EUR) economic considerations, such as well cost, operational well cost, recovery, volume of drilling water per well, volume of fracking water per well. Lastly, the rapid deployment of the DAS model can provide guidance to governments for future bid rounds, exploration planning and prioritization.

Estimating accurate volumetric concentrations of hydrocarbon and water in a producing reservoir is a critical component of predicting well performance, designing well placement and field development planning. Core testing procedures and petrophysical models in unconventional shale reservoirs have always faced the challenges of establishing representative in-situ water and hydrocarbon saturations. When using existing techniques of core calibrated petrophysics, actual well production often varies significantly from expectations. This has a serious impact on the development of major U.S. unconventional plays such as the Eagle Ford, Midland Basin and Delaware Basin among many others. Core taken from these formations enables better understanding what fluids are present and in what quantities. Changes in pressure and temperature as rock is taken from downhole, handled and transported to a laboratory facility, affects the contents of the pore system. Some of the in-situ fluids in the pore space gets volatilized and show up as void space in laboratory measurements. Standard practice calls for treating this void space as previously occupied by oil. Therefore, estimates of hydrocarbon filled porosity are made using the volume of oil extracted from the rock during testing (whether thermally or via solvents) combined with the volume of void space measured. Water filled porosity is assigned a value based on the actual water measured from the rock during the extraction process. However, fluid phase behavior in nano-pore systems is not very well understood. Pore wettability and permeability are also important factors that may control what fluids are lost from the system. Given these uncertainties, the assumption that void space is associated with volatilized hydrocarbon does not hold true. This manuscript will show several experiments including: comparisons between preserved and non-preserved samples, re-testing old core to measure fluid changes with time, nuclear magnetic resonance (NMR) scans, flow-through and fluid imbibition studies among others. NMR T1T2 logs will be used as a downhole water saturation reference. It is shown from these experiments that void filled porosity is usually occupied by formation water. Additionally, log interpretations calibrated to this new water saturation will be shown and compared to well performance. The actual well production agrees well with the above new interpretation.

The realization of Triassic Yanchang 7 member shale resource potential in Ordos basin is a significant exploration success within the last few years, and the tight sandstone, in the upper section of Chang 7 member received much of the drilling focus as horizontal development expanded in Ordos basin. In order to discuss the prospecting potential of shale oil in Chang 7 3 sub-member, South Fringe of Ordos Basin, organic geochemical parameters, quantitative characterization of different occurrence oil and assessment for Shale oil potential producibility have been carried out, based on core observation, microphotographs, Rock-Eval pyrolysis, thermovaporization in different temperature ranges and etc. The results indicate that the shale system in Chang 7 3 sub-member is a high-quality source rock with stable distribution, medium degree of thermal evolution, high TOC content and good kerogen types. Three different lithological associations are identified according to lithology, TOC and mineral content, and thereby a series of key parameters are calculated including total oil content, free oil content and its ratio that normalizes free oil content to total oil content, OSI index. In conclusion, zone 2, composed of tuffaceous fine-grained sandstone and mudstone, is characterized by "high total oil content, high free oil content and its ratio, high OSI index and high physical properties", which proves the favorable target for shale oil, whereas zone 1 is characterized by high total oil content but relative low free oil content and OSI Index; for zone 3, both total oil content and free oil content are much lower than the other two lithological associations. Introduction These years shale oil resource has been explored and developed on an industrial scale in North America, which lead the revolution of shale oil in the world. As the second sedimentary basin in China, Ordos basin is enrichment with oil and gas. The Yanchang formation in late Triassic is one of the most important oil-bearing systems and the Triassic Chang 7 source rocks are regarded as target interval during shale oil exploration and development. Due to technological advance, horizontal wells and stimulated reservoir volume (SRV) has been developed and applied in the depression area, targeting at tight sandstone in the upper section of Chang 7 member (Chang 71 and Chang 72 sub-member). A huge exploration success is achieved and individual wells have got high production. As a result, four pilot development regions are chosen and built, which revealed a good potential of shale oil resource in the central basin (Fu et al., 2019). In order to accelerate the exploration process in the south of Ordos basin, two appraisal wells have been drilled and good oil show is encountered in the bottom of Chang 7 member (Chang 73). And meanwhile two existing wells, such as J-3 and J-4, are chosen to test productivity through CO2 coupled fracturing technology and hydro-fracturing technology, both of which are landing in the lower organic-rich shales. J4 is stimulated to yield oil flow with an average rate of 8.3t/d, and cumulative oil production is approaching to 1700t, while J-3 has got a lower productivity through CO 2 coupled stimulating. Though systematic core wells and productivity tests have provided solid foundations for further study, however, exploration and development for lacustrine shale system in china is still in its infancy, compared with 30 years’ technical research and actual practices for American marine shale system(Li, 2017).

The Uinta Basin is known for its lacustrine depositional environment and its high lateral variance that makes it challenging to predict and characterize petrophysical properties. In this study, a formation evaluation workflow is presented that extracts geochemical and geomechanical data for a lateral section of horizontal well. Then the obtained data was benchmarked to production, completion and core laboratory testing data to identify the best landing targets. For this case study, four wine rack placed wells were analyzed for geochemical and geomechanical properties. Wine rack wells were drilled in Uteland Butte and Wasatch formations within the Uinta Basin. First, high-resolution drilling cuttings were collected at pilot wells. Cuttings were analyzed for mineralogy using X-ray diffraction (XRD), elements using wavelength dispersive X-ray fluorescence (XRF) and total organic carbon (TOC) using pyrolysis. Subsequently, considering the geomechanical aspect, a stress profile was generated from well log data by assuming isotropic material. Young’s modulus and Poisson’s ratio values obtained from rock mechanic testing on the vertical core as well as DFIT results were used to validate the model. Utilizing the core’s mechanical properties and mineralogy, rock physics modelling was used to find the best theoretical bounds. The bounds can be implemented to predict mechanical properties using mineral composition from cuttings samples. Afterward, the stress profile from cuttings analysis can be compared with the geophysical log. It was identified that the rock physics model of core data follows the Reuss-bound trend. It explains that the rock is layering horizontally, validating the isotropic assumption when calculating the rock physics model. Since the mineral composition of cuttings samples match the core, the Reuss model can be applied to the cuttings data to calculate rock mechanical properties and the stress profile. The generated stress profile from the geophysical log has less variability than the rock components, as the Biot’s coefficient is usually assumed as a constant. On the other hand, using a rock physics approach, the Biot’s coefficient can be predicted. In the Wasatch Formation interval, the stress profiles from both the geophysical log and rock components show that there is a stress barrier, which is more pronounced in the latter. Completion strategies on the four horizontal wells are similar with the Uteland Butte wells showing the highest cumulative production. The low flowable hydrocarbon index explains the lower production on Wasatch wells. Moreover, the observed stress barrier prevents vertical growth of the hydraulic fracture which leads to less access to hydrocarbon. This study shows that cuttings analysis provides valuable information and better decision input to identify more productive wells.

The presented study focuses on the Permian Delaware Basin Wolfcamp A (WCA) formation with the objectives: 1) to understand how completion practices and geology influence the productivity of existing wells using machine learning technique and state-of-art model explainer - SHapley Additive exPlanations (SHAP); 2) to predict the EUR of future wells and assess the technically recoverable oil in a focus study area (5,000 sq. mi, ~40% of the Delaware Basin by area) of Delaware WCA formation. Our results show that hydraulic fracturing water use intensity (HFI), true vertical depth, lateral length, sandstone facies, and formation thickness are the top five variables driving the first year productivity (stb/1000ft). For the existing producing wells in the AOI, median EUR using multi-segmental Arps model is ~30% higher than median EUR using the physics-based model (475 thousand barrels (Mstb) vs. 320 Mstb) and such differences decrease as the observed production period increases, indicating the large uncertainty that decline curve analysis will bring to EUR estimation for both existing wells and future wells. Our predictive and technically recoverable resource analyses assume that future wells will be drilled under the current technology in terms of lateral length, HF water use, proppant use, and well spacing. Based on the first year productivity analysis and the average decline curves predicted by Arps and physics-based model, the expected EUR of future wells in the AOI ranges from 140 Mstb to 780 Mstb. The technically recoverable oil in the AOI is in range between 10 billion barrel (Bstb) to 22 Bstb and the corresponding area wide recovery factor is in a range between 9% and 19%. Introduction Production of oil and natural gas from U.S. unconventional resources has soared in recent years with the Permian becoming the largest producing basin. The Permian Basin is less mature compared to other unconventional plays in the U.S., such as Bakken and Eagle Ford plays, in part due to the vast size of the area, thickness of organic-rich deposits and multiple economic targets. In 2019, oil production from the Permian Basin accounted for ~35 percent of the total U.S. crude oil production and ~15% percent of total U.S. natural gas production (EIA, 2019a, 2019b). While production is expected to slump in 2020, the recent high rate of oil and gas production from the Permian raises the question: how long can recent production rates be maintained, irrespective of changing oil prices with time? To help answer the question, we need to understand what factors influence production. With that understanding, and by developing an understanding of the remaining inventory, we can better forecast future production in the undeveloped area of the basin and develop a view of remaining potential.

In 2015-2016, the Resource Assessment team from the Bureau of Economic Geology of the University of Texas at Austin conducted its analysis of the Williston Basin and Eagle Ford play, providing the estimates of the resource in place, technically recoverable resource, and production outlook projections. The outlook projections for each play were found to be heavily dependent on the price environment and technological advances. Almost four years later, we revisit our results to check the accuracy of our original predictions, identify missing factors influencing each play development, and suggest ways to improve outlook modeling. Notably, we focus on changes in individual well’s productivity and suggest a new procedure to improve productivity predictions and the accuracy of our outlook projections. Random forest machine learning algorithm is applied to the original database updated with additional geologic, production, and completion data. We also present a discussion on the importance of drilling financing modeling, a new block of the outlook model built to capture, explain, and enhance projections of both play’s development dynamics. Introduction The oil and natural gas industry has witnessed a long history of production forecasting and production outlook projections. While forecasts or predictions suggest what one can expect in the future, projections play a different role. Projections allow us to investigate how the future may develop would the prices, technology, costs, and other relevant factors change. Hence, projections serve as an essential tool to investigate various questions: e.g., how much industry should push in technological progress to increase its profitability by X%; how resilient are operators to price or regulatory risks; or for how long can a play survive until it is exhausted given a specific price assumption. The nature of the questions suggest that projection models may affect the involved company’s behavioral, technological and market characteristics. As a result, the underlying assumptions and mechanisms in the models would become obsolete. To keep outlook models and projections relevant, regular validation analyses are needed. Such examinations would reveal whether underlying assumptions and parameters are still true and model’s mechanism captures the industry’s behavior, i.e. investing strategies. Namely, we have to consider whether additional modeling blocks are required to replicate operator’s decisions. In this context, our paper explores what, why, and how one should change and update in the Williston Basin (also referred to as Bakken) and the Eagle Ford play outlook models owing to the developments in the past 4 years.

Maximizing well construction return on investment is essential in unconventional shale formations, it is therefore important to have consistent key performance indicators, which enable optimization of completion and fracturing design parameters. To this end, PIs (Productivity Index) and the SRVs (Stimulated Reservoir Volumes) were measured on over sixty MFHW (Multi-Fractured Horizontal Wells) in Bakken shales, thereby quantifying the impact of horizontal lateral length, fracture spacing and volume of fluid pumped in each perforation cluster. The PI was measured during BDF (boundary dominated flow) and in undersaturated conditions, thereby eliminating the effect of free gas on inflow productivity. To eliminate the dependence on fluid properties (e.g. formation volume factor), the downhole flowrate was measured directly utilizing ESP (Electrical Submersible Pump) properties. This method had the additional benefit of providing the required data frequency, resolution and repeatability to analyze bilinear and linear reservoir flow regimes. This technique also eliminates the time lag between downhole pressure measurements and surface rate metering. The downhole real-time flowrate measurement proved to be a consistently reliable method for visualizing when the well was in bilinear, linear and boundary dominated flow regimes. This real time data and workflow also enabled calculation of the real-time average SRV static pressure, which therefore made calculation of the PI and SRV relatively simple using physics-based workflows as opposed to correlations. As expected, the length of the horizontal drain had a quantifiable positive impact on both the SRV and the fracture flow PI, which was compared to the known analytical expressions for both these performance indicators to provide a forecasting tool. The SRV also showed an increase when the volume of proppant pumped per perforation cluster was also increased. The PI, on the other hand, showed an increase when the spacing between perforation cluster was reduced, which effectively increased the number of induced fractures for a given length of horizontal drain. Finally, the trends in rates of depletion and PI established a database of production profiles for known completion parameters, which enabled production forecasting using analytical tools as opposed to empirical decline curve models. The availability of high-quality real time downhole pressure and liquid rates enabled rapid well performance measurement, which not only enabled optimization of future completion designs, but also provides a means of evaluating well placement and completion and fracturing execution practices by comparing achieved well performance against a known historical base line. This data collection and measurement techniques is key to both maintaining and improving the ROI of well construction in unconventional wells.

A key component of an unconventional reservoir development is 3D characterization. A necessary precursor for any seismic inversion work is an inversion feasibility study. We shall demonstrate a best practice inversion feasibility workflow that has several key components: regional geology, petrophysics, rock physics, and geophysical analysis. The study shows an integrated approach using spatially diverse well data to cover the entire Delaware Basin, focusing on Avalon and Bone Springs formations. The results show the ranking of petrophysical properties that contributes to the changes in elastic properties. A good relationship was established between TOC, vclay and porosity to elastic logs using both conventional and unconventional RPMs. A class-IV AVO response was observed in the Avalon formation. Finally, our analysis showed that a depth trend based 1D Bayesian classification using bandlimited log data was able to separate organic rich high TOC facies from siltstones and carbonates. To conclude, an integrated approach involving geology, petrophysics, rock physics and inversion feasibility study increased our understanding of the basin and set path for further analysis. The results from inversion feasibility can be used to understand what facies and how much resolution can be resolved from an inversion which is further used to guide the drilling direction and landing zones. The workflow outlined in this study potentially can lead to a 3D inversion analysis, reservoir property estimations from seismic, TOC mapping and finally for finding sweet spots and better drilling/landing zones in the subsurface. Introduction Unconventional oil and gas production have increased dramatically in the last decade, and in the U.S., the Permian Basin is the most prolific of all the basins. The Delaware Basin located in the western part of the Permian Basin has become one of the most active drilling sites with multi-stacked plays (Mire et al. 2017). Most of the production comes from the Permian-aged Avalon, Bone Springs and Wolfcamp formations. These formations are comprised of a heterogeneous mixture of organic rich mudrocks, siltstones and carbonates (Nester et al., 2014). Due to the complex nature of these rocks, it is advantageous to understand and extract useful information from available data resources from all disciplines. Hence, a collaboration to perform an integrated approach between different disciplines is crucial to effectively find solution for complicated technical challenges in the Delaware Basin (Hoang et al., 2019; Anantharamu and Del Moro, 2019). The knowledge of geology and petrophysical analysis enhances our understanding of the basin and its mineral constituents. A proper rock physics analysis is extremely important for establishing a link between elastic properties and reservoir parameters, which can later be extrapolated to 3D domain using seismic and inversion workflows.

The Lower Barmer Hill (LBH) Member of Barmer Hill Formation is the major regional source rock in Barmer basin of Rajasthan and has sourced nearly all the discovered fields. Our previous studies helped to identify the geochemical potential of the LBH as a shale play. Its considerable thickness (50m-800m), high organic richness (6-14 wt.%) and optimum thermal maturity as indicated by vitrinite reflectance (VRo up to 1.7%) makes it a potential unconventional shale play. However, many other questions need to be answered before exploration wells can be drilled. In this paper, we have addressed those important questions and the associated workflow for answering them, with an emphasis upon 1) delineating the prospective areas, 2) estimating prospective resource volumes in these areas, and 3) estimating the chance of commerciality. We have adopted a play-based approach to identify prospective areas in the northern part of the basin. The LBH shale was divided into two play types (oil and gas) based on thermal maturity ranges of 0.7-1.1% VR o and 1.1-2% VR o respectively. The less prospective areas were eliminated by applying global cut-offs for thickness (>30m) and TOC (>3 wt.%). Finally, the fault segments and the gross depositional environment (GDE) map guided the subdivision of each play type into play segments. A total of 8 play segments (five oil and three gas play segments) were delineated for further exploration. We then estimated the hydrocarbons-in-place and prospective resources of each play segment. Each play segment was subdivided into sub-play segment polygons based on five different thermal maturity windows corresponding to different hydrocarbon phases. The probability distribution of in-place volumes and technically recoverable resources (TRR) for individual sub-play segment polygon was generated using a reservoir hydrocarbon pore volume and recovery factor approach. Next, we compute the minimum breakeven estimated ultimate recovery (EUR) on a single well basis assuming an economic hurdle of zero NPV 10 and production type curves from North American analog shale plays. The chance of meeting or exceeding this EUR for the average well (economic chance of success or ECOS) was then computed for each sub play segment. The 1U, 2U, and 3U Prospective Resources for the play segment were estimated by probabilistically aggregating the TRR distribution of its’ constituent sub-play polygons incorporating risk dependencies. The aggregated Prospective Resources numbers and the chance of success, along with other strategic parameters, help to rank the 8 play segments to high-grade projects for exploration drilling.

Acid fracturing can be potentially used in calcareous shale/mudstone reservoirs to stimulate the unpropped fractures. The unpropped fractures are created in large amount during hydraulic fracturing and provide considerable productivity, which, however, fails to sustain in the production stage. To understand how acid fracturing affects the fracture performance in laboratory is a critical step before field trial and can provide valuable guidance for fracture design. In this paper, we benchmarked the performance of acid-treated to brine-treated unpropped fracture using a "half-core" approach. The half cores to compare split from one shale core had very close mineralogy, they were treated with brine or acid independently and then form a composite core with a stainless steel half for conductivity measurement. This approach minimized the effect of mineralogy anisotropy in shale and cyclic stress that involved in traditional methods and enabled more meaningful comparison. Preserved Eagle Ford shale categorized into low, medium and high carbonate content were used; uniform and non-uniform flow patterns were compared in the treatment. Fracture conductivities and their response after cyclic stress, etched surface profiles and mechanical properties were systematically investigated. With uniform flow, acid resulted in lower fracture conductivity and restored conductivity after cyclic stress than brine. Both conductivities further decreased with carbonate content. On the contrary, non-uniform flow led to a higher acid-treated than brine-treated conductivity. The difference reached two orders of magnitude at reservoir pressure in sample with high carbonate content. The restored conductivities after acid treatment were higher than or close to that after brine treatment. Non-uniform etching was shown critical in a successful stimulation of unpropped fractures, and it could be achieved by viscous fingering or presence of carbonate veins. Most samples after acid etching developed isolated pits with depth within 50 µm on surface, channels were only observed for the sample with carbonate content etched with non-uniform flow and the one with calcite vein. Acid treatment reduced hardness in all samples, and the reduction tended to increase with the original carbonate content in samples. Mineralogy impacted the unpropped fracture conductivity, even in the cases of brine treatment, where no chemical reaction happened. We experimentally demonstrate the feasibility of acid fracturing in shale and provide insight to shale candidate selection for acid fracturing.

In the petroleum industry, accurately predicting production potential of undrilled unconventional horizontal wells requires the use of highly complex models and is a constant research in progress. Deterministic approaches such as decline curve analysis (DCA) are used in most cases to estimate production potential due to ease of implementation. The main disadvantage of using DCA by itself is that it requires an existing well to forecast the production, which is very expensive to drill. This paper shows the procedures for building a flexible machine learning based decline curve-spatial method that can be easily used to predict the estimated ultimate recovery (EUR) of newly proposed wells without the requirement of costly data or other time consuming methods. A type of artificial neural network (ANN) called feed forward neural network (FFNN) was used as the machine learning method during this process. In order to achieve this goal, production and well data were collected from public domain sources. Power law exponential (PLE) DCA method was implemented on a portion of the existing wells with sufficient production history. The data then was divided into training and test sets. The training set was fed into the ANN model and the results were compared with the results obtained from other inherently spatial methods such as universal kriging, geographically weighted regression (GWR), and generalized additive model (GAM). Finally, the EUR of the new wells were compared to the original training and test data to observe any discrepancies in the prediction and necessary adjustments to the model hyperparameters were made when there were discrepancies. Analyzing and observing all the results from the various combinations of methods indicates that using ANN without any spatial correlation is a less reliable method to estimate the production potential of new wells in the Marcellus shale gas reservoir when compared to other inherently spatial methods. This study shows that it is necessary to include spatial correlations between wells in the EUR prediction of new wells in the ANN models. The predictive strength of other spatial methods were within a reasonable range. One of the main input parameters that may increase the accuracy of the model are completion parameters. However, most of these completion parameters that will affect the production are very difficult and expensive to obtain. These parameters which could not be included as inputs can increase the accuracy of these models when available. This experimental study indicates that there are many other possible variations of this method other than the methods discussed in this paper. It also shows that the spatial correlation of wells is highly important when predicting the new well production in unconventional reservoirs. This is a flexible method, which can be easily modified to find better predictive methods for other areas of interest.

One of the least expensive ways to improve the production of a horizontal well is to target the best rock. Though operators have developed robust petrophysical methods for selecting targets, regional-scale evaluations of the impact of that target selection could be improved. This type of information could expand capabilities to objectively identify analogs that were not previously obvious. This larger analog list could then be used to improve production, benchmark offset operators, and customize spacing/stacking configurations. Here, we quantify the impact of target on production from the Midland Basin. We apply a decision trees-based machine learning algorithm on a regional Midland dataset composed of production, geologic grids, completions header, and detailed geologic targets, totaling over 7,000 wells. The target data contains the location of the well within the zone, reflected as a fraction from 0 (top of zone) to 1 (bottom of zone). We leverage SHAP values (Shapley Additive exPlanations) to interpret the impact of target on expected production. The impact of position in zone varies from 13% for the Wolfcamp A to 5% for the Wolfcamp B and 7% for the Lower Spraberry Shale. This represents approximately 20,000 barrels two year cumulative oil for Wolfcamp A wells, for an net present value uplift potentially over $100,000/well. The best target within the Wolfcamp A and Wolfcamp B are usually near the top of the zone, though some location and target combinations, such as Irion County Wolfcamp B, show best production near the middle of the zone. We also show that with some targets, increased oil production is associated with higher water production, and in others, with lower water production. Selecting the best target and geosteering to keep in that zone are among the most cost-effective means to increase production, though changes to landing zone must also take into consideration the impact of spacing degradation and sharing of production from other zones. For instance, a Wolfcamp B well may see highest production if landed near the top of the zone, though that may increase spacing interference with Wolfcamp A neighbors. This study provides guidelines to inform landing, geosteering, and spacing/stacking planning decisions.

The U.S. Geological Survey (USGS) recently completed an assessment of water and proppant demands associated with petroleum production from the Upper Cretaceous Eagle Ford Group in South Texas. The water and proppant assessment builds on the 2018 petroleum assessment conducted by the Energy Resources Program of the USGS, and the assessed water and proppant quantities correspond with the undiscovered technically recoverable resource estimated in the 2018 petroleum assessment. We will present the results and a summary of the methods used in the water assessment which include estimates of water needed for drilling and cement, water needed to hydraulically fracture wells, proppant needed to perform hydraulic fracturing, and produced water from the Eagle Ford Group. The impacts of water demand and the requirements of wastewater disposal are key issues in planning and developing unconventional (continuous) petroleum resources in rural areas such as arid South Texas. While the number of annually drilled oil and gas wells has declined in recent years, the amounts of water and proppant needed to hydraulically fracture each new well have increased over time. This is partly related to an increase in lateral lengths but also due to an overall increase in water volume per lateral foot. Proppant demands have also increased, while proppant-to-water ratios have remained relatively constant over the last six years. Each new well adds to the total water from formation that is produced and requires either recycling or wastewater disposal. The intent of this assessment is to provide geology-based estimates of the water and proppant resources potentially needed to extract the estimated undiscovered technically recoverable petroleum from the Eagle Ford Group. By linking estimates of future water and proppant demands and produced water volumes with undiscovered petroleum resources, we provide information to support informed planning. Introduction The Upper Cretaceous Eagle Ford Group in South Texas is one of the largest producing continuous (unconventional) accumulations of oil and gas in the United States (Whidden et al., 2018). In order to produce hydrocarbons from this low permeability formation, each well is hydraulically fractured, which requires large amounts of water and proppant (primarily sand). Water is also used during the drilling process for drilling mud and cement for the well casing. Throughout the lifetime of the well, wastewater is produced along with oil and gas. Some of the wastewater consists of hydraulic fracturing fluid that returns to the surface, known as flowback water. The rest of the wastewater consists of brine that is produced from the formation over the lifetime of the well. All of this wastewater requires recycling or disposal.

Decline curve analysis has been the mainstay in unconventional reservoir evaluation. Due to the extremely low matrix permeability, each well is evaluated economically for ultimate recovery as if it were its own reservoir. Classification and normalization of well potential is difficult due to ever changing stimulation practices. The standard methodology for conducting decline curves gives us parameters associated with total contact area and a hyperbolic curve fit parameter that is disconnected from any traditional reservoir characterization descriptor. A new discrete fracture model approach allows direct modelling of inflow performance in terms of fracture geometry, drainage volume shape, and matrix permeability. Running such a model with variable geometrical input to match data in lieu of standard regression techniques allows extraction of a meaningful parameter set for reservoir characterization. Since the entirety of unconventional well operation is in transient mode, the discrete well solution to the diffusivity equation is used to model temporal well performance. The analytical solution to the diffusivity equation for a line source or a 2D fracture operating under constrained bottomhole pressure consists of a sum of terms each with exponential damping with time. Each of these terms has a relationship with the constant rate, semi-steady state solution for inflow, although the well is neither operated with constant rate, nor will this flow regime ever be realized. The new model is compared with known literature models, and sensitivity analyses are presented for variable geometry to illustrate the depiction of different time regimes naturally falling out of the unified diffusivity equation solution for discrete fractures. We demonstrate that apparent hyperbolic character transitioning to exponential decline can be modeled directly with this new methodology without the need to define any crossover point. Each exponential term in the model is related to the various possible interferences that may develop, each occurring at a different time, thus yielding geometrical information about the drainage pattern or development of fracture interference within the context of ultralow matrix permeability. Prior results analyzed by traditional decline curve analysis can be reinterpreted with this model to yield an alternate set of descriptors. The approach can be used to characterize the efficacy of evolving stimulation practices in terms of geometry within the same field, and thus contribute to the current type curve analyses subject to binning. It enables the possibility of intermixing of vertical and horizontal well performance information. The new method will assist in reservoir characterization, evaluation of evolving stimulation technologies in the same field, and allow classification of new type curves.

In unconventional plays, operators commonly use scaling factors to adjust actual or expected production between groups of wells with different completions designs-e.g., increasing proppant loading from 1500 lbs/ft to 1750 lbs/ft should see 6% production uplift. Though these scalars are easily deployed, they suffer inaccuracy in the time domain away from the date at which the analysis was anchored. Here, we present a machine learning-based study of the Bakken-Three Forks play of the Williston Basin, showing that large completions designs have the biggest impact on production between IP days 90-180, with the impact steadily decreasing through time afterwards. This method can be used to build scaling factors for any completions or spacing parameter by using SHAP values (SHapley Additive exPlanations), which isolate the contribution of each feature on the model prediction. Proppant loading, fluid loading, and stage length all show strong variation in scalar impact through time. All three parameters show diminishing impact over the life of the well, with large designs showing approximately 55% uplift in rates over average designs at IP90 but only 30% uplift over small designs by IP720. In contrast, the relative importance of inter-well spacing and geology increases steadily through time. SHAP values offer a powerful method to extract scaling factors from a tree-based machine learning model. Because they can be incorporated into the model-building pipeline, they remove the need to run synthetic cases or build partial dependence plots. Applying time-dependent scaling factors when making well predictions or building type curves will result in a more accurate production profile, improving decision making no matter whether the operator prefers payback or rate of return economic metrics. These methods can also be used to help answer to what extent intense designs or tight spacing accelerates or improves recovery. Introduction Engineers making decisions on unconventional development plans often rely upon scaling factors to generate type curves for a range of potential designs. This approach is often taken when implemented designs within a given area do not fully sample the possible design choices, such as in a step-out area where none of the first-generation wells were completed with large proppant volumes. Even in a well-developed area, an operator may have only a handful of well-studied and validated type curves, yet the unique permutations of reasonable design choices can stretch into the hundreds or thousands (e.g., five potential proppant per foot values, three potential fluid per foot values, three stage lengths, five inter-well spacing choices and three lateral lengths results in 675 possible combinations). Scaling factors bridge this gap by starting from the well-understood base production expectation (e.g., a type curve, EUR estimate, or point cumulative production value) and scaling the production up or down by multiplying by a scalar, commonly derived from descriptive statistics of grouped wells (Curtis and Montalbano 2017, Srinivasan et al. 2018, Al-Alwani et al. 2019)

In shale reservoirs, the stimulated reservoir volume (SRV) is comprised of multiple propped hydraulic fractures in the presence of a matrix and/or natural fracture system. Each SRV component is well known to exhibit stress dependent permeability such that there may be significant loss in permeability with increased drawdown. The concern about SRV damage due to aggressive flowback strategies has led to a variety of heuristic flowback guidelines of "x psi/day" across various operators and shale plays. However, conservative flowback strategies may also result in productivity loss due to water block and capillary trapping resulting from ineffective unloading of treatment water. Real time flowback management based on reservoir physics is therefore necessary for a balanced flowback strategy aimed at optimal SRV integrity, EUR, and well economics. Additionally, since maintaining SRV integrity is important from early transient flow to late boundary dominated flow, a flow regime agnostic approach is especially beneficial. Based on the dynamic flowing material balance, this paper presents a real time SRV integrity monitoring approach using a flow regime agnostic dynamic productivity index. Field applications for flow back diagnosis and rate acceleration are presented using examples from the Woodford, Eagle Ford, Permian, and Haynesville shale plays. Introduction Shale reservoirs are well known to exhibit stress dependent permeability. The effective stress change results from a decrease in pore pressure and is related to the absolute stress on the rock as in equation 1 below. Stress dependent permeability is therefore synonymous with pressure dependent permeability in this paper. (equation) Jones (1988) presented a two-point experimental determination of the inverse relationship between matrix permeability and net confining stress for pressures up to 10,000 psi. Kwon et al. (2004) experimentally developed a cubic law correlation for matrix permeability reduction with confining pressure. Crawford et al. (2019) demonstrated via uniaxial experimental setup that that classic net confining stress approach for rock stress measurements may overestimate the permeability modulus by up to 100%. They also demonstrate permeability hysteresis typical in stress dependent permeability systems. Yilmaz and Nur (1985) had previously introduced the permeability modulus as a measure of the sensitivity of permeability to pressure change and is defined as:

With signs of the shale boom slowing, the need to make informed decisions on asset development becomes increasingly critical in a competitive landscape. Yet the data for informed decision making is often spotty, particularly with regards to one of the industry’s greatest hurdles: maximizing stimulated reservoir volume with minimal investment. For many, the problems that underpin this hurdle are clear: improper well spacing, frac hits/stress shadowing, unsuitable connectivity between benches, and interwell interference/communication all well-established contributors to poor productivity and a non-ideal stimulated reservoir volume (SRV), a definitive strategy for the "best" way to develop different plays has not proven as obvious. One area that has seen sustained interest is the SCOOP/STACK play of the Anadarko basin. Since 2012, this geological region has proven one of the lowest-cost, highest margin plays in the U.S. With multiple stacked reservoirs, the effective stratigraphic trap of the area creates a continuous petroleum system with multiple development opportunities. In this study, non-radioactive liquid chemical tracers were pumped alongside stimulation treatments for dozens of wells placed in the Woodford, Meramec, and Osage layers. These tracers served to uniquely "tag" downhole oil and water phases, allowing the operator to quantitatively track the production of oil and water as well as their point of origin. On an individual level, these liquid tracers served as an effective production log, measuring changes in behavior over time. On a field level, however, the application of liquid tracers made it possible to evaluate bench-to-bench communication and the effect different completion designs had on interwell communication/interference. Multiple variables were considered, including the effect of geological features, frac order, well spacing, and parent pump-in protection to achieve an optimal completion strategy. This work shares many of the high-level lessons learned, providing a beneficial ROI by rapid fine-tuning of well spacing based on ongoing tracer communication over time.

The Canning Basin, located on the northern coast of Western Australia, is one of Australia’s largest basins at over 400K sq. km (figure 1), two times the size of the Permian Basin of Texas and New Mexico. It is adjacent to the prolific offshore Carnarvon Basin. A limited number of wells have been drilled with only marginal success on the basin margins bounding the northern deep Fitzroy Trough depocenter. Prograding Devonian and Carboniferous marine carbonates and siliciclastics were deposited in a transitional ramp setting in the Fitzroy Trough; defined source rocks with TOCs of .5 to 4.25% lie within this depocenter but limited information is known about the potential deeper sources. Multiple tectonic phases altered this basin, creating an extensive set of both transpressional and extensional fault systems. In the Canning Basin, the Australian government estimates potential tight gas resources are 74 TCF, and that an additional 70-150 TCF of shale gas resources are geologically and technically producible. This is approximately equal to the USGS assessed resource size for the entire US Marcellus shale gas system. Unconventional discoveries like those found in North America have shown that there is significant potential for basin-centered gas around the Fitzroy Trough. Due to Australia’s needs for natural gas, both to feed the declining conventional feedstocks for export, as well as for meeting domestic energy needs, several vertical wells have been stimulated in the last 10 years demonstrating the gas potential within the basin-centered gas play. Using prior work that was undertaken to develop a geological understanding of the Canning Basin, a US operator has initiated a new evaluation, including a large 3D seismic program (circa ∼200 sq. miles), with a commitment to future testing, drilling and stimulation of unconventional wells to help better define and understand the play with an eye toward economic production. Introduction Using the wealth of prior work that was undertaken to develop an understanding of Canning Basin geology, Black Mountain Exploration has initiated a fresh, robust evaluation of the basin-centered play within a large exploration permit along the northern margin of the Canning Basin within the Fitzroy Trough (Figure 2). This evaluation has included extensive utilization of the existing data within the northern part of the basin and will include a future acquisition of a large 3D seismic program. With this evaluation is a commitment to future testing, drilling and stimulation of unconventional wells to help better define the resource play. A review of the geochemical, petrophysical and vintage 2D seismic data establishes an extensive basin-centered gas fairway beginning at 2000m and continuing to over 4500m in depth. This fairway steps down from the basin margins into the depths of the Fitzroy Trough and is analogous and correlative to a similar petroleum system that hosts significant gas discoveries on the opposite margin of the trough.

Cyclic gas injection (huff-and-puff) in hydraulically fractured wells has been successfully applied, at single well or pad level, as an EOR method in tight unconventional basins such as Permian and Eagle Ford. Since 2014, a considerable amount of literature and research has been devoted to feasibility, implementation and optimization of gas huff-and-puff in fractured unconventionals. Most of these studies have been devoted to oil reservoirs and few publications have addressed gas condensates. Furthermore, to the authors’ knowledge, all the publications addressing the suitability of huff-and-puff EOR in the Montney are focused on individual well studies and none consider the performance of the process on a pad-level basis. This study attempts to study the application of huff-and-puff on a typical, stacked Montney pad and propose alternatives to increase recovery and improve the economics of the process. One of the drawbacks of cyclic processes is the unproductive time that is dedicated to injection and soaking. In this paper we evaluate the feasibility of implementing a sequential huff-and-puff strategy where wells in a pad convert to injector and producers sequentially to avoid unproductive time. The produced gas is reinjected back into the reservoir to eliminate the need to bring an external EOR agent, while taking advantage of the heavier fractions still left in the stream. Since compression is one of the highest costs associated with huff-and-puff EOR in unconventionals, injecting in one well at a time with a compressor that is continuously active will help in reducing the CAPEX significantly. To perform this study, publicly available data was used to construct a geomodel of typical reservoir in the liquids-rich portion of Montney. Geomechanical and natural fracture parameters were incorporated into the model. A 4-well pad was considered, and typical completion and pumping schedules were used to predict hydraulic fracture propagation and geometries representative of typical stimulated volumes in this play. A compositional numerical reservoir simulator was then used to forecast liquid recovery under different huff-and-puff scenarios. The results of this study show that under favorable stimulation conditions (planar hydraulic fractures with minimum frac-to-frac interaction), applying huff-and-puff in a sequential schedule under a zero net gas production strategy, can significantly improve pad condensate recovery through re-vaporization and frac-to-frac displacement mechanism. Cash flow projections of the zero net gas production huff-and-puff scenario indicate that such a strategy yields a better economic outlook when compared to the one obtained under primary depletion; this is true when the oil price is higher than $30/bbl. The implementation of the scenarios studied in this paper require a proper understanding of the hydraulic fracture geometry. If there is a greater degree of uncertainty on the hydraulic fracture characteristics, or the fractures are not entirely planar and there is increased interaction between them, it might be more advantageous to apply a pad-level injection/production huff-and-puff strategy.