|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Abstract The objectives of this paper are to summarize effective Reserves estimation methods for use in unconventional reservoirs, and to propose systematic procedures for classification of Resources other than Reserves (ROTR) volumes. We propose optimal timing for application of decline curve analysis (DCA), rate transient analysis (RTA), and reservoir simulation. Using these techniques, we provide results for one well from a 38-well database in the Permian Basin wells (TX USA). We then describe how the volumes are classified and categorized and how those volumes move between Reserves and ROTR as more information becomes available. We begin with the analysis of well performance, where we specify the information that is necessary for each estimation method. We then suggest procedures to identify the flow regimes using diagnostic plots, provide guidance on the application of multi-segment DCA models, and finally suggest procedures for the application of RTA and reservoir simulation. We continue with progress toward Reserves classification, starting with suggested procedures to reclassify Prospective Resources as Contingent Resources (upon discovery). We provide post-discovery guidance on development and commerciality for the project maturity sub-classes (within the Contingent Resources classification). We explain that “established technologies” must be technically and economically viable before they can be used for development decisions. And finally, we examine requirements to remove contingencies so that the volumes can be reclassified properly as Reserves. Our major suggestions for well performance analysis are, first, that the multi-segment DCA approach is most effective in unconventional reservoirs when specifically relevant models are used for transient flow and boundary-dominated flow. Furthermore, we suggest that RTA using analytical models expands possibilities of forecasting for changes in well conditions and for well spacing studies. Though time and computationally time consuming, compositional simulation is required for confident analysis of near-critical reservoir fluids. For movement of resources toward Reserves, we suggest that there is no linear path to define the movement from Prospective to Contingent Resources, though there are certain criteria which must be met for a given project. Certain contingencies, such as price of oil and available technologies, dominate the classification of resource volumes. This paper provides a visual representation of when to use each Reserves estimation method depending on available data. We present a thorough analysis of best practices for each Reserves estimation method. We provide graphical representation of the movement between Prospective to Contingent Resources categories, the progression in chance of development and commerciality within project maturity sub-classes for Contingent Resources, and the contingencies that must be resolved to move from Contingent Resources to Reserves. Finally, we present an explanation of the criteria that must be met before volumes can be reclassified and/or recategorized from undiscovered to discovered.
Abstract The first objective of this work is to determine the volume of hydrocarbon that can be moved from Resources other than Reserves (ROTR) to Reserves, or from Proved Undeveloped Reserves (PUD) to Reserves based on well placement. The second objective is to create a model that incorporates the production history and forecasted estimated ultimate recovery (EUR), in this case by implementing multi-segment decline curve analysis (DCA) as presented in URTeC 336 (Moridis et al.(2), 2019). To perform this analysis, we selected 38 wells from a Permian Basin dataset available to Texas A&M University. The first portion of this work involves running a sensitivity analysis to determine the spatial well relationships that may trigger movements in certain regulatory frameworks. A successful well may promote the offsetting 2P wells to PUD wells. We incorporate the methodology in SPEE Monograph 3 (2013) for estimating PUD volumes beyond immediate offset locations that can be used to estimate the Reserves and possibly Contingent Resources in some situations. The question we aim to answer is: How do we move the PUDs to proved developed producing (PDP) Reserves? In the second part of this work, we create a model which includes the production history and the forecasted EURs. As time moves forward, continuity and consistency must be maintained across the model. Assume the following scenario: we plan to move a volume "x" from 1C Contingent Resources to1P Reserves, but we can only book 0.7 x as 1P Reserves. The model must reflect the fraction of the volume x that was actually moved and how it depends on, for example, commodity price contingencies. The remaining 0.3 x volume that was not classified as Reserves must be accounted in the model. The continuity of the model through time will track the volumes, and it needs to be able to do so consistently.
Abstract Over the past decades, different researchers have proposed various decline relations attempting to model the rate/time behavior of unconventional reservoirs in order to estimate reserves and forecast future production performance. Reserves estimation is a process that is thoroughly renewed during the life of a reservoir. Its accuracy depends on the amount of available data and the method of forecast. Forecasting production and estimating ultimate recovery (EUR) in unconventional reservoirs have long been problematic. Unconventional reservoirs exhibit fracture-dominated flow regimes and rarely reach boundary-dominated flow (BDF), even over several years of production. The reserves estimation in these reservoirs is not straight forward like in the conventional ones because of the variation of reservoir properties and completion type. Many challenges are facing the forecasting of such reservoirs due tohigh initial rates, sharp decline in the transient flow period, and shallow decline resulting from BDF. For that, it is difficult to match the whole production trend of an unconventional wells using a single decline curve relation. Traditional decline curve such as Arps' method have been successfully used for forecasting production and estimating EUR of conventional reservoirs. However, using the traditional Arps' with low permeability reservoirs yields over-forecasted reserves. Arps' model can be adapted for estimating reliable reserves of unconventional reservoirs by using different values of b-factor based on the flow regime. Different relations and models have been presented to model the production behavior in shale gas reservoirs as alternatives for the traditional Arps' decline model, such as Modified Hyperbolic decline, Power Law Exponential Decline (PLE), Stretched Exponential Decline (SEPD), Logistic Growth Model (LGM), and Duong method. All of them are based on empirical observations of a particular scenario, and thus often provide different forecasts. The application of those methods is depending on the behavior of each relation. This work presents a new look at decline curve methods for reserves estimation in unconventional gas reservoirs. Recommendations are also presented based on practical applications, which might help in understanding the behavior of such problematic calculations. Additionally, a recommended workflow is presented for better application of decline curves in estimating unconventional reserves using a short period of production.
Ambrose, Ray J. (Devon Energy and The University of Oklahoma) | Clarkson, C. R. (University of Calgary) | Youngblood, Jerry (Devon Energy) | Adams, Rod (Devon Energy) | Nguyen, Phuong (Devon Energy) | Nobakht, M.. (Fekete Associates Inc. and University of Calgary) | Biseda, Brent (Devon Energy)
Abstract Low-permeability (tight) and shale (gas and oil) reservoirs have emerged as a significant source of energy in North America. Recent advances in technology, such as long horizontal lateral/multi-lateral drilling combined with hydraulic fracturing, and new surveillance techniques, have enabled commercial production from ultra-low permeability reservoirs, previously considered source or cap-rock, not reservoirs. Forecasting well production for reserves estimation, hydraulic fracture stimulation optimization, and development planning remains a challenge because of complex reservoir behavior and flow geometries associated with current wellbore architectures/stimulation treatments used to exploit tight formations. Depending on the completion design, transient flow periods can last for weeks to years, and hence traditional methods requiring boundary-dominated flow are strictly inapplicable for most of the commercial life of many wells completed in tight formations. Recently, several analytical (type-curve, flow-regime analysis and simulation) and empirical approaches have been introduced to match and forecast tight reservoir production. The challenge is to develop routine techniques that can be used to forecast tight formation production, while adequately addressing the complex physics of the problem. In this work, we build on recent attempts to combine analytical and empirical methods ("hybrid" methods) for forecasting tight/shale gas reservoirs completed with multi-fractured horizontal wells. We forecast the homogenous completion (equal hydraulic fracture length) case using established analytical procedures for transient linear flow (pre fracture interference), combined with the Arps decline curve for late-time (boundary-dominated) flow. We also examine the heterogeneous completion (unequal hydraulic fracture length) case to establish the impact of heterogeneities on decline characteristics post fracture-interference. Finally, we present an innovative method for designing hydraulic fracture and well spacing.
Abstract Shale gas currently provides 20% of domestic supply, is targeted by half of the gas-directed drilling rigs, and represents the large majority of domestic resources. However, modern shale plays, their development strategies and their engineering analysis are young by comparison to those of conventional reservoirs. Uncertainty in shale gas reserves has significant implications at both the micro and macro levels. Conventional reservoir engineering tools must be viewed as potentially inadequate (or even inappropriate) for the evaluation of shale gas performance primarily because of the extremely low aggregate permeability of these systems, but also because of other unique aspects of the systems. Reservoir modeling (simulation) has an important role as an assessment and prediction tool; however, the character of the reservoir (induced and enhanced natural fractures) must be considered, as well as the geological and fluid characteristics. Rate-transient analysis (modern decline analysis) techniques are also more rigorous and have been expanded and adapted to fit the uniqueness of shale gas production. Application of each method for shale gas is discussed, including methods and limitations. These two techniques more closely represent the physics of shale gas production, but their implementation is often prohibitive. By way of necessity, much engineering evaluation is performed using Arps decline curve analysis. This technique is argued by some to be inappropriate due to a lack of theoretical support and demonstrated tendency to over-estimate reserves in tight gas systems. Given the limitations, practical methods exist to reduce error associated with its use. A newer decline method, power-law exponential, is also investigated.