The Upper Assam basin is a major onshore sedimentary basin located in the Assam-Arakan geological province in the north-eastern part of India. This case history of a high permeability, stratified sandstone reservoir of the Lower Eocene formation in Upper Assam basin illustrates how integrated pressure transient analysis coupled with geological, geophysical, petrophysical and production information can be used for construction of dynamic fluid flow models to provide better reservoir description for reservoir simulation study. Chronostratigraphic and litho-stratigraphic layering of reservoir units for this field done earlier through seismic and log interpretation suggested isolated units. Such a model led to the computation of high primary recovery factors. Independent analysis of well tests was found to yield conflicting results, especially with respect to outer boundary configuration. The reservoirs comprising of fine to medium grained sandstones with interbedded shale/carbonaceous shale are located at around 3600 m depth, generally very thin (1–5 m) and exhibit rapid lateral variation making well-to-well log correlation unreliable in many instances. Consistent modeling of flow barriers through analysis of all available well tests and integration of the same with other geo-scientific data led to a more reliable and consistent interpretation of field observations and facilitated refinement of reservoir description. A 3D-3P reservoir simulation study carried out with this model facilitated in establishing the fact that several sand units/reservoirs in the field under study are in pressure communication with a common aquifer system. This information allowed more reliable estimates of in-place reserves and provided good data for fine-tuning the reservoir model to arrive at reasonable history match. The new model used in reservoir simulation is much more plausible and brings out the importance of using dynamic information for reservoir characterization. Performance prediction of the field based on the new model aided in making realistic assessment of future production profiles and proper planning for field installation of formation water disposal and artificial lift facilities. Introduction Reservoir characterization is the process of defining reservoir properties and geological conditions for evaluating reservoir performance and forecasting future behavior. Between the micro and mega scales of reservoir heterogeneity, it is important to describe and analyze those reservoir characterization parameters that impact fluid flow. This leads to an economically optimized and contextually correct reservoir-engineering program. The ultimate goal of most reservoir engineering programs is to develop a mathematical model that closely reproduces known reservoir behavior. Such a model can be used as a forecasting and reservoir management tool. These models use large scale averaging to define grid block properties. Well testing is one of the most important means of measuring the properties of a reservoir at such a scale. In recent years, well testing has undergone significant technological changes to evolve from a technique for evaluating permeability and skin to diagnosing reservoir flow models1–2 and assisting reservoir management3. The Dikom oilfield is an example of a geologically complex field. Crude oil production is obtained from about 3500 m [11500 ft] deep, over-pressured reservoirs located in multiple, stacked, relatively thin sands of about 2 to 5 m [7 ft to 16 ft] thickness. Formation permeability varies from as low as 50 md to as high as 6000 md. Thin sands defy reliable sand to sand correlation and identification of reservoir units. Accurate interpretation of sand development pattern is beyond the resolution of seismic data acquisition due to large depths and thin sands. Lithostratigraphic correlation based on well logs is problematic due to very thin sands and extreme heterogeneity. The increased dependence on transient well testing as a tool for reservoir characterization was, therefore, spontaneous.
In this era of cutting-edge technology where stochastic and geostatistical modelling is the order of the day, conventional layer-cake modelling for highly heterogeneous reservoirs in two fields have yielded very good results. The performance predictions of reservoir simulation studies based on the conventional geological model of these fields are in close conformance with the actual production behaviour thus substantiating the efficacy of such type of modelling. The results of these studies recommended that additional wells be drilled for augmenting production and recovery. Mapping of advancing edge-water front from simulation based on the 3D seismic derived geological model helped in optimal positioning of the development well locations. Although the role and accuracy of geostatistical and stochastic modelling can not be denied, conventional layer-cake modelling, within certain limitations, is a fairly reliable and cost-effective reservoir characterization tool. Introduction The two fields under study viz. Dikom-Madarkhat and Kathaloni-Bachmatia, discovered in 1990–91, are prolific producers and have their hydrocarbonaccumulations in thin, layered reservoirs inter-bedded with shale/mudstone of Eocene sandstones in Assam-Arakan geological province of North-East India. The production from these two fields presently account for around 25% of the total production (8400 tonnes/day) of the company. The basic objective of these studies was to build dynamic models to chalkout optimum field development plans and assess alternative exploitation schemes subject to their financial feasibility. Due to the heterogeneous nature of the reservoirs, sand-to-sand correlation was a major challenge and the problem was aggravated by sparseness of conventional cores, PVT and pressure transient survey results. However, the best possible correlation using 3D-seismic data, well logs, pressure-production and available well test analysis data was established for constructing the geological models which were eventually subjected to fluid flow simulation. This paper deals with these two case studies where conventional geological modelling for reservoir characterization was resorted to and satisfactory results were obtained thereof. Dikom-Madarkhat Field — Field Overview The Dikom-Madarkhat-Kuhiarbari oilfield established commercial oil production for the first time from the Lower Eocene formation in Upper Assam in April 1990 with the drilling of well N-438. Madarkhat, a satellite structure of the Dikom main structure and located in its south-west vicinity, was identifiedon the basis of close-grid 2D seismic data acquired during 1991–93 and commercial oil was discovered in well M-1 drilled in 1996. A review of close-grid 2D seismic data in Dikom-Madarkhat area revealed that Madarkhat structure is connected to the Dikom structure by a narrow north-south trendingridge. As seen from Figure 1, Kuhiarbari is another satellite structure developed to the east of Madarkhat structure with a minor intervening saddle.The first well Kuhiarbari 1 was drilled during October-November 2000 and produced clean oil on testing. A pressure transient survey carried out in December 2000 in this well indicated pressure level similar to that of D-15(360–365 ksc). The initial pressure in this well was of the order of the pressure of N-438 Block thus indicating communication with the main Dikom-Madarkhat structure. Till date, a total of 22 wells have been drilled in the Dikom-Madarkhat-Kuhiarbari field of which 18 wells have been completed as producers. Wells D-4 and 11 did not encounter any commercial hydrocarbon prospects whereas D-1and D-5 could not be completed as oil producers due to blowout and casing related downhole problems respectively. The remaining 16wells are currently producing at a total rate of around 1380 m3/daywith about 40% water cut.
TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractA multi-disciplinary approach had been adopted to resolve the exploitation and development strategy for Deohal and its extension oilfield having mainly the deep Eocene clastic reservoirs which is geologically complex, over pressurized stack of thin sands [by 30-40 Kg/cm 2 (assuming 0.1 Kg/cm 2 /m hydrostatic gradient] interbedded with shale/ carbonaceous shale. An accurate delineation of individual sand in such reservoirs is beyond the resolution of seismic as these occur at a depth of 3600 to >4000 m and its thickness vary from 2 to 4 m only. The Lithostratigraphic correlation based on well log is often unreliable/ difficult due to thin and extreme heterogeneous nature of the sediments. Moreover, areal extent of reservoir as a single unit is difficult to ascertain. The petrophysical properties viz. porosity and permeability substantially deteriorate with increase in depth i.e. below 4000 m.Initially 3D seismic data was showing a broad faulted anticline with fault 500 m to 800 m away from the crestal part. Analyses of pressure transient data showed barrier nearby and reinterpretation of 3D seismic data confirmed the presence of minor faults having limited extension.The Paper presents how transient well test data in conjunction with static 3D seismic data, wireline log and dynamic pressure-production data have helped to workout development strategy for a geologically complex and heterogeneous Lower Eocene thin sand reservoir in one of the oilfields in Upper Assam Basin, India.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.