Africa (Sub-Sahara) Gas was discovered at two separate levels in the Mronge-1 well in Block 2 offshore Tanzania. The discovery is estimated at between 2 and 3 Tcf of natural gas in place, bringing Block 2's estimated total in-place volumes up to 17 to 20 Tcf. Statoil (65%) operates the Block 2 license on behalf of Tanzania Petroleum Development Corporation, and partners with ExxonMobil Exploration and Production Tanzania (35%). Oil was discovered at the Agete-1 exploration well on Block 13T in northern Kenya. The well, drilled to a total depth of 1929 m, encountered 330 ft of net oil pay in good-quality sandstone reservoirs. Tullow Oil (50%) is the operator with partner Africa Oil (50%). Asia Pacific Indonesia announced plans to offer 27 oil and gas blocks in 2014 in regular tenders and direct offers.
Africa (Sub-Sahara) A drillstem test was performed on the Zafarani-2 well--located about 80 km offshore southern Tanzania. Two separate intervals were tested, and the well flowed at a maximum of 66 MMscf/D of gas. Statoil (65%) is the operator, on behalf of Tanzania Petroleum Development Corporation, with partner ExxonMobil Exploration and Production Tanzania (35%). The FA-1 well--located in 600 m of water in the Foum Assaka license area offshore Morocco--was spudded. The well targets Eagle prospect Lower Cretaceous resources. Target depth is 4000 m. Kosmos Energy (29.9%) is the operator, with partners BP (26.4%),
Africa (Sub-Sahara) Eni discovered gas and condensate in the Nkala Marine prospect offshore Congo. The discovery could hold from 250 MMBOE to 350 million MMBOE in place, the company said. In a production test, the Nkala Marine 1 discovery well in the Marine XII block yielded more than 10 MMcf/D of gas and condensate. Eni is the operator with a 65% interest in the block. The remaining shares are held by New Age (25%) and Societé Nationale des Pétroles du Congo (SNPC) (10%). Sonangol and Total will break ground on a deepwater oil pumping project that will increase Angola's production by more than 30,000 B/D.
Reservoir Mapping While Drilling tool, consisting of a Deep Directional Electromagnetic Propagation Resistivity (DDEM) Logging While Drilling (LWD) tool and associated imaging software can detect bed boundaries and map reservoir bodies laterally beyond the wellbore being drilled. It has been successfully deployed to resolve and overcome geological and reservoir uncertainties when drilling wells offshore Malaysia. In Case Study-1, the DDEM tool was used to locate and navigate the drilling borehole assembly within a turbidite target reservoir successfully. In addition, the Reservoir Mapping While Drilling tool was able to detect and map a hydrocarbon bearing sand ten metres below the original borehole. In Case Study-2, the DDEM tool was used to identify the current Gas Water Contact (GWC) in a carbonate field which was experiencing high water cut early in the life of the field. In Case Study-3, the DDEM tool was used to determine the Top of Carbonate (TOC) in a carbonate gas field. It was critical that the top of carbonate be identified correctly, as the surface seismic data was relatively poor and was not able to pinpoint the TOC accurately. In the first case study, the hydrocarbon pay sand would have been completely missed if standard LWD tools alone were used to drill the well in a mature field. In the second case, the DDEM tool was able to locate and map the current Gas Water Contact (GWC), which was found 35 metres below the wellbore, in the carbonate gas field. It was found that the water was being produced through a karst zone, although all the previous production wells had been drilled horizontally way above the original GWC. In the third case, the DDEM tool was able to detect the Top of Carbonate, thereby allowing to set the casing shoe above the TOC. The plan was to set the intermediate casing shoe just a few metres above the TOC to avoid encountering severe mud losses when drilling through the carbonate reservoir. This paper will discuss the various steps involved in planning, designing and drilling of these wells using the DDEM tool with the associated reservoir mapping software. The methods used in imaging the reservoirs of interest and the bed boundaries and results obtained will also be discussed.
Kyi, Ko Ko (PETRONAS) | Latiff, Nazri Abdul (PETRONAS) | Yen, Kok Kwi (PETRONAS) | Saadon, Danial (PETRONAS) | Rahim, M Ikhlas (PETRONAS) | Jaafar, Juhaidi (PETRONAS) | Afandi, Tomi (PETRONAS) | Fei, Ng Kiang (PETRONAS)
Tango Field, located offshore Sabah in East Malaysia, is a mature field which has been producing oil and gas for more than forty years. This field has many fault blocks, thus creating barriers to fluid and pressure communication between different fault blocks. Furthermore, the reservoir sands are turbidite sands which are difficult to correlate across the whole field. Being fan lobes, it is not easy to target these sands in drilling development wells. As part of the campaign to improve recovery and sustain production, two infill wells were drilled during 2014, by sidetracking two existing wells from the Tango-B Platform, which is located in the western part of the field. The target reservoirs are M1 and M2 sands, which still carry some upside potential based on the latest review of the field performance. To properly target and penetrate these sands in the planned wells, the Reservoir Mapping While Drilling LWD (DDEM) tool, in combination with standard triple combo LWD (Logging While Drilling) tools, was deployed. This is to ensure that the well trajectory stays within the targeted sands and the bed boundaries are detected long before the drill bit exits the sand body. Unlike previous deep reading LWD resistivity tools, the DDEM tool is a Deep Directional Electromagnetic Propagation tool which has the capability to see about 30 meters laterally beyond the wellbore. While drilling the first well, the target sands were penetrated as planned. However, there was a pleasant surprise where a new hydrocarbon sand was detected by the DDEM tool about 10 meters below the wellbore. The DDEM reservoir mapping software was used to image the newly found sand body. Based on this new finding, the drilling Bottom Hole Assembly was pulled back and the hole was side-tracked to target this new sand, which was successfully penetrated and completed. This new sand, which would not have been discovered with standard LWD tools has increased the well production by a factor of two or more. Being a turbidite sand, it was not picked up on the surface seismic section. The reservoir mapping software technology, together with the deep sensing resistivity imaging LWD tool, was instrumental in finding the new hydrocarbon sand which has substantially increased the production of Tango Field.
Cominelli, Alberto (Eni E&P) | Casciano, Claudio (Eni E&P) | Panfili, Paola (Eni E&P) | Rotondi, Marco (Eni E&P) | Del Bosco, Paolo (Schlumberger) | Trajtenberg, Horacio Damian (Schlumberger) | Thompson, Alan Mclure (Schlumberger)
The oil industry is developing more and more complex reservoirs, often lying in difficult environments like deep or ultra-deep water. At the same time, brown fields are systematically studied to implement IOR and EOR processes to increase the ultimate recovery factor. In this context one of the challenges is to simulate highly detailed models and provide robust answers for corporate decision making processes. Because conventional reservoir simulators are not very well suited for these purposes, new generation reservoir simulators are tempting solutions. During recent years eni is implementing a step-change in the way reservoirs are modeled and simulated, deploying a new generation, high resolution simulator for the most critical and complex assets. The purpose is twofold: computational efficiency on one side and enabling the development of more accurate models on the other one. The process is run in a selective manner, aimed at identifying opportunities when conventional simulators do not meet expectations.
In this paper we present the methodology used by the Company in the selection of field cases, together with the results achieved for some of the most interesting and complex assets. In particular, comparative results, with respect to conventional simulators, are presented for: deep-water reservoirs, a tight oil development, CO2 injection schemes, and a large scale heavy oil project. The analysis is performed using key computational and engineering performance indicators. The deployment is run in cooperation with the technology provider: cases, logic, issues and solutions are discussed together in a critical manner.
The process is run on the basis of long term corporate objectives, targeting the simulation of EOR processes and complex assets in a computationally efficient and accurate manner.
There is a common consensus in the oil industry about the fact that reservoirs are becoming more and more difficult. Exploration is targeting complex reservoirs, like deep water and ultra-deep water, with larger and larger investments often based on information collected during relatively short appraisal phases. The performance of IOR/EOR recovery processes heavily depends on reservoir heterogeneity, small scale geological features and difficult to model structural frameworks, and this is driving reservoir modeling towards detailed geological descriptions, which represent a challenge for reservoir simulation. At the same time the industry demands an integration of reservoir simulation with surface activities, both gathering networks and process facility, to predict production under real life conditions. These trends are expected to grow in the years, and the need to simulate production from unconventional assets, such as shale gas/oil or oil shale, represents a further stimulus. At the same time more accurate modeling is challenged by the need to anticipate as much as possible the start-up of new projects to increase net present value and sustain large capital investments. In this context reservoir simulation is often a real computational bottleneck and dynamic models are often over-simplified to meet economic time scales.
While many companies are hunting for elephants, exploration results in deep offshore plays over the past years show the increasing trend away from giant-field discoveries toward smaller fields in the 50 - 100 million bbl range, which tend to be geographically dispersed. These resources need to accumulate to a critical mass, a global threshold, to justify an economically viable development. This is not only a question of volume but also of geographic location of the discoveries, and the threshold, of course, also depends on economic factors.
A methodology is proposed to evaluate the potential of a block to lead to a multiprospect development and to optimize exploration and appraisal. It is illustrated by a real deepwater case study including five discoveries and four prospects, 10 - 30 km apart. The practical approach taken is to define circles or ellipses on a map representing potential hubs and their catchment areas. For each area, a global resource threshold is defined by analogy with other regional developments or by a detailed economic study of representative cases (not discussed in the paper).
A probabilistic model of the resource base is derived from the geological assessments of discoveries and prospects. It is entered into Monte Carlo simulation to generate a large number of scenarios representing exploration outcomes and discovery volumes, which are stored in a scenario database. This allows a probabilistic evaluation of the performance of an exploration and appraisal program, using specially developed indicators such as the cumulative discovered P50. Intelligence is introduced in the process by evaluating after each well the probability of meeting the threshold with the remaining wells. If it is low (10% or less) the program is stopped, which has a great impact on risked economics.
The main results of the analysis are the economic decision tree with the probability of a development decision and the P90/P50/P10 of the developed resources; the number of wells in the dry branch of the tree, which actually is not a fixed number but a probability distribution; the definition of a firm and contingent well program; and an optimum drilling order, which may also reduce the well count.
The evaluation of vertical sweep has been a long standing problem when determining waterflood efficiency in massive problem when determining waterflood efficiency in massive sandstones. As a result of reservoir heterogeneities and gravitational effects the vertical distribution of water near an injection well is typically different than its distribution in the reservoir. A thermal modeling approach has been successfully applied to this problem in the Sadlerochit sands at the Prudhoe Bay field to attain a greater understanding of the vertical sweep efficiency in a portion of this reservoir.
A thermal simulator was used to construct a 3-0 model of one half of a 320 acre Inverted 9-spot waterflood pattern in the Flow Station 2 waterflood area of Prudhoe Bay. The model was history matched to a temperature profile taken in a replacement injector drilled 720 feet (220m) away from Its original location. The amount and location of reservoir cooling observed is a function of the water volume that has passed through it at a given location and its associated conductive and convective cooling. The replacement well modeled was drilled 4.75 years after injection had been initiated in the original well. The analysis thus provides an understanding of the vertical water distribution in the reservoir 720 feet (220m) from an injector, 4.75 years after the initiation of secondary recovery.
The producing sand modeled is approximately 170 feet thick and is characterized by a high permeability (0.1-5.0 darcy) thief zone at its top and 0.5 to 1.0 darcy sands below. In this waterflood pattern approximately 62% of the injectant had entered this thief zone. Thermal modeling indicated that 60% of the water had slumped from this thief interval into the sands below, 720 feet (220 m) from its injection point. This result indicates that although the thief zone has taken a high percentage of the water injected into the pattern, the vertical percentage of the water injected into the pattern, the vertical distribution of that water in the reservoir is far different from its distribution at the sand face as measured in the wellbore. By quantifying the volume of water fiat slumped out of the thief zone, an estimate of vertical permeability was determined and our understanding of vertical sweep efficiency greatly enhanced.
The Prudhoe Bay Oil Field on the Alaskan North Slope is the largest producing oil field in North America with over 21 Billion STB oil originally in place. The main reservoir interval is the Triassic Ivishak sandstone of the Sadlerochit group. This main oil-bearing formation is at a depth of about 9000 fl. The rock quality is generally high with permeabilities in the range from 10 md to 5 darcies. Two permeabilities in the range from 10 md to 5 darcies. Two recovery mechanisms operate in the field. Gravity drainage is the principle process under the original gas cap. Water is currently being injected to promote oil production in peripheral parts of the reservoir. The waterflooding is being peripheral parts of the reservoir. The waterflooding is being performed using variations of inverted nine spot patterns with performed using variations of inverted nine spot patterns with interwell distances of about 1500 ft.
The occurrence and extent of water and oil gravity segregation i.e. slumping in Prudhoe Bay waterfloods has long been a issue of high uncertainty and conjecture. Field surveillance data has been misleading and at best speculative in establishing a understanding of the vertical sweep mechanism.
Historically waterflood injection has been dominated in Prudhoe Bay by thief zones as was anticipated by original Prudhoe Bay by thief zones as was anticipated by original design studies. To maximized oil recovery it is necessary to produce as much of the waterflooded interval as possible. As a produce as much of the waterflooded interval as possible. As a result, it is inevitable that larger and larger quantities of water will be produced. Consequently, any knowledge of the factors controlling waterflood sweep efficiency will allow these reserves to be produced while handling the lowest volume of water. Recently a replacement water injection well has provided much needed insights into the vertical sweep efficiency in the reservoir.
P. ***** P. *****
This paper presents the results of the application of a mixed-boundary pressure transient testing model to Prudhoe Bay wells. The proposed model considers a Prudhoe Bay wells. The proposed model considers a constant pressure upper boundary and a no-flow lower boundary. Reservoir anisotropy and partial penetration were considered in the model. These boundary conditions simulate the pressure transient testing in the Prudhoe Bay unit of the Alaskan North Slope where the Prudhoe Bay unit of the Alaskan North Slope where the oil column of the Sadlerochit formation is overlain by a huge gas cap.
A set of type curves for gas cap dominated data analysis is presented. These type curves should complement those presented by Kuchuk and Kirwan for no-flow upper and lower boundaries. A correlation for predicting the beginning of the constant pressure predicting the beginning of the constant pressure boundary effect is provide This correlation can be used to estimate the period to record analyzeable early-time data during the design of a pressure transient test. Also, it is a useful tool for validating the results of type curve analysis of data obtained from gas cap dominated transient tests.
Examples of pressure buildup tests conducted in the Western half of the Prudhoe Bay unit are analyzed. The results obtained are used to discuss the problems of single well pressure transient analysis in gas cap dominated reservoirs.
Most models presented for pressure transient analysis over the years have the common feature of simulating no-flow upper and lower boundaries. For the most part, this has been advantageous because most reservoirs are, in fact, bounded above and below by impermeable strata. The need for other boundary conditions arises when either the upper or lower boundary is not an impermeable boundary but rather a constant pressure boundary. Such is the case with many wells in the Prudhoe Bay unit of the Alaskan North Slope.
The oil column in a significant portion of the Sadlerochit formation is overlain by a huge gas cap. Pressure buildup tests run in these gas cap dominated Pressure buildup tests run in these gas cap dominated wells are often difficult to interpret because the gas cap acts like a constant pressure boundary that flattens the post wellbore storage portions of typical Horner plots. post wellbore storage portions of typical Horner plots. Attempts to analyze the gas cap dominated portions of the data as infinite acting will result in both high permeabilities and skin factors. Typically, the influence of permeabilities and skin factors. Typically, the influence of the afterflow extends well beyond the beginning of the gas cap effects. The need for early time data analysis becomes important and this has been suggested by several authors, notably, McKinley et al, Streltsova Adams, Kuchuk and Kirwan, and Brown and Mao.
In this study, we present a method for analyzing pressure transient tests using type curves calculated with pressure transient tests using type curves calculated with mixed boundary conditions. To simulate the pressure transient tests conducted in a Prudhoe Bay well with a gas cap, we modeled the gas cap as a constant pressure boundary. The heavy oil/tar zone that underlies the field is considered as a no-flow lower boundary.
This model is derived by the method of Green's functions. It is similar to the model of an unconfined aquifer developed by Hantush and Jacob in the hydrology literature. Recently, a similar model has been derived by Theuveny et al. Unlike Theuveny et al, in the present work the convolution of the pressure function present work the convolution of the pressure function was evaluated entirely in the Laplace domain and then it was inverted numerically into the real space. The proposed model is validated by comparison against proposed model is validated by comparison against previously published solutions. previously published solutions. A correlation for estimating the beginning of the constant pressure boundary effect is provided. A set of pressure buildup tests obtained from the western region pressure buildup tests obtained from the western region of the Prudhoe Bay unit are analyzed to demonstrate the use of the mixed boundary type curves.