Abstract: In this work, a comprehensive field study was conducted to assess the production condition of oil field A in southwestern Iran. The reservoir is a mature, naturally-fractured carbonate, vuggy formation. The production began in 1966 with an initial reservoir pressure of 3,800 psia. The gas injection into the field commenced in 1999 representing nearly a 20-year delay in pressure support; currently gas is being injected through seven injectors. By 2009, 1.68 billion STB (stock tank barrel) had been produced through 49 producers with a current average reservoir pressure of 2,470 psia at 1,800 meters sub-sea. The biggest production challenge is the additional gas production and gas coning which can hamper the improvement of production strategies and further development of the field. In order to gain a proper understanding of the possible production challenges, a thorough study was conducted on the effect of gas injection on fluid pressures, fluid contact levels, and reservoir properties. Also, the effectiveness of the gas injection in enhancing production has been evaluated. A series of data of well tests, PI (productivity index) tests, well logs, core, PVT (pressure, volume and temperature relationships), production logs, and geology together with the simulation results have been used. The data of the integrated production logging was used as a screening tool and the results were interpreted and presented for four horizontal and highly-deviated wells. Using PLT, the advantages, limitations, and effects of the gas injection on field production performance were revealed. Finally, a practical solution to overcome the challenges encountered in improving the production was proposed.
　　Key words: Naturally fractured reservoir, fracture intensity, production logging, gas injection.
　　Conversion Factor
　　1 cP = 1.0000E-03 Pa·s
　　1 Darcy = 9.8692E-13 m²
　　1 ft = 3.0480E-01 m
　　1 psi = 6.8948E+03 Pa
　　1 bbl = 1.5899E-01 m³
　　1 lbm = 0.453592 kg
　　1. Introduction
　　In order to forecast the production of a mature field and have a proactive scheme to improve the existing production strategies, in-depth analyses of the field and the problems associated with each individual well should be the first essential step under consideration in the field development stage. This approach should include different methods that are applied to gather the required information of both the problematic wells and full field histories from techniques such as well testing, well logging, production logging, and etc.. PLT (production logging tools) yield reliable and real-time information of each well which can be implemented for whole field monitoring, including down-hole and surface measurement of pressures, temperatures and flow rates. This is especially important in the case of the targeted oil field in this study because it produces from a vuggy, Asmari formation with highly connected fractures and a large initial gas cap. In order to obtain an accurate representation of the reservoir drainage pattern, PLT can be applied to obtain the flow rate profile along the well. In turn, information on the drainage pattern will affect decisions on how to plan infill drilling or remedial actions and provide guidance for workovers and improved oil recovery projects. Meanwhile, average pressure and the PI (productivity index) of each producing or injection interval can be determined through zone-by-zone measurement of pressure and flow rate [1-7].
　　This paper investigates the production history of an Iranian field and the status of the gas injection project into its Asmari formation by analyzing the production challenges of select individual wells from the field. In order to preclude any disparities across the field, it was divided into three sectors from the north, south and west. Variation of fluid pressure and fluid contact levels for different sections are presented that correspond to the time period from the inception of the gas injection program to the present. Meanwhile through utilizing PLT, the production of each well has been monitored and the interpreted results are provided. Finally, a practical solution to improve the production has been introduced. PLT was run to evaluate the free gas production, flow contribution of the open-hole sections, gas entry detection to determine how production can be optimized, and to obtain required data for a full field development study. Selected field examples are represented by four horizontal and deviated open-hole wells that are exhibiting excess gas problems which are located in different parts of the field. Besides PLT, the results of well tests, PI tests and other production data have been applied to investigate the status of gas injection and production condition of the field.
　　
　　Fig. 1 Cross-sectional view of the targeted field.
　　2. Field Production History
　　The field was discovered in 1964 and contained a large gas cap in the south Zagros Mountains. It produces from the Asmari limestone and was estimated to have 12.29 billion barrels of OOIP (oil original in place).The field’s reserves were estimated at 3 billion barrels of 34.2o API (American Petroleum Institute) oil and 9 TCF (trillion cubic feet) of gas [8]. Fig. 1 shows a cross-sectional view of the targeted field and its surrounding. The field began producing oil in 1966. In 1970, the oil production had peaked to 380 MSTBD (thousand stock tank barrels per day); however, during operational tests in 1992, it was producing at 80 MSTBD. Later production was suspended due to reservoir formation damage. The output capacity was eventually raised to 110 MSTBD and since 1999, the start of an immiscible gas injection in the Asmari reservoir has maintained the production rate at around 100 MSTBD. The field originally had a 1,024 m oil column of 34.2° API oil. The initial WOC (water-oil contact) was at 2,066 m elevation and the initial GOC (gas-oil contact) was determined to be at 1,042 m. By 2006, the oil column had been reduced to less than 500 m as the WOC had risen to 1,798 m and the GOC dropped to 1.594 m. The cumulative oil production at this time was 1.5 billion barrels.
　　The initial reservoir pressure was 3,800 psi, which remained nearly constant during the early life of the field. The pressure declined rapidly to 2,680 psi during 1970 to 1978 as can be observed in Fig. 2. Between 1978 and 1990, the pressure declined more slowly due to curtailing production. Since 1999, the reservoir pressure has been maintained at 2,400 psi through the implementation of the gas injection scheme. The field has been plagued by an extensive fracture network, resulting in the production of excess gas. Subsequent simulation strategies recommended high-flow-rate gas injection while limiting the production rate. Currently around 300 MMSCFD gas has been injected through seven injection wells.
　　3. Asmari Reservoir of Field A
　　Asmari reservoir of the field A is a fine-grained, consolidated reservoir, and is characterized as a highly fractured, vuggy formation with low porosity and permeability. According to recent PI test results, the average permeabilities in north, west and south sectors are 730, 180 and 490 mD, respectively. Average porosity of north and west parts is 13%, while that of south sector is around 10%.
　　Oil flow is through reservoir fractures which are horizontally and vertically connected. However, due to intense fracturing, the porosity and permeability are greatly enhanced in some areas resulting in porosities up to 25% and permeabilities in excess of 100 mD. The Asmari reservoir has some fine fractures which make connection between the pore volume across and along the reservoir. The fractured carbonate reservoir is produced under natural depletion mechanism for its entire production life. Most parts of the Asmari reservoir have already been depleted and the field is in dire need for an enhanced oil recovery technique to further secure production rates. Fig. 3 shows a structural contour map of the top of the Asmari formation in the oil field A with location of the four selected wells.
　　
　　Fig. 2 Oil pressure and daily oil production rate since the start of production of the field and before and after gas injection.
　　4. Production Logs as One of the Indicators in Fractured Reservoirs
　　Reservoirs that contain naturally-occurring fractures quite often exhibit significantly improved reservoir fluid flow either in the form of increased anisotropic reservoir permeability and/or revealed as additional reserves [9]. Although there exists some techniques for dynamic fracture indication such as PI or well testing, in PLT, flow meters are utilized to directly detect a well’s production from one individual fracture or an intensely fractured interval. The presence of a major feature such as a fault or intensely fractured zone can be indicated by a great influx of fluid over a short interval. Alternatively, a gradual flow over a longer interval could be indicative of a small diffuse fracture network.
　　
　　Fig. 3 Structure contour map on top of the Asmari formation, field A, faults shown as heavy black lines.
　　5. Fracture Intensity of Field A
　　In field A, the occurrence of natural fractures is clearly related to the structural deformation. Also the faults that are traversing the hinge of the structure exhibit strong control on the corresponding fracture intensity. The area of maximum fracture intensity is typically near the hinge of the fold, with local maxima being situated in the immediate proximity of the faults. Average fracture permeability was found to be 5,500 mD, with maximum values in excess of 15,000 mD near the faults. Fracture porosity varies between 0.28 near the faults, down to 0.15 in the unfaulted portion of the fold, resulting in an average fracture porosity of 0.22.
　　
　　Fig. 4 Variation of fluid level contacts and oil production rate of field A (Asmari reservoir) before and after gas injection.
　　6. Fluids Level Contacts
　　In order to evaluate the effect of injected gas on reservoir net pay, the variation of GOC and WOC has been presented in Figs. 4-6 demonstrating the change in GOC and WOC, respectively in three different sectors of field A. Average oil and gas pressure of 2005 psi were used to calculate the GOC and WOC in the north, south, and west regions of the field; they were found to be 1,601 msubsea, while WOC was 1,896, 1,886 and 1,885 msubsea in north, south and west side of the reservoir, respectively. As can be observed in Fig. 4, the thickness of the reservoir has been decreased sharply when first brought on production, but later it reduced steadily by decreasing the oil flow rate.
　　
　　Fig. 5 Gas oil contact change since the start of gas injection for different sections of field A.
　　
　　Fig. 6 Water oil contact change since the start of gas injection for different sections of field A.
　　7. Fluid Pressures
　　According to Fig. 2, the oil pressure showed a severe reduction during the second stage of the field history, reiterating the need of applying an improved oil recovery project at this time; however, since the end of the second development period until the present, the pressure has remained nearly constant with only a small decrease during the fourth production period. In 2005, static pressure surveys are using a reference depth of 1,800 msubsea. The oil pressure was found to be 2,481, 2,474 and 2,456 psi in the north, south and west regions of the field, respectively (Fig. 7). While, gas pressure was measured using an 1,100 msubsea reference and found to be 2,166, 2,161 and 2,137 psi for north, south and west regions of reservoir, respectively (Fig. 8).
　　Similarly, the water pressures were measured at 2,150 msubsea by static pressure test and found to have values of 2,930, 2,950 and 2,950 psi in north, south and west section of the reservoir, respectively(Fig. 9). Although the amount of injected gas was below the forecasted rate, the variation of oil and gas pressure clearly shows the effectiveness of the gas injection scheme to keep the production rate almost constant. Fig. 10 compares the gas injection rate with the forecasted amount during the production period of 1999 to 2005.
　　
　　Fig. 7 Oil pressure change since the start of gas injection project for different sections.
　　
　　Fig. 8 Gas pressure change since the start of gas injection project for different sections.
　　8. Application of Production Logging in the Field A
　　The PLT string developed by Schlumberger Company was utilized on the subject wells and includes: basic measurement tools (telemetry, gamma ray, casing collar locator, temperature and pressure), gradiomanometer tool (density, deviation), Ghost (gas holdup optical sensor tool), in-line spinner and the flow-caliper imaging tool (directional full-bore spinner, X-Y caliper, local electrical probes and Deft(water holdup tools)). Among the above mentioned tools, Ghost is used to distinguish between gas and liquid by means of optical sensors. In the gas-liquid mixtures, the optical signal reflected by the probe is used to determine the gas holdup and a gas bubble count, which is related to the gas flow rate. The Ghost holds four probes mounted on a four arm centralizer. Each probe is on the inner diameter of a centralizer blade, thus providing some protection. This configuration is effective in giving good coverage of the wellbore in vertical, deviated and horizontal wells. Production logging is the primary reason for running such sensors in order to determine the source and rate of produced fluids in a well. In high GOR (gas oil ratio) wells, the temperature data can be useful in increasing the confidence of identifying gas-entry intervals, especially for the first entry and for significant gas entry from one interval. The temperature drop data supports the entry-interval determinations. In the flowing condition, the production logs are applied to obtain the production rate of each zone using a spinner graph. The locations of the producing zones are obtained from the temperature log, down-hole fluid properties by calculating pressure and finally volumetric flow in two-phase flow and entry points in three-phase flow through fluid density tools and probes. In shut-in conditions, the spinner curves are used to confirm the validity of the flowing spinner calibration and from temperature gradient, evidence of cross flows or leaks between layers can be determined. In horizontal wells, due to fluid segregation during shut-in conditions, the fluid density/probes accurately determine the density of each phase and by computing pressure, permeability and skin can be calculated [1]. In addition to flow rate calculations from full field studies, another important use of PLT is extracting real time fluid properties which can be obtained precisely on the well site at logging time. In the current field cases discussed here, the calculated surface and down-hole PVT data are presented in Tables 1 and 2. Calculated oil and gas flow rate and the contribution of each producing layer are also presented in Table 3 (all intervals are based on measured depth).
　　
　　Fig. 9 Water pressure change since the start of gas injection project for different sections.
　　
　　Fig. 10 The amount of injected gas versus forecasted rate.
　　Table 1 Surface PVT data obtained for subject wells (PLT results).
　　
　　Table 2 Down hole PVT data obtained for subject wells (PLT results).
　　
　　Table 3 Calculated flow rate for each producing interval (PLT results).
　　
　　9. Field Examples
　　Well-27; History: This well was drilled and completed with 6.125 inch open-hole horizontal producer with 211 m in length (2,367-2,578 m) and total depth of 2,578 m. Flowing and static bottom hole pressures were determined to be 2,449 and 2,461 psi, respectively.
　　Results: According to the PLT interpretation results, the main oil producing interval is 2,561-2,578 m with 73% contribution in oil production. There is also free gas production with oil in this interval and the estimated GOR is 870 SCF/STB. Table 3 shows oil and gas flow rates calculated for each producing interval and their contribution in well production. The interval 2,367-2,475 m has the most contribution to total gas production with a GOR of around 4,500 SCF/STB. Fig. 11 shows Ghost and Deft readings, as well as oil and gas production contribution for the whole open-hole interval. There is water recirculation above 2,425 m, gas holdup in the upper side and water holdup in the lower side of the open-hole interval. Density readings should have a decreasing trend towards the top. As the pressure decreases more, additional gas will be liberated from oil, and density decreases more; while in the case of water entry or circulation, the density at the lower depth is slightly higher than the density at greater depth, as can be observed from Fig. 11. Based on Fig. 12 and from the Ghost reading in the interval 2,440-2,570 m, one probe is in the gas phase almost continuously, one probe exhibits gas occasionally, and the other probe is in liquid phase. In the interval 2,420-2,440 m, two probes are in gas phase almost continuously. Above 2,420 m, all probes see a mixture of oil and gas in the presence of re-circulating water. The total oil production rate was 2,699 STBD (stock tank barrel per day) and total GOR was 1,850 SCF/STB.
　　Well-43; History: This well was drilled and completed with 6.125 inch open-hole, deviated (39°) producer with a 157-m length (2,383-2,540 m).
　　Results: The main oil producing interval is 2,417-2,432 m (55%) and 2,460-2,465 m (23%).
　　
　　Fig. 11 Deft and Ghost probes readings, and production contribution of producing zones of well-27.
　　
　　Fig. 12 Ghost (gas holdup) readings along with well-27 horizontal trajectory for three intervals of 2,420 to 2,440, 2,455 to 2,510 and 2,510 to 2,578 meters.
　　Free gas is produced at a rate of 491 MSCFD(thousand standard cubic feet per day) from the interval at 2,396-2,403 m. There may be additional gas production from the interval at 2,383-2,392 m (see Table 3). No considerable water production or circulation has been detected. Total flow rate was 5,578 STBD oil and 3540 MSCFD gas. Calculated GOR is 635 SCF/STB. Besides temperature log, temperature deviation can be used to detect gas and liquid entry. Fig. 13 clearly shows how temperature, spinner and gas hold-up tools in PLT can help to recognize oil and gas production in the well.
　　Well-52; History: This well was drilled and completed with 6.125 inch open-hole horizontal producer with a 406-m length (2,284-2,690 m).
　　Results: Oil and gas is mainly produced from an interval below 2,480 m. Gas is channeling behind the casing and is produced from the casing shoe as shown in Fig. 14. The contribution of each phase is as follows: below 2,620 m the fluid consists of 30% oil and 40% gas; the 2,450 to 2,600 m section contribution is made up of 53% oil and 60% gas and 2,300 to 2,450 m section indicates only oil production consisting of approximately 17% of the total contribution (Table 3). There is standing water at the hill section of the horizontal drain. This section was placed at the deepest part of the wellbore. Flow regime is between slug and bubble flow. The estimated total oil production rate and GOR were 2,158 STBD and 957 SCF/STB, respectively. Due to the location of the Ghost probe, gas might not be detected in the locations 90 degrees and above due to its location far away from upper side of the hole. Lighter phase (gas) will flow on top and faster in the 89 degree position than in 90 degree. Also there may be gas holdup in the upper side and water holdup in the lower side of the open-hole interval.
　　Well-64; History: This well was drilled and completed with 8.5 inch open-hole horizontal producer with 260 m length (2,303-2,563 m).
　　
　　Fig. 13 Production logging tools reading in flowing condition of well-43 (application of spinner and temperature logs).
　　
　　Fig. 14 Well-52, tool readings along with well trajectory and imaginary holdup profile from Deft and Ghost.
　　
　　Fig. 15 Well-64, production contribution of producing intervals in downhole conditions and tool readings in flowing condition.
　　Results: Production of gas at the initial 25-m of the open-hole interval was detected. No water production/circulation was detected, but at the end of open-hole interval there is stationary water (as determined from the Deft reading). The interval from 2,310-2,320 m produces gas and oil (contribution from oil production: 61%); the 2,470-2,482 m interval is an oil producer (contribution on oil production: 39%) and other intervals do not have significant contribution in production (Table 3). Gradiomanometer readings below 2,450 m are not reliable since the tools are in the horizontal position. Both gradiomanometer density and Ghost readings are indicating no gas production below 2,320 m. Total oil measured was 3,330 STBD. Fig. 15 shows production contribution of producing intervals at down-hole conditions and PLT reading in flowing condition.
　　10. Results and Discussion
　　A gas injection strategy was implemented to maintain reservoir pressure at approximately 2,800 psi. Although the amount of injected gas increased after starting the project, the pressure continued to decline because of accelerated oil production; without the implementation of an appropriate pressure maintenance program to revitalize the reservoir’s energy, oil production will continue to decline and the reservoir pressure will fall accordingly. As a result, the oil column will be reduced by the expanding gas cap, lowering the GOC and increasing the chance of more unwanted gas production. According to the simulation study, in the case of injecting 400,000 cubic feet of gas per day, the field production rate can be stabilized at 150 MSTBD for approximately 13 years, with the cumulative oil production reaching approximately 3 billion barrels at the end of life of the field corresponding to a 26.7% recovery factor. Therefore, considering the 1.56 billion barrels of oil produced up to 2004, the field can be expected to produce another 1.4 billion barrels oil due to the pressure maintenance program. A 20 year delay in commencing gas injection in field A has caused the reservoir pressure to decrease by 1,023 psi, which has shrunk the oil column to about 365 m, as well as many other problems in putting new oil wells on production. Due to the large number of vertical and horizontal fractures in the reservoir rock, drilling operations in the Asmari formation generally occurs with high fluid loss. Therefore, drilling must be implemented under balanced and is recommended until reaching a stable pressure and oil column, any drilling of new well should be stopped. Referring to the results obtained from PLT analysis of selected wells, there is gas producing through all open-hole intervals and most have additional and unexpected free gas production. In addition to the change in GOC, the evidence of increased gas pressure confirms the existence of a highly connected fracture network throughout the reservoir which further accelerates gas coning growth and enlarges the gas cap of the field. The selected wells in this study have been chosen from different sections of the field and all show high amounts of gas production. Therefore, it can be concluded that the natural fractures have been developed throughout the reservoir. In order to better describe the reservoir, it is recommended to apply fracture indicator tools, seismic or coring from the selected wells.
　　11. Conclusions and Recommendations
　　In general, oil production of the north and south sectors of field A are better than the west section. Substantial pressure loss over the field history and the highly fractured and vuggy formation characteristic cause significant challenges for further field development and remediation of current production problems.
　　The reservoir rock permeability is low and to further improve the oil production, gas injection should be increased to a maximum capacity while allowing the oil in the matrix enough time to drain to the oil column. Conversely, further increases in oil production rate will reduce the oil column to a greater degree and due to large fracture network in the reservoir, the oil producing wells will connect to the enlarged gas cap, intensifying the problem of gas coning in the field. Generally, temperature and gradiomanometer profiling are the most effective tools for identifying gas entry and locating standing water levels. Ghost and Deft are very useful sensors in detecting gas and water, and calculating multi-phase flow rates. PLT results confirm the growth of gas coning in the field and all the selected wells show significant production of free gas throughout the producing intervals.
　　It is recommended to decrease oil production to compensate for the energy loss and implementation of a gas injection strategy must be continued to prevent further pressure decline. Also, any infill drilling must be done utilizing the under balance drilling technique. Applying fracture indicator tools, seismic and coring is highly recommended to have an accurate and better understanding of the intensity of the fracture network of the targeted field.
　　Acknowledgments
　　The authors would like to thank National Iranian Oil Company (NIOC-NISOC) for providing the required information to conduct this analysis on field A.
　　References
　　[1] T.O. Allen, A.P. Robert, Production operation, well completion, workover and stimulation, Oil & Gas Consultants International, Tulsa, Vol. 1, 2002.
　　[2] H.A. Al-Ali, S.P. Salamy, S.A. Haq, The challenges of detecting gas entries in horizontal well by using integrated production logging tool, in: 2000 SPE/Petroleum Society of CIM International Conference on Horizontal Well Technology, Calgary, Alberta, Canada, Nov. 6-8, 2000.
　　[3] M.S. Al-Ismail, M. Zeybek, O. Kelder, A.S. Al-Muthana, S. Mubarak, A. Abdulrahman, Advances in integrated horizontal production logging in open hole completions, in: 2005 SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, Oct. 9-12, 2005.
　　[4] A. Carnegie, Advanced horizontal well production logging—An Australian offshore example, in: 1999 SPE Asia Pacific Oil and Gas Conference and Exhibition, Jakarta, Indonesia, Apr. 20-22, 1999.
　　[5] C. Lenn, F.J. Kuchuk, J. Rounce, P. Hook, Horizontal well performance evaluation and fluid entry mechanism, in: 1998 SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, Sep. 27-30, 1998.
　　[6] D. Meunier, M.P. Tixier, J.L. Bonnet, The production combination tool—A new system for production monitoring, Journal of Petroleum Technology 23 (5)(1971) 603-613.
　　[7] D. Vu-Hoang, M. Faur, R. Marcus, J. Cadenhead, F. Besse, J. Haus, et al., A novel approach to production logging in multiphase horizontal wells, in: SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, Sep. 26-29, 2004.
　　[8] A. Shadravan, M. Nabaei, M. Amani, Dealing with the challenges of UBD implementation in the southern Iranian oilfields, in: SPE/IADC Middle East Drilling Technology Conference & Exhibition, Manama, Bahrain, Oct. 26-28, 2009.
　　[9] R.A. Nelson, Geologic Analysis of Naturally Fractured Reservoirs, Gulf Professional Publishing, BP Amoco, Boston, TX, 2001.