Summary This paper covers the successful pilot field application of polymer gels for reservoir conformance improvement in the ongoing CO2 injection project at Bati Raman heavy-oil field in southeastern Turkey. Bati Raman is a naturally fractured carbonate reservoir in which the heterogeneities and the unfavorable mobility ratios between CO2 and the heavy oil cause inefficient sweep of the reservoir. These conditions prompted the pilot application of a conformance-improvement fracture-plugging (flowing) gel system in three wells in July 2002. Based on injection tests performed in the field, approximate treatment volumes were estimated to be on the order of 10,000 bbl for each well. Volumes actually pumped ranged from approximately 6,500 to 11,000 bbl. All three of the wells showed a gradual increase in injection pressure during treatment, indicating a decrease in injectivity index as treatment progressed. During one treatment, an offset producer experienced changes in fluid level consistent with rapid pressure transmission via the connecting fracture early in the treatment, with later loss of such communication. This behavior provides direct evidence of fracture plugging during treatment (Lane 2002). A mechanistic semianalytical model based on previously published laboratory work (Lane and Seright 2000) obtained a good match with the field data. The rate/pressure data were fed into the model, and effective fracture widths were backcalculated. Comparisons of results with the Formation MicroImager (FMI) log findings are explained. Gel-monitor well responses were scaled based on field data using a Fetkovich type decline-curve analysis. These studies enabled the incorporation of the effect of reservoir heterogeneities on the gel propagation radius so that future gel-treatment design parameters could be optimized. Pre and post-treatment CO2 injection pressures and the rates are as shown in Table 1. Sweep efficiency was increased as defined by produced oil/injected gas ratio. The 1-year average post-gel oil rate from 19 offset producers is 720 STB/D, as compared with apre-gel oil rate of 645 STB/D. The rate of increase from the treatments is thus 75 B/D, or 12%, which indicates a payout time of 12 months. Keeping this enlightened approach and seizing on the key concepts, four more CO2 injector wells were treated in 2004 to follow up on the encouraging results.

With 1.85 MMM bbl OOIP, the Bati Raman field is the largest oil field in Turkey. After its discovery in 1961, the field was put on stream for primary production until 1986. The recovery factor was only 2% after twenty five year production mainly due to low oil gravity. The well-known immiscible CO2 flooding project commenced in 1986, and the recovery factor reached 5% at the end of 2007. The recent steady decline in production entails the implementation of new development plans and this paper summarizes these efforts. After reviewing the performance of the current CO2 injection, short and long term development strategies were discussed. Short term plans include the continuation of the CO2 project in the areas where it is still viable. Some parts of the field are under WAG process. To improve the recovery in the short run by a better sweep (or displacement), a chemically augmented water injection process was proposed in those areas. Potential chemicals (surfactants and alkalis) were tested for wettability alteration and IFT reduction applying static (spontaneous) imbibition experiments. The best performing chemicals were determined for the field pilot after an economic analysis. In addition, the possibility of steam injection into the field was evaluated for the long run. Due to extreme heterogeneity and fractured structure, crestal steam injection that uses steam as heating rather than a displacement agent was proposed. An analytical study for the optimization of steam injection was provided. To determine the locations for the above listed processes, an extensive reservoir characterization study was performed using dynamic and very limited static -well- data. Using well recorded primary (1961–1986) and CO2 production data (1986–2007), fracture swarms were mapped. In this process, the changes in the initial production rate and GOR over different time periods were considered. The quickest decline in the initial rate and the lowest GOR areas correspond to highly fractured regions. Highly -vertically- fractured areas (typically the crest) were determined for potential steam injection. This analysis also helped detect high quality matrix areas as candidates for chemically augmented WAG.

No abstract

Raman field is a naturally fractured limestone reservoir located in southeastern Turkey. The field is a heavy oilfield with 18o API gravity oil and has strong aquifer pressure support. There are about 140 producers and daily oil production is about 6,000 bbl/d. However, average water cut has exceeded 90% in recent years because of the fractures communicating between the aquifer and the oil zone, which required some remedial treatment such as polymer gels to reduce the WOR. As it is well known, gel treatments have become a more convenient method as they can penetrate deep into the reservoir without a complete shutoff. As a pilot application, deep penetrating gels were used in seven wells. Wells with different behavior were selected as candidates on purpose, in order to see the effect of gel treatments on reducing WOR. The main purpose of the treatments was to increase oil recovery, with water shutoff was considered as a secondary benefit. Treatments were performed in September 2007. Very favorable results have been seen in the eight months following the treatments. Although gross production rates were generally reduced to get the most benefit from the gel treatments, still, pre-treatment oil rates from 0–14 bbl/d were increased to 6–91 bbl/d along with average water cut decrease from 97% to 80%, and even up to 40% in one well. These successful results will also provide the necessary clues for improving the design and field application of gel treatments which are to be extended on a field wide basis in the near future. Introduction Raman oil field, which is the first discovered oil field in Turkey, is located in Southeast Turkey (Figure 1). Raman is the second biggest oil field in Turkey with 600 MMbbls of original oil in place (OOIP). A total of 232 wells were drilled through May 2008. Eight are horizontal, 13 are deviated and 20 are dry holes. 6,000 bbls per day of oil is produced from 140 oil producers. 75,000 bbl per day of produced water is injected to aquifer by 13 water disposal wells. The main productive zone in the Raman field is the Mardin formation. However, the Garzan formation is also productive areally depending on the evolution of fractures and porosity in the east and north of the field. Because of low porosity and permeability the first 3–10 meters of the Mardin formation is non-productive. Under this non-productive zone there is, on average, a 20–25 meter productive zone in the Mardin formation. The porosity changes from 14–20 % in the Mardin formation productive zone and 10–25 % in the Garzan formation productive zone. The production mechanism is a strong water drive. Original reservoir pressure is about 1300 psi at the -200 m datum depth. Pressure is now 1100–1150 psi areally. The original WOC is -300 m. Some reservoir and fluid properties are given in Table 1. Because the reservoir is heavily fractured, in some wells the water-cut (wc) value increases rapidly due to fractures between the aquifer and the production zone. These types of wells were abandoned in a short time or are still being produced with very high wc values. Cumulative production from these wells is much lower than the average.

This paper covers the successful pilot field application of polymer gels for reservoir conformance improvement in the ongoing CO2 injection project at Bati Raman heavy-oil field in Southeastern Turkey. Bati Raman is a naturally fractured carbonate reservoir where the heterogeneities and the unfavorable mobility ratios between CO2 and the heavy oil cause inefficient sweep of the reservoir. These conditions prompted the pilot application of a conformance improvement fracture-plugging(flowing) gel system in three wells in July 2002. Based on injection tests performed in the field, approximate treatment volumes were estimated to be on the order of 10,000 bbls for each well. Volumes actually pumped ranged from ~6,500 – ~11,000 bbls. All three of the wells showed a gradual increase in injection pressure during treatment, indicating a decrease in injectivity index as treatment progressed. During one treatment an offset producer experienced changes in fluid level consistent with rapid pressure transmission via the connecting fracture early in the treatment, with later loss of such communication. This behavior provides direct evidence of fracture plugging during treatment. A mechanistic semi-analytical model based on previously published laboratory work (Seright, et.al) obtained a good match with the field data. The rate-pressure data were fed into the model and effective fracture widths were back calculated. Comparison of results with the FMI log findings are explained. Gel monitor wells responses were scaled based on field data using Fetkovich type DCA. These studies enabled the incorporation of the effect of reservoir heterogeneities on the gel propagation radius so that future gel treatment design parameters could be optimized. Pre and post-treatment CO2 injection pressures and the rates are shown in the following table:Table Sweep efficiency was increased as defined by produced oil/injected gas ratio. One year average post-gel oil rate from 19 offset producers is 720 stb/d as compared with pre-gel oil rate of 645 stb/d. Rate of increase from the treatments is thus 75 bbl/d or 12 %, which indicates a payout time of 12 months. Keeping this enlightened approach and seizing on the key concepts, 4 more CO2 injector wells are slotted for treatment in 2004 to follow up on the encouraging results. Introduction B. Raman field is the largest oilfield in Turkey, having an estimated 1.85 billion barrels of heavy oil reserves. Its reservoir rock is heterogeneous fractured, vugular limestone. The field was first placed on production in 1961 and had produced 1.5 % of its reserves by 1986, when TPAO began immiscible CO2 injection. Up to 2003, 5% of the reserves could be produced, which is still an unexpected low value. Production rate has declined drastically since 2000. TPAO is seeking the most applicable methods to impede or reverse the decline. Polymer gel treatments were an obvious EOR method to increase CO2 sweep efficiency. The behaviour type of the field after the CO2injection application had started was analyzed by using oil rate, injection pressures, gas-oil ratio (GOR) and gas utilization factor history, and 3 different behaviour types were observed (Figure 1). The years 1986–1993 are defined as the fill-up period where the injected gas fills the fractures and vugs.