No abstract

The Ghawar field in Eastern Saudi Arabia contains the largest accumulation of carbonate reservoirs in the world. The majority of wells in the field produce from the Arab-D reservoir, an Upper Jurassic limestone sealed by anhydrite. Oil production from the field started approximately 55 years ago. Water injection started in the 1970's. Long before water injection was considered for the reservoir, the evaluation of wettability was considered essential. Our present day evaluation of Arab-D wettability takes into account a long historical record of wettability measurements and production history. The procedures, results and caveats of the original measurements have changed slightly but they also show a strong consistency fifty years later. Wettability indices obtained from initial tests, Amott, and USBM methods generally indicate neutral to slightly oil-wet character for cores processed and tested in a preserved state. Comparisons with restored state cores did not indicate major differences. Over the years fluids used in coring operations and core preservation have shown little impact on the observed results. Local variations in wettability indicating mixed wettability and oil-wet tendencies can be observed when tar is present in a significant amount and in areas high on structure. The combination of methods from advanced SEM observations, to qualitative contact angle measurements, to relative permeability results all point to a common wettability value. Introduction It has become an evident that about 50 % of the world proven oil reserves are contained in carbonate reservoirs.1 The wetting properties of carbonate reservoirs are fundamental to the understanding of fluid flow in all aspects of oil production, and can affect the production characteristics greatly during water flooding. So, knowledge of the preferential wettability of reservoir rock is of utmost importance to petroleum engineers and geologists. Due to this importance, many reviews of wettability and its effect on oil recovery have been conducted.2–4 Carbonate reservoirs are heterogeneous in nature due to the wide spectrum of environments in which carbonates are deposited and subsequent diagenetic alteration of the original rock fabric. These heterogeneities and effect of wettability on residual oil saturation, capillary pressure, electrical properties, relative permeability, and oil recovery encouraged many researchers to perform various studies to characterize and evaluate wettability of carbonate reservoirs. In the past, many engineers assumed that most reservoir rocks are water-wet. The reasons for this conviction are the work of Leverett5 and test methodology of determination of wettability after thoroughly cleaning cores that were likely to have been contaminated and exposed to air. The paper published by Treiber et.al.6 was the major breakthrough in showing that the large numbers of carbonate reservoirs are oil-wet. Consequently, various studies showed that the wettability of carbonate rocks is oil-wet, neutral or mixed.7–9 This paper provides detailed study and survey of wettability evaluation for Arab-D carbonate reservoir (Upper Jurassic), Saudi Arabia. The wettability results presented in this paper combine data obtained from various quantitative and qualitative methods over fifty years using preserved and restored core material. The studied areas are Uthmaniyah, Hawiyah, and Haradh. Arab-D Reservoir The Ghawar field in Eastern Saudi Arabia contains the largest accumulation of carbonate reservoirs in the world. The majority of wells in the field produce form the Arab-D reservoir, an Upper Jurassic limestone sealed by anhydrite. The Arab-D reservoir was discovered in 1948. Following further separate discoveries along the structure's main axis, five areas were quickly identified as parts of giant Ghawar oil field (Fig. 1): from north to south they are Ain Dar, Shedgum, Uthmaniyah, Hawiyah, and Haradh. At the Arab-D level, the field is NNE-trending composite anticline 230 km long and about 30 km wide.10 The largest oil accumulations occur in the lowest grainstone cycle of the Arab Formation, the Arab-D reservoir. The vertical oil column reaches a maximum of 1,300 ft. The oil-saturated interval extends about 250 ft below the anhydrite that separates the Arab-D reservoir from overlying Arab-C carbonate beds (Fig. 1).

The collection of core samples from a reservoir is essential for its proper characterization. Core data play very important roles in the following: log calibration, well treatment and performance prediction, reservoir modeling and simulation, and production and development planning.Coring is time-consuming, expensive and risky.It is difficult, if not impossible, to retrieve whole cores from horizontal and maximum reservoir contact (MRC) wells — which form the bulk of wells being drilled at present. As an alternative to coring, the use of cuttings has been proposed by many, as cuttings are always available from wells. Aside from some loss, contamination and depth-uncertainty (typically within 10 ft), cuttings show strong potential to represent all the formations encountered during the drilling of a well. This paper discusses the capabilities of various laboratory instruments to extract reliable density and porosity data from cuttings. The methodology used involves full characterization of a carbonate core plug using several laboratory tools followed by crushing the plug into cuttings of different mesh sizes. Samples of different mesh sizes were then scanned with a CT-scanner and bulk densities and porosities were calculated using advanced histogram-based analysis techniques. Results for individual cuttings sizes were compared against those for the whole plug. Additional data on the same cuttings were generated using micro-CT, Environmental Scanning Electron Microscope (ESEM), Nuclear Magnetic Resonance (NMR) spectrometer and APEX (Apparatus for Pore Examination) mercury porosimeter, which were also used for comparison. A set of procedures was developed for generating porosity data based on image processing of SEM and micro-CT generated images. The work helped establish the size of about 2.5 mm as the limit of detectability of medical-based CT-scanners to generate reliable density and porosity data from drilled cuttings. Comparison between the micro-CT and ESEM data showed the range of application for each tool in determining porosities from cuttings.Comparison between the pore size distributions generated by the NMR and APEX mercury porosity instruments and their visual comparisons with ESEM images provide important insights for the pore network modeling efforts. Introduction Subsurface geologic samples are usually of two types - cores and cuttings.Coring involves lowering of a special bit and core barrels into the well to retrieve rock samples.Since only about 60 ft of core is usually retrieved at a time, several trips may be required to collect a sufficient number of cores from the reservoir sections of a well.The requirements for drilling at a slower rate than usual (to have good recovery) and changing core barrels/bits with additional trips (trip time is not a productive time) make the coring job very expensive, especially in the case of offshore wells.Operational difficulties and risk of pipe sticking makes it almost impossible to cut cores from horizontal or maximum reservoir contact wells.Core recovery is also usually poor in weaker (unconsolidated) or more soluble rocks and some damage to the cores take place during transportation and storage.Wireline percussion coring provides an alternative to rotary coring operations but the success of such an operation is not well established and except for hard and consolidated formations, the recovery is generally very poor.

The M_1 nested bimodal pore system is prevalent in many large limestone oil reservoirs in Saudi Arabia. Within this pore system is contained a large portion of these fields' oil in place. Very low initial water saturation in these large structural relief carbonate reservoirs results in oil emplaced into pores controlled by M macropore throats and also into pores controlled by much smaller Type 1 micropore throats. Approximately, seventy-five percent of the M_1 oil portion is stored in the macropore system and about 25% is stored in the Type 1 micropore system. This prevalent M_1 petrophysical rock type (PRT) is an example of nested bimodal pore system consisting of an instance from the distribution of Macro possibilities (M porositon) and an instance from the Type 1 micro porositon distribution. The maximum pore-throat diameters of the Type 1 micro porositon are on the average 53 times smaller than the M macro porositon average maximums. M porosity average is 17% with a mean maximum pore-throat diameter of 58 microns. The Type 1 microporosity average is 5.6% with a mean maximum pore-throat diameter of 1.1 microns. Thus, common in Arab-D carbonate reservoir matrix is a bimodal pore network with a very large hydraulic contrast between a fine network of well-sorted tubular Type 1 micropore throats, connected and adjacent to a network of much larger diameter moderately-sorted M macropore throats. In a previous publication by Clerke, it was shown that the very small micropore throats' contribution to the total permeability is commonly below the resolution and reproducibility of the permeability measuring device when in the presence of many much larger pore throats. The micropore network is permeable if only at a small value. For the two phase flow occurring in a waterflood for oil recovery, the M_1 PRT requires an understanding of the two phase recovery processes in each pore subsystem considering capillarity in the combined pore network. This paper demonstrates that the Type 1 micropores are themselves a permeable network to water and to both oil and water when under waterflood. Hence for our carbonate reservoirs, "pores with throat diameters less than one micron when filled with oil in a bimodal M_1 pore system contribute to oil recovery through a time dependent spontaneous imbibition process and thereby contribute to oil recovery by waterflood." Further, it is demonstrated that the multimodality porositon classification proposed by Clerke are a form of dynamic rock type that classify the position and the type of internal pore level capillarity spatial gradients that affect ultimate oil recovery. New high-precision laboratory data has been obtained at very low phase pressure: water imbibition into oil saturated M_1 pore systems at near zero phase pressures (spontaneous imbibition) and dispersion of D2O into water filled M_1 pore systems. These pore systems can now be analyzed to obtain the magnitude, direct time dependence and scaling behavior of this important and previously overlooked portion of the total carbonate oil recovery by waterflood.

No abstract