Search

The article presents a new method of determining the residual water content and water saturation of the Bazhenov rocks formation (unconventional reservoir), which is contingent on the synchronous thermal analysis integrated with gas FT-IR spectroscopy and mass spectroscopy. The studies were executed on extensive factual core material. The combination of thermal and spectrometric methods for the identification of gases which are released during heating of core samples facilitated to analyze the dynamics of water release and proposed methods of its separation accordingly by the degrees of connectivity.

The application of coupled horizontal wells and the multistage fracturing technology to shale formations makes it viable to develop such hydrocarbon resources. In this paper, we focus on the water saturation distribution within hydraulic fractures and the uncertainty of water saturation distribution of the matrix around the hydraulic fractures resulted from fluid injection during a fracking process. Arp's decline curve is introduced to depict initial water saturation distribution within the hydraulic fractures. Through modifying the decline index, it is easy to obtain a range of initial water saturation decline patterns which will lead to different water production profiles. Typical decline patterns such as linear, exponential, harmonic, and hyperbolic declines are used. As for the initial water saturation distribution of the matrix around the hydraulic fractures, the Stimulated Reservoir Volume (SRV) concept is implemented to identify the matrix region. Through reservoir simulations, we find that the initial water saturation within the hydraulic fractures mainly contributes to the early water production. Various saturation distribution models result in about 25% differences in the total water production. On the other hand, a leak-off effect, which increases the initial water saturation of the matrix around the hydraulic fractures, contributes to the long term water production. This study provides insights about the uncertainty of water production during the development of shale gas reservoirs and guidelines for the history matching of water production rates.

Low-salinity waterflooding (LSW) is an enhanced technique that improves oil recovery by lowering and optimizing the salinity of injected water. Previous research demonstrated that the initial water saturation and oil aging during LSW affect oil recovery. This work aims to validate and quantify these effects. This laboratory study investigated the effect of injected brine, initial water saturation, and crude oil aging on relative permeability and low-salinity waterflooding performance in sandstone reservoirs. A set of comprehensive coreflood tests have been conducted using 3″ Bandera sandstone cores to examine the effect of initial water saturation and injection brine on the relative permeability curves. Additionally, six coreflood experiments were conducted to examine the effect of the connate water salinity variation on the relative permeability end points. Four spontaneous imbibition experiments were performed using 6 in. length Buff Berea sandstone samples to investigate the effect of initial water saturation and aging on the performance outcome of low-salinity waterflooding. Two spontaneous imbibitions were performed using 20 in. length core samples. The objective of these two experiments was to study the effect of initial water saturation distribution across the core on the oil production using X-ray computed tomography. Results obtained from coreflood and spontaneous imbibition tests suggest injected brine and initial water saturation have predominant influence on the oil recovery performance. As the initial water saturation increased from 23 to 41%, the oil recovery increased from 43.8 to 48.4% of OOIP using low-salinity waterflooding. Comparison between the high-salinity relative permeability and low-salinity water relative permeability showed that the end-point water relative permeability slightly decreased for the cores after using low-salinity brine as injected brine compared to the end-point water relative permeability after flooding with high-salinity brine for Bandera sandstone. The end-point relative permeability to water increased from 0.07 to 0.39 when the reservoir connate water salinity increased from 4,633 to 174,156 ppm for injected brine salinity of 500 and 5,000 ppm, respectively for Buff Berea sandstone. The un-aged core plugs with higher initial water saturation provided oil recovery higher than that with lower initial water saturation by 5% of OOIP for the low-salinity waterflooding.

Some difficulties are frequently associated with the integration of seismic and flow simulation datasets, especially related with the low vertical resolution of the seismic data and the uncertain estimations in the areas between the wells. One challenge is then the integration of both datasets in different scales, in order to take advantage of their characteristics. The present study proposes a redistribution of the reservoir saturation estimated with 4D seismic inversion methods, combining the information provided by the flow simulation in order to improve the quality of the estimations and their resolution. The methodology comprehends a saturation redistribution algorithm that is applied to each reservoir block combining the information of two saturation maps: one derived from the 4D seismic data and the other estimated by the flow simulator. The final saturation estimation follows the vertical distribution given by the simulation data but keep the average behavior observed in the 4D seismic. In order to have a better control of the results, the methodology is applied to a synthetic dataset that includes a reservoir model with different grid resolutions: geomodel, simulation model and seismic model. Two different case studies present the main results of the proposed methodology to improve the saturation predictions. The first case study represents an ideal solution being assumed that the base model (simulation model) is very similar to the reference model (geomodel that represents the true answer), remaining a few differences due to the different scales. In the second case study, the base model is selected between multiple realizations during the uncertainty reduction process. The results shows that is possible to improve the resolution of the saturation variation maps computed from 4D seismic data allowing the identification of new fine scale heterogeneities and providing better estimations of the saturation changes due to production. This methodology can also give additional clues in future history matching procedures through the identification of critical regions. With the continuous calibration of the models during a history matching process the results obtained with the redistribution method tend to improve, approaching the results obtained in the first case study. The proposed redistribution method combines the best characteristics of the seismic and simulation data. It includes the higher sensibility of the seismic data to identify the areal distribution of the main anomalies and inserts the higher sensibility of the simulation data regarding the identification of vertical water flow trends due to gravitational effects. Thus, the procedure introduces, in the maps provided by 4D seismic, new information regarding the injection/production patterns that become more and more reliable as we approach the wells.

Initial water saturation (Swi) in a gas reservoir is an important parameter for inplace resource (GIIP) and ultimate reserve estimation, which in turn impacts the economic decision. Swi estimation in a low resistivity deep-water clastic reservoir is more challenging because limited well data in less number of (costly) appraisal wells in a large area. Conventionally, Swi is computed from open hole logs and validated with core plug data to reduce the range of uncertainty in estimation. But this standard methodology fails when resistivity log derived Swi shows a variance with the saturation measured from the core plugs and increases the range of uncertainty. Sometimes, log derived Swi shows high water saturation in the gas bearing interval due to low anomalous resistivity which is beyond correction. Saturation height modelling is an age old solution for this sort of problem and the same was attempted first to estimate Swi. But single saturation height function does not represent all the rock types of the reservoir and replicate log derived Swe curve. Petrophysical rock typing was carried out using porosity and permeability from the core plug in the first step and then by using the concept of flow zone index (FZI) and rock quality index (RQI). FZI-RQI based rock types were able to characterize the reservoir in a better way in 3D-Geologic model and also able to separate the different behaviours of Capillary pressure curves. Two saturation height functions were made after characterizing good rock and poor rock type, which were tested in the dynamic model for flow simulation and recovery factor estimation successfully. This innovative concept of FZI-RQI based petrophysical rock typing (PRT) and saturation height modelling finally added value to recreate the Swi curve against the low resistivity pay interval and to reduce the range of uncertainty in Sw estimation.

Characterizing hydrocarbon distribution is one of the most critical and challenging steps, because the initial hydrocarbon saturation determines the original hydrocarbons in place and also subsequent reservoir numerical simulation during the field development stage including history match, production forecast, sensitivity analysis, etc. Accurate hydrocarbon distribution knowledge is vital for optimum reservoir characterization, hydrocarbon exploration and production. For an unconventional reservoir, the distribution of hydrocarbon and water is abnormal and complex compared with a conventional reservoir. The characterization and representation of the reservoir requires a multi-discipline team incorporating geology, geophysics, petrophysics, core analysis, reservoir engineering, etc. to enhance the understanding of reservoir heterogeneity and the target area selected for new development wells during the development phase.

Accurate estimation of permeability is essential in reservoir characterization and in determining fluid flow in porous media to optimize the production of a field. Some of the available permeability prediction techniques — e.g., Porosity-Permeability transforms and more recently artificial intelligence and neural networks — are encouraging but still show only moderate to good match to core data. This could be due to limitation to homogenous media while the knowledge about geology and heterogeneity is indirectly related or absent. The use of geological information from core descriptions, e.g., Lithofacies, which includes diagenetic information, show a link to permeability when categorized into rock types exposed to similar depositional environments. The objective of this paper is to develop a robust combined workflow integrating geology and petrophysics and wireline logs in an extremely heterogeneous carbonate reservoir to accurately predict permeability. Permeability prediction is carried out using pattern recognition algorithm called multi-resolution graph-based clustering (MRGC). We will bench mark the prediction results with hard data from core and well test analysis. As a result, we show how much better improvements are achieved in permeability prediction when geology is integrated within the analysis. Finally, we use the predicted permeability as an input parameter in J-function and correct for uncertainties in saturation calculation produced by wireline logs using the classical Archie equation. In conclusion, a high level of confidence in hydrocarbon volumes estimation is reached when robust permeability and saturation height functions are estimated, in conjunction with important geological details that are petrophysically meaningful.

Good estimation of water saturation is necessary for successful estimation of reservoir properties as it minimizes the error in initial oil-in-place calculations. Electrical measurements on core plugs have been used to predict water saturation by analyzing the parameters of Archie's equation. Presently, several methods such as conventional, CAPE (1, m, n), CAPE (a, m, n) and 3D methods have been used to analyze the parameters. However, the accuracy of these methods has become inadequate for optimal estimations. Recent successful applications of Artificial Intelligence (AI) techniques in petroleum engineering have confirmed the capability of AI to handle such non-linear and complex industrial problems. Interestingly, the new paradigm of employing AI for water saturation estimation has not been sufficiently addressed in the petroleum engineering technology application literature. This study focuses on the use of the predictive capabilities of Artificial Neural Networks and the Fuzzy Inference Engine. 378 data samples obtained from the laboratory measurements of electrical properties of 41 core plugs taken from Carbonate reservoir rocks in the Middle East were used for the implementation of the proposed techniques. Using the popular stratified data sampling method, 70% of the data points were used to train the AI models while the remaining 30% were used for validation and testing. The comparative results showed that the AI models performed better with higher accuracy and lower errors than those obtained with the current methods. Based on this result, we conclude that Fuzzy Logic exhibits a robust predictive capability for the estimation of water saturation from electrical measurements by providing a good match with the core values and giving better error distribution than the other AI and current methods.

We present an inversion-based method for petrophysical interpretation in a vertical or slightly deviated well that integrates flexural wave dispersion data from a sonic dipole tool, and resisitivity data from a tri-axial array-induction tool. At each logging depth, the inversion solves for radial profiles of water saturation and porosity extending several feet from the wellbore. Radial changes in water saturation represent the effect of mud-filtrate invasion, while radial changes in porosity represent the effect of mud-fines migration or mechanical breakdown of rocks. Petrophysical transforms specific to the formation lithology are used to map water saturation and porosity to elastic velocities and resistivity, which constitute the input to electromagnetic and sonic tool response simulators. Radial profiles of formation properties are estimated by minimizing the mismatch between the measured and simulated data using an inversion workflow. The workflow is designed taking into account complementary sensitivities of the flexural and induction measurements to improve inversion robustness. We benchmark the inversion workflow for two fluid-lithology configurations relevant to deepwater offshore environments. The first configuration corresponds to gas-bearing sandstone drilled with oil-base mud, where the effective elastic properties are governed by the Gassmann fluid substitution petroelastic transform. The second configuration corresponds to oil-bearing carbonate drilled with oil-base mud, governed by the Self-Consistent Effective Medium (SCEM) petroelastic transform. For various near-wellbore invasion and mechanical alteration scenarios, the inversion reconstructs the radial formation properties extending up to five feet from the wellbore. The results demonstrate the efficiency, robustness, and accuracy of the inversion for application to challenging formation evaluation scenarios.

Complex variations in pore geometrical attributes, which in turn, defines the existence of distinct zones or hydraulic unit within a mappable geological facie or “rock type (RRT)' which have similar dynamic behavior is “masked” when “rock types (RRT's)” in reservoir simulators are assumed to be different geological facies of similar dynamic behavior. To effectively and accurately distribute water saturation (Swi) in the cells of dynamic model, capillary pressure (Pc) vs.Swi and relative permeability curves must be assigned such that they honor the pore system structure as well as the saturation from well logs. Therefore it might not be enough to assign one SHF for a given geological RRT. These poses a challenge, for example, in Niger Delta sandstone reservoir comprisingmainly alternating deltaic sandstone with shale suggesting two or three consistent RRT's with wide variability in Swi. The proposed approach tends to develop a correlation that will consistently replicate the behavioral relationship between reservoir quality index ((RQI=0.0314K/∅ϵ)), effective water saturation S* = (Sw − Swirr)/(1 − Swirr) and Height above free water level(HAFWL) of both routine core and log data. This correlation is used to develop drainage log derived Pc and relative permeability curves for each class of RRT. Rock classification employed in this work was based mainly on irreducible water saturation(Swirr) classes. The SATNUM keyword in ECLIPSE was used to assign each SHF and relative permeability curve to the grid cells of the dynamic model based on the rock classification. The developed SHF yielded an excellent match, both vertically and laterally, with log derived Swi and within 1% difference between static and dynamic hydrocarbon-in- place for some Niger-delta reservoirs studied. Also, a plot of Swi generated from the newly developed SHF at well points on the Y-axis and those from well logs on the X-axis yielded a 45° straight line through the origin with an R-square of 0.9935. The dynamic model stability and running performance was greatly improved using the developed SHF.

Modeling of enhanced / improved oil recovery processes that takes into account mass transfer between phases depends on the correct prediction of thermodynamic properties and composition of phases in equilibrium. In a steam flood process, the mutual water / hydrocarbons solubility is a function of the temperature, pressure and fluid composition of the system and may not be negligible at flooding conditions. The most used equations of state (EoS) in the petroleum industry fail to accurately correlate saturation properties of polar substances that self-associate through hydrogen bonding and as a consequence, do not calculate the distribution of the components among equilibrium phases precisely. In this paper, we present the development of the Association Peng-Robinson and Association Soave-Redlich-Kwong equations of state. The proposed equations of state are composed of two parts, one physical (the original cubic equation of state model) and one chemical (an empirical chemical reaction term which accounts for the self-association of a component) and can be used to model systems in equilibrium that contain one associating component such as water or alcohol. In the extended equations of state which includes the self-association chemical reaction, the degree of the molar volume polynomial is increased from its normal value of three to six but, in general, there are only three positive roots. The chemical part of the extended EoS includes three parameters that can be adjusted to match data, the entropy and enthalpy of the association chemical reaction and one free parameter. The fugacity calculations for the extended EoS can be split into physical and chemical parts, where the physical part has exactly the same form as the original equation of state fugacities. In order to estimate the new parameters of the proposed equations of state, the differences between the experimental and calculated saturation pressure and saturated liquid molar volume of pure water are minimized using a Particle Swarm Optimization (PSO) algorithm. The equations of state presented here enhance the original ones through the addition of a chemical part to deal with self-associating polar components. The average relative deviation between the experimental and calculated saturated data using the association forms of the Peng-Robinson and Soave-Redlich-Kwong equations of state are smaller than those obtained from the original models. The chemical reaction approach is robust and improves the prediction of thermodynamic equilibrium properties of self-associating pure components by adding only three adjustable parameters, where two of them have a clear physical meaning (self-associating reaction enthalpy and entropy).

Roseneath and Murteree organic shales in Nappamerrie trough are part of most prolific unconventional gas reservoirs in Cooper Basin, South Australia. The cumulative effect of rocks, minerals, fluids and organic content in these shale gas reservoirs has been investigated and incorporated using Archie's equation and other resistivity models such as Indonesian, Siamandoux and total shale model for water saturation evaluation. In this study, wireline logs have been calibrated with QEMSCAN and FIB/SEM images. We have found that the shale content in Murteree formation is around 50% using QEMSCAN analysis. Also, it is found that using Steiber formula, the shale volume is close to 50%. Therefore, the Steiber formula is selected as the proper correlation to estimate shale content for Murteree formation. Furthermore, the free porosity of Roseneath and Murteree shale was measured to be 2% based on QEMSCAN and CT scanning. Then, Using Wyllie formula for sonic porosity with Hilchie's correction factor, the porosity was also determined close to 2%. Thus, Wyllie formula with Hilchie's correction factor was also found to be the appropriate correlation for porosity calculation in Roseneath and Murteree shale. For water saturation, both Simandoux and Indonesian correlations were shown to give reasonable results. However, it is found that if free porosity is more than 9%, both total shale and Archie models can also give acceptable results. The sensitivity of true resistivity in water saturation calculation was investigated as well. Furthermore, the evaluation of Archie's parameters (cementation exponent, m, tortuosity factor, a, and saturation exponent, n) were examined. Also, FIB/SEM images show that brine and organic matters are both attached to the clay particles which have caused a modified true resistivity for the shale.

Mixed wettability is a common concern in heterogeneous reservoirs. The problem from a practical point of view is the general lack of water saturation exponent (n) laboratory values in rocks with mixed wettability. To contribute to the solution of this concern, we utilize electromagnetic mixing rules to develop a new petrophysical model capable of estimating the water saturation exponent (n) and water saturation (Sw) in heterogeneous reservoirs with mixed wettability. The reservoirs are represented by dual porosity (matrix and fractures or matrix and isolated porosity) and triple porosity (matrix, fractures, and isolated porosity) models. Results from the models developed in this work compare well with core data. We have used the same electromagnetic mixing rules in the past for successful determination of the porosity exponent (m) in heterogeneous reservoirs. The addition in this paper offers a means for calculating the water saturation exponent (n) and makes the electromagnetic mixing rules a powerful tool for complete formation evaluation of heterogeneous reservoirs with complex pore structures and mixed wettability. It is concluded that the new petrophysical models developed in this study provide a significant contribution to the determination of water saturation exponent (n) and water saturation (Sw) in rocks with mixed wettability and hence lead to improved values of hydrocarbon in place estimates.

Porosity, permeability and water saturation cannot be effectively estimated from conventional logs by using the classical methods in tight sandstone reservoirs due to the complicated mineral components, pore structure and strong heterogeneity that caused by the diagenesis in the Chang 8 Formation of northwest Ordos basin, In this paper, based on the difference of diagenetic facies, effective formations are classified into three types, the relationships between conventional density log and core derived porosity, core derived porosity and permeability are established in every type of formation. In the meanwhile, the laboratory resistivity experimental results are also classified based on the diagenetic facies, and the cross plots of porosity and formation factor, water saturation and resistivity index are drawn to obtain the rock resistivity parameters, separately. These results illustrate that for formations with different diagenetic facies, the matrix densities are diverse, the relationships between core derived porosity and permeability are also different, and they contain different rock resistivity parameters. By using the diagenetic facies classification method and chart proposed by Xiao (2012), the effective Chang 8 Formation is classified into three types, and the corresponding models are applied to calculate porosity, permeability and water saturation. Comparisons of estimated porosity, permeability and water saturation by using the proposed techniques and the core derived results illustrate that the proposed techniques are valueable in tight sandstone reservoir parameters evaluation.

Accurate water saturation estimation in thin carbonate reservoirs is one of the major formation evaluation challenges in horizontal wells. This is mainly due to the strong shoulder bed effect of thick dense sections on the resistivity logs recorded in thin porous reservoir layers, particularly in horizontal wells. The performed work implies correction of the high resistivity log data acquired while drilling horizontal well sections by means of resistivity modeling. This has been achieved by building a forward model for the resistivity data acquired in the deviated section of the overlying reservoirs in water injector wells, where the porous section is thick and the shoulder bed effect is minimal. The resistivity model construction is based on the relationship of the deep and shallow Laterolog resistivity to the neutron porosity log responses measured in the thick reservoir sections. The established model equations are used to construct deep and shallow resistivity logs in the horizontal sections of thin reservoirs by resistivity inversion using the neutron porosity log data.Several well data sets have been tested and the results were complimented by test data, formation tester fluid sample analysis, Dean Stark measurements and core data. The results obtained confirmed that the resistivity inversion technique is applicable to compute water saturation in both the flushed zone (Sxo) and the un-invaded zone (Sw) for horizontal well sections drilled across carbonate reservoirs. The application of the inversion process allows generating resistivity log responses corrected for the tight shoulder bed effects, and providing more accurate water saturation estimation.

One major problem in the numerical simulation of the water-alternating-gas (WAG) process is the uncertainty associated with the changes in three-phase relative permeability (kr) values in the sequence of drainage and imbibition cycles known as cyclic hysteresis. In this work we have investigated the effect of cyclic injection of water and gas on three-phase kr of water, gas and oil under both water-wet and mixed-wet conditions. Two different sandstone cores with one order of magnitude difference in absolute permeability (65 mD and 1000 mD) have been used in the coreflood experiments.In the base study, WAG injection started with water flooding (I) followed by gas injection (D) and this cyclic injection was repeated three times (IDIDID). To investigate the effect of immobile water saturation, the IDIDID injection scenario was repeated with lower immobile water saturation. To study the effect of saturation history, another WAG experiment was performed which started with the primary gas flooding (D) followed by water injection (I) and this cyclic injection were repeated as DIDIDIDI. For all of the experiments, three-phase relative permeabilities were obtained analytically from the coreflood data using an extension of Buckley-Leveret formula to three-phase flow.The results show irreversible kr hysteresis loops for water and gas and saturation history dependency for oil in processes involving cyclic injection under three-phase flow conditions. Generally, krg reduced during successive cycles for all WAG experiments. For the IDIDID injection sequence and in water-wet system, it was found that krg shows larger hysteresis effect for the case with higher immobile water saturation. krw hysteresis was larger in the DIDIDIDI injection scenario compared to the IDIDID case. krg and krw hysteresis effects were larger for the 1000 mD core compared to the low permeability (65 mD) sample. The most important hysteresis effect for both krg and krw was observed for the transition from two-phase to three-phase. As WAG cycles were repeated, krg and krw hysteresis reduced became minimal after the 2nd water injection in IDIDID and 2nd gas injection in DIDIDIDI injection sequences.Comparison of our experimental observations with predication of existing models highlights some serious shortcomings of reservoir simulators in simulation of oil recovery processes involving three-phase flow and flow reversal. The paper also offers some novel insights into cyclic hysteresis behaviour and offers explanations based of our understanding of the pore-scale and core-scale displacement mechanisms involved in three-phase flow and WAG injection.

Proper petrophysical evaluation of carbonate formations, offshore Abu Dhabi is a difficult process considering the number of challenges to resolve. Lithology mainly consists of a combination of dolomite and calcite but also contains anhydrite which must be accounted for to get an accurate porosity. Resistivity measurements are affected by invasion and by the very high shoulder bed resistivity so computing formation resistivity can only be done through resistivity modeling and inversion, and once formation porosity and resistivity are properly computed, it is possible to compute an accurate formation saturation only if the Archie parameters cementation and saturation exponents m and n are properly defined. We will show how to resolve these challenges by acquiring and integrating in an advanced workflow a modern suite of logs including density-neutron-resistivity-gamma ray along with a multi-frequency dielectric measurement. We will also show how to confirm the formation saturation in selected zone in an Archie independent manner by combining the dielectric log and pump-out formation tester. The integration of the log data with core analysis results in a very comprehensive petrophysical evaluation of the formations encountered.

Within an oil reservoir the water saturation height functions can vary strongly, especially for carbonate rocks. These variations can be significant and difficult to estimate. The amount of hydrocarbons in place, the prediction of recoverable oil, the recovery process and the future plans of developing such reservoirs depend on many factors, one of which is the accurate modeling of water saturation. The Khafji carbonate reservoir is a heterogeneous reservoir with two different types of oil: light oil in the top of the reservoir and heavy oil in the bottom of the reservoir. The challenge of water saturation modeling is primarily in the heavy oil zone, where conventional height function techniques produces poor match against measured water saturation logs. Alternative methods were utilized in order to obtain good match in both light oil and heavy oil columns. A workflow has been created in order to overcome these challenges. Laboratory derived capillary pressure curves were used to establish water saturation height relationships as a function of rock type (RT). Additionally, a Flow Zone Indicators (FZI) analysis was used as a basis for rock typing. Then a J-function derived from capillary pressure data for each rock type or hydraulic flow unit (HFU) was used to generate saturation height function for each RT. The generated saturation undergone via several iterations to match the large span of openhole electric water saturation logs above the free-water level (FWL). The saturation profile generated by this workflow shows a good match to the measured Sw electric logs, and the calculated fluid volumes are in agreement with company's approved reserves estimation.

Water saturation (Sw) estimates are of prime importance for reserves estimation, reservoir development, and reservoir management. Traditionally, Sw has been derived from the Archie equation with formation resistivity or thermal neutron capture cross-section (Sigma, Σ in stand-alone mode. Either the resistivity or Σ approach requires good knowledge of formation water salinity which can be difficult, particularly in instances of unknown and/or mixed salinities. When resistivity and Σ logs are unaffected by fluid invasion, one can simultaneously compute Sw and salinity from the two measurements as has been done recently using wireline logs acquired in flowing wells. This paper addresses 2 drawbacks of wireline resistivity-sigma technique: first, it is not always practical to flow the well, and second is the concern about the Σ shallow depth of investigation (DOI). The first drawback might be answered by using LWD to mitigate fluid invasion issues instead of logging a live well. The second drawback is addressed by taking advantage of the newly-developed LWD multiple depths of investigation (MDOI) Σ measurements (shallow Σ, medium Σ, and deep Σ) to determine the presence of invasion. If invasion has occurred, via a new MDOI inversion process analogous to the resistivity step-profile inversion, we estimate both invaded and virgin zone Σ for use in petrophysical applications. We analyzed LWD MDOI resistivity and sigma logs acquired in an oil-producing reservoir drilled with a low-invasion water-based mud for a remaining Oil Saturation (ROS) study. The coring operation has allowed some invasion to take place, which can clearly be seen from the separation of MDOI sigma curves. The inverted (true) Sigma allows meaningful Sw comparisons between LWD, wireline and core analysis. The errors on estimating water saturation and salinity using shallow Σ, medium Σ, deep Σ, and true Σ are illustrated to highlight the need to correct for invasion before doing quantitative formation analysis with Σ. The results indicate the viability of using LWD measurements to determine ROS in flooded reservoirs with mixed/unknown salinities.

Despite the great effort made to characterize reservoir rock, proper techniques in grouping rock type, in predicting permeability, and also in estimating water saturation from reservoirs that exhibit high variability in permeability-porosity relationship are still debatable today. In addition, core sampling for both routine and special core analysis may be biased in spite of the wide number of existing rock types, mostly in the case of heterogeneous rock reservoirs. The main objective of this work was to develop a new mathematical expression for water saturation prediction, in the case study Urdaneta-01 heavy oil reservoir, in Maracaibo Lake basin, Venezuela, based on an alternative approach of traditional Leverett J-function. This new equation was developed by introducing two additional power-fit coefficient as function of water saturation (for specific capillary pressure values), Leverett J-function slope (from all values of capillary pressures test), and height above free water level (FWL). In addition, an automatic Rock Type Index was used to accurately estimate reservoir permeability. A reliable water saturation profile according to rock type was obtained from the new water saturation approach. At least 3 to 5% difference between water saturation from log-based traditional techniques and that of the alternative approach (new expression for Sw) was obtained, mostly in heterogeneous intervals. Also it was determined that the innovative rock type indicator provides a better permeability estimation and rock description because of the high definition of rock type obtained from this new technique.

A 3D integrated saturation model was built for the Sierras Blancas Formation of the Neuquén Basin, Argentina. The saturation model was based on core, logs and seismic data. History match of reservoir pressure and well productivities were taken into account to accurately determine the gas in place and productive reservoir boundaries, specifically using 3D seismic water saturation in the gas condensate formation. The Sierras Blancas Formation is an eolian deposit. In tight, wet and diagenetically altered regions, the seismic inversion porosity and acoustic impedance based models were not adequate to describe the gas in place distribution. Further, the effective gas permeabilities in the tighter part of the reservoir are a strong function of the initial water saturation as evidenced by fewer condensate and water blocking problems of horizontal wells that navigated through low water saturation, high permeability regions. Any relationship between seismic impedance and porosity correlation degraded in areas affected by secondary diagenetic processes therefore necessitating the use of a saturation parameter. 30 vertical wells that had DT curves were selected based on their production and spatial location in order to establish a correlation between log saturation and seismic attributes. Seismic saturation cubes were generated by multiattribute seismic analysis and resampled into the simulation scale model. Log saturations were then co-kriged with the 3D seismic saturation. Water saturations obtained from the initialized simulation scale model were compared with the 2D saturation logs, the 3D seismic and the geological model scales. An objective function was defined to match the 3D seismic water saturation with the initialized simulation model water saturation. Model parameters were iterated until a satisfactory match with the initialized simulation model was obtained. By focusing the saturation match at the initialization stage, seismic, geological, petrophysical and SCAL models were ensured to be consistent prior to the full history match. Well history matching was consequently achieved much more simply and quickly. This paper presents a new detailed methodology of 3D pseudo-seismic water saturation generation, modeling and simulation used to accurately define OGIP, the productive boundaries of the reservoir, and to design trajectories for new horizontal wells.

Near-miscible gas injection represents a number of processes of great importance to reservoir engineers including hydrocarbon gas injection and CO2 flood. Very little experimental data is available in the literature on displacements involving very low-IFT (interfacial tension). In this paper, we present the results of a series of two-phase and three-phase gas injection (drainage) and oil injection (imbibition) core flood experiments for an gas/oil system at near-miscible (IFT= 0.04 mN.m‒1) conditions. Two different cores; a high-permeability (1000 mD) and a lower permeability (65 mD) core were used in the experiments and both water-wet and mixed-wet conditions were examined. The results show that despite a very low gas-oil IFT, there is significant hysteresis between the imbibition and drainage oil and gas relative permeabilities (kr) curves in the 65mD core. Hysteresis was less for 1000mD core (compared to the 65 mD core) but it still could not be ignored. Near-miscible kr hysteresis was significant for both water-wet and mixed-wet systems. Presence of immobile water in the water-wet cores improved oil relative permeabilities but reduced gas relative permeabilities in both imbibition and drainage directions. As a result, oil recovery for gas injection experiments improved when the rock contained immobile water. Both oil and gas relative permeabilities reduced when the rock wettability was altered to mixed wet from water wet and as a result, oil recovery by gas injection in the mixed-wet rock was less than that obtained under water-wet conditions. We offer explanations for these observations based on our understanding of the pore-scale interactions and mechanisms, the distribution of fluid phases and their spreading bahaviour. The results help us better understand the impact of some of the important parameters pertinent to kr and its hysteresis especially in very low IFT gas-oil systems and mixed-wet rocks. Understanding these effects and behavior is important for improved prediction of the performance of gas injection and water-alternating gas (WAG) injection in oil reservoirs.

Sequential waterflooding refers to cycles of flooding for which initial water saturation is re-established after a waterflood by flow of crude oil followed by further waterflooding. In previously reported examples of sequential waterflooding, cores were neither cleaned nor re-aged at high crude oil saturation between floods. Numerous core floods with different rock types showed significant decrease in residual oil saturation from one flood to the next (Loahardjo et al. 2010a). Systematic decrease in residual oil saturation by sequential waterflooding was confirmed by nuclear magnetic resonance imaging measurements of in-situ saturations (Loahardjo et al. 2010b). In this work, sequential waterflooding has been demonstrated for outcrop Berea sandstones of medium and low permeability: both showed reduction in residual oil saturation with each sequential flood. The tests have been extended to include aging periods at either residual oil saturation or initial water saturation. Aging at residual oil saturation (high water saturation) resulted in an increase in oil recovery for the subsequent flood, with larger increases observed for extended aging times. After an extended period of aging at initial water saturation (high oil saturation), decreased oil recovery was observed. The variations in oil recovery are consistent with changes in wettability that depend on oil saturation during displacement and subsequent aging conditions.

A thorough understanding of foam fundamentals is crucial to the optimum design of foams for improved/enhanced oil recovery. This study, for the first time, presents anomalous foam fractional flow solutions which deviate significantly from the conventional solutions at high injection foam qualities, by comparig Method of Characteristics and mechanistic bubble-population-balance simulations. The results from modeling and simulations based on coreflood experiments revealed that (1) there exist three regions: region "A" with relatively wet (or high fw) injection conditions where the solutions are consistent with the conventional fractional flow theory; region "C" with very dry (or low fw) injection conditions where the solutions deviate significantly; and region "B" in between which has a negative dfw/dSw slope showing physically unstable solutions; (2) for dry injection conditions in region "C", the solutions require a constant state (IJ) between initial (I) and injection (J) conditions, forcing a shock from I to IJ by intersecting fractional flow curves, followed by spreading waves or another shock to reach from IJ to J; and (3) the location of IJ in fw vs. Sw domain moves to the left (or toward lower Sw) as the total injection velocity increases for both weak and strong foams until it reaches limiting water saturation. Even though foams at high injection quality are popular for mobility control associating a minimal amount of surfactant solutions, foam behaviors at dry conditions have not been thoroughly investigated and understood. The outcome of this study is believe to be helpful to successful planning of foam field I/EOR applications.

The Archie’s equation lost its role in tight gas sands due to the complicated pore structure and strong heterogeneity. It’s a challenge to determine the input parameters in the Archie’s equation. In this paper, 36 core samples, which were drilled from tight gas sands in China, are chosen for resistivity and NMR laboratory measurements. Based on the experimental study of these core samples, the influence factors to electrical properties are concluded to reservoir porosity and the proportion of small pore components. When the porosity is higher than 25%, the relationship between the porosity and the formation factor illustrares a power function, this is coherent with the classical Archie’s equation. When the porosity is low, the statistic line of the porosity and the formation factor bend to the left. The relationship between the porosity and the formation factor is not a simple power function, the parameter of m is various and relevant to porosity. The relationship between the water saturation and the resistivity index is divergent, the saturation exponent n varies from 1.63 to 3.48. After analyzing the corresponding NMR laboratory measurement for the same core samples, an observation can be found that the saturation exponent is relevant to the proportion of small pore components. When core samples are dominant by the small core components, the corresponding saturation exponent is high, vice versa. To estimate reservoir initial water saturation accurately, the pore structure information must be considered.

Abstract Artificial Intelligence techniques have been used in petroleum engineering to predict various reservoir properties such as porosity, permeability, water saturation, lithofacie and wellbore stability. The most extensively used of these techniques is Artificial Neural Networks (ANN). More recent techniques such as Support Vector Machines (SVM) have featured in the literature with better performance indices. However, SVM has not been widely embraced in petroleum engineering as a possibly better alternative to ANN. ANN has been reported to have a lot of limitations such as its lack of global optima. On the other hand, SVM has been introduced as a generalization of the Tikhonov Regularization procedure that ensures its global optima and offers ease of training. This paper presents a comparative study of the application of ANN and SVM models in the prediction of porosity and permeability of oil and gas reservoirs with carbonate platforms. Six datasets obtained from oil and gas reservoirs in two different geographical locations were used for the training, testing and validation of the models using the stratified sampling approach rather than the conventional static method of data division. The results showed that the SVM model performed better than the popularly used Feed forward Back propagation ANN with higher correlation coefficients and lower root mean squared errors. The SVM was also faster in terms of execution time. Hence, this work presents SVM as a possible alternative to ANN, especially, in the characterization of oil and gas reservoir properties. The new SVM model will assist petroleum exploration engineers to estimate various reservoir properties with better accuracy, leading to reduced exploration time and increased production. 1. Introduction Petrophysical properties such as porosity and permeability are two important properties of oil and gas reservoirs that relate to the amount of fluid in them and their ability to flow. These properties have significant impact on petroleum field operations and reservoir management. They both serve as standard indicators of reservoir quality in the oil and gas industry (Jong-Se, 2005). Porosity is the percentage of voids and open spaces in a rock or sedimentary deposit. The greater the porosity of a rock, the greater its ability to hold water and other materials, such as oil. It is an important consideration when attempting to evaluate the potential volume of hydrocarbons contained in a reservoir (Schlumberger, 2007a). Permeability is the ease with which fluid is transmitted through a rock's pore space. It is a measure of how interconnected the individual pore spaces are in a rock or sediment (Schlumberger, 2007b). It is a key parameter associated with the characterization of any hydrocarbon reservoir. In fact, many Petroleum Engineering problems cannot be solved without having an accurate permeability value. Many reports such as Ali (1994) and Mohagheh (1994) have featured the successful application of Artificial Neural Networks (ANN) as the pioneer Artificial Intelligence (AI) technique in oil and gas reservoir characterization over the years. Despite this, ANN has been reported to have some drawbacks (Petrus et al., 1995). The recent introduction of Support Vector Machines (SVM) that is based on the concepts of Tikhonov Regularization and Structural Risk Minimization (SRM) was introduced to overcome some of the limitations of ANN. Many reports such as such as Anifowose and Abdulraheem (2010); and Helmy et al. (2010) have presented SVM as a promising predictive technique in a good number of applications. This paper focuses on the study and analysis of the comparative performance of ANN and SVM in the prediction of porosity and permeability of some Middle East and American oil and gas reservoirs. To achieve this aim, Section 2 presents a succinct survey of ANN and SVM. Section 3 describes the experimental methodology, structure of datasets and the evaluation criteria for the study. Section 4 presents the results of the study with a detailed discussion while conclusion is presented in Section 5.

Discrepancies in terms of hydrocarbon initially in place volumes (HIIP) between static and dynamic models might take place because of nonlinear dependency of the saturation height function (SHF) versus porosity and averaging of height above free water level. Upscaling tends to eliminate the high and low porosity values in favor of the average porosity, which might lead to substantial changes in the resulting water saturation. Furthermore, pressure and compositional variation with depth in the dynamic model might lead to substantial contribution to the discrepancies, independent of upscaling.We present a procedure to address the above issues and provide full consistency between static and dynamic models in terms of HIIP, without the use of capillary pressure (Pc) scaling by the reservoir simulator. A "preprocessing" workflow is used to slightly reassign the Pc curve of each grid block so that a very close match is obtained between the two models in terms of HIIP.Results show that HIIP obtained from the dynamic model through equilibration using the proposed procedure are within 1% of those obtained using the static model without having to use Pc scaling in the reservoir simulator which warranties realistic and physically valid Pc curves in the reservoir dynamic model initialization process.

The flow of nitrogen foam in Bentheimer sandstone cores previously saturated with a surfactant solution has been investigated experimentally. The displacement process was visualized with the aid of a CT scanner. The CT data were analyzed to obtain water saturation profiles at different times. The pressure drops measured over core segments were recorded to determine the foam mobility. It was found that foam undergoes a transition from a weak to a strong state at critical gas saturation of Sgc = 0.75±0.02. This effect could be interpreted successfully in terms of the surge of the yield stress of the foam at this value. It is suggested that a confined jamming is most likely responsible for the mobility transition.

The commercial viability of SAGD process is negatively affected by several undesirable reservoir features like pronounced heterogeneity, low vertical permeability, thick and areally extensive shale barriers and steam thief zones. The efficiency of SAGD projects is also affected by the presence of mobile water saturation in the target zone. Although the presence of small mobile water saturation is not considered harmful, reservoirs with high mobile water saturation may be poorly suited for the SAGD process. Nonetheless, SAGD remains the only practical technology for in situ extraction of oil from oil-sand reservoirs, even when mobile water is present. This raises the question of how much mobile water is a show stopper. To investigate the effect of mobile water saturation on SAGD performance, high pressure physical model experiments were carried out. Different levels of mobile water saturations were established in the model by modifying the packing and saturating techniques. SAGD experiments were conducted by injecting superheated steam at controlled rates and producing the oil from the production well at constant pressure. The injection rate was selected to keep the pressure difference between the injector and producer at a low level. The oil production behavior was analyzed to evaluate the effect of water saturation on the thermal efficiency of the process. Based on the results of low (immobile) and high (mobile) water saturation experiments it was observed that oil recovery factor droped by 7% and Cumulative Steam Oil Ratio increased by 50 percent when initial water saturation was increased from 14.7% to 32%.

The Camisea-area gas reservoirs are world-class reservoirs with production from 15 wells producing more than 1.2 Bcf/day and 85000 BCPD from the jungles of Peru. These reservoirs consist of stacked dune and marine sands that have undergone substantial diagenesis. The salinities in these reservoirs are low by "Archie-standards". Extensive coring has been done in all of the reservoir intervals in the three main fields of San Martin, Pagoreni and Cashiriari. These cores show macro, meso and micro porosity distributions. The upper-most sands, Cretaceous Vivian and Chonta are shallow marine and tidal depositions, while the pre-Cretaceous Nia and Permian Noi-Ene sands are continental deposits mainly composed of eolian facies. Resistivities vary widely and gamma ray responses are both high and low in reservoir intervals. In some intervals there is very poor correlation between the gamma ray and the very low clay volumes due to feldspar content. The micro-porosity creates a "low-resistivity" pay effect and irreducible water content in the reservoir that is not related to clay porosity. High iron residues also may cause some complications for the interpretation. A combined approach of cores and logs was used to minimize uncertainty and maximize confidence in the results. Special core analysis was done to analyze the Archie exponents of "m" and "n" in the different formations. Core capillary pressures were also measured in conjunction with nuclear magnetic resonance core measurements to analyze irreducible water volumes. All three measurements were used to guide computation of water saturations from the logs in these environments, formation-by-formation. The objective was to use the core analysis parameters as a guide to validate computation of saturations and reserves in the most appropriate and accurate manner. Small errors in saturation would cause very large uncertainties in producible reserves from this mammoth field. Examples of the special core analysis results, SEM analysis, complex pore structure, and log analysis approach will be used to show the integration of the data to make the most accurate computation of reserves.

Through the years there has been the problem of characterising reservoirs properly with insufficient data. This is because such data were either not acquired due to operational difficulties, cut budgets or needs not anticipated or where acquired were not of the right quality and thus not useful in building reservoir models that could be used to predict reservoir performance. One such key shortage is relative permeability data from SCAL. The impact of this shortage is being felt as the Niger Delta reservoirs increasingly move to the "brown" category over time and the need for reliable reservoir models increases. A dependable reservoir model requires a robust definition of the relative permeability. To address this, a new opportunity that recycles an old technique is being proposed as an effective means of obtaining dependable and representative relative permeability data. The method leans on the abundance of log data for saturation evaluation, permeability estimation using techniques already proven and published. These techniques provide a platform for deriving water saturation and permeability as continuous curves with the result that irreducible water saturation and consequently relative permeability for water and hydrocarbons respectively can be derived as curves across reservoir intervals using expressions earlier published by Park Jones (1944/45). The resulting Rel-Perm versus Water saturation plots agree well with good data from SCAL showing that the technique can be deployed reliably where SCAL data is non existent or judged to be of poor quality. The technique also ensures that Rel-Perms are derived from data from subject reservoirs. This method has been applied successfully in reservoirs studies with the impact that it facilitated improvements in turn-around times and quality of history matching in reservoir simulation. Two such results are presented in this paper.

A petrophysical refresh of the bpTT fields in the Columbus Basin in Trinidad has been carried out. The objectives of the refresh were to provide continuity and consistency in petrophysical interpretations in this mature basin where over the years multiple vendors and differing interpretational approaches have been employed. In an effort to create a more robust core data set, data gaps were identified in the existing core analyses and supplemental analysis performed. The core data set was expanded to include Co/Cw measurements on plugs from 2 wells to augment legacy data to investigate log-based water saturation methods. New models were developed for permeability and water saturation and each of these models were calibrated against the core dataset. Permeability was re-evaluated with the new model being based on core-derived measurements and tuned to dynamic well test data to incorporate upscaling heterogeneities. Both log-based and core-based water saturation models were explored. The new core conductivity measurements provided support for the log-based method selected. Air-brine capillary pressure data have provided a key input to the development of a new saturation height function. The match between the new saturation height function water saturation and that derived from resistivity-based saturation is good, reinforcing its validity.

Belayim Land field is one of the oldest and largest oil fields in Egypt. It produces oil from many sandstone reservoirs. Oil is produced from 4-30 ohm-m resistivity in Belayim Formation, while to the contrary, in Rudeis Formation, 70–80 ohm-m resistivity is producing water. Rock characterization program was planned for Rudeis Formation using core material. It includes porosity, permeability, thin section description, SEM, wettability, relative permeability and capillary pressure by mercury injection to obtain pore throat size distribution. The wettability of the reservoir rock was determined by Amott method and proved to be strongly oil wet. The wettability controls the position of fluids within the pore network, where in oil-wet the water saturation distributed as poorly connected droplets within the macro-pores and filled the micro-pores. To simulate the saturation history of the reservoir, restored-state samples at irreducible water saturation were flooded with water to perform electrical resistivity measurements. Non-linear curves were obtained during the electrical resistivity measurements. Two saturation exponents (n) values were proposed to reflect the mechanism of forming continuous water film from the water droplets. The inflection point between the two (n) values was used to determine the resistivity cutoff below which water is produced. In addition, it represents the water saturation cutoff which was confirmed from the calculation of fractional flow of water from relative permeability measurements. The production data confirm the use of variable (n) to monitor the water saturation in oil-wet reservoirs.

Future workflows for water saturation interpretation in heterogeneous intervals, should take the heterogeneity into account in a robust manner. We want fine-scale geology to be modelled realistically, down towards mm-scale when appropriate, and populate these models with properties from pore-scale modelling. The geologically realistic models can then be upscaled for effective properties for different rocktypes. Accordingly, Archie's law can be used for water saturation interpretation without breaking its assumptions. The work presented is a study in which we want to validate such a workflow, and assess the predictivity to be expected. Two core plugs from a heterogeneous reservoir have been studied in a steady-state flow experiment while simultaneously measuring both resistivity and in-situ saturation in 3D by CT-imaging. Resistivity was measured in a 4-electrode set-up connected to a multi frequency impedance meter. The experimental design yielded data where now resistivity, relative permeability and fine-scale saturation are all coupled. 3D digital models of the plugs have subsequentially been constructed from CT porosity and populated with properties from pore-space modelling. This enables us to compare the fine scale saturation modelling, which is a difficult task for this type of sedimentary rocks, to the measured water distribution in the plug. Fine-scale saturation also determines upscaled resistivity, which we can model digitally using pore-space models and compare with the measurements. Our findings indicate that the workflow is viable, and that we are able to obtain predictive power from pore-scale modelling and fine-scale geological modelling.

Quantifying the uncertainty in the volumetric estimation of original oil in place (OOIP) is an important process in evaluating the field potential and hence in designing the proper and the most economical subsurface and surface facilities to produce the field reserves. This uncertainty in the OOIP estimate results from uncertainty in reservoir areal extent, net reservoir thickness, porosity, and hydrocarbon saturation. In this work, a methodology is presented to assess the uncertainty in the hydrocarbon saturation estimated from open hole logs using the commonly used empirical and theoretical shaly sand models.This technique is based on development of water saturation sensitivity charts for the most commonly used water saturation models (Simandoux, Indonesian, Waxman & Smits, Dual Water, and Effective Medium) due to the uncertainty in the different input parameters to each model separately. Both analytical and numerical error analysis techniques were used to develop these charts and hence used as a forward tool to quantify the uncertainty in the hydrocarbon saturation due to the uncertainty in the core and shale properties.Fifteen wells with 1300 shaly sand points from Alam Bewab formation, in Western desert of Egypt, were used as the data base in generating these sensitivity charts. The uncertainty in input data was assumed from ± 5 to ± 15%. The results showed a significant range of uncertainty in hydrocarbon saturation estimate from ± 2% reaching to ± 75% in some models.General water saturation sensitivity charts were developed, for each model, based on the above mentioned database and validated mathematically. These charts can be easily used to predict the uncertainty in hydrocarbon saturation estimate due to uncertainty in the input data. In addition, it can be considered as a useful screening tool to select the best saturation model to be used depending on the input data uncertainty.

Accurate predictions of water saturation were achieved through an integrated process involving Petrophysics and core analysis. During reservoir evaluation, log derived water saturation inaccuracies were found due to inconsistency with production data. The conventional resistivity methods for calculating water saturation did not fully resolve this inconsistency and hence a different approach was required. The inaccuracies associated with Sw prediction from Archie's resistivity are due to a number of factors including mud invasion, presence of a dual pore system and resistivity tools physics. In an effort to determine the true in situ water saturation and make more accurate predictions of oil in place a special study was performed. Such study involved: Cutting low invasion core in the reservoir.Using a tracer in the drilling mud to help differentiate formation water from invasion fluidsTaking Dean-Stark vertical plugs from the cores at the rig site and preserving them for laboratory analysis.Analyzing the fluids in the plugs to determine the degree of invasion.Use of plug analysis results to predict more accurate water saturation.Use vertical and horizontal resistivity measurements to identify the resistivity anisotropy and therefore, provide more accurate water saturation as compared to the conventional resistivity approach.Aquire RST-sigma log to help in data integration.Improved oil in place estimations. The tracer allowed the core analysis to determine the percentage of invaded fluids in the plug sample and more accurately predict the true in situ saturation state of the reservoir. The results showed that the dual pore system and layering have played an important role in the invaded fluid distribution and quantity. The new saturation model allowed for corrections to log measurements providing a better representation of the reservoir properties. The use of vertical resistivity was found to be important to correct for layering and allowing to obtain results closer to the direct core measurements. The new model estimates were in much closer agreement with production test data and this resulted into improved oil in place estimations. This paper, will discuss the methodology and data integration of Sw determination and how it lead into critical development decisions.

Peripheral water flooding has been the preferred pressure maintenance tool for many gulf carbonate reservoirs over the past 30 years. Due to uneven sweep and pressure distribution, this technique has given way to pattern floods in several gulf fields. As these new floods are established, it is important to understand the water saturation between wells to properly manage the sweep and recovery. In 2007, ADCO initiated water injection (WI) and WAG pilots to test the recovery strategy. The pilot employs advanced geophysical and modeling tools to measure formation properties at the wells and between wells; this paper discusses the WI pilot. Among the novel techniques applied is the crosswell electromagnetic method, which measures the interwell resistivity distribution between observation wells at the pilots. Interwell resistivity data can be used to infer the water saturation distribution because of the sharply different electrical resistivity between injected water and oil bearing reservoir rock. By allowing an evaluation of the water distribution long before the injected fronts reach producers or observers, a better and more rapid understanding of the pilot arises from the crosswell electromagnetic technique. In this paper, we briefly describe the pilot design, describe the detailed geological model and show results from the initial set of baseline and time lapse EM data sets from the water injection pilot. The images highlight the influence of background geological constraints on the flow.

The second Archie equation relates water saturation to formation resistivity index through a power function. This important relationship has been widely used to evaluate hydrocarbon saturation. However, many rocks don't obey this empirical rule. The best Archie fit to these data may not comply with physical bounds and create significant bias in the computed hydrocarbon saturation. This paper develops a new method for water saturation estimation based on the equivalent rock element model (EREM) that has been demonstrated to work well for rock transport properties to the first order effect. EREM contains two orthogonal pore components, one parallel and the other perpendicular to the direction of transport. Pore structure efficiency is defined as the volume ratio between the two components. When pores contain conductive and non-conductive fluid phases, the pore structure efficiency for the conductive phase changes with saturation due to fluid and mineral properties and pore structure variations. The conductive phase may become discontinuous below a certain critical saturation due to pore structure, wetting characteristics, interfacial tension, etc. The proposed algorithm incorporates this critical saturation phenomenon and relates capillary pressure data with electrical measurements. Approximating the pore structure efficiency of conductive phase as a function of saturation, we formulated a theoretical relationship between resistivity index and water saturation. For some rocks, additional conductive meachanisms may exist in addition to electrolytic conductivity, in which case the Archie's equation would not apply. Many proposed formulas can be found in the literature regarding non-Archie rocks, but each only works well for a particular type of rocks. This paper demonstrates that the EREM approach can account for the additional conductivity and work well regardless of rock type. This new approach expands the applications of the equivalent rock element model. It has several advantages over existing methods. (1) The approach is based on a simple physical model that reflects the two main components in a pore structure and accounts for the first order transfer effect. This ensures compliance with physical boundary conditions and increases the predictiveness. (2) The innovative inclusion of critical water saturation avoids the underestimation of water saturation at low water saturations. It links electrical measurements with capillary pressure measurements and allows the two observations to cross-validate and complement each other. (3) The proposed method is demonstrated to match core measurements from different rock and pore types with a single model.

An accurate determination of oil reserve either in virgin or in developed reservoirs represents the main task of petrophysicist and reservoir engineer. Input parameters of Archie's water saturation model in clean formation or suitable shaly water saturation model must be determined as accurately as possible. Archie's parameters a, m and n are the most important parameters to be estimated affecting determination of water saturation, in addition to the empirical constants required for selected empirical shaly water saturation model in case of shaly formations. This paper presents a new technique to determine Archie's parameters called 3-D Technique. This technique is based on the three dimension plot of three parameters; formation water resistivity, water saturation and formation porosity. A comparison study with conventional technique and CAPE technique is conducted to outline the accuracy level of each technique defining Archie's parameters a, m and n. Two field examples are given to show the applicability of three techniques and also to demonstrate the impact the uncertainty of Archie's parameters values on the accuracy of calculated water values using Archie's model or shaly water saturation model.

Interpretation problems are commonly associated with calculating water saturation in non-homogenous shaly sand reservoirs. Recent studies on the electrical anisotropy of shaly sands have shown that the level of our understanding our ability to correctly and accurately evaluate complex shaly sand reservoirs can be greatly improved. The model ("Total Expansible Clay Model") developed in this work is similar in form to the recent shaly sand Dual Water Model, except that the clay effect have been defined by a volume-conductivity transform rather than the usual volume-porosity transform. The model is experiment-based and designed to quantify the effects of the mineralogical and electrical properties of clay minerals on well log signatures. The Total Expansible Clay Model evaluates these effects via direct measurement of independent mineralogy and cation exchange capacity of the clay minerals within the reservoir sands. The model integrates the following as an effective basis for characterizing shaly sand reservoirs: Rietveld based siroquant assay for quantitative X-ray diffraction (XRD) used in determining mineral percentages from standard XRD trace patterns, Cation exchange capacity (CEC) used to determine the quantity of cations involved in the exchange at the shale-water interface, Porosity, permeability, density and resistivity measurements, Overall, the correlations drawn from this model yield improved results for total water saturation (especially in low porosity and highly shaly formations), which appear consistent with those calculated earlier using well known water saturation models (Dual Water and Juhasz). A total of 23 core plugs from two wells in the Cliff Head Field, Perth Basin, were analyzed for this study.

Determination of original hydrocarbon in place, OHIP, is a vital task in petroleum development. The estimation of appropriate rock and fluid properties is a requirement, with irreducible water saturation, Swir, being one of the key parameters. A representative Swir is also required when conducting multi-phase experiments, for example relative permeability determination. Furthermore, Swir has an influence on residual oil saturation during tertiary recovery. Conducting primary drainage capillary pressure experiments to measure Swir is the industry-established practise. Such experiments tend to require considerable resources and long time periods to complete. As a consequence, a limited number of representative core plugs are typically considered, often leading to data gaps for some facies within a reservoir. In such situations, an empirical model may be useful in predicting Swir. In a recent study, artificial neural networks have been applied successfully to the estimation of irreducible water saturation (Swir) for Australian formations. The model demonstrates a superior performance when compared with other, conventional models. This paper features the translation of the artificial neural network model into a simple mathematical equation that is suitable for quick hand calculation. Moreover, a new semi-empirical model to predict Swir is presented, containing five adjustable constants. The optimal values for these constants were obtained by minimizing the calculated error utilizing a genetic algorithm. Both neural network and semi-empirical models were developed, calibrated and validated by using an extensive data set gathered for Australian hydrocarbon basins.

In a large clastic reservoir of Saudi Arabia, rock typing by rock-quality-index (RQI) and flow-zone-indicator (FZI) proved to be an effective technique to develop porosity-permeability transforms for 8 lithologic facies in a reservoir model. Moreover, capillary pressure and relative permeability curves could be grouped into defined rock types as well. This technique provided an effective tool to distribute permeability, initial water saturation, relative permeability, and residual oil saturation per lithologic facie. Conventional core porosity and permeability data were used to determine FZI for each core. The FZI was manipulated mathematically to group into hydraulic units (HU) in distinct integers. Each HU was then correlated to corresponding defined lithologic facies. The resulted porosity-permeability transform for each facie was applied to distribute permeability in the reservoir model. Relative permeability and capillary pressure curves also clustered together in corresponding HUs. Confidence levels in distribution of initial water saturation and residual oil saturations were increased due to this HU rock typing. Series of J function and relative permeability curves for 8 facies were applied in the 1.4 million gridblocks reservoir simulation model. Distributed permeability and initial water saturations were validated against pressure transient analyses, core and well log data, and proved to be in excellent agreements. As a result, the uncertainty associated with petrophysical parameters was greatly reduced. The simulation model developed with the combination of HU rock typing and lithologic facies provided a higher history matching statistic (~70%) for 500 wells in a couple of weeks using assisted history matching method where facies-based petrophysical and aquifer parameters were globally modified in very narrow ranges. This paper presents through a case study the methodology applied for generating permeability and water saturation distribution by lithologic facies and hydraulic units and its impact on simulation history match and future prediction.

Inaccurate modeling of fluid flow near-wellbores is commonly recognized as shortcoming of numerical reservoir simulators. After water breaks through, the well's inflow involves two or three fluids flowing at velocity exponentially increasing with reducing distance to the well. Understanding of the oil/water inflow to wells and possible improvement of its simulation is the objective of this study. Current commercial simulators disregard the effect of dispersion due to high flow velocity. They only consider effects of viscous and gravity forces, capillary pressure and sometimes Non-Darcy flow effects. We postulate that a process of transverse immiscible dispersion should be considered in evaluating the oil/water transition zone around a well and the production water cut. Transverse dispersion is a process of mixing two fluids in the direction perpendicular to the segregated flow of two phases. In the process, one phase enters the stream of another phase and contributes to the flow characters and the saturation distribution change. Away from the well where the flow velocity is low, the effect is small and overshadowed by the capillary pressure effect. It, however, may significantly increase as the two-phase flow is approaching well since the transverse dispersion coefficient is a function of velocity. Thus, at the well, transition zone size and distribution might be significantly affected by transverse dispersion resulting in water production larger than results of current simulators. In this paper, an analytical model of transverse dispersion in porous media is derived and used to study various factors influencing the dispersion. The results show that radial distance and mechanical dispersion coefficient are essential to transverse dispersion. It also shows that transverse dispersion controls transition zone growth at the bottom of well's completion where vertical gradient of horizontal velocity is the largest – up to 0.25 ft/s-ft.

Resistivity data is normally used to evaluate water saturation using porosity values from porosity logs (neutron and density). Determination of initial oil (gas) in place is based on hydrocarbon saturation, porosity and thickness obtained from openhole logging data for a given drainage area. It is important not only to determine the initial hydrocarbon in place, but also to define the existing hydrocarbon moveability, indicating the recoverable hydrocarbon and its type. This paper presents a new approach of hydrocarbon moveability factor (HCM). This factor is derived from shallow and deep resistivity data. The relation F = a/ϕm is correct in water saturated zones, in partially saturated zones this relation becomes invalid and it will give the apparent formation resistivity factor (Fa). Based on this idea the hydrocarbon moveability factor (HCM) has been derived. With scale goes from 0.0 to 1.0. It is found that for HCM less than 0.75, hydrocarbon is moveable and for HCM greater than 0.75, the hydrocarbon is immoveable. When HCM is less than 0.25, the moveable hydrocarbon is gas and for HCM greater than 0.25 and less than 0.75, the moveable hydrocarbon is oil. Field examples have been analyzed with the HCM factor. These field examples demonstrated the contribution of HCM in the field of hydrocarbon type identification and determination of hydrocarbon moveability from openhole resistivity logging.

Low field nuclear magnetic resonance (NMR) relaxometry has been successfully used in the past to perform in-situ estimates of oil and water content in unconsolidated oil sand samples. This work has intriguiging applications in the oil sands mining and processing industry, in the areas of ore and froth characterization. Studies have previously been performed on a database of ore and froth samples from the Athabasca region in northern Alberta, and preliminary results have been encouraging. In this paper, supporting data is presented and refinements suggested to the previous algorithms, to improve the oil and water saturation predictions. A suite of real and synthetic samples of bitumen, water, clay and sand have been used to investigate the physical interactions of the different components on the NMR spectra. An automated algorithm is used to separate the oil and water NMR signals, and this algorithm is tested against samples both from this zone and from other heavy oil fields. Moreover, preliminary observations regarding spectral properties indicate that it may be possible in the future to estimate the amount of clay in the samples, based upon shifts in the NMR spectra. NMR estimates of oil and water content are fairly accurate, thus enhancing the possibility of using NMR for both in-situ oil sands development and in the oil sands mining industry.

Seismic data provides a unique source of information widely used for reservoir characterization. High resolution 3D seismic impedance field is critical for building a model of facies distribution. 4D seismic time-lapse can reflect the water saturation difference (time-lapse), hence could point towards potential facies (dis)continuities not originally apparent from the static data. An interactive procedure is proposed to improve the prior facies model by spotting areas of large discrepancies between the recorded 4D seismic data and the corresponding forward simulated time-lapse data. Various indicators of discrepancy are proposed. The subsequent correction honors the prior geological scenario. It is fast because it does not call for any iterative optimization.

Typical petrophysical deliverables for volumetric and modeling purposes are net reservoir, porosity, permeability, water saturation and contact locations. These data are usually provided without quantitative determination of their uncertainties. Current computing power renders it now feasible to use Monte-Carlo simulation to determine the uncertainty in petrophysical deliverables. Unfortunately, quantitative uncertainty definition is more than just using Monte-Carlo simulation to vary the inputs in your interpretation model. The largest source of uncertainty may be the interpretation model itself. This paper will use a variety of porosity interpretation models to illustrate how the impact of each input on the uncertainty varies with the combination of input values used in any given model. It will show that use of the incorrect model through oil and gas zones may give porosity estimates with Monte-Carlo derived uncertainty ranges that exclude the actual porosity. Core data provides the best means of quantifying actual uncertainty in the petrophysical deliverables. Methodologies for deriving uncertainties quantitatively by comparison with core data will be presented. In the absence of core data, interpretation models should have been tested against core data through the same or similar formations nearby. Monte-Carlo simulation can then be used as an effective means of quantifying petrophysical uncertainty. Comparisons between the core comparison and Monte-Carlo techniques will be made, showing that similar results are achieved with the appropriate interpretation models. The methodologies described in this paper are straightforward to implement and enable petrophysical deliverables to be treated appropriately in volumetric and modeling studies. In addition, quantification of petrophysical uncertainty assists in operational decision-making by letting users know how reliable the numbers produced actually are, and what range of properties is physically realistic. Such work also allows the key contributions to uncertainty to be defined and targeted if overall volumetric uncertainty must be reduced.

The "genetic petrophysics" approach for predicting petrophysical parameters using genetically focused neural nets (GFNN) has only recently been introduced. The approach only requires a minimum number of core plugs, along with associated wireline log data, from a chosen representative genetic unit (RGU). This case study has successfully developed and tested this new methodology to predict residual water saturation. Combinations of wireline logs and core data from a short 7m RGU interval in a North sea well were used to train the GFNN predictors. These were then applied to predict the residual SW throughout the whole logged section in the training well and adjacent wells in the same field. Traditional hydraulic unit analysis provided the basis for selecting the minimal training plugs. Only 4 core plugs were finally required to represent the hydraulic units in the RGU and provide good results. This approach is very cost effective in terms of core material and computing time. Presently we have only tested this approach in oil bearing shoreface reservoirs. Thus, it is recommended that this approach be tested in other environments.

The modeling of initial water saturation distribution in simulation models could be achieved by one of two techniques. The upscaling the log interpreted water saturation distribution of the finer 3D geological model constitutes the first technique. The challenge in this technique is to accurately describe the important rather thick transition zone of carbonate reservoirs. A large and non-achievable number of control points (wells) are needed. Rock type controlled restored state oil-brine capillary pressure, Pc, curves is the second technique. This results in an accurate picture of oil reserves providing that the rock types are distributed correctly. Significant numbers of oil-brine Pc measurements are needed however, to address all rock types, wettability, porosity and permeability ranges. Frequently, not all the necessary Pc curves are available. To fill this gap, reservoir characterization data along with log-derived water saturation, Sw, data are utilized to generate representative Sw-height functions for all rock types of a carbonate oil reservoir. Only cored wells are employed, ensuring the most accurate rock-type profiles. To match the span of Sw at a given depth, sorting by porosity, permeability and reservoir quality index within each rock type is carried out. Describing the large span of log-derived Sw in the transition zone is especially challenging. The technique assumes the log-derived saturations are correct and emphasize the need for laboratory tests of resistivity indices and cementation factors representative of the dominant rock types for accurate log interpretation. The attainability of a good matching between the log-derived and the Pc-derived Sw and bulk volumes of water depth profiles for the cored and un-cored wells validated the modeling process. This is supported by statistical analysis and mapping of the water saturations in all layers of the simulation model.

The second Archie equation relates water saturation to formation resistivity index by a power function. This important relationship has been widely used to evaluate hydrocarbon saturation. However, many rocks don't obey this empirical rule. The best Archie fit to these data may not comply with physical bounds and create significant bias in computed hydrocarbon saturation. This paper develops a new method for water saturation estimation based on the equivalent rock element model (EREM) that has been demonstrated to work well for rock transport properties to the first order effect. EREM contains two orthogonal pore components, one parallel and the other perpendicular to the direction of transport. Pore structure efficiency is defined as the volume ratio between the two components. When pores contain conductive and non-conductive fluid phases, the pore structure efficiency for the conductive phase changes with saturation due to fluid and mineral properties and pore structure variations. Approximating conductive phase pore structure efficiency as a function of saturation, we formulated a theoretical relationship between resistivity index and water saturation. The conductive phase may become discontinuous below a certain critical saturation due to pore structure, wetting characteristics, interfacial tension or a combination of different factors. The proposed algorithm readily incorporates this critical saturation phenomenon and explains the empirical relationships between resistivity index and water saturation for both Archie and non-Archie rocks. It fits core measurements better than Archie's second equation as demonstrated in our real examples and provides more accurate estimation for water saturation. This new approach expands the applications of the equivalent rock element model.