Abstract In current economic and environmental contexts, the optimization of long, horizontal well completion and the maximization of individual well performance are becoming increasingly important. The challenge is to be able to start improving the production efficiency while designing an adapted completion for each well without compromising the project economy. The cost-effective formation evaluation technique described in this paper allows rapid identification of dynamic heterogeneities along the reservoir after the drilling of a horizontal well. This key information then can be used to optimize well completion and treatment. This new approach, called WTLog, combines well testing and logging techniques and was introduced initially for the optimization of unconventional well completion (Jacques et al., 2019 and Manivannan et al. 2019). The log begins by circulating a low-viscosity liquid that can be injected in the formation through the mud cake. The brine circulation operation is run at the end of the drilling phase, after reaching TD of the drain while maintaining a constant wellhead pressure at the wellhead. The constant pressure control can be applied without a specific additional choke device when Managed Pressure Drilling (MPD) is used to drill the formation section. The inlet and outlet flowrates are measured accurately, and their difference corresponds to the apparent formation-injection rate. The depth of the interface between the two liquids inside the borehole is estimated from the flowrates and pressure measured at the wellhead. Combining these data allows derivation of the low-viscosity/liquid-injection profile along the open hole. A permeability log then can be derived by inversion. Well Test Logging has been applied successfully for the first time on two horizontal wells in a conventional carbonate reservoir. The interpretation results were benchmarked to static conventional openhole logs and validated against the data log obtained by the dynamic production log tool (PLT) performed after well start-up. This technique opens new perspectives for optimizing well completion in these carbonate-fractured formations for which porosity logs might not be a good permeability indicator and where conductive fractures seen on image logs are not always indicative of future production.

Abstract Offshore exploration requires the evaluation of hydrocarbon presence, estimation of volumes in place, and flow potential. To this capacity, formation testers are widely used to determine static data such as reservoir fluid gradients and reservoir pressure, obtain fluid samples, and to assess reservoir connectivity. Dynamic data, acquired with interval pressure transient testing and well testing techniques, are used to assess reserves and productivity. However, these evaluation techniques provide dynamic data at different resolution and length scales, and with different environmental footprint, cost, and operational constraints. A new wireline formation testing technique known as deep transient testing (DTT) has been introduced, which combines high-resolution measurements, higher flow rates, and longer test durations to perform transient tests in higher permeability, thicker formation, and at greater depth of investigation than with previous formation testers - without flaring and at a low carbon footprint. The platform combines advanced metrology with extensive automation to generate unique, real-time reservoir insights. Traditionally, pressure transient analysis and well deliverability predictions were produced through an analytical framework. Today, deep transient testing measurements are interpreted, and placed in reservoir context, in real-time by integration with geological and reservoir models. These steps can be performed from any wellsite utilizing cloud-based resources. Products such as reservoir fluid compressibility, saturation pressure, equation of state (EOS) models, well productivity, or minimum connected volumes are integrated in real-time interpretation utilizing numerical analysis. The digital infrastructure enables key reservoir insights to be shared between all stakeholders in a transparent and collaborative environment for both operational control and rapid decision making. This paper presents a case study where the new DTT technique was combined with numerical analysis and real-time integrated workflows to characterize a multilayer reservoir in a recent discovery in deepwater Mexico. During the drawdown phase of the DTT operation, real-time downhole fluid analysis was used to determine the fluid composition, density, viscosity, compressibility, and saturation pressure. These fluid properties were then used to generate and tune an EOS model. Accurate drawdown flow rate measurements and the subsequent pressure transients were combined with the fluid model and geologic model to enable integrated pressure transient history matching. The resulting calibrated numerical model honors the fluid measurements and geologic model and was used to predict the permeability profile, zonal producibility, and the volume of influence of the test.

Abstract During well planning, drillers and petrophysicists have different principle objectives. The petrophysicist’s aim is to acquire critical well data, but this can lead to increased operational risk. The driller is focused on optimizing the well design, which can result in compromised data quality. In extreme cases, the impact of well design on petrophysical data can lead to erroneous post-well results that impact the entire value-chain assessment and decision making toward field development. In this paper, we present a case study from a syn-rift, Upper Jurassic reservoir in the Norwegian Sea where well design significantly impacted reservoir characterization. Three wells (exploration, appraisal, and geopilot) are compared in order to demonstrate the impact of overbalanced drilling on well data from both logs and core. Implications for reservoir quality assessment, volume estimates, and the errors introduced into both a static geomodel and dynamic reservoir simulation are discussed. This case study highlights the importance of optimizing well design for petrophysical data collection and demonstrates the potential for value creation. Extensive data collection was initially carried out in both exploration and appraisal wells, including full sets of logging while drilling (LWD), wireline logging, fluid sampling, and extensive coring. Both wells were drilled with considerable overbalanced mud weights due to the risk of overpressured reservoirs in the region. The log data was subsequently corrected for significant mud-filtration invasion, with calibration to core measurements guiding the interpretation. Geological and reservoir models were built based on results from the two wells, and development wells were planned accordingly. A thorough investigation of core material raised suspicion that there could also be a significant adverse effect of core properties resulting from overbalanced drilling. The implications were so significant for the reservoir volume that a strategic decision was made to drill a geopilot well close to the initial exploration well, prior to field development drilling. The well was drilled six years after the initial exploration phase with considerably lower overbalance. Extensive well data, including one core, were acquired. The recovered core was crucial in order to compare the reservoir properties for comparable facies between all three wells. The results from the core demonstrate distinctly different rock quality characteristics, especially at the high end of the reservoir quality spectrum. Results of the core study confirmed the initial hypothesis that overbalanced drilling had significantly impacted the properties of the core as well as the well logs. The study concluded that the updated reservoir model properties would significantly increase the in-place volumes compared to the pre-geopilot estimate. This study shows how well design adversely affected petrophysical measurements and how errors in these data compromised geological and reservoir models, leading to a suboptimal field development plan that eroded significant value. This example provides a case study that can be used to improve the well design so that petrophysicists and drillers can both be part of the same value creation result. Future work will include further laboratory investigations on the effects of high overbalanced drilling on core and possible “root causes” for compromised core integrity.

Abstract The gas present in the Valhall overburden crest area interferes with the seismic data and obscures the fault detection (minor faults). Spatially resolving fractures and fracture network is essential for subsurface understanding and future well placement in this field, and it is a critical input to the dynamic reservoir model. Additionally, mapping the fracture network in poor permeable reservoir formation beyond the wellbore is crucial to identify completion intervals to maximize productivity/injectivity, and hence field value. The well 2/8-F-18 A was drilled on the crest of the Valhall field as a pilot water injector in Lower Hod formation, where core and data analysis formed the foundation for a future potential 11 well development. The well is placed in the southern section of the Valhall crest, and no major faults or strong amplitude features were mapped out in the overburden via surface seismic before drilling. In this case study, an integrated workflow is proposed and tested within the reservoir formation to identify “sweet” (permeable and fractured) zones beyond the wellbore. This is achieved using borehole acoustic data combined with image and ultrasonic imaging to characterize fracture networks beyond the borehole wall. The sonic imaging workflow identifies reflection events from fractures and faults and provides the true dip, azimuth, and location in 3-dimensions. This data is complemented by nuclear magnetic resonance (NMR), dielectric and spectroscopy data to understand reservoir petrophysics. NMR-derived permeability has also been evaluated for identifying high permeable zone in this formation, which primarily focuses on intergranular permeability of the formation a few inches away from the borehole wall. Reservoir textural heterogeneity and fractures beyond the wellbore wall make this method difficult to estimate or enhance the effective permeability estimate. The baseline assumption for the NMR permeability estimation is also not valid in Hod formation; the Timur and SDR equation needs significant change to match core permeability. Hence, the primary aim is to identify a fracture network that will help support water injection and maximize hydrocarbons production through them. The goal is to establish a workflow from the learnings of this study, performed on the pilot well, validate its findings with the near-field data (core, imaging, and ultrasonic), and optimize it if needed (described in the methodology section). The developed workflow is then intended to be used to optimize the placement of future wells. The results achieved from the integrated workflow identified a key fault and mapped it approximately 23 meters away on each side of the borehole. It also captures acoustic anomalies (high amplitudes), validated based on near-field data, resulting from a fracture network potentially filled with hydrocarbons. The final results show the sub-seismic resolution of the fracture and fault network not visible on surface seismic due to the gas cloud above the reservoir and frequency effect on the surface seismic when compared to borehole sonic data. Evidently enhancing the blurred surface image, which helps enhance the structural and dynamic model of the reservoir.

Abstract Legacy crushed rock analysis, as applied to unconventional formations, has shown great success in evaluating total porosity and water saturation over the previous three decades. The procedure of crushing rock into small particles improves the efficiency of fluid recovery and grain volume measurements in a laboratory environment. However, a caveat to crushed rock analysis is that water and volatile hydrocarbon evaporate from the rock during the preparatory crushing process, causing significant uncertainty in water saturation assessment. A modified crushed rock analysis incorporates nuclear magnetic resonance (NMR) measurements before and after the crushing process to quantify the volume of fluid loss. The advancements improve the overall total saturation quantification. However, challenges remain in the quantification of partitioned water and hydrocarbon loss currently derived from NMR spectrum along with its uncertainty. Furthermore, pressure decay permeability from crushed rock analysis has been reported to have two to three orders of magnitude difference between different labs. The calculated pressure decay permeability of the same rock could even vary several orders of magnitude difference with different crushed size, which questions the quality of the crushed pressure decay permeability. In this paper, we introduce an intact rock analysis workflow on unconventional cores for improved assessment of water saturation and enhanced quantification of fast pressure decay matrix permeability from intact rock. The workflow starts with acquisition of NMR T 2 and bulk density measurements on the as-received state intact rock. Instead of crushing the rock, the intact rock is directly transferred to a retort chamber and heated to 300 °C for thermal extraction. The volumes of thermally-recovered fluids are quantified through an image-based process. The grain volume measurement and a second NMR T 2 measurement are performed on post retort intact rock. The pressure decay curve during grain volume measurement is then used for calculating pressure decay matrix permeability. Total porosity is calculated using bulk volume and grain volume of the rock. Water saturation is quantified using total volume of recovered water. In addition, the twin as-received state rocks are processed through the crushed rock analysis workflow for an apple-to-apple comparison. Meanwhile, pressure decay permeability is cross-validated against the steady state permeability of the same sample. The introduced workflow has been successfully tested on different formations, including Bakken, Bone Spring, Eagle Ford, Cotton Valley, and Niobrara. The results show that total porosities calculated from intact rock analysis are consistent with total porosities from crushed rock analysis, while water saturations from the new workflow are average 8%SU (0.2–0.7%PU of bulk volume water) higher than those from the prior crushed rock workflow. The study also indicated that for some formations (e.g., Bone Spring) the fluid loss during crushing process is dominated by water, however, for some other formations (e.g., Bakken), hydrocarbon loss is significant. Pressure decay permeability quantified using intact rock analysis is also confirmed within an order of magnitude of steady state matrix permeability.

Abstract The economic feasibility of a well drilled in tight carbonates is extremely dependent on the level of fracture permeability; hard and dense carbonate formations may not be considered as net pay without the presence of fractures. The evaluation of fractures is a key to reservoir effectiveness characterization for well drilling, completion, development and stimulation of fractured reservoirs. While knowledge of the geological conditions and regional stress is helpful to estimate the characteristics of the natural fracture system in a given reservoir, the true extent of the natural open fracture system in any specific location is typically unknown. Several methods are available to the industry to identify natural fractures near the wellbore, including acoustic and resistivity image logs. In some cases, the poor-quality results of these techniques do not provide reliable information and such data cannot be available in all the wells. When minor downhole losses are accurately detected, it is possible to locate and characterize the natural open fractures intersected by the drill bit while drilling operations. The differential flow (Flow-out minus Flow-in) and the Active Volume System are continually monitored during drilling and integrated with drilling and hydraulic parameters. These readings are processed in a computer-based, data-acquisition system to form a compensated delta-flow signal that identifies the occurrence of downhole fluid losses. The differential flow is measured accurately through a dedicated Coriolis type flow-meter with a Limit Of Detection up to 10 l/min. By accurately detecting and measuring the downhole micro-losses instantaneously at the surface, the responses would be compared to predefined models for fracture characterization; that enables identification of different types of fractures (open natural, induced fractures). The system can detect very fine micro-fractures that might not be visible with wireline images; fracture density plots can then be created to highlight the fracture concentration along the well. Drilling deep wells in Kuwait is challenging due to high pressure, high-temperature formations, with the Bottom Hole Pressure of +15kpsi and Bottom Hole Temperature of +150 Centigrade degrees. In conventional surface systems, the loss detection relies on the Active Volume System and the Paddle type Flow-out sensor; however, these systems usually fail to identify the minor mud losses associated to open fractures. Especially for active pits with a big surface, it is almost impossible to identify few millimetres of mud level decrease and during fluid transfers, mud conditioning will make the job even more difficult to identify minor losses. With flow paddle type of sensors, the flow out information is not displayed as a calibrated value but rather as a percentage of full scale, which can be difficult to interpret. Instead, dedicated Coriolis type flowmeters properly installed, can identify flow rate changes accurately, regardless of any transfer of mud, water or diesel between pits. By applying this technique, it is possible to identify fractures while drilling in different types of wells, such as vertical, highly deviated and horizontal. The data were validated initially through core and image logs and further applied in next drilling campaigns.

Abstract With this paper, we demonstrate how CoreDNA, a trans-disciplinary suite of high-resolution, non-destructive measurements performed on whole cores at the onset of core analysis programs, helps operation geologists and petrophysicists with an innovative, cost effective and objective way to characterize the reservoir quality of highly laminated hydrocarbon-bearing formations where the standard practice (systematic plugging every foot) fails to provide a correct estimate. The case study focuses on core data from three wells intersecting formations characterized by very thin (millimetre-scale) sand and clay/silt laminations where the resolution of conventional wireline and lab gamma ray logs were not sufficiently sharp for an effective evaluation of reservoir quality. Although a high volume of routine core analysis data was already available for these wells, the remaining uncertainty on reservoir evaluation was deemed high enough by the study team to motivate the acquisition of additional data comprising ultra-high resolution pictures (1.8μm/px) and topographic maps created from micron-accurate laser scans. We explain how continuous profiles of grain size indicators could be used for the prediction of permeability variations across these laminated formations and for the definition of a permeability cut-off for the identification of poor vs good reservoir ratios compatible with the reservoir characteristics. CoreDNA test procedures are specifically designed to greatly accelerate the deliverables of core analysis, so that petrophysical evaluation may start right from the moment cores arrive from the well site, which is usually month before routine core analysis results are known. In the context of this paper, CoreDNA results were confirmed a-posteriori by the permeability measured on plugs samples from the two first wells. In the third well however, some marked differences were observed: although permeability ranges were found similar by the two methods, the distribution of permeability values obtained from routine core analysis conducted according to standard guidelines (one sample per foot) gave a more optimistic picture of permeability (90% rock above the 1mD cut-off) than the alternative approach based on high resolution continuous grain size data (70% rock above the 1mD cut-off). From the above findings, we conclude that a standard 1-ft interval for plug acquisition is not enough to fully characterise the distribution of permeability in highly laminated formations. Alternatively, a continuous profile of permeability index based on high resolution grain size measurements offers a fast and cost-efficient solution to obtain representative reservoir quality data, which enable objective well and reservoir management decisions few days after barrel opening without compromising core integrity for further studies.

Abstract Mud filtrate invasion is a complex and time-dependent process. During the process, a zone of finite size around the wellbore (invasion zone) in which a portion of the initial pore fluids have been displaced by the mud filtrate is gradually generated. As a result, the petrophysical and fluid properties of the formation in this zone will be inevitably altered, and sometimes tend to be quite different from their initial values. Petrophysicists and logging analysts have long considered mud filtrate invasion as a nuisance due to its troublesome effect on formation properties and logging measurements, especially on resistivity logging measurements. Note that even deep reading resistivity logging may not see deep enough (beyond the invasion zone), and need to be corrected. Therefore, simulation of mud filtrate invasion under near reservoir conditions is crucial for an in-depth understanding of its physics and effects on logging measurements, and hence for logging interpretation and formation evaluation. Otherwise, this will produce substantial errors in determining initial formation properties, and estimating hydrocarbon reserves and well productivity. To date, most researchers have done a number of works on mud filtrate invasion on the basis of physical simulation at core plug scale. However, conducting invasion experiment on core plug has intrinsic limitations. Firstly, the cylindrical shape of core plug determines that the seepage form of mud filtrate within it (horizontal linear flow) is completely different from that (plane radial flow) in the actual downhole environment, thereby causing a poor representation of the filtration law observed in the experiment. Secondly, due to the small size of core plug, it is almost impossible to monitor the radial resistivity variation for reflecting the dimension and geometry of the invasion zone. To overcome the limitations, a large-sized formation module (sectorial block structure, 55.9 cm in radial depth, and 10 cm in thickness) made by sandstone outcrop was introduced in this paper. Compared with core plug, as a novel type of experimental equivalent, the formation module is larger in size, greater in saturation capacity, and much more similar to the in-situ formation. Its structure can ensure the seepage form of mud filtrate within it is exactly the same as that in the actual downhole environment. Its large size is able to provide enough space and radial distance to follow the entire invasion process from beginning to dynamic equilibrium. The dynamic processes of long-term water-based mud filtrate (WBMF) invasions were duplicated realistically in laboratory. During the whole experimental period, the dynamic invasion data (including radial formation resistivity profile and filtration rate) can be uninterruptedly real-time acquired, thereby investigating and comparing the phenomenon of WBMF invasion in the formation modules with different physical properties. Finally, by combining physical and numerical simulation, the invasion characteristics of WBMF in high-permeability and tight sandstone reservoirs under in-situ formation conditions were quantified. The results obtained in this paper provide an experimental basis and theoretical support for enlightening novel simulation methodologies of mud filtrate invasion, revealing invasion mechanisms, and establishing invasion correction model for electric logging, etc.

Abstract Building reliable subsurface models requires detailed knowledge of both the rock and fluids involved. One critical petrophysical property determining the viability of a development is the hydrocarbon saturation. In 3D geological models, the saturation is populated via Saturation height models and free fluid levels. In populating a 3D model with meaningful properties, measurements at various scales are integrated. Core measurements acquired at resolution far superior to that used in the 3D models require a change of scale-upscaling step. The process of accurately predicting water saturation in the upscaled model is not trivial. Here we follow this process by employing a saturation height model (SHM) at different scales in relationship to various permeability realizations. Multiple choices available as inputs into the SHM in various ranges of sensitivity with respect to the free water level position as well as different rock quality are looked at. Various degrees of heterogeneity are studied by using synthetic data, the saturation prediction accuracy based on upscaled input rock properties (like arithmetic/geometric and harmonic upscaled permeability) is investigated. For homogeneous rocks a workflow is detailed with the purpose of detecting the upscaling limits highlighting the possible errors that might appear in the upscaling process. A counterintuitive result is that in the transition zone (the focus of this work) permeable rocks are more prone to errors than the less permeable ones. We also conclude that no alteration of the SHM is necessary in the upscaling process. Given the fact that rock quality enters the SHM and that permeability upscaling follows a route that ultimately attempts to honor well performance, a natural question is what the relevance of such a permeability model as input for the SHM is. Our results highlight the best choices for an upscaled SHM input (upscaled) permeability- not necessarily the upscaled permeability used in history matching. Smallest errors are shown to be resulting from using geometrical or 1/3 power law upscaled permeability.

Abstract Reliable and real-time assessment of directional permeability and saturation-dependent capillary pressure are utterly important because they significantly affect the exploitation strategies. Conventional well-log- based methods (e.g., NMR-based, saturation-height analysis, resistivity-based, correlation-based) are either highly dependent on calibration efforts or rely on model parameters which are difficult to obtain in real-time and make them dependent on core measurements. Moreover, most conventional methods for assessment of directional permeability and saturation-dependent capillary pressure fail in the presence of multi-modal pore-size distribution. Recent publications suggested that integration of transverse Nuclear Magnetic Resonance ( T 2 NMR) and resistivity measurements enables assessment of pore-throat-size distribution as well as permeability and capillary pressure. However, the reliability of these methods is questionable in rocks with complex/multi-modal pore geometry. The objectives of this paper include (a) reliably estimating a variable constriction factor (a geometric parameter which relates the pore- and throat-size) in rocks with complex pore geometry to accurately quantify pore geometry, which is the main contribution of this work, (b) developing a new rock physics workflow for integrating NMR and electrical conductivity for assessment of permeability and capillary pressure that takes into account a variable constriction factor, and (c) verifying the reliability of the introduced workflow using core scale measurements. The proposed workflow starts with calculating pore-body- size distribution from NMR T 2 distribution. Then, we combine electrical resistivity and pore-size distribution to estimate the distribution of constriction factor in the pore structure. Next, we determine pore-throat- size distribution using the estimated variable constriction factor. We then introduce a new permeability model which takes variable constriction factor into account. The inputs to the permeability model include throat-size distribution, tortuosity, and porosity. Finally, we calculate saturation-dependent capillary pressure using the estimated throat-size distribution. We successfully verified the reliability of the introduced workflow in the core-scale domain in carbonate rock samples with complex pore structure. The permeability estimates obtained by the new workflow yielded less than 7% average relative error when compared against core measurements. We also observed a good agreement between the throat-size distribution and capillary pressure estimated from the new workflow and the ones acquired from MICP (mercury injection capillary pressure) measurements. Results also confirmed that integration of a variable constriction factor improves directional permeability estimates compared to cases where an effective constriction factor was used to quantify pore-throat size distribution in rocks with multi-modal pore-size distribution.

Abstract Relative permeability is one of the most important petrophysical parameters to evaluate a reservoir’s production during primary and subsequent secondary or enhanced oil recovery processes. Yet measured relative permeability data for tight oil reservoirs are very scarce to find in the literature, mainly because the measurement is difficult and time consuming to make. In this paper the protocol and results of water/oil, surfactant /oil, CO 2 /oil, and N 2 /oil relative permeability are presented, and compared to the digital core analysis results where wettability was set to water-wet or mixed-wet, as well as the Brooks-Corey model. Amott-Harvey wettability index was measured to explain the differences. The target formation is a sandstone tight oil formation located in Songliao Basin, China. Its permeability is mostly in the 0.01-5mD range. Core and oil samples from the target formation were used in the wettability and relative permeability determination. Relative permeability was measured at reservoir conditions using a customized core flow setup. Core samples were cleaned then wettability restored. To match the reservoir fluid viscosity and avoid changing wettability, stock tank oil was blended with kerosene to reservoir fluid viscosity at reservoir temperature. Relative permeability was measured using the unsteady-state method. Amott-Harvey wettability index was measured on core samples from the same formation at reservoir temperature. Amott-Harvey wettability index results show that the restored wettability ranged from water-wet to oil-wet, with most samples being mixed-wt. The addition of non-ionic surfactant promoted wettability change toward more water-wetness. However, anionic surfactant had little effect on reversing wettability. Oil relative permeability (Kro) results obtained from the digital rock analysis (DRA) assuming uniform water-wetness are consistent with relative permeability calculated from mercury injection capillary pressure using Brooks-Corey model. When wettability of the digital rock model was set to mixed-wet, the resulted Kro matches the measured Kro of a sister plug to the sample used to build the digital rock model, which is consistent with the wettability measurements. The addition of surfactants increased both water and oil relative permeability through wettability alteration and IFT reduction. CO 2 flood was conducted as an immiscible flood due to reservoir pressure lower than MMP. CO 2 flood left high residual oil saturation compared with water floods. N 2 flood left even more oil behind compared with CO 2 flood. Relative permeability provides key input parameters for formation evaluation and the subsequent EOR processes such as huff-n-puff operations. There are very little published relative permeability data for tight oil reservoirs. This work extends the relative permeability database, and is a starting point for future EOR work.

Abstract In carbonate reservoirs, the estimation of a reliable permeability log is a long-standing problem mainly because of the inherent multi-scale heterogeneities. The conventional approach relies on core-calibrated algorithms applied to open-hole (OH) logs. In general, this static log-based prediction uses to underestimate the actual dynamic performance of the wells and an ad-hoc integration with production logging tool (PLT) and well test (WT) analyses represents a required step to correct the initial estimation. However, it is critical, and at once challenging, to define the relation between dynamic-based corrections and OH characterization outcomes. An elegant solution is here proposed that makes use of predictive analytics applied on special core analyses (SCAL), nuclear magnetic resonance (NMR) log modeling, and multi-rate PLT/WT interpretations. The methodology is presented for a complex oil-bearing carbonate reservoir and it starts with an advanced NMR characterization performed downhole for more than 100 wells, and after a rigorous calibration with SCAL. The main outputs are a robust porosity partition (in terms of micropore, mesopore and macropore contributions), and a physics-based permeability formula. Although the match with core data demonstrates the reliability of the applied NMR rock characterization, log permeability underestimates the actual dynamic performances obtained from WT, as expected. At the same time, multi-rate PLT data from more than 150 wells are used to compute an apparent permeability value for each perforated interval, automatically consistent with the associated WT interpretation. Finally, both static and dynamic characterization outputs are used as inputs for a dual random forest (RF) template. In detail, the first RF algorithm learns through experience how NMR porosity partition and core-calibrated permeability are related to PLT/WT apparent permeability values, after considering the proper change of scale. Next, the second RF is utilized to estimate the uncertainty associated to the previous step, still in a completely data-driven way. Hence, the so-defined dual model provides a continuous automatic flow-calibrated permeability log, together with its confidence interval, directly from static NMR responses. The presented methodology allows dynamic data to be incorporate efficiently into a static workflow by means of a pure data-driven analytics approach. The latter is able to shed light on the statistical relationships hidden in the available datasets, thus leading to a more accurate permeability estimation. It is also shown how this provides fundamental information for perforation strategy optimization and reservoir modeling purposes in such carbonate rocks.

Abstract Rock typing in carbonate reservoirs has always represented a difficult challenge due to rock heterogeneity. When interpreting electrical logs, the thick carbonate formation can leave an impression of a homogenous environment; however, looking at core analysis and mercury injection capillary pressure (MICP) data, reservoir heterogeneity can be determined. This complexity of the formation characterization presents challenges in reservoirs that contain tilted water/oil contact (WOC). Tilted WOC discovers hydrocarbon saturation below the free-water level, and different events during geological time can contribute to this specific fluid accumulation. Knowledge of the fluid distribution is needed to understand the mechanisms of oil entrapment, oil volumetrics, and potential recovery mechanisms involved in reservoirs under this wettability and WOC conditions. This case study will describe the workflow used to characterize and model an atypical regime like non-water wet formations in reservoirs with tilted WOC. In this study, a combination of electrical logs, core analysis (lithofacies, poro-perm, MICP), and customized workflow was used to characterize, classify, and map facies. Capillary pressure information and formation tester data were integrated and compiled for each facies. Moving forward, a new method was developed to model saturation height functions representing non-water wet formations and tilted WOC phenomena. Fluid and saturation properties are estimated and assigned to each reservoir point and after reservoir rock types (RRT) were defined. This method has been validated by applying the new approach to actual well data. The drainage capillary pressure (Pc) lab data in the reservoir intervals with established conventional WOC complemented interpretation results derived from acquired logs; however, for the reservoirs zones with identified tilted WOC, correlation and matching Pc lab data with logs was not possible. The new method provides saturation properties in formations with complex fluid-rock interactions and phenomena. This work introduces a novel approach to estimate saturation height functions and saturation distribution for reservoirs with complex fluid-rock interaction and distribution, such as non-water wet formations in tilted WOC conditions.

Abstract Spatial anisotropy and heterogeneity in petrophysical properties can significantly affect formation evaluation of hydrocarbon bearing formations. A common example is permeability anisotropy, which is a consequence of the depositional mechanisms of sediments. Additionally, the variation in spatial distribution of rock components and the effect of post-depositional processes on the physical and chemical structure of the rock constituents can strongly impact the directional dependency of petrophysical, electrical, and elastic properties. Therefore, image-based quantification of spatial distribution of rock constituents can be used for anisotropy evaluation. Assessment of anisotropy has been previously accomplished through use of pore-scale images. However, the discrete nature of this images gives a narrow picture of anisotropy in larger scales. Whole-core computed tomography (CT) scan images, despite revealing the distribution of rock components at a coarser scale, provide a continuous medium for anisotropy estimation. Assessment of anisotropy using three-dimensional (3D) CT-scan data and incorporation of that information in well-log-based formation evaluation is, however, not widely studied or practiced in the petroleum industry. The objectives of this paper are (a) to develop a method to quantify anisotropy utilizing whole-core 3D CT-scan image stacks, (b) to provide a semi-continuous measure of rock anisotropy, and (c) to show the value of the proposed method by means of estimation of directional-dependent elastic properties. First, we pre-process the raw whole-core CT-scan images to remove undesired image artifacts and to generate an image containing pixels representing only the recovered core material. Then, we segment each whole-core CT-scan image stack into distinctive phases. Then, we conduct numerical simulations of electric potential distribution in conjunction with streamline tracing techniques to quantify the electrical tortuosity of the continuous phase in each cartesian direction. We employed the tortuosity distribution values in each direction as a measure of rock anisotropy. Finally, we use a simulation model to estimate direction-dependent elastic properties. We applied the introduced method to dual energy whole-core CT-scan image stacks acquired in a siliciclastic depth interval. Estimates of rock anisotropy obtained using the proposed method agreed with the observed visual distribution of the segmented phase and the observed heterogeneity in available slabbed whole-core photos and 2D CT-scan images. Additionally, estimation of directional-dependent elastic properties demonstrated the value of the proposed method. Anisotropy results coincided with directional-dependent estimation of elastic properties. We observed measurable anisotropy in the 3D CT-scan image stacks, which is important to be quantitatively taken into account in petrophysical/ mechanical evaluation of this formation. A unique contribution of the proposed workflow is the use of core-scale image data for anisotropy estimation and the continuous nature of the anisotropy estimates when compared with workflows employing only pore-scale image data. It should also be noted that the proposed method can potentially be employed to identify the optimum locations to acquire core plugs for further assessment of rock anisotropy.

Abstract Matrix wettability is a key driver in relative permeability and, hence, a critical factor controlling imbibition and drainage at UR fracture-matrix interfaces as well as enhanced oil recovery (EOR). In this study, we (1) adapt and apply the NMR-based wettability index (NWI) methodology of Looyestijn et al. (2006) to a variety of unconventional twin samples undergoing, respectively, spontaneous imbibition with oil-displacing-water and water-displacing-oil and (2) compare the robustness of this method among a variety of samples pairs and also to other NMR-based wettability methods. The samples analyzed cover a range of rock types, major formations, maturity and content of organic material. All displayed unique time-lapse wettability profiles and steady state NWI values. This work advances our previous works (Dick et al., 2019; Kelly et al., 2020) on this subject, where the viability of the methodology was established on end-member pilot samples, towards applicability as a UR SCAL method. The NWI methodology predicts T 2 spectra using linear combinations (mixing) of “end-point” T 2 spectra. The mixing ratios yielding the closest match to the measured spectra are then used to compute a wettability index. These mixing ratios were validated against (1) mass-balance calculations, (2) repeat experiments with heavy water (D2O) instead of H2O and (3) measured T 1 -T 2 maps, enhancing confidence in the robustness of the method. Our comparisons show that alternative approaches representing the T 2 spectra through a single mean T 2 value or T 2 peak-fit, fall short, especially in tight rocks where fast relaxation rate components tend to skew harmonic mean T 2 values and also in samples where oil and water peaks are not clearly resolved. Full spectrum-based methods, akin to Looyestijn’s, appear more robust and stable over a much wider range of reservoir conditions. Repeated NMR acquisition throughout our long-term imbibition experiments shows that time-lapse NWI methodology probes the effects of rock properties, saturation changes, and injected fluid chemistry (enhanced oil recovery strategies) on wettability alteration. Additionally, this NWI study quantifies the wide variation in wettability among unconventional samples.

Abstract The understanding of residual saturation in an oil field in mid-development is essential for estimating the cumulative production achievable, optimizing the future production mechanisms planned for infill targets, development of adjacent reservoir levels and optimizing the design of future facilities. The ACG (Azeri, Chirag, Gunashli) field is a giant oil field located about 120 km offshore in the South Caspian Sea, Azerbaijan. The field consists of multiple stacked clastic reservoirs including the Fasila and Balakhany formations, each with variable oil water contacts, and variable presence and fill level of gas caps. The Fasila reservoirs have been nearly fully developed. Both down flank water injection and crestal gas injection have been employed to drive oil towards producers. These two processes result in different residual oil “trapping” mechanisms which have been explored by logging and coring. Future development of overlying reservoirs can be optimized if we understand the effectiveness of these mechanisms to improve oil recovery and understand produced fluid compositions to enable facilities optimization to handle them. Established techniques to measure the residual oil saturation in a live field depletion, such as conventional open hole logging, pulsed neutron logging and direct core measurements have been employed. This paper investigates the methodology of each technique and the comparison of the magnitude and uncertainty of the saturations obtained. The sands in the ACG main reservoirs are relatively massive and high Net-to-Gross (NTG), however their clay content and distribution is quite variable leading to a range of rock types which behave differently under fluid sweep, and the presence of both intra reservoir sealing shales and lateral sand quality variations lead to a complex pattern of sweep behavior. It was considered that conventional core would be the principle measurement, with the most direct estimation of downhole fluid conditions as well as achieving all other coring objectives. Core was acquired on two pilot wells, one behind the water flood front and another behind the expanding crestal gas cap. Several innovative core analysis techniques were employed. A full conventional log suite was acquired in both wells as well as an open hole pass of a multi detector pulsed neutron log in the crestal gas swept well. The analysis of all this data has led to some interesting conclusions. Previous core flood experiments had led the team to believe gas is more efficient than water in terms of lowering residual oil saturation and reaching higher recovery factors. The new core demonstrated that such low residual oil saturations are achieved more slowly than originally thought, though it didn't change the view of efficiency of gas displacement relative to water. It is also likely that reservoir heterogeneity has had a bigger impact on the variation in residual oil saturation between layers than reservoir quality itself.

Abstract The necessity of knowing formation pressure is crucial to classifying pressure regimes for better understanding in well planning and to de-risk potential abnormal pressure conditions before any future field development wells are drilled, consequently minimizing operational cost. Wireline formation pressure testing has been a useful and reliable technology, that has evolved to confront the challenge of ultra-low permeable reservoir conditions by innovating and improving pump capability, accuracy in pressure measurements, automated control and the implantation of Formation Rate Analysis (FRA) intertwined with an Artificial Intelligent tool. In any pressure testing, the key factor is to be able to withdraw volume from the formation to generate a disturbance on formation pore pressure that a pressure gauge can measure. In the past this has been a difficult task in ultra-low permeable zones. The new generation of wireline tools are capable of withdrawing volume from ultra-low permeable reservoirs, with mobilities lower than 0.01mD/cP. This has been made possible by utilizing control of the pump speed down to 0.0003cc/s which then gives the operator the ability to test ultra-tight formations without the need for inflatable packers. By pulling down the pressure at an extremely low rate and using Artificial Intelligence to control the rate by knowing the formation rate, a proportional amount of volume can be extracted without calling it a tight test. During the operation by observing the rate, and making sure the pump is not oscillating, which indicates the formation rate is lower than the lowest rate the pump can withdraw, the test can be validated for formation flow and the pressure transient of the build – up can be analysed to confirm that at least spherical flow is observed. Once reservoir communication has been confirmed and by analysing drawdown and build-up pressure versus volume withdrawn and implementing the FRA equation, the reservoir pressure can be back calculated by considering isothermal compressibility and FRA slope. This paper highlights the best technical approach to quality check and quality control these tests, showing examples of various wells, where the technique has been used to predict a formation pressure, which can be used for further use for field development, drilling optimisation and production profiles. These pressures would never have been possible using static rates and volume.

Abstract The filtrate contamination cleanup time on a complex carbonate well using a traditional wireline formation tester (WFT) tool can vary from a couple of hours to over half a day. The method proposed aims at reducing operational time to collect a low-contamination formation fluid sample by determining regions with a smaller depth of invasion using a forward model simulation that considers static and dynamic formation properties to predict the radial profile of invasion. The mud filtrate invasion process was modeled considering the static and dynamic properties of the near-wellbore region in an industry reference reservoir simulator, and it integrates three mechanisms for fluid flow: Darcy’s law, material balance, and capillary pressure. The physical robustness of the reservoir simulator was united to a data-driven model to reduce the computational cost. This proxy model is based on a trained neural network with a broad range of scenarios to predict the numerical simulation results with high accuracy. The invasion estimation from the model is then used to predict the filtrate cleanup time using an industry consolidated numerical modeling. One of the variables influencing most of the cleanup time is the depth of mud filtrate invasion. Thus, reducing this time is a determinant for the WFT operational efficiency. The model for mud invasion has been successfully tested on a complex carbonate well, and the results for the depth of mud invasion were comparable to the results obtained with a commercial data-driven inversion using multiple resistivity channels. The estimated cleanup time using the results of depth of invasion predicted by the forward model has been compared and matched with real carbonate sampling stations, and there was a high correlation indicating that zones with lower depth of invasion required less cleanup time. Besides, using the history-matched cases, different WFT technologies such as single and radial probes, focused, unfocused, and dual-packer WFT inlets were evaluated, showing a high potential for reduction of operational time when properly planned and selected for the specific type of reservoir. The proposed methodology is a viable method for understanding the clean-up behavior in different reservoir scenarios using different WFT technologies. The innovation of this method relies on the data calibration using basic and advanced petrophysical properties through a data-driven model based on a trained neural network to reduce the uncertainty in the predicted invasion radial profile and the WFT cleanup time. The reliability on the theoretical results was increased using real data calibration, and this calibrated theoretical model has been used to guide the sampling depth selection, saving operational time.

In this paper, we examine fluids interpretation techniques in a prolific oilfield in offshore West Africa. The reservoir rocks are dominated by Cretaceous limestone, with a small fraction of dolomite and siliciclastic minerals. Due to concerns of radiation hazard, the drilling team has selected a sourceless logging program, consisting of LWD NMR, resistivity and formation tester, to log the reservoir section in 6.5" holes. Therefore, standard log interpretation, which relies on multi-mineral analysis, is no longer viable. The purpose of this study is to answer questions related to asset appraisal and development with these limited measurements. Whole cores were collected from both of the geological structures of the oilfield and the lab measured porosity, permeability, water salinity, Archie m and n and Dean-Stark Sw. Comparison of core and NMR log indicates that NMR total porosity is not affected by hydrocarbon in the pore space. We use a statistical method called factor analysis to deconvolve independent fluid modes, such as clay bound water, capillary bound water and oil+oil-based mud (OBM) filtrate, from the T2 distribution. The number of modes to solve for is determined by principal component analysis. The free fluid T2 cutoff is chosen based on the identified modes. The NMR irreducible water saturation ( S wirr ) computed with this cutoff agrees with Dean-Stark Sw, measured on core samples assumed to be fully invaded by OBM. Continuous Sw is calculated with Archie’s equation with lab-measured parameters and validated against Dean-Stark Sw above the transition zone. The Timur-Coates model is used to estimate matrix permeability, using core-calibrated multiplier and the T2 cutoff from factor analysis. The permeability, Sw and S wirr curves are then used to compute continuous effective permeability to water and oil. The first application of this interpretation workflow is to confirm the free water level (FWL) derived from pressure gradients. We found the Sw profile largely controlled by heterogeneity in rock textures. Good-quality rocks have negligible transitions zones and contain little free water above FWL. Poor-quality rocks have longer transition zones, but the relative permeability to water is too low for the water to flow, as confirmed by production. Pressure depletion suggests excellent connectivity within the reservoir, so these poor-quality rocks are considered a local feature. Log analysis confirms the reservoir-wide FWL, which translates to a significantly increased OOIP over initial estimation. The second application is perforation design. Zones with good porosity and low mobile water volume are selected for perforation and a safe distance is maintained from FWL. As a result, all producer wells exhibit zero water cut.

The calculation with the Archie equation of water and oil saturations from electrical resistivity logs in water flushed reservoirs has historically been challenging. Previous publications on experiments conducted at the laboratory have demonstrated that, in waterflooded reservoirs, the imbibition regime impacts the resistivity index and may lead to an Archie saturation exponent "n" different from the SCAL drainage measurements (referring to the "n" hysteresis between drainage and imbibition observed in oil-wet rock samples). While core, log and reservoir techniques can approach the residual oil saturation (SWTT/NMR/core Sor), they are not commonly used in the realms of a deep offshore field development for operational and cost reasons. This paper presents the results of a case study, offshore Africa, from a highly deviated well that intercepts turbiditic sands (high porosity, oil-bearing) irregularly swept by viscosified polymer-water (injected nearby). Thanks to in-situ sampling of the injected fluid, the water saturation was derived from the sigma log (neutron capture cross section) in order to pinpoint the undisturbed zones and then, by comparison with the resistivity saturation equation, estimate the magnitude of change of the "n" parameter in the flushed zones. Beyond the added value of the so-called old sigma log, which uncertainties and limitations will be discussed, the extra-information gained by knowing both the polarity and the magnitude of change of the "n" parameter gives access to an indication of the insitu change of reservoir properties. The polymer additives present in the injected water may be the cause of the deduced electrical hysteresis and can lead to insightful understanding of reservoir behavior in water flooding context. Such results finally lead to a better estimation of the remaining hydrocarbon in place. INTRODUCTION In development context (water injection, EOR…), the reconciliation between resistivity derived oil saturation and other techniques (sigma, C/O, NMR, Dean-Starks) often seems impossible if using the drainage saturation exponent " n ". But appropriate imbibition measurements are often not available. Then, in case of strong drainage-imbibition " n " hysteresis, gross overestimation of the remaining oil saturation may result from resistivity interpretation in waterflooded reservoirs. Through a comparison between the resistivity derived water saturation and the neutron capture (Sigma) saturation, we first demonstrate that results are in reasonable agreement in pay intervals and aquifers, thanks to favorable conditions regarding this type of electrical logging. We then present in a second example crossing water injected areas the expected remaining oil saturation in best sand facies where no " n -hysteresis" is observed. We then show results from an observer well where sand bodies have been early waterflooded by water with polymer and both resistivity and Sigma logs are available but do not reconcile. By comparing water saturations obtained from the two methods, several observations are made on the sweep efficiency at macroscopic scale and the remaining oil saturation as shown by the neutron capture measurement. If reservoir properties have changed during this full drainage-imbibition cycle (polymer injection), it must be addressed with reasonable hypotheses, amongst those the probable difference in rocks wettability between selected examples but also a possible change of pore surface properties (interfacial tension) possibly due to the polymer.