A New Approach for Low-Resistivity Sand Log Analysis

Mathematical modeling for prediction of the distribution and mixing of mud filtrates in the reservoir formation surrounding a well being drilled and the filter cake build-up and its effect on the mud filtrate invasion into the formation, as well as the resulting formation damage effects is presented. The model can simulate the single and two-phase flow situations in the formation with water or oil based drilling mud cases. External particle invasion during the early period of filter cake initiation and its effect on the formation damage by particle migration and retention is considered. An application involving radial flow is given. This model is a useful tool for accurate interpretation of oil-mud-fluid system saturations and resistivity profiles which are essential for the development of efficient well-log interpretation.

A new generation Wireline Formation Tester (WFT) was used to perform fluid sampling in a subject field as part of an appraisal drilling campaign targeting significant undeveloped conventional oil reserves. The field consists of free gas caps with associated saturated oil intervals that appear as thin rims around anticlinal flanks. Resistivity log analysis predicated a mud filtrate invasion volume of 89L. Approximately 150% of the expected invasion volume (128L) was pumped during sampling. Laboratory results showed that the sampled fluid was still mud filtrate, despite pumping an extra 50% of the estimated volume required for hydrocarbon breakthrough. This paper presents a mathematical model for predicting filtrate invasion and post drilling WFT cleanup performance. A new methodology is developed to incorporate critical shear stress concepts into mud filter-cake thickness and filtrate invasion calculations. The mathematical model is used to provide input data for Eclipse reservoir simulator (E100) and an analysis of the field case. The model can also be applied as a predictive tool for filter-cake development, filtrate invasion and pump-out volumes required during fluid sampling. In addition, the model can be used to design optimal mud properties that minimize filtrate invasion and extra treatment of the invaded zone post-drilling. The mathematical model is integrated with reservoir simulation to model the cleanup process. The mathematical model developed in this study predicted an invasion volume of 528L, which is significantly greater than what was estimated based on resistivity logs (89L). This implies that filtrate invasion volume in the field case was underestimated. A sensitivity study showed that filter-cake permeability was the greatest determinant of filtrate invasion volume, and that radius of filtrate invasion was the greatest control on mud filtrate cleanup during sampling. Reservoir simulation provided a good match between model simulation results and field data.

The nuclear magnetism log (NML) ha recently been upgraded to a workable and reasonably trouble-free service. Although the first NML was run commercially in 1960 by Perforating Guns Atlas Corp., the service did not become routinely adequate until 1987. The NML has been one of the great hopes for significant improvement in log interpretation and should now provide the next significant improvement in log interpretation. The NML instrument measures the nuclear magnetism of protons. The proton is a particle identical to the nucleus of the hydrogen atom. The -proton is Mm a magnet; when placed in a magnetic field, it tends to slip with the magnetic field like a compass needle. The NML instrument in its simplest form has a large coil that produces a magnetic field when electrical current is passed through this coil. The p in the formations around the wellbore are aligned by this magnetic field. The borehole (mud) is "doped" with something like magnetite to eliminate the borehole influence. When the magnetic field is released by stopping current flow m the coil, the protons tend to reslip with the earth's magnetic field. During the process of realigning, the protons give back the energy that it took to slip them. This energy induces a signal in the same coil used to create the magnetic field. The process is very inefficient. Free-Fluid Index The most important parameter measured by the NML is the free-fluid index (FFI). The initial magnetization of the formation is a function of the number of aligned protons. This is the FFI, which is the number of protons that are free to be oriented, scaled to porosity units. Protons adsorbed on, or in close proximity to, a solid surface like sand and shale are not free to be oriented. FFI is obviously related to the porosity and surface area of the solid rock material. Intuitively, FFI must then be related to porosity and the amount of water that is coating the surfaces of the reservoir rock-the irreducible water saturation. In rocks, FFI may be thought of as a true effective porosity because it is a measure of the pore space through which fluids will flow. Although irreducible water saturation is influenced by hydrocarbon density (particularly gas), it is a useful parameter. In physical terms, irreducible water on is the water saturation you would obtain in the laboratory if you passed oil or gas through an initially water-saturated permeable rock for weeks or months. The water remaining in the rock is held by surface and capillary forces. Irreducible water saturation is controlled by the pore sizes and the surface area of the rock. Artus proposed that the irreducible water saturation (Siw), porosity (0), and FFI (IFF) were related as ..........................................(1) Siw is usually related to permeability and porosity by one of several empirical correlations. Unfortunately, these correlations require an increase in porosity with increased permeability, which does not always occur. Siw is the parameter that indicates the maximum water saturation a reservoir rock can contain and produce only oil or gas. A high Siw implies that the well will produce few, if any, fluids. A low Siw implies that the reservoir rock can produce hydrocarbons if hydrocarbons are present. Siw is related to effective permeability, which is the permeability to a particular fluid with more than one fluid permit in the formation. This is the condition we always have in the reservoir and is thus the important key to hydrocarbon productivity. A knowledge of Siw makes a good estimate of the fluids to be produced from a reservoir possible. For example when a calculated water saturation, Sw, from a conventional set of well logs is 40% and the Siw is 40%, the interval will produce only hydrocarbons. If, for the same formation, an Siw, of 50% is calculated, you will probably produce hydrocarbons and water. If, for the same formation, you calculate an Sw of 70% or more, you will probably produce water. SW, is an indication of the magnitude of Sw, needed to produce water-free production. It is also indirectly an indication of the median grain size of a granular-type rock (see Hilchie). An independent source of Sw could help define the parameters needed to calculate Sw from conventional logging suites in reservoir where no water zone is present. For example, a conventionally calculated Sw, that is less than Siw in an interval that produced only hydrocarbons indicates a need for better log-interpretation parameters. Fig 1 is a good example. This published log, unfortunately, is not complete (no spontaneous potential), and is apparently a Cretaceous section (probably from Alberta). Each horizontal lineon this metric depth scale represents 5 m (16.4 ft). The sandstone Intervals A and B are the main concern. The formation water resisitivity, Rw, is quoted as 0.2 to 0.5 . m . which is not of much value. The water saturations for Intervals A and B, using the resistivity values from the deep induction and the porosity values from the density and neutron log, calculate about the same, as the resistivities and porosities are about the same. P. 273^

This paper describes the field applications of low invasion coring using waterbase coring muds in several Alaskan and lower 48 sandstone reservoirs. This is a companion paper to SPE 20413 which presented the laboratory development of the tow invasion coring method. High coring rates exceeding 200 feet/hour (61 m/hr) with 100% core recovery were achieved with the new low invasion PDC core bits. In most applications, this technique yields a 2.5 inch (6.35 cm) diameter uninvaded core center in a 4 inch (10.16 cm) diameter core. The invasion of the outer part of the core has been observed in the field as a color change caused by lower oil saturation. Tracers confirm that when coring rate exceeds 80 ft/hr (24.4 m/hr), approximately 80% of the samples taken from the core center have no mud filtrate invasion for permeabilities greater than 50 md. Connate water saturation measured in offsetting wells by oilbase cores agree with low invasion waterbase cores. This demonstrates the possibility of using low invasion coring for log calibration and other reservoir engineering applications where water saturation is needed. These core samples are also useful for laboratory testing to measure relative permeability and other wettability sensitive parameters. Field muds with 5 cm3 API fluid-loss may be used in such coring operations as successfully as more costly oilbase coring fluids. The combination of high coring rate (reduced rig time) and use of field muds (reduced mud cost) can have significant impact on coring cost.

Determination Of Irreducible Water Saturation Using Magnetic Resonance Imaging Logs (MRIL): A Case Study From East Texas, USA Abstract Control of water production and identification of pay zones with high irreducible water saturation are two formation evaluation problems in the United States, the Middle East, and other fields around the world. These problems have delayed completion decisions and have burdened field management with additional expenses. This paper discusses a case study from East Texas where key field problems such as these could not be solved with conventional logs alone. In the subject field, water production is almost impossible to predict from conventional openhole logs. Some zones with relatively low water saturation, as calculated from the logs, produce with a relatively high water cut, primarily because much of the water is movable. However, other zones with high calculated water saturation produce water free hydrocarbons. The difficulty in predicting water production arises from the petrophysics of the producing formations, which are low-permeability, medium- to fine-grained shaly sands. Where grains are small, the formations have high surface-to-volume ratios that result in high irreducible water saturation and, thus, lowered bulk resistivities. Hence, the use of resistivity logs as pay indicators can cause low-resistivity pay zones to be overlooked and, consequently, net field pay to be underestimated. In the last few years, nuclear magnetic resonance (NMR) logs have shown great promise in solving problems of formation evaluation that could not be directly resolved with conventional logs. The capability of NMR logs to differentiate between the immovable and movable water has added reserves of up to 10% in many fields where, although water saturation was high, much of the water was immovable. NMR logs have also been capable of providing better formation permeability than conventional logs, a feature that can save time and expense in well-completion decisions. Because of these two capabilities, it seemed appropriate to apply NMR technology to the subject field. The tool selected for use in the field was an improved version of a mandrel-type, centralized, magnetic-resonance imaging device known as the MRIL@ C-type tool. A series of logs were run in several wells. Accurate determination of movable water from these logs allowed operators to (1) better select pay zones to minimize water production and (2) eliminate the expense of unnecessary conventional logs. The MRIL tool is now run as a regular service in this field. This paper includes examples of MRIL logs, discusses possible applications in the Middle East, and presents recommendations derived from NMR logging experience. Introduction In formation evaluation, induction and resistivity logs are the main hydrocarbon identifiers because of the resistivity contrast between high-resistivity hydrocarbons and low-resistivity formation brine and clay minerals. If, however, a clean zone exhibits low resistivity because of formation brine, these two logs are incapable of indicating water mobility. Because of this limitation, many potentially productive hydrocarbon zones with high irreducible water saturation are bypassed because of fear of excessive water production. A major problem in East Texas is identification of low-resistivity pay zones. Because of the very fine sand grains in this area, formation surface-to-volume ratios are large; thus, formations can hold large volumes of water. P. 155^

The Xu-White clay-sand mixture model simulates the effects of porosity, clay content, water saturation and invasion fluids on sonic wave velocities. Its predictions compare well with observed P- and S-wave sonic logs. P-wave logs can be predicted from a suite of logs including at least {gamma}-ray, density and deep resistivity logs. The model provides three ways of predicting S-wave velocity: (1) from porosity and shale volume, (2) from the P-wave sonic log and shale volume, and (3) from the P-wave sonic log and porosity. Comparison of the results from these methods not only helps improve the prediction but also provides a method of quality control. This paper illustrates application of the model in improving the tie between a well-log synthetic seismogram and seismic data at a well. It is well known that the S-wave log from a full waveform tool is unreliable over soft-formation intervals where the S-wave velocity is lower than the P-wave velocity of the drilling mud. Comparison between the predicted and measured S-wave logs demonstrates that the full-waveform S-wave transit time is severely underestimated in soft shale formations. The V{sub p}/V{sub s} is then underestimated, which was found to significantly affect seismic modelling. The effects of watermore » saturation and mud filtrate invasion were investigated. Comparisons between the measurements and our predictions show that the fit between measured P-wave sonic log and the prediction assuming a 100% mud filtrate saturation is better than that when water saturation has been considered, implying that the propagation paths of the measured P- and S-wave velocities are largely in the flushed zone. The well tie was considerably improved after editing of the sonic log guided by the predicted P-wave sonic and model-based correction of the log for the effect of fluid content. Use of the predicted S-wave sonic for the soft shale formation gave more diagnostic modelling of the pre-stack seismic data.« less

Summary This paper shows that mud-filtrate invasion prior to drillstem testing in low-permeability gas formations can cause significant reduction of gas flow rate observed during the test. In addition, apparent gas permeabilities determined from test data can be much lower than the true values. These conclusions are supported by simulator results and field data. Introduction Interpretation of drillstem tests (DST's) in low-permeability gas formations is difficult. and this difficulty leads to uncertainty in identifying formations that, when stimulated, could become commercial completions. Operators have tong suspected that mud-filtrate invasion into these low-permeability formations could contribute to these uncertainties because of significant alterations in relative permeability and capillary pressure. These uncertainties led us to a study of the effects of mud-filtrate invasion on response in a DST; this paper reports the results of that study. In this investigation, we used a two-phase, two-dimensional, fully implicit reservoir simulator to model DST's. The input data for the simulator were obtained from special core analyses. openhole tops, and DST's from the Falher sandstone in western Canada. Model Description The two-dimensional, two-phase model used in this study has been described previously in the literature. The model simulates the flow of gas and water in the formation using a simultaneous determination of saturations and pressures from finite-difference approximations of the gas and water flow equations. Capillary pressure and relative permeability data are input into the model; the model is therefore capable of simulating imbibition, water blocking, and cleanup effects. These effects are potentially important in DST performance. To simulate a DST, uniform initial pressures and saturations were established in the model for all reservoir cells. To model the imbibition of filtrate into the formation prior to the DST, the pressure in the wellbore was set equal to (and maintained at) the hydrostatic mud pressure, and the water (filtrate) saturation was maintained at 100%. Water saturations and pressures around the wellbore increased during the imbibition period as filtrate moved into the reservoir. To model a mud filter cake, the permeabilities in the formation adjacent to the wellbore were reduced. Following the imbibition period, a flow period was simulated. During the flow period, the pressure in the wellbore was maintained at a fixed pressure and the wellbore volume was set equal to the value of a typical drillpipe open to flow. Following the flow period, a pressure buildup test was simulated, with the wellbore volume set equal to a typical volume below the shut-in valve in a DST tool assembly . JPT P. 299^

Throughout the subsurface of the Arabian Peninsula, the approximately 460ft thick, Devonian Jauf Formation generally consists of well-compacted, low-porosity sandstones and shales, but it also includes friable and highly porous sandstones which form significant gas and condensate reservoir intervals. The mineralogy and pore properties of these reservoir intervals at the Hawiyah field (part of the giant Ghawar structure) were studied by integrating petrographic data with petrophysical measurements of reservoir sandstone samples. The reservoir sandstones are mainly composed of quartz arenites containing small amounts of altered potassium feldspar grains, authigenic illite and chlorite. Based on the pore types, which reflect the habits of the intergranular clays, three reservoir sandstone types have been defined: Type A, characterized by macroporosity; Type B, with microporosity; and Type C, with combined laminations of Types A and B. The dominance of pore-lining clay (as in Type A) or pore-filling clay (as in Type B) is the principal factor controlling the petrophysical properties of the samples. Types A and C sandstones contain macro pores, but irreducible water saturation is high (25 to 45%) compared to clean samples elsewhere, because of the presence of micropores associated with clay. In Type B sandstones the irreducible water saturation is commonly greater than 40% because all the pores spaces are in the microporosity range. The irreducible water saturation in Type B sandstones increases rapidly as porosity decreases. When porosity is less than 10%, the corresponding permeability is 0.2 mD, but no economic production can be expected because water saturation is as high as 100%. In the producing intervals, authigenic clays result in low electrical resistivity due to high water saturation; however, water-free gas is produced.

Oil and gas fields in offshore East Kalimantan, Indonesia have a characteristic of having stacked hydrocarbon pays. Most of oil and gas are produced from Miocene-age Mahakam Delta sediments. Numerous oil and gas reservoirs from several depositional sequences demonstrate low-resistive pay characteristics. The low resistivity can be attributed to the presence of conductive water, clays or the effect of thin bed. This paper presents two approaches, to be run in addition to the conventional wireline logs, in order to help in concluding the questionable pays. Employing one or both approaches can vastly improve the interpretation of low resistive formations. Both methods have been tested and tried in more than 30 wells, with the majority of the wells used only the new wireline formation tester tool. Utilizing new measurements from nuclear magnetic resonance tool is the first approach used in the evaluation of low-resistive reservoirs. The tool can provide the information about the amount of irreducible water and movable fluid. The data are then combined with log interpretation from conventional wireline data to determine the amount of hydrocarbon more accurately, and to find more producible oil and gas behind the pipe. With this data, a relationship between low resistivity and irreducible water saturation can be possibly made. The second approach is the use of a new wireline conveyed formation tester tool. This new tool can identify formation fluid downhole. It eliminates the necessity to bring fluid samples to the surface for hydrocarbon confirmation, therefore the formation evaluation time is greatly reduced. Presence of oil, water or gas - interpreted from the resistivity reading, temperature and infrared spectrometry - is monitored in real time at the surface. Introduction Unocal Indonesia Co. is operating several offshore oil and gas fields (i.e. Attaka, Sepinggan) in East Kalimantan, Indonesia. A typical characteristic of the fields is the presence of numerous sandstone reservoirs from 2,000 feet to 10,000 feet subsea depth. The reservoir thickness ranges from several inches to more than 100 feet. It's not uncommon to have a well with tens of reservoirs with a combined hydrocarbon pay of more than 100 feet. Evaluating all reservoirs during logging usually do not resolve the fluid saturation of all of them. Wireline logs provide a starting point in the formation evaluation. However, the formation evaluation is complicated by the presence of numerous low-resistive reservoirs, among other things. Getting surface sample of all the questionable pays is not practical nor economical considering the expensive rig time and the logging cost. The objective of the approaches is to minimize the bypass of low-resistive pay zones and to maximize the recovery of oil and gas from the low-resistive formations. Geology of East Kalimantan Fields Unocal's fields in East Kalimantan, Indonesia are located within the Kutei Basin which is characterized by the Samarinda Anticlinorium, a series of northeast-southwest trending anticlines. The basin is surrounded by Kuching High in the north, a rift with North Makasar Basin in the east, Adang Fault as southern boundary with Meratus Uplift and Paternoster Platform and Barito Shelf in the west. Major hydrocarbon traps are generally controlled by the structure, closures and faults. The productive reservoirs are mainly sandstone of Tertiary sediments of Middle Miocene to Pliocene age. Oil and gas have been produced primarily from the deltaic sands which are normally thick, clean, blocky and highly resistive. P. 173^

Modeling of porosity and permeability for hydrocarbon Exploration: A case study of Gongola arm of the Upper Benue Trough

The Yamasaki fault is a transcurrent fault extending over a distance of 80km in the western part of Japan along which microearthquakes occur densely. Electromagnetic investigation has been conducted in this area. The results of electrical resistivity measurements conducted during the period from 1975 to 1978 are presented in this paper.Resistivity measurements were made by several different methods, such as direct current methods and electromagnetic induction techniques. It has been revealed by the direct current methods that there is a belt along the fault in which the electrical resistivity is low, ranging from 100 to 400ohm-m, whereas the resistivity is higher than 10, 000ohm-m far outside. This belt is about 1km wide at some places. Inside this low resistivity belt, there run streaks of extremely low resistivity, such as 20ohm-m. The locations of these streaks correspond to the surface traces of the fault plane. Magnetotelluric investigations for a frequency range of 8 to 20Hz have indicated that the low resistive belt is circumscribed by a zone with intermediate resistivity of 1, 000ohm-m. Outside the area, on the other hand, the resistivity is higher than 10, 000ohm-m. From the investigations so far conducted, it may be said that the low resistive zone has a three-fold structure consisting of an inner low resistive belt of 100 to 400ohm-m, the outer zone of 1, 000ohm-m, and narrow streaks of very low (20ohm-m) resistivity inside the inner belt.The low resistive zone is suggestive of existence of a fractured zone with high water content along the fault. The difference between the low and the relatively low resistivity zone might arise from difference in degree of fracture and degree of water content. On the other hand, extremely low resistivity streaks perhaps indicate the surface traces of the fault plane along which clay mineral saturated with water has been developed.

The increased usage of the FoRxo, Guard and gamma combination of logs in salty muds has led to the development of interpretation techniques which are fast and simple. This is possible because:The effects of bed thickness on both FoRxo and Guard logs are minimal and in moot cases may be neglected.With salt muds, true formation resistivity, Rt, is linearly and directly proportional to recorded Guard log resistivities over a wide range of conditions.Likewise, invaded zone resistivity, Rxo is approximately equal to recorded FoRxo resistivities. This direct reading of these two vital parameters makes possible a fast but accurate visual location of all zones of interest. The same characteristics of the logs also make possible simplified interpretation curves to calculate both porosity and wager saturation. Inasmuch as shaly sands are best measured without the disturbing influence of fresh waters, their interpretation is greatly enhanced by this system of logs. For this technique to be used, the following conditions must be fulfilled:The zones of interest must be invaded by mud filtrate.The ratio of resistivities of mud filtrate to connate water must be known and the ratio must not exceed 10.Mud filtrate resistivity must be known.Mud cake thickness must be small. Field examples are given which exhibit the inherent features as well as the disadvantages of the proposed techniques. Simplified charts for interpretative procedures are also presented. In conclusion, the FoRxo, Guard and gamma logs provide a detailed log of the lithology from which porosity and water saturation may be calculated in detail by a simple interpretation method when certain conditions exist. The logging system also provides an improved method of interpreting shaly sands when both formation water and mud filtrate are highly conductive.

A core study was conducted to investigate the return-permeability and changes in the irreducible water saturation in Sandstone and Carbonate cores, when exposed to filtrate from actual oil-based and water-based drilling mud. Results indicate that the oil-based filtrate contains sufficient oil wetting surfactants and emulsifiers to reverse the wettability of rock surfaces from water wet to oil wet. The oil wet rock surface in the low permeability matrix, reduced the effective oil permeability drastically. In the high permeability matrix, the change in wettability to an oil wet system reduced the irreducible water saturation, resulting in an increase in the effective oil permeability. This effect was observed for Sandstone as well as the carbonate cores (1). Clay swelling, dispersion and ion exchange combined with water blockage are believed to control the changes in flow properties for the Sandstone core samples. The introduction of colloidal particles and the dissolving of calcite are believed to be the mechanisms controlling the changes observed in the flow properties for the carbonate cores (1).

Summary This paper provides a new workflow for delineation and evaluation of low-permeable and low-resistivity reservoirs of clastic formation. These reservoirs were previously considered as water-bearing, however well-tests showed significant oil inflow from them. Thereby, this research is aimed at more detailed study of the formation based on new approaches to rock typing and oil saturation assessment. In order to perform core-derived rock typing and characterization we applied alternative FZI, calculated as function of porosity and irreducible water saturation, together with modern techniques for microstructural analyses, such as NMR and CT. That also enabled to explain the formation’s low permeability and low resistivity. For well logs interpretation we proposed technique for permeability calculation as a multivariate parameter. The resulted permeability log matched very well with core-derived permeability. Application of supervised ML algorithms, supported by thorough petrophysical interpretation, allowed reliable prediction of log-derived rock types in other wells, where preliminary interpretation was not performed. The oil saturation was assessed using Archie, Dual Water and Waxman-Smits equations, incorporating cation exchange capacity for each rock type. The developed methodology enabled detection of previously missed productive intervals, that resulted in increase of the productive thickness by 4 times and linear hydrocarbon reserves by 3 times.

Electrical properties are important parameters to quantitatively calculate water saturation in oil and gas reservoirs by well logging interpretation. It is usual that oil layers show high resistivity responses, while water layers show low-resistivity responses. However, there are low-resistivity oil zones that exist in many oilfields around the world, leading to difficulties for reservoir evaluation. In our research, we used digital rock technology to study different internal and external factors to account for low rock resistivity responses in oil layers. We first constructed three-dimensional digital rock models with five components based on micro-computed tomography technology and x-ray diffraction experimental results, and then oil and water distributions in pores were determined by the pore morphology method. When the resistivity of each component was assigned, rock resistivities were calculated by using the finite element method. We collected 20 sandstone samples to prove the effectiveness of our numerical simulation methods. Based on the control variate method, we studied the effects of different factors on the resistivity indexes and rock resistivities. After sensitivity analyses, we found the main factors which caused low rock resistivities in oil layers. For unfractured rocks, influential factors arranged in descending order of importance were porosity, clay content, temperature, water salinity, heavy mineral, clay type and wettability. In addition, we found that the resistivity index could not provide enough information to identify a low-resistivity oil zone by using laboratory rock–electric experimental results. These results can not only expand our understandings of the electrical properties of low-resistivity rocks from oil layers, but also help identify low-resistivity oil zones better.