Over the years reports have been made of anomalous wireline induction log responses in some shale formations. Anderson (Anderson, 2000) suggested an extremely large dielectric effect as an explanation. As shale formations such as the Woodford, Bossier, and Haynesville plays have gained attention as reservoir rocks, it is important that we understand the reasons for these anomalies. This paper examines resistivity wireline logs run in the Bossier Shale, Louisiana with several instruments based on different physical principles which indicate that a large dielectric effect cannot be the only cause of the anomalous responses. The following logs were acquired: Array Induction log, frequency 10 kHz to 150 kHz Multi-Laterolog, frequency 30 Hz to 200 Hz Dielectric log, frequency 200 MHz Multi-Component Induction log, frequency 20 kHz to 220 kHz Accurate tool responses in ‘typical’ shale above the Bossier formation all provide evidence that the tools are functioning properly. A pure dielectric effect would linearly increase with frequency (smallest for the low frequencies and largest for the high frequencies). The observed behavior indicates that the low- and medium-frequency tools show similar anomalous response, while the high-frequency dielectric tool is unaffected. The multi-component induction response, however, shows a strong azimuthal effect where the horizontal ZZ component is always affected. One vertical component (XX or YY) is also affected, while the other vertical component appears normal. The affected vertical component alternates as the tool rotates.
High data quality and improved environmental corrections of array induction logging tools have expanded their operating range into wells drilled with high-salinity water-based mud. Multi-laterolog tools are preferred for these borehole conditions when formation resistivity is high, but if the ratio of formation-to-mud resistivity is moderate, array induction tools can be used successfully. A fairly large region of formation resistivity and formation-to-mud resistivity contrasts exists where you would expect either array induction or multi-laterolog devices to perform well. However, each responds differently to borehole effects, near-borehole effects, and invasion. Data from both array induction and multi-laterolog devices was acquired in an offshore well from China with low-to-moderate formation resistivity drilled with high-salinity water-based mud. In some sections of the well the drilling assembly produced a "spiral" effect with small, periodic changes in borehole size. The majority of the well had no invasion but there was shallow conductive invasion in some permeable intervals. The differences in sensitivity to shallow conductive invasion and the significant effect of a very conductive "spiral" borehole on the array induction responses is illustrated and explained by modeling. A filtering technique to eliminate the spiral effect and improve the responses is described and illustrated. The multi-laterolog tool has no sensitivity to the spiral borehole shape. The determination as to whether an array induction or a multi-laterolog is the preferred resistivity device should be based on more than the value of formation resistivity or the contrast between mud and formation resistivity. In some conditions one has preference, while in other conditions it would be advisable to acquire both.
The Dual Laterolog (DLL) is one of the most commonly used resistivity logging devices in environments with highly conductive (salt-saturated) drilling mud and high formation resistivity. However, this device produces only two apparent resistivity logs with different radial depths of investigation, and it has a relatively low vertical resolution. Several array laterologs have been developed during the 1990s to improve both the vertical and radial resolution of measured formation resistivity. Each of these devices provides several depths-of-investigation measurements. However, most of these measurements, especially the shallow ones, are subject to borehole and eccentricity effects in very conductive muds and large boreholes. These effects can be severe. Therefore, conventional borehole corrections based only on borehole size, mud conductivity, and a fixed value of tool eccentricity (typically used for DLL measurements) become ineffective. A newly developed array laterolog device, Multi Laterolog, is based on the Dual Laterolog principle but with four independent focused measurements. This device is effectively shorter in total length than the traditional DLL and provides higher vertical resolution. Four measurements with different depths of investigation yield a detailed radial resistivity profile. A hardware-based feedback loop is employed to ensure the required focusing. To overcome the problems of the borehole and eccentricity effects on shallow measurements, we have developed a model-based adaptive borehole correction (ABC) technique. It is based on a radial 1-D inversion approach that properly corrects laterolog measurements for the borehole effect including determining and accounting for the unknown tool eccentricity. At every logging depth, we determine four earth-model parameters: tool eccentricity (Ecc), formation resistivity (Rt), invasion zone resistivity (Rxo), and invasion length (Lxo). After that, we apply the model-based correction by modifying the tool response in a borehole with mud resistivity Rm to a response in a borehole with virtual mud resistivity equal to Rxo (or to Rt if the formation is not invaded). This approach guarantees that in uninvaded intervals all four laterolog curves are properly "stacked", while in invaded intervals they show a reliable and consistent resistivity profile. The forward modeling used in inversion is based on the five-dimensional interpolation of pre-calculated look-up tables of various eccentered tool responses. Therefore, the inversion is very fast and the new borehole correction algorithm can be applied in real time. As a by-product, we output the length and resistivity of the invaded zone in invaded intervals. Thus, this correction scheme can provide Rt, Rxo and Lxo for thick beds together with borehole- and eccentricity-corrected curves at the wellsite. We illustrate the advantages of the Multi Laterolog and the model-based adaptive borehole corrections on synthetic data and on field data examples. Introduction The objective of resistivity logging is to determine hydrocarbon saturation from the true formation resistivity (Rt). In complex logging environments, such as thinly bedded reservoirs and invaded formations, the traditional Dual Laterolog measurements usually do not provide sufficient information to extract true Rt. The laterolog technique was first introduced more than a half century ago (Doll 1951), and the Dual Laterolog (DLL) tool was developed two decades later (Suau et al., 1972). Since its introduction the Dual Laterolog has been used as a standard formation resistivity logging device in conductive (salt-saturated) drilling mud and hard rock (high formation resistivity) environments. However, it can produce only two apparent resistivity logs at fixed radial depths of investigation for a given resistivity contrast. The vertical resolution of the deep and shallow Dual Laterolog measurements, RD and RS, is about two feet. Thin beds are assuming more importance as potential reservoirs, and the vertical resolution of RD and RS is increasingly recognized to be insufficient for an unambiguous evaluation of these beds. Also, with just two measurements, RD and RS, it is impossible to solve unambiguously for formation and invasion resistivities (Rt and Rxo) and the invasion length (Lxo). In addition, the Dual Laterolog deep measurement is subject to the Groningen effect when low-resistivity reservoirs are below very resistive beds.
A new wire-line high-definition formation resistivity imaging instrument employing a 'two-electrode' measurement configuration was developed for application in low-resistivity formations drilled with non-conductive (oil-based) mud. The new instrument and measurement principles are described with modeled synthetic responses. Several field log examples with image interpretation are shown.The high-spatial resolution of a 'two-electrode' arrangement has been well documented for nearly 25 years in conductive boreholes [1] and has been used in oil-based mud (OBM) for more than a decade [2]. However, the lower formation resistivity limit in OBM often precluded realizing the benefits in many important offshore deep-water plays, such as offshore East Malaysia and the Gulf of Mexico. The new instrument employs multi-frequency impedance measurements to overcome this low-resistivity low-contrast (LRLC) formation limitation.The imaging device measures electrical impedance across electrodes and produces two image logs: an image of the real part of impedance that is calibrated to yield formation resistivity and an image of the imaginary part of impedance that conveys information about image quality and borehole rugosity. The instrument employs six individually articulated pads, each containing ten sensor electrodes that provide a 79% surface coverage in an 8-in borehole. Typical spatial-resolution is better than 0.8-in vertically and 0.3-in azimuthally, providing a high-definition image of the near-wellbore formation. Pad-to-pad vertical offset is less than 6 inches. This short offset means there is never any pad overlap and guarantees high coverage of borehole surface while the instrument rotates during logging.Presented are field examples from typical logging environments found offshore Malaysia. Detailed geologic features such as turbidite thin-beds, slumping, debrites, faulting and fracturing in low resistivity formations are clearly resolved in highdefinition image logs. The high-definition images deliver information necessary for detailed sedimentological studies, reducing uncertainty when calculating hydrocarbon saturations in LRLC sandstone successions, and provide more precise and accurate net-to-gross calculations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.