Formation imaging aided by combining acoustics, resistivity

Simultaneous rock and fluid properties evaluation

Imaging logs provide detailed information about the rocks in formations surrounding an open hole and about the pipe and casing condition in cased holes. Because most oil and gas wells are drilled using drilling fluids, which do not transmit visible light, the images are not optical.

The most common formation imaging tools either "see" through mud by using sound instead of light, or record electrical resistivity by positioning small electrodes against the borehole wall, thereby excluding the mud column. Imaging logs generate more data than most other logging techniques.

Resistivity images typically contain four orders of magnitude more data per ft per well than a conventional gamma ray log. This enormous quantity of information requires substantial computer power to process and display and the power of the human eye and brain to interpret.

Image analysis joins the power of automatic algorithms to the experience of a human interpreter in obtaining a rich variety of information on petrophysical properties and geologic features in the vicinity of a borehole. In 1995, Western Atlas developed a new openhole tool, the Star Imager (SM). that combines acoustic and electrical resistivity imaging in a single tool.

Acoustic images respond most strongly to rock properties, while resistivity images respond most strongly to fluid properties within the rock. The acoustic image is obtained with a device similar in design to Western Atlas' Circumferential Borehole Imaging Log (CBILSM) instrument and is typically recorded at 250 samples per level and 30 or 60 levels per foot. The resistivity image is obtained with a set of six pads (24 buttons per pad), and is typically recorded at 120 sampling levels per foot.

There are several advantages to recording both types of images in a single run. The most obvious is that the combination offers the full 360° coverage of the acoustic image together with the high dynamic range and resolution of the resistivity image. The use of both types of images reduces the possibility of missing important features such as thin beds, most easily seen in the resistivity image, and superficial features such as vugs and breakouts, better seen in the acoustic image. There are other less obvious advantages to combining the two techniques. To understand these advantages better, let us take a closer look at the flow of formation image data after acquisition.

Processing steps

Imaging data undergoes three major processing stages after acquisition:

Corrections or automatic processing procedures are applied to the data to eliminate the tool geometry and borehole environmental effects, including noise reduction and filtering.
Analysis of features such as bedding planes and fractures are identified and measured on the corrected images.
Interpretation or determination of the origin and meaning of the observed features is made.

Some image data corrections are specific to the type of image while others apply to both. Specific corrections to acoustic image data include:

Despiraling
Despiking
Eccentricity
Ellipticity.

Despiraling eliminates the effects of the tranducer's continuous rotation as the tool moves vertically up the borehole. Despiking removes anomalous acoustic traveltime data. Eccentricity and ellipticity corrections eliminate variations in the traveltime and amplitude data that arise when the tool is not exactly in the geometric center of the borehole. They also provide information about borehole geometry. Resistivity image data corrections include:

Resistivity calculation
Button-balancing
Geometric corrections.

Resistivity calculation yields a value of resistivity from raw measurements, gain data, various recorded tool voltages, and electrical configuration of the tool. Button-balancing eliminates the effects of systematic differences between button readings. Geometric corrections remove the effects caused by the offsets between rows of buttons on a pad, the offset between consecutive pads, the swing-arm effect, and the angular coverage of the pads.

All image data are depth corrected for the tool's irregular movement as it is pulled uphole. Depth correction uses recorded acceleration data to calculate the true depth of the tool for every measurement. The orientation correction rotates the image so that its edge is always aligned with geographic or magnetic north, or with the low or high side of the borehole.

The use of both images results in better depth and orientation corrections. Both images are recorded by different sensors in the same tool string and are separated by about 20 ft. If the tool's movement is not regular, the irregularities will show at different depths in each image before the depth correction and orientation corrections are applied. These irregularities in the image can be used not only to verify the quality of the corrections, but also to improve the corrections beyond what can be achieved with accelerometer and orientation data only.

Data analysis

The identification and measurement of image features combine automatic algorithms with manual editing and picking techniques performed interactively on a computer. The Western Atlas Vision (TM) software package performs these tasks. Furthermore, the software allows the user to display formation images together with conventional logging curves that give a log analyst additional information about the rocks and fluids surrounding the borehole.

The user first enhances the images by applying filters and selecting color scales. Next, significant features such as beds, fractures, breakouts, and vugs are marked. The program automatically calculates and records the orientation of the marked features including the dip and dip directions of fractures, bedding planes, and the orientation of breakouts. Other modules allow the user to extract quantitative properties from the images, which permit physical rock properties to be inferred.

One of the most common tasks of image data analysis is to trace, on a computer screen, inclined planar features such as bedding planes and fractures. These features appear on the acoustic and resistivity images as sinusoid-like curves, if the borehole is cylindrical.

In the case of the acoustic image, the radius of the cylinder is known.

The orientation of the plane can be determined by a simple geometric calculation. The problem with resistivity image devices is that the tool "sees" a variable and somewhat unpredictable distance into the formation, so the radius that must be used in this calculation is not known. This radius varies with depth and the type of feature observed and cannot be calculated reliably.

When dipping features such as fractures or thin beds are recorded, it is possible to calculate their attitude correctly because the acoustic device provides an image taken at the actual borehole radius. Once the attitude of a planar feature is known, it is possible to back-calculate the resistivity image depth of penetration.

The calculation involves simple spatial geometry. Depth of penetration is useful because it tells what rock volume is represented in the resistivity measurement. In conjunction with resistivity measurements from other tools, knowing the penetration depth allows the log analyst to determine the extent to which borehole fluids invaded the formation and the resistivity of the invaded zone.

Yet another advantage of combined image analysis is that fracture aperture and penetration can be better estimated using both images. For open fractures (fractures not filled with mineral deposits), the acoustic amplitude image is more sensitive to fracture aperture, while the resistivity image is more sensitive to fracture penetration. Traveltime can "infer" the open or closed condition of fractures, although it is somewhat erratic. Both amplitude and traveltime data are complemented by resistivity data when interpreting fractures. Filled fractures can be discerned better, and sometimes their fill identified when both images are used.

When trying to identify rock types, the combined interpretation of two physical properties again provides more information than the sum of each independently. Two measurements provide a 2D scale in which data clusters more readily correspond to mineral composition. The improved value is similar to that obtained from conventional density and neutron logs. In addition, formation images provide textural information that is extremely valuable in identifying rocks and estimating their properties.

Data interpretation

Data interpretation constitutes the "geologic objective" of imaging. This step gives geologic meaning to the features identified in the analysis stage. For example, the log analyst can determine borehole stresses from the orientation of fractures and breakouts previously identified in the analysis stage. Also, the analyst can determine the depositional and structural causes of stratigraphic dip directions measured on the images.

Combined imaging presents certain advantages not only in the analysis stage but in the interpretation stage as well. The study of borehole stresses, for example, requires the precise measurement of borehole breakouts - best performed on acoustic images - as well as the identification of fractures, some of which may only be apparent on resistivity images.

In stratigraphic interpretations, the acoustic image can provide information not obtainable by other means, even though the resistivity image is usually preferred due to its higher resolution and penetration. Formation image analysis and interpretation are complemented by core evaluation. Though more difficult to obtain than image logs, cores provide actual samples of the rocks observed on the images.

Coring versus imaging

Cores have some advantages over formation images. They can be examined more closely and their physical properties measured in a laboratory. On the other hand, formation images are easier to obtain than cores; they provide a less disturbed view of the rocks. Cores are usually fragmented during extraction, and their fluid content and stress state changes more drastically than that of the rocks in the borehole wall.

Formation images already contain measurements of some physical properties. Cores and conventional downhole logging measurements have long been used to complement each other. Physical properties measured from cores such as rock type, porosity, and permeability provide a means to improve downhole measurements through proper calibration. In the case of formation images, these calibrated downhole measurements are also far more detailed than would be possible with conventional logs.

Authors

Dan Tetzlaff holds MS and PhD degrees in Applied Earth Sciences from Stanford University. He currently leads the Image Prcessing team for Western Atlas Logging Services.
Eric Paauwe holds BS and MS degrees in geology from the University of Amsterdam (The Netherlands). He joined Western Atlas in 1996 as Senior Geoscience Advisor for the Geoscience and Interpretation group in Houston.

Editor's Note:This article is an abbreviated version of a technical paper appearing in Western's in-house publication, inDepth, (v. 3, no. 1). The full-length version is available from the authors (Houston).