CHAPTER TWENTY-EIGHT: FRACTURED RESERVOIRS 1
Fracture Identification

Table of Contents
28.00 Introduction To This Chapter
28.01 Definition of Fractures
28.02 General Methods For Identification Of Fractures
1. Drilling Characteristics
2. Sample Descriptions
3. Inflatable Packers
4. Drill Stem Tests
28.03 Fracture Identification From Core Analysis
28.04 Fracture Identification From Spontaneous Potential Logs
28.05 Fracture Identification From Caliper Logs
28.06 Fracture Identification From Micro Resistivity Logs
28.07 Fracture Identification From Dipmeter Logs
28.08 Fracture Identification From Density, Neutron, and PE Logs
28.09 Fracture Identification From Gamma Ray Logs
28.10 Fracture Identification From Resistivity Logs
28.11 Fracture Identification From Temperature Logs
28.12 Fracture Identification From Sonic Logs
28.13 Fracture Identification From Sonic Waveform Logs
28.14 Fracture Identification From Formation Microscanner Logs
28.15 Fracture Identification From Borehole Televiewer Logs

Case Histories (These will open in a new page)
28.16 Classic Example
28.17 Austin Chalk Example
28.18 Fractured Shale
28.19 Vertical Fracture in Vertical Hole
28.20 Vertical Fracture in Horizontal Hole
28.21 Overthrust Example

28.22 In Conclusion
28.23 Exercises for Chapter Twenty-Eight
28.24 Bibliography for Chapter Twenty-Eight

Continue to Chapter Twenty-Nine (includes Fracture Intensity, Porosity, Permeability)
Continue to Chapter Thirty: Dual Porosity Model for Fractured Reservoirs

Publication History: This Chapter is an updated version of part of Chapter Nine of The Log Analysis Handbook - Volume Two, originally self-published in 1990 as a seminar handout and workbook.

CHAPTER TWENTY-EIGHT: FRACTURED RESERVOIRS 1
Fracture Identification

28.00 Introduction To This Chapter
This Chapter covers fracture location and identification from conventional open hole logs. Chapter Twenty-Nine covers straight-forward quantitative and semi-quantitative methods for evaluating fractured reservoirs. The Dual Porosity Model is covered in Chapter Thirty. Rock Stress and Mechanical Properties are the topic for Chapter Twenty. These four Chapters comprise a mini-course in Fractured Reservoir Theory and Practice and should be read as a set.

Natural fractures in reservoir rocks contribute significantly to productivity. Therefore, it is important to glean every scrap of information from open hole logs to locate the presence and intensity of fracturing. This Chapter deals with fracture identification from open hole logs and calculation of fracture intensity and fracture porosity.

Even though some modern logs, such as the formation micro-scanner and televiewer, are the tools of choice for fracture indicators, many wells lack this data. Thus all known fracture location techniques are described.

Naturally fractured reservoirs contain secondary or induced porosity in addition to their original primary porosity. Induced porosity is formed by tension or shear stresses causing fractures in a competent or brittle formation. Fracture porosity is usually very small. Values between 0.0001 and 0.001 of rock volume are typical (0.01% to 0.1%). Fracture-related porosity, such as solution porosity in granite or carbonate reservoirs, may attain much larger values, but the porosity in the actual fracture is still very small.

Fracture analysis literature in the 1970’s suggested that fractures might contribute as much as a few to several percent porosity. More modern work using fracture aperture calculated from resistivity micro-scanner logs indicates much lower numbers. To appreciate this, consider fractures with 1 millimeter aperture spaced 1 meter apart. This gives a porosity of 0.001 fractional (0.1%). This is a very large open fracture. Most are only microns in width, so even 10 fractures of 10 microns each only give 0.0001 fractional porosity (0.01%).

The term “secondary porosity” also includes rock-volume shrinkage due to dolomitization, porosity increase due to solution or recrystalization, and other geological processes. “Secondary porosity” should not be confused with “fracture porosity”. Porosity formed in this way can be determined from modern log suites without difficulty (see Chapter Seven), except for porosity formed by fractures, which is too small to detect with conventional logs.

Fracture porosity is found accurately only by processing the formation micro-scanner curves for fracture aperture and fracture frequency (fracture intensity). All other methods, including the well known “dual-porosity” model, are extremely inaccurate. These models either over-estimate fracture porosity by several orders of magnitude, or cannot be applied because the log data does not fit the model. All published models are described in this Chapter and the student or practitioner can decide whether or not to use them.

The effect of fracture porosity on reservoir performance, however, is very large due to its enormous contribution to permeability. As a result, naturally fractured reservoirs behave differently than un-fractured reservoirs with similar porosity, due to the relative high flow capacity of the secondary porosity system. This provides high initial production rates, which can lead to extremely optimistic production forecasts and sometimes, economic failures when the small reservoir volume is not properly taken into account.

Reservoir simulation software that accounts for the fracture system is often termed a “dual porosity” model. While this is strictly true, it would be better to think of them as “dual permeability” models, since the fracture permeability fed by the matrix or reservoir permeability is far more important than the relative storage capacity of the fractures and matrix porosity. A reservoir with only fracture porosity is quickly depleted; a decent reservoir in the matrix rock feeding into fractures will last much longer.

In order to understand the behavior of naturally fractured reservoirs, estimates must be made of hydrocarbons-in-place within both the primary (matrix rock) and secondary (fracture-only) porosity systems. To do this, we must first be able to detect the existence of fractures. Therefore, this Chapter covers fracture detection from the usually available conventional logs, as well as the method used to partition porosity into primary and fracture components. The effect of this partitioning on the Archie water saturation equation is also described. Modern methods for quantifying fracture porosity directly from micro-scanner logs are also discussed.

28.01 Definition of Fractures
A fracture is a surface along which a loss of cohesion in the rock texture has taken place. A fracture is sometimes called a joint and, at the surface, are expressed as cracks or fissures in the rocks. Figure 28.01 shows the prominent features of a fracture. The orientation of the fracture can be anywhere from horizontal to vertical. The rough surface separates the two faces, giving rise to fracture porosity. The surfaces touch at points called asperities. Altered rock surrounds each surface and infilling minerals may cover part or all of each surface. Minerals may fill the entire fracture, converting an open fracture to a healed or sealed fracture.

FIGURE 28.01: Fracture Porosity Definitions

Fractures are caused by stress in the formation, which in turn usually derives from tectonic forces such as folds and faults. These are termed natural fractures, as opposed to induced fractures. Induced fractures are created by drilling stress or by purposely fracturing a reservoir by hydraulic pressure from surface equipment (see Chapter Twenty). Both kinds of fractures are economically important. Induced fractures may connect the wellbore to natural fractures that would otherwise not contribute to flow capacity.

Natural fractures are more common in carbonate rocks than in sandstones. Some of the best fractured reservoirs are in granite - often referred to as unconventional reservoirs. Fractures occur in preferential directions, determined by the direction of regional stress. This is usually parallel to the direction of nearby faults or folds, but in the case of overthrust faults, they may be perpendicular to the fault or there may be two orthogonal directions. Induced fractures usually have a preferential direction, often perpendicular to the natural fractures. A schematic diagram of these relationships is shown in Figure 28.01, bottom right.

A fracture is often a high permeability path in a low permeability rock, or it may be filled with a cementing material, such as calcite, leaving the fracture with no permeability. Thus it is important to distinguish between open and healed fractures. The total volume of fractures is often small compared to the total pore volume of the reservoir.

Most natural fractures are more or less vertical. Horizontal fracture may exist for a short distance, propped open by bridging of the irregular surfaces. Most horizontal fractures, however, are sealed by overburden pressure. Both horizontal and semi-vertical fractures can be detected by various logging tools.

The vertical extent of fractures is often controlled by thin layers of plastic material, such as shale beds or laminations, or by weak layers of rock, such as stylolites in carbonate sequences. The thickness of these beds may be too small to be seen on logs, so fractures may seem to start and stop for no apparent reason.

To be an aid in production, fractures must be connected to a reasonable hydrocarbon bearing reservoir with sufficient volume to warrant exploitation. If there is no reservoir volume, a lot of fractures won’t help much unless there is sufficient fracture related solution porosity to hold an economic reserve. This can be determined by normal log analysis techniques. In reasonable non-fractured reservoirs, it is usually possible to estimate permeability, and hence productivity (see Chapter Ten), but this is not always possible in fractured reservoirs. Although both the presence of fractures and the presence of a reservoir can be determined from logs, a production test will be needed to determine whether economic production is possible. The test must be analyzed carefully to avoid over optimistic predictions based on the flush production rates associated with the fracture system. Local correlations between fracture intensity observed on logs and production rate are also used to predict well quality.

Sometimes the primary reservoir and the fracture system may be so poorly connected that they are saturated with different fluids. Production from fractures full of hydrocarbons in a water bearing formation may initially be very good but very short lived. A more desirable scenario is a primary reservoir with appreciable hydrocarbon saturation and a fracture system that is full of water close to the borehole, showing invasion and hence good permeability, but full of hydrocarbon in the virgin formation.

This situation can be recognized with fairly simple log analysis techniques. Two examples are shown in Figure 28.02. At lower left is a detailed log of the raw dipmeter curves, showing a conductive streak on pad one, followed by another on pad 3 as the tool rotates as it goes up the hole. This is a single simple semi-vertical fracture filled with conductive drilling fluid in an otherwise non-conducting rock.

On the right side of Figure 28.02 is a Laterolog with a shallow resistivity device (solid curve) showing very conductive streaks on the shallow curve, not seen by the other curves. This is a good indication of fractures or washed out borehole, as the crossover would be the reverse of this in normal porosity. These and other techniques will be described more fully later in this Chapter.

FIGURE 28.02: Dipmeter and Laterolog conductive streaks indicating open fractures

28.02 General Methods For Identification Of Fractures
Most well logs respond in some way to the presence of fractures. Each major log type is discussed in the following sections with respect to its fracture response. Not all logs detect fractures in all situations, and very few see all fractures present in the logged interval. Bear in mind that other borehole and formation responses will be superimposed on each log. Moreover, it is not normal to analyze a single log in isolation, but to review all log curves together to synthesize the best, most coherent, result.

Because we are stuck with the existing logs in the well files, this Chapter covers the assessment of fractures from all these commonly available logs, even though image logs are usually the tool of choice today. On new wells in which fractures may be significant, we would run the correct log suite for fracture identification. Depending on local experience, this would be one or more of those on the following list.

Usually at least two of these would be run for confirmation, but the microscanner or televiewer are often sufficient when run alone. Caliper, gamma ray, porosity, and resistivity logs are usually available as well, so there is no shortage of data!

There are a number of mechanical methods for locating fractures, which will not be discussed further in this Chapter, except for the brief outline given below. There is extensive literature on the subject, especially on well testing, which should be referred to. However, the log analyst needs to be aware of the possibility and confirmation of fractures from these sources:

1. Drilling characteristics: occurrence of lost circulation or mud loss, abrupt drilling breaks, bit bouncing or torqueing, mud weight reduction, well kicks, oil on the mud pit surface, large de-gasser volumes, oil or gas shows on mud logs, calcite in well cuttings coming from fracture incrustations or veins may be indications of fractures. A review of the well history file is an important source of knowledge for the log analyst.

2. Sample descriptions: observation of fractures, slickensides, calcite in healed fractures, blocky or fissile texture may indicate fractures.

3. Inflatable packers: an impression of the borehole wall can be imprinted on the rubber when the packer is set in place. If fractures are present, they will be seen, but there is no way to tell if they were induced by drilling or were present before drilling.

4. Drill stem testing: analysis of pressure transient data from flow and buildup tests has been used extensively to indicate the presence of fracturing.

28.03 Fracture Identification From Core Analysis
Conventional core analysis can provide much information about fractures. Visual observation of open and healed fractures, stylolites, slickensides, fracture density, and fracture dip angle can be made at the wellsite or in the laboratory. They can also be described from core photographs under natural light as in Figure 28.03, and when oil is present, under ultraviolet light.

FIGURE 28.03: Core examination and core description show fractures

If the core itself is not available for direct observation, you may find clues in the core analysis report or core descriptions, where the words fractured, frac, rubble, or lost core are clues to the presence of fractures. Descriptive information may not be transferred into all data bases, so it pays to check the original documents.

A high permeability value in an otherwise low permeability environment is another clue, as is an asterisk in the permeability column, indicating a fractured core sample in which permeability could not be measured. Large differences between maximum horizontal, minimum horizontal, and vertical permeability also may indicate fractures not seen by eye.

The dip direction and strike can be determined, if the core has been oriented with directional data. Cores can also be oriented by comparing observed bedding plane and fracture dips with those from a dipmeter analysis or by paleomagnetic orientation. Bedding plane and fracture plane dip angle and direction are determined by tracing the visible portion of the plane on the core surface with a goniometer, a fancy word for a three axis (X-Y-Z) digitizer. Originally mechanical devices, they are now electronic and interfaced to computers for calculation of the best fit dip plane and subsequent presentation of data listings and displays.

Special core analysis techniques are used in fractured reservoirs, in addition to those normally performed to obtain porosity, permeability, and electrical characteristics. The major technique involves epoxy injection at formation pressure to fill all pores and open fractures. Whole slabs or thin sections are viewed or photographed under plane or ultraviolet light. Pore structure, fracture connections to isolated pores, fracture intensity, width, extent, and direction, and anhydrite or calcite filled fractures are recorded. Results are plotted versus depth and on rose diagrams, as illustrated in Figure 28.04.

FIGURE 28.04: Epoxy filled thin section for fracture location

X-ray tomography of cores, a non destructive, non invasive technique, may see fractures not visible to the naked eye. Open fractures with sufficient width and macro porosity appear as dark (fluid filled) pixels on the computer screen. Both horizontal slices and vertical reconstructed slices can be viewed on the computer screen. Micro porosity and narrow fractures may not be distinctly visible because the resolution is on the order of a millimeter or larger, not as good as the photomicrographs described above. Examples are shown in Figures 28.05 and 28.06.

FIGURE 28.05: X-Ray tomography for fracture detection

FIGURE 28.06: More X-Ray tomography for fracture

Analysis by color coded partitioning of histograms is commonly used to highlight particular features, such as porosity, mineralogy, or invaded fluids. Since the color coding used to represent the X-ray count rate is proportional to density, healed or filled fractures are easily identified when the filling mineral is different than the matrix. Fractures and porosity are shown in black on Figure 28.05, calcite filled fractures as yellow, and dolomite matrix as orange.

Figure 28.06 portrays an anhydrite filled fracture (blue) with vuggy porosity (black) and mud invasion (also blue). The matrix, dolomite, is color coded yellow. Colors will vary under user control, so be sure you understand where the color break points are and what they are supposed to represent. Some ambiguity may exist, as in this last example. Visual examination is used to corroborate problematic situations.

Invasive X-ray tomography, using an aqueous sodium iodide solution to fill pores and fractures, is also used. A pressure sleeve is required, but resolution is better than the native state method, due to the high X-ray response of the injected fluid.

28.04 Fracture Identification From Spontaneous Potential Logs
The spontaneous potential normally does not develop well in carbonate rocks, due to high resistivity and the long distance to a nearby shale. However, some SP excursion is usually seen opposite very porous or permeable carbonate zones, or opposite lower porosity fractured zones. Fracture detection by the SP is possible in a low porosity or low permeability bed if fracturing has occurred and if the fractures contain a formation water of a different salinity than that in the borehole. Development of an SP is not a direct measurement of porosity or permeability.

The SP is a voltage generated by electrochemical reactions between the mud filtrate, formation water, and a nearby impermeable shale barrier. No fluid movement takes place. In addition, a streaming potential can be generated when mud filtrate passes through the mudcake. It is primarily dependent upon mud resistivity and differential pressure. As differential pressure increases, streaming potential increases for a constant mud resistivity. Either the normal SP or the streaming potential can be indicators of permeability and fractures. A streaming potential only exists while fluid is flowing and is not normally seen in a stable wellbore.

An example is shown in Figure 28.06A, left hand side. Depths are in meters and grid lines are two meters apart. The fracture zone below 75 meters is indicated by the shallow resistivity reading significantly lower than the deep. Over the same interval a small SP development is superimposed on a straight line SP with a slight drift to the left. When evaluating SP responses for fractures, remember that the higher the oil saturation (lower water saturation), the more the SP will be depressed. Very small excursions of the log curve may be meaningful.

FIGURE 28.06A: Minor SP development in fractured zone

The SP deflections to the right at 76, 84, and 132 meters may be caused by a streaming potential due to mud filtrate flow into the formation at these depths. This is not certain. Usually streaming potentials are larger and cover longer intervals. These anomalies may be caused by telluric currents, northern lights, rig power bumps, or nearby welding.

Another SP method is to compare the character of the SP to that of a gamma ray log over the zone. If the SP develops in a zone which shows relatively high radioactivity on the gamma ray, this could be an indication of a permeable fractured zone in which uranium salts have been precipitated.

Many factors influence the SP and it is difficult to identify fractures directly using this method alone, but often it aids in confirming the possibility of a fractured zone. Care must be taken not to interpret random variations or drift in the SP baseline as evidence of permeability.

Having detected the fractures, it is useful to count the footage, or meters, of the fractured interval. In the example in Figure 28.06A, the zone from 75 meters to the bottom of the log, or about 95 meters, is fractured. About 16 meters of this shows little resistivity crossover and little SP deflection, giving a net fractured interval of 79 meters.

28.05 Fracture Identification From Caliper Logs
In competent formations, the borehole will often become oblong when it intersects a fracture. This looks like a hole washout on a two or three arm caliper, but a 4 or 6 arm caliper will show the oblong shape. The long axis of the hole is usually parallel to the strike of folds or faults.

This method is best when a sensitive caliper with sharp wall contacts is used. Calipers recorded with most surveys are not very sensitive and serve purposes other than measuring the hole size. Their design does not allow for detection of small abrupt changes. Examples are the three arm bow spring type calipers recorded with sonic and density log which provide centralization as well as hole size measurements. The caliper logs which are most helpful are recorded with the dipmeter, microlog, and modern dual axis calipers on density neutron logs. Special purpose, very sensitive, calipers are available from most service companies.

The caliper recorded with the microlog is designed to float on top of the mudcake. It will respond and measure the thickness of the mudcake, instead of measuring borehole rugosity. The presence of mudcake should be more conclusive of permeability and possible fracturing than rugosity alone. Dipmeter pads are pressured to cut through mudcake and usually measure the rough hole if it is present. Other dipmeter curves are also used to identify fractures.

Mud rings sometimes form even in front of impermeable zones. Therefore, mudcake indicators of permeability must be confirmed with another log if possible.

Rough, large, or irregular borehole in otherwise competent rock usually indicates fractures. Mudcake opposite very low porosity usually indicates fractures. Hole caving due to stress release is very common, but open fractures are not always present. Not all washouts indicate fractures; shale, salt, and unconsolidated sands often erode, but their presence can usually be distinguished by other log characteristics.

FIGURE 28.07: Dipmeter dual axis caliper shows oblong hole in fractured reservoir

A good example is shown in Figure 28.07. Zone A has a round hole, roughly in gauge, indicated by the four arm caliper, and the dipmeter curves show no fractures. Zones B and C show significant hole elongation on the caliper. Fractures are inferred from this and confirmed by the dipmeter curves. Fracture orientation is roughly NE - SW. This information would help determine well spacing; offset wells to the NW or SE would have to be closer than those in line with the fracture orientation.

Remember that a two arm caliper would probably see the long diameter. The enlarged hole is a clue for fractures if the other log curves indicate competent rock. A three arm caliper would average the two diameters, and the hole enlargement may not be as obvious.

Zone E again shows a round hole, this time oversize, indicating a washout in un-fractured rock. This is probably a shale zone, which would easily be confirmed by a gamma ray or SP log. Shales can erode to an oblong hole, especially in deviated holes or in folded or faulted areas.

28.06 Fracture Identification From Micro Resistivity Logs
Micro resistivity logs, such as microlog and micro SFL, indicate fractures by showing low resistivity spikes opposite open fractures, and high resistivity spikes opposite healed fractures and tight or highly cemented layers. In older wells, the microlog, caliper, and an ES may be the only logs available for use in fracture detection, and for porosity and permeability for that matter.

The microlog is one of the most conclusive indications of permeability. When the dotted curve (2 inch micro normal) reads higher than the solid curve (1 inch micro inverse), there is some permeability. This effect is called “positive separation”. The caliper curve accompanying the microlog resistivity measurements is also available for mudcake location and thickness determinations. Mudcake is another indication of permeability.

FIGURE 28.08: Micrologs show fracture location

In Figure 28.08, left, there are three porous and permeable layers with positive separation and mudcake, one near the top and one near the bottom of the log section, and another just below the 7300 foot grid line. The lower layer has a single sharp, low resistivity, spike. This is a fracture or a thin conductive shale streak. The balance of the logged interval is impermeable and probably shale, which could be confirmed by the SP or GR logs.

A newer microlog run in combination with a proximity log is shown in Figure 28.08, right. The permeable zone contains three distinct fractures with several more tiny conductive spikes that could indicate fractures. Only one is seen by the proximity log.

If the micro resistivity curves are smooth, permeability is due to porosity; if low resistivity spikes are present, fractures are indicated. The microlog is a very reliable fracture indicator, but like all single pad devices, it only sees a few of the fractures. If the zone is known to be a carbonate or tight sand, and the hole size is larger than bit size, an elongated hole and probable fractures are indicated.

28.07 Fracture Identification From Dipmeter Logs
High resolution dipmeters with 4, 6, or 8 micro-conductivity log curves, 2 or 3 opposed calipers, plus directional and orientation data can indicate fractures by visual observation of log curve characteristics and from individual dip magnitude and direction calculations. Hole enlargement in a preferential direction discussed in the previous Section, usually caused by fractures, is easily displayed from the multi-arm caliper data, as illustrated in Figure 28.07.

Each dipmeter pad provides a recording of changes in resistivity which occur along the borehole, usually related to porosity variations, bedding planes, or fractures. One pad is selected as reference and its position relative to north is continually recorded. The other pads are numbered clockwise looking from the top down. This determines the orientation of all the pads. Dipmeter tool design is described in more detail in Chapter Twenty-Six.

The log is analyzed in a similar fashion to a micro resistivity log. However, four, six, or eight pads and better focusing make the dipmeter a popular choice in modern wells, because it is more sensitive and covers more of the borehole wall. The more elaborate micro-scanner log has superceded the dipmeter log in many areas and most comments about the dipmeter also apply to the micro-scanner.

FIGURE 28.09: Dipmeter curves show horizontal fractures or bedding planes

Semi-horizontal fractures appear as a short conductive anomaly on all four curves. Examples of these sharp conductive spikes are shown on a 4-pad dipmeter in Figure 28.09. Individual spikes represent bedding planes or semi-horizontal fractures. Fracture intensity counts are made by counting the number of spikes per unit length. Modern thought now suggests that there is no such thing as a horizontal fracture; they are considered to be poorly indurated laminations. Regardless of their proper name, they often contribute to well performance and are easily found with the dipmeter.

FIGURE 28.10: Dipmeter curves show semi-vertical fractures

Semi-vertical fractures usually cause a relatively long conductive anomaly on two opposite pads, or on one pad if the fracture is off axis enough to be missed by the opposite pad. A typical vertical fracture is shown by the shaded portions of Figure 28.10. Additional examples are on Figure 28.07. This kind of analysis is normally done on expanded scale playbacks of the raw dipmeter curves. Notice that the grid lines in these examples are two feet apart, displayed at 1:20 or 1:40 depth scales. Total length of vertical fractures compared to total interval is a useful measure of vertical fracture intensity.

The fracture anomaly may disappear or jump from one curve to the next as either the fracture or the tool rotates around the borehole. An example was shown earlier in Figure 28.02, lower left, where a single vertical fracture appears first as an anomaly on curve 1, then on curve 3 twelve feet above, after the sonde rotated 180 degrees. Remember that the pads see less than 50% of the borehole wall, so the gap in coverage makes the fracture look like two unconnected ones. Because of this, repeat passes should be made in zones of scattered fracturing to provide better detection.

When all four pads show conductive streaks over a long vertical interval as in Figure 28.11, right side, a badly broken, rubble zone can be inferred.

FIGURE 28.11: Fracture Identification Log (FIL) presentation of dipmeter curves

To amplify the fracture detection capability, the dipmeter curves may have to be rerun or replayed with a different scale to show all non-fractured zones as saturated (ie., displaying a constant maximum resistivity). The log should be recorded in the usual way to get the best dipmeter data. Then with the aid of the computer in the logging truck, the curves can be displayed in different formats to emphasize fractures. On existing logs, this can be done in the computer center if the data tapes can be located.

An easy way to analyze fractures with the dipmeter consists of comparing the values of one pad to values of the other pads by replaying adjacent curve pairs on top of each other. The curves are normalized in tight, high resistivity zones. The magnitude of the separation of the curves provides a qualitative indication of the fracture intensity. This visual overlay technique has been dubbed the Fracture Identification Log (FIL) by Schlumberger. An example is given in Figure 28.11, left side, where the shaded areas represent vertical fractures. Many other semi-horizontal fractures and permeable bedding planes are also present, and contribute to production.

FIGURE 28.12: Fracture Identification Log (FIL) in Austin Chalk

Another example, from the Austin Chalk (Figure 28.12), shows the heavily fractured upper zone, the poorly fractured lower zone, and an intervening zone with no fractures. Notice that the bedding planes in the shales look a lot like fractures, so a preliminary screening to identify shale zones is absolutely necessary. People looking for oil or gas in fractured shale can use this technique to great advantage.

The FIL presentation can be made for most types of dipmeters if the data tapes can be located. For older, pre-tape, dipmeters eyeball techniques must be used. The Stratigraphic High Resolution Dipmeter (SHDT), with 8 conductivity curves, is a little more difficult to use than standard high resolution tools. Vertical fractures may influence both electrodes on a single pad. An FIL presentation can be made by turning off one curve from each pad or by presenting two sets of FIL overlays. The GEODIP presentation (see Chapter Twenty-Six for details) may show numerous unconnected bed boundaries in fractured zones. SYNDIP will often show non planar dips or no dip correlations at all (bubble coding).

Since pad orientation is known from the directional data, the fracture azimuth can be determined. This will indicate the preferential permeability direction. The azimuth of pad 1 is recorded directly on low angle dipmeters, but the magnetic declination must be taken into account.

Where:
AHD = azimuth of hole deviation (degrees)
AZ1 = azimuth of pad number one on log (degrees)
FAZ = azimuth of fracture (degrees)
MAGD = magnetic declination (degrees)
PAZ = azimuth of pad one relative to true north (degrees)
PAD# = pad number on which fracture anomaly occurs
RBR = relative bearing azimuth on log (degrees)

Under normal conditions, it is easy to read AZ1 on the log opposite the fracture, add the magnetic declination, and put the result in the range 0 to 360 degrees, using mental arithmetic. Figure 28.11, left side, shows the pad azimuth for a number of fractures showing the preferential direction to be in the range 170 to 200 degrees, or roughly south. Fractures to the north are expected also, but are not as obvious. This may be due to off axis fractures or poor pad pressure in that direction caused by hole deviation or bad tool maintenance.

28.08 Fracture Identification From Density, Neutron, and PE Logs
If the density log shows high porosity spikes that are not seen by the neutron log, usually fractures, large vugs, or caverns exist. Broken out borehole also causes the same effect, but fractures are often present when this occurs. Both cases are shown in Figure 28.13.

Because the density tool only looks at a small fraction of the borehole circumference, only a few of the fractures present will be logged. The depth of investigation is rather shallow, so mudcake and borehole rugosity can have an appreciable effect on the total measurement, despite the fact that it is a pad type contact device with some borehole compensation applied.

FIGURE 28.13: Density log spikes show fractures

Large density correction values in competent rock, especially when weighted muds are used, is a fracture indicator. The fracture network usually does not increase the total porosity appreciably, but the resultant increase in compensation, due to the rugosity, mudcake, or fluid in the fractures, provides an indication of fracturing. See Figure 28.13 again.

Both density and density correction curves show fractures better if the log is recorded with a short time constant. This makes the log look noisy and possibly useless for its normal purpose. The time constant on existing logs can only be changed by reprocessing raw count rate data from the original data tape.

Large PE values, greater than 5.0 cu., especially when weighted muds are used, is a fracture indicator. Barite has a very large photoelectric cross section, 267 as compared with 5.0 for limestone and 3.1 for dolomite. Thus the PE curve should exhibit a very sharp peak in front of a fracture filled with barite loaded mud cake. In Figure 28.14, two very sharp peaks on the PE curve correspond to fractures. The density correction curve also has a bump for the presence of heavy mud. Corroboration from other sources is essential. In light weight muds, an abnormally low PE value, less than 1.7, indicates, fractures, bad hole condition, or coal.

FIGURE 28.14: PE curve shows fractures in barite weighted mud

The compensated neutron log looks at the entire circumference of the well bore, but is usually decentralized to minimize borehole effect. It is not a useful fracture indicator by itself. However, neutron porosity values are often compared with other sources to indicate either lithology or the possibility of fractures.

The sidewall neutron log sees only a small portion of the borehole wall and may be affected by borehole break out in the same way as the density log. Break out is often associated with fractures.

No one would go out of their way to run a density neutron log combination to identify fractures. However, it is the most common log suite run today, and must be used if no other fracture logs have been run. Fortunately the resistivity log, which is nearly always available, can also help identify fractures and this helps confirm density and PE anomalies.

28.09 Fracture Identification From Gamma Ray Logs
The natural gamma ray spectral log provides a quantitative measurement of the three primary sources of natural radioactivity observed in reservoir rocks: potassium, uranium, and thorium. The usual gamma ray log records the sum of these three radioactive sources. This log should not be confused with the (induced) gamma ray spectral log, which is a form of pulsed neutron log run in cased hole to evaluate lithology and water saturation.

Most productive formations show a low content of all three radioactive isotopes. The radioactivity associated with potassium and thorium is normally attributed to clays in the formation. Since uranium salts are soluble in both water and oil, zones of high uranium content indicate fluid movement, subsequent mineral deposition, and thus a probable zone of permeability, usually a fractured zone.

An Austin Chalk example is given in Figure 28.15. Here the upper section is heavily fractured, the middle is not fractured, and the bottom fractured in a few places. In nearly all cases the uranium curve shows high radioactivity at the same depths as the sonic amplitude and sonic variable density log indicate fractures. In some cases, uranium may be present in the porosity without a fracture, as in the shaded portions of the example. In others, there may be fractures with no uranium (see arrow in Figure 28.15). Just as with any fracture location method, there is no absolute guarantee of identifying all fractures.

FIGURE 28.15: Natural gamma ray spectral log (KUT log) shows fractured zones in Austin Chalk

If uranium data is not available, the apparent shale volume from SP, gamma ray, and density neutron crossplot are compared. If the gamma ray derived shale volume is higher than the others, uranium in fractures may be suspected.

Sandstones are sometimes radioactive because of clay or feldspars, not fractures. This can be confirmed by sample descriptions.

In thinly laminated sand-shale series, the zone will appear radioactive due to the shale, but may also contain uranium in fractures. To locate fractures in these beds using the gamma ray method, calculate shale volume independent of the natural radioactivity, then compare this to the actual radioactivity, some of which may be due to uranium in fractures. If a spectral log is available, the assessment is easier.

The natural gamma ray spectral log is one of only three methods that can be used in cased hole to locate fractures. The others are the array sonic and temperature logs, to be described later.

CAUTION: In some areas, fractures are never radioactive, so this method is not always suitable.

28.10 Fracture Identification From Resistivity Logs
On older wells that do not have porosity, caliper, or gamma ray logs, the shallow resistivity log is used to help find fractures. The shallow resistivity log may read the resistivity of drilling mud in washed out borehole sections caused by the presence of fracturing. Check the log heading and compare the mud resistivity, corrected for the temperature of the borehole, with the actual log reading. If they are similar in clean formations, a large borehole may be suspected. Use the SP to find clean zones and use the deep resistivity to check the resistivity of shales or water zones.

Another method, applicable to both old and new logs, is to look for cross over of the shallow and deep resistivity. If mud resistivity is less than the formation resistivity, as is true in many cases, then the shallow resistivity curve will cross over the deep resistivity in a fractured interval and read lower resistivity, due to invasion of the fractures. Normally the shallow curve reads higher than the deep, except in salt mud systems. The shallow curve may also appear noisy or spiky. Review the example in Figure 28.02, right side, and Figure 28.16.

FIGURE 28.16: Shallow resistivity cross over shows fractures

Remember that the deep resistivity logs are averaging 5 or more vertical feet of rock and that the shallow sees about 1.5 feet, so the differences between the two logs is subdued by this. In thinly laminated shaly sands, the cross over is probably due to shale, not fractures. Check the sample descriptions.

For improved resolution, an even shallower focused measurement can be made with a proximity, microlaterolog, or micro spherically focused log, and compared with the deep resistivity log.

All are pad type instruments and survey a smaller portion of the borehole, but all have been successfully used to aid fracture detection. Pad type devices do not see the entire borehole, so only a few of the fractures are logged. However, if the borehole is oval because of fractures, most of them will be seen because they are located on the long axis of the hole, where the pad rides. The sharp conductivity anomalies may, at first be confused with a loss of pad contact. Check to see that the tool is reading higher than mud resistivity.

The dual laterolog has been designed to provide resistivity measurements in wells drilled with highly conductive drilling fluids. However, it is the contrast between the resistivity of the formation, and the resistivity of the drilling mud that is most important. When the ratio of formation to mud resistivity is greater than about 50, better results can be expected from the laterolog than with induction or electrical surveys. Due to the high resistivity of most tight fractured reservoirs, the dual laterolog microlaterolog combination is the preferred resistivity log for such zones.

Focused logs such as the laterolog, spherically focused log, and even the 16 inch normal on old ES and induction logs read resistivity vertically through the formation. Induction logs read horizontally. Therefore there can be a significant difference between the two log readings when vertical fractures are present and invaded with low resistivity drilling fluid. Normally, the induction log ignores any vertical fractures and hydrocarbon filled horizontal fractures. On the other hand a horizontal fracture filled with conductive mud or formation water may produce a rather large conductivity anomaly. When a laterolog or micro device reads less than an induction resistivity in the same zone, vertical fractures are indicated. The induction and ES are not useful in salt mud, so the technique is not available in that case.

If the dual laterolog is used, as recommended above, we lose the advantage of being able to compare the horizontal measurements with the vertical measurements, but the crossover effect due to invasion is still common. Note that recent resistivity literature states that all these tools actually measure vertically, contrary to previous conventional wisdom. The newest induction logs, dubbed 3-D resistivity logs, truly measure both vertical and horizontal resistivity. Comparisons of curve response assists in solving anisotropic resistivity problems, as in laminated shaly sands, and may help determine fracture orientation.

Both methods are more conclusive in thick beds. Since invasion improves the responses, any fracture system detected should be permeable. Thin or high angle beds, along with the normal effects of borehole fluid and size changes on the resistivity measurements, should be considered when using this method.

28.11 Fracture Identification From Temperature Logs
Mud fluid invasion into a fractured zone can lower its temperature. If logged before it can return to the geothermal temperature, the presence of fractures or, at least, invasion can be confirmed. It is possible that the invasion is merely a function of porosity, but usually the effect is smaller than for fractures. Figure 28.17 shows two clear anomalies opposite two fractured zones.

Gas evolving into the mud system, often from tight fractured reservoirs, may be seen if the mud system is static and under balanced for sufficient time. The cooling anomaly should disappear above and below the fracture zone, and will disappear everywhere in a few hours if no additional flow or invasion takes place.

FIGURE 28.17 Temperature log may locate fractures

In perforated cased holes, and in open hole or barefoot completions, an injection profile can be run by increasing pressure on the well head and then logging several passes with a temperature log spaced over a few hours. The pressure will force fluid above the zone downward, injecting cooler fluid into the formation. The larger temperature anomalies are often associated with fractures or the best permeability zones.

28.12 Fracture Identification From Sonic Logs
There is much literature concerning the effect of fractures on acoustic wave propagation in porous and fractured rock. Unfortunately, much of it is theoretical and not always supported by field examples; often it is contradictory. Nevertheless, the sonic log is the best fracture finder in older wells because other, more modern, methods were unavailable at the time.

Today, dipmeter and formation micro-scanner images provide more information, but at higher cost, so sonic logs are still used extensively for fracture identification. The modern full wave or array sonic and dipole shear sonic tools provide much new information, including shear wave travel time and amplitude plus full wave-train digitization This allows the wave train to be further processed.

In theory, the normal compressional interval transit time is little affected by fractures so long as there is a free matrix path between transmitter and receivers, as would be expected for vertical fractures. In practice, large vertical and most sub-horizontal fractures, create cycle skipping on the compressional transit time curve on all sonic logs that rely on detection of the first energy arrival. This is due to reduction in amplitude of the sound pulse by reflection at the fracture face, and by destructive interference caused by other propagation modes generated at the fracture. In addition, refraction caused by near vertical fractures diverts energy from the receivers, again reducing amplitude.

Cycle skipping makes the sonic travel time too long. Thus simple theory is overwhelmed by the complexity of sound transmission in a heterogeneous medium.

28.18: Sonic log cycle skips may indicate fractures

On the array and dipole shear sonic logs, travel time is found by waveform correlation and not by first arrival detection. Therefore, it is less likely to skip a cycle due to low amplitude. Amplitude curves are presented as a matter of routine, so fractures can be identified by low compressional and shear amplitudes. Sonic curves on the array or dipole sonic can disappear or be shown as straight lines where amplitude is too low to obtain a waveform correlation.

Cycle skipping is an excellent fracture indication in hard formations. The example in Figure 28.18 is a good illustration of what to expect. Shallow resistivity crossover might help confirm fractures in a typical well with only an induction and sonic log.

Gas in the formation or in the mud, poor borehole conditions, and poor tool condition or recording parameters, especially on older logs, may also cause skipping. Tool centralization is also important; compressional amplitude can be reduced to less than 20% of normal with the tool only 1 inch off center. This can cause skipping (see Figure 28.19, bottom). Note that most modern sonic logs are designed to avoid cycle-skipping so this identification technique may not be useful in many newer wells.

FIGURE 28.19: Compressional and shear sonic amplitude versus fracture orientation and tool centering

The cause of the skipping can be checked; if the skips occur only in a competent zone and not in the surrounding shales, gas in the formation or fractures are the only possibilities. Cycle skipping is more common on long spaced than on short spaced sonic logs in hard formations, because of lower sound amplitude on longer tools. The reverse is often true in softer sands and shales, due to rock alteration near the wellbore.

In contrast, shear energy is, theoretically, strongly reduced by both horizontal and vertical fractures, but not much by fractures between 35 and 75 degrees (see Figure 28.19, top left). In practice, fluids and fluid filled fractures do not conduct shear waves, and shear arrivals are strongly attenuated in fractured zones (Figure 28.19, right). On full wave or array sonic presentations, the absence of shear arrivals or straight line segments on shear travel time logs are sure signs of fractures. The Stoneley wave amplitude is also strongly reduced by fractures.

To differentiate between fractures and other causes of skipping, a number of different logging tool designs and presentations of sonic data have been developed. Special presentations include the sonic amplitude, sonic wavetrain, and variable intensity (variable density) displays, discussed below.

The sidewall acoustic instrument was introduced to improve the bed resolution and measurement of acoustic properties, but it was not widely available or used. It is an acoustic pad device containing one transmitter and two receivers designed to reduce attenuation in the borehole and through the rock. The distance from the transmitter to the first receiver is 9 inches, and the spacing between the receivers is 6 inches. These dimensions allow for better bed definition for porosity measurements and improved wave forms for fracture studies.

Fractures are more readily identifiable from this short spaced measurement than from devices which measure and average a longer distance. However, the measurement is affected by borehole rugosity and only surveys a small portion of the borehole circumference. It is best suited for thin bedded formations.

28.13 Fracture Identification From Sonic Waveform Logs
The elastic constants of rocks determine the velocity of sound waves (see Chapter Twenty). Compressional waves travel faster than shear waves and theoretically can be seen separately on a display of sonic waveforms. Waveforms are a record of sonic energy level versus arrival time. In practice the two waves, and others, interfere with each other to some extent.

The amplitude of both waves are affected by the rock type, porosity, borehole rugosity, tool centralization, formation fluid, and fracture size and orientation. The fractures may be only those induced near the borehole wall by drilling or may be in-situ. Closed fractures reduce the amplitude less than open fractures. Refracted waves traveling other than direct ray paths can also reduce amplitude and give false impressions of fracturing.

The usual way to record these amplitude values is to present the amplitude of the first energy arrival, which is from the compressional wave, in the form of a log curve, or to present the entire wavetrain, or both. On the newer array and dipole shear sonic, the shear amplitude is also displayed. On older logs, some attempts were made to measure shear travel time and amplitude by adjusting gate times and trigger levels on the instrument panel. These logs were not too reliable, so take care if trying to use them.

FIGURE 28.20: Sonic amplitude log may indicate fractures

Numerous versions of these logs have been developed over the years, with little standardization. Names such as Micro-seismogram, Fracture Finder, 3-D Velocity, Acoustic Parameter, Shear Sonic, Variable Density, and Frac Log were used by various suppliers. We will use the generic term sonic amplitude log to cover all of these.

The sonic amplitude log is a curve representing the first arrival energy, measured in milli-volts. Energy varies with many factors, so absolute values mean little, but low amplitude often means fractures. All the things that cause cycle skipping, described above, cause low amplitude, so fractures are only one possibility. A sample is shown in Figure 28.20. This log is usually combined with a gamma ray, caliper, and a wavetrain presentation, as shown.

The sonic wavetrain log is a display of the recorded energy presented as wiggly trace signatures, usually one for every 6 inches to 2 feet of borehole. The variable intensity display, sometimes called a variable density log, displays the same waveform information, but the amplitude of the positive peaks are shaded gray or black and negative peaks are white. When plotted continuously, dark and light bands representing peaks and valleys are displayed versus depth. Conventions have varied, and arrival time has been plotted increasing right to left or left to right, with the latter used today.

FIGURE 28.21: Sonic waveform presentation

Two waveform logs, with associated gamma ray logs are shown in Figure 28.21. Compressional amplitudes are lower than shear in most cases, but two areas on the left hand log show reduced amplitude on both compressional and shear waves, indicating fractures. Notice that waveform arrival time increases from left to right.

In Figure 28.20, a standard variable density (VDL) presentation is given. Arrival time increases from right to left in this example. Fractured intervals have low compressional amplitude on the amplitude curve, and dark patterns on the waveform at the onset of energy (arrows). The amplitude log also exhibits cycle skipping (S symbol on log).

Reflections from fractures cause changes in amplitude and travel time of the main signal, and some waves arrive at later times, out of phase, thus causing irregular interference patterns on the waveform. Usually chevron patterns spanning several feet can be seen, indicating reflections from near horizontal fractures. Chevrons are difficult to see at best, and are not necessary diagnostic tools. Low amplitude is all that is needed. Other interfering effects, such as Stoneley waves and rough borehole cause jittery patterns. Vertical fractures create less disturbance.

On modern logs, the shear travel time and shear amplitude are recorded, along with complete waveform displays and other diagnostic curves. Stoneley wave travel time and attenuation are also shown. Colour images of the waveform correlation amplitude or colour versions of the waveform display are now common (see Chapter Three for tool details).

Note that single receiver travel time may vary, often indicating poor tool centralization.

FIGURE 28.22: Circumferential sonic log for fracture detection is a rare commodity

A circumferential sonic log has also been developed but was not widely available. Sound pulses travel around the borehole wall and are attenuated most by vertical fractures, due to reflection at the fracture surface. Few examples exist outside the well logging literature. Both the sidewall and circumferential sonic rely on waveform analysis for fracture identification. By alternating between the two transmitters, four separate wavetrain or variable intensity displays are created, one for each quadrant around the hole. Figure 28.22 displays the geometry and output log for this tool.

Evaluation of any acoustic measurement is still complicated because many factors other than a fracture system can cause attenuation or distortion of the wave. Washout zones should be identified before a fracture interpretation is made because they give similar responses. In some shales, the compressive amplitude is larger than the shear amplitude, which again looks like a fractured zone. A gamma ray or SP log should be used to identify such zones.

FIGURE 28.23: Dipole shear sonic logs offer many display options

On the full wave or array sonic log, we can measure travel time and attenuation of the compressional, shear, and Stoneley wave energies, instead of merely the compressional energy as on conventional sonic logs.

FIGURE 28.24: Shear attenuation may locate fractures or vuggy porosity

These attenuations result primarily from the large contrast in acoustic impedance between the rock matrix and the fluid in the fractures and in porosity. As compressional and shear waves traverse a fracture their energies are significantly attenuated with the greatest attenuation occurring to the shear wave. This is illustrated in Figure 28.23 and 28.24. Remember that high attenuation is equivalent to low amplitude. Attenuation is measured in decibels per foot or per meter (db/ft or db/m).

Another cause of energy reduction is poor acoustic coupling in zones with vuggy porosity. This attenuation is due to acoustic wave scattering as it is being transmitted through the vuggy porosity. Analysis of acoustic energies must be supported by porosity information to distinguish this situation. Acoustic energy is not severely attenuated by normal intercrystalline porosity.

FIGURE 28.25: Waveform correlation map shows attenuation in fractures

Suitable processing of the digitally recorded waveforms can enhance the visibility of fractures. One example is to plot the velocity cross correlations to observe the compressional, shear, and Stoneley energy on a time versus velocity crossplot. The peaks of the contoured correlation amplitudes show where the sonic energy is located . Figure 28.26 illustrates a comparison of a fractured and un-fractured zone, showing the loss of shear energy as fracture intensity increases. Note also that the log curve disappears (see left hand track) because no energy is being received at the tool. The gap in the log can be drawn as a straight line. This loss of data is equivalent to cycle skipping on older logs.

Another method involves filtering the waveforms to enhance the chevron patterns caused by mode conversion interference. This is similar to F-K or velocity filtering on seismic data. The dipole array sonic sharpens the chevron patterns naturally, due to the different propagation path of the directional acoustic beam compared to the omni-directional pattern of the monopole array sonic (Figure 28.23, right). Stoneley reflection coefficients, computed from adjacent Stoneley velocities, also help to pinpoint fractures.

28.14 Fracture Identification From Formation Micro-scanner Logs
The formation micro-scanner (FMS) or the newer formation micro-imager (FMI) is an array of electrodes on pads used to produce an electrical image of the formations seen on the borehole wall. On earlier tools, the image arrays were on only two of the four pads, so several logging passes of the tool had to be merged together for better borehole coverage. Using this technique, from forty to eighty percent wellbore coverage could be achieved. Newer tools now have four or eight active imaging pads, reducing the need for repeat passes to obtain 100% coverage of the borehole wall.

In addition to the array electrodes, the tool also has ten standard dipmeter electrodes (8 measure electrodes plus 2 speed buttons) as well as a directional cartridge containing accelerometers and magnetometers for orientation input to the standard dip computations.

The electrical images are made by applying a gray scale to the resistivity wiggle-traces produced from the electrodes on the tool. In this way, low resistivity zones appear dark and high resistivity, low porosity intervals appear white. Since the array on each pad is two and a half inches wide, irregular features, such as vugs and fractures, show up as dark spots and lines on the images. Colour tones may be used instead of grey.

FIGURE 28.26: Formation micro-scanner shows fractures and bedding planes

The image depth scale is usually 1:20 or 1:40, and the X axis is scaled from -180 to +180 degrees around the borehole, putting North in the middle of the track. Examples are provide in Figures 28.26 and 28.27. A dramatic near vertical fracture can be seen in the image at the top of Figure 28.26. Two vertical scales are used: one for reconnaissance and one for detail evaluation. Fracture orientation is roughly NNW - SSE dipping at more than 80 degrees. Other images on these two figures illustrate induced fractures, borehole breakout, inter-bedding laminations, slump brecchia, vugs with fractures, and stylolites.

FIGURE 28.27: Formation micro-scanner shows porosity features sometimes

Fractures or bedding planes can be identified by connecting the linear features to form a sinusoid on the image. The sinusoid can be analyzed to find the angle of dip:

Where:
Y = peak to peak distance of the sinusoid (millimeters)
D = hole diameter (millimeters)

Since Y is measured on a plot or CRT, it must be transformed into actual wellbore distance by multiplying the measured distance by the plot scale. Note also that near vertical fractures will appear near vertical on the plot and do not form sinusoids. Fracture orientation is determined by the azimuth of the sinusoid troughs, read from the scale at the top of the image.

Fractures should produce a higher contrast anomaly than other porosity features because the fractures are flushed with conductive borehole fluid and there is exaggeration of the anomaly due to breakout of the wellbore on the fracture. The fractures are sometimes masked, however, by extremely conductive vugs, so both the gray scale images and the electrical wiggle-trace data are analyzed to identify fractures. Resolution of the micro-scanner is about 10 mm, but contrast between fractures and rock is so good that thinner events, as thin as a few microns, can often be seen.

Micro-scanner images give a very good visual correlation to core and allow the interpretation of small and large scale sedimentary features in the formations. The identification of fractures, along with fracture orientation, and the ability to differentiate them from high angle bedding features is possible.

FIGURE 28.28: FMI log in fractured granite reservoir showing computed dip angle and direction

Further processing of the images to generate fracture frequency and fracture aperture is now routinely applied to the newest formation micro-imaging (FMI) logs. Older logs can be reprocessed for frequency and aperture only if data tapes still exist. The product of frequency and aperture is fracture porosity.

28.15 Fracture Identification From Borehole Televiewer Logs
The borehole televiewer image is similar in appearance to a formation micro-scanner, but uses an ultrasonic derived, directionally oriented, 360 degree view of the borehole wall. Such an image, created by a conventional televiewer, has sufficient resolution to see major fracture systems in good hole conditions. The hole must be round, smooth, and filled with light weight mud to get really good images. The tool must be well centered. These requirements are not met in most fractured zones, but logs are still run for fracture identification and they are useful in many cases. Trade names for these tools are not as well known as others: CBIL (pronounced Cybill) stands for Continuous Borehole Image Log and UBI for Ultrasonic Borehole Imager. Versions of these tools are also used for cement evaluation in cased holes.

The televiewer log of the wellbore is a representation of the amount of acoustic energy received at the transducers, which is dependent upon rock impedance, wall roughness, wellbore fluid attenuation, and hole geometry. For example, a smooth surface reflects better than a rough surface, a hard one better than a soft one. A surface perpendicular to the transducers reflects better than one that is skewed. Therefore, any irregularities such as fractures, vugs and irregular porosity will reduce the amplitude of the reflected signal.

FIGURE 28.29: Older acoustic televiewer log (left) and interpretation image (right)

The resolution of the tool allows us to determine events of about 10 mm in width. Fractures are often accentuated in the wellbore by the drilling process, which breaks out the fracture on both sides of the opening. If it were not for this breakout, most fractures would not be seen by the televiewer as their width is commonly less than 1 mm. An example of an actual image from an older televiewer log, and an interpretive sketch with artificially enhanced resolution, is shown in Figure 28.29. The formation microscanner is much more sensitive to fractures than the televiewer. The electrical conductivity of the fluid in the fracture is 1000 or more times higher than the surrounding rock, compared to about 4 times for acoustic televiewer signals.

In addition to the amplitude image, the travel time image is also recorded on modern logs. This is the travel time from tool to wellbore wall and back to the detector through the mud. This image is effectively an acoustic caliper log, and is used to locate breakouts.

Considerable research is being conducted to enhance the televiewer images, using both arrival time and amplitude of the sound waves, plus computer methods for image enhancement, especially edge enhancement to resolve fractures and bed boundaries. Modern televiewer logs can be used effectively in more rugged boreholes than older versions because of the new processing techniques. Be aware of the age of the log before you start your analysis.

Since the televiewer image is oriented to magnetic north, we can determine the dip direction of a fracture or bedding plane from the azimuth of the troughs of the sinusoid. The dip angle can be calculated from the same equation as given for the microscanner.

CAUTION: The direction scale on the top of the image varies between service companies. One uses a scale with North in the center of the image (same as for FMS and FMI), another puts South in the center.

Figure 28.30 provides a televiewer and core photo of the same fracture. The sinusoidal shape of the fracture trace is very obvious. In this image, South is in the center of the track and the fracture is oriented N 70 E, with a thinner, steeper fracture at N 45 W.

FIGURE 28.30: Core photo (left) and televiewer image (right) of fractured interval

Fracture identification is easiest when several detection methods are combined. This is illustrated in Figures 28.31 and 28.32, where sonic variable intensity and televiewer images are used. If density of the rock is also measured, numerous elastic properties of the rock can be derived, which are useful in hydraulic fracture design and sanding studies. The mathematics for this work is covered in Chapter Twenty.