e-dipmeter surveys e1-general principles Introduction
Since its introduction in the 1930s, the dipmeter di pmeter tool has found steadily increasing application in the petroleum industry. Used initially in exploration, the tool helped to locate and identify the major features of geologic structure serving as oil traps. As techniques became more refined and interpretation i nterpretation became more secure, the dipmeter’s range of applications expanded, making it the principal logging tool for describing internal lithologic features and the sedimentological processes responsible for them. The current emphasis on investigating sedimentary bedding conditions has further enhanced the utility of the dipmeter log. The high sampling density of 120 readings per foot of borehole depth makes the t he dipmeter tool virtually the only logging device that can supply the petroleum geologist with detailed i nformation on finestructured sedimentary beds in the subsurface. The dipmeter tool measures conductivity or resistivity changes, hole size, and sonde orientation-nothing more, nothing less. It does not directly measure the dip of bed boundaries or the dip of lithology li thology changes. The conductivity changes are input into a computer program that correlates the recorded wiggle traces and computes apparent dip from the correlations. Computed dips are then corrected for sonde til t and converted into true dips. The true dips are plotted pl otted and used to make inferences of structural dips, bed geometries, and depositional environments. Dips displayed on the tadpole or arrow plot pl ot result from a combination of the original depositional dips, differential compaction and structural rotation during subsidence, and postdepositional deformation. AS is true with other logs, information i nformation other than that contained on the dipmeter log is required to make the best interpretation. The minimum mi nimum required input from the geologist is to describe missing sections and depositional environments. The more information available, the better the dipmeter interpretation. The dipmeter tool operates on the following principle. A bedding surface cutting across a borehole at some angle causes microresistivity changes to be recorded at different depths on the individual dipmeter curves, which are recorded from electrodes on pads located at various circumferential positions around the borehole. Figure 1 shows a borehole intersected by a steeply dipping, thin resistive bed.
1
Figure 1
Note that as the four pads ascend the hole, each measure electrode contacts the thin bed at a different elevation, giving gi ving rise to displacements, or shifts, between curves. The depth differences, or displacements between the curves, depend upon the dip magnitude and direction, or azimuth, of the bedding beddi ng surfaces. Mathematical correlation methods are applied to measure these displacements, either individual features or short intervals being matched together. The dip and azimuth of t he bedding can then be computed, and corrected for the effect of the deviation of the borehole. It should be noted that formation dip di p computations with the conventional 4-curve tool require that a bedding plane pl ane be crossed by at least three of the four pads, since three points are needed to define a plane. pl ane. This creates the constraint that pad-topad correlation must be established between the resistivity curves recorded by at least three of the four pad electrodes. el ectrodes. Generally, in well-bedded or laminated formations, the recorded data allow the determination of formation dip and azimuth. Pad-to-pad correlations are limited for many stratigraphic studies, however, because of the fi ne detail associated with sedimentary features. Eight-curve and microelectric scanning tools incorporate a
2
number of major improvements over the 4-curve tool to overcome this limitation, and are specifically applicable to sedimentary studies. Although the newer tools are replacing the 4-curve tool, many hundreds of t he 4curve logs have been run in the past and will continue to be used for geologic and production studies. Therefore, for completeness, the 4-curve tool and field log will be discussed first and the 8-curve dipmeter and formation imaging measurements will be covered later in more detail.
Tools Available
A number of dipmeter tools are available. Three-arm dipmeter tools were used for many years, but these have now been entirely superseded by four-arm and six-arm tools. Figure 1 illustrates a commonly used four-arm dipmeter tool.
Figure 1
All currently used dipmeter tools have the following common characteristics: characteristics: · the orientation section measures tool deviation from vertical, tool azimuth with respect to north, and the orientation of the reference electrode pad to either north or the low side of the hole
3
· the caliper section measures two or more hole diameters · the microelectrode array records the resistivity of the formation in the very localized area where the pads contact the formation; · the gross correlation device, such as a moderately deep resistivity curve or a gamma ray or SP curve Until recently, orientation was measured using a pendulum to indicate deviation from vertical and a magnetic compass to indicate tool rotation relative to magnetic north. Recently introduced tools use flux gate magnetometers, gyroscopes, and/ or accelerometers to deduce the tool position and orientation. The microresistivity pads carry small "button" electrodes for water-base muds and "knife-edge blade" electrodes for oil-base muds, although the latter are not always very effective. In the field the norm is to supply a 5-in. print of the orientation curves, the correlation traces, and the caliper curves. All data are recorded on magnetic tape. On the rare occasions when it may be desirable to compute di p results from the film rather than from the digital data tape, a film on a very expanded scale (60 in. = 100 ft) is required. Figure 2 illustrates the far more detailed 60-in.
4
Figure 2
dipmeter presentation. The 4-Curve Dipmeter Tool
The 4-curve device uses four identical microresistivity electrodes mounted on four pads. The four caliper arms are actuated hydraulically from the surface with a force sufficient to maintain good pad contact with the borehole wall under most conditions. The resistivity measurements are sampled 60 times per foot, or every 0.2 i n. Such detail is essential, because even 1° of structural dip may be significant in determining the location of hydrocarbon traps. A 1° dip across an 8-i n. borehole causes a shift of 0.14 in. between curves. The electrodes are small enough to resolve fine structure with linear dimensions down to about 0.4 in. (1 cm). Because dipmeter correlations depend on variations in resistivity, the circuitry for the electrode output is arranged so that the curve deflections are proportional to the electrode current. Current varies widely according to the contrast between the resistivity of the formation in front of the electrode and the formation surrounding the sonde. The curves are recorded with a "floating zero" on a nonlinear scale designed to accommodate large variations in local resistivity.
5
Figure 3 shows the four primary dip curves.
Figure 3
On this expanded depth scale, it is apparent that a consistent shift occurs between any two curves. The shifts in this case result from bedding planes intersecting the borehole at an angle of approximately 30°. This angle i s the dip with respect to a plane normal to the instrument axis. The cable speed at the surface is measured, but the velocity of the downhole tool may be different and may alternately accelerate and decelerate with changes in friction because of the elastic properties of the cable. It is important for purposes of dip computation that the instantaneous velocity of the tool be known throughout the logging run. A fifth electrode (known as the speed button) provides for this correction. The curve recorded by this electrode should very closely correlate with the curve recorded by the electrode mounted below it on the same pad, and thus yield a displacement equal to the separation between them. However, if the instantaneous tool velocity varies from the constant surface cable speed, this apparent displacement also would vary, and velocity corrections must be made.
6
Without knowing the orientation of the tool in space, the computed dip would be the slope of a geologic feature relative to the plane defined by the four resistivity pads. To convert this angle to true dip, three continuously measured angles are required: · deviation of the tool from the vertical (inclination) · hole-drift azimuth · azimuth of Electrode No. 1 from magnetic north The deviation and the first of the two azimuths are measured directly. A relativebearing measurement is also made (the angular rotation about the axis of the tool of Electrode No. 1 from the upper generatrix of the hole), and it is from this angle and the azimuth of Electrode No. 1 that the hole-drift azimuth is computed. The relationship is: hole-drift azimuth = azimuth pad 1 - relative bearing Deviation and relative bearing are measured with pendulum systems, and the azimuth of Pad 1 with a magnetic compass. True north is the reference for the orientation of the tool. True north and magnetic north are frequently different; this difference is referred to as magnetic declination. Maps showing current values of magnetic declination are available. At point A on such maps, magnetic north is 20° east of true north; therefore, 20° must be added to the magnetic-north bearing to obtain the orientation of the tool with respect to true north. East declination refers to conditions in which magnetic north is east of t rue north. East declination requires that the declination value be added to the magnetic-north azimuth measurement to obtain orientation with respect to true north. West declination refers to conditions in which magnetic north is west of true north and requires that the declination value be subtracted from the magnetic -north azimuth measurement.
True dip magnitude and the downdip direction with respect to true north is calculated from all of the previously mentioned acquired data-i.e., di p curve shifts, caliper measurements, deviation, deviation azimuth, and azimuth of Pad 1. The 4-Curve Dipmeter Field Log
At the wellsite, a field monitor log is recorded for each run of the tool . By carefully monitoring the four dip correlation curves on this log, the field engineer can ensure the reliability of the final computation. The log heading provides a review of definitions of the various angles measured and calculated for the tool. The choice of low-angle or hi gh-angle unit affects those definitions and calculations. The low-angle unit is for holes as much as 36° from vertical, the high-angle for holes up to 72° from vertical.
7
The angle called azimuth is: · the clockwise angle between magnetic north and the horizontal projection of the arm carrying the reference electrode (No. 1) for a low-angle unit. · the clockwise angle from north to the horizontal projection of the axis of the tool-called DHD on the log-for a high-angle unit. The relative-bearing angle is measured clockwise from the high side of the tool to the reference electrode. Azimuth and relative-bearing traces should move roughly parallel to each other in a low-angle unit. The major part of the log, the right-hand side, contains the four correlation curves. The log heading shows the relative position of each curve and i ndicates the direction in which resistivity increases. On the far right-hand side of the log are the two caliper curves, showing the hole diameter between Pads 1 and 3 as a dashed line and that between Pads 2 and 4 as a solid line. The depth scale appears in the center column of the field log. Definitions of Formation Dip
The dipmeter survey records ways in which subsurface layers of rock have been deposited and subsequently moved. The raw data consists of orientation information, showing where the downhole tool is located with respect to vertical and geographic coordinates; and correlation information, used to determine the attitude of bedding planes with respect to the tool. The field log does not indicate formation dip. Computer processing of the raw data is required before any geological information can be extracted. The two important computer-processed parameters, bed-dip magnitude and dip azimuth, yield a great deal of valuable information when studied with regard to how these parameters vary with depth. Dip angle is the angle formed between vertical and a normal taken from a bedding plane. Thus, a horizontal bed has a dip of 0° and a vertical bed has a dip of 90° (see Figure 1 ).
8
Figure 1
The dip azimuth is the angle formed between geographic north and the direction of greatest slope on a bedding plane. Dip azimuth is conventionally measured clockwise from north, so that a plane dipping to east has a dip azimuth of 90°, and one to west 270° ( Figure 2 ).
9
Figure 2
Dipmeter surveys have a variety of applications. At the lowest level, the raw data may be used to compute (1) a deviation survey, (2) true vertical depth, (3) the integrated hole volume (as an aid to fracture detection) and (4) thin -bed definition. At the intermediate level, computed dipmeter results may be used to determine the gross geologic structural features crossed by the wellbore, sedimentary details within sand bodies, the depositional environment, and true stratigraphic and vertical thicknesses. At the highest level, computed dipmeter results from many well s may be combined to produce structural cross sections and trend surface maps. The most important applications of the dipmeter survey are in exploration drilling, t o help identify local structure and stratigraphy, and in development drilling, to h elp map the productive horizons and indicate direction to follow for further fi eld development. Introduction
The primary, and sometimes the only, use of a dipmeter is for determining structural dip. Structural dip is the attitude of the formations resulting from tectonic
10
movements. Structural dip information might be used by the geologist for possible whipstocking or deviating the present well or to locate a future well updip or downdip. Structural dip determination from logs is not always obvious. It is possible to have two equally plausible trends; when this occurs, additional information is necessary to determine the most probable trend. In tensional areas, such as the U.S. Gulf Coast, offshore West Africa, and portions of the North Sea, structural dip consists of a dip trend extending at least a thousand feet. The trend would remain constant or change gradually, unless a fault or unconformity is crossed. Thrust provinces tend to exhibit more stages of local structural deformation than tensional areas. This increased structural deformation is due to tectonic or major erosional events, and it negates the thousand-foot structural dip rule. As a general rule, structural dip extends horizontally no farther than it does vertically. When determining structural dip, use the trends with the greatest vertical extent. In addition to green groups, which indicate structural dip, red and bl ue groups are also useful for determining the direction of structural dip. Red and blue groups are particularly helpful when dip magnitude is low (about 1° or 2°). At low angles there is often a choice of trends; the most probable trend matches the majority of red and blue dip groups. Low-energy environments allow deposition of horizontal sediment layers. The dip of layers that have undergone only structural uplift indicates the structural dip. Determining Structural Dip
To determine structural dip from an arrow plot, examine the reduced scale tadpole plot for zones of low dip scatter. Use either the 1-in. or the 2-in. scale. The zones of least scatter are derived from sediment layers deposited in low -energy environments, and they produce dips indicating the structural dip. From the zones of least scatter, pick a dip trend extending as far vertically as possible; this is the approximate structural trend. Next, use the 5-in. scale to determine the exact dip magnitude and azimuth of the trend ( Figure 3 ). Dips plotted on the reduced scales are pooled; therefore, any trend determined from the 1-in. or 2-in. plots would be slightly in error.
11
Figure 3
Unless the logged section is short, there may be several structural dip trends on the log. Structural dip changes indicate sections missing due to faulting or unconformities, or indicate the end of periods of postdepositional uplift. It is important to determine the exact location of dip changes. Sometimes the point of change can be determined exactly; in other conditions it may be di fficult or impossible to determine. One technique for locating points of change is t o determine the obvious dip trends above and below the point of change, then extend both trends toward each other as far as possible using isolated dips for support ( Figure 4 ). The point of change is located between the two extended trends. This technique does not l ocate the exact point of change, but it does better define the zone in which the change occurs.
12
Figure 4
Hole Deviation as a Dip Indicator
Hole deviation may be used in some instances as a dip indicator. The hole tends to drift or walk when dip is present. The following general rules can help in i dentifying structural dip. Compacted Formations: Compacted formations cause the bit to walk or drift updip in a hole drilled with mud. Updip drift occurs as the bit attempts to align perpendicular with the dip of the bedding planes. When air or gas is used for drilling, the hole usually drifts downdip. Uncompacted Formations: Less compacted formations are more complex, but in uncompacted formations the hole generally drifts downdip. In one offshore area the hole drifts downdip to about 6000 ft, then clockwise along strike as the zones become more consolidated. The clockwise drift results from bit rotation. Near 12,000 ft, the bit encounters more compacted beds, and the bit drifts updip. Unless controlled, the hole follows a U-shaped path.
Prior knowledge of hole-drift tendencies can save rig time; the surface location can be offset relative to the proposed bottomhole location, reducing the need to control the parameters that affect drilling rate.
13
Faults and Unconformities: Whenever a fault or unconformity is encountered, the bit will create a dogleg. The Plio-Pleistocene example in Figure 5 illustrates the effect of a change in formation compaction on direction of hole drift.
Figure 5
There is a down-to-the-south-southeast growth fault at 8200 ft. Structural dip is to the north-northwest on both sides to the fault. On the downthrown side of the fault the hole drifts east-northeast or 90° clockwise from the downdip direction. The upthrown drift is south-southeast or updip. The hole-drift direction changes across the fault because of an abrupt change in formation compaction. Flat Structural Dip: When flat or almost-flat structural dip is encountered, the hole slowly spirals through 360°. A complete rotation may require up to 1000 ft of depth. Low Structural Dip
Low structural dip is indicated by a tadpole cloud with its left edge at the zero dip line. This is illustrated in column B in Figure 6 .
14
Figure 6
If the dip trend is flat, some dips would have magnitudes of a few tenths of a degree and very few actual zero dips would be computed. Five or six tadpoles per hundred feet would be near zero (less than 1°). Not every interval would contain these few very low dips, since the beds were not deposited flat. The directions of the red and blue dip groups also indicate the presence of very l ow dip trends. An area that was flat during deposition would have red, blue, and green dip groups lacking a common azimuth. Do not overlook a low dip trend when a few, almost-flat dips are present. Column C in Figure 6 contains a low (2°) southeast trend. When an obvious trend is present, honor it. Difficult Environments
The most difficult environments for determining structural dips are from sediments deposited in shallow water and on the continental slope. Both environments produce a high degree of dip scatter.
15
In the shallow water environment, the scatter results from the initial high-angle depositions, reworking by waves, and bioturbation. The scatter from beds deposited on the continental slope results from post-depositional deformation. Fishing operations increase the difficulty of determining structural dip because of the damage they cause to formations near the borehole. The dipmeter is a shallow investigation tool, and its measurements are made from the zone that is damaged during fishing jobs. Formation damage increases the scatter on the tadpole plot; the greater the formation damage, the greater the dip scatter. Zones of l east scatter with a 2° or 3° magnitude variation may exhibit 10° or more after a fishing job. Wells drilled with mud weights that were too heavy exhibit the same damage pattern. The 8-Curve Dipmeter Tool
The 8-curve tool emits a current from the entire lower section of the sonde into the formation. A small portion flows from the electrodes to record the mi croresistivity dip curves. The rest of the current serves to focus this small electrode current, providing a measurement with very good vertical resolution. Comparison of the detail of the microresistivity curves with cores shows the resolution to be on the order of 0.4 i n. (1 cm). All current is returned to the metal housing of the tool string above the insulating sleeve. The inclinometry cartridge fits inside the top of the sonde. Its axis is accurately aligned with that of the sonde and includes a triaxial accelerometer and three singleaxis magnetometers. The four arms that carry the measure electrodes have a maximum diameter of 21 in. A simplified mechanical linkage is used so that the el ectrodes describe arcs of circles as the caliper arms extend. The opposite arms are linked, making the sonde selfcentralizing in the hole. In an oval hole, however, each pair of arms opens to a different diameter, and so the electrodes on them are noncoplanar. This noncoplanar geometry is accounted for in the computation process for dip calculations. The 4curve tool design uses a more complex arm geometry to keep all electrodes coplanar. The bottom of the sonde, where the dipmeter pads are mounted, is decoupled from the weight of the electronics and communications cartridges by means of a flex joint. Using a cross-linked arm arrangement, it remains centralized in holes where the deviation is less than or equal to 70° (with the pad pressure control at its maximum). The centralization assures tangential contact between pads and the borehole wall, ensuring that the electrodes on the pad maintain good f ormation contact. The formation-imaging tool also uses this sonde design. Figure 1 shows a comparison of the measuring electrodes on the 4-curve tool, the 8curve tool, and the 2-pad and 4-pad formation-imaging tools.
16
Figure 1
For the 8-curve tool there are two measure electrodes on each of the four pads. The short spacing (3 cm) between the side-by-side electrodes results in a better curve likeness than that from the pad-to-pad configuration. This enables a larger number of high-credibility correlations to be made, with the result that shorter correlation intervals can be used to measure displacements between the side-by-side curves while maintaining a sharp and unambiguous curve match. By using processing methods that exploit the improved data-collection capabilities of the 8-curve tool, a fine vertical resolution of dips is achieved. The 2-pad formation-imaging pad has the two side-by-side electrodes, plus an array of 27 resistivity buttons for detailed formation scanning. The 4-pad version has 16 electrodes per pad. With previous pad-to-pad configurations of the 4-pad device, the lower li mit for meaningful interval correlations was on the order of one dip computation per foot. Using the side-by-side correlation technique, this can be reduced to about 3 in. under favorable conditions, thus enabling more i nformation on sedimentological dips to be derived. The mechanical inclinometer in the 4-curve tool has been replaced in the newer tools by a triaxial accelerometer and three magnetometers. The three-axis accelerometer
17
is housed in a single unit. The Al, A2, and A3 axes correspond to Pad 1, Pad 2, and the tool axis direction, respectively. Accelerometer information is used to derive tool axis deviation and make speed corrections to the recorded curves. The magnetometer has a separate unit for each of the above axes. By measuring the direction of the earth’s magnetic and gravity fields in relation to the tool axis, azimuth information is obtained. The inclinometer gives accurate tool-deviation (0.2°) and tool-azimuth (2°) information. Also, since there are no moving parts, there are no problems caused by friction or inertial delays as there were with earlier mechanical designs. The response time of the system, therefore, is very fast, so that any sudden tool movement is recorded and taken into account during the processing of dip results. At the wellsite, the computation program uses the microresistivity information from the two additional electrodes (or speed buttons) to perform the speed correction. At the computing center, additional processing is performed and the speed correction is further refined. The accelerometer data are first used to correct the eight dip curves and the two speed curves for the effect of irregular tool movement. The displacements with the speed curves are then used to remove any remaining minor speed fluctuations. The original dip curves can than be corrected to their tr ue downhole depths. The 8-curve tool has a sampling rate of 0.1 in., as compared with 0.2 in. for the 4 curve tool. The total current (called Emex) that is sent into the formation is automatically controlled by the surface computer to allow for major changes in formation resistivity. In this way the microresistivity curve activity is maintained in both highand low-resistivity zones so that good correlations can be made. In addition, the microresistivity curves may be played back and re-scaled at the wellsite or computing center to remove the visual effect of variation in Emex current. This ensures that information about grain-size or textural change in the formation is not obscured, as might be the case on the original raw data curves. The 8-Curve Dipmeter Field Log
A real-time field log is recorded during the logging runs. After listing details concerning the tool and recording system, the log heading also identi fies the various curves and scales. The following curves are presented: Hole Deviation. This is computed from sonde deviation using values of sonde length and cartridge standoff. Either the hole or sonde deviation can be presented (default is the tool deviation calculated with zero standoff). Hole Azimuth. Displayed on a -40° to 360° scale. Pad 1 Azimuth. Displayed on a -40° to 360° scale, this curve shows the azimuth of Pad 1.
18
Relative Bearing. Displayed on a -40° to 360° scale, this curve is presented as a cross-check between Pad 1 azimuth (P1AZ) and hole azimuth (HAZI). The relationship is RB = P1AZ - HAZI Dip Curves. These are the eight raw microresistivity curves before any Emex correction. The speed curves are not presented. Emex Curves. Both Emex current and voltage are displayed. As an aid to the field engineer, they allow the operation of the Automatic Emex Control to be monitored during logging. Calipers. Two caliper diameters set at 90° to each other are presented on a linear 20-in. scale.
The field log is readily used to evaluate the data quality. Dip curves should be visually similar in detail and activity. Any departure from this norm may signal unusual conditions or faulty tool operation. The user of computed data is encouraged to study the curves carefully when judging the quality of the computations.
Dipmeter Computation Given that a plane cutting a wellbore produces resistivity anomalies at slightly differing depths on the wall of the borehole facing up- or downdip, the computation of dip and dip azimuth is reduced to a problem of trigonometry. Any plane can be uniquely defined by three points in space. A four-arm dipmeter provides four points. If the bedding planes are uniformly thick and plane at the intersection with the wellbore, only three of the available four points are necessary to compute a dip. When one of the correlation traces is substandard due to hole conditions or recording techniques, the fourth trace allows a margin of safety. Parts (a) and (b) of Figure 1 show a cross section of a borehole with a four-arm dipmeter tool, and a schematic of the correlation curves that might be recorded.
Figure 1
19
A comparison of displacements of an anomaly on two correlation curves is key to computing the formation dip. Figure 2 illustrates a dipping plane cutting across a borehole and the expected displacements.
Figure 2
The starting point for dip computation is thus the correlation of one trace to another in order to discover the relevant displacement. The correlation process can be made optically using the 60 in. per 100 ft record and a special apparatus known as an optical comparator, or it can be done by computer. Optical correlation is rarely used anymore since it requires a skilled specialist, takes time, and makes no allowance for tool acceleration and deceleration. Computer-based correlation can be made using a variety of techniques, such as pattern recognition, Fourier analysis, and conventional correlograms. The most commonly used technique builds correlograms. Three parameters are used to control the correlation process, as illustrated in Figure 3 .
20
Figure 3
They are the correlation interval, the search angle, and the step distance. Correlation intervals may range from a few inches to several feet, depending on the information sought. For detailed stratigraphy with high-quality raw data, a correlation interval of 3 in. to 2 ft may be used. For standard work, 2 ft to 6 ft is good, while for structural information, 6 ft to 18 ft will do. The search angle defines how far up and down the hole to seek a correlation and, depending on the hole si ze, reflects the analyst’s guess of the highest expected dip. The step distance defines the depth increment to be used between rounds of correlations. This is usually set to half the correlation interval. Thus, a dipmeter computed on 4 ft x 2 ft x 35° means a correlation interval of 4 ft was used with a step of 2 ft and a search angle of 35° Since only three points are required to define a plane, a four-arm dipmeter survey forms an overdetermined system. Any three curves of the four can provide a dip. Three items may be selected from a choice of four in twelve ways. Potentially, therefore, many dips may be computed at the same depth. In practice, it is found that they do not all agree. For the same reason that fourlegged stools tend to wobble on an uneven floor, but three-legged stools do not, a number of dips are possible simply as a result of nature not providing us with bedding planes that are perfect planes at the scale of one borehole diameter. Add to this the effects of borehole rugosity, floating pads, and the like, and the result is a scatter of possible dips. The choice of the correct dip then becomes an exercise in common sense. In general, this exercise has come to be known as "clustering." Simply stated: If at any level in the well the majority of the possible dips agree with each other and agree with the majority of the dips at adjacent levels in the well, then those are the most probable dips to use. The criterion for judging the worth of any type of dipmeter computation is, of course, its ability to reflect the known geologic facts.
Computing Dip In the early days of the dipmeter, operators made dip measurements directly from readouts similar to the modern field log. Conductivity curves were recorded in much greater detail at a scale of 1:20, or 60 in. = 100 ft.
21
Each curve feature is the signature of a geologic event in the depositional sequence through which the tool passes. The same event can often be recognized in each of the eight curves, though depth may vary because of dip. By measuring the displacement of the event between each of the curves and knowing the precise depth scale, the actual displacement may be read in inches or fractions of inches of borehole. The dip angle relative to the plane of the electrodes can be calculated trigonometrically. Hole deviation and direction, the orientation of Pad 1, the true dip angle, and direction relative to a horizontal plane can also be calculated. Computer processing of dipmeter data has completely replaced the manual method for normal applications, but the basic principles have remained. Visual correlation and inspection of detailed logs is still useful in quality control and in studies of fractures and other specific geological events. In the following discussion of dip computation systems, references are made to examples of dip results in order to show the effects of computation type, tool type, and computation parameters. Here we provide an explanation of the presentation method.
Other Presentations Several approaches for processing raw dipmeter data and for displaying the results are available. The choice of system or systems to use should be determined by the type of problem to be solved-structural, stratigraphic, or (as is often the case) both. In addition to the various arrow plots, azimuth-frequency diagrams, and formation-imaging displays that have been described and illustrated, a number of other graphic and tabular presentations are available from dipmeter data. The more popular ones are covered in the dipmeter interpretation sections of this manual.
Interpretation and Applications
Once a dipmeter has been computed, a number of ways of presenting the answers is available. These include: · tadpole or arrow plots · SODA (separation of dip and azimuth) plots · listings · azimuth frequency plots · histograms · polar plots · stick plots · stratigraphic plots A typical tadpole plot is shown in Figure 1 .
22
Figure 1
The dip magnitude is read from the position of the base of the tadpole on the plot. The dip azimuth is read by observing the direction in which the tail of the tadpole points. The azimuth convention is to measure angles clockwise from north. Thus a north dip points uphole, an east dip to the right, a south dip down-hole, and a west dip to the left. SODA plots separate dip and azimuth as distinct points on separate tracks of the answer plot. Listing In addition to the dip and dip azimuth, these listings may include further details such as dip quality and hole volume. Azimuth frequency diagrams (or rose plots) present statistical information regarding some depth interval in the well, usually 100 ft or 500 ft. Within that interval a polar plot is built with the number of dips having a dip azimuth of a particular direction plotted in a circular histogram. These are most useful for making a qui ck scan of the geologic column for trends in dip direction wi th depth. Conventional histograms of both dip and dip azimuth can also be presented ( Figure 2 ).
23
Figure 2
Polar plots can be built in two ways. One way, the rose plot, has already been described. Another way is to scale the plot with zero di p at the outside and 900 at the middle. Thus the azimuth of the lowest dips becomes more apparent. This type of plot, popular for picking structural dip, is il lustrated by Figure 3 .
Figure 3
24
Stick plots ( Figure 4 and Figure 5 ) show a series of short lines l ines inclined to the horizontal.
Figure 4
Each line represents the dip angle as projected in some line of cross section.
25
Figure 5
A stick plot can be oriented whichever way the geologist wishes. If the orientation is changed, the new axes must be relabeled. It is normal to distort the horizontal and vertical scales on these plots to fit the geologist’s mapping requirements. Stick plots, normally used in multiwell projects to draw cross sections, are particularly helpful where the interwell correlation is not immediately obvious from conventional logs. Stratigraphic plots attempt to give a visual representation of the bed stratigraphy. Each dip may be represented by the trace t race of the bedding plane on the borehole wall. If the trace could be "unwrapped" and laid on a flat surface, a sine wave would be visible, its amplitude a reflection of the dip magnitude and its low point an indication of the dip azimuth. Figure 6 illustrates such a plot.
26
Figure 6
Dipmeter plots may be interpreted by observing the variation of di p and dip azimuth with depth in conjunction with the openhole logs. Here color h elps highlight certain types of patterns. Conventionally, dips of more or less constant azimuth that show an increase in dip magnitude with depth are colored red; those that show a decrease in dip magnitude are colored green. Figure 7 illustrates these patterns.
27
Figure 7
Broadly speaking, dip interpretation may be split into two parts, structural and sedimentary. Gross structural characteristics, such as unconformities, folds, anticlines, and synclines, produce patterns that vary gradually over hundreds of feet. Sedimentary characteristics, such as crossbedding, only appear within sedimentary beds and are localized to a few feet to tens of feet. To become familiar with some of these patterns and their associated geologic features, six cases may be considered. Presentation of Dip Data
The basic method of presentation of computed dip answers is the arrow or tadpole plot. Each tadpole consists of a dot with an attached tail. In Figure 8 the position of the top dot shows a dip magnitude of 20°.
28
Figure 8
Magnitude is the dip angle with respect to horizontal. The tail of the tadpole always points in the downdip direction in this example-N60E, or 60° east of north. The computed dipmeter result is composed of many, often thousands, of tadpoles. From the tadpoles it is possible to recognize changes in dip and direction up and down the well. Changes in magnitude and direction are shown as depth increases. During the computation process, the computer outputs quantities that are used to qualify the sharpness or reliability of the correlation. This determination of answer quality is represented on the tadpole plot in three basic codes. Solid tadpoles represent answers of high accuracy and confidence. Hollow tadpoles represent answers of a lesser degree of the same. No tadpoles, or blank zones, are intervals for which actual correlations were sufficiently in doubt that a decision could not be reached. This method of plotting enables the user to make a judgment on data quality. Figure 9 is a typical tadpole plot over 40 m of hole.
29
Figure 9
Note the solid tadpoles, hollow tadpoles, and blank zone, as previously described. The second set of tadpoles to the far right indicates the hole-drift angle from vertical and the direction of drift. This information can be very useful in interpreting dip data and will be addressed later. An azimuth frequency plot (also known as a rose diagram) is shown on the same track as the dip tadpoles. Each of these plots represents azimuth distribution of all dips between the arrowheads A and B. From a series of these plots over a long interval, one may recognize major direction changes without studying the tadpole plot in detail. The curves on the left of the figure are the two calipers and a computed resistivity. Gamma ray curves may also be displayed. The calipers are a useful indicator of difficult logging conditions, particularly poor pad contact due to hole irregularities. The calipers may also show an enlarged hole where the borehole intercepts a fault or fractured zone. The resistivity curve can be used to positively tie the computed dip plot on depth with other openhole logs. Tadpole Plot Characteristics
30
Figure 10 is a dipmeter plot of a section with excellent parallel bedding, less welldefined bedding, and a blank zone, where no correlations could be f ound.
Figure 10
Note that a consistent trend of hollow tadpoles can gi ve a high-quality interpretation although each individual dip may not in itself imply high accuracy; this is the case within the top 15 m of the log. The general appearance of the dipmeter plot when variables such as tadpole scatter, tadpole quality, and other trends are considered reflects changes in bedding characteristics that are functions of depositional environment, tectonics, diagenesis, rock stress, and other useful geologic factors not deduced from most other logging devices. Indeed, the sequence of those observable characteristics often can be repeated from well to well as consistently as can li thologic sequences, and can provide additional geologic information about an area. Note that during interpretation of any dipmeter plot, th e major influence on the quality of the tadpole is the rock characteristic. Poor bedding may be influenced by any of the following: · lack of textural or mineral stratification
31
· small-scale heterogeneities--e.g., concretions, cross-laminations · bioturbation · diagenesis--e.g., dolomitization of limestones or cementation of clastic rocks resulting in obliteration of original bedding · deformation by creep, slumping, diapirism, or plastic flow · fracturing due to tectonic stress and movement · rubble in fault zones · in some cases, swelling of clay-rich formations adjacent to the borehole by absorption of drilling fluids or modifi cation of the rock stress by the drilling process · dips paralleling the hole axis From the appearance of the plot we can infer formation characteristics related to sedimentary and tectonic processes that further enhance the overall interpretation. 1. Folded Structure. Figure 1 shows a folded structure.
Figure 1
Note that in the shallow part of the well, dips are moderate and to the north.
32
In the deeper section, the well has crossed the axial plane of the fold and dips are more pronounced and to the south. At the point the well crosses the axial plane, dips are flat. It is here that a hydrocarbon trap exists. From the dips on the flanks it is possible to compute both the tilt of the axial plane and the plunge of the fold. 2. Unconformity. Figure 2 illustrates an unconformity.
Figure 2
A series of sediments in the deeper part of the well was originally deposited flat. Thereafter, these sediments were tilted and then eroded and a new set of beds deposited. At the interface between the old and new sediments, there is an abrupt change of dip. 3. Faults. Faults may be picked from dip patterns by observing the drag patterns, if present, on either side of a fault. Figure 3 shows a normal fault with drag.
33
Figure 3
Above the intersection of the wellbore with the fault, a red pattern will develop (dip increasing with depth). Below the intersection of the wellbore with the fault, a blue pattern will develop (decreasing dip with depth). At the intersection of the wellbore with the fault plane, the dip of the fault plane itself may be seen occasionally. Note that the fault dips down in the direction of the azimuth of the drag pattern. It thus strikes perpendicular to that direction. 4. Current Bedding. Among the sedimentary details that may be inferred from a dipmeter plot i s the direction of transportation of sediments by streams. Figure 4 shows the sort of pattern to be expected in such a case.
34
Figure 4
Here, blue pat- terns develop with the dip azimuth in the patterns pointing downstream. Depending on where the well is drilled, it may be of interest to move upstream toward the source or down- stream to finer sediments or broader deposits. 5. Channel Cut and Fill. A common type of deposit results when a channel is cut and refilled with reservoir sand. A red pattern will develop together with a characteristic Sp shape, broadening to the base. In drilling such plays, it is useful to know in which direction the channel extends and in which direction it thi ckens. Note that the well in Figure 5 was drilled off the axis of the channel.
35
Figure 5
Had it been drilled to the north, a thicker section of sand would have been found. To move to the center of a channel, therefore, offset the well in the same direction that the red pattern tadpoles point. To follow the channel, move at right angles to the red pattern dip azimuth, in this case either east or west. 6. Buried Bar with Shale Drape. Another common feature is a buried bar over which subsequent shale deposits have been draped. Here, dips within the sand body decrease with depth (blue), but, above the sand body, dips in the shale increase with depth (red) ( Figure 6 ).
36
Figure 6
The SP usually shows a characteristic pattern, broad at the top. To follow the bar, wells should be offset at right angles to the dip azimuth seen within the bar. To drill a thicker section, a well should be offset in the opposite direction to the dip seen in the bar. Another application of the dipmeter survey is the detection of fractures. There are many methods available for fracture detection, but no single method by itself is completely reliable. The use of the dipmeter for fracture finding, then, is just one of many methods, and should be used to complement the others. The theory is very simple. A fracture may be invaded with mud filtrate and therefore offer a less resistive path to electric current. If one of the dipmeter pads happens to lie in front of a fracture, it will record a low resistivity value. Another pad at the same depth may not be in front of a fracture and will record a higher resistivity. Thus, comparison of adjacent pad traces should reveal the presence of a fracture if the two resistivity values are different. Curves can be displayed in various ways to highlight such differences. Figure 7 shows one such presentation.
37
Figure 7
Note that since the orientation of the dipmeter tool is known, the orientation of the fracture can be deduced. Good dip information requires good raw data. To ensure such data the following guidelines are suggested:
Recondition the hole prior to running the dipmeter. Use a swivel-head adapter to reduce tool rotation while logging.
Log at 1800 to 2400 ft/hour to reduce tool jerking. Slow down even more if the tension on the line is erratic.
Reject sections of log where the tool rotates once in less than 60 ft of hole.
Make repeat sections and/or overlaps of 100 ft to 200 ft every time the l ogging is stopped for film or tape changes.
Inspect the raw log for dead correlation curves, insensitive curves, stuck calipers, etc. As a last resort, three good correlation curves are sufficient, but four are much better.
Carefully inspect the orientation curves for nonsense readings, such as a hole deviation less than zero.
38
On a computed log, check the dip against field controls for consistency. Many dipmeter surveys have been off by 90° or 180° due to i ncorrect pad wiring or erroneous computation.
E2-computation Computation Methods
One method used to obtain dip information from the raw data involves correlating intervals of the dip curves. To a mathematician, a correlation coefficient is a measure of agreement between any two curves. Numerically, coefficients may run from zero (representing two completely dissimilar curves) to one (representing two identical curves). The computer calculates the similarity between a section of one curve and an equal section of a second curve. The length of the interval on the first curve is the correlation length or interval. The computer then moves the first curve by some small, preset increment and recomputes the coefficient. This process is repeated many times. When plotted with respect to depth, the resultant series of coefficients forms a function called the correlogram. This correlogram shows a peak value where the curves have the best fit with each other ( Figure 1 ). The position of this peak with respect to the center of the interval chosen on the first curve is the shift, or displacement, between curves.
39
Figure 1
The process is repeated for all curve pair combinations at that depth; the result is the relative position of correlated points around the borehole, which (when combined with the other measurements such as tool orientation, drift, and caliper data) are used to calculate the dip answer for that depth. A new interval is then chosen on the first curve at a distance equal to the step distance from the previous round of correlations just described, and the process is repeated to produce another dip answer displaced in depth from the previous one by an amount equal to the step distance. This step distance is normally chosen to be some fraction (usually 25 to 50%) of the correlation interval. During the curve-to-curve comparison it is essential to prescribe for the computer the distance up and down the second curve to which the first curve is to be compared. This distance is fixed by the choice of the input parameter called search angle. Search angle is chosen according to the dip environment. For low structural dip areas, a 45° search is common, as most stratigraphic dips fall within that range. In tectonically disturbed areas, higher search angles may be required. The choice in such circumstances must be guided by both local knowledge and close inspection of the dip curves. Large displacements may be visually evident and an approximate dip range may be estimated. The user of the computed data should be aware of a particular characteristic of the interval correlation system. In order to prevent some data from not being used in the computation, the step distance is normally (as mentioned above) less than the correlation interval. This may allow a dominant anomaly (a large sharp peak or trough) to influence the dip answer for each step i n which it is included in the correlation interval. This can cause two or more adjacent dips to be essentially identical, giving the user the impression that several parallel beds exist when in fact there may be only one. For example, a 25% step may produce four similar dips from one anomaly, a 33% step may produce three similar dips, and a 50% step may produce two similar dips. If the user is aware of the parameters used for the computation he will recognize the duplications and interpret the dips correctly. However, if the effect is not desirable, a method called pooling may be used to present the results. In pooled plots, adjacent dips within a very small solid angle (2° to 3°) are presented as one dip answer. Dips that do not pool are still presented, so that no data is discarded. Figure 2 shows another interval with both the unpooled and pooled results side by side.
40
Figure 2
Note the groups of four dips on the unpooled data set that appear as single dips on the pooled result. Also evident is the marked decrease in dip density in the pooled data for the upper half of the log. This can be a desirable presentation, particularly when plotting data on reduced scales, such as 1:600 or 1:1200, for structural dip analysis. Computation Parameter Selection
There are three basic types of interpretation problems that users of dipmeter data may wish to solve: · structural interpretation · large-scale stratigraphic features · maximum detail, very fine stratigraphic features, as observed on detailed core inspection.
41
It is often desirable to interpret a combination of the above from a single dipmeter log. As a result, a variety of systems have evolved to handle widely different requirements. The most commonly used and generally applicable approach is the correlation interval system described earlier. For analysis of structure and large-scale sedimentary features, a 4-ft correlation interval and a 1-ft or 2-ft step i s usually the first approach to analysis. For special applications or difficult logging conditions, other values of these parameters may be more useful. In fact, if the user of the data is specifically interested in one of the three interpretations mentioned above, parameters must be chosen to optimize that result. Therefore, it is important to understand how the tadpole pl ot is affected by the choice of these parameters. For each step, a single dip answer is produced, and all the data within that correlation interval are used to obtain that single dip. A 4-ft interval may contain from 0 to more than 100 correlations, due to bedding contrasts, but only a single dip is calculated, based on the best fit of the correlation curves. Large correlation intervals tend to smooth the dip results. Short correlation intervals allow the system to find more detailed results. Figure 3 contains a 4-m section of dipmeter computed in a sand section using several correlation intervals.
Figure 3
42
Note that although the dip direction trend is similar in each, the implied crosssectional view of the formation is significantly different. Plot A clearly shows detailed internal sedimentary structures with a much better suggestion of environment than do B and C. Plot B retains most of the characteristics of Plot A, but with some apparent averaging and smoothing at dip magnitude boundaries. Plot C suggests large-scale, almost parallel crossbedding. This plot fails to indicate the more complex internal sedimentary structures evident on the A pl ot. It is apparent from comparing these three computations that the choice of the computation parameters should be influenced by the type of information required to support exploration and production programs. Although the basic principles described in the foregoing apply to all correlation interval techniques, algorithms differ significantly for different tool types, allowing the best adaptation to the data obtained by the tool.
Dip Computations with the 4-Curve Dipmeter Tool For the 4-curve tool, two correlation techniques are available to determine the magnitude of the dip and the azimuth of its direction: interval correlation (CLUSTER* Program), as described above; and feature correlation (GEODIP* Program), where individual peaks and troughs are first classified by size, shape, and other characteristics, and these features are matched from curve to curve, taking into account certain constraints. The objective of the latter method is to adapt the program to variations in bedding frequency and thickness, with the result that dip computations are made at points on the dip curves rather than over preselected intervals. This system then frees the computation from a fixed interval constraint, and allows computation of dips of i ndividual bed boundaries. Note: Throughout this document an asterisk (*) indicates a mark of Schlumberger. The overlapping correlation sequences of CLUSTER processing are an improvement over previous programs, but it still has the disadvantage of a fixed, rigid correlation "window," unresponsive to variations in the density of geologic data in the curves. A close study of dipmeter curves shows that many curve features or elements are identifiable from curve to curve. As shown in Figure 1 , these features have various thicknesses (from 1 in.
43
Figure 1
to several feet), amplitudes, and shapes. Each feature may be considered to be the signature of a geological event in the depositional sequence of the formation. Moreover, the dip of the bed boundaries is not necessarily constant, and sometimes varies rapidly. In the GEODIP program, each of the four dip curves to be correlated is mathematically decomposed into a depthordered sequence of ranked elements. In feature extraction, which is the first phase of the program, elements such as peaks, troughs, spikes, and steps are identified in the curves. Each feature has one or two boundaries and a set of parameters that describe its shape. In the second phase, the GEODIP program attempts to match elements of one curve with similar elements of the others according to the following logic:
By a built-in order of precedence (e.g., first large troughs, then large peaks, then medium troughs, and so on), the program first evaluates higher-order features, then when necessary also evaluates lower-order ones. This is done during multiple passes through the four sets of elements.
Because geologic strata are deposited in succession, their boundaries do not cross. So, i f event A appears above event B on one curve, it cannot appear below event B on another. This is the rule of noncrossing correlations.
44
If no correlation can be found within the specified search angle among all four curves, the program lowers its standards and looks for 3-curve correlations instead. Planarity is monitored continuously, and if it fails to meet preset standards, the program makes no attempt at 4-curve dips, but computes the four different 3-curve dips and displays them all. Because the program works from identifiable features on the curve, each one corresponds to a geologic event and the density of the output data depends on the density of geologic information at that level. This makes GEODIP processing particularly successful in fine-structured sedimentary sections and for definition of l ithological changes, such as scour surfaces. The calculation of dip angle at each depth is from displacements measured on boundaries rather than on feature centers. These boundaries are shown on the correlation curves of a GEODIP log. They are themselves useful features for interpreting lithology, as Figure 2 suggests.
Figure 2
Determining Data Quality The geologic validity of each dip determination may be tested in several ways. Closure If displacements are determined between each adjacent pair of curves, taken cylindrically (1-2, 2-3, 3-4, 4-1) they should have an algebraic sum of zero. (Moving from one electrode to the next, you should return to where you began after making a traverse of all four electrodes.) This condition is called perfect closure. Small
45
closure errors may be due to inaccuracies in the computed displacements; large closure errors in dicate that one or more of the correlations are in error. Planarity Another test is for planarity, the condition that the four points should define a plane. After four displacements have been calculated, the lines joining diametrically opposed electrodes may fail to in tersect, if there is an anomaly in the calculation or in the bedding.
For the 4-curve tool, the geometry of the pad lin kage ensures that distances between opposing adjacent pairs remain equal. Displacements computed from opposite pairs of curves (h 1-2 and h 3-4 for example) must therefore be equal but opposite if the bedding surface is planar. (The line segment connecting Pads 1 and 2 on the dipping plane parallels and equals in length-but is oppositely directed to-the line segment connecting Pads 3 and 4, for example.) For perfect planarity, h 1-2+h3-4 = 0 and h2-3 + h4-1 = 0. Likeness A third test is for likeness, a quality derived from the correlogram, to compare the similarity of the curves. The highest correlation coefficient computed over the search interval is the likeness of the two curves, and the trial displacement of that maximum is the displacement retained for that interval of the curves. Since more than one cross correlation is required to compute a dip, the credibility of the dip answer is roughly proportional to the lowest likeness of all the correlations used.
Despite these tests, the results sometimes show excessive scatter that is n ot of geologic origin, particularly when shorter correlation lengths are selected to improve resolution. The CLUSTER program reduces the scatter in the output by statistically reducing the data. It is assumed that random noise does n ot repeat itself through small changes of the correlation environment. Thus, at a given level the redundancy inherent in having four correlation curves allows the curves to be grouped in various combinations in a search for consistency. In addition, coherence between consecutive overlapping levels above and below each point in the hole is checked. The program computes correlations between five of six possible pairings of the four curves, taken two at a time. To define a plane, any two of these pairs must have one curve in common. The CLUSTER program, working with this output, considers eight such solutions. Each of the eight yields a solution for the true dip plane, and generally each is slightly different. Calculations from an adjacent level yield another set of eight solutions. Since the correlation interval is greater than the step distance, neighboring correlation intervals overlap. Comparison of dips from several overlapping levels (eight solutions from each level) shows statistical scatter among the different solutions, but there should be a tendency for many of them to "cluster" near some numerical value. When several solutions (not all from one level) fall within an acceptable range of values, the program quotes the value for the group, rejecting those that scatter outside. As a result, legitimate dip trends can be sorted from noise.
Computing Dip with 8-Curve Data This section discusses the methods developed specifically for processing 8-curve data using the principles of interval and feature correlation, the presentation of the results, and the presentations available at the wellsite and at the computing centers. The determination of formation dip measurements using the 4-curve dipmeter tool depends on the bedding plane being detected by at least three of the four measure electrodes. This, in turn, implies that the formation is wellbedded or laminated. Unfortunately this is not always the case, and for many formations pad-to-pad correlations are impossible to establish, making sedimentary studies difficult or impossible. Also, pad-to-pad correlations may be difficult in highly dipping formations or in highly deviated holes. The 8-curve tool was designed specifically to overcome this limitation by providing two microresistivity curves, 3 cm apart, on each of the four pads. The density of the results is an order of magnitude higher than with previous 4-pad hardware and processing. In addition, the improved sonde velocity correction, using accelerometer data to compute instantaneous sonde speed and length of travel along the borehole, greatly increases the coherence of the results and helps salvage data affected by severe hole conditions. The processing methods discussed here have been developed to take advantage of the tool improvements. They provide three independent computations of formation dip and allow adaptation of the interpretation of th e results to the specific problem of interest (e.g., structural, sedimentary, geometry of the sand body). Programs for computing dip from 8-curve measurements include the basic interval correlation program, called mean square dip (MSD), which uses all 28 possible cross correlations to compute 28 displacements (if all are successful). Since only two adjacent displacements are needed to define a plane, considerable redundancy has
46
been built into the measurement system. The program thus tries to find a "best fit" plane that satisfies most of the displacements. A second interval correlation method called continuous side-by-side (CSB) is also used. It only considers displacements computed from the side-by-side buttons on the pad. These four computed displacements represent the apparent angle of the set of bedding planes that cut across the borehole. Finally, feature correlation is provided by the LOCDIP* computation. These pad-to-pad correlations are made over short intervals centered on bed boundaries, as defined by the major inflection points on the microresistivity curves. This method is used to identify and then correlate major individual curve features. The correlation lines are displayed with the actual microresistivity curves i n a way similar to the GEODIP computation and presentation.
Mean Square Dip (MSD) Processing At any one depth level, there are 28 possible cross correlations for the 8-electrode measurements, as compared to six for the 4-curve recording. As in 4-curve processing, the correlation method for the eight curves requires defining an interval length, a step, and a search angle; however, there is a significant difference in the way the cross correlation is made. In the standard interval correlation program, a specific interval of a reference curve is defined and then slid along the interval of the matched curve. For the 8-curve dipmeter tool, the MSD method considers the same depth interval on each curve and uses only the data within that interval to make correlations. In the case of low apparent dip, nearly all the data points within the interval are considered when the correlation is made. As the apparent dip increases, fewer and fewer points enter into the correlation. A limit is imposed when the search angle is increased until only half the points in the intervals are being used. This corresponds to an apparent dip of about 72° in an 8-in. borehole with a 4-ft correlation interval. In areas where high dips (or high apparent dips due to deviated hole conditions) are expected, this limitation can be overcome by displacing the curves by a known amount before cross-correlations are attempted. The amount of the curve displacement or shift would be that corresponding to the displacement one would expect if the actual dip plane were the same as the assumed or "focusing" plane. Hence, the net displacement used in the dip computation is the interval shift plus the displacement computed between the curves after the shift. The focusing plane can be chosen as either
a fixed plane defined by the analyst (default is a horizontal plane), or a plane defined by a previously computed dip
For moderate structural dip computations, experience has shown that the following input parameters are usually satisfactory:
interval length, typically 4 ft.
step distance, expressed as a percent (usually 50%) of interval length-(e.g., for a 4-ft interval, step distance would be 2 ft)
search angle; 45° usually find most dips relative to a horizontal plane
The MSD program, then, is primarily used to determine structural dip by finding strong planar events crossing the borehole. The button-button displacements are computed and the best-fit plane through these displacements is found. This initial best-fit can then be refined by an iterative process in which points beyond k (which varies from 2.5 to 1.4) standard deviations from this initial best-fit plane are rejected, and a best-fit plane through the remaining points is calculated. An empirical quality factor is assigned to the final best-fit plane. This factor, ranging from 0 to 20, is a function of the number of i terations made and the final number of displacements retained. There is no vertical continuity logic or clustering routine in the MSD computation; each level is autonomously processed. The redundancy available (28 possible displacements, when two are enough to define a dip) reduces the possibility of producing mathematical dips or noise correlations.
47
Continuous Side-by-Side (CSB) Processing Continuous side-by-side (CSB) processing is a unique feature of the 8-curve measurement and takes advantage of the fact that there is great similarity between the two microresistivity curves recorded by each pad since the two measure electrodes are separated by a horizontal spacing of only 3 cm. With side-by-side correlations, CSB processing is able to define formation dip that may not be apparent on pad-to-pad correlation. Even more important, the CSB program is responsive to the fine bedding structure of the formation, making it particularly effective for defining stratigraphic features. This is illu strated in Figure 1 , where the curves recorded by Pads 2 and 3 are shown for 12 ft of hole.
Figure 1
Side-by-side correlations are shown as thin lines, and, for reference, the pad-to-pad correlations found for the same interval are shown as thick lines. From this example, we see that the number of side-by-side correlations is approximately an order of magnitude greater than the pad-to-pad correlations, and that the resolution is on the order of a few inches. Another important feature, due to the proximity of the buttons on the pad, is that the displacements found by side-by-side correlations are much smaller than pad-to-pad displacements. This allows the measurement of very high dips that are not detected by pad-to-pad correlation. For such cases, once credible dips are found by CSB processing, they can be used as input to the focusing option for the MSD program. Figure 2 shows a conventional pad-to-pad MSD correlation for a case of high apparent dip.
48
Figure 2
The well is deviated about 35° to the southwest, in the same direction as the regional structural trend (30° to 40°). Thus, a given bedding surface would cut the borehole high on the northeast side and low on the southwest side. Obviously, getting a good correlation is difficult, although the quality of the dip curves and the borehole condition is excellent. Figure 3 shows the results obtained with side-by-side CSB processing.
49
Figure 3
In this case, the 3-cm spacing of the buttons allows an unambiguous correlation to be made. In the standard CSB computation, each pair of microresistivity curves (e.g., buttons 1-lA) is cross-correlated using short correlation intervals of 12 in. or less, and under favorable conditions even 4 in. or 3 in. The step distance can be taken equal to half or three-quarters of the correlation interval. This gives a vector parallel to the dip plane. Under ideal conditions (planar beds) another vector is found at the same depth by cross-correlating the microresistivity curves of an adjacent pad (e.g., buttons 2-2A). These two vectors are then used to define a dip plane. With only four side-by-side correlations, a cross-check is needed to verify that the bed is indeed planar. If it is, then displacements obtained using microresistivity curves from opposite pads (e.g., buttons 1-lA, 3-3A) should be equal in value but opposite in sign, and the dip can be obtained from any two orthogonal pairs at that depth. If this is not the case, however, a window is opened around the level under examination, and the vertical continuity of the displacements a certain number of levels above and below it is checked. The pad showing the best vertical continuity is kept. A similar procedure is then followed for Pads 2 and 4 and, again, the pad showing the best vertical continuity is kept. The orthogonal pair showing the smoothest continuity within the window is used for dip computation. In order to evaluate the credibility of the dip, a quality value ranging from 0 to 20 is assigned to each dip according to the vertical continuity and the quality of the correlograms at the various levels or depths. If the environment of deposition produces little contrast between beds or the formation is highly crossbedded with sequences terminating over lateral distances of the same order as the borehole diameter, then pad-to-pad correlation may be difficult or impossible due to curve dissimilarity. CSB provides an excellent solution to this problem.
50
Correlation intervals as small as 2 in. have been matched with detailed core i nformation, although 6-in. to 1-ft correlation intervals are most commonly used. Figure 4 shows the detail available from the CSB as compared to visible core features. To make this comparison the CSB was processed with a 6-in.
Figure 4
correlation interval and a 2-in. step and then plotted on a scale one-quarter of full size in order to match with the core photographs. Good dip agreement is apparent. Note the low contrast on the dip curves correlating to the fore-sets in the lower one-third of the photo. The truncation visible on the core is also evidenced on the dip plot. Such detail would not be possible with standard pad-to-pad correlation systems. The good likeness of the side-by-side curves is useful in cases of high apparent dip. Under these conditions it becomes difficult to find an u nambiguous curve match between the pads. Use of the side-by-side configuration allows reliable measurement of displacements between the curves from the same pad and computed dip values.
LOCDIP Computation As discussed earlier, inflection points on the microresistivity curves describe geological events in the depositional sequence of the formation. The purpose of the LOCDIP program is to detect the geological events, or boundaries, and where applicable to associate a dip precisely at that boundary independent of dips at other depths. Instead of correlating intervals of curves, it detects features (inflection points) on each curve and attempts to link these around the borehole, in a manner somewhat similar to GEODIP processing. There are, however, some important differences:
51
To be retained as a LOCDIP result, an event must be recognized on at least seven of the eight microresistivity curves; GEODIP logic requires only three out of the four curves. Thus, LOCDIP logic is more demanding than GEODIP logic.
A measurement of the planarity is derived for each of the possible dip planes at any level. The retained value corresponds to the surface that best approximates the set of these planes. By convention, a perfectly planar surface has a planarity of 100.
Some events are recognized on only a few of the dip curves. In this case, the available correlations are traced across the applicable curves, with an "options" notation of "F" (fracture) or "P/L" (pebble or lens) for single-pad events or two/three-pad events, respectively. These interpretations, however, are not to be considered as certain, but rather as possible.
The processing of the 8-curve data is designed to extract the maximum amount of dip information from the raw curves. A well may present several interpretation problems due to variations in lithology and bedding characteristics. A single computation system may not offer the total solution. It is useful, therefore, to be able to combine the results of several types of computation in one presentation.
DUALDIP* Presentation The DUALDIP presentation for the 8-curve dipmeter tool allows results from more than one computation to be combined. Figure 5 is an example of multiple computations on a short section.
Figure 5
52
In the figure, the dips on the l eft side are side-by-side (CSB) results with a correlation length of 8 in. and a step of 4 in. This produces three dips per foot, or about 10 dips per meter. The tadpoles on the right are of two types. The round-headed tadpoles were computed from pad-to-pad correlations with a correlation interval of 4 ft and a step of 2 ft. This is the MSD computation. The triangular-headed tadpoles are LOCDIP computations, also known as pad-to-pad feature correlations. These dips usually correspond to the more prominent bed boundaries, and are computed by the earlier mentioned pattern-recognition system. For each LOCDIP computation which used seven or ei ght of the dip curves, a solid correlation line is drawn on the plot showing exactly where the bed boundary was interpreted. For each of these correlations a local dip is shown. If fewer than seven curves are correlated, then the correlation is shown as a dotted line, but dip is not computed. This presentation not only gives a visual impression of the frequency of stratification and its planarity and parallelism, but it also allows the user to judge the validity of the correlations. This is of particular value in detailed studies of sedimentary features. All three systems may not, nor should they necessarily, give the same dip answer. This characteristic can be used to great advantage in interpreting sedimentary features, particularly thin, highly bedded clastics. In Figure 5 , the two local dips at A and B correspond to the top and bottom of a distinct sedimentary unit. They suggest the boundaries both dip at 1° northerly. All the finer bedding within these boundaries produced CSB or round-headed dips consistently north-northeast between 4° and 10°. The internal bedding indicates sediment transport direction from south-southwest to north-northeast, with topset and bottomset surfaces approximately 1° northerly. The CSB result is different from that obtained from LOCDIP and MSD processing, whose computation system is restricted to major events, which can be correlated from pad to pad. The CSB logic favors events with some continuity; individual single events are less likely to be computed, particularly where both types are visible within the correlation interval. This tendency for different systems to favor different types of bedding planes has been very useful, particularly in the interpretation of fluvial environments. Note also that the 4-ft MSD correlation showed the dip at C to be southwest about 90 and consistent over 4 ft. This is easily explained, considering the previous discussion of overlap effects, and it is supported by the LOCDIP computation at that depth. This boundary presents a dominant anomaly to the 4-ft correlation system, and for fine stratigraphy would be misleading by itself. When all bedding features, large and small, are parallel, all systems should give the same answer as at D.
Formation-Imaging Tool
Successful dipmeter interpretation depends greatly upon the accurate evaluation of geological features. The application of the classic dip patterns is a relatively s imple matter when geological events such as current bedding or lateral accretion are known. In many complex environments this is a severe problem. A whole core over the zone of importance solves these problems, but whole core availability is t he exception rather than the rule. Formation imaging provides a continuous oriented borehole representation that can be used in conjunction with a whole core or, in most cases, by itself to evaluate geological events. Interpretation The goal of formation-image interpretation is to characterize formation properties to assist sedimentological interpretation, determine the presence of permeability paths and permeability barriers, help calculate net pay, plan perforation and fracturing, and to help decide whipstocks and where to dril l next.
Formation images must always be interpreted after lithology has been fairly well defined, so supplemental data are usually necessary to enhance the confidence of
53
image interpretation. As with other dipmeter interpretations, the more supplemental data available, the better the interpretation. Measurement The clustered microresistivity buttons on two or four of the microscanner pads provide a continuous electrical image of the borehole wall. The pads are oriented at right angles to achieve a three-dimensional perspective. These resistivity data are then mapped to a gray-scale or color "corelike" borehole wall image. This allows fine-scale features to be described through essentially the same interpretation procedure as that used in the examination of slabbed cores. The images characterize many types of structural and stratigraphic features. These oriented features, combined with a conventional dipmeter plot, are used to evaluate these events to extend the reservoir geometry beyond the wellbore .
Images of the rock formation exposed by the wellbore are processed from the microresistivity traces. Each image pad covers 2.8 in. of the borehole wall. Thus, 22% of an 8-in. borehole can be imaged with two pads and 44% with four pads on each logging pass. This coverage can be increased with multiple logging passes. The tool also contains a triaxial accelerometer and three magnetometers for orientation and to enable speed corrections to be made on the acquired data. Presentation of Images
Several presentations are available for displaying the data. The vertical scale provides the most striking difference between the formation-imaging presentations and other logs. The normal detail scale for logs i s 1:240, while the formation images are presented on a 1:5 scale. The standard presentations can be broadly classified into two types: straight-line images and azimuthal images. Straight-Line Images A straight-line presentation shows the images in a stationary horizontal scale ( Figure 1 ).
54
Figure 1
This presentation is divided into several sections. The left section contains the dept h scale, the pad orientation, and the borehole deviation. The long arrow on the tadpole indicates the direction of borehole drift; the body of the tadpole indicates the magnitude of deviation by its position on the horizontal scale. The small arrow shows the azimuth of Pad 1. The next section contains the caliper and resistivity correlation curves. The calipers from Pads 1-3 and 2-4 are shown. The resistivity curve is used only for correlation and not for quantitative purposes. Pads 3 and 4 of t he 2-pad tool provide the image. Both the raw microresistivity traces and the processed images are presented. The microresistivity traces are from the 27 image buttons. The image traces are computer enhanced using 16 gray levels; they range from white (resistive) to black (conductive). Another popular presentation is shown in Figure 2 .
55
Figure 2
In this example, the formation images are displayed on the same depth scale as the dipmeter log. This scale is not as effective for identifying individual sedimentary features but is better for displaying the overall features of a zone and showing how they relate to dip patterns. Azimuthal Images A BORMAP presentation is shown in Figure 3 .
56
Figure 3
The horizontal scale shifts according to the respective azimuths of each pad. Thus, multiple passes can be merged to portray a more complete pi cture of the wellbore. In this example, images from two logging passes (from a tool with two imaging pads) were merged to cover approximately 44% of the well-bore. There are vugs present at 4208.7 ft and at 4210.4 ft. This presentation is very effective for secondary porosity evaluation and for sedimentary structure identification. Image-Examiner Workstation
Image interpretation can be enhanced by means of a computer workstation equipped with image-examiner processing programs. This allows such interactive processing features as · scale changes of both the vertical and horizontal, to enhance the interpretation · a display of other logs for correlation on the same scales · graphic enhancement of specific features, such as bedding, texture, vugs, and fractures
57
· dip computation of bedding surfaces, fault planes, and fractures · correlation of images to whole core sections, extending the interpretation to noncored sections · orientation of cores from features present in both the core and the formation images · quantification of images (such as sand count and calibration to core porosity) to increase interpretation accuracy Dip Computation/Thin Bed Definition Computation of the dip magnitude and azimuth of specific beds is essential to many interpretations and can be performed on an image-examiner workstation. An example is shown in Figure 4 .
Figure 4
The magnitude is measured from horizontal (0°) to vertical (90°). The azimuth of the downdip direction is measured from true north. The thin sand shown at 6969 ft dips to the northwest. A sine wave is fit through both the upper and lower surface of the sand, indicating a 39° dip magnitude and an azimuth of 317°. These dips are "true
58
dip", since hole deviation is compensated. Apparent dips may be presented if a direct comparison with a whole core is required The actual thickness of the sand stringer, measured be-the sine waves, is 1.61 ft.
E3-dip patterns Introduction
There are several graphical methods of plotting dip computations. This chapter covers interpretation rules based on the common "tadpole" plot. The head of the tadpole indicates dip magnitude and is plotted on a dip scale ranging from 0° to 90° versus depth. The tail of the tadpole, which points in the downdip direction, is plotted on a compass rose (north, up; east to the right; south, down; and west to the left). The two or more tadpoles forming a group are derived from the internal structure of sediment layers deposited in a single depositional environment. All dips on a tadpole plot can be assigned to one of four basic groups. These groups are the building blocks used to create megapatterns. Mega-patterns, lithology, and knowledge of the depositional environment are used to make interpretations. Dip Groups
The four dip groups are the red (slope), blue (current), green (constant), and random. These basic groups are the building blocks of megapatterns, which are used to identify missing or repeat sections and to interpret stratigraphy. The red, blue, and green patterns are illustrated in Figure 1 , which shows the borehole, formationimaging, and dipmeter tadpole plots for each.
Figure 1
Red (Slope) Groups
59
Red dip groups are composed of two or more adjacent tadpoles with constant azimuths and downward increasing magnitudes. These groups are generated from sediments deposited on a sloping surface or from sediments with dips that have been altered by postdepositional movement. Most red groups result from downdip thickening. Beds deposited on a sloping surface thicken and become wedge-shaped in the downdip direction; therefore, the dip direction of red groups indicates the direction of thi ckening. Blue (Current) Groups
Blue dip groups are composed of two or more adjacent tadpoles with constant azimuths and downward-decreasing magnitudes. These groups are generated mainly from sediment layers deposited as foreset beds. The dip di rections of the foresetgenerated blue groups indicate the directions of current flow during deposition. Some blue groups are generated by weathering beneath erosional surfaces; this process creates downward-flattening features. Green (Constant) Groups
Green dip groups are composed of two or more adjacent tadpoles with constant magnitudes and azimuths. These groups are derived from parallel crossbeds or from sediments that were deposited flat and have subsequently undergone structural uplifting. Green groups are the only dip groups indicating structural dip today and are the groups sought within zones of least scatter. Random Groups
Random dip groups are composed of adjacent tadpoles with random magnitudes and azimuths. These groups are derived from sediment layers deposited in high-energy environments, such as shallow water less than 50 ft deep; from layers that have undergone reworking by bioturbation; and from layers that have undergone postdepositional movement. Megapatterns
These dip groups are used as building blocks in identifying megapatterns. Megapatterns, lithology, and depositional environment information are used for determining the location of attributes of faults, unconformities, and stratigraphic features. Mega-Red Dip Patterns
The mega-red dip pattern is a family of the basic dip groups characterized by an increasing downward magnitude trend and a constant or gradually rotating general azimuth ( Figure 2 ).
60
Figure 2
Individual basic dip groups may exhibit dips that do not match the general trend. The features in the subsurface that create mega red patterns on the dip plot include distortions near a fault plane, sand bars, beach ridges, reefs, and channels of all types. Normal Faults Two distortion types, rollover and drag, may be present near a normal fault. Rollover, with dip into the fault, results from sediments slumping into the downthrown side of a fault that was active at the seafloor during the time of deposition.
Drag zones contain beds dipping in the same direction as the fault plane. The megared dip pattern results from friction between the active downthrown block and the passive upthrown block. Most of the distortion occurs in the active or down -thrown fault block; however, upthrown drag is occasionally noted. Mega-red dip patterns are not always found near the fault plane. Some faults have no downthrown or upthrown distortion; in these situations there is no dipmeter indication unless there has been tilting of one of the fault blocks. Reverse Faults Reverse thrust fault usually exhibit drag on both sides of the fault. The drag zone in the overthrust block creates a mega-red pattern dipping in the direction of overthrust. The downthrown dip pattern, if one exists, is a mega-blue pattern. If drag is present on only one side of the fault, it occurs on the more active, overthrust side.
61
If a missing or repeat section is at or near the base of a mega-red pattern, the pattern probably results from some type of distortion near the fault plane. If there is no indication of a nearby missing or repeat section, then the mega-red pattern probably has a stratigraphic origin. Additional information from other logs about the depositional environment and lithology is necessary to determine the stratigraphic feature generating a mega-red pattern. Mega-Blue Dip Patterns
Mega-blue dip patterns are formed when the dip magnitudes of famili es of basic groups decrease downward but their azimuths remain the same or rotate slowly ( Figure 3 ). As is true of mega-red patterns, a few individual basic dip groups may exhibit random azimuths. Also, local data about depositional environment and lithology obtained from other logs are required to make a stratigraphic interpretation.
Figure 3
Mega-blue patterns result from foreset deposition, weathering under erosional surfaces, and compaction caused by the sinking of a relatively dense mass, such as a sand or coral reef, into softer underlying beds. Dip direction of foreset beds indicates
62
the direction of sediment transport or current flow. The dips created by compaction indicate the direction to the thickest portion of the overlying mass. Fore-set deposition occurs in delta-dominated environments, tide/ wave-dominated environments, longshore current sand waves, submarine fans, tidal flats, and at or near the axes of any type of channel. Identifying Megapatterns
Basic dip groups that do not form megapatterns terminate at or near a vertical or slightly inclined line ( Figure 4 ).
Figure 4
Basic dip groups that form megapatterns terminate at successively higher magnitudes-e.g., higher downward for a red, higher upward for a blue--for the length of the pattern. If the deepest dip group of a megapattern has an azimuth different from the azimuth of the general pattern, the azimuth of the pattern, not the basal group, should be used.
Theoretical Dip Patterns
63
A series of theoretical dip-versus-lithology patterns can easily be created for any speci fic environment. Since the same dip patterns can be created by different stratigraphic features, the theoretical sketches are grouped by depositional environments (nonmarine, deltaic, interdeltaic, and deepwater) . A missing and repeat dip response is also included. The interpretation process can be carried out step by step in the following sequence:
Determine structural dip.
Delete structural dip if necessary.
Identify and describe the attributes of missing and repeat sections.
Make stratigraphic interpretations using lithology and knowledge of the depositional environment.
If independent knowledge of the depositional environment is unavailable, local "rules of thumb," using such parameters as bound water resistivity, shale resistivity, and dip scatter, can be used as environmental indicators. Structural Dip In order to represent structural dip today, any bedding plane or sediment layer must have been deposited flat and have undergone only structural upli ft since the time of deposition. Structural dip trends are selected from zones of least dip scatter, since such zones are most likely to have been deposited in a low-energy environment and thus are most likely to represent structural dip today. A good rule of thumb is to assume that structural dip trends picked from the dipmeter display extend h orizontally no farther than they do vertically. Dip trends that extend 1000 ft or more can usually be extended as far horizontally as the closest offset well. However, if numerous faults and unconformities are present, it may be impossible to find a dip trend that e xtends 1000 ft vertically.
Structural dip should be deleted before fault and stratigraphic interpretations are made if the dip magnitude in the zone of interest is less than th at of structural dip, or if the dip azimuth in the zone of interest is different from that of structural dip.
Missing or Repeat Sections
After determining and (if necessary) deleting structural dip, the next step is interpreting missing and repeat sections. Missing sections result when normal faults, angular unconformities, disconformities, or diastems are present. Repeat sections result from compressional faulting and folding. Since stratigraphic features and faults can generate identical dip patterns, an independent input as to locations of probable missing sections is desirable before making missing-section interpretations. Normal faults may generate red di p patterns that dip either toward or away from the fault plane. Dip patterns on the downthrown side of growth faults, which result from rollover into the fault, dip toward the fault plane. The vertical extent of such patterns can be used as a minimum fault displacement indicator. Nongrowth normal faults that occurred after some formation compaction had taken place create red dip patterns that dip in the same direction as the fault plane. These result from a drag zone immediately downthrown to the fault. Reverse and thrust faults, which generate "right-side-up" repeats on the logs, create red-over-blue dip patterns. The patterns dip in the direction of overthrust, and the fault plane is located at their junction. Overturned folds also create log repeat sections, but one repeat is a mirror image of the other.
64
From the bottom up on Figure 1 , the first missing and repeat section is a diastem or disconformity.
Figure 1
Since the angular difference across such features is less than one-half of a degree, they are not easily recognized on dipmeter plots. The small blue pattern shown beneath the missing section is the result of some type of weathering. The next repeat section results from an overturned fold. The log response of the repeat section produces a mirror image with the repeat section upside down with respect to the first log response. In this example, there is a dip reversal across the fold; this is not always the case. A reverse or thrust fault also produces a repeat log response, with the repeat right side up with respect to the first log response. Both the red pattern in the upper, or overthrust, block and the blue pattern in the downthrown block are the result of drag. The dip direction of the overthrust red pattern is the same as the direction of over-thrust (to the east in this example). The next upward dip decrease is the result of a period of postdepositional uplift that created a portion of the underlying 25° northeast structural trend. There was no
65
erosion of the uplifted beds. Deposition, which continued without a break, then produced onlapping beds. The overlying 20° east structural trend was produced by a later period of uplift. Such features are common in sediments deposited in deepwater environments. The next upward dip increase, from 20° east to 10° east, occurs across an angular unconformity. The blue dip pattern drawn below the unconformity results from some type of weathering and occurs most of the time. Since this small blue pattern is identical to patterns produced by stratigraphic events, it should not be considered a diagnostic unconformity indicator. There is independent input that a fault exists within the section of 10° east structural dip. Since there are no associated red or blue patterns, this is a middle-aged normal fault that has no distortion near the fault plane. However, a sudden structural dip change occurs when one fault block has been tilted. There is also independent input that another fault is l ocated just uphole. A red dip pattern, which terminates at the probable fault location, is present in thi s example. If the vertical extent of the red pattern is more than 200 ft, the pattern is almost certainly the result of dip into the downthrown side of a growth fault The dip direction is toward the upthrown block (example: upthrown to the northeast). If the vertical extent of the red pattern is l ess than 200 ft, the red pattern may result from either rollover into a growth fault or drag on the downthrown side of a later fault. When the pattern results from drag, the dip direction is toward the downthrown side and normal to the fault strike. Continental Environment
Figure 1 illustrates some continental environment depositional features and their associated dip patterns.
66
Figure 1
From bottom up, the group of tadpoles indicating east structural dip is derived from sediments deposited essentially flat in an upper delta pl ain environment. Sands deposited in such an environment may contain secondary porosity because some plant-produced acids are capable of dissolving sand grains. Flood-plain sediments produce a "bag-of-nails" dip scatter. Few (if any) dips reflecting structural dip are found within such sediments. Next is an eolian sand. The illustrated dip patterns have constant (angle of repose) dip trends underlain by blue patterns. This is a typical dip response from transverse and barchan dunes. The dip direction indicates the prevailing wind direction at the time of deposition (from west to east in this example). Longitudinal dunes produce red or blue patterns whose dip directions are normal to the prevailing wind direction. Dome and parabolic dunes produce mainly red patterns dipping in the prevailing wind direction. Swamp or marsh deposits generally produce blank zones because bedding planes have been destroyed by rooting and bioturbation.
67
Stream channels filled with clay plugs produce red dip patterns within shale sections. The patterns dip toward the channel thalweg. When stream channels are filled with sand instead of clay, possibly during a marine transgression, a red dip pattern found at the base of the sand dips toward the thalweg and normal to the strike of the channel (example: thalweg to the northeast, and northwest-southeast strike). This dip pattern is overlain by a blue pattern whose dip is 90° from that of the underlying red pattern. This dip direction indicates the current flow direction within the channel (example: direction of flow to the southeast). Point-bar sands exhibit a number of internal blue dip patterns whose dips are i n the direction of current flow. A red pattern that dips toward the thalweg may also be present just above the point bar. If the beds that produce the blue dip patterns are thicker than 3 ft, the blue patterns probably result from accretion depositions that dip toward the thalweg rather than from trough crossbeds that dip down-current. Continental Shelf, Delta-Dominated Environment
In the example in Figure 1 , delta-dominated means that some, if not all, of the stratigraphic features deposited in a deltaic environment were preserved in their original forms rather than in reworked forms.
Figure 1
68
The bottom sand is channel-like and was formed by the compaction of underlying muds. All dips of the red dip pattern (faulting has been eliminated) found within the sand dip toward the axis and normal to the strike of the sand. Because of compaction of the sediments below the sand, a blue dip pattern dipping toward the channel axis is usually found beneath the sand in the underlying shales. Other logs exhibit gradients (downward-decreasing resistivity, increasing interval transit time) in the underlying shales. Sands formed by compaction may be more than 2000 ft thick. Crevasse splays generate blue dip patterns pointing in the direction of current flow (example: direction of flow to the southeast). A sand deposited as a distributary mouth-bar and topped by a scour channel exhibits a red-over-blue dip pattern that dips in the same direction. The blue pattern dips in the direction of current flow (example: direction of flow to the east-southeast) and the red pattern di ps toward the scour channel axis (example: axis to the east-southeast), which usually has a very limited areal extent. In general, when adjacent red-over-blue patterns dip in the same direction, the red pattern can be ignored. Whenever a distributary mouth-bar sand undergoes shallow-water reworking, a bagof-nails dip scatter is produced. Such sands tend to be clean with good porosities and permeabilities. When all the original depositional features of a distributary channel are preserved, they produce a red dip pattern at the base of the sand, overlain by a blue pattern. The pattern azimuths are 90° apart. The red pattern dips toward the channel axis and normal to the channel strike (example: axis to east and north-south strike) The blue pattern dip indicates flow down the channel (example: flow from north to south). A distributary mouth-bar produces a blue dip pattern whose direction is that of current flow (example: flow from northwest to southeast) . When the blue pattern magnitude variation is 10° or more, the distributary mouthbar tends to be elongated in the direction of dip (inertia-dominated environment). When the dip variation is less than 10°, the distributary mouth-bar tends to be fan or crescent shaped (friction-dominated environment). Distributary mouth-bars and crevasse splays look the same on dipmeter plots. Continental Shelf, Tide- or Wave-Dominated Environment
Figure 1 illustrates some of the stratigraphic features and associated dip patterns that are found in a continental shelf, tide- or wave-dominated environment. Many of these features are the result of reworking of previously deposited del taic sediments.
69
Figure 1
At the bottom of the figure, parallel seaward-dipping cross-beds are produced by beach rock that forms in a carbonate environment at the saltwater-freshwater interface along shorelines. An oolitic bar is identified by a red pattern immediately above the bar (assuming, of course, that it was not penetrated on the crest). The red pattern di ps toward the pinch-out and normal to the strike of the bar (example: pinchout to the northeast, and northwest-southeast strike) . Dips within the oolitic bar are immaterial. A reef also exhibits a red pattern above its top and a blue pattern in the underlying beds. Few, if any, meaningful dips are found within reefs. The overlying red pattern dips toward the pinchout and normal to the strike of th e reef. The blue pattern, which results from compaction, dips toward the thicker part of the reef (example: pinchout is to the east-northeast and the reef strikes north-northwest, southsoutheast) . A buried beach ridge exhibits a red dip pattern immediately above the top of the ridge and numerous dips within the beach-ridge sand. The red pattern di ps toward the shaleout and normal to the strike of the beach ridge (example: shaleout to northeast, and northwest-southeast strike) .
70
A sand bar that formed at the wave breakpoint also exhibits a red dip pattern above the sand but few dips within the sand (reworking increases the electrical homogeneity) . The red pattern dips toward the shaleout and normal to the strike of the bar (example: shaleout to the northeast, and northwest-southeast strike) . In Figure 2 the bottom sand was deposited as a slip-face sand on the landward side of a beach.
Figure 2
The internal blue dip pattern dips landward and normal to the beach strike (example: land to west, and north-south beach strike). The next sand was deposited as beach dunes and contains varying dips resulting from festoon crossbedding. Formations on the berm crest of a beach can be deposited flat and would later indicate structural dip. Runnel sands may exhibit blue patterns derived from mega-ripples whose dip azimuths approximate the beach strike. Small-scale ripples may produce either blank zones or random dips. The example beach strike is north-south, indicated by southdipping blue patterns derived from megaripples.
71
A beach-face sand contains seaward-dipping parallel cross-beds (example: parallel crossbeds dipping 5° east indicate that seaward was to the east during deposition). Upper shore face sands contain a few parallel seaward-dipping crossbeds (example: 1° and 2° east dips indicate that seaward was to the east during deposition). Lower shore face sands contain mainly blank zones and random dips that result from high energy environments and extensive bioturbation. Longshore current sand waves exhibit blue dip patterns dipping in the di rection of transport and parallel to the nearby fossil shoreline (example: dips to south indicate transport from north to south along a north-south striking shoreline). A tidal flood delta, or washover, fan generates landwarddipping blue dip patterns (example: west-dipping blue patterns indicate that land was to the west during deposition). Ebb deltas produce seaward-dipping blue patterns (example: east dip indicates that seaward was to the east at the time of deposition). Deepwater Depositional Environment
Figure 1 illustrates the sedimentary features found in deepwater (continental slope and deeper) sediments.
72
Figure 1
Often, postdepositional movement occurs within sediments deposited on the continental slope. This produces a bag-of-nails dip appearance. Structural dip is extremely difficult to determine from such intervals. Deposition at the distal end of submarine fans produces alternating sand-shale layers that later can become low-resistivity pay zones. Dips recorded in this environment indicate structural dip. The midfan portion of a submarine fan produces blue patterns that indicate sediment transport directions (example: transport direction was north to south) . Debris flows produce blank zones or zones of random dips. A submarine channel penetrated near the edge exhibits a red pattern that dips toward the channel axis and normal to its strike (example: axis to the east, and north-south strike) . A nearthe-axis location within a feeder channel produces only blue patterns, which indicates flow down the channel (example: south-southwest dipping blue patterns indicate flow from north-northeast to south-southwest) . A feeder channel penetration between the axis and channel edge produces the "blue over-red with axis 90° apart" dip pattern combination. The red pattern dips toward the channel axis and normal to the channel strike (example: axis to the east and
73
north-south strike). The blue pattern dip direction indicates the flow direction down the channel (example: flow direction was from north-northeast to south-southwest) . These theoretical patterns show all of the original dip patterns intact. In practice, portions of the original patterns may have been destroyed by reworking. Also, random dips that behave like noise are scattered throughout the patterns. Exercise No. 1
Figure 1
The upper 3 m of the log in Figure 1 are in interbedded shales and silts. The lower 4 m are mostly sand in a fluvial environment. This exercise requires the student to study the dip curves closely, from a standpoint of similarity between adjacent side-by-side electrodes and similarity from pad to pad. Also study the dip results from each of the three systems: CSB, MSD, LOCDIP. Study the comparison of 5-inch correlation CSB, LOCAL DIP, and 1-foot correlation MSD. What are the bedding characteristics for each of the four intervals?
74
Solution Interval 1
Curve pairs vary from similar to unlike, and the CSB results reflect this fact. Pad-to-pad similarity is quite poor, causing the MSD dip scatter. Bedding is probably irregular and of very short l ateral extension. There is some stratification, however, as indicated by the similarity of side-by-side curves. Interval 2
The lower 2 m of this section are well-bedded with small curve contrast. Agreement between systems is fair, implying some consistency in direction. At the arrow, note that LOCDIP and MSD point north at 9°, whereas CSB shows SW crossbedding over that section. This i s an excellent example of a dominant anomaly (see correlations) influencing the dips over the complete 1-ft correlation interval on the MSD, and the similar LOCDIP response. Interval 3
Well-bedded, with good basic agreement among systems. Interval 4
Poor bedding, with noncorrelational conductive anomalies. These are pyrite blebs, very small but very conductive.
E4-structural dip interpretation Structural Dip Interpretation
Structural dip changes (and the lack of such changes) are good indicators of the type of structure present ( Figure 1 ). The following are guidelines for interpreting structure based on structural dip changes.
75
Figure 1
Structural dip decreases upward in structures uplifted contemporaneously with deposition. Constant dip over an interval indicates postdepositional structural uplift. Structural trends that decrease to zero dip and reverse magnitude and azimuth indicate structures with tilted axes. Deviated holes create the same effect by penetrating different parts of the structure being explored. Structural dip changes over short intervals indicate numerous faults. The beds between two faults only a few hundred feet apart commonly exhibit different dips from beds above and below the two bounding faults as a result of tilting. If structural dip is changing rapidly in the horizontal direction, it is dangerous to extend the structural trends very far horizon-tally. Only the geologist can decide how far the trend may be extended. When the dip of a structure is changing, the feature interpreted as structural dip is the dip of a plane tangent to the mapping horizon. Salt Domes
Intrusive masses of salt form domelike features by penetrating overlying normally bedded sediments. Figure 1 is a sketch of a typical salt dome.
76
Figure 1
A number of faults are present, most of which dip toward salt. Unconformities and pinchouts are common, as are steep dips near the flanks of the salt dome. If the top of the dome is shallow enough, it may be overlain by caprock. Not all domes resemble the one shown. Other features that l end themselves to dipmeter interpretation may be present; these are presented on the foll owing pages. Overhangs Figure 2 illustrates a well that penetrated salt far below an overhang.
77
Figure 2
Note the following: · Dips are generally highest closest to salt. · Dips increase as an overhang is approached from above. · Dips, then, decrease below the overhang. There is another downward increase as the well approaches the main salt stock. One of the uses of the dipmeter on wells drilled near a salt dome is to indicate the presence of overhangs, which warrant further investigation by an ULSEL survey. The ULSEL device is an electrical logging system with long electrode spacings allowing formation investigation up to 2000 ft from the wellbore. ULSEL measurements combined with induction log and dipmeter data provide the information necessary to compute the distance, direction, and profile of the nearest salt dome.
78
Vertical and Overturned Beds Vertical, near-vertical, and overturned beds are found near salt domes and in areas of over-thrusting. Straight holes are rarely drilled through vertical beds. The apparent dip has a computed value of less than 90°. The dips become vertical only after correction for sonde tilt.
The steepest dips near a salt dome are generally found under an overhang, and some beds may be overturned indicating a horizontal and vertical component to salt movement ( Figure 3 ). The illustrated well was sidetracked under the overhang, and it penetrated increasing easterly dipping vertical beds, overturned beds, and, finally, high easterly dips again.
Figure 3
Pre-Salt Uplift Growth Faults Another cause of dip into salt is the presence of a large pre-salt uplift growth fault. The dip into the downthrown side of the growth fault can override any uplift-created dip away from salt. This feature occurs on the south flank of the dome illustrated in Figure 4 .
79
Figure 4
Gouge Zones Some salt domes are covered by a thin gouge zone, usually less than 100 ft thick. These gouge zones contain a mixture of the various sediments the dome has penetrated. When the resistivities of the normally pressured, bedded shales around a dome are approximately 1 ohm-m, the gouge resistivity averages approximately 1.2 ohm-m. Gouge is a mixture of sands and shales, and it has a "hashy" appearance on the SP and short-spaced resistivity curves.
A blanket of diapiric clay is sometimes found draped around one flank of a salt dome. This is a high-pressure, low-resistivity clay. Resistivities within Gulf Coast diapiric clay domes are commonly less than 0.5 ohm-m. Dips within gouge zones and diapiric clays tend to be random or nonexistent. Clay Domes
Clay domes are formed in the same manner as salt domes. Source beds are masses of low-density shales. The density of these shales can be less than the density of salt: 2 g/cc versus 2.16 g/cc. These low-density shales floated upward through zones of weakness to form clay domes. The penetration of younger overlying beds created dips away from the clay dome. In the northern Gulf of Mexico the top of a clay dome is indicated by a downward decrease in resistivity. The half-ohm shale point was used as an indicator of the t op of the clay dome in the Eugene Island Block 198 field.
80
Resistivities within domes may be as low as 0.2 ohm-m in the U.S. Gulf Coast region. In Nigeria, a 1 ohm-m value is more common. It is currently more difficult to identify clay domes than it was in the 1960s. At that time, a constant dip trend matching the dip of the domal surface was recorded within the dome. As the dome was approached from above, the dip trend increased in magnitude, just as if the flank of a salt dome were being approached. After the clay dome was penetrated, a constant dip trend was usually recorded. This is illustrated in Figure 5 .
Figure 5
Since the late 1960s clay dome dips have become more elusive. Instead of constant dip trends within the dome, only blank zones are found on dip plots. One explanation for this change has been advanced by a major company geologist. He suggests that the current lack of dip data within the clay dome results from formation damage caused by increased mud weights. Dips detected within clay domes were probably derived from cleavages or compaction surfaces, not from bedding planes.
81
Dips are still found within high-pressure, low-resistivity shales in their normal stratigraphic position. After shales have been uplifted, they may be more susceptible to mud-weight induced damage. Structural Dip Deletion
Formation dip results from the original depositional dip, compaction and postdepositional deformation, and structural uplift or subsidence. The magnitude and direction of structural dip are removed before making fault or stratigraphic interpretations. · If the dip in the zone of interest is less than the structural dip, structural dip should be deleted from each of the dips on the tadpole plot. · If the dip in the zone of interest is equal to or greater than structural dip, but with a different azimuth, structural dip should be deleted. Results of Dip Deletion
Figure 1 is an actual dipmeter plot that illustrates the results of structural dip deletion.
82
Figure 1
The dips opposite the pay zone are less than structural dip, so structural dip should be deleted before attempting a stratigraphic interpretation. After deleting the 22° of north-northwest structural dip, the dips in the zone of interest form a southsoutheast dipping red pattern. If structural dip is not deleted prior to stratigraphic interpretation, the interpretation will be in error. Instead of being a fan deposited by a north-northwesterly flowing current, the sand was deposited as fill within an east-northeast, west-southwest striking channel, with the axis lying to the south-southeast. Benefits of Dip Deletion
Structural dip deletion serves as an indicator that the correct structural trend was identified and deleted. The structural dip on Figure 2 was selected as 35° at an azimuth of 90° down to 7150 ft.
Figure 2
83
Below 7150 ft, the structural dip was selected as 35° with an azimuth of 117°. After a structural dip of 35° at 117° was deleted over the entire interval, an apparent northeast structural dip trend remained above 7150 ft. Almost all apparent structural trends disappeared below 7150 ft. This indicates that 35° at 117° was the correct structural dip below 7150 ft but incorrect for the interval above. Another deletion pass was made over the entire interval to delete 35° at an azimuth of 90°. The apparent trend above 7150 ft disappeared, indicating that the correct structural dip was deleted. The incorrect structural dip deletion below 7150 ft produced an apparent southwest structural trend. Another benefit of structural dip deletion is the identification of dips resulting from erroneous correlations. These dips tend to be higher than structural dip, and they typically remain unreasonably high after deletion. The Process of Dip Deletion
If the magnetic recording tape is available, structural dip deletion is a relatively easy process, and a tadpole plot with structural dip removed can be quickly generat ed. If the answer tape is not available, the processing must be recomputed, or a "stereo net" or hand calculation must be performed. Programs are available for the HP-25, HP-41C, HP-75, and the TI-59 calculators. For logs with more than a few points requiring structural deletion, log recomputation is strongly recommended . Deleting Uplift Effects
Gulf Coast salt domes may have undergone several periods of uplift, both contemporaneous and postdepositional. Dips have reversed as the salt being uplifted at one location masked the dip from a nearby salt spine that had been uplifted earlier. Prior dips in directions different from those of current dips indicate the existence of fossil structures in the area. These structures may still be productive. To determine structural dip at any specific time, the effect of structure must be removed a single uplift at a time. The shallowest structural dip should be removed first. The remaining dips indicate the attitude of beds prior to the youngest uplift ( Figure 3 ).
84
Figure 3
After selecting a new structural trend for the shallowest remaining interval, delete the trend. The remaining dips indicate the attitude of the beds at the time of the second-youngest uplift. This process is continued until the end of the dipmeter log is reached.
E5-fault interpretation Introduction
Faulting occurs when beds are in tension or under compression. Such forces produce normal faults and reverse or thrust faults. In areas that have undergone mainly tension (such as the northern Gulf of Mexico), almost all of the faulting is normal. In areas that have undergone both earlier tension and later compression, both normal and reverse/thrust faults may be present in the same well. For dipmeter interpretation, input of the local geology is required to define the actual model. In order for a fault to be detected by the dipmeter, either some sort of distortion must be present near the fault plane or one fault block must be tilted more than the other. When tilting is present, the location of the fault is indicated by a sudden change in dip magnitude and/or direction. Typical forms of distortion near both tensional and compressional faults are shown in Figure 1 .
85
Figure 1
Each of these is covered in this section. Normal faulting (beds in tension) is discussed first and reverse/thrust faulting (beds under compression) last. Beds in tension break, and one block slides downward relative to the other. The block that moves downward is usually called the downthrown block. The upthrown block appears to have been passive during much or all of the faulting process. The attributes of a normal fault include directions of dip and strike, hade, dip angle of the fault plane, and vertical and horizontal displacements (throw and heave). Beds that are in compression break, and one block overrides the other. The overriding block is called the overthrust block; the other block is called the downthrown block. If the fault angle (with respect to the horizontal) is less than 45°, the fault is called a thrust rather than a reverse fault. Attributes of a reverse or thrust fault are dip of the fault plane, strike and over-thrust directions, and displacement. Growth Faults
Faults that were active during the time of deposition are called growth faults. The downward-moving block provided a low area that acted as a sediment sink and accumulated thicker layers of sediments than the equivalent upthrown zone.
86
Tension-created slumping into the downthrown side of the fault also aided the downthrown thickening processes. Downthrown thickening, which begins some distance from the fault, increases toward the fault plane, with the maximum amount of thickening found immediately downthrown. This thickening into the fault, plus sinking of the increasingly heavier downthrown side of the fault, produces a rotation effect that also increases the dip of the beds into the fault. Rollover
The cumulative effect of downthrown thickening, slumping, and rotation, which is called rollover, produces a trend of downward-increasing dips that dip toward the upthrown fault block ( Figure 1 ).
Figure 1
This trend terminates at, or shallower than, the fault plane. It i s this dip trend that allows growth faults to be located, and their attributes i dentified, by the dip-meter tool. The downward-increasing dip trend produces a red dip pattern whose azimuth is toward the upthrown fault block and normal to the strike of the fault. Although not routinely found associated with growth faults, strike slip movement rotates the azimuth in a direction opposite to that of block movement. The vertical extent of the red pattern can be used as an indicator of the minimum displacement of the fault. Displacement is usually greater than the vertical extent of the fault; i t is rarely less. Subsidence Effect
87
When the rate of deposition is greater than the rate of subsidence, a system of progressively younger faults in a seaward direction is created. When the rate of deposition equals the rate of subsidence, a fault with a very large displacement is produced, assuming of course that the system is stable for a considerable period of time. When the rate of deposition is less than the rate of subsidence, progressively younger faults are created in a landward direction. Bed Thickness
Figure 2 is a cross section illustrating the effect of a growth fault on bed thickness.
Figure 2
Wells 1-3 penetrated the upthrown block of a down-to-the-east growth fault. Both sands A and B are the same thickness in both wells. Part of sand A is faulted out in in Well 4, while sand B, which whi ch is still located upthrown, remains the same thickness. Well 5 penetrated sand A in a downthrown position in the rollover zone, so the sand is much thicker than its upthrown equivalent. Sand B was faulted out of Well 5. Well 6 penetrated both sands in a downthrown position within the rollover zone, so they are thicker than their upthrown u pthrown equivalents. The downthrown thickening continues to decrease to the east until Well 9 is reached. This well is located l ocated beyond the eastern
88
limit of the rollover zone, so both sands are the same thickness as their upthrown equivalents. Growth Fault Examples
Figure 3 is an example of a large growth fault which cuts the Vicksburg formation of a South Texas well.
Figure 3
The fault, which cuts the well at a depth of 14,890 ft, is i s downthrown, or dips, to the southwest. Therefore the rollover zone, which dips toward the upthrown block, dips to the northeast. The rollover zone (the zone that creates a downward-increasing dip trend) extends upward to 13,750 ft; the minimum displacement di splacement of the fault is approximately 1000 ft. Figure 4
89
Figure 4
is an example of an offshore Louisiana fault whose displacement is smaller than that of the fault in Figure 3 . This offshore Louisiana Miocene example illustrates the rollover created by a small growth fault. The fault, which by correlation has a displacement of only 120 ft, is located at 12,638 ft and is i s downthrown, or dips, to the south-southeast and strikes normal to the pattern dip direction, or eastnortheast west-southwest. Because Because of shattering of sediments near the fault plane, only a scattering of dips were recorded immediately downthrown. Some fault examples show even more extensive shattering and washing out of the hole on the downthrown (most active) side of the fault plane. The dip trend which begins at the base of the blank zone is recorded from the upthrown block, so the fault cut is no deeper than the bottom of the blank bl ank zone. Usually it is picked at the base of the blank zone. One of the "eyeball" indicators of a possible missing section, which is present in these examples, is a borehole dogleg. Any time the bit crosses a formation compaction change it reacts by creating a change in the amount and/or direction of well drift. Since compaction changes are almost always present across a fault or unconformity, a change in well drift azimuth and/or magnitude can (but does not always) indicate the presence of a fault or unconformity.
90
As long as a fault continues to grow, downthrown thickening is produced. This in turn produces an increasing-with-depth red dip pattern. During periods of relative nongrowth resulting from changes in depositional processes, the beds are rotated at a constant rate. This in turn produces constant dip zones within rollover-created red dip patterns. Figure 5 is an example of such a fault.
Figure 5
The increasing-with-depth rollover zone begins at about 8700 ft. From 9100 to 9250 ft the dip trend remains constant. From 9250 down to 9350 ft the dip again increases downward, indicating a period of renewed growth. Structural Dip Imprint
The dips from a zone of distortion are changed when structural dip is imprinted on them, so it may be necessary to remove structural dip before determining the attributes of a fault, just as it is necessary when making stratigraphic interpretations. Figure 6 is a theoretical example of the appearance of a dip-meter plot when structural dip is imprinted over dips created by rollover.
91
Figure 6
(A) shows a west-dipping red dip pattern created by rollover into a down-to-the-east growth fault. (B) shows that a moderate amount of east structural dip was added to the west dipping red pattern of (A). The resultant dip pattern, moving down the hole, is a decreasing dip trend or blue pattern. The dip decreases to zero, then increases in the opposite direction until a maximum is reached at the fault. A typical red pattern is formed below the zero crossing point. Below the fault cut, only east structural dip is seen. (C) illustrates an even stronger east structural dip imprinted over the west-dipping red pattern. At the point where the west-dipping structural dip and the strong east structural dip start to oppose each other, a decreasing dip trend, or blue pattern, begins. In this case, the trend decreases down to the fault cut but never quite reaches zero dip. As soon as the fault is crossed, only strong east structural dip is recorded. Figure 7 is a dipmeter example of the first type of imprint, where a strong red pattern opposes moderate or low structural dip in the opposite direction.
92
Figure 7
Structural dip is about 10° south-southeast. This opposes the northwesterly dipping red pattern dipping into a down-to-the-southeast growth fault. The resulting dip patterns are ones of decreasing dips (blue pattern) from 13,700 ft down to the zero crossing point at 13,790 ft, then ones of i ncreasing dips (red pattern) in the opposite direction down to the fault cut at the base of the blank zone at 13,880 ft. Deviated Wells
Deviated wells are sometimes drilled parallel to fault planes. As a well periodically gets closer to the fault and, in some instances actually bumps the fault, the dips increase and then decrease. Some dip scatter is created by formation shattering and hole conditions near the fault. Platform wells may be deviated in a direction and an angle such that they cross normal faults from the upthrown to the downthrown sides instead of i n the usual manner. Post-Depositional Pre-Compacted Faults (No Distortion)
93
Normal faults that occur after deposition but before formation compaction usually exhibit no distortion near the fault plane. Such faults can be recognized on the dipmeter plot only if a change in structural dip occurs across the fault. Because there is a change in the degree of formation compaction, the borehole doglegs even though there is no distortion of the beds near such a fault. The sudden downward decrease of structural dip is one of the "eyeball" indicators used to differentiate between faults and unconformities. Most of the time, i n areas that have not undergone strong tectonic deformation, the downward decrease indicates faulting. In order to have lower dip below an unconformity, two different centers of uplift are required. Instead of downthrown rotation, some faulted areas have undergone rotation of the upthrown fault block. This creates a sudden structural dip increase in a downward direction. This is the same dip pattern created by the presence of an angular unconformity or a rapid, postdepositional structural uplift. Therefore, other information is needed to determine which of the three features is present when a sudden downward increase in structural dip is noted on the dipmeter plot. A lack of distortion near a fault plane can occur with both tensional and compressional faults. Therefore, unless there is a structural dip change at the fault cut, faults of this class cannot be seen on dipmeter pl ots. Post-Compaction Faulting (Drag)
Normal faulting that takes place after some degree of deformation has occurred usually develops drag, or beds dipping in the same direction as the fault, near the fault plane. In some areas, drag is found only on the downthrown side of the fault; in others, drag may be found in beds on both the downthrown and upthrown sides. Since the relative motion of the upthrown and downthrown fault blocks creates a drag zone whose dip is in the same direction as that of the fault plane, the maximum dip of the resulting red dip pattern may be used as a minimum dip of the fault plane. Local experience is used to determine whether or not th e maximum dip of a draggenerated red pattern is in fact a reasonable value for the dip of a fault plane. Figure 1 is a theoretical example of a normal fault with drag only on the downthrown side.
94
Figure 1
The red pattern, which was generated by the downthrown drag zone, dips in the same direction as the fault and normal to the strike of the fault. The maximum dip of the pattern may be used as the minimum dip of the fault plane. As happens with a growth fault, the hole doglegs within a hundred feet or so of the depth at which the fault cuts the well. Since drag is present only on the downthrown side, structural dip is recorded on the upthrown side of the fault. Figure 2 illustrates a normal fault with drag in both the upthrown and downthrown beds adjacent to the fault plane.
95
Figure 2
The dip in the downthrown block created a dip pattern similar to the one in the previous example. However, drag, which is also present in the upthrown beds, creates a pattern of downward decreasing dips or a blue pattern. The fault cuts the well at the junction of the two patterns. Once again, the maximum dip of the red pattern may be used as a minimum dip of the fault plane. The amount of rollover present on the downthrown side of a nonburied growth fault decreases upward. It disappears at a point corresponding to the time at which the fault ceased to be active. The amount of drag created by any period of movement remains relatively constant over the entire interval. The termination point may be a point corresponding to the end of the active faulting period, or, if buried, to an unconformity. Faults with Hybrid Dip Patterns near the Fault Plane
Some faults begin as growth faults with downthrown rollover zones. Either continued movement along the fault plane or movement that began after compaction occurred then created a downthrown drag zone. Since rollover and drag-generated dips oppose each other, dip patterns like those illustrated in Figure 1 are created by continuing or later fault movement.
96
Figure 1
A red, or downward-increasing, dip pattern begins at the point at which the hole penetrated the rollover zone. The dips increase down to the point at which the dragzone dips begin to oppose the rollover dips. The trend then decreases downward to the zero crossing point. Below that point, another red pattern dipping in the same direction is formed. The dips continue to increase in magnitude down to the fault cut. On the upthrown side of the fault the dips may return immediately to a structural trend, or indicate an upthrown drag zone. Buried Faults
Growth faults die out gradually in an upward direction. Postdepositional faults may extend to the surface, where they create cliff-like scarps, or they may end abruptly at an erosional surface. Such faults are called buried faults. In addition to ending abruptly at an unconformity, disconformity, or diastem, buried faults may change displacement across deeper unconformities. The buried -fault creation process is illustrated in Figure 1 .
97
Figure 1
First, a fault extending to the surface is formed. Later, erosion removes the elevated portion of the upthrown block; the land surface is once again level across the fault zone. The original amount of uplift is labeled a. Still later, deposition begins again, and horizontal sediment layers are deposited above the erosional surface. Subsequently, movement again occurs along the fault plane. This movement creates a displacement labeled b. The displacement of the beds below the unconformity is now a + b. Erosion has removed the beds that were originally displaced by amount a. Therefore, only displacement b extends across the unconformity on the upthrown side. If erosion occurs again, the beds that were uplifted above the surrounding land surface will be eroded to a flat surface. The displacement below the shallowest unconformity is equal to b. The displacement below the deepest or oldest unconformity equals a + b. This cycle may be repeated. Figure 2 illustrates a buried fault example from eastern Venezuela.
98
Figure 2
A down-to-the northwest growth fault is located at 3842 ft. The fault terminates at a depth of 3760 ft, which i s the unconformity separating the Paleozoic from the Lower Cretaceous. This growth fault was originally active in Paleozoic time. Any scarp that exi sted was eroded before Lower Cretaceous sediments were deposited above the unconformity. No subsequent movement occurred along the fault plane. Since both types of distortion (rollover and drag) commonly found near normal faults create similar dip patterns, some local knowledge is useful in determining which type to use when making an interpretation. In any given area one typ e of distortion is found near most of the area faults. For example, in the northern Gulf of Mexico and in Nigeria, rollover is the dominant distortion type. In Mississippi and North Louisiana, drag is most often found on the downthrown side of normal faults. Here are some rules of thumb for determining the type of distortion present near a normal fault: · If the vertical extent of the downthrown mega-red dip pattern is more than 200 ft, rollover is assumed. Normal fault drag rarely extends vertically more than 200 ft.
99
· When the vertical extent of the mega-red dip pattern is l ess than 50 ft, drag is assumed. In areas where rollover dominates, this assumption can lead to incorrect interpretation about 30% of the time, since many small growth faults do exist. When the vertical extent of the mega-red pattern is between 50 and 200 ft, use the dominant type of distortion known to exist in the area. Deviated wells may cause the extent of the dip pattern to be expanded by 50% or more. Hole deviation must be taken into account when using the vertical extent of a mega-red pattern as input into one of the rules of thumb. Semicontemporaneous antithetic fault systems that help accommodate rotation are often found associated with growth fault systems. These faults usually exhibit downthrown drag that creates red dip patterns dipping toward their downthrown blocks. The dipmeter example in Figure 3 shows three such faults.
Figure 3
These faults are down-to-the-northwest so the northwest dipping red dip patterns found on their downthrown sides are the result of drag rather than rollover. These antithetic faults are dipping into a large down-to-the-southeast growth fault which is below the total depth of this well.
100
Compressional Faults
Faults that result from compressional forces may, depending on the fault angle, be called reverse or thrust faults. The fault angle of reverse faulting is 45° or more, while thrust fault angles are less than 45° The main form of distortion found near a reverse or thrust fault is drag on both sides of the fault. Drag, which is the result of movement of compacted beds, may be additionally modified by horizontal movement, or strike slip. The compressional fault attributes that may be available from dip-meter plots are depth, strike, direction of overthrust, and fault angle. Figure 1 illustrates the expected dip patterns near compressional faults.
Figure 1
Such faults commonly show up very well on dipmeter plots. A mega-red dip pattern is usually found in the overthrust block. Its azimuth i s in the direction of overthrust, assuming, of course, that no strike-slip has occurred. The downthrown pattern is one of downward-decreasing dips, or a blue pattern. These dips also point in the direction of overthrust. Both dip patterns are the result of drag on both sides of the fault. The fault is located at the junction of the red and blue patterns.
101
Figure 2 is a dipmeter example from western Venezuela showing a reverse fault.
Figure 2
Structural dip above 7800 ft is 12° southeast. From 7800 to 8450 ft westerly dipping beds in the overthrust drag zone oppose the southeast structural dip. This produces a decreasing-with-depth pattern down to the zero crossing point at 8030 ft; the pattern then increases downward to the fault cut. The dip azimuth reverses across the zero crossing point. The maximum dip of this example is recorded at the fault. The blue pattern generated by the downthrown drag zone decreases rapidly. The dip patterns near major compressional faults are rarely symmetrical. The overthrust pattern usually has the greatest vertical extent. If the displacement is small (i.e., less than 100 ft), the red and blue dip patterns tend to be more nearly symmetrical. The dip direction on both sides of the fault is the same as the direction of overthrust, which is to the west in this example. The strike of the fault, north-south, is normal to the direction of dip patterns. In this example, the direction of structural dip i n the overthrust block and the direction of overthrust are opposite, so the dip trend decreases to the zero crossing
102
point and then increases in the opposite direction. Had the di rection of structural dip and of overthrust been the same, the dip trend would have continued to increase in the overthrust drag zone. Horizontal movement of one block relative to the other (strike-slip) may also occur. The drag-created dip patterns, which dip in the direction of over-thrust, would be modified by any horizontal movement. When such movement occurs, the drag dip patterns are rotated in the trailing direction, which is opposite to the direction of movement, and so no longer indicate the direction of overthrust. Both compressional faults and overturned folds create repeat sections on logs. If the repeat is right side up, it is the result of faulting. If one repeat is upside down or a mirror-image of the other, it is the result of folding.
Exercise No. 1
Figure 1
See Figure 1 . There is a missing section in this well between 14,000 and 14,100 ft.
103
Where is the dip change that indicates the location of the fault? Where is the top of the rollover zone or top of the mega-red dip pattern? What indicator suggests the presence of rollover rather than drag? What is the minimum displacement of this growth fault? In what direction is the fault downthrown? What is the strike? Is a dogleg present?
Solution The dip change corresponding to the base of the rollover zone is at 14,049 ft. The top of the mega-red dip pattern, which corresponds to the top of the rollover zone, is at 13,680 ft. The vertical extent of the red pattern is more than 200 ft; therefore, rollover rather than drag is present. The minimum displacement of the fault, which is equal to the extent of the red pattern, is 370 ft. The rollover zone dips into the fault; therefore, the fault is downthrown to the south and strike is EW. There is a dogleg. The hole is vertical at 13,000 ft and drifts to the SE a maximum of 3-1/2° at 14,000 ft. The hole then begins to straighten.
D6-reef interpretation Introduction
Buried topography may significantly influence the thi ckness, sedimentation, and dip attitude of beds overlying topographic features. One of the first stratigraphic applications of dipmeter data was to determine the positions of well s drilled on buried topographic features. The factor contributing to the interpretation of these situations is the drape of beds over the underlying buried topography. Although this chapter deals primarily with reef interpretation, many of its basic principles apply to
104
buried ridges, knobs, and depressions covered in the next chapter. The differences between them lie in the application of the interpretation. The dip of reef surfaces is interpreted from the drape of sediments over the reef, particularly where the reef underwent considerable vertical growth. Dipmeter data obtained very near the reef or on the reef slope exhibit dip anomalies that help to describe the reef slope. The magnitude of the dip at any point above a reef varies depending on the following: · the slope of the reef surface · the height of the reef above the surrounding platform · the distance of the point above the reef surface · the type of rock above the reef · the total historical overburden · the position over the reef at which the measurements were taken (crest, flank, or toe)
Figure 1 shows the cross section of a barrier reef complex.
Figure 1
Dipmeter interpretation will be described for wells drilled
105
· in pinnacle reefs overlain by shale · in pinnacle reefs overlain by low-compactibility formations · on the forereef slope · on the crest of the barrier complex The dip patterns may be somewhat different in each of these cases because of the relief of the feature as well as the lithology of the enclosing formations. Well-correlation and dip data have shown that the slopes on reef flanks can vary from 2° or 3° to as high as 45°. As evidenced by current reefs, higher slopes are possible, but are not generally observed in the subsurface. It is possible that erosion of steep and irregular slopes prior to burial produced reef fl anks of reduced dip angle. Large accumulations of reef talus material near some reefs may support this premise. Of the factors influencing the dip above a reef, the most important are reef topography and the compactibility of overlying beds. If all dip patterns above reefs conformed to the same model, interpretation would be straightforward. Unfortunately, identical dip patterns may imply quite different reef slopes in different environments. Figure 2 is a photograph of a pinnacle reef in a Cambrian zone in West Texas.
Figure 2
Some significant features show in the enclosing beds. Note the drape of the overlying beds in the flank position with the direction of dip sharply away from the reef mass.
106
It is useful to note that subsidence of the mass into the underlying plat form has caused dip below the reef to be toward the reef mass itself. Finally, within the reef there is generally a lack of distinct stratification. Comments on Reef Interpretation
The tectonic and geological history of an area generally determines a maximum height of buried topography above a reference datum. This is particularly true of reefs, because the controlling conditions for vertical accumulation may prevail over a large portion of a basin. Further, unfavorable conditions cause cessation of growth over large areas, and the limited vertical height becomes common to many reefs simultaneously. Therefore, a maximum expected height of reef crest above a datum may be well established. Reef falls, or pinnacles that for whatever environmental reason ceased growth early, may have any thickness less than maximum. The length (and shape) of the slope pattern in wells drilled on the flank position may be interpreted qualitatively as a guide to the vertical size of the reef. "Typical" dipmeter patterns for the area are very useful in this respect. The accuracy of interpretations discussed in this chapter depend largely upon knowledge of sediment compaction around reefs and other buried topography. The relatively simple procedures employed to estimate compaction may not apply directly to all areas. Extrapolation of dip assuming a linear slope may be used best where experience with seismic and well-to-well correlations support this approach. Extrapolating updip beyond the elevation geologically possible or likely for a particular feature could be misleading and expensive. The basic concept should prevail, however, and persons employing these basics in their analyses are advised to consider all available data and experience.
Reef Interpretation The pattern of drape over a reef may be determined by a number of factors, including
107
compaction
compaction with deposition
solution of surrounding salts
solution with deposition of overlying sediments
gypsum to anhydrite conversion
combinations of the above
Compaction In most circumstances, compaction plays the key role in causing drape over buried topography. It is useful therefore to refer to a model to understand and interpret the dip patterns. The simplest model, and one with good independent support from well correlation, is ill ustrated in Figure 1 , which shows a simple reef mass of constant slope of 30° surrounded by shale.
Figure 1
If we know the compaction factor of the shale, and we assume that all shale compaction occurred after deposition and that the reef is rigid, then we can accurately calculate the present attitude of the shale bedding plane. For this model it is assumed that present shale thickness is 50% of the original precompaction thickness; therefore, the resulting compaction factor is 0.5. From these assumptions we can conclude that the dip of the shale bedding plane is the angle of the tangent, which is 50% of the tangent of the reef angle. The equation is -1
tan (0.5 tan 30°) = shale dip = 16.1° The general equation is -1 shale dip = tan [(1 - C) tan reef dip] where:
108
, and 1 - C = compaction Solving for the reef dip, which cannot be directly measured, we have
-1
reef dip = tan
reef dip = tan -1
In this example, with a measured shale dip of 16.1°, we can calculate the reef dip to be 30°. Considering the range of local changes in compaction due to lithological changes, locally changing reef slope, or the fact that compaction may not be totally postdepositional, a simple solution is to divide the shale dip by an estimate of compaction. reef dip = = 32° As shale-dip values increase, as in the case of ver y steep-sided reefs, the sim plified solution becomes less accurate and the general equation should be used; however, slumping, fracturing, and sliding may render interpretation more difficult and the precision of reef slope less significant. In the previous simplistic model where all compaction was assumed to occur after deposition, the theoretical dip pattern would be a constant 16.1°. Where beds are now essentially parallel, they may be considered to have been paralleled during deposition and therefore equally compacted. This model is applicable in these cases. In general, however, the drape of beds over a reef produces a red pattern on the dipmeter plot if the well i s drilled in the flank position. The existence of the red dip patterns implies that compaction cannot be assumed to be postdepositional except over limited intervals where the dip magnitude is relatively constant. In this case, the compaction may only be invisible over short intervals because of the low rate of change of dip with depth. Provided other factors remain constant, reefs with large relief tend to produce long red patterns above; those with low relief produce shorter patterns. In any case, it is the analysis of the red pattern that allows us to calculate the reef slope.
Estimating Height of Nearby Reefs Two basic dip patterns have been observed in shale-enclosed reefs. The first is a long, slowly increasing red pattern. The dip of the shale above the reef is some fraction of the reef dip, and the reef dip is estimated using the earlier derived equation. The i nterval from 1315 to 1365 m shows little dip change, which indicates that it was deposited prior to most of the compaction process. This zone is therefore a candidate for the simple model approach. The second dip pattern observed in shale-enclosed reefs is characteristic of steep-sided, probably curved surfaces of high-relief pinnacles ( Figure 2 ).
109
Figure 2
Statistically, this pattern has two sections. The lower section is characterized by a sharp slope pattern immediately above the reef. The upper section i s a long and slowly decreasing slope pattern. Where the sharp slope pattern and the slowly decreasing dip pattern join is the approximate height of the nearby reef.
Estimating the Reef Slope The reef slope should be estimated using two methods. First, applying your knowledge of compaction to the upper dip section, calculate the dip. Second, extrapolate the lower red dip pattern to the reef surface. This dip is a good estimate of the reef dip at the contact, but it may not persist over long horizontal distances. If the two dips agree within a few degrees, confidence in the answer is high. If the two dips do not agree within a few degrees, they at least establish a range of possible dip. Knowledge of the seismically defined size and shape should be integrated into the final solution. Where overlying beds are of low compaction, the overlying dip also is l ess. For example, a formation with compaction of 20% and a dip over the reef of 4° would imply a reef dip of approximately 20°. In this case the red pattern would not be nearly as striking as in the previous examples. Some interpreters may be tempted to find the exact depth of the contact, and they may seek a particular tadpole to define the dip of the surface. This approach can give quite erroneous results, because of the local irregularities existing on any weathered surface. These irregularities may be of a size on the order of the borehole diameter. In this case the dip at the contact would be e ntirely misleading, and the general dip trend is better defined from dips sufficiently above the surface, because the small features would have been compensated by sedimentation and compaction.
110
As a general rule, dip patterns should be extrapolated horizontally in the same order as the vertical length of the pattern. This implies that any single- dip tadpole should not be extrapolated beyond the borehole.
Reef Detrital Material Dip within a reef is not generally very well ordered. There is one exception, however: reef detrital material. Reef detrital material is often found in accumulations near the base of the reef, and dip patterns within this material may have the appearance of foreset bedding. This bedding may have dips greater than the reef dip, but the direction should be generally downslope. This information is particularly useful where it supports draping dip in overlying beds and in situations of low reef dip or low compaction. The only indication of this detrital material may be from the dip-meter plot, as there is little mineralogical distinction between detritus and the main reef mass. This information may be significant when estimating the depth of the reef top, particularly where the detrital section is quite thin.
Reefs Surrounded by Salt Reefs surrounded by salt are not likely to exhibit strong dips in the overlying sediments. If salt solution occurs simultaneously with or subsequent to deposition of these sediments, a dip pattern is produced. This pattern is determined by the rate and timing of the removal of salt. If salt removal commenced after deposition of some of the overlying beds, then these beds would have collapsed to more or less conform to the reef surface. Their dip would then be equal to the reef dip, and a pattern of essentially constant magnitude would be formed. This is illustrated in interval A in Figure 3 .
Figure 3
111
If the beds of i nterval B were being deposited during the removal of salt, a red pattern over that interval would be produced on the dipmeter plot. Interval B could be expected to thicken in the downdip direction. The dip pattern in the figure would be further modified by compaction during and after deposition, but the basic pattern would be recognizable. Reef dip in this case would be approximately equal to or slightly greater than the constant dip value of interval A. If salt removal occurred contemporaneous with the deposition of some of the overlying beds, most of the interval above the reef would exhibit a red pattern, as illustrated in Figure 4 .
Figure 4
The dip of the reef would approximately equal the trend of the red pattern extrapolated to the reef contact. Extrapolation of red patterns of drape over reef or weathered surfaces is necessary, because dips near the surfaces may be difficult to ascertain due to bedding destruction by fractures, slump, or sliding and local irregularities of the surface.
Dipmeter Interpretation Figure 5 illustrates the dipmeter pattern of a well drilled in the flank position of a reef where salt removal around the reef played a significant role in the final dips of the overlying beds. The pattern may be analyzed as follows:
112
Figure 5
Interval A contains a section of relatively constant dip with an average value of 15° to 18° east. This interval was probably deposited prior to salt removal, and it represents the minimum dip of the reef. Interval B contains a red pattern that indicates the period of salt removal. Interval C contains a long, gentle red pattern that finally disappears well above the top of the figure. This pattern is probably a reflection of compaction during deposition, and it would be superimposed on the patterns of intervals A and B.
Because regional dip in this area is less than 1° to the southwest, the consistent east dip above 3900 ft is interpreted as part of the overall drape on the reef. The long drape feature, over 1000 ft in length, suggests that the reef feature is not small. Based on the seismic interpretation and the dip data, it was decided to whipstock the well to contact the reef 300 ft to the west. The result was to gain 90 ft of el evation on the reef. A straightline correlation between the two contacts implies a reef dip of 16-1/2°, approximately the mean value of the tadpoles in interval A.
Exercise No. 1
113
Figure 1
Figure 1 shows a Devonian reef which has a long, li near slope in this area. Regional dip is 1° SW. Compaction is 40 to 50%. What is the reef dip? What is the direction of dip? How far and in what direction must an offset be drilled to gain maximum reef?
Solution The long red dip pattern terminates at 6° just above the reef talus interval and would extrapolate to the reef surface at about that angle. If compaction = 50%,
reef dip
114
If compaction = 40%,
reef dip Direction of dip = NE. Additional reef available is 190 ft. If dip = 12°, go 890 ft. If dip = 15°, go 710 ft.
Calculated from offset = As shown in the accompanying figure, the actual offset well gained 160 ft in 700 ft offset from a reef dip of 13°. Compaction of the shale calculates to be 40% from the data in these wells.
E7-unconformity interpretation Introduction This chapter discusses two distinctly different types of topography. The first, buried unconformities, refers to dipmeter interpretation of beds overlying topographic features such as ridges, knobs, or depressions. The second, angular unconformities, involves the identification and interpretation of sediments deposited on eroded surfaces, where the underlying beds are tilted with respect to the overlying beds.
Buried Unconformities Dipmeter interpretation in traps formed below topographic features is similar to the approach used for reefs. Compaction of overlying sediments and extrapolation of dip patterns are handled in a similar fashion. Where the trap is formed above the unconformity, however, more data are re quired to describe the geometry of the reservoir. Figure 1 is an example of such a case.
Figure 1
115
The unconformable surface is Precambrian granite. Traps may be formed where sand has been deposited above or along the flanks of granite ridges or knobs. The overburden, whether shale, sand, or evaporites, drapes over the highs. The dip data on the flanks typically exhibit a red pattern terminating near the unconformity. This pattern points in the direction of thickening of the overlying sediments, and therefore in the direction of potentially thicker reservoir rock.
Determining Reservoir Geometry Figure 2 is a schematic of a cross section and associated formation dips.
Figure 2
Note that the dip of the sand top is less than that of the unconformity. These two dips are significant in determining reservoir geometry. In this example, the sand thickens in the east direction, but it is structurally lower. In the west direction the sand would be higher, but thinner or nonexistent. Although the sand thins updip, the pay section thickens in this example. The slope of the top of the sand may be determined directly from the dipmeter plot, but the slope of the base must be computed with the same logic described in the reef section. From these two dip values, the sand thickness in both the updip and the downdip directions can be calculated. Experience has shown that a compaction of 0.5 may be used in this area. Therefore,, the dip of the unconformity would be roughly twice that of the reservoir top. With knowledge of any water contact, marginal or wet wells may be whipstocked or offset to encounter the optimum hydrocarbon section.
116
The left side of Figure 3 shows an induction log and computed dipmeter through a section overlying a granite unconformity.
Figure 3
This unconformity is anomalously higher than was expected. Zone A, the prospective zone, contains no sand development. The dipmeter plot shows a coherent S80E dip. The dip is at an average of approximately 5°, and, applying 50% compaction, it is calculated that the granite drops off to the east at a rate of 10°. Regional dip in this area is less than 1°. The east dip trend gradually decays to regional dip far above the top of the granite. This long red pattern can be interpreted in the same manner as that for reefs. From local data, the operator could surmise that sand is developed and hydrocarbon does exist in the area. Based on this information, the operator decided to whipstock the hole downdip. The induction log on the whipstock hole is shown on the right side of the figure. Seismic information and experience with other wells suggests that an additional thickness of 30 ft for zone A would increase the chances for reservoir development.
Determining whipstock Distance and Direction Assuming linear slopes, the following is the equation for determining additional thickness: q2
q1
additional thickness = D (tan - tan ) where: q2 is unconformity dip q1 is top dip
117
D is horizontal distance
In this case, the unconformity dip was calculated to be 10°. The zone-top dip read from the plot was 5°. Using the equation, the additional thickness of 30 ft would require that the offset distance D be 337 ft, provided the offset direction is directly downdip. The well was whipstocked 500 ft S35E, approximately 45° further south than the dipmeter plot indicated. If the strike of the high feature persists in the N10E-S10W direction, as was suggested by seismic, then the 500-ft offset at S35E would contact the unconformity at the same depth as a 350-ft offset at S80E ( Figure 4 ).
Figure 4
The cross section between the original and the whipstock holes shows an increase of 34 ft in zone A thicknessonly 4 ft more than predicted. Approximately 20 ft of the additional thickness was composed of sand, all of it oil bearing. Assuming that the slopes are approximately linear and that the feature is e longated, the linear interpolation from the 350-ft whipstock produces a 9° dip for the unconformity and a 3.5° dip for the zone top. These figures are not in great variance with the dips calculated from the dipmeter plot. In this case, it was essential for the operator to know that the granite feature was elongated in the N10E-S10W direction in order for a S35E whipstock to be successful, since that was not exactly the dipmeter-indicated direction. If a feature is more round than elongated, "splitting the difference" between seismic- and dipmeter-indicated offset directions can lead to very unsatisfactory results. The dipmeter data, if of good-quality tadpoles and persistent patterns, is a specific measurement of a specific geographic location, and should provide better results than other methods that may be more i nterpretive. Compaction Estimation In wildcat areas the compaction factor must be estimated, and experience shows that a figure of 40 to 50% generally works well. However, developed areas with whipstocked holes or multiple wells
118
drilled on the same feature (or similar features) provide the information to back-calculate compaction values. The previous case contained excellent data for calculation of compaction. Using the logs from Figure 3 , correlations A, B, C, and D indicate that very little thickening occurred downdip through that section. The major thickening was in the interval between the unconformity and correlation A; thus, the dip of correlations A to D can be ascribed to compaction of the sediments below A, and correlations A to D were deposited basically flat and parallel. The following method for estimating compaction applies equally well to reefs or unconformities. Figure 5 shows a rigid sloping mass penetrated by two wells, A and B.
Figure 5
Log correlation is made on a bed as close as possible to the rigid mass in Well A. The following assumptions are then made:
The correlated bed was flat at the time of deposition.
Compaction rate of sediment a in the high well was identical to that of sediment e in the low well.
All compaction occurred after deposition of the correlated bed.
Based on those assumptions, the following compaction estimate can be made: If C = ratio of present to original thickness, then, from the data in the figure,
119
And
Also, h+a=e Therefore,
Then,
By definition, compaction = 1 - C Applying this principle to the in duction log example in Figure 3 : d = 45, b = 14, and h = 54 C calculates to be 0.57, and the compaction, 1 - C , is 0.43, or 43%. Multiplying the zone-top dip by 2 to obtain the unconformity dip would have given a fair approximation. This type of analysis should only be applied to similar lithological sequences in the same general area. It is not necessary to use two wells on the same feature, as long as the li thological sequence and compaction can logically be assumed to be the same. Inaccuracies would be encountered in situations with significantly nonlinear surfaces or with extrapolations beyond those that are geologically sound.
Angular Unconformities
Angular unconformities are created when Previously deposited sediments are removed by erosion and redeposited in another area. Erosion is commonly considered to be a land-surface phenomenon, but it can also occur in a subaqueous environment, where wave action and currents erode sediments. Angular unconformity implies an angle difference of One-half degree or more between the beds below the erosional surface and those above. If the angle is less than one-half degree, the beds are essentially parallel, and the separating surface is called a disconformity. Figure 1 illustrates the creation of an angular unconformity.
120
Figure 1
Sediments are initially deposited in layers. The layers are shown horizontal in the figure, but bedding planes may be deposited at any angle from flat to the angle of repose. Next, structural uplift elevated the layers from an environment of deposition into one of erosion. Some sediments were removed and redeposited elsewhere, and the erosional process produced a more or less flat surface. Conditions changed again and deposition recommenced on th e erosional surface. There is a difference in attitude between the underlying beds, which were tilted by structural uplift, and the overlying beds, which were deposited flat. This process may be repeated. The end result of deposition -up-lift-erosion-deposition cycles is two or more dip trends decreasing upward in steps. Unless the area underwent major tectonic activity, the usual dip pattern across an angular unconformity is higher dip below, lower dip above. A di p below an angular unconformity that is lower than that above requires two centers of upli ft. Figure 2 is a dipmeter log example through an angular unconformity.
121
Figure 2
Structural dip below the unconformity is 18° south; structural dip above is 9° south. This example also illustrates a feature found beneath erosional surfaces much of the time: a blue dip pattern resulting from weathering below the erosional surface. This zone may contain fractures that flatten downward or beds steepening immediately below the erosional surface as a result of swelling clays or overburden removal. Localized steepening beds can be seen in outcrops. When a topographic surface is developed by erosion, drape of the overlying beds can occur, as illustrated in Figure 3 .
122
Figure 3
The drape generates a red pattern terminating at or above the unconformable surface. Depending on the depositional environment, a blue pattern may be present instead. Exercise No. 1
123
Figure 1
Figure 1 is an example of a major unconformity. Where is the unconformity? Is a weathered zone present? Is a dogleg present? What is the structural dip at the top of the example? What is the structural dip at the bottom of the example? What is the structural dip between the top and bottom of the example?
Solution The major dip increase in this example occurs at 11,450 ft, where the structural trend increases from 3° SE to 14° SE.
124
There is a weathered zone. The high-angle blue dip pattern immediately below the unconformity was caused by weathering. The other blue patterns result from stratigraphy. Structural dip at the top of this example is flat. There is another angular unconformity at 11,150 ft, where the dip trend changes from flat to 3° SE. There is a hole dogleg between 11,100 and 11,200 ft.
E8-depositional interpretation Eolian environment Introduction
A dune is a hill of sand, deposited by wind, that rises to a single summit and possesses a slip face. Dunes may be various sizes and shapes depending on wind conditions, sand type, and sand supply. Dunes may be oriented perpendicular to the prevailing wind (e.g., barchan and transverse dunes), parallel to the prevailing wind (e.g., seif or longitudinal dunes), or they may acquire complex formations (e.g., dome-shaped or star-shaped dunes). Dunes are the most impressive and important feature of a desert environment. They are also important geologically. The Nugget formation of the western United States, the Norphlet formation of the U.S. Gulf Coast, and the European Rotliegendes formation form important hydrocarbon reservoirs. The eolian Botucatu of Brazil is a large freshwater aquifer. Much of the following information is based on the work of Reineck and Singh. (See Depositional Sedimentary Environments, 1980, New York: Springer-Verlag.) Figure 1 illustrates some typical dip patterns in eolian environments.
125
Figure 1
Parabolic Dunes
Parabolic dunes are U-shaped sand ridges with their concave side toward the wind. Parabolic dunes are associated with blow-up features. The middle part of the parabolic dune moves forward ahead of the arms, which are believed to be hi ndered by vegetation. The characteristic dip pattern of parabolic dunes is a red pattern at the center of the dune dipping in the direction of the prevailing wind ( Figure 1 ). The dips found near the tips of the arms may be skewed more than 90° from the direction of the prevailing wind. Foreset laminae of parabolic dunes are low-angled relative to other dune types. The foreset laminae are characteristically concave-downward as a result of slip-face shape and the presence of vegetation. The azimuth spread of the di p of foreset laminae is rather large-up to 200°. Barchan Dunes
126
Barchan dunes are crescent-shaped sand mounds occurring as isolated bodies, in chains, or in colonies of individual dunes coalescing into complex forms. B archan dunes are formed by a unidirectional wind, and they migrate by sand avalanching on the slip face. The extremities or horns of a barchan extend forward and downwind, as the horns migrate more rapidly than the main body. Simple barchans may be made complex by the coalescence of many sand dunes. In regions where the wind blows periodically from directions other than that of the prevailing wind, small, oblique slip faces may be produced, but the general dune form and direction of movement are retained. When interpreting eolian zones, note that the structural dip i s at the left edge of the tadpole cloud unless the log is from an area that has undergone appreciable structural uplift. The general dip direction is in the direction of the prevailing wind. Near the horns, dips are less and may be almost 90° to the direction of the prevailing wind. The zones of crossbedding dips with constant magnitude near the center of barchan dunes reflect the angle of repose during deposition. The angle-of-repose zones are underlain by fore set-generated blue dip patterns. The minimum dip found at the base of these blue patterns reflects the dip of interdunal layers and approximates structural dip. The foreset laminae of crossbedded units in barchan dunes are mainly planar (tabular types with a dip from 20 t o 35° in the central part). Figure 2 is a dipmeter log through the eolian Rotliegendes sand in Holland.
127
Figure 2
The dip patterns are typical for barchan-type dunes. Dome-Shaped Dunes
Dome-shaped dunes are low, circular sand ridges lacking a well-developed, downwind, steep slip face. Dome-shaped dunes develop when dune height i s checked by a strong, unobstructed wind. The characteristic internal structure displays low-angled foreset laminae. Dome-like dunes produce red dip patterns similar to parabolic dunes. The central portion of the dome contains dip in the direction of the prevailing wind. The dip left and right of the center may be skewed as much as 75° to the prevailing wind. Transverse Dunes
Transverse dunes are elongate, almost straight sand ridges p erpendicular to the predominant wind direction. These ridges are regularly spaced and are separated by broad interdune areas that may have developed as inland sabkhas.
128
Transverse dunes originate in areas of inland sabkhas, where the damp sabkha surface inhibits the growth of barchan horns. When the interdune sabkhas eventually disappear, a sand sea with transverse dunes may be produced. Dip patterns produced from transverse dunes are similar to the patterns found near the center of barchan dunes; the patterns consist of zones of angle-of-repose dips underlain by foreset-generated blue patterns. Crossbedded units are mostly of the planar-tubular type. The foreset laminae are relatively long, even, and high-angled. The azimuth spread of foreset laminae dip is probably less than that of all other types of sand dunes, with one well-developed maxima in the direction of the prevailing wind. Longitudinal Dunes
Seif or longitudinal dunes are elongate, continuous, serrated, straight sand ridges. Their long axes parallel the prevailing wind direction. Several seif dunes commonly occur as a series of long parallel ridges separated by broad interdune areas. Sand is deposited alternately on opposite sides of the sand dunes. Crossbedding dips are normal to the elongation of the sand ridge; therefore, the two maxima of highangle foresets are almost l80° apart. Locally, some low-angle bedding is present, especially in the lower part of a seif dune. It has been suggested that the most important factor in generating seif dunes is the existence of a strong wind with a uniform direction. The higher the wind velocity, the larger the seif dune and the greater the interdune spacing. All other conditions being equal, barchan dunes develop at lower wind velocities than seif dunes. Seif dunes may be modified to barchan dunes if wind velocities are not strong enough to maintain the seif dune form. The depositional pattern in seif dunes produces red and blue dip patterns with an azimuth normal to the prevailing wind. Occasionally an azimuth reversal occurs within the blue patterns. Whalebacks
Whalebacks are large-scale features associated with seif dunes. They are platforms of rather coarse-grained sediments left by the passage of a series of seif dunes along the same path. The platform and the sides are composed of horizontally bedded sediments with crossbedded seif dune sediments below. Wadis
Wadis are predominantly dry desert streambeds that are only active following sporadic, but often heavy, rains. Wadis are better developed near hills where the rainfall is slightly higher. Wadis are characterized by sporadic and abrupt fluvial activity and by a low water-to-sediment ratio. Deposition by flash floods is very rapid because of the sudden loss of velocity as the water i s absorbed underground. Most
129
wadis diverge downslope and deposit the bulk of their sediment in fan-shaped bodies at the downstream limit of the flow. Wadi channels are not permanent, and they may be filled by their own detritus or by wind-blown sediments. During subsequent seasons, a new channel system is likely to cut into the older sequences. Wadi channels produce dip patterns similar to those found in braided streams. Small ripples, megaripples, and plane beds are the bedforms developed in wadi channels by the variable flow conditions. Deposits within the wadi channels may be conglomeratic and fanglomeratic. During certain phases of flow, the sediment transported through the wadi may resemble mud flows. The nature of the sediment is strongly controlled by source rocks and the availability of various grain sizes. Wadi deposits may lack pebbles and may contain only well-sorted sand. The deposits produce ripple and horizontal beddings. Desert Basins
Desert basins represent areas of inland drainage with water flowing towards the center. Basins are often low depressions resulting from deflation of tectonic origin. Water accumulates in these low-lying areas, producing shallow, ephemeral lakes. The larger examples may be semipermanent desert lakes. Inland sabkhas are formed when sediments are subjected to wetting by inflowing wadis or ground seepage, subsequent drying, and deposition of damp, salt-encrusted sediments. In deflation hollows, where the water table is higher than the ground surface, a small lake may develop. Sand dunes may be drowned and preserved as a consequence of a rising water table caused by seepage and inflowing water. Abundant detrital sediment is brought to desert lakes and inland sabkhas during floods. As current velocity is almost nil, the deposition of silt and clay occurs from suspension, and individual thin beds may contain graded bedding. Gypsum, halite, and other evaporite minerals are commonly associated with these deposits. The uppermost clay layers may crack and curl during dry seasons, and these features may be preserved if covered by blown sand. Detrital sediment is rarely deposited in lakes resulting from groundwater seepage; instead, salt pans are built. Some windblown detrital sediment may be incorporated as thin layers or impurities within the chemical precipitates. Sediments of inland sabkhas are usually parallel-bedded with silty and clay-rich layers alternating with thin, sandy, gypsum or gypsiferous clay layers. These sediments are deposited as inflowing wadi sediments settle from suspension, or as wind-blown sediments are captured by adhesion ripples on the sabkha surface. Bedding is better developed in desert lake sediments than in inland sabkha sediments. Sabkha sediments sometimes generate only blank zones on the dipmeter plot.
130
Deltaic environments Introduction
Only a small percentage of modern coastlines are delta-dominated at any given ti me. Most coastlines are located in i nterdistributary environments where sediments deposited by older deltas are undergoing reworking and redeposition. Deltas are constructed where rivers enter the sea. Where long-shore currents are weak and abundant sediments are available, deltas prograde seaward, forming elongate or birdfoot deltas. The modern Mississippi Delta is a classic example of a birdfoot delta. Strong longshore currents prevent or retard seaward progradation, and the resulting deltas form cuspate-arcuate shapes. Deltas discharge seaward through active distributaries. Fan, crescent, or elongate sand bodies called distributary mouth bars or distributary front sands are deposited seaward of the mouth of each distributary. These and the following features are illustrated in Figure 1 .
Figure 1
131
During periods of high water, breaks occur in the natural levees formed along the distributary channel margins. Discharge through breaks or crevasses in the natural levees forms crevasse splays. Crevasse splays have the same shapes as distributary mouth bars. As the distributary channel progrades, bodies of water between distributary channels are constrained by sedimentary deposition into interdistributary embayments. Figure 2 is an example of dipmeter plots from a deltaic environment and a tide/wave-dominated environment.
Figure 2
Distributary front deposition at rates of tens of feet of sediment per year exist. The associated rapid burial and subsidence appear important in sediment preservation because they prevent reworking of the sediments by waves, tides, and currents. Identifying the Environment
132
It is possible to confuse a thin eolian sand section with a deltaic sand; therefore, recognizing a fossil delta depends in part on local l ocal knowledge that the sediments under investigation were deposited in a marine environment. If it is known that deltaic conditions existed during the deposition of a zone of interest, log character can be used to determine the probability of preserved deltaic sediments. A strong family of mostly blue dip di p patterns is a good indicator. The blue patterns would be intermixed with a few red patterns with azimuths 90° from the blue patterns. Funnel-shaped SP and gamma ray curves are indicative i ndicative of preserved deltaic sediments; however, a funnel shape alone does not identify a deltaic environment. Shale resistivities may provide clues on a strictly local basis to i ndicate that the zone of interest was deposited in i n a deltaic environment. Identifying Deltaic Features
Once it has been determined that a zone of interest was deposited in a deltaic deltai c environment, the data should be compared to a generalized deltaic model. If the entire deltaic system was preserved, which is unlikely, the system would consist of the following: · distributary channels · distributary mouth bars · crevasse splays · longshore current sand waves · marshes These features become the pieces of the jigsaw puzzle you wish to solve. In the worst-case scenario, scenario, the entire delta would have been reworked, and all of the pieces would be missing. Usually, however, several of the pieces are present. They may be from adjacent parts of the puzzle, or they may fit fi t randomly into the model with no adjacent pieces. There are several facts to help solve the puzzle. If the zone of interest was deposited during a deltaic period of deposition, strong dip patterns can be assigned a deltaic, rather than a reworked, origin. Also, the location of land during the time of deposition is known, at least approximately. Logs are responding to only fragments of each deltaic feature, not the entire system. Distributary Channels
If a complete depositional sequence were preserved, the dip patterns on the following illustrations would be seen. Figure 3 shows the expected dip patterns within a distributary channel.
133
Figure 3
When the channel is penetrated at or near its axis, axi s, only blue dip patterns, indicating flow down the channel, are recorded. The south-dipping blue pattern on this figure indicates flow down a north-south n orth-south striking channel. When the channel is penetrated near the edge, only red dip patterns, dipping toward the channel axis, are found. Current velocities are lower near the channel edge; therefore, only laminar deposition occurs. Between the two zones, a red-blue dip pattern combination is usually found. The basal layer of fill mimics the dip of the surface it is deposited on; therefore, the drape over the sloping surface of the channel cut creates a red patte rn dipping toward the channel axis. This red pattern (or patterns) is overlain by blue patterns with a dip azimuth 90° to the underlying red pattern. These blue patterns result from flow down the channel. Foreset beds deposited by sediments transported down the channel are formed after the basal portion of the channel is fil led and leveled. Distributary Mouth Bars
Within the distributary mouth bar seaward of the distributary channel mouth, only blue dip patterns would be recorded ( Figure 4 ). These patterns indicate the direction of sediment transport.
134
Figure 4
The dip magnitude spread of the family of blue patterns is an indicator of the type of depositional environment and the probable sand geometry. If the magnitude spread of the family of dips is greater than 10°, the sand was probably deposited in i n an inertia-dominated environment, and the shape of the distributary mouth bar is probably elongate. If the family magnitude spread is 10° or less, the environment was friction-dominated, and the shape of the distributary mouth bar is probably fanlike or crescent. The subsurface deltaic sediments usually consist of a stack of fossil delta remnants rather than sediments deposited by a single active delta. Dips belonging to patterns measured in the subsurface tend to be steeper than their original depositional angles. This steepening is probably the result of compaction. Discharge Direction
The discharge directions of a delta are not always directly seaward. Some active distributaries of the modern Mississippi Delta discharge to the north-northeast-not to the south-southeast, which is the main direction of progradation.
135
Figure 5 is an example of a preserved distributary mouth bar from the East Cameron Block 270 field.
Figure 5
The distributary prograded to the northeast, a direction similar to the Main Pass distributary of the modern Mississippi Delta. Deltas may prograde almost across the continental shelf, as has the Mississippi Delta. Distributary Channels and Distributary Mouth Bars
Dip meters run in wells that penetrate both distributary channels and distributary mouth bars create the patterns shown in Figure 6 .
136
Figure 6
Location A: A red pattern resulting from drape over an east-west striking channel overlies a blue pattern generated by distributary mouth bar sands transported from west to east. Location B: The deepest blue pattern indicates distributary mouth bar sands. The overlying red pattern indicates drape over the base of the di stributary channel. The shallowest blue indicates flow down the channel. Location C: There is no red pattern indicative of drape at this location. Relative to the information from the other wells, it is possible to identify this as the channel axis. The underlying blue pattern indicates a distributary mouth bar sand. The overlying blue pattern indicates flow down the channel.
This sequence is repeated on the opposite side of the channel, with red patterns dipping to the north (Locations D and E). Cuspate-Arcuate Deltas
137
Rivers create cuspate-arcuate deltas by discharging their fresh water and sediments seaward, but the strong longshore currents transport the marine sediments in the direction of current flow, subparallel to the fossil coast-line. Figure 7 is an example of a cuspate-arcuate delta from the Bekapai field, Mahakam Delta, Kalimantan.
Figure 7
Strong longshore currents transported sediments that were carried to the sea by the ancestral Mahakam to the southwest. The dominant dip is southwesterly dipping blue patterns. Creation and Destruction of a Delta
The sediments deposited at a river mouth create increasing resistance to flow. Eventually, the river follows the path of least resistance and changes course. When the sediment supply to the delta is eliminated, deposition ceases, and destruction begins. Deltaic sediments exposed on land and the seafloor are attacked by rains, waves, currents, and tides. These destructive forces remove, re-sort, retransport, and redeposit the previously deposited sands and clays in new forms. The amount of a fossil delta that is preserved depends on many variables: the depth of subsidence, the period of deltaic deposition, the thickness of the deltaic column,
138
and the amount of protection from the open sea. One estimate by a knowledgeable geologist, Dr. John Kraft, estimates a worst-case preservation rate of less than one percent. The remaining 99% may be transported by waves, tides, and currents to be redeposited in an interdeltaic environment. Figure 8 illustrates the dipmeter response in zones where some of the original bedding planes were destroyed by reworking.
Figure 8
Interdeltaic environment Introduction
Many sediments deposited in an interdistributary environment by waves, tides, and currents were originally deposited within deltaic environments. Later reworking provided the raw materials for the interdeltaic deposition. Dipmeter logs that were run through sediments deposited in i nterdeltaic environments tend to look rather sparse. They contain blank zones resulting from bioturbation and rooting, and open tadpoles from low-quality correlations.
139
As is true in other marine environments, the direction to land during the time of deposition is a key direction ( Figure 1 ). This information allows tentative identification of landwarddipping foresets deposited within tidal flood deltas, washover fans, and slipface deposits.
Figure 1
Other transport directions indicated by blue patterns are seaward-dipping ebb delta sands and sand waves deposited by longshore currents, paralleling the coast. Beach sands dip seaward on their front portions and landward on their slip -face portions. Tidal flat sediments exhibit blue patterns dipping i n opposing directions as a result of landward- and seaward-dipping foreset beds. Tidal channels in microtidal and mesotidal ranges generate red patterns dipping toward their axes. In a microtidal range (less than 2 m), any ebb delta present would be small ( Figure 2 ).
140
Figure 2
In microtidal environments, tidal inlets with pronounced flood deltas on their landward side exist. In a mesotidal range (between 2 and 4 m), a prominent ebb delta would be formed. In a macrotidal range (more than 4 m), a tidal estuary would be formed. Macrotidal estuaries contain sand bodies elongate in the directions of tidal flow. Ebb Delta
In the ebb delta shown in Figure 3 , it is assumed that land is to the west and the coastline strike is north-south.
141
Figure 3
A dipmeter log run at location A, in the southern portion of the ebb delta, would contain southeasterly dip, not directly seaward in an easterly direction. A dipmeter log run from a well drilled at location B would exhibit seaward or east dips. A well drilled at location C would penetrate both the marginal flood channel and the underlying ebb-deposited sediments. Foresets dipping back into the tidal channel were deposited on the flood tide; as a result, they dip to the southwest. Beds deposited during the ebb dip to the northeast. Tidal Channel
Tidal channel dip patterns resemble patterns from other channel types. The basal layer of channel fill mimics the dip of the surface on which it is deposited. The channel base is a sloping surface except at the axis; therefore, red dip patterns are created with azimuths toward the channel axis and normal to the channel axis strike. After the fill smoothes and levels the channel base, foreset beds with dip along the channel are deposited. The type of deposition preserved-flood or ebb-depends on the location within the channel.
142
Sand waves formed within tidal channels may contain a large amount of shell hash. If preserved and buried, these waves would generate seaward-dipping blue dip patterns. In Figure 4 , the well location B is at or near the channel axis.
Figure 4
Only sand-wave foresets would be deposited, because of the relatively flat underlying surface. Flood Deltas
Deposition within flood deltas occurs as landward-dipping foresets. Similar landwarddipping foresets are found within washover fans and slipface sands. Dipmeter logs run at locations A, B, and C on Figure 5 contain blue dip patterns dipping to the northwest, to the west, and to the southwest, respectively.
143
Figure 5
Particular caution must be exercise in the interpretation of dipmeter logs run through sediments deposited in a tidal environment. The most significant dips are de rived from sediments deposited within the flood delta, the ebb delta, and the tidal inlet. Swash Bar and Recurved Spit
Two groups of dips that may produce conflicting interpretations are swash bars and recurved spits. Swash bar dips create land-ward-dipping blue patterns, which can be mistaken for flood delta dips. This can lead to an offset seaward of the terminal lobe. The best approach to identifying swash bar deposits is to expect them t o be preserved near the tops of tidal sands or carbonates; therefore, beware of blue patterns existing only in the top of a tidal sequence. Landward-dipping blue patterns from flood deltas should extend throughout most of the sand or carbonate under study. Recurved spit dips are the other set of problem blue dip patterns. They tend to dip away from the inlet and can contribute to offsets in the wrong direction. These dips can be recognized by their dip in the direction of coastline strike. Longshore Current Sand Waves
144
Longshore current sand waves are composed of fore set beds that generate blue dip patterns paralleling the fossil coastline. Deep water contains longshore currents strong enough to redistribute sands previously deposited by turbidity flows. To identify sand waves, one must (1) know that the sequence being interpreted is from an interdeltaic depositional environment and (2) determine the direction to land during the time of deposition. Beach Sands Shoreface Sands
Shoreface sands were deposited between the beach and a water depth of 20 m, the fair-weather wavebase. These sands were deposited in a high-energy environment, and few, if any, of the bedding planes were deposited flat. After deposition, bioturbation occurred, destroying or distorting the original bedding. Dipmeter logs run through shore face sands record a few widely scattered dips and blank zones because of bioturbation. Bioturbation decreases in the shallowest portion of the shore face zone; therefore, more dips are recorded as the mean l ow water line is approached. Beginning at depths of about 5 m, some low-angle, seaward-dipping crossbeds were deposited and preserved. These beds initially dip seaward 1° or 2°. Flaser bedding is also present in the lower shoreface zone. Beachface Sands
Beachface sands were deposited as parallel crossbeds dipping seaward plus or minus 5°. Runnel
Deposition within a runnel may appear as megaripples dipping parallel to the beach, small ripples, or laminations. Preserved megaripples generate small blue dip patterns best identified by CSB computation. Small ripples usually create blank zones or false correlations; however, they can be identified on the multisensor dipmeter output of the 8-curve tool. Berm Crest
Deposition on the berm crest is essentially horizontal. If preserved, the beds would indicate structural dip. Dunes, which also form on the berm crest, contain festoon cross-bedding, which generates a wide dip scatter. Back Beach
If the back beach escaped bioturbation by fiddler crabs or their ancestors, it would contain landwarddipping foresets that generate blue dip patterns. These landwarddipping patterns are the best indicators for determining the strike of a fossil beach.
145
Washover Fans
Washover fans generate landward-dipping blue patterns similar to slip face foresets and flood delta foresets. The character of other log responses provides clues for distinguishing these features. Washover fans were deposited over marsh deposits by a catastrophic event. This process did not allow for appreciable sorting. Flood deltas were deposited in a subaqueous environment with winnowing before final deposition. Slipface sands were deposited on a land surface containing some plant material; this, in turn, created a rooted layer. The rooted layer generates blank dip zones and is electrically more homogeneous than undisturbed bedding. Barlike or Convex-Upward Sands
Barlike or convex-upward sands may be formed at the wave break point or as beach ridges. There is one distinct difference between these two types of sand Break pointbar sands are winnowed until there is little internal electrical contrast; therefore, dipmeter logs exhibit mostly blank zones. In contrast, beach ridges exhibit many internal dips ( Figure 6 ).
Figure 6
146
When either type of sand is penetrated on the flanks, the drape the overlying beds creates a red dip pattern just above the sand A fault can create the same dip pattern; therefore, faulting must be ruled out before any stratigraphic interpretation is attempted The direction of the red pattern is toward the shaleout and normal to the strike of the bar or beach ridge. If the bar or ridge is penetrated at or near the crest, no drape would be present, and the sand would appear blanket-like. These same guidelines can be applied to oolitic bars. The drape extends beyond the limits of barlike sands. A red pattern in the silty zone, where a bar should have been located, dips away from the bar, and it can be used to determine the direction of sidetrack. In some cases only blue patterns dipping toward the shaleout are found above a bar ( Figure 7 ). These patterns tend to be components of a very subtle red dip pattern, and may be partially related to slump of the clays deposited above the bar.
Figure 7
147
Deepwater environment Introduction This chapter addresses the processes of deposition and the resulting dip patterns encountered in deepwater environments. The processes of mass transportation are able to move, transport, and lay down sediments between their zone of origin and a topographically lower zone under the influence of gravity. Generally, these mechanisms provide intermittent and catastrophic transfers of large amounts of sediments, which are deposited at or near the base of a slope. Mass transport consists of rockfalls, slides and slu mps, and gravity flows ( Figure 1 ).
Figure 1
Rockfalls Rockfalls are formed by free-falling bodies of sediments accumulating at the bases of fault scarps, canyon floors, and other steep slopes. The deposited sediments generally exhibit distinct limits, but no bedding. The dimensions of clasts that form rock falls vary from sand-size to blocks measuring several tens of meters. The clasts are in contact and generally contain intergranular porosity.
148
The sequences resulting from submarine rock falls are often related to forereef escarpments or platform edges. On slopes in deep-sea environments, rockfalls may contain abyssal sediments. The accumulation of sediment blocks caused by rock slides can only occur at the foot of strongly inclined slopes, which are often characteristic of carbonate margins.
Slides and Slumps Subaqueous slides occur when a mass of semiconsolidated sediments moves along a basal shear surface. These slides are able to transfer considerable masses (up to tens of cubic kilometers) of sedimentary materials from the inner or outer continental platform to the abyssal plain. Any internal bedding characteristics of the mass are preserved during movement. Slides can be divided i nto translational or glide and rotational or slump types. The basal shear surface of a glide i s a plane of slightly undulating surface paralleling the stratification. In a slump, the concave shear surface permits rotation of the slump block. As a slump block moves down a slope, compression occurs at the foot of the block, and tension occurs at its rear. Compression produces thrusts and folds, and tension produces normal faults and open cracks. The central part of the block is generally not deformed. Slump blocks penetrated in the su bsurface are not always easy to identify. They may appear on the dipmeter as an isolated trend. When this occurs, the most probable explanation is either a tilted fault block (most commonly found between two nearby faults) or a slump block. If faulting can be ruled out, then the slump block becomes the most probable explanation.
Gravity Flow Sediment gravity flow is a general term for flows of mixed sediments and fluids in which the bedding coherence is destroyed and the individual grains move in a fluid medium. This includes mud flows or debris flows, grain flows, liquefied flows, and turbidity flows. Mud flows exhibit essentially plastic behavior with the muddy carrier phase creating sediment coherence. The matrix containing the clasts is the main driving and lubricating force behind the flow. The dynamics of grain flows are governed by the reciprocal interaction of clasts. This granular interaction causes sandy flows to exhibit plastic mechanical behavior rather than fluid behavior. In contrast, liquefied flows, fluidized flows, and turbidity currents exhibit a fluid behavior. Grain flows consist of cohesion-less sediment supported by dispersive pressure. This process requires steep slopes for initiation and sustained downs lope movement. Liquefied flows consist of cohesionless sediment supported by upward displacement of fluid as loosely packed structures collapse. The sediments settle into tightly packed textures. Liquefied flows require slopes of greater than 3°. Fluidized flows consist of cohesionless sediment supported by upward motion of escaping pore fluid. These flows are thin and short-lived. Turbidity current flows contain clasts supported by fluid turbulence. These flows can move long distances on lowangle slopes.
Submarine Channel-Fan Complex Figure 2 illustrates the features found during the growth of submarine fans.
149
Figure 2
Of these features, the obvious deepwater features interpretable by dipmeter logs are debris flows, which result in blank zones; feeder channels, which produce typical red dip patterns at the base and blue patterns with a 90° azimuth difference above; and midfans, which generate blue dip patterns. Outer fan sediments generate structural dips. A submarine channel-fan complex can exhibit the same features as a delta complex, including natural levees. Submarine feeder channels are cut by downs lope sediment flows and later filled ( Figure 3 ).
150
Figure 3
As with other types of channels, the basal layers of fill mimic the dip of the underlying surface. Deposition on a sloping surface produces a red pattern dipping toward the channel axis and normal with the channel strike. After the bottom was filled and leveled, foresets dipping down the channel were deposited; these, in turn, generated blue patterns dipping 90° from the u nderlying red patterns. In the midfan portion of the system, only blue patterns dipping in the direction of sediment transport are detected. Few obvious foreset beds are found within midfan outcrops, and this raises the question of what the dipmeter tools are measuring. It is possible the dipmeter sensor is detecting some type of permeability change associated with timelines or climbing ripples. Permeability changes do not always have a visual representation and may appear only on X-ray photographs. In the outer fan portion of the system, only structural dips are detected because deposition was essentially horizontal. This is an environment in which the deposition of alternating laminations of sand and shale may become low-resistivity pay zones.
Transport Directions A common feature of deepwater sands is that transport directions are not directly offshore. Some sediments were transported parallel to the continental shelf while others were transported back into land ( Figure 4 ).
151
Figure 4
Landward transport can be a function of seafloor topography or it can be initiated by the presence of a down-tothe-basin growth fault. Whatever the cause, inshore transport in deepwater depositional environments does occur.
Deepwater Longshore Currents In some areas, considerable numbers of blue dip patterns indicate sediment transport parallel to the slope. This is a result of the reworking of previously deposited sediments by deepwater longshore currents. This is another environment conducive to the deposition of alternating sand-shale laminations.
Submarine Canyons Submarine canyons exhibit alternating up and down canyon sets of blue dip patterns generated by deepwater tidal action ( Figure 5 ).
152
Figure 5
Submarine canyon fill sands closely resemble tidal sands on the dipmeter plot. The fill sands contain many blue patterns dipping both up and down the canyon. These patterns were probably generated by deepwater tidal action within the canyon. Canyon fill sands may be up to a thousand feet or more in thickness. They may also exhibit indicators of compaction underneath-e.g., downward-decreasing resistivity or increasing interval transit time gradients.
Turbidity Flows Turbidity flows produce complex sand packages containing multiple depositional units. Separate reservoirs may be present, though sand-to-sand contact seems probable. Figure 6 illustrates dip patterns encountered in sand packages produced by turbidity flows.
153
Figure 6
This package is made up of at least six submarine fans, two scour channels, and a sediment layer deposited in the upper portion of a submarine feeder channel. In other areas, this portion of the channel is filled with conglomerate. Shale layers are not required to separate one reservoir from another; an inch or so of silt suffices.
Debris Flows Debris flows are best recognized by the dual-dip curves themselves, since few (if any) meaningful dips are produced. Some correlations not extending around the four pads are seen on the presentation, but no tadpoles are produced. Conglomerates can produce these features.
Feeder Channels The depositional environment of the sand at 6400 ft in High Island Block 560-561 is a continental slope environment; therefore, feeder channels would be the most probable feature ( Figure 7 ).
154
Figure 7
The expected dip model would be a red dip pattern at the base of the sand section, with blue patterns above. The red pattern azimuth is toward the channel axis and normal to its strike. The blue dip patterns indicate flow down the channel. The azimuth of the blue patterns is approximately 90° from the azimuth of the red patterns. The dipmeter log on Well 4 of High Island Block 561 exhibits the expected dip patterns for a filled feeder channel. The basal red pattern dips to the northwest, which is the direction of the channel axis. The overlying blue patterns dip to the southwest, which indicates flow down the channel from northeast to southwest. The lower portion of the example shows the same dip pattern combination from the sand at 8900 ft. In this example the red pattern dips to the north; therefore, the channel axis lies north of the well, and the channel strike is west to east. The east-dipping blue pattern indicates sediment transport down the channel from west to east. The relative magnitudes of the dip patterns in these examples indicate their approximate positions within their respective channels. The sand example at 6400 ft contains several blue patterns, but only one red pattern; this indicates a position near the channel axis, where the blue patterns dominate. Had the location been nearer the channel axis, only blue dip patterns would have been present. The thin 8900-ft sand contains a strong red pattern and a weak blue pattern. This indicates a position near the edge of the channel, where current velocities were lower and drape over the underlying surface was the dominant type of deposition. Transport in this channel was from west to east paralleling the fossil coastline. This orientation may result from flow parallel to a down-to-the-south growth fault system in the area.
Dip Scatter as an Environment Indicator
155
Dip scatter results from both the original attitude of deposition and postdepositional deformation. The products of both processes are diagnostic depositional indicators. The following comments on scatter are conf ined to a marine environment. The marine ecological zones, their defining depths, and their location on the continental shelf and slope are illustrated in Figure 8 .
Figure 8
The original concept of less scatter on the lower continental slope was due to a lack of paleo-calibrated dipmeter logs run through lower sl ope sediments. Later observations of dipmeter logs run in paleo-identified lower slope sediments confirmed that some of the greatest sediment jumbles exist at the base of the continental slope. Dip scatter is best used with shale resistivities, density-neutron responses, and other indicators. When deltaic deposition is preserved in its original form, it can mask effects of the surrounding depositional environment. Indicators are more obvious in a tide/wave-dominated environment than in a delta-dominated environment ( Figure 9 ,
156
Figure 9
Figure 10 , Figure 11 ).
157
Figure 11
158
Figure 10
Inner neritic deposition in a tide/wave-dominated environment generates a 40° dip s catter and blank zones. Some scatter results from a high initial angle of deposition, but much of it is the result of bioturbation. Bioturbation produces zones of no correlation or z ones where miscorrelations are probable. Near the 20-m boundary between the inner and middle neritic zones, the amount of bioturbation and the corresponding dip scatter decrease. The scatter across the middle neritic zone ranges from 20° on the shoreward side to 3° on the seaward side. Local experience allows additional subdivision of the middle neritic zone into 50- to 100-ft, 100- to 200-ft, and 200- to 300-ft ranges. Dip scatter in outer neritic sediments ranges from none, where parallel laminations exist, to 2°. Sediment spreading by long-shore currents in this zone can produce laminated, low-resistivity pay zones.
Continental Slope Sediments Dipmeter logs run through continental slope sediments tend to be difficult to interpret because of postdepositional deformation by downslope creep, slump, and fracturing. Shales, in particular, may be so severely deformed that few meaningful dips can be computed. Sometimes the only intact bedding planes are found within sands; when this occurs, the sand dips are used for determining structure. Shale resistivities can provide clues to the proximity of sand bodies associated with shorelines or deltaic depositions. Shale resistivities are partially a function of grain size; therefore, the presence of silt-sized particles increases the resistivity values. Assuming a model progressing from sand to silt to clay, the presence of i ncreased silt creates higher shale resistivity values.
159
In the northern Gulf of Mexico, shale resistivities less than 0.8 ohm-m usually indicate deposition in a slope or abyssal depth range. There are, however, exceptions to this general rule. Dip scatter of 60° on the continental slope results from postdepositional deformation. The scattered dips result from deformation; they are not related to structural or stratigraphic dips. The dip scatter again decreases to a maximum of 2° in the abyssal range. Some sediment transport by deepwater longshore currents also occurs in this environment.
Sea Level Fluctuations A Pleistocene example illustrates changes in scatter resulting from sea level fluctuations during glacial and interglacial periods. During the interglacial periods, sea level is high and deposition occurs in low-energy environments-probably the outer shelf. This permits layer-cake deposition with dip variations less than 3°. During glacial periods, the sea level drops, and deposition occurs in inner and midshelf environments. The environmental changes increase dip scatter considerably.
Compaction Features Many thick, channel-like sands were formed by compaction, not by the cut-and-fill process. Sands deposited on a mud bottom gradually sank downward, compressing and dewatering the underlying muds. Shales formed from compressed muds exhibit downward-decreasing resistivity gradients and downwardincreasing interval transit time gradients. Density-neutron log response gradients are also present( Figure 12 ).
Figure 12
160
The dip pattern resulting from compaction is a mega-red pattern with interspersed blue groups dipping in the same direction. No right-angle relationship exists between the azimuth of the red and blue dip groups, as it does in features resulting from the cut-and-fill process.
Deepwater Chalks Localized dipmeter interpretation rules are occasionally convenient. The following set of rules was developed for the deepwater chalks of the Norwegian Central Graben. Figure 13 illustrates an Ekofisk chalk example.
Figure 13
In developing these rules, i t was noted that chalk wells whose dipmeter logs exhibited many blank or scattered dip zones and dip patterns were better producers than wells containing zones exhibiting mainly structural dips. To quantify the interpretation process, multipliers or weights of 4, 2, and 1 were assigned respectively to blank or scatter zones, red patterns, and blue patterns. These arbitrary weights are based on the permeability of each type of zone. Blank or scatter zones result primarily from chalk debris flows or conglomerates that contain the highest permeability; therefore, they were assigned the highest weight.
161
Red dip patterns represent beds draped over a sloping surface. These draped layers permit laminar flow, which has a lower permeability; therefore, red patterns were assigned a weight of two. Blue dip patterns indicate foreset-generated crossbeds cutting across the reservoir at some angle that interfered with flow into the well. Blue dip patterns have the lowest permeability and the lowest weight factor.
Reservoir Quality Factor The following equation was developed to determine whether a chalk interval is capable of commercial production. The potential for commercial production was called the quality factor. The quality factor is the proportion of the total footage under study contributed by each type of zone multiplied by the weight factor for the zone. The equation is given below:
where: FD = the total footage of dipmeter blank or scattered dip zones Fr = the total footage of dipmeter red patterns Fb = the total footage of dipmeter blue patterns. Using this approach on operator data, it was discovered that chalk intervals such as the Ekofisk, Tod, or Hor with quality factors of 2.6 or greater contained intervals capable of commercial production. For the commercial threshold of 2.6 to be a reliable indicator, the interval must be sufficiently thick. Quality factors can be contoured on both regional and fieldwide bases for the Central Graben area. Other weight factors could have been chosen that would have worked as well. The value of the commercial threshold would have changed. Similar techniques may have applications in areas where sandstones have undergone some downs lope creep and slump or shallow-water working.
Fluvial channel deposits Interpretation of Fluvial channels
The following are the five basic steps in interpreting a dipmeter log for a fluvial channel: · Determine the structural dip (and delete it if necessary). · Determine the stratigraphic encasement. · Define the depositional environment. · Orient the sand trend. · Locate the offset. Structural Dip
Structural dip is determined from the shales above and below the channel. These dips may be different, since channels frequently occur as unconformities. The shale
162
above the channel usually reflects the structural dip required for interpretation. In very low-angle situations, the stratigraphic gain may be more important than structural dip. The influence of coalescing channels repeatedly affects the structural position of fluvial channels. Generally, if structural dip is greater than 4°, then the structural dip should be deleted. Sometimes, if the dip magnitudes are low (less than 10°), then even 2° di p should be removed. Stratigraphic Encasement
The stratigraphic encasement is the interval in which the channel facies occur, including both the sand and shale components of a channel. Detailed correlations with offset logs are used for defining the stratigraphic encasement. Under special circumstances, the channel abandonment facies, or clay plug, can be identi fied from higher gamma ray or more resistive shale log responses. Red dip patterns, reflecting compaction features, may also be used to define channel facies. In all cases, identifying the interval is critical to the interpretation. Depositional Environment
Success in defining the depositional environment depends on the geologist’s input, core and sample data, log responses, formation images, and dipmeter arrow plots. Local knowledge of the geology is very important in identifying the environment. Cores and samples are an integral component in new areas and are always useful in any area. Electrical formation images are a valuable aid to the interpretation of the thi n, fluvial sand zones. Log responses are generally used to identify a fining -upward sequence, which infers a channel system. Distinguishing a braided stream from a meandering stream is only possible when very simple depositional sequences are penetrated. The braided stream contains several fining-upward sequences within the sand. A meandering stream contains one overall fining-upward sequence. This becomes very complex when the borehole penetrates several coalescing units. Coalesced point bars occurring in meandering streams may be interpreted as braided streams. Dipmeter patterns are very similar in braided and meandering stream environments. Families of red and blue dip patterns with 90° azimuth differences typically occur in both environments ( Figure 1 ).
163
Figure 1
Braided streams can sometimes be recognized by the identification of transverse bars within the stratigraphic encasement. Channel Orientation
Channel orientation from dipmeter patterns is usually determined by the following priorities: 1. strong blue 2. strong red 3. weak blue 4. weak red 5. erosional or drape Channels are generally elongated parallel to the blue patterns and perpendicular to the red patterns.
164
Locating the Offset
Considerations for locating an offset include reservoir geometry, reservoir quality, structural position, surface restrictions, and secondary-recovery prospects. Reservoir quality is determined primarily by permeability, bed thickness, and porosity. In a point bar, for example, the coarser, better-developed sand i s generally near the thalweg, and fine-grained, poor-permeability sand is generally near the inside bank. Also, the leading edge of a point bar usually has better sand quality than the trailing edge. Structural position is critical when a water or gas contact has been penetrated. The structural position often depends on compaction over coalescing channels and stratigraphic gains within the channel system. Optimum location of an injection or a production well in a secondary-recovery project is dependent on the relative position of the well in the reservoir. A well near the leading edge of a point bar, for example, usually depletes quickly on primary production. The well can only produce from one direction, toward the mi ddle of the point bar. This well can be used quite effectively as an injection well. Required Accuracy Fluvial channels are usually narrow in width. This requires high accuracy in the measurement of the channel orientation. Typically, fluvial channels have a productive width of approximately 40 times their productive thickness. A 10 ft thick channel sand has an estimated productive width of 400 ft.
An azimuth diagram, as shown in Figure 2 , can help in defining the channel orientation.
165
Figure 2
There are several segments to the diagram. 1. Depths are determined from the logs and the dipmeter plot. 2. The geologist and the dipmeter interpreter agree up on the type of deposition. 3. Structural dip is determined from the surrounding shale sections. If the dip above and below the sand is different, the sand is assumed to be deposited at an unconformity and the structural dip above the sand is recorded. 4. Confidence rating is a means to rank the quality of the interpretation. The rating is from A (highest) to D (l owest): A = strong blue and red, B = strong blue or red, C = weak blue or red, and D = erosion, drape, or intuition. 5. Orientation is shown as a line along the sand trend. 6. Current flow is the arrow on the end of the orientation line.
166
7. Channel thalweg is the arrow in the center of the azimuth diagram. Determination of Well Position in a Point Bar
A relation between the blue and red dip patterns allows the determination of relative position in a point bar ( Figure 3 ). Blue pattern azimuths are usually parallel to the sand axis, since current flow is across the point bar. Red pattern azimuths are generally perpendicular to the sand axis and point toward the channel thalweg.
Figure 3
For wells located on the leading edge of a point bar, blue and red pattern azimuths are normally greater than 90° in angle difference. When a well is located midpoint, the blue and red pattern azimuths are approximately 90° different (perpendicular) to each other. For wells positioned on the trailing edge of a point bar, the blue and red pattern azimuths are usually less than 90° in angle difference. Figure 4 shows a dipmeter plot through a Cretaceous sand interval in a fluvial meander channel.
167
Figure 4
The strong NE red and SE blue dip patterns show the channel thalweg to be N67E, a current direction of S47E, and an orientation of N47W-S47E. The angle di fference between the red and blue dip pattern azimuths is slightly less than 90°. This indicates the well position to be on the trailing edge of a point bar.
Exercise No. 1 This exercise uses a classic deltaic example. The sand, shown in Figure 1 , from 6744 ft to 6900 ft was deposited in a deltaic environment.
168
Figure 1
In which part of the delta complex was this sand deposited? What is the strike of the sand? In what direction is the thickest part of the sand body? What was the direction of current flow? Was the entire sand deposited as one feature, or was there more than one feature deposited? This sand was deposited as fill within a distributary channel. The strike of the channel is NE-SW. The axis lies to the NW of the well. Current flow was down the channel from NE to SW. There is more than one channel present.
169
The main channel is below 6784 ft. This is the feature to consider when offsetting the well. Above 6784 ft, the current flow diminished as the channel began to fill with sand, and channel switching occurred. There is another minor channel between 6784 and 6761 ft. Its strike is also NE-SW, and its axis lies to the NW. Flow was from the NE to SW. The few scattered dips within this interval indicate some reworking.
Solution This sand was deposited as fill within a distributary channel. The strike of the channel is NE-SW. The axis lies to the NW of the well. Current flow was down the channel from NE to SW. There is more than one channel present. The main channel is below 6784 ft. This is the feature to consider when offsetting the well. Above 6784 ft, the current flow diminished as the channel began to fill with sand, and channel switching occurred. There is another minor channel between 6784 and 6761 ft. Its strike is also NE-SW, and its axis lies to the NW. Flow was from the NE to SW. The few scattered dips within this interval indicate some reworking.
170
Exercise No. 2
Figure 1
In Figure 1 , the sand between 3810 and 4060 ft was deposited in an interdeltaic environment. What type of sand is it? What are its attributes?
Solution This sand is the product of previously deposited deltaic sediments reworked by waves, tides, and currents. The top of the sand is now barlike, and it shales out to the NE. The strike of the sand is NW-SE. The blank zone near the top results from shallow-water reworking and bioturbation.
171
Exercise No. 3
Figure 1
Figure 1 represents a Pennsylvanian sand deposited in a fluvial meander channel. What is the current flow direction? What is the thalweg direction? What is the channel orientation? Construct the azimuth diagram. What is the position of this well on the point bar?
Solution The current direction is south.
172
Channel thalweg is N41W. The orientation is along the blue pattern azimuth (NS) The angle difference between the red and blue pattern azimuths is greater than 90°. This indicates that this well is on the leading edge of the point bar.
ADDITIONAL READING Bigellow, E. L. (1982). “Application of Dip-Related Measurements to a Complex Carbonate-Clastic Depositional Environment”. SPWLA, The Log Analyst , v.23, no. 2, p. 9-30 Bigellow, E. L. (1985 ). “Making More Intelligent Use of Log Derived Dip Information”. Part IV. SPWLA, The Log Analyst , v.26, no. 4, p. 21-43 Bigellow, E. L. (1991). “Log-Derived Dip Data Successfully Delineates East Texas Paluxy Reservoirs”. Society of Professional Well Log Analysts Logging Symposium, 32nd , Midland , Tex. , 1991, Transactions, p. V1-V16. Bricaud, J.M., and Poupon, A., (1959) . “Continuous Dipmeter Survey, the Poteclinometer and the Micro- Focused Devices”. World Petroleum Congress Meeting, 5th , New York , 1959, Proceedings, p. 1-9. Chauvel Y., Seeburger D.A., and Orjuela A.C. (1984) . “Applications of SHDT* Stratigraphic High Resolution Dipmeter to the Study of Depositional Environments”: Society of Professional Well Log Analysts Annual Logging .Symposium, 25th , New Orleans, La., 1984, Transactions, p. G1-G23. Chemali R., Su S.M., Goetz J.F. (1989). “Measuring Rxo and Dip in Oil Based Mud with the Six Arm Dipmeter”. Society of Professional Well Log Analysts Logging Symposium, 30th , Denver , Colo. , 1989, Transactions, p. O1-O25. Chemali R., Su S.M., Goetz J.F., Maute R.E., Osborn F.F. (1990). “Methods for Improved Dip Determination in Water Based Mud with the Six Arm Dipmeter”. Society of Professional Well Log Analysts Logging Symposium, 31st, Lafayette , La. , 1990, Transactions, p. O1-O25. Cox, J.W. (1970). “The High Resolution Dipmeter Reveals Dip-Related Borehole and Formation Characteristics”. Society of Professional Well Log Analysts Annual Logging Symposium, 11th , Los Angeles , Calif. , 1970, Transactions, p. D1-D25. Davis R.J., Johnson C.A. , Gilreath J.A. (1990). “Interpretation of Depositional Systems in the Lower Silurian Medina Group of Western New York ” . Society of Professional Well Log Analysts Logging Symposium, 31st , Lafayette , La. , 1990, Transactions, p. N1-N25. Dumont A., Kubacsi M., Chardac J.L. (1987) “The Oil-Based Mud Dipmeter Tool” . Society of Professional Well Log Analysts Annual Logging Symposium, 28th , London , England , 1987, Transactions, p. LL1-LL15. Enderlin M.B., Kratochvil T. (1998). “Get The "Rhythm of the Rocks" and Find the Dip”. In Transactions of the SPWLA (Society of Professional Well Log Analysts) Thirty-Ninth Annual Logging Symposium , May 26-29, 1998: Houston , Tex. , Society of Professional Well Log Analysts, Paper HHH, 11 p. Etchecopar A., Dubas M.O. (1992). “Methods for Geological Interpretation of Dips”. Society of Professional Well Log Analysts Logging Symposium, 33rd , Oklahoma City , Okla. , 1992, Transactions, p. J1-J21. Fitzgerald D.D., Theriod J.C., York P.L. (1980). “Dipmeter Validity in Deviated Boreholes”. SPWLA Log Analyst, v. 21, no. 3, p. 8-18.
173
Georgi D.T. (1986). “Guides For The Interpretation of Dipmeter Fracture Logs”. Society of Professional Well Log Analysts Annual Logging Symposium, 27th , Dallas , Tex. , 1986, Transactions, p. TT1-TT17. Gilreath J.A., Maricelli J.J. (1971). “Detailed Stratigraphic Control Through Dip Computations”. Bulletin of the American Association of Petroleum Geologists , v. 48, no. 12, p. 1902-1910. Goetz J. F., Prins W. J., Logar J. F. (1977) . “Reservoir Delination by Wireline Techniques”. SPWLA, The Log Analyst , v.18, no. 5, p. 12-40 Goetz J.F. (1987). “New Angles and Dimensions in Dipmeter Interpretation Opened by the Six Arm Dipmeter”. Society of Professional Well Log Analysts Annual Logging Symposium, 28th , London , England , 1987, Transactions, p. JJ1-JJ25. Hepp V., Dumestre A.C. (1975). “Cluster--A Method for Selecting the Most Probable Dip Results from Dipmeter Surveys”. Society of Petroleum Engineers of the American Institute of Mining , Metallurgical, and Petroleum Engineers Paper SPE 5543, 17 p. Holt O.R. (1972). “Structural Geologic Considerations in Diplog Interpretation”. Society of Professional Well Log Analysts Annual Logging Symposium, 13th, Tulsa , Okla. , 1972, Transactions, p. Q1-Q30. Holt O.R., Schoonover L.G., Wichmann P.A. (1977). “True Vertical Depth, True Vertical Thickness and True Stratigraphical Thickness Logs”. Society of Professional Well Log Analysts Annual Logging Symposium, 18th , Houston , Tex. , 1977, Transactions, p. Y1-Y19. Jurado M. J., Bartrina T., Lamela M., Ruiz J. L. (2003). "Integrated Use of Dipmeter and Conventional Logs for Characterization of Fluvial Deposits of Triassic Argilo Greseux Inferieur Formation within the Rhourde El Khrouf Field, Berkine Basin ( Algeria )". AAPG International Conference . Paper _83673. Barcelona , España. Kerzner M.G. (1998). “Optimized Speed Correction”. In Transactions of the SPWLA (Society of Professional Well Log Analysts) Thirty-Ninth Annual Logging Symposium , May 26-29, 1998: Houston , Tex. , Society of Professional Well Log Analysts, Paper MMM, 10 p. Keskes N., et al (1987). “Application of an Interactive Statistical Classification System to the Analysis of High Resolution Dipmeter Curves”. Society of Professional Well Log Analysts Annual Logging Symposium, 28th , London , England , 1987, Transactions, p. O1-O24. Koepsell R.J., Jenson F.E., Langley R.L. (1989). “ Gulf Coast Fault Orientation Determined by Formation Imaging Techniques”. Society of Professional Well Log Analysts Logging Symposium, 30th , Denver , Colo. , 1989, Transactions, p. VV1VV25. Langford R.P., Grigsby J.D., Howard W.E. (1992). “Use of the Enhanced Density and Microresistivity Logs in Interpreting Diagenetic Facies in Tertiary Gulf Coast Sandstone”. Society of Professional Well Log Analysts Annual Logging Symposium, 33rd , Oklahoma City , Okla. , 1992, Transactions, p. BB1-BB16. Marshall A.G.B. (1976). “The "Cluster" Program Applied to Dipmeter Interpretation in South East Asia ”. South East Asia Petroleum Exploration Society , v. 3, p. 50-72. Matthews R.R., et al (1965). “Supplementary Computer Programs for Dipmeter Analysis”. Society of Professional Well Log Analysts Annual Logging Symposium, 6th, Dallas , Tex. , 1965, Transactions, v. 2, p. D1-D19. Moore E.J., Desai K.P. (1970). “A Multi-Pad Acoustic- Resistivity Dipmeter Tool”. SPWLA Log Analyst , v. 11, no. 2, p. 45-46. Moran J.H., Coufleau M.A., Miller G.K., and Timmons J.P. (1961). “Automatic Computation of Dipmeter Logs Digitally Recorded on Magnetic Ta pes”. Society of Petroleum Engineers of the American Institute of Mining , Metallurgical, and Petroleum Engineers Paper SPE 174, 19 p.
174
