DSI* Dipole Shear Sonic Imager. Eccentralized Dual ...... borehole sonic and density measurements and the ..... processe
SCIENTIFIC DRILLING
The KTB Borehole—Germany’s Superdeep Telescope into the Earth’s Crust
The drill bit has stopped turning and the KTB project is winding down. Germany’s superdeep borehole is complete. How and why was it drilled? And what have the scientists achieved so far?
Kurt Bram Johann Draxler Gottfried Hirschmann Gustav Zoth NLfB-KTB Hannover, Germany
Hamburg
Hannover
Stephane Hiron Clamart, France Miel Kühr Windischeschenbach, Germany
For help in preparation of this article, thanks to Peter Kehrer, NLfB-KTB, Hannover, Germany; Horst Gatto, Jochen Kück, Manfred Neuber, Thomas Röckel, Michael Sowa and Klaus Sturmeit, Windischeschenbach, Germany; Karl Bohn and Ignaz de Grefte, Schlumberger Wireline & Testing, Windischeschenbach, Germany; Pete Howard, Schlumberger Wireline & Testing, Houston, Texas, USA; Jörn Lauterjung, GFZ Institute, Potsdam, Germany; N. Riquier and Alain Poizat, Schlumberger Wireline & Testing, Abbeville, France. In this article CSU (wellsite surface instrumentation), FMI (Fullbore Formation MicroImager), Formation MicroScanner, GLT (Geochemical Logging Tool), Litho-Density, NGS (Natural Gamma Ray Spectrometry) and Sidewall CoreDriller are marks of Schlumberger. VAX is a mark of Digital Equipment Corporation.
4
Thermal gradients, heat production, stress fields, fluid transport, deep seismics and deep resistivity are all of great interest to earth scientists. Studying these fundamental topics helps them unravel the mysteries of weather fluctuations, the distribution of mineral resources, and natural disasters such as earthquakes, volcanoes and floods. Rock outcrops, river gorges and cliff faces provide visual evidence to interpret deep probing measurements such as seismics, magnetics and gravimetrics. Commercial mining and drilling have also guided scientists, giving tangible connections to surface observations (above). However, drilling has been used
specifically for scientific research only within the last thirty years. The internationally funded Ocean Drilling Program (ODP) was started as part of a worldwide effort to research the hard outer layer of the Earth’s crust called the lithosphere.1 Results from this project have been dramatic, providing real evidence of continental drift and plate tectonics. The lithosphere is made up of six major and several minor rigid moving plates. New oceanic crust is formed and spreads out at mid-ocean ridges and is consumed at active plate margins—subduction zones—where it sinks back into the Earth’s mantle. This process takes up to a few hundred million years. Continents are different. They are made of lighter rock and are not easily recycled,
Oilfield Review
Frankfurt
KTB
Munich
■ The lithosphere beneath Germany. Without superdeep boreholes earth scientists rely on surface observations to paint a picture of the structure of the lithosphere.
allowing them to achieve ages of 4 billion years.2 They also provide the vast majority of the world’s resources, so it is vital to understand their structure and development. One way of doing this is to extend the work started by ODP to the continent. KTB— which stands for Kontinentales Tiefbohrprogramm der Bundesrepublik Deutschland, or German Continental Deep Drilling Program—is drilling one of a handful of boreholes specifically for continental scientific research. This article looks at the major drilling achievements of KTB, at the Schlumberger wireline logging contribution and at some of the main areas of research. The project was initiated in 1978 by a working group of the German Geoscientific Commission of the German Science Foundation. The group discussed more than 40
possible drillsites in Germany, eliminating all but those with the broadest possible research potential. Two sites were chosen for further studies: Haslach in the Black Forest region of South Germany and Windischeschenbach 80 km [50 miles] east of Nürnberg in Bavaria, southeast Germany. In 1985, the Federal Ministry for Research and Technology gave the final approval for the KTB deep drilling program and both sites were comprehensively surveyed. Both geology and the expectation of a lower formation temperature gradient favored the Windischeschenbach site. The site is located on the western flank of the Bohemian Massif about 4 km [2.5 miles]
1. Anderson RN, Jarrard R, Pezard P, Williams C and Dove R: “Logging for Science,” The Technical Review 36, no. 4 (October 1988): 4-11. The lithosphere forms the hard outer layer of the earth. The rocks below are hot enough to deform freely and comprise the asthenosphere. The lithosphere slides over the asthenosphere with little resistance. As the rocks of the lithosphere cool, they thermally contract, increasing their density. At some stage the lithosphere becomes denser than the asthenosphere and starts to bend and sink back into the Earth’s interior. A subduction zone occurs where the descending lithosphere meets an adjacent overlying lithosphere. 2. Turcotte DL and Schubert G: Geodynamics: Applications of Continuum Physics to Geologic Problems. New York, New York, USA: John Wiley & Sons, 1982.
5
Bohemian Massif Other Variscan basement outcrops Basalt
Saxothuringian Paleozoic
Cretaceous
Erbendorf greenschist zone
Jurassic
ZEV gneisses
Permian and Carboniferous Granite
Hamburg Berlin
Franconian line
ST Praha
RH
ZEV metabasites ZEV- Zone of Erbendorf Vohenstrauss
Nürnberg MN KTB wellsite
NW NE
RH Rhenohercynian zone ST Saxothuringian zone MN Moldanubian region
5
km 0
Two-way transit time, sec
1.0
2.0
3.0
4.0
5.0
■ Location of KTB wellsite. The site is located on the western flank of the Bohemian Massif about 4 km east of the Franconian line. The local surface geology is shown along with the three-dimensional (3D) surface seismic time slice beneath the KTB site. This was generated from data recorded after drilling the pilot hole (KTB-VB). Labeled are major seismic reflectors—SE1 corresponds to a major fault zone associated with the Franconian line. This was crossed by the superdeep hole (KTB-HB) at the interval 6850 to 7300 m [22,575 to 23,950 ft]. The inset shows Variscan basement outcrops in Middle Europe, including the Saxothuringian and Moldanubian units, and the position of the KTB site in the southeast corner of Germany close to the Czechoslovakian border.
6
east of a major fault system—the Franconian line (left ).3 Scientists also believe it lies at the boundary of two major tectono-stratigraphic units in Central Europe—the Saxothuringian and Moldanubian.4 This boundary—which they hoped to cross—is regarded as a suture zone formed by the closure of a former oceanic basin 320 million years ago. This process gave rise to a continent-continent collision—forming a mountain chain and the present-day Eurasian plate. The mountains have long since eroded away, exposing rocks that were once deeply buried. Therefore, this area is ideal for the study of deep-seated crustal processes. In addition, geophysical surface experiments have shown that the area around the drillsite has unusually high electrical conductivity and strong gravimetric and magnetic anomalies, which deserve closer investigation. The scientific challenges for the KTB project all contribute towards understanding the fundamental processes that occur in continental crust. Among these are earthquake activities and the formation of ore deposits. The primary objectives, therefore, were to gather basic data about the geophysical structure below the KTB site, such as the magnitude and direction of stresses, so that the evolution of the continental crust might be modeled. Information about thermal structure—temperature distribution, heat sources and heat flow—was also needed to understand chemical processes such as the transformation to metamorphic rock and the mineralization of ores. Fluids also play an important role in temperature distribution, heat flow and the various chemical processes, so measurements of pressure, permeability and recovery of fluids found were also important. The overriding goal of the KTB project was to provide scientists with a permanent, accessible, very deep hole for research. With a budget of 498 million Deutsche Marks [$319 million]—provided by the German government—the initial target was to drill until temperature reached about 300°C [572 °F]—expected at a depth of 10,000 m to 12,000 m [32,800 ft to 39,370 ft]—the estimated limit of borehole technology. This includes drilling hardware, drilling fluid chemistry, cementing as well as the downhole instrumentation required for the various scientific experiments. Many technical spin-offs developed from the project.
Oilfield Review
To accomplish these objectives, a new drilling technique was developed that combined rotary drilling and sandline core retrieval techniques (right ). A modified land rig used a high-speed topdrive to rotate internal and external flush-jointed 51/2-in. outside diameter mining drillstring in a 6-in. borehole. This drillstring provided enough clearance inside to allow 4-in. cores to be cut and pulled up to surface through the drillpipe by sandline—eliminating round trips to recover cores. A solids-free, highly 3. Massif is a block of the Earth’s crust bounded by faults or flexures and displaced as a unit without internal change. 4. A tectono-stratigraphic unit is a mixture of lithostratigraphic units resulting from tectonic deformation. The Saxothurigian is a low pressure-high temperature unit still showing sedimentary structures. The Moldanubian is a low pressure-high temperature unit with relics of two older phases of higher pressure. The sedimentary structures have almost disappeared. 5. Rischmüller H: “Example for Advanced Drilling Technology,” Oil Gas 16, no. 4 (1990): 16-20.
January 1995
1994 KTB Oberpfalz HB scientific drilling
KTB Oberpfalz VB 1989 scientific drilling
1985 Mirow, GDR scientific drilling
1985 Kola SG 3, USSR gas Exploration
Zistersdorf UT-2, Austria 1983 gas exploration
Deep sea drilling project, 1976 Atlantic, Spain, scientific drilling
1974 Bertha Rogers, USA gas exploration
Munsterland 1962 gas exploration
Heide, Holstein 1938 oil exploration
Poruschowitz, Silesia 1893 coal exploration
1858 Wietze, Lower Saxony oil exploration
It is not common to drill through surficial crystalline rock especially when the drilling conditions are unknown. Kola SG 3, on the Kola peninsula near Murmansk, Russia, is one exception. It is the world’s deepest borehole, but not an ideal role model ( right ). After 15 years of drilling, at an untold cost, the borehole reached a depth of 12,066 m [39,587 ft]. Years later it was deepened to 12,260 m [40,200 ft]. The project management team, having studied the Russian project, decided to first drill a pilot hole—KTB Vorbohrung (KTBVB).5 This was spudded on September 22, 1987. The objectives for the pilot hole were as follows: •Acquire a maximum of geoscientific data, from coring and logging the entire borehole, at low cost and minimum risk before committing to an expensive heavy rig and superdeep hole. •Minimize core runs and logging in the large-diameter, straight vertical upper section of the superdeep hole. •Analyze the temperature profile for planning the superdeep hole. •Obtain data about problem sections with inflow or lost circulation, wellbore instabilities and breakouts. •Test drilling techniques and logging tools in preparation for the superdeep hole.
Neusalzwerk, Westfalia 1847 salt exploration
Drilling the Vorbohrung—Pilot Hole
■ Depths of historically important boreholes in Germany and around the world.
35 695
2003
3818
3930
4000
5966
8008 8553 9583 9101
Depth, m
1
12,260
2
3
4
5
■ Sandline coring technique. Four-in. cores are cut with high-performance, impregnated diamond core bits (1). As the cores are cut, they slide inside the core catcher barrel (2 and 3). When coring is complete, the drillstring is picked up off bottom to break free the core (4). A rig line and latching tool are run through the drillstring and latched onto the core barrel. The barrel is pulled out-of-hole through the internally flush mining drillstring (5). A new core barrel is run back into the drillstring and landed above the core bit to continue coring.
7
Schlumberger Logging Tools Tool
KTB-VB
Acoustic TeleScanner*
KTB-HB High-Temp.
✓
AIT* Array Induction Imager Tool
✓
ARI* Azimuthal Resistivity Imager
✓
Array-Sonic*
✓
Back Off
✓
✓
Casing Collar Locator
✓
✓
✓
Cement Bond Log
✓
✓
✓
CERT* Correlated Electromagnetic Retrieval Tool
✓
CET* Cement Evaluation Tool
✓
CNL* Compensated Neutron Log Tool
✓
✓
✓
DIL* Dual Induction Resistivity Log tool
✓
DLL* Dual Laterolog Resistivity tool
✓
✓
✓
DSI* Dipole Shear Sonic Imager
✓
✓
Eccentralized Dual Axis Caliper
✓
✓
FGT* Formation Gamma-Gamma Tool
✓
✓
✓
FMI* Fullbore Formation MicroImager Formation MicroScanner*
✓
✓
FPIT* Free Point Indicator Tool
✓
✓
GCT* Guidance Continuous Tool
✓
✓
Germanium Detector Tool
✓
GLT* Geochemical Logging Tool
✓
✓
GPIT* inclinometry tool
✓
✓
High Resolution Temperature
✓ ✓
HLDT* Hostile Litho-Density Tool Litho-Density*
✓
✓
MicroSFL* tool
✓
✓
Multifinger Caliper Tool
✓
✓
NGS* Natural Gamma Ray Spectrometry tool
✓
✓
RFT* Repeat Formation Tester
✓
SAT* Seismic Acquisition Tool
✓
Severing Tool
✓ ✓
Sidewall CoreDriller* SLT* Sonic Logging Tool Spontaneous Potential
✓
Stratigraphic High-Resolution Dipmeter
✓
TDT* Thermal Decay Time
✓
Variable Density*
✓
✓
✓ ✓
✓
✓
✓
✓
✓
KTB Logging Tools Auxiliary Measurement Sonde1
✓
✓
✓
Borehole Geometry Tool1
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Colliding Detonation Drill Collar Fluid Sampler Formation
Cutter2 ✓
MicroScanner*1
Gamma Ray1
✓
✓
✓
Single Shot
✓
✓
✓
Temperature Tool
✓
✓
✓
6-Arm Caliper
✓
Self Potential
lubricating mud system had to be used, because of the small clearance between the flush external surface of the drillstring and the borehole wall. This coring method worked well until February 1989 when excessive corrosion in the pipe joints required replacing the mining string with conventional 3 1 / 2 -in. externally upset drillpipe and core barrels. Coring operations had to be interrupted on other occasions—three times for directional drilling to bring the hole back to vertical and twice for sidetracking, because of lost bottomhole assemblies after unsuccessful fishing. However, a total depth (TD) of 4000 m [13,124 ft] was reached for KTB-VB on April 4, 1989, after 560 days of drilling and logging. More importantly, 3594 m [11,790 ft] of cores were recovered—a recovery rate comparable to those achieved worldwide in easier formations—and the hole was extensively logged with many different instruments (left and next page ). The drilling experience in KTB-VB proved invaluable to the planners of the superdeep borehole. For example, they encountered areas of borehole instability across fault zones; they had to modify the mud system to account for water influx and water-sensitive rock; they had numerous breakouts caused by the relaxation of stressed rock; and the formation dipped more steeply than predicted making it difficult to keep the hole anywhere near vertical. In total, the pilot hole presented a greater challenge for drillers than expected. For the next year, many experiments and measurements—such as hydrofracs, production tests and extensive seismic work— were carried out in and around KTB-VB. In April 1990, the hole was finally cased and cemented. Meanwhile, plans continued for construction of a new rig to drill the superdeep borehole about 200 m [656 ft] away.
■ Logging tools used by KTB. The logging tools listed in the tables were run in the pilot hole (KTB-VB) and the superdeep hole (KTB-HB) and were provided by various logging companies, universities and institutes. KTB bought several tools, some of which were developed for the project.
* Mark of Schlumberger 1 Bought from or developed by Schlumberger for KTB 2 Developed by SwETech (Bofors) for KTB
Oilfield Review
Other Companies’ Logging Tools Company
Tool
KTB-VB
EDCON
Borehole Gravity Meter
✓
Prakla-Seismos3
Geophone Survey
✓
Moving Source Profile
✓
Vertical Seismic Profile
✓
Leutert
✓
✓ ✓
Fluid Sampler
✓
High Temperature Fluid Sampler PREUSSAG
KTB-HB High Temp.
Fluid Sampler
✓
✓
Gyroscope
✓
✓
Steering Tool
✓
✓
Scientific Directional Drilling Company
Gyroscope
✓
Multi Shot
✓
Western Atlas
Multiparameter Spectroscopy Instrument Carbon/Oxygen Tool
✓
Pulsed Neutron Decay Time
✓ ✓
Segmented Bond Tool ✓
Z-Density Tool 3 Now part of Schlumberger Geco-Prakla
University and Institute Logging Tools University or Institute
Tool
KTB-VB
KTB-HB High-Temp.
Berlin University, Berlin, Germany
Heat Conductivity Logging Sonde
✓
Braunschweig University, Braunschweig, Germany
Fluxgate Magnetometer
✓
Bundesanstalt für Geowissenschaften und Rohstoffe (BGR), Hannover, Germany
3D Magnetometer
Deutsche Montan Technologie (DMT), Bochum, Germany
Borehole Televiewer
✓
✓
Eötvös Lorant Geophysical Institute, Budapest, Hungary
Induced Polarization
✓
✓
Frankfurt University, Frankfurt, Germany
Dipole-Dipole Redox Potential
✓ ✓
Geostock, Malmaison, France
Geophone
✓ ✓ ✓
Göttingen University, Göttingen, Germany
Gradient Magnetometer
✓
Lawrence Livermore National Laboratory, Livermore, California, USA
Geophone
✓
Los Alamos National Laboratory, Los Alamos, New Mexico, USA
Fluid Sampler
✓
Munich University, Munich, Germany
Magnetic Susceptibility
✓
✓
Niedersächsisches Landesamt für Bodenforschung (NLfB), Hannover, Germany
Induced Polarization
✓
✓
Petrodata, Zürich, Switzerland
Variable Amplitude Logging
✓
Scientific Academy, Swerdlowsk, Russia
Magnetometer
✓
January 1995
✓ ✓
✓
9
Drilling the Hauptbohrung
■ UTB 1. This is the largest land rig in the world, capable of drilling to 12,000 m [39,500 ft] and of handling extra-length drillpipe racked in stands of 40 m [130 ft]. The rig has many advanced features, meets high environmental standards, including full electric drive, a large degree of automation and soundproofing.
6. Wohlgemuth L and Chur C: “The KTB Drilling Rig,” Oil Gas 16, no. 4 (1990): 42-46. 7. Sperber A: “The Casing Concept for the Main Well,” Oil Gas 16, no. 4 (1990): 21-23. 8. Chur C, Engeser B and Oppelt J: “ Vertical Drilling Concept for the Main Well,” Oil Gas 16, no. 4 (1990): 26-29.
10
The superdeep hole—KTB Hauptbohrung (KTB-HB)—was spudded on October 6, 1990, and reached a TD of 9101 m [29,859 ft] on October 21, 1994. To drill to this depth in only four years required the design and construction of the largest land rig in the world—UTB 1 (left ).6 This rig could handle 12,000 m of drillpipe that required a maximum hook load of 8000 kN [1,800,000 lbf] —more than three times that of the rig used to drill KTB-VB. Mechanical wear and tear was expected to correspond to drilling 30 deep conventional wells, with over 600 round trips. Reducing trip time called for radical rig design. Using 40-m [130-ft] stands of drillpipe instead of the standard 27-m [90-ft] stands saved 30% of the time. However, long drillpipe stands meant the rig had to be 83 m [272 ft] high. To further increase trip efficiency, an automated pipe handling system was installed (right ). This consisted of a 53-m [174-ft] high pipehandler that grasped and lifted stands of drillpipe between the rotary table and star-shaped fingerboard for stacking in the derrick. Pipe connections were made by an iron roughneck that replaced traditional pipe-spinners and make-up tongs. Automatic slips gripped the pipe in the rotary table as connections were made and hydraulically operated elevator clamps did the derrick man’s job. More time was saved as pipes were connected or disconnected by retracting the traveling block out of the way as it traveled up or down the derrick. This allowed the pipehandler to operate while the traveling block was moving. The entire operation was controlled by a driller, pipehandler operator and two floormen. Only the floormen worked outside on the rig floor—the other two sat inside a control room at consoles equipped with video screens and gauges (next page, top). Computers controlled many of the operations of the pipehandler. Using a pipe conveyor to lift single pipes between rig floor and pipe racks saved additional time. Borehole torque and drag as well as the strength of the drillpipe are decisive factors when it comes to reaching great depths quickly and safely. Torque while drilling and excessive hook loads when pulling the string are caused by lateral forces and friction between drillstring and borehole wall. These two factors are increased by the weight of drillstring, borehole inclination and severity of any doglegs. Borehole trajectory influences not only drillstring design,
19 meters
Pipehandler
53 meters
Pipe manipulator arm
Traveling block with retractor system
Pipe conveyor
11 meters
■ Pipehandling. Fast round-trip operation was one of the main specifications for UTB 1. Three mechanical aids helped. First, the 53-m [174-ft] high pipehandler takes stands of drillpipe and stacks them in the derrick. Second, to allow continuous pipehandler operation during round trips, the traveling block is retracted out of the way as it moved up or down the derrick. Third, a pipe conveyor picks up and lays down single joints of drillpipe between rig floor and pipe racks.
but also any proposed casing scheme. A slim-clearance casing scheme cuts down on rock volume drilled, but also requires a near vertical borehole to minimize friction.7 Without active steering the borehole would start to build angle as was proved in KTB-VB. So KTB commissioned Eastman Christensen —now Baker Hughes Inteq—to develop a self-steering vertical drilling system (VDS) (next page, bottom ).8
Oilfield Review
Drilling instrumentation
DC motor controls Pump volume and pressure gauges Driller
Pipehandler
Drilling parameters such as hook weight, weight-on-bit, rotary speed and torque Pipehandler
■ Armchair drilling. The driller, surrounded by instruments and video monitors, controls drilling operations from the comfort of a control room overlooking the rig floor. The driller is joined by the pipehandler operator when pipe is tripped in or out of the borehole.
VDS-1 / VDS-2
VDS-4
VDS-3 / II Motor universal joint
Motor drive sub Nonrotating housing
Data pulser
Internal shock absorber Motor drive shaft Motor section
Rotating part
Rotating shaft Top stabilizer
Bearing assembly Top stabilizer
Sensors, electronics, valves, batteries
Rotating shaft Sensors, electronics, valves, batteries
Sensors, electronics, valves, batteries
Extendable ribs
Internal steering
Drill bit
Drill bit
Extendable ribs on bearing assembly
Drill bit
■ Vertical Drilling Systems (VDS). The VDS systems consist of positive-displacement motors to drive the drill bits, battery-powered inclinometers to measure deviation, and hydraulic systems to steer the drill bit. Any drift away from vertical is measured by the inclinometer and fed back to the hydraulic system. Two steering systems were developed. The first holds extendable ribs against the formation using mud pressure (left). Deviation corrections are made by releasing pressure in one rib, causing the whole assembly to move towards that rib. The second system is steered internally (center). Pistons move the gimbal-mounted rotating shaft, which is connected to the bit, to make any deviation adjustments. Another variation, VDS-4, reverted back to extendable ribs (right). (Courtesy Baker Hughes Inteq.)
January 1995
11
■ Lithology comparison between KTB-VB and KTB-HB and an overview of drilling and casing KTB-HB. The differences in lithology between the two boreholes—which are only 200 m apart—highlight the complex structure being drilled (left). Drilling difficulties required cement plugs to be set to correct for deviation and for sidetracks to be drilled after unsuccessful fishing operations (middle). Extra casing strings had to be run to protect crumbling borehole (right).
12
The VDS system consisted of a positivedisplacement motor to drive the drill bit, a battery-powered inclinometer to measure deviation and a hydraulic system to adjust the angle of the drill bit to correct for deviation. Two hydraulic systems were used: the first system operated external stabilizer ribs that pushed against the borehole wall moving the whole VDS assembly back to the vertical; the second system used internal rams to move the shaft driving the drill bit back to vertical. As long as battery power was maintained to the inclinometer, both systems operated automatically. Inclination, and other parameters such as temperature, voltage and systems pressure, were transmitted to surface by a mud pulser to monitor progress. The first 292 m [958 ft] of KTB-HB were drilled with a 171/2-in. bit and opened up to 28 in. before setting the 24 1/2-in. casing (previous page ). To meet the requirements of a vertical hole, a 2.5° correction to deviation was made as the hole was widened. The next section was drilled with a 171/2-in. VDS system to 3000 m [9840 ft] and completed at the end of May 1991. Teething problems with prototype VDS systems meant using packed-hole assemblies (PHAs) during maintenance and repair. Even so, average deviation for this section was less than 0.5°. The same strategy was used for the 143/4-in. hole—alternating between the improving VDS systems and PHAs. A high deviation buildup from 5519 to 5596 m [18,107 to 18,360 ft] during one PHA run led to the borehole being pulled back and a correction made for deviation. The hole continued on course to 6013 m [19,728 ft] where 133/8-in. casing was set in April 1992—horizontal displacement at this stage was less than 10 m [33 ft]. Drilling continued with VDS systems and PHAs and 121/4-in. bits. Within this section 45.7 m [150 ft] were cored, including 20.7 m [68 ft] with a newly developed, large-diameter coring system that gave 91/4-in. diameter cores. However, in July 1992 at 6760 m [22,179 ft], the bit became stuck. Eventually, after an unsuccessful fishing operation, the hole had to be plugged back to 6461 m [21,198 ft] and sidetracked. In March 1993, over an interval of 6850 to 7300 m [22,474 to 23,950 ft], a major fault system was crossed. The VDS system could not control deviation over this interval and another correction had to be made. This system was thought to be an extension of the main fault that lies along the boundary between sedi-
January 1995
ments to the west and metamorphic rocks to the east—the Franconian line. Along this fault system a displacement of more than 3000 m occurred, showing a repetition of drilled rock sequences. This signaled the start of the most difficult drilling yet and additional funds had to be provided by the German government to complete the project —bringing the total cost to DM 528 million [$338 million]. At 7490 m [24,573 ft], when the horizontal displacement was only 12 m [39 ft], the VDS system was abandoned, as borehole temperatures became too high for the electronics. The hole then started to deviate north (below ). Within the main fault system the borehole became unstable and more breakouts occurred. While tripping out-ofhole from 8328 m [27,323 ft], the drillpipe became stuck at 7523 m [24,682 ft]. Jarring eventually broke the downhole motor housing allowing the pipe to be pulled out but leaving behind a complicated fish. Several attempts to retrieve the fish failed and the
hole was finally plugged back to the vertical section—at 7390 m [24,245 ft]—and sidetracked. Drilling again proved difficult and so a 95/8-in. liner was set at 7785 m [25,541 ft] in December 1993 to protect this hardwon section of hole. Difficult drilling continued with a 81/2-in. bit down to 8730 m [28,642 ft]. Borehole instability prevented further progress and a 75/8-in. liner was set in May 1994. To bypass the unstable section, a sidetrack was made at 8625 m [28,297 ft] through a precut window in the liner. Funds to continue drilling were now running low and a decision was made to stop 476 m [1561 ft] later on October 12, 1994. More than four years after spudding, the hole had reached 9101 m with a final bit size of 61/2 in. However, the borehole had not finished with the drillers yet. Attempts to lower logging tools into the open hole failed. The last section had to be re-drilled and a 51/2-in. liner set, leaving only 70 m [230 ft] of open hole for the wireline loggers and other scientific experiments.
■ Superdeep borehole projection. The main graph shows the horizontal projection of the final three openhole sections drilled to complete KTB-HB—inset are the vertical projections. Down to 7490 m [24,573 ft], the horizontal projection wanders around the origin, indicating that the hole is nearly vertical. At this depth, the borehole temperature became too high to continue with the VDS drilling system and deviation built up rapidly. The hole followed a course labeled Hole 4 until drilling difficulties caused it to be abandoned. A new hole was started in the vertical section—Hole 5. After the casing was set, the last openhole section was drilled—Hole 6. The final directional survey, conducted at 9069 m [29,754 ft], showed a horizontal displacement of only 300 m [1000 ft].
13
■ Field laboratory core measurements—detailed density profile (left) and electrical conductivity measurements (right).
Data Collection and Analysis
The main center of scientific activity at KTB was the field laboratory with a staff of 40 including resident scientists and technicians (see “KTB Logging Center,” page 16 ). Here, experiments were performed on cores— mainly from the KTB-VB—drill cuttings and gas traces from the shale shakers, sidewall cores from the Schlumberger Sidewall CoreDriller tool, rock fragments from the drillpipe-conveyed cutting sampler and fluid samples collected during pump tests and downhole. The field laboratory provided cataloging and storage facilities and a data base of basic information such as petrophysical properties, mineralogy and lithology needed for further experiments (above ).9 More detailed long-term experiments were conducted at universities and research centers in 12 countries. Nearly 400 logging runs were made in KTB-VB—the pilot hole—with every available borehole instrument (page 8 ). And 266 runs were made in KTB-HB—the superdeep hole.10 The wealth of data acquired in the field lab allowed a rare opportunity to calibrate borehole log responses to core data in crystalline rocks—as opposed to sedimentary environments where their response is well known—satisfying one of the main objectives of KTB-VB. The formations that were cored and drilled consisted of metamorphic basement rocks—principally gneisses and amphibolites.11 Initially cores and rock fragments— from cuttings—were photographed and cataloged according to depth recovered. Microscopic analysis of thin sections assisted recognition of mineralogy and microstruc-
14
Core photograph
Core wrapping
BHTV image
Formation MicroScanner image
BHTV Transparency overlay 10 20
938
938
939
939
940
940
30 40 50 180 N E
0 180 S W N E
0 180 S W N
■ Post-orientation of cores. The cores are released from the core barrels on the catwalk, reassembled in storage boxes and marked by a black solid line and a red dashed line to show which end is top. They are then photographed in the field laboratory (far left). A photocopy is then taken by rolling the cores over the copier (left). This photocopy may then be compared to borehole images recorded by the Borehole Televiewer (BHTV) (right) or the Formation MicroScanner tool (far right) and the orientation of the black line, drawn on the core, calculated.
Oilfield Review
ture and assignment of rock type.12 By mapping the macroscopic structure and orienting it with borehole logs such as the FMI Fullbore Formation MicroImager image or Borehole Televiewer (BHTV) image, a structural picture of the borehole was gradually built up (previous page, bottom ). Petrophysical parameters, such as thermal conductivity, density, electrical conductivity, acoustic impedance, natural radioactivity, natural remanent magnetism and magnetic susceptibility were also routinely measured. In addition to determining the strength of rock samples, scientists made highly sensitive measurements of expansion of the cores as they relaxed to atmospheric pressure. Geochemists at the field laboratory performed detailed core analysis using X-ray fluorescence for rock chemical composition and X-ray diffraction for mineralogy.13 This analysis allowed a reliable reconstruction of the lithology. After comparing logs with cores, scientists at the Geophysical Institute at the University of Aachen were able to distinguish 32 distinct electrofacies corresponding to 32 minerals. This enabled borehole logs to contribute to and refine the lithological profile of the superdeep borehole, established from cutting samples and the limited cores available (right ). One contributor to the success of the logging operation was the GLT Geochemical Logging Tool. This provided concentrations of 10 elements present in rock: silicon, calcium, iron, titanium, gadolinium, sulfur, aluminum, potassium, uranium and thorium. Another tool with a semiconductor detector—germanium—was also used, which gave a higher sensitivity and provided the additional elemental concentrations of sodium, magnesium, manganese, chromium and vanadium. 14 By combining the GLT results with other measurements, minerals such as pyrite, pyrrhotite, magnetite and hematite could be quantified. Older logging techniques also proved invaluable. Abnormal Spontaneous Potential (SP) deflections occurred across mineralized fault systems. Other SP deflections combined with low mud resistivity readings from the Auxiliary Measurement Sonde (AMS) occurred at zones of water influx. When the AMS resistivity showed only mud and the SP showed a deflection, this was regarded as an indicator for mineralization. Uranium tends to concentrate at graphite accumulations so the uranium reading from the NGS Natural Gamma Ray Spectrometry tool was used as a graphite indicator.
Cutting
500
EFA log
Cutting
EFA log
1500
Cutting
EFA log
■ Comparison of Electro-Facies Analysis (EFA) log and lithology from cuttings. The EFA log was developed by scientists at the University of Aachen by comparing core data with logs from KTB-VB. They were able to distinguish 32 electrofacies corresponding to 32 rock types. The EFA log—produced from logs from KTB-HB—compares well with the lithology profile from cuttings.
2500
Resistivity Gamma ray Paragneiss 1000
2000
Catacl. paragneiss Metabasite Catacl. metabasite Plagioclase gneiss Cataclasite Alternation gneiss/amph. Hornblende gneiss Calc-silicate rock Lamprophyre
9. Emmermann R, Lauterjung J and Umsonst T: KTB Report 93-2: Contributions to the 6th Annual KTBColloquium. Geoscientific Results. Stuttgart, Germany: Schweitzerbart, 1993. 10. Bram K and Draxler JK: KTB Report 93-1: Basic Research and Borehole Geophysics (Report 14). Stuttgart, Germany: Schweitzerbart, 1993. 11. Gneisses are banded rocks formed during high-grade regional metamorphism. Included in this group are a number of rock types having different origins. Gneissose banding consists of the more-or-less regular alteration of schistose and granulose bands. The schistose layers consist of micas and/or amphiboles. The granulose bands are essentially quartzofeldspathic and may vary from 1 mm up to several centimeters in thickness. There are various types of gneiss depending on the rock origin. Paragneiss is from a sedimentary parent and orthogneiss from an igneous parent. Amphibolite is a metamorphic rock composed mainly of feldspars and amphibole, a group of inosilicates.
12. For more detail: Emmermann et al, reference 9: 67-87. 13. Emmermann et al, reference 9: 529-534. 14. Grau JA, Schweitzer JS, Draxler JK, Gatto H and Lauterjung J: “Elemental Logging in the KTB Pilot Hole—I. NaI-based Spectrometry,” Nuclear Geophysics 7, no. 2 (1993): 173-187. Schweitzer JS, Peterson CA and Draxler JK: “Elemental Logging with a Germanium Spectrometer in the Continental Deep Drilling Project,” IEEE Transactions on Nuclear Science 40, no. 4 (August 1993): 920-923.
(continued on page 19)
January 1995
15
KTB Logging Center
■ KTB logging center and Schlumberger crew. The logging center (top left) comprises an enclosed area— housing the logging unit—numerous offices for KTB, guests and Schlumberger personnel, a computer room and various workshops, including a wellequipped electronics laboratory (bottom left) and sonde laboratory (bottom right). Schlumberger Engineer Miel Kühr, Operator Karl Bohn and Electronics Engineer Ignaz de Grefte were assigned to KTB operations for the entire seven-year project (top right).
The KTB logging center is every wireline engi-
The winch unit houses the CSU wellsite surface
wiring to run through a tunnel below the rig yard
neer’s dream come true. Situated 60 m [200 ft]
instrumentation computer and a silent power
exiting at the capstan unit. The cable passes
from the rig, the logging unit is housed in a cov-
pack, which provides hydraulic power for the
around the capstan before going up the outside of
ered enclosure, providing space for calibration
winch and electrical power for the instruments
the rig. Here it passes over the upper sheave
and operation checks of logging tools (above).
(next page). The winch is extensively modified to
wheel attached to a retractable jib. When logging
Portable units off the enclosure provide mainte-
cope with the high cable tensions encountered
is required, the jib is extended out over the rig
nance workshops for electronics and hydraulic
during logging. A high-strength drum holds
floor so that logging tools may be connected to
sondes. Several offices are provided for the KTB
around 9500 m [31,200 ft] of cable, and a cap-
the cable and lowered into the borehole. Built
logging staff and Schlumberger personnel as well
stan at the foot of the rig reduces cable tension
into the rig floor is a series of tool magazines.
as a computer room equipped with a micro-VAX
from a maximum of 90,000 N [20,000 lbf] to a
These contain logging tools in a state of readi-
III for interpretation and presentation of results.
normal spooling tension of 4500 N [1000 lbf].
ness to run into the borehole.
The offices also provide workplaces for scientists
Other modifications allow additional tensiome-
from universities who run their own logging tools
ters and depth measuring systems to monitor the
temperatures approaching 260 °C requires spe-
into the borehole using the logging unit.
cable at different points of the rig-up.
cial hardware—cable, logging head and logging
The logging cable is permanently suspended in the derrick to save as much rig-up time as possi-
Logging at depths of 9100 m [30,000 ft] and
tools—to withstand ultrahigh temperatures and pressures. The cable also has to have the
ble. The winch is housed in a cellar in front of the logging unit allowing logging cable and ancillary
16
Oilfield Review
strength to pull the logging tools back to surface.
Capstan and hydraulic motor
At 9100 m, the normal logging tension is 67,000 N [15,000 lbf], right on the maximum safe pull for the special high-strength cable used. This cable has high-tensile steel armor wires and is thicker than standard to provide the strength. It has special insulating materials at the business end that allow logging at temperatures up to 260 °C for short periods of time.
Upper sheave wheel on retractable jib
If the borehole had gone any deeper, a twocable approach to logging would have been used. At 10,000 m, the high-strength logging cable would be in danger of breaking under its own weight—even accounting for the buoyancy effect of the mud. To reduce cable weight, a smaller
Logging unit housing CSU and silent power pack
diameter, lighter, high-temperature cable hooked up to a second winch would have been used to lower logging tools into the borehole for the first 3200 m [10,500 ft]. At this depth, the smalldiameter cable would be connected to the highstrength cable of the main winch and the journey into the hole continued. With this tapered cable configuration, the overall weight of cable is decreased, reducing the tension, and the highLogging unit
strength cable is at the top of the hole where the tension is greatest. This technique was used twice, but only with 1500 m [4921 ft] of smalldiameter cable. This approach was taken by Rus-
Yard
■ Permanent logging rig-up. The logging cable is rigged up permanently to allow logging at short notice and save rig time. The cable drum is in a cellar below ground level, so that the logging cable and ancillary cables can pass through a tunnel to the rig. At the side of the rig, the cable passes around a capstan—to reduce cable tension—and then up the outside of the derrick to a retractable jib holding the upper sheave wheel. Logging tools are pulled up to the rig floor during other rig operations and lowered into tool magazines. When the rig is ready for logging, the jib is lowered, aligning the logging cable with the borehole. Then logging tools are connected to the cable and winched out of the magazines before entering the borehole.
Cable
Capstan
sian well loggers to log the 12-km [7.4-mile] Kola borehole using a three-conductor cable. A special oil-filled logging head was developed by Schlumberger for the KTB project. This had
high-temperature Formation MicroScanner tool,
one component that should have been changed
high-temperature feed-throughs and special O-
which was upgraded to 260 °C (next page, left).
could result not only in a malfunction but also in
rings, and provided the connection between cable
The first task in modifying this tool was to pro-
destruction of expensive equipment.
and logging tool.
duce a list of components to upgrade. Several
Temperature limits on the mechanical aspects
components, such as the pads containing the
of the tool were relatively straightforward to over-
Although there are several standard high-temperature logging tools available, tools were
button electrodes, were not changed, but could
come. However, the electronics were of major
upgraded especially for KTB. One example is the
be used only once. Other components, such as
concern. Normally these operate up to 175 °C
the hydraulic motor that opens and closes the sonde calipers, could still be used more than once. Mechanical maintenance of such high-temperature tools has to be meticulous—using even
January 1995
17
Logging cable
Telemetry cartridge
Dewar flask
[350 °F]. To keep the temperature within this
On the first logging run at TD, the maximum tem-
limit meant housing them inside a Dewar flask
perature recorded was 240 °C [464 °F] and on
(below, right). The outside temperature could be
the last run, this reached 250.5 °C [483 °F]—
as high as 260 °C with the inside remaining
confirming earlier calculations. At the end of
below 175 °C for up to 8 hours.
each logging run the Dewar flasks were cooled down slowly by blowing air through to avoid
The cooling effects of mud circulation during
thermal shock.
drilling were calculated to be about 50 °C [90 °F]
Apart from Schlumberger logging tools, sev-
at TD. When circulation stopped, the temperature would gradually climb, giving a window of 36
eral universities developed equipment for their
hours for logging before it exceeded tool ratings.
own experiments in the borehole (see page 9).
Auxiliary measurement sonde Insulating sleeve
External temperature = 260 °C for 2.3 hours 280
Auxiliary measurement electronics
Near power supply Near inclinometer External temperature Computed
240
Nonmagnetic housing 25 kpsi at 300 °C
Gamma Ray
Thermal stopper
Temperature, °C
Dewar flask 200
Computed
160
120
Tool power on 80
Controller cartridge
Dewar flask Dewar flask
Tool power off
40
0 0
Standard 175 °C electronics Inclinometer
180
360
540
720
Exposure time, min
■ Temperature profile inside Dewar flask. Logging tools are heat-tested in ovens. With the oven temperature maintained at 260 °C for 2.3 hours, the temperature remained below 120 °C [248 °F] inside the Dewar flask.
Dewar flask Acquisition cartridge Caliper with button electrode pads
18
■ High-temperature Formation MicroScanner tool. The tool was developed for KTB and has a temperature rating of 260 °C. Standard electronics—rated to 175 °C [350 °F]—are protected inside Dewar flasks and sealed at the ends by thermal stoppers. Standard mechanical components may withstand this temperature, but the button electrode pads are changed after each logging run.
Oilfield Review
Surprises—Some Welcome, Some Not
Both boreholes yielded unexpected results for the scientists. Geologists had formed a picture of the crust at the Windischeschenbach site by examining rock outcrops and two-dimensional (2D) seismic measurements. At a depth of about 7000 m [22,966 ft] they had expected to drill through the boundary between two tectonic plates that collided 320 million years ago, forming the Eurasian plate. However, this boundary was never crossed, and the geologists have had to redraw most of the subsurface picture. Other unexpected results include core and log evidence for a network of conductive pathways through highly resistive rock, and in rock devoid of matrix porosity, an ample supply of water. Look at these findings in more detail:
N
460 m
nLithology and structure revealed by KTB-VB. Simplified lithology of KTB-VB (left) shows the alternating layers of metamorphic rock. The structural model (right) shows formations dipping 50° to 75° southsouthwest over the first 3000 m, followed by a rotation of the dip to the east with a much shallower dip of 25° in the fold hinge. These models were built from interpretation of core data and borehole images such as Formation MicroScanner images. The formation appears to have twisted and piled up.
1160 m
1610 m
2470 m 2580 m
Seismic Investigations—During the project, surface and borehole seismic measurements helped visualize the structure below the KTB site. The original picture had been formed from 2D seismic work undertaken before drilling. But the structural profile of KTB-VB showed a more complicated subsurface. Instead of a nappe unit, the formation followed a more tortuous path (right ).15 After KTB-VB was completed in April 1989, a year was spent on major seismic evaluation. The seismic work, under the joint responsibility of KTB and DEKORP —German Continental Reflection Seismic Profiling—was performed by Prakla-Seismos—now part of Geco-Prakla. This included a 3D survey over an area of 19 by 19 km [11.8 by 11.8 miles], vertical seismic profile (VSP) and moving source profile (MSP), using geophones in KTB-VB, and two wide-angle 2D seismic surveys with an offset of 30 km [18.6 miles] using vibrators and explosives as sources. The evaluation, conducted by a number of German universities and their geophysical institutes, utilized acoustic impedance calculated from borehole sonic and density measurements and the acoustic measurements made on cores in the field laboratory. The seismic processing was complicated by the tortuous
January 1995
3575 m
Alternating layers of gneisses and amphibolites Paragneisses Metabasite Amphibolite and hornblende gneiss
structure and the large seismic anisotropy. The results, however, gave a much clearer picture than the earlier 2D work and accurately predicted the major fault system drilled through between 6850 to 7300 m (page 6 ).16 It is now known that the borehole remained inside the Zone of Erbendorf Vohenstrauss (ZEV), a small crystalline unit tectonically placed between the Saxothuringian and Moldanubian units. There are indications that these metamorphic units of the Bohemian Massif have been uplifted 10 km [6.2 miles] since Variscan time—about 300 million years ago—and eroded to the present day surface.17 Future experiments have been designed to measure seismic anisotropy at greater depths, the spatial extension of seismic reflectors—such as the “Erbendorf” structure at a depth of about 12 km [7.4 miles]—and
the detailed velocity distribution between the two boreholes using seismic tomography. Seismologists will also take advantage of the superdeep borehole KTB-HB by recording downhole seismic waveforms emitted by earthquakes. In this way, surface noise will be reduced and the frequency content of the signal preserved. 15. Wall H, Röckel T and Hirschmann G: “Structural Information from FMI in Crystalline Rocks. Results from the KTB Drilling Site,” Document du BRGM 223 6th International Symposium—Continental Scientific Drilling Programs—Paris, France, April 1992. Nappe is a sheet-like rock unit emplaced by thrust faulting, recumbent folding or both. 16. Emmermann et al, reference 9: 137-164. 17. Indications come from the analysis, for example, of detrital muscovites from the sedimentary basin west of the Bohemian Massif and the study of eroded sediments. Typical techniques used in geochronology are the determination of cooling ages by radiometric dating.
19
– E 00 10
+
80 0E
60 0E
Naabdemnreuth
40
–
+
+ +
0E
18. Cataclastic rock is rock that has undergone mechanical breakage usually by dynamic metamorphism or faulting. 19. Emmermann et al, reference 9: 165-218.
0E
Electromagnetics—One of the reasons for choosing the Windischeschenbach site was to investigate the origin and nature of a lowresistivity layer recorded by surface measurements that appeared to be 10 km below the Earth’s surface. This is not unique to southern Germany, as similar layers are found in many continents around the world. To unravel the mysteries of this conductive layer, scientists pursued many different angles. Conductivity measurements on cores from KTB-VB showed high resistivity as expected in crystalline rocks. But then highly conductive graphite-bearing faults and cataclastic zones were found at various depths up to 7000 m [22,970 ft].18 These were also seen on borehole logs where abnormal SP deflections of more than 200 millivolts (mV) coincided with the graphite. Other logs, such as induced polarization—where the decay of a voltage applied at a surface electrode is measured downhole—showed conductive pathways potentially formed by veins of graphite and/or sulfides. At a much larger scale, when the KTB-HB was at a depth of 6013 m [19,730 ft] a dipole-dipole experiment was carried out. This consisted of using the casing from both holes to inject current into the formation (above, right ). The resulting potential field was measured around the borehole. Any changes in potential indicated a connection of an electric conductor to one of the casings, supporting the theory for a conducting layer extending over a distance of several hundred meters. The results showed that the conducting layer coincided with graphite deposits in a north-south striking fault system—the Nottersdorf fault zone. The faults from this system crossed KTB-VB at about 250 m [820 ft] and KTB-HB at about 1500 m [4921 ft]. Further experiments are planned to investigate the depth, thickness, electrical anisotropy and source of the high conductivity layer still believed to be at 10 km.
20
0E
–
20
–
0S
Reference electrode S 00
KTB-HB
E1 E2 E3
2
E4
0S
KTB-VB
40
Polarity I + – Polarity II – + 0S
60
U
22A 0
0S
80
Nottersdorf 00
2000
I
4V
KTBVB
KTBHB
S
m
10
4000
■ Dipole-dipole experiment. Current is injected into KTB-VB and KTB-HB casing. This sets up a potential field between the two. Surface electrodes (E1, E2, E3, E4) positioned hundreds of meters away are used to measure the distribution of the potential field with respect to a reference electrode. This distribution indicates positions of conducting layers.
MPa
Stress and Deformation—One of the goals of earth science is earthquake prediction, and ultimately reduction in earthquake risk. The physics of earthquakes requires an understanding of the movement of tectonic plates, the forces involved and role the crust plays in transmitting those forces. Many scientists think that the top 10 km of crust is brittle and carries most of the stress that moves the entire 100-km [62-mile] thick continental plates. They also believe that, with increasing depth, the crust becomes ductile and cannot support the stress. KTB research may help clarify the transition from brittle to ductile.
6000
Preliminary work in the two KTB boreholes has already determined the orientation of the local stress field.19 The four-arm caliper, resistivity imaging tools, such as the Formation MicroScanner tool, and acoustic imaging tools, such as the BHTV, were used to calculate the stress direction from analysis of two types of failure: shear failure of the borehole wall—called breakouts—and drilling-induced tensile failures. The former occur at an azimuth orthogonal to the orientation of the maximum horizontal stress. The latter are near-vertical fractures in the borehole wall in the direction of the maximum horizontal stress (next page, left). These fractures were easily identifiable on the cores cut in the KTB-VB and were oriented using Formation MicroScanner and BHTV images (page 14, bottom). The maximum horizontal stress is oriented to N 150° ± 10° E from surface down to 6000 m [19,685 ft]. To obtain the stress magnitude, hydrofrac experiments were carried out in both boreholes at various depths in conjunction with geoscientists at the Universities of Bochum,
Oilfield Review
Before Hydrofrac Orientation North 0
180
After Hydrofrac Orientation North
360
0
180
360
Temperature, °C
Depth m 0
150
300
Depth, m
0
1000
2000
2852
January
20-21
1990
March
2-4
1991
June
1-4
1991
October
11-17
1991
March
13-20
1992
February
7-12
1993
February 17, March 2 1993
3000
July
27
1993
August
12
1993
August
15
1993
January
2
1994
February
28
1994
4000
5000
Temp. grad.= 0.0276 K/m
6000
2853
■ Temperature profiles recorded in KTB-HB. The dashed line shows the extrapolated formation temperature estimated from borehole temperature profiles. Estimated bottomhole temperature (BHT) is between 260 °C and 270 °C [500 °F and 518 °F] at 9101 m [29,859 ft]. BHT calculations at other depths are marked. The borehole was cooled by mud circulation in the lower part of the hole and heated up in the upper sections. Time-lapse profiles, such as those recorded in March 1992 when the hole was at 6013 m [19,728 ft], show temperature increasing.
7000
8000
Stationary temperature measurements BHT calculations 9000
10,000
2854
■ Formation MicroScanner images before and after a hydrofrac in KTB-VB. The Formation MicroScanner image recorded after the hydrofrac clearly shows nearvertical induced fractures around an azimuth of about 150°.
and Karlsruhe, Germany and at Stanford University, California, USA. By fracturing the formation, the minimum and maximum principle stresses were determined. These and earlier tests in KTB-VB confirmed that the strength of the rock was increasing with depth, supporting the theory that the upper crust is strong enough to carry most of the stress of tectonic movement. Very recently, a hydrofrac experiment was carried out at 9000 m [29,528 ft] and is being evaluated.
100 °C 200 °C 300 °C
Thermal Studies—Of the many processes occurring within the continental crust, most are temperature dependent. Mapping the temperature distribution and measuring heat production, heat flow and thermal conductivity are therefore a vital part of understanding these processes. During the initial temperature mapping, KTB-VB held the unwelcome surprise that the formation temperature gradient was higher than anticipated. This disappointing result
meant that 300 °C—the set limit of current technology—would be reached at about 10,000 m—much shallower than originally predicted. Temperature measurements were carried out in the two boreholes during regular logging campaigns (above ). These were used to estimate true formation temperature. The borehole is cooled during drilling, by up to 70 °C [158 °F] in the deepest sections of KTB-HB. Formation temperature is obtained by recording several temperature profiles at preset time intervals as the hole heats up again and extrapolating these profiles to infinite time on a logarithmic plot. Each temperature profile was recorded during the first wireline logging run. This helped avoid another complication, disturbing the mud temperature profile by the logging tools. A wireline tool was even modi-
21
Caliper
Depth, m
fied at KTB with the temperature sensor mounted on the bottom of the tool to provide the least disturbance and give the best possible result. Temperature data provided an opportunity to measure heat production and conductivity. In addition, thermal conductivity measurements were carried out in the field laboratory on cores cut from the boreholes. From the NGS and Litho-Density data, heat produced by radioactive decay was calculated—for metabasites the results were 0.5 micro-Watts per cubic meter (µW/m3) and for gneisses 1.6 µW/m3. The final temperature profile has yet to be extrapolated from the data obtained so far. Experiments will continue to examine temperature distribution, heat production, heat flow and thermal conductivity.
Max. 300
mm
Caliper 800 10000
23.3.92 ppm
18000
Min. 300
Caliper mm 800 10000
24.3.92 ppm
18000
Bit size mm 300 800 10000
1.4.92 ppm
18000
28.12.92 ppm
18000
10000
3000
Lithology
Salinity NaCl equivalent
FS FS FS
FS FS 4000
FS
FS FS
Hornblende gneiss Alter. Seq. Musk bio. gneiss Gran. sill. bio. gneiss Catacl. gneiss Amphibolite Catacl amph. Lamprophyre Catacl. Alter. Seq.
■ Fluid detection and sampling. Pulling the drillpipe out of hole without refilling the borehole decreased the hydrostatic head, provoking inflow from the formation. A fluid sampler (FS) —run in combination with the Auxiliary Measurement Sonde (AMS) tool— collected samples. The AMS tool was used to detect changes in mud resistivity (track 2). Where this occurred, a fluid sample was taken (track 2). Also displayed are hole size (track 1) and lithology (track 3). All curves displayed were acquired in open hole.
5000 FS
Ca, Na, Cl
N, He, CH4 6000
Fluids—The scientists at KTB expected deep crystalline rock to be bone dry, but to their surprise, water influx occurred at several depths from open fractures. Sonic, Formation MicroScanner and BHTV data were used to detect the fractures. As fresh mud was used for drilling, any saline water inflow would cause a decrease in mud resistivity. This could easily be seen from mud resistivity measurements made by the AMS tool (above right ). These zones were allowed to produce by dropping the mud level, enabling a fluid sample to be collected by a wireline-conveyed sampler run in combination with the AMS tool. Tests showed the water had not leached down recently from surface. Further tests will be performed to ascertain the origin and composition and investigate fluid-rock interaction. During a two-month pumping test 275 m3 [1730 bbl] of salt water were produced from an open fracture system at the bottom of KTB-VB. Further evidence showed the extent of the fluid network. During a production test at 6000 m in KTB-HB, the fluid level in KTB-VB dropped. When the 13 3/8-in. casing in KTB-HB was cemented, there was an increase in fluid level in KTB-VB. These two events confirmed hydraulic communi-
22
Casing shoe
Casing shoe
7000
cation and allowed an estimate of permeability of the fracture system between the two boreholes.20 Natural causes of fluid movement became apparent when pressure sensors deployed in KTB-VB recorded changes in pressure due to earth tides caused by the gravitational pull of the moon. Fluids play an important role in the chemical and physical processes in the Earth’s crust, influencing mineral reactions, rheological properties of rocks and melting and crystallization processes. To aid further scientific research into these processes longterm pumping tests are planned between KTB-HB and KTB-VB to measure hydraulic communication, identify fluid pathways and collect additional fluid samples. The Future for Superdeep Boreholes
In 1996, the KTB boreholes will be handed over to GeoForschungsZentrum (GFZ), a German government-sponsored geoscientific institute based in Potsdam, Germany.
GFZ will continue the work started by KTB and the site will become a laboratory for deep measurements. Although the rig will be dismantled, the derrick will remain as a monument to KTB’s achievements. These achievements have inspired scientists all over the world to look again at superdeep boreholes with renewed enthusiasm. Many potential sites whet the appetite: the San Andreas fault zone, California, tops the list for studying earthquake activity; for volcanic studies, the Novarupta Vent in Alaska, USA; subduction zones at the Izu peninsula, Japan or the Hellenic subduction zone, Crete; and there is no larger continental collision zone than the intracontinental thrust in the Nanga Parbat region of the Himalayas. It is now up to scientists to convince governments to support international continental scientific drilling with the necessary funding. —AM 20. Bram and Draxler, reference 10: 337-365.
Oilfield Review
SEISMICS
Reducing 3D Seismic Turnaround
Imagine breaking your leg and having an X-ray, only to be told that the image won’t be ready for interpretation for a year or more. Until recently, seismic surveys suffered from similar delays. But thanks to breakthroughs in acquisition, processing and communication, 3D seismic turnaround time—time from the first shot to the beginning of interpretation—has been reduced from years to weeks. Chris Beckett Tim Brooks Gregg Parker Houston, Texas, USA
tion. Unlike 2D seismic, which grew from the exploration market into development, 3D seismic has grown in the opposite direction. Companies are discovering that early acquisition of 3D data reduces finding costs and overall project costs.2 Interpreted seismic data are essential for intelligent bidding on acreage. And some exploration contracts now require a 3D survey before drilling. This expansion into exploration, along with decreases in the cost of seismic acquisition and processing, has raised demand for 3D seismic data. This increased demand has forced service companies to reduce turnaround time— without sacrificing quality. This article looks first at the dramatic improvements in marine turnaround time, then at the steps being taken to significantly reduce turnaround in transition zone and land surveys.
Alan Strudley Robin Walker Stavanger, Norway
There are two main reasons oil and gas producers worry about the time spent on 3D seismic acquisition and processing, called turnaround time.1 First, in the oil and gas business, as in every business, time is money. The more time spent on drilling, logging and well completion, the longer the delay in production and the lower the profit. Add the time to acquire and interpret seismic data before drilling, and the delay in bringing reserves to the surface may grow beyond the schedules and budgets of many production managers. Second, and special to the oil and gas business, saving time can make the difference between being able to do business and not. Development contracts worldwide require oil companies to drill within a specified time. The clock starts ticking once acreage is licensed. A 3D seismic survey planned, acquired, processed and interpreted in advance arms developers with tools for intelligent well placement, yielding higher production from fewer wells. More 3D seismic surveys are also being commissioned for exploration, in addition to field development, their initial applica-
For help in preparation of this article, thanks to Gary Biro, Geco-Prakla, Dallas, Texas; Adrian Bligh, Scipio Brook and Colin Hulme, Geco-Prakla, Gatwick, England; Helge Bragstad, Peter Canter, Olav Lindtjorn and Odd Olav Vatne, Geco-Prakla, Sandvika, Norway; Bill Chapman, Conoco Inc., Ponca City, Oklahoma, USA; Kim El-Tawil, Tom Neugebauer and Mike Spradley, Geco-Prakla, Houston, Texas; Bill Fraser and Bruce Hinton, Hunt Oil Company, Dallas, Texas; Jakob Haldorsen, Per Helgaker, Hans Klaassen, Dietmar Kluge, Claus Schnellbacher, Tony Woolmer and Mike Worthington,
Geco-Prakla, Hannover, Germany; Hal Harper, Conoco, Midland, Texas; David Etherington-Brown, Johannes Hvidsten and Phil Selley, Geco-Prakla, Stavanger, Norway; Kristian Kolbjørnsen, Saga Petroleum, Sandvika, Norway; Bård Krokan, Norsk Hydro, Stabekk, Norway. In this article, Charisma, Digiseis-FLX, LINK, Monowing, Olympus-IMS, TQ3D, TRILOGY, TRINAV, TRIPRO, TRISOR and Voyager are marks of Schlumberger. RISC 6000 is a mark of International Business Machines Corporation. SPARCstation 20 is a mark of Sun Microsystems, Inc.
1. For the purpose of this article, 3D turnaround time is defined as the time from first shot to the end of processing. Some companies use the term “cycle time.” Oil companies include the survey planning before acquisition and the interpretation after processing, so their turnaround time is from decision to shoot to selection of drillsite. However, planning and interpretation are often outside the responsibility of service companies, so they remain outside the definition used here. 2. Chisolm G: “Advances in Delivering 3D Data to Customers,” presented at the PETEX 94 meeting on Techniques For Cost-Effective Exploration & Production, London, England, November 16-18, 1994.
Robin Bjoroy Dominique Pajot Paul Taylor Gatwick, England David Deitz Unocal Lafayette, Louisiana, USA Terje Flaten Lars Jan Jaarvik Statoil Stavanger, Norway Ian Jack Keith Nunn BP Exploration Stockley Park, England
January 1995
The Marine Story
Three years ago, a marine survey of 500 km2 [193 sq miles] took a year or more to be acquired and processed. Today, through a combination of new technologies, turnaround time for similar surveys can be as lit-
23
25
Months
20
15
10
5 0
Surveys prior to 1992
1992
1993
1994
■ Geco Gamma towing six streamers. Inset graph shows reduction in 3D marine seismic turnaround for a 500-km2 survey. Turnaround data from smaller or larger surveys are scaled up or down, accordingly, for the purposes of this graph. The last eight surveys, with dramatically lower turnaround, were processed with only a subset of the data. (Graph courtesy of BP Exploration.)
1987
Year
1991
1993
Acquisition Aquisition Positioning Positioning Processing Processing
1994
0
10
20
30
Time, weeks
40
50
■ Concurrence of the three phases of marine turnaround —acquisition, navigation positioning and processing. Overall turnaround has been cut as individual phases have been shortened and all phases now occur concurrently. Time corresponds to turnaround for a 500-km2 survey.
tle as nine weeks (above ). Technologies responsible for this dramatic reduction vary from faster acquisition capacity to highspeed links with shore-based computers for real-time, full-scale processing (left ). Today seismic vessels can acquire data 12 times faster than they could in the early 1980s, thanks to multielement acquisition—multiple air gun sources, multiple receiver streamers and even multiple vessels. 3 Prior to 1984, vessels towed one source array and one 3-km [1.9-mile] streamer (next page, bottom ). This configuration evolved to two streamers and two sources per vessel by 1986, quadrupling the area covered with each traverse, and decreasing the cost per unit area. In 1990, streamer length started to increase, also 3. For reviews of the state of marine seismic acquisition six years ago: Backshall l, Donohue R, Jamieson G, Kilenyi T, Naylor R, Staughton D, and Walker C: “Marine Seismics in Cameroon,” Oilfield Review 1, no. 1 (April 1989): 26-34. Hansen T, Kingston J, Kjellesvik S, Lane G, l’Anson K, Naylor R and Walker C: “3-D Seismic Surveys,”Oilfield Review 1, no. 3 (October 1989): 54-61.
24
Oilfield Review
decreasing costs. By 1991, there were two sources firing alternately to three streamers, and by 1992, there were four streamers. In 1994, the Geco Gamma acquired the world’s first survey with six streamers. And in a continuing quest for greater capacity, contractors are now building or refurbishing seismic vessels to tow 8 to 12 streamers. A challenge in designing vessels for multistreamer acquisition is to keep all the streamers uniformly separated while maintaining vessel speed. Streamers are separated with a deflector, which steers outer streamers away from their normal stream lines (right ). Most streamers follow angled slabs—paravanes—which deflect the streamer outward, but also create drag on the vessel. Each 3-km deflected streamer may exert up to 12 tons of drag, forcing the vessel to consume more fuel to maintain speed. Eight to twelve streamers, with paravanes deflecting the outer ones, would act like a sea anchor, creating enough drag to stop an ordinary vessel. One contractor, PGS Exploration, is designing a more powerful vessel to address this problem. Rather than design a larger, more expensive vessel to tow more streamers, GecoPrakla has designed the Monowing deflector. Acting like an airplane wing flying through
Monowing
Air gun arrays
water, this “lifts” the streamers apart, and results in a 500% increase in lift-to-drag ratio compared to conventional deflectors. The reduced drag increases acquisition efficiency, and also safety. The lower tension in the lead-in, or tow cables, between the vessel and the streamers, reduces the chance of a tow cable snapping and flapping back to hit the vessel. And unlike other deflectors, orientation of the Monowing can be controlled remotely, to act as a rudder for the streamer. This allows streamer spacing to be controlled from the vessel, and permits individual streamers to be spooled in for repairs. The Monowing deflector has already been deployed in the Irish Sea and West Africa, to tow six streamers. It is being
■ Seismic vessel towing six streamers with Monowing deflectors (top). Monowing towing technology steers streamers and reduces drag for increased towing efficiency and safety.
tested with five streamers at extra-wide 150-m [492-ft] spacing, making the 600-m [1980-ft] swath acquired in a single vessel pass the widest ever. Streamers themselves have also been upgraded. In earlier, analog streamers, hydrophones were wired to the streamer cables and the analog signal transmitted up the streamer and then digitized. There may have been signal leakage in the streamer, or cross-talk, in which a signal from one hydrophone gets mixed with that from another. With digital streamers, the signal is recorded digitally so cross-talk is eliminated. Digital streamers are also more reliable, resulting in less downtime and better turnaround.
■ Single-streamer and six-streamer acquisition. Single streamer (left, side view) acquires data from a narrow swath beneath the vessel. Six streamers (right, front view) acquire six times as much data in a wide swath.
January 1995
25
Front network
3000-m distance
■ Front and tail positioning networks. Geographic positions of every seismic source and receiver in the survey are determined using floats instrumented with Global Positioning Systems (GPS), acoustic range measurements, compass data and bearings and ranges from lasers. Streamer length is not to scale.
Tail network
Source Streamer Float Compass Gyro Hydrophone
While multielement acquisition has played the leading role in reducing acquisition time, it has created a new challenge in reducing overall turnaround time. Data can arrive at a staggering 5 MBytes/sec and some of it must be processed before the next shot is fired—about every 10 seconds —if the processing is to keep pace. Rising to the challenge is concurrent processing, a combination of onboard processing and high-speed communication with onshore computers and decision makers.
26
To achieve minimum turnaround time, two sets of data—source signature quality and survey position—must be processed between shots. The source is a cluster of different-sized air guns. On Geco-Prakla vessels the air guns are controlled by the TRISOR module of the TRILOGY integrated acquisition and processing system. This module fires the air guns in a sequence that is tuned to their sizes. As the size of the gun increases, so does the time from firing to maximum pressure. The TRISOR controller synchronizes the guns’ pressure maxima, giving a stronger source signal. TRISOR hardware also monitors source output to check the quality of each shot.
TRISOR sensors, located within one meter of the air guns, communicate with the vessel through fiber-optic connections, and are packaged based on concepts from Anadrill’s measurements-while-drilling (MWD) technology. In this hostile environment, near a high-energy source and sustaining at least 500,000 shocks per year, the rugged construction that ensures reliable MWD also helps reduce seismic turnaround. To maximize vessel uptime, errors such as a gun going off at the wrong time, or not at all, must be detected immediately. Then processing specialists can determine whether the shot must be retaken, or whether the recorded signal satisfies the geophysical objectives of the survey. If the signal is sufficient, time is saved. If insufficient, time is still saved, because a seismic line can be quickly reshot while the vessel is still over the survey area. The second set of data that must be processed between shots is survey position coordinates, called navigation data. Navigation data describe the position on the earth of every source and receiver point in the 3D survey. The data come from relative position measurements made with every shot as the vessel is in motion. The position of the vessel relative to satellites is determined using the Global Positioning System (GPS).4 Geographic positioning with GPS is a relatively new technique, more reliable and available than traditional radio positioning, and can fix locations to within two meters. The insea positions of the seismic sources and receivers are computed using directions from compasses mounted on the streamers and distance information—ranges—provided by acoustic sensors and lasers distributed in networks across the ends of the streamers (left ). The TRINAV module of the TRILOGY system collects the compass, laser and acoustic signals, detects transit times, processes them for range, computes the network node positions, calculates source and receiver positions and stores the results in a data base before the next shot is fired. The number of sensor data measurements— including compass data, laser ranges and bearings, satellite and radio position signals —used in such a calculation has grown from 15 in the days of single source and single streamer, to more than 350 now with dual sources and eight streamers ( next page, bottom ). 4. For a review of applications of GPS: “Talking Satellites,” Oilfield Review 4, no. 4 (October 1992): 70-72. 5. For a review of 3D marine seismic processing: Boreham D, Kingston J, Shaw P and van Zeelst J: “3D Marine Seismic Data Processing,” Oilfield Review 3, no. 1 (January 1991): 41-55.
Oilfield Review
Checking that the positions fall within the project specifications is a daunting task, and one whose automation has further reduced turnaround time. Until recently, this was done subjectively by navigation analysts, visually checking plots and position listings. Now, computed positions are quality assured using position acceptance criteria (PAC), automating the time-consuming task and slashing weeks off turnaround. The PAC are established by comparing the range in question to the range of the last shot. If the two are within a predefined threshold, the range is accepted. Deviations are flagged by the computer, making them easy to spot. As recently as 1993, some contractors made range measurements during acquisition and calculated rough initial positions, but waited until their return to shore to verify the calculations and link—merge, in seismic-speak—the seismic data traces with the corresponding source and receiver posiNavigation Processing Turnaround 14 12
Weeks
10 8 6 4 2 0 1992
1993
Area of survey NORWAY Bergen
Stavanger
N O R T H
N
tions. Three years ago, contracts typically allowed six to eight weeks for this process, but a difficult job could take six months. Now, the final position data can be made available in three hours (left ). While navigation data are being collected and processed, the seismic traces are beginning their journey through data processing. Essentially any processing offered by onshore processing centers can be supplied onboard. The entire processing chain is too elaborate to detail here. 5 But a few key steps, and how they are being streamlined to help reduce turnaround, are examined in the following case study.
1994
Year
In the summer of 1994, Statoil, in partnership with Saga and Mobil, conducted a 3D turnaround pilot project in block 33/6 of the
Navigation Sensor Data Per Shot 350
Sensor readings
* = Two-vessel acquisition
*
*
250
*
200 150 100 50 0 2D
1x1
1x2
2x2
2x2
2x3
3x3
2x4
2x5 2x6
First number is number of sources per vessel. Second number is number of streamers per vessel.
January 1995
■ Location of 3D seismic reducedturnaround pilot survey for Statoil, Saga and Mobil. The vicinity of the 3D survey (red outline) had substantial previous 2D coverage (black lines).
A Turnaround Breakthrough
■ Evolution of navigation data turnaround. Since 1994, positioning data can be fully processed onboard.
300
S E A
U. K.
2x6
2x7
2x8
■ Number of sensor measurements used in marine positioning calculations for different acquisition scenarios. The calculation must be made between shots—about every ten seconds—for onboard processing to keep pace with acquisition. Acquisition scenarios range from 2D to dual sources with eight streamers.
Norwegian North Sea (above ). The area had already been traversed with 2D lines. The acreage covered in the 3D survey was an extension of a play concept that had proven prolific to the south—the oil basin contains the Statfjord field, estimated at more than 3.5 billion barrels of recoverable oil, and the Snorre field. The 33/6 area will be part of concession round 15, recently announced by the Norwegian government. With this survey already acquired, processed and interpreted, the oil companies, acting individually, can make better decisions about how to bid for acreage. The goal of the pilot project was to turn around the 313-km2 [120-sq mile] survey in seven weeks. With conventional technology, such a survey would take 18
27
Low-frequency Noise
High-frequency Noise
■ Examples of lowfrequency (left) and high-frequency noise (right) detected with onboard processing. Low-frequency noise is caused by ocean swells during bad weather. High-frequency noise is caused by reverberations between sea surface and sea bottom, enhanced at particular water depths.
0
1000
Two-way time, msec
2000
3000
4000
5000
6000
Trace number
Trace number
Sail-line number
Raw Root Mean Square Noise
Filtered Root Mean Square Noise
40
40
35
35
30
30
25
25
20
20
15
15
10
10
5
5
1
1 400
350
300
250
200
150
100
50
1
400
350
300
Shot number
■ Noise level of survey data, before (left) and after filtering (right). High-level noise related to bad weather appears as red and orange bands along sail lines. Noise concentrated in one area, but spanning several lines (yellow upper left), is generated by a change in subsea topography. After filtering (right), the noise is still apparent, but within acceptable limits.
28
weeks: 6 for acquisition, then at least another 12 for processing. Executing such a tightly constrained survey requires exact planning. Survey design, acquisition parameter selection and choice of processing chain were given special attention by Statoil and Geco-Prakla geophysicists. In addition to these standard steps, during the planning phase it was recognized that to minimize turnaround time, both Statoil and Geco-Prakla would have to reevaluate accepted working practices: Statoil agreed to hold decision-response time to 12 hours, and Geco-Prakla agreed to increase computer and communication resources that would allow more rapid acquisition and processing. The Geco-Prakla vessel, Geco Gamma, was equipped with the latest technology for the job. Gamma had the TRILOGY system for onboard navigation and seismic data processing, and access to INMARSAT, the international marine satellite system. Three IBM RISC 6000s were installed to handle the near real-time processing, reproducing the software and hardware of an onshore processing center. The data would travel directly from the acquisition system to the memory of the TRIPRO onboard processing system. The plan called for crucial data to be transmitted via satellite and land lines to the Statoil office in Stavanger, Norway, where a work-
250
200
150
100
50
1
Shot number
-128
0
127
Amplitude
Oilfield Review
■ Effects of fold—the number of traces summed to create one stacked trace—on stack quality. In the survey design stage, Statoil and Geco-Prakla geophysicists considered processing lower-fold data to speed turnaround, but tests showed that only full-fold data would give acceptable results. Stack processing run on test lines shows that 4-fold stacking gives low signal-to-noise ratio and unclear reflections (top). Increasing the stack to 12-fold improves the visibility of reflections, but does not adequately suppress reverberations, called multiples, in the lower part of the section (middle). Full-fold, or 30fold stacking, produces a high-quality section (bottom).
Poststack Filtered Stack
Prestack Filtered Stack
Two-way time, msec
3500
4000
4500
600
550
Trace number
600
550
600
Trace number
550
Trace number
■ Testing effect of noise reduction after and before stacking. Stacked section of unfiltered data (left) shows noisy portion near center. Filtering the section poststack (middle) retains much of the same noise. Prestack filtering (right) cleans up the noise and produces an acceptable stacked section.
Time, msec
4-fold Stack
3500
4000
12-fold Stack
Time, msec
6. Stacking is the summing of traces with reflections that have a common subsurface point. The number of traces summed is called fold. Stacking reduces the amount of seismic data by a factor of the fold and increases signal-to-noise ratio. Processing takes longer when performed on prestack data, because of the greater data volume.
Unfiltered Stack
3500
4000
30-fold Stack
Time, msec
station was installed with the same processing and interpretation software. The first shot was fired on June 22, 1994, with the vessel towing two air gun clusters and four 3000-m [9840-ft] streamers spaced 75 m [246 ft] apart. The survey was 11 km [6.9 miles] wide and was completed in 38 vessel passes, making 293 lines. Some of the first lines were shot in bad weather, which created low-frequency swell noise, above the tolerance level set in the presurvey plan (previous page, top ). When that level is exceeded, many oil companies choose to shut down acquisition, and the vessel stands by, at up to $30,000 per day, waiting for weather to calm. But onboard processing showed that the noise could be filtered out, though the filtering would have to be done prestack (right ).6 By monitoring signal quality onboard, and processing the acquired, subspecification data in real time, Geco-Prakla geophysicists were able to decide that the processing scheme would tolerate the noisier data (previous page, bottom ). This eliminated the need to reshoot five or six lines, saving $70,000. The savings paid for the added cost of equipping the vessel with the RISC 6000s, and cut two days off the turnaround. Early in the planning, the team considered undertaking onboard processing of reducedfold data. But tests conducted prior to acquisition indicated that the reduced fold would give inadequate imaging of subsurface reflectors, so full, 30-fold data were processed onboard (right ). One of the crucial phases of the survey was the construction of the earth velocity
3500
4000 600
800
1000
1200
1400
Common depth point number
January 1995
29
Onboard Velocity Picking and Quality Control A
B
D
C
1545
1000
velocity, m/sec
Time, msec
1000
2000
3000
3500 2000
4000
1000
2000
3000
4000
Velocity, m/sec
1
2
3
4
5
6
7
Velocity function
Common depth point number
■ Onboard velocity picking and quality control. Stacking velocities computed by onboard processing are displayed as contours, the peaks of which can be identified in an interactive velocity picking window (A, black squares). Picks from the previous location are white squares, and picks from the next location are pink squares. Seven time-velocity curves, called velocity functions, are plotted as black curves. The picked velocities, applied to one common midpoint gather before stacking, yield flat reflections (B). The seven velocity functions are applied to one gather, yielding seven panels. The velocities are correct when they give clear, flat reflectors (C). Overlaying the stacked section on a color plot of the velocity field provides a quality check: changes in velocity coincide with major reflections (D).
model that would be used to stack and later to migrate the data.7 Geco-Prakla geophysicists analyzed velocities on 18 seismic lines selected at 500-m [1640-ft] intervals, and transmitted their results via satellite to Stavanger (above ). Statoil geophysicists loaded the data on workstations in their offices and worked weekends to monitor data quality and relay decisions on the quality of the velocity picks back to the vessel. A velocity model for the 3D volume was then built onboard. The last major step before stacking—3D dip moveout processing (DMO)—was also completed onboard for the 30-fold data. This process corrects for the reflection point smear that results when events from dipping reflectors are stacked (right ). The final stack volume was being built as soon as the last shot was fired, and inline migration begun while the vessel was steaming back to port. The computers and processing specialists were flown to Stavanger, where the final processing was completed three weeks later. Data quality was equivalent to that of a normal onshore processing job, and no immediate reprocessing was scheduled. Seven weeks after the first shot was fired, a Charisma workstation-ready tape was produced, waiting to be interpreted (next page ).8
30
Fasttracks and Quicklooks Migrated trace
Zero offset trace
Source
Midpoint trace
Receiver
MIG DMO NMO
Original data NMO –Normal moveout correction DMO–Dip moveout processing MIG–Migration
■ Effect of dip moveout (DMO) processing. In the case of a dipping reflector, DMO processing is required to correctly position a reflection signal. DMO is applied after normal moveout (NMO) correction and before migration (MIG). [Adapted from Sheriff RE: Encyclopedic Dictionary of Exploration Geophysics. Tulsa, Oklahoma, USA: Society of Exploration Geophysicists (1991): 89.]
Reduced-turnaround surveys are evolving rapidly, and the amount of processing that goes into each survey varies.9 Specialists divide reduced-turnaround surveys into two categories: fasttracks and quicklooks. Fasttracks are fast, fully processed surveys, like Statoil’s 33/6. Quicklooks are surveys that process a subset of the full data set—called low-fold—or that simplify processing, such as skipping dip moveout processing. Quicklooks give interpreters a head start on interpretation, allowing earlier exploration or development decisions and identifying areas that deserve more detailed processing. BP Exploration has conducted four such surveys offshore Vietnam with GecoPrakla, using onboard processing of navigation, low-fold data and widely spaced streamers to speed turnaround. In one case, BP had farmed into a prospect—taken over a license relinquished by another operator—with only two years remaining. At the time, the planned 3D survey would have taken six months for full-fold processing, compared to 11 weeks for a low-fold interim data cube. By getting the data earlier, BP interpreters were able to spend more
Oilfield Review
time understanding the prospect before the spud date deadline. Quicklooks can be considered preliminary or intermediate results, with potential to benefit from later reprocessing. One example is a 700-km2 [270-sq mile] exploration survey shot and processed onboard by Geco Resolution for Mobil in Papua New Guinea. Only portions of the survey were processed with full fold, saving some of the exploration money for drilling and development. Today, quicklooks and fasttracks alike are possible only if the onboard processing sequence is nearly set in stone during presurvey planning with tests on prior 2D data. If acquisition conditions require processing modifications, some, such as noise attenuation, can be accommodated during the survey. Further reductions in turnaround will come from improvements in all stages of acquisition and processing. Geco-Prakla researchers are looking into improved algorithms for navigation and seismic data processing. New, high-density data storage media now being introduced will mean
fewer tapes created, and speed data transfer wherever tapes are required.10 The wider availability of high-speed communication links such as LINK 100, which enabled the fast turnaround of the Statoil 33/6 survey, will make shorter turnaround the norm rather than the news. Geco-Prakla’s LINK 100 telecommunication service uses very small aperture terminal (VSAT) satellite technology to transmit data to office-based users via leased land lines or SINet, the Schlumberger Information Network.11 However, the greatest potential for improvement in 3D turnaround, lies not in marine seismic, but in land and shallow-water, or transition zone (TZ), environments. The Onshore Challenge
Today, turnaround for 3D land and TZ surveys can be only unfairly compared with that for marine surveys. The main difference is in acquisition, which in some cases may take 50 times longer on land than at sea. There is also little formal data on the trends in turnaround for land and TZ surveys, because no two surveys can be com-
pared. In the relatively constant marine environment, where every survey has roughly the same sources, receivers, subsurface and acquisition geometry, surveys of different sizes and from different areas can be scaled up or down for the purposes of keeping statistics. However, on and near land, every survey is different, and turnaround comparisons from one area to another may be meaningless. The environment may vary from swamp to arctic tundra, from desert to jungle. Sources, receivers and acquisition geometries come in as many combinations as there are environments. But in spite of the absence of statistics, land and TZ turnaround are improving. Paralleling improvements in marine turnaround, TZ and land surveys are seeing more reliable acquisition hardware, faster acquisition through multiple sources and more receivers, and real-time verification of source and receiver positions. The following two sections describe case studies—first transition zone, then land—to demonstrate some of the latest techniques to shorten turnaround.
327
7. Migration is a processing step that uses earth velocity information to position reflections at their true locations. 8. A Charisma workstation is one of GeoQuest’s seismic interpretation systems. For more information on interpreting seismic data: James H, Tellez M, Schaetzlein G and Stark T: “Geophysical Interpretation: From Bits and Bytes to the Big Picture,” Oilfield Review 6, no. 3 (July 1994): 23-31. 9. Hardy R and Haskey P: “The Changing Role of OnBoard Processing,” paper B029, presented at the 56th EAEG Meeting and Technical Exhibition, Vienna, Austria, June 6-10, 1994. Johnson DT, Bradshaw DG, Early RG and Done WJ: “Processing Marine 3D Seismic Data on Board During Acquisition,” paper B028, presented at the 56th EAEG Meeting and Technical Exhibition, Vienna, Austria, June 6-10, 1994. Taylor P and Keggin J: “Onboard Processing Can Be Done,” The Leading Edge 13 (November 1994): 1103-1105. Thornton RI, Reilly JM, Millard P and Johnson ML: “Real Time Offshore 3D Processing—A Case History,” paper B030, presented at the 56th EAEG Meeting and Technical Exhibition, Vienna, Austria, June 6-10, 1994. 10. Helical-scan magnetic tape—also known as VHSformat video tape—can store 25 GBytes on a cartridge, and will soon be able to store 365 GBytes. 11. SINet is managed by Omnes, a joint venture between Schlumberger and Cable & Wireless.
427
b
m
u en
lin
In
er
527
627
727 827 941
Two-way time, msec
2896
3144
3392
3640
3888
4136 4352 3007
2759
2511
2263
2015
1767
1519
1271
1023
Crossline number
■ Final 3D-migrated seismic data volume. The data cube is shown with a “chair” cut on a Charisma workstation. High-amplitude reflections are displayed in red and blue.
January 1995
31
water. Processing such surveys takes extra steps to account for the different responses of the various receiver types. The hydrophones used in the North Freshwater Bayou were attached to the DigiseisFLX system, a new, flexible transition zone acquisition system developed by GecoPrakla ( bottom ). Each Digiseis-FLX data acquisition unit (DAU) is a floating instrumented tube, tethered to an anchor and connected to four hydrophone groups (next
page, top ). Up to 1536 channels have been recorded in real time without reaching the limits of the system. This large number of channels allows for flexibility in arranging source-receiver combinations, often without moving the DAUs. Seismic data are transmitted to the acquisition boat using radio frequencies that can be adapted to avoid conflict with other radio activity. The Digiseis-FLX system presents advantages over other TZ equipment, called bay ■ Location of North Freshwater Bayou 3D transition zone (TZ) survey. IC O CEA N
CANADA
UNITED STATES
ANT Louisiana Texas
ppi River
MEXICO
Mississi
The North Freshwater Bayou in southern Louisiana, USA, was the site of a 3D survey demanding exceptional turnaround (right ). The acreage covered leases operated by Unocal and Exxon. Unocal was drilling at the time of the survey, and planned at least one additional well. Drillers, heading for a deep target below 4.0 sec two-way travel time, wanted to confirm the location of the target before reaching total depth. The challenge was to complete acquisition between the July 15 end of the alligator breeding season and the October 15 start of duck migration—a 13-week window of opportunity. Survey planners designed a 79-sq mile [200-km2] survey to be processed in two phases. Processing began on an 18-sq mile [46-km2] priority area, while acquisition continued over surrounding acreage. The shallow-water environment allowed an all-hydrophone acquisition. Some TZ surveys cross the line between water and land, and require a combination of receivers— geophones on land and hydrophones in the
ATL
Transition Zone
Alabama Mississippi
3D survey area
GU
■ Data acquisition units (DAUs). Each DAU consists of an instrumented tube, a square floatation pad and an antenna. DAUs are maintained and transported on the floating base camp.
32
EX OF M LF
ICO
cable. Bay cable consists of a 1/3-in. [0.8cm] diameter instrumented cable, two to three miles long, that lies on the sea bottom. The cable can shift with currents, and can be damaged by boat propellers and sharp coral. While radiotelemetry avoids these problems, the added flexibility creates a new problem, synchronization: each unit must record at exactly the same time. The Digiseis-FLX system uses a patented synchronization method, achieving an accuracy significantly higher than other radiotelemetry systems. Another innovation that contributes to the speed of the survey is the method with which the source explosives and the hydrophones are emplaced. The technique—ramming—is like using a hypodermic needle to inject a source or receiver into the earth. Ramming sources in soft transition zone cuts down on the time required to drill
Oilfield Review
source holes. On land, drilling crews typically drill 100- to 180-ft [30- to 55-m] deep shot holes in advance of the acquisition crew. Equivalent results are obtained with 40- to 50-ft [12- to 15-m] deep ram holes. Ramming not only takes less time, but it also costs less. Deep holes cost about $300 per hole to drill, while ramming costs about $75 per hole. Ramming hydrophones to a uniform depth of 20 ft [6 m] below sea level results in better receiver coupling and higher quality data. The main limitation of ramming is the restriction to unconsolidated earth. Not all the North Freshwater Bayou turnaround speed came from fast acquisition. Geometry verification—much like navigation data processing in the marine environment—carried out in the field, cut weeks off the normal processing time. Geometry verification, a feature of the Voyager mobile data processing system, checks that the source and receiver positions attributed to every shot record are correct. Usually this is checked back at the office after acquisition has been completed and the crew has left, but fixing errors after the fact is time-consuming. In some cases, entire land surveys have had to be reshot—a turnaround nightmare. One error typically encountered in geometry verification is a mistake in the identification of shot-point location. This can occur when the source, say a vibrating truck (vibro for short) is at the wrong location, can’t get to the right location, or if the location is missurveyed. It can also occur if receiver loca-
■ Transition zone crew deploying DAU from environmentally-friendly air boat.
tions are missurveyed, or if the wrong receivers are active. These mistakes can be detected quickly by applying some simple processing at the base camp, after the day’s acquisition (left ). The process is called linear moveout, or LMO. LMO compares arrival times recorded for a given source-receiver geometry to those expected for the same geometry, assuming a constant velocity subsurface. If the source and receivers are in the right places, the
■ Floating base camp. The camp can be towed between locations. Not having to dismantle and set up camp every day speeds turnaround.
January 1995
33
LMO process yields seismic traces with first arrivals aligned in time. Any other pattern of first arrivals indicates a mistake in the source-receiver geometry (below ). This technique was used in the Unocal survey to quickly verify geometry in the field. Catching errors with the crew still on site permits corrective action. Shot and receiver locations can be resurveyed if necessary to revise the location data base. Without this field verification, errors may be detected weeks or months later. Then, processing specialists would have to test several possible geometries in hopes of discovering what really happened, spending time and adding uncertainty. Verifying the geometry in the field saves up to four weeks in the office. With much of the time-consuming work out of the way, the computing center proceeded with the rapid disk-to-disk processing on a Sun SPARCstation 20. The fully processed 3D cube was ready three weeks after acquisition, in time for interpreters to use. Interpretation of the seismic volume signaled drillers that their target would be productive. Unocal interpreters were able to use the seismic data to confirm the quality of their next well location and plan at least one additional deep well at greater than 20,000 ft [6090 m].
Reducing Turnaround on Land
Three-dimensional surveys on land encounter many of the same difficulties as in transition zones, with the added problems of access, topography and extreme temperatures. All of these make for longer acquisition campaigns and more difficult processing. Under fair marine conditions, multielement acquisition can collect more than 75 km2 [29 sq miles] per day. Under extreme land conditions, such as −40° C [−40° F] arctic surveys, acquisition may proceed at less than 1 km2 [0.4 sq mile] per day. Land surveys of 1500 km 2 [586 sq miles] have taken up to 4 1/2 years for acquisition. The potential for improvement in land 3D turnaround is undisputed.12 In land surveys more than other types, presurvey planning is the key to minimizing turnaround. 13 Time spent planning and designing is more than compensated by time saved acquiring data. With a given set of equipment, say a certain number of geophones and people, one plan might achieve 150 to 200 shots a day, while a suboptimal plan with different shot and receiver line spacing may collect only 100 shots a day. The most time-consuming tasks in acquisition—be they laying out receivers, drilling shot holes, repairing damaged cables or advancing to the next vibro location—must
Geometry Error Two-way time, sec
0 0.1 0.2 0.3 0.4
Corrected Geometry Error Two-way time, sec
0 0.1 0.2 0.3 0.4
■ Field processing for source and receiver geometry verification. Linear moveout (LMO) processing, applied immediately after the day’s shooting, detected an error in the position of the shot fired to four receiver lines, as indicated by warped first arrival times (top). The next day, the shot point was resurveyed, and the new location input to LMO processing. The flat arrival times indicate correct geometry (bottom). 12. Jack I and Nestvold W: “3D Seismic—The Next Step,” Keynote Address at PETEX 94 meeting on Techniques For Cost-Effective Exploration & Production, London, England, November 16-18, 1994. 13. For a review of 3D seismic survey planning: Ashton CP, Bacon B, Mann A, Moldoveanu N, Déplanté C, Ireson D, Sinclair T and Redekop G: “3D Seismic Survey Design,” Oilfield Review 6, no. 2 (April 1994): 19-32.
34
be identified and minimized to reduce turnaround. In the following examples of 3D land surveys in Texas, such bottlenecks were identified during presurvey planning and circumvented in novel ways. Rough Terrain Turnaround
The Val Verde basin in Texas, USA is at the edge of the Sierra Madre mountains that extend north from Mexico ( next page, bottom ). The basin is a hot play for gas, with some wells in the region producing more than 7 MMcf/D. The terrain is extremely rough, with steep-edged mesas and incised canyons (next page, top ). Several 3D surveys in the area have contributed to the continuous improvement of field operating procedures. In one case, Conoco joined forces with Hunt Oil to acquire the Geaslin survey in the summer of 1994. Both companies had a short fuse: they had to evaluate their leases and make decisions for an early 1995 drill date. The survey design specified the number and location of shot points, but the short turnaround and high cost ruled out dynamite as a source, because too much time would be taken to drill shot holes. Vibro sources were available—four vibrating trucks at 12.5-m [41-ft] spacing constitute one source—but the terrain presented mind-boggling logistics: in some cases it would take four hours for a vibro trip up and down a mesa (next page, middle ). The solution was to use two sets of buggy vibros, or eight in all, similar to a dual-source marine survey.14 While one set was shaking in the valley, the other set would work its way up a mesa. Similar dual-source vibro operations have been extremely successful in desert areas, such as Egypt and Oman, where there are no obstructions. In this case they allowed acquisition of 60 sq miles [153 km2] in 65 days. As in all land jobs, darkness presents too many hazards, so the crew operates only during daylight hours. Evenings were well spent, though, running geometry verification on the day’s acquired data. One of the goals of the next shift was to have that day’s geometry checked and attached to the seismic traces, usually by midnight. That way, geometry problems could be fixed the next day, before the receivers were moved.
14. Buggy vibros are vibros equipped with wider than normal tires. This allows access to rugged terrain while causing less damage to the environment.
Oilfield Review
■ Rough terrain of Geaslin survey. The challenge of moving equipment on and off mesas was met by use of two sets of vibrator sources.
IC O CE
AN
CANADA
ATL
ANT
UNITED STATES
Louisiana
MEXICO
New Mexico Texas
3D survey area
MEXICO
■ Buggy vibrator source—vibro for short. Four such trucks shake in series to create a single source. Each vibro weighs 50,000 lbs [22,700 kg], is 30 ft [9 m] long and 10 ft [3 m] wide.
GULF OF MEXICO
■ Location of 3D land surveys Val Verde County, Texas.
January 1995
35
Processing the data from the Geaslin survey proved to be a great challenge. Val Verde basin is notorious for bad data. Highvelocity carbonates near the surface deflect much of the source energy away from deeper layers; receiver and source coupling to the surface varies with location; and the rugged relief introduces high residual statics—differences in seismic travel time through surface topography. After four months of testing and processing, including 3D DMO and migration, the processing was complete. The next step is interpretation, in preparation for a possible 1995 drill date. In the nearby Brown Bassett survey for Mobil, acquisition time was further shortened by the use of helicopters to move cables, recording boxes and geophones up and down the mesa and canyon walls. Three hundred “helibags”—net bags for transporting material—helped the crew complete the 60-sq mile [153-km2] acquisition in significantly less time than usual.
Windmill Receiver Array
Keeping track of all the information pertinent to a land survey is often the most timeconsuming job, and steps are being taken to shorten it and make fuller use of all the information available. The Olympus-IMS information management system, now in use by Geco-Prakla in Germany, is designed to do just that. The Olympus-IMS system colocates in a single data base the many types of data that must be handled in a land survey. Previously, every type of data had its own data base: the planned survey layout, the actual surveyed receiver and source point locations, shot hole drilling data, shooting schedule data and the recorded seismic trace data were handled by different software. The new integrated system minimizes the number of data handling steps, reducing errors and improving turnaround. The system will also link directly with processing software to allow field processing for geom-
Areal Source Array Pos 1 Vibro 1 x
2.5 m
What’s Coming to Land
etry verification and further processing steps. The Olympus-IMS system will be available in Australia and Texas by the middle of 1995. Further improvements in land turnaround will come from improvements in hardware and communication. In the most adverse conditions, a good crew may spend as little as two to three hours shooting out of ten spent in the field. In these circumstances, a small amount of time spent trouble-shooting equipment faults can have a considerable impact on turnaround. Geco-Prakla engineers are developing more reliable hardware, to reduce the amount of time spent looking for and repairing flaws in geophones, cables and connectors. Today, each receiver point marked on a map consists of up to 72 individual geophones, whose signals are combined to yield a less noisy signal at a central location, or source point ( below ). Up to 140,000 geophones will have to be repeatedly picked up, put down
Linear Source Array
Pos 1
Pos 1
Pos 1
x
x
x 12.5 m
120°
Receiver point
13
.75
Vibro 2 x
x
Vibro 3 x
x
x
x
Vibro 4 x
x
x
x
m
x
x
Source point
x x x x x x x x x x x x x x x x 3.125 m
Source point 46.875 m
12.5 m
■ Typical patterns for receiver and source arrays. When geophysicists talk about a receiver or source position, they nearly always mean the central position of an array of receivers or sources. Arrays are designed to attenuate surface noise. Up to 72 receivers can be arrayed around the central position, and up to 20 individual source positions can be summed to make one source point.
36
Oilfield Review
and maintained in the course of a 3D survey. Efforts are also underway to find new ways to acquire the same amount and quality of data with fewer receivers, cutting survey time. Improved communications will also cut turnaround time. Increased use of GPS is decreasing the time spent surveying positions for land source and receiver points. Surveying with GPS is faster and easier to check than traditional theodolitic surveying, and leaves less room for human error. Placing GPS units on vibro sources helps keep track of actual source locations and reduces location error. For arctic land surveys, snow streamers have been developed in collaboration with Norsk Hydro as substitutes for hand-placed geophones in an effort to increase acquisition efficiency. Geco-Prakla engineers have tested snow streamers in six programs, acquiring 1200 km [750 miles] of 2D data. Efforts are also underway to minimize environmental impact, which in arctic environments must be included as part of turnaround—a single drop of oil spilled must be recovered before the crew moves.15 Connecting land crews via satellite to SINet, the Schlumberger Information Network, will give better day-to-day contact with office bases, speeding equipment and supply requests and allowing interaction with processing centers. The first such satellite link has been made in Venezuela, and others are planned. Moving more processing to the field will further reduce turnaround for both land and transition zone surveys. Parameter testing, noise attenuation and velocity picking can be done with today’s field processing tools. But full concurrent processing, as performed in marine surveys, is still a dream for land. Land acquisition, more so than marine, is a three-dimensional problem: sources are not aligned with receiver lines, and more time is needed to acquire enough seismic traces to process one part of the 3D volume. At best, processing through to stacking could lag acquisition by a few weeks, but the difficult task of computing residual statics before stacking cannot begin until all the data are in. Advances may come from taking a new view of 3D land surveys—planning, acquiring and processing with a truly three-dimensional view—rather than simply repeating a series of two-dimensional snapshots.
January 1995
The Role of Integrated Services in Reducing Turnaround
Marine, TZ and land 3D surveys are sure to find further turnaround improvement in the common ground of integrated services. In an integrated-service survey, planning, acquisition, processing and project management are delivered by one service company. Traditionally, the oil company plans the survey, then one contractor acquires the data and another processes it. Time is wasted transferring data and responsibility between parties. Geco-Prakla has developed an integrated service for 3D surveys called TQ3D—Total Quality 3D. Larger in area than most surveys, TQ3D projects can cover leased and open blocks. A TQ3D project may be operated from 100% proprietary to 100% nonexclusive, or anywhere in between. Data acquired on a proprietary basis become the property of the operator. Large projects can involve several operators. Data acquired on a nonexclusive basis become the property of Geco-Prakla, and may be licensed. The turnaround improvement achievable through integrated services is remarkable. A mixed proprietary-nonexclusive TQ3D for BP in UK block 47/10 was started and completed in November 1994. Geco Topaz acquired the 230-km2 [89-sq mile] survey in three weeks. While full-fold data were being acquired, a 20-fold data volume was partially migrated onboard, and processing was completed onshore. Processed data were sent to the GeoQuest Data Services group via SINet, and converted to Charisma workstation format. Total project time was four weeks. Thirty-four such marine surveys have been completed, and 21 more are in progress, covering a total of 43,000 km2 [16,800 sq miles]. Integrated services are also reducing land survey turnaround. Land surveys, with their difficult logistics, benefit from the approach a committed team brings to a project. In addition to survey design, acquisition and processing, land surveys require obtaining permission to access an area from those who own and live on the land. The project can run more smoothly when a single contractor coordinates every phase. One such project in Africa turned around a 50-km2 [20-sq mile] survey in seven months, from planning through installation of processed data onto an interpretation workstation. Two other projects are in the survey design stage.
For marine, transition zone and land 3D survey turnaround, the journey is not finished, but the direction is clear. Marine surveys are in the lead, having made tremendous progress in the last three years by converting acquisition, positioning and processing to a set of parallel tasks. Some marine surveys that would have taken 10 years to acquire and process using 1980s technology are now completed in months. Some say marine turnaround is no longer an issue, that any marine survey can now be turned around within any explorationist‘s time constraint. Land and transition zone surveys, while lagging their marine counterparts, have made inroads with field processing, innovative sources and survey designs that optimize available equipment. These and further improvements will contribute to minimizing turnaround time, allowing oil companies to spend less time waiting for information and more time using it. —LS 15. For more on arctic surveys: Meyer H, Read T, Thomas J, Wedge M and Wren M: “Environmental Management in the Arctic,” Oilfield Review 5, no. 4 (October 1993): 14-22.
37
INTERPRETATION
Classic Interpretation Problems: Evaluating Carbonates
Mahmood Akbar Mario Petricola Mohamed Watfa Abu Dhabi, United Arab Emirates
In recent years, there has been a small revolution in our ability to evalu-
Mohammed Badri Mouhab Charara Cairo, Egypt
reveal many of their secrets.
Austin Boyd Denver, Colorado, USA
Carbonate reservoirs account for 40% of today’s hydrocarbon production, and because of several elephant fields in the Middle East they are expected to dominate production through the next century. Therefore, understanding carbonate reservoirs and producing them efficiently have become industry priorities and are likely to remain so. Current efforts in carbonate exploitation focus on correctly targeting new wells, frequently horizontal, to optimize production from untouched reserves and on ensuring that massive water injection schemes deliver an effective sweep of the reservoir. In support of these efforts, geoscientists are trying to decipher the enigma of carbonate rock’s complex pore space and understand how permeability barriers and conduits affect
Bruce Cassell Roy Nurmi Dubai, United Arab Emirates Jean-Pierre Delhomme Clamart, France Mike Grace Dallas, Texas, USA Bill Kenyon Ridgefield, Connecticut, USA Jon Roestenburg Jakarta, Indonesia
ate carbonates. Novel technology and a community of interpreters determined to crack one hard nut are forcing carbonate reservoirs to
For help in preparation of this article, thanks to Hervé Anxionnaz, Schlumberger Wireline & Testing, Clamart, France; Jean-Louis Chardac, Schlumberger Wireline & Testing, Dubai, UAE; Bob Dennis, Schlumberger Wireline & Testing, Wattayh, Oman; Philip Cheung, Arnaud Etchecopar and Ollivier Faivre, Schlumberger Wireline & Testing, Clamart, France; Brian Hornby, Schlumberger Cambridge Research, Cambridge, England; Rachel Kornberg, GeoQuest, Gatwick, England; William Murphy and Pabitra Sen, Schlumberger-Doll Research, Ridgefield, Connecticut, USA; Oberto Serra, Consultant, Paris,
38
reservoir behavior. This article tracks the interpretation process, from carbonate rock description and petrophysical log evaluation to new techniques for measuring permeability downhole and mapping large-scale flow conduits and barriers. Carbonates for Beginners
Carbonates and sand-shale rocks, or siliciclastics, are worlds apart. Whereas siliciclastic rocks are composed of a variety of silicabased grains that may have traveled hundreds of miles from their source, carbonate rocks mainly consist of just two minerals—calcite and dolomite—and remain near their point of origin.1 Carbonates form in shallow and deep marine settings, evaporitic basins, lakes and windy deserts.2 Most of the carbonates formed in past have shallow France; Julian Singer, Schlumberger Wireline & Testing, New Delhi, India; and Eric Standen, Schlumberger Wireline & Testing, Montrouge, France. In this article, ARI (Azimuthal Resistivity Imager), CMR (Combinable Magnetic Resonance tool), DSI (Dipole Shear Sonic Imager), EPT (Electromagnetic Propagation Tool), FMI (Fullbore Formation MicroImager), FracView, GeoFrame, MDT (Modular Formation Dynamics Tester) and TDT (Thermal Decay Time) are marks of Schlumberger.
Oilfield Review
60°
40°
20°
0°
20°
40°
60°
Reef
Shelf carbonate
Deep carbonate
Carbonate oil province
nDistribution of carbonates. Areas of modern sediments in shallow water are shown in yellow and orange, modern deepwater sediments in black-hatched lines. Prolific carbonate oil basins are shown in red-hatched lines. [Adapted from Wilson, reference 2; Blatt H, Middleton G and Murray R: Origin of Sedimentary Rocks. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. (1972): 410; and Perrodon A: Dynamics of Oil and Gas Accumulations. Pau, France: Bulletin des Centres de Recherches Exploration-Production Elf-Aquitaine (1983): 215.]
marine origins, but the most widespread type of modern carbonate is formed in deep water (above ). Silica-based rocks generally stand up to the rigors of geologic time, undergoing only minor alteration, or diagenesis. Their depositional record is preserved, with bedding planes on outcrops and subsurface correlations between wells clearly recognizable. The grains are regularly shaped, and the pore space, though complicated, remains intergranular. Carbonate rocks, on the other hand, are chemically unstable and undergo substantial alteration such as mineral dissolution and dolomitization—the replacement of calcium carbonate by magnesium carbonate. Carbonates house a jumble of complex particles, including a huge variety of biological origin, and an even more complex pore space (Photographs, reprinted with permission from Scholle, reference 3, are positioned throughout this article).3 This complicates the tracking of facies across a carbonate reservoir and the assessment of the productivity of a given carbonate formation.
January 1995
0.23 mm
nUnfilled interparticle porosity (black).
Holocene oolite, Great Salt Lake, Utah, USA. Cross-polarized light photograph.
The typical carbonate rock is made of grains, matrix and cement. Grains are either skeletal fragments of small organisms or particles precipitated from calcium-rich water. The latter includes a variety of small, accretionary grains identified according to their size, origin and internal structure. Matrix is the lithified mud of deposition that fills most of the space not occupied by grains. In carbonates, fine mud has several sources—chemical precipitation, breaking of skeletal material into finer material,
1. An exception is the less-abundant class of rocks called calcareous sandstones, or carbonate arenites, which form when carbonate rock is broken up by wind or water, then transported and deposited. Calcareous sandstones exhibit many of the structural and petrophysical characteristics of siliciclastic sandstones, while retaining carbonate mineralogy and microporosity. 2. For a review of carbonate geology: Wilson JL: Carbonate Facies in Geologic History. Berlin, Germany: Springer-Verlag, 1975. Reeckmann A and Friedman GM: Exploration for Carbonate Petroleum Reservoirs. New York, New York, USA: John Wiley & Sons, 1982. Tucker ME and Wright VP: Carbonate Sedimentology. Oxford, England: Blackwell Scientific Publications, 1990. 3. Scholle PA: A Color Illustrated Guide to Carbonate Rock Constitutents, Textures, Cements and Porosities. Tulsa, Oklahoma, USA: American Association of Petroleum Geologists, 1978.
39
remains of algae, and others. On lithification, mud becomes a very fine-grained calcite called micrite. Cement describes crystalline material that forms in most of the space remaining between grains and matrix or between grains themselves, binding them. Cement may have a variety of crystal sizes depending on its composition, the conditions of crystallization and the spaces to be filled. Crucial to the interpretation of carbonates is classifying the numerous ways grains and matrix coexist. Progress in categorizing these complexities surged in the late 1950s because of pressure within oil companies to better understand their carbonate assets. The classification that has stood the test of time most successfully is by Robert Dunham.4 Dunham classifies a spectrum of rock types based on the internal structure and texture of the rock (below ). Mudstone consists mainly of matrix in which relatively few grains are suspended. Wackestone is also matrix-supported but has more grains. Packstone has enough grains for them to start providing support—matrix fills the remaining nonpore space. Grainstone has plenty of grains providing support and includes progressively less matrix. Finally, boundstone describes carbonate rocks in which the original material provided support during deposition, such as in reefs. Crystalline describes rock that has lost its depositional fabric because of diagenetic recrystallization, for example, dolomitization. Dunham’s classification provides some clue to the energy of deposition. The mudbased mudstone and wackestone are Mudstone Wackestone Packstone
Less than More than Grain10% grains 10% grains supported Mud-supported
(black) in oolite. Upper Mississippian Pitkin Limestone, Oklahoma, USA. Crosspolarized light photograph.
deposited in low-energy settings. Packstone and grainstone would appear to be from high-energy deposition, but given significant diagenesis, these grain-supported rocks could equally well have been deposited as mud-supported agglomerations and then through compaction and chemical alteration transformed to their present state. The difficulty in classifying carbonates to reflect both their current state and depositional history demonstrates how dominant diagenesis is in forming the final carbonate rock. Diagenesis may be divided into five main mechanisms: compaction—the reduction of pore space in response to tighter grain packing as overburden increases; carbonate degradation—the destruction of carbonate material through chemical dissolution and micritization, the transformation of large crystals into small ones; carbonate aggradation—the construction of carbonate material through precipitation of cement between grains, and recrystallization, such as the
Depositional Lacks mud Original and is grain- components texture not supported were bound recognizable together
Original components not bound together during deposition
40
nCalcite-filled interparticle porosity
Grainstone Boundstone Crystalline
Contains mud, clay and fine silt-size carbonate
Depositional texture recognizable
0.07 mm
nDunham’s classification of carbonates, based on the internal structure of the rock.
replacement of limestone by dolomite; stylolitization —the formation of stylolites, irregular planes of discontinuity between rock units due to compaction-related pressure solution; and fracturing—the planar breaking up of rock mass due to stress. Time and diagenesis generally work against the preservation of porosity (next page ). Young carbonates usually have porosities around 60%. Old carbonates have just a percent or two. Reservoir carbonates survive with porosities of 5 to 15% largely because the presence of hydrocarbon impedes further destruction of porosity. The typically prolonged and extensive diagenesis in carbonates also usually obscures the provenance and history of the rock. Reservoir Description
How does the reservoir geologist use these descriptions to help plan the optimum exploitation of a carbonate reservoir? Identifying and classifying carbonates are crucial in two key tasks. First, assessment of the reservoir’s paleoenvironment builds a broad understanding of likely reservoir geometry. Then, detailed well-to-well correlation of lithofacies helps construct a detailed threedimensional picture. Clues to paleoenvironment come from every available source—seismic surveys, outcrop studies, cuttings and core analysis, and logs, including those from the latest generation of electrical imaging tools, which can capture the wellbore likeness to a resolution of about 5 mm [0.19 in.] The main paleoenvironment indicators are: •Lithology—This provides a general idea of the depositional setting. The presence of clastic rocks indicates an external source of sediment, while their absence indicates an environment free of external influence. •Rock texture—The Dunham classification of texture provides some idea of the energy of deposition. Grain size variations also point to the sequence of deposition. For example, a fining upward sequence may indicate a relative sea level rise, or marine transgression. A coarsening upward sequence probably indicates a relative sea level drop. •Sedimentary structure—Large-scale sedimentary structures are more difficult to see in carbonates than in siliciclastics, but when identified, they offer powerful clues to the depositional environment. Examples are crossbeds in eolian dunes or solu-
Oilfield Review
Formation of pores
Destruction of pores
Depositional environment Synsedimentary cement Micrite
Internal sediment
Lime mud Microdebris Peloids
1. Initial Porosity
Boring organisms
Marine waters Cement 2. Early Aragonite Diagenetic MagnesiumPorosity calcite en t
m ics
Recrystallization Intercrystalline
Dissolution Vugs Channels Original shell
Ov e
Geologic time
rb u rde n&
tec ton
d
rly
Ea
Ce
Fresh water Calcite
Burrowing sis organisms
ne
e iag
High energy Framework Intraparticle Interparticle Low energy Fenestral Intramicrite
Buried unfilled
3. Pressure-and TemperatureRelated Porosity
Compaction
Filled before burial
Tectonic activity Fracture
Recrystalliz
Pressure solution
Buried filled
Dissolve shell material
ation
Infillings
4. Erosional Porosity
Bury in sediment
ria is
bu
Fill cavity
di
ag
en
es
te La
Fill cavity
Dissolution Fissures Vugs Caverns
l
Calcite spar
Fracture Breccia Joints
0
25
50
Porosity, %
75
100
nCreation and destruction of carbonate porosity due to compaction and diagenesis, as a function of age. Inset—Transformation of a shell to create various molds and casts, as it undergoes different combinations of burial, filling and dissolution. (Adapted from Reeckmann and Friedman, reference 2.)
January 1994
tion grooves caused by irregular dissolution of the surface of a carbonate rock. •Biofacies—Identification of the wide variety of skeletal grains, burrows and molds may pinpoint precise geological time and settings. Ages and habitats of hundreds of carbonate creatures have been tabulated for this purpose. •Nonskeletal content—Grains formed by precipitation and accretion provide a powerful indicator of depositional setting. For example, homogeneous pellets develop in quiet lagoons, while concentrically layered ooids occur mainly in active, shallow environments. •Authigenic minerals—The cement and minerals that form in the rock after deposition provide some additional clues. The presence of pyrite suggests reducing (deoxidizing) conditions; glauconite indicates marine conditions; organic matter indicates little reworking. Slowly evidence accumulates, and the origin and evolution of the reservoir become less conjectured and more certain. In comparison, the mapping of lithofacies is a detailed, nuts-and-bolts task, but no less challenging. Unlike siliciclastics, carbonates usually carry no bedding signatures that allow ready well-to-well correlation across a field. Even dipmeter correlations across the borehole can be elusive. There are exceptions. Eolian carbonate deposits, for example, display continuity and bedding signatures exactly like their siliciclastic equivalents. More frequently, though, carbonates exhibit a mix of features such as fractures, breccia, stylolites and vugs. Helping the reservoir geologist recognize and catalog these features in wells drilled
4. Dunham RJ: “Classification of Carbonate Rocks According to Depositional Texture,” in Ham WE (ed): Classification of Carbonate Rocks. Tulsa, Oklahoma, USA: American Association of Petroleum Geologists, 1962.
41
XX87 XX15
XX88
XX17
Depth, ft
Depth, ft
XX89
nMottled fabric with thin producing zone at XX88 ft. Dark color is nSwarms of stylolites in a Mississippian carbonate, with two interpreted as mud-filled porosity. Light color is grains and matrix. core points visible.
XX92
XX93
Depth, ft
XX94
42
nSubhorizontal stylolites (wide dark bands) and inclined fractures (narrow dark lines) in a Middle East carbonate formation.
with water-base mud is the FMI Fullbore Formation MicroImager tool, which provides a picture of most of the borehole with 192 small current-emitting buttons mounted on four pads and four flaps. In the images, light color denotes high resistivity, indicating rock grains or hydrocarbon-filled pores, and dark color indicates low resistivity such as water-filled pores or shale. The images are no substitute for core analysis, but rather a complement to them. Other evidence is frequently needed to corroborate an interpretation, for example to decide whether a dark patch is porosity or shale. However, an experienced interpreter of FMI images can glean strong evidence of numerous types of carbonate features down to the centimeter scale (this page and next page ).5 A recent trend in FMI interpretation has been toward quantitative analysis of the images. One processing method automatically extracts five facies types based on a textural classification by Nurmi et al.6 The facies are uniform zones of constant conductivity or resistivity; layered zones of alternating conductive and resistive layers; zones 5. Serra O: Formation MicroScanner Image Interpretation. Houston, Texas, USA: Schlumberger Educational Services, 1989. 6. Nurmi R, Charara M, Waterhouse M and Park R: “Heterogeneities in Carbonate Reservoirs: Detection and Analysis Using Borehole Electrical Imagery,” in Hurst A, Lovell MA and Morton AC (eds): Geological Applications of Wireline Logs. London, England: Geological Society of London, 1990.
Oilfield Review
XX59 96.2
96.4 XX60
Depth, m
96.6
XX61
Depth, ft
nLarge calcite concretion, that may help elucidate depositional environment. XX26
nUnconformity between overlying shale and mid-Cretaceous carbonate. Large, irregular dark features below the unconformity are voids created by extensive dissolution during subaerial exposure. XX27
18.2
Depth, ft
XX28
18.6
nSecondary micrite formation in the form of large, light-colored areas.
XX84
nBreccia caused by collapse of a limestone cavern, indicated by interconnected channels between sharp fragments of rock. This type of breccia porosity often makes for prolific production.
January 1995
XX85
XX86 Depth, ft
Depth, ft
19.0
nMottled fabric of an Upper Cretaceous carbonate reservoir in the Middle East. 43
with interwoven or contiguous conductive areas, interpreted as interconnected porosity; resistive zones with isolated conductive areas, interpreted as nonconnected pores; and conductive zones with isolated resistive areas, caused for example by nonconducting calcite or anhydrite nodules. Such a zonation can be rapidly calculated from the images and lends itself readily to facies mapping across a field (below ). A more advanced processing method actually delineates identifiable objects such as rock grains or pores on the images.7 This is quite a challenge because picking the edge of an object depends somewhat on overall image intensity, which varies. The solution is to equate object boundaries with inflection points in image intensity. This approach is incorporated in SPOT—Secondary Porosity Typing—prototype software running on Geo-
Depth, ft
A
B
C
Quest’s GeoFrame platform. Current SPOT processing can yield the boundaries of both resistive (light color) and conductive (dark color) features (next page, top). In tests made on laboratory rock samples bored with “pores” of varying but precisely known diameter, the processing has given accurate and consistent pore delineation. Once resistive and conductive features are delineated, then all manner of quantitative information can be computed, such as their average size, the spatial density of the features, the total area on the image covered by the features, and the degree to which like features are connected. We will later Cond. Area C
THLY 0 t f 0.1 Cond. Area B REPC 0 m m h o 250 COPC 0
%
100 0 m m h o 250 Resistive with isolated conductive areas Layered
Resistive with isolated conductive areas XX05
Conductive with isolated resistive areas
XX10
nImplementation of Nurmi’s porosity classification through automatic processing. The FMI image is processed to yield resistive and conductive segments (B). Segments that are not continuous across the image are eliminated (C). Conductive area within each processed image is plotted as a log (blue for B, red for C). Average conductivity for resistive segments in B (green) and for conductive segments in B (orange) are shown with the variation in thickness of these segments across the image width (black).
44
0.26 mm
nUnfilled intraparticle porosity (black)
within a large coral fragment. Holocene back-reef beach sediment, Belize, British Honduras. Cross-polarized light photograph.
address how these new parameters may contribute to understanding of rock porosity and permeability. The average sizes of resistive and conductive features have recently been used to help identify Dunham rock types in an Occidental Oil Company field in Indonesia and thus contribute to facies mapping (page 46, bottom).8 In this interpretation, resistive features correspond to carbonate coral framework or grains, while conductive features correspond to pores or micritic matrix. On a log, the average sizes of the two types of features are played back together. The interpretation proceeds by noting the separation between the curves and also their absolute magnitudes. Mudstone is interpreted when separation is at a maximum. This occurs when the average size of conductive features peaks —that is, micritic cement dominates—and the average size of resistive features—or grains—drops. Wackestone is interpreted when the average size of resistive features increases, while the average size of conductive features remains about the same. Packstone is interpreted when the average size of resistive features peaks. And finally, grainstone is interpreted when the sizes of conductive and resistive features become equal. This broad-brush methodology has been verified against macrofacies descriptions from cores in two wells in the field. Furthermore, the frequency with which the two curves mirror one another appears to indicate the frequency of a complete depositional cycle—from low-energy mudstone to high-energy grainstone. 7. Delhomme JP: “A Quantitative Characterization of Formation Heterogeneities Based on Borehole Image Analysis,”Transactions of the SPWLA 33rd Annual Logging Symposium, Oklahoma City, Oklahoma, USA, June 14-17, 1992, paper T. 8. Roestenburg JW: “Carbonate Characterization and Classification from In-Situ Wellbore Images,” presented at the 23rd Annual Convention of the Indonesian Petroleum Association, Jakarta, Indonesia, October 4-6, 1994.
Oilfield Review
Number of resistive spots per ft
Orientation 0
30
60
90
120
150
180
210
240
270
300
330 360 0
Percentage of image covered by resistive spots,%
50 0
50
Depth, ft
XX62
XX63
XX64
XX65
XX66
nSPOT processing on FMI images delineates either resistive or conductive inclusions, or spots. The inclusions can then be analyzed to provide quantitative parameters such as median size of inclusion, density of inclusions per foot or meter, average area percentage of inclusions, and even a porosity estimate and connectivity parameter.
0.24 mm
0.24 mm
0.26 mm
nCross-polarized light photographs showing porosity. Left: Reduced interparticle and intraparticle porosity (black) in foraminifera and mollusks. Pleistocene Key Largo Limestone, Florida, USA. Middle: Intercrystal porosity (black) in a fine- to medium-crystalline replacement dolomite. Middle Eocene Avon Park Limestone, Florida. Right: Moldic porosity (black). Pleistocene Miami oolite, Florida.
January 1995
45
0
40 -2
2 0
100
Diagenesis
Core Description
Percent Recovered
Synthetic Seismic Trace
Depth, ft
Gamma Ray
Log Facies
– +
Sonic-Rt Overlay 1.0
Ohm-m
100
145 BOPD XX50
Packstone XX60
1660 BOPD
Wackestone XX70
Grainstone XX80
nFacies interpretation according to Dunham’s classification using statistical calibration between log data and petrologic analysis from cores in an offshore carbonate field, India. The calibration allows facies to be mapped across the field using other logged wells without cores. Well 1
Well 2 Conductive Average Size
Conductive Average Size FMI Image Resistive Average size mm2 30000
30
mm2 30000
Depth, ft
30
FMI Image Resistive Average size
Petrophysical Evaluation
XX70
W ackestone
XX72
Grainstone
XX74
Grainstone
Mudstone
nFacies interpretation in an Occidental Oil Company carbonate field in Indonesia using SPOT processing to automate Dunham’s classification from FMI images. Dunham rock type is interpreted by comparing the relative magnitudes of the average size of resistive and conductive inclusions. Interpretation has been verified from cores. 46
Without images, mapping facies following gamma ray and other log signatures can often prove unreliable. The safest bet, short of coring every borehole, is to collectively interpret all available log data, initially calibrating the interpretation results to core data. An example of this approach can be found in a study by the Indian Oil and Natural Gas Commission (ONGC) and Schlumberger that recently addressed a complex Middle Eocene carbonate formation in offshore India.9 In this study, the first step was to identify facies in the four cored wells according to Dunham’s classification. This required the analysis of 120 thin sections, 12 polished sections and 6 scanning electron microscope images. This petrologic interpretation was then integrated statistically with five log measurements made in the same wells— density, neutron porosity, sonic travel time, gamma ray and saturation. Matching the log measurements to the facies descriptions revealed clear links between weighted combinations of log data and the Dunham classification (above, left ). However, rather than Dunham’s four, the logs recognized five facies types, the last of which always occurred within a wackestone zone but at depths where no core was retrieved. This facies was termed wackestone+. With logs calibrated to a core facies description, a facies interpretation could be made directly from logs in all the remaining uncored wells, and then facies mapped between wells.
Facies determination from logs is hard enough, but the challenge of establishing petrophysical parameters such as saturation and permeability is even more daunting. The reason lies squarely with the complex diagenesis and resulting convoluted pore systems of most carbonate rocks. Log analysts divide porosity into primary and secondary components, with primary existing at the time of rock formation and secondary appearing as the rock matures and diagenesis prevails. The more detailed classification of Choquette and Pray exposes the immense diversity in both shape and size of carbonate pores (next page, top ).10 The variety in pore type explains why permeability answers remain so elusive. Vugs and their cousins may make for high porosity, but a consistent pore connectivity, usually taken for granted in sandstones, may or may not be present. Worse yet, the chaos
Oilfield Review
Fabric-selective
Not fabric-selective
Interparticle
Fabric-selective or not
Fracture
Breccia
Channel
Boring
Vug
Burrow
Cavern*
Shrinkage
Intraparticle
Intercrystal
Moldic
Fenestral
Shelter
*Cavern applies to man-sized or larger pores of channel or vug shapes
reigns at all scales. In sandstones, small 1/2-in. [1.27-cm] plugs bored from cores usually provide samples homogeneous enough for estimating average permeability. In carbonates, however, sometimes not even a whole piece of core can be regarded as representative. The discrepancy between permeabilities measured at different scales may be related to heterogeneity or to anisotropy.11 The only sure way of estimating reservoir-scale permeability is by using wireline, drillstem or production tests. This was the approach taken in the second phase of the Indian study, in which nine well tests in two wells established a link between carbonate facies type and permeability. Each carbonate facies type was allowed a permeability value, to be determined. Then, for each test, the well’s flow capacity calculated during the test was matched with the sum of the individual flow capacities of the well’s various facies types. Each facies’ flow capacity was the product of the facies type’s unknown permeability and its cumulative thickness in the well. The result of the match was a range of permeabilities for each facies type, two types—grainstone and wackestone+—being particularly permeable. Production logs in one well confirmed the productivity of wackestone+ (right ). A much earlier study, predating imaging technology, also recognized clear differences in permeabilities of the rock types of the Dunham classification. This study first
January 1995
Cumulative Flow 100
%
Entry
Facies
0 100 % 0
Grainstone
XX60
Packstone Wackestone +
Wackestone
nMatching well tests with Dunhamstyle evaluation in an Indian offshore carbonate field to provide permeability values for main facies types. Cumulative flow profile (left) and flow contribution from each layer (middle) confirm high permeability of wackestone+ (right).
XX70
Depth, ft
Growthframework
nChoquette and Pray’s classification of carbonate porosity: Fabric-selective porosity includes interparticle porosity occurring between grains; intraparticle porosity occurring within original skeletal grains; intercrystal porosity occurring within crystallized micrite and/or dolomite; moldic porosity that results from the dissolution of grains; large-scale framework porosity called fenestral porosity that usually results from dissolution of algal mat deposits; shelter porosity that describes pore space resulting from shelter offered by large overlaying grains; and growth-framework porosity that is the natural outcome of organic processes such as coral reefs. Nonfabric-selective porosity includes fracture porosity; channel porosity caused by extensive leaching; vug porosity that results from extensive dissolution of material and retains no evidence of the original host grain; and man-size cavern porosity that results from highly extensive and prolonged leaching. Porosity that could fall in either camp includes breccia, boring, burrowing and shrinkage porosity.
9. Vashist N, Dennis RN, Rajvanshi AK, Taneja HR, Walia RK and Sharma PK: “Reservoir Facies and Their Distribution in a Heterogeneous Carbonate Reservoir: An Integrated Approach,” paper SPE 26498, presented at the 68th SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, October 3-6, 1993.
10. Choquette PW and Pray LC: “Geologic Nomenclature and Classification of Porosity in Sedimentary Carbonates,” AAPG Bulletin 54 (February 1970): 207-250. 11. Ayan C, Colley N, Cowan G, Ezekwe E, Goode P, Halford F, Joseph J, Mongini A, Obondoko G, Pop J and Wannell M: “Measuring Permeability Anisotropy: The Latest Approach,” Oilfield Review 6, no. 4 (October 1994): 24-35.
47
and currently extends to the use of neural networks that attempt to mimic and even improve on our inherent ability to recognize patterns in diverse data.13 The other approach is to somehow directly measure something about the rock’s pore space, ideally from logs, and then tie this in with sought-after petrophysical parameters such as saturation and permeability. To this end, the newest measure comes from FMI images, again thanks to SPOT processing. The proportion of an image delineated as pore space leads directly to a new estimate of porosity, subject of course to the interpretation that dark areas of the image are indeed pores. In a well drilled through a carbonate reservoir in Italy, SPOT-derived porosity compares well with porosity conventionally interpreted from neutron and density logs (below, right ).
used a rudimentary log interpretation method to distinguish one rock type from another (below, left ).12 Once the type was identified, a relevant porosity-permeability relationship was applied at each depth to calculate permeability from porosity logs. The procedure resulted in far better agreement with core permeability measurements than had previously been obtained. In general, there are two ways to establish elusive petrophysical parameters such as permeability from log data. One is to link the parameter statistically to log data, calibrating the link with measurements of the parameter made in the field or laboratory. The calibration can be in just one well or an entire field. An example is the Indian offshore study where well test results were linked to a statistically derived facies interpretation. The variety of such statistical methods is immense
In much of the logged interval, the two porosities agree well, while elsewhere porosity derived from the FMI images is substantially less than conventional porosity. This could be due to the FMI tool responding only to pores larger than the 5-mm resolution of the tool and missing smaller intergranular and micritic pores. Interestingly, zones where the two porosities differ coincide with zones flagged with a secondary porosity index by the SPOT interpretation. Another SPOT calculation is connectivity, an elaborately conceived but necessarily limited attempt to quantify the degree of connection between pores identified on images. A limitation is imposed because two-dimensional images can say only so much about three-dimensional connectivity. Nevertheless, SPOT connectivity has successfully predicted the productivity of oil
me le
dia
iam
rtic
25
Conductive Area % 0
Pa
le d rtic Pa
P Limack e mston ud es,
ite lom e o n y d to gg kes u V ac W
0.1
Depth, m
00 µ ter
r= ete
eter diam ticle Par
=1
0µ = 50
rock efal in re nels
1
0
0 25 0
Secondary Porosity Index Variable m
50 5
Moldic grainstone
G dia rains me ton ter e, de por clin e ing
Permeability, md
10
FMI Connectivity
Porosity
X40
Leac
hed
chan
100
300 µ
1000
X50
ore
p alk ch µ h 1 olit cc er < Co met dia
X60
Wackestones Mudstones
0.01 0
10
Fractures
Porosity, %
20
Compaction and cementing
30
Leaching
nPermeability-porosity relationships for carbonate rocks following Dunham classification. X70
48
nFrom a carbonate field in Italy: an FMI image (left track) allows SPOT calculation of percentage area of conductive objects. Comparison (middle track) between SPOT porosity (orange curve) and neutron-density porosity (black curve) shows good agreement. Archie’s exponent “m” (blue curve), computed from Rxo and EPT Electromagnetic Propagation Tool logs, and connectivity of pore space computed from FMI image (gold curve) also correlate (right track), except where there is secondary porosity (yellow).
nData from Ordovician carbonate oil reservoirs in Oklahoma and Texas, USA, indicating a correlation between well production and connectivity derived from FMI images. The data pertain to wells known not to intersect fracture and stylolite systems.
500
Oil production, BOPD
400
300
200
100
0 0
8
4
12
16
20
24
28
FMI Connectivity
.5 5
φ nc
8
1
=
5
φf=
=
10
10
10
=0
φf=
6
φ nc =
2.5
4
3.0
2 φ nc = .5 =2 φ nc
φf =
k = 126.7 φ m R 2 millidarcies,
2
φ nc
1
,
20
φ nc
in which Rt , Rw and φ are, respectively, the water-filled formation resistivity, connate water resistivity and porosity.14 Early on, researchers realized that the cementation exponent captured something about the pore space, particularly its tortuosity, and thus could serve to estimate permeability as well as interpret resistivity logs. Several theoretical expressions for permeability based on m have been developed, this being a recent example:
2.5
Non connected porosity
0.8
Porosity, p.u.
φm
2.0
Fracture porosity
.1
Rw
1.5
0.5
φf = 1
Rt =
Cementation exponent, m 1.0
φf = 0
and gas wells in Texan and Oklahoman Ordovician carbonates with vuggy, connected porosity (above ). Without images, the commonest approach to pore geometry lies through consideration of Archie’s law with its cementation exponent m:
30 40
in which R is an “effective” pore radius in microns.15 The exponent m measures reasonably constant at about 2 for sandstones, as it does for similarly constructed oolitic carbonates. But otherwise in carbonate rock, it wanders all over. In fractured carbonate rock m tends to 12. Nurmi RD and Frisinger MR: “Synergy of Core Petrophysical Measurements, Log Data, and Rock Examination in Carbonate Reservoir Studies,” paper SPE 11969, presented at the 58th SPE Annual Technical Conference and Exhibition, San Francisco, California, USA, October 5-8, 1983. 13. Wiener JM, Rogers JA, Rogers JR and Moll RF: “Predicting Carbonate Permeabilities from Wireline Logs Using a Back-Propagation Neural Network,” Expanded Abstracts, SEG 61st Annual International Meeting and Exposition, Houston, Texas, USA, November 10-14, 1991: 285-288. Mohaghegh S, Arefi R, Ameri S and Rose D: “Design and Development of an Artificial Neural Network for Estimation of Formation Permeability,” paper SPE 28237, presented at the SPE Petroleum Computer Conference, Dallas, Texas, USA, July 31August 3, 1994.
January 1995
1, and in rocks with nonconnecting vugs m rises to 3, 4 or higher (below, left ).16 A particularly copious study on Qatar carbonates by Focke and Munn shows not only how m varies with porosity—it varies a great deal—but also how that functionality depends on permeability.17 The challenge in using m to evaluate a carbonate therefore depends on being able to reliably estimate the exponent at any depth, rather than use an arbitrary value, generally 2, derived from observations on sandstones. Guidelines for achieving this were first offered by Lucia of Shell Oil in 1981.18 Using samples from carbonate reservoirs in Texas, USA and Alberta, Canada, Lucia noted that m depended unambiguously on the proportion of the rock’s porosity coming from unconnected vugs. Estimate that from core samples, he suggested, and a likely m could be derived for selected intervals in the well. But a more versatile method was soon devised that permitted estimating m foot by foot. This made use of a new logging measurement—high-frequency electromagnetic propagation travel time, or tpl .19 Like the resistivity log, tpl responds to water-filled porosity, but does so without an exponent. Combining resistivity and t pl therefore allows elimination of porosity for a continuous evaluation of m. The results of such an m computation transformed the accuracy of carbonate evaluation in a number of Middle East fields (next page, top ). The methodology was later extended to take advantage of yet another wireline measurement, the TDT Thermal Decay Time log, permitting the continuous evaluation of not just m, but also the saturation exponent n.20 The exponent n appears in Archie’s law
50
Fractures
Vugs
nConjectured variation of Archie exponent m in fractured and vuggy rocks. (Adapted from Watfa and
Rt =
Rw
φ m Swn
,
adapted for hydrocarbon-bearing rock:
Nurmi, reference 16.)
14. “Archie’s Law,” The Technical Review 36, no. 3 (July 1988): 4-13. 15. Watfa M and Youssef FZ: “An Improved Technique for Estimating Permeability in Carbonates,” paper SPE 15732, presented at the 5th SPE Middle East Oil Show, Manama, Bahrain, March 7-10, 1987. 16. Watfa M and Nurmi R: “Calculation of Saturation, Secondary Porosity and Producibility in Complex Middle East Carbonate Reservoirs,” Transactions of the SPWLA 28th Annual Logging Symposium, London, England, June 28-July 2, 1987, paper CC. 17. Focke JW and Munn D: “Cementation Exponents (m) in Middle Eastern Carbonate Reservoirs,” paper SPE 13735, presented at the Middle East Oil Technical Conference and Exhibition, Bahrain, March 1114, 1985.
18. Lucia FJ: “Petrophysical Parameters Estimated from Visual Descriptions of Carbonate Rocks: A Field Classification of Carbonate Pore Space,” paper SPE 10073, presented at the 56th SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, October 5-7, 1981. 19. Freeman DW and Henry KC: “Improved Saturation Determination with EPT,” paper SPE 11466, presented at the SPE Middle East Oil Technical Conference, Manama, Bahrain, March 14-17, 1983. Amin AT, Watfa M and Awad MA: “Accurate Estimation of Water Saturations in Complex Carbonate Reservoirs,” paper SPE 15714, presented at the 5th SPE Middle East Oil Show, Manama, Bahrain, March 7-10, 1987. 20. Watfa M: “Using Electrical Logs to Obtain the Saturation Exponent (n) in the Archie Equation,” paper SPE 21415, presented at the SPE Middle East Oil Technical Conference and Exhibition, Bahrain, 49 March 2-5, 1994.
0
Secondary Porosity, p.u.
25
Lithology Analysis 0
%
100
Cementation Exponent, m Fluid Volume 7.00
100.00
0.00
%
Depth, ft
2.00
Residual oil
XX00
Moved oil
Limestone
Dolomite
[Reprinted from Nurmi et al.: “Typecasting Heterogeneities,” Middle East Well Evaluation Review, no. 10 (1991):16-32.]
XX50
Water
nCarbonate rock’s resistivity-versussaturation behavior for three carbonate samples, indicating dependence on pore-size distribution. The straight-line behavior at high saturations is likely dominated by large pores, while the behavior at low saturations is dominated by small pores. Resistivity index is defined as the ratio of rock resistivity at an arbitrary saturation to rock resistivity when water saturated. (Adapted
10 8
Resistivity index
6
4
2.5
1.5
0.04 1 50
Microns 60
63
from Bouvier, reference 22.) 70
Water saturation, %
50
nArchie exponent “m,” determined from EPT and resistivity logs, compared with secondary porosity, determined by subtracting sonic porosity from neutron-density porosity—in the Arab carbonate formation. In the pure limestone zone at the top, the two parameters match well. In the dolomite zone at the bottom, they diverge, indicating the presence of unconnected vuggy porosity.
80
90
100
in which Sw is water saturation. Like the exponent m, n also runs into trouble in carbonates, sometimes varying dramatically from the conventionally assumed value of 2.21 This has been shown in several sets of experiments on cores, the most recent by Bouvier (below ).22 Petrophysicists suspect the likely cause of discrepancy is tiny micropores in the micritic matrix. Most probably, these small pores still contain original water while the large pores contain oil. It is also probable that the micrite remains water wet, while the grains have become oil wet. Both phenomena would explain why carbonate formations producing only oil sometimes exhibit low resistivities more characteristic of a water-bearing formation. Essentially, the water-filled micropores provide a short-circuit to the survey current. In summary, the classic Archie approach for analyzing the complex pore geometries of carbonate is fraught with obstacles, which have been only partially overcome. New Logging Techniques
Today, two new techniques—nuclear magnetic resonance (NMR) logging and Stoneley wave logging—offer new perspectives on carbonate permeability and pore structure. The theoretical foundations for both techniques have been known for years, but until recently neither has received adequate technical implementation. That is changing with the introduction of the CMR Combinable Magnetic Resonance tool and the DSI Dipole Shear Sonic Imager tool. In nuclear magnetic resonance, sharp magnetic pulses are used to momentarily reorient hydrogen molecules away from the ambient magnetic field direction. After each pulse subsides, the hydrogen molecules realign themselves with the ambient field, oscillating about it as they do so. Observing these oscillations permits measuring how many hydrogen molecules relax after the imposed magnetic pulse and also the rate at which they realign to the ambient field, called the relaxation. The implications for logging are dramatic. The measurement of how many hydrogen molecules relax provides a measure of porosity, and the relaxation times indicate the size of pores containing the hydrogen molecules. Relaxation times are short in small pores because the hydrogen molecules are near the grain surface where interaction with surface charges speeds relaxation. Relaxation times are longer in large pores. Measuring the spectrum of relaxation times—so-called T 2 relaxation
Oilfield Review
NMR Measurement
Signal amplitude
1
0 0
100
200
300
400
Time, msec T2 Spectrum
Signal distribution
8
0 0.1
1.0
10
the formation. This rules out the possibility of a borehole signal, a problem that plagued earlier technology that used instead the much weaker and unfocused earth’s field. Eliminating the borehole signal used to require the expensive and unpopular technique of doping the entire mud column with magnetite. The new tool’s depth of investigation is about 1 in. [2.5 cm], and a dead zone directly in front of the pad avoids most effects from mudcake or rugosity. Vertical resolution is just 6 in. [15 cm], facilitating comparisons with the high-resolution FMI logs. Recent CMR logs run in carbonate formations in West Texas coupled with laboratory measurements on cores from the wells illustrate exciting possibilities for overall petrophysical evaluation. 24 The formations in question are partly dolomitized carbonates with a good deal of nonconnected vuggy porosity. In addition, silt layers create vertical permeability barriers. The main interpre-
tation challenge is to estimate at any depth what percentage of porosity actually contributes to production. This requires being able to discount the minute pore space in the silt and also any vuggy porosity that is not connected. T2 spectra were measured on water-saturated cores both before and after they had been centrifuged to expel all producible water (below ). Before centrifuging, the spectra show water-filled porosity covering the full range of pore sizes, while spectra after centrifuging no longer show the large pore sizes, since the water has been expelled from them. Equating the porosity difference between the two spectra with the volume of water expelled during centrifuging established a T2 cutoff of 95 msec to divide large from small pores. Applying this cutoff to spectra measured by the CMR tool provided an estimate of small-pore porosity that correlated well with silt intrusions evaluated
1000
100
T2, msec Core Sample 1
Pore Geometry
nT2 spectra for three samples during laboratory centrifuging to drive water from the pores. As centrifuging continues for a total duration of three days, water is gradually expelled from large pores, leaving some water in the small ones.
Porosity volume
5.9 p.u. 0.2 md
Core Sample 2 7.1 p.u. 1.1 md 10
1
100
Pore diameter, microns
nDecaying signal amplitude from a downhole nuclear magnetic resonance measurement (top) is transformed to a T2 relaxation spectrum (middle), and then interpreted in terms of rock pore geometry (bottom). (Adapted from Morriss et al.,
Core Sample 3 12.2 p.u. 7.5 md
0 hours
reference 23.)
times—resulting from each pulse promises to give an indication of the range of pore sizes in reservoir rock. In sandstones, comparisons between T2 relaxation times and mercury porosimetry, a standard lab technique for evaluating pore sizes—pore neck sizes to be precise—are generally good (above ).23 This indicates that in sandstones, there is a predictable relationship between pore and pore neck sizes. Researchers are conducting similar measurements on carbonates, but results so far have not shown the same predictable relationship. The CMR measurement is made from a pad-type tool with permanent magnets that provide an ambient field focused entirely into
January 1995
1.5 hours 6 hours 3 days
1
10
100
1000
10,000
T2, msec
21. “Archie II: Electrical Conduction in HydrocarbonBearing Rock,” The Technical Review 36, no. 4 (October 1988): 12-21. 22. Bouvier L: “Les Saturations en Réservoir Carbonaté à Double Porosité: Réconciliation des Mesures Laboratoires et Diagraphies,” Pétrole et Techniques 375 (October 1992): 21-24.
23. Morriss CE, MacInnis J, Freedman R, Smaardyk J, Straley C, Kenyon WE, Vinegar HJ and Tutunjian PN: “Field Test of an Experimental Pulsed Nuclear Magnetism Tool,” Transactions of the SPWLA 34th Annual Logging Symposium, June 13-16, 1993, Calgary, Alberta, Canada, paper GGG. 24. Chang D, Vinegar H, Morriss C and Straley C: “Effective Porosity, Producible Fluid and Permeability in Carbonates from NMR Logging,” Transactions of the SPWLA 35th Annual Logging Symposium , Tulsa, Oklahoma, USA, June 19-22, 1994, paper A.
51
25. Kenyon WE, Takezaki H, Straley C, Sen PN, Herron M, Matteson A, Petricola MJ: “A Laboratory Study of Nuclear Magnetic Resonance Relaxation and its Relation to Depositional Texture and Petrophysical Properties—Carbonate Thamama Group, Mubarraz Field, Abu Dhabi,” paper SPE 29886, presented at the SPE Middle East Technical Conference and Exhibition, Bahrain, March 11-14, 1995. 26. Winkler KW, Liu H-L and Johnson DL: “Permeability and Borehole Stoneley Waves: Comparison Between Experiment and Theory,” Geophysics 54 (January 1989): 66-75. 27. Petricola M and Frignet B: “A Synergetic Approach to Fracture and Permeability Evaluation from Logs,” paper SPE 24529, presented at the 5th Abu Dhabi Petroleum Conference and Exhibition, Abu Dhabi, UAE, May 18-20, 1992. 28. Cassell B, Badri M and Faulhaber J: “Permeability Prediction Based on Anelastic Attenuation Using Dipole Shear and Low Frequency Monopole Sources in a Carbonate Reservoir in Saudi Arabia,” presented at the GEO-94 Middle East Geosciences Exhibition & Conference, Bahrain, April 25-27, 1994.
0.18 mm
nMoldic porosity (purple). Formerly a fos-
siliferous micrite, the micrite has been replaced by dolomite and the more soluble calcitic fossils dissolved to leave a moldic porosity. Upper Eocene Ocala Limestone, Florida. Gypsum plate added.
up and down the borehole. The Stoneley wave preserves most of its energy in the borehole, but in permeable formations some energy is attenuated when wave pressure pushes fluid from the borehole into the formation, similar to a quick, small-scale well test. This slows the velocity of the wave by an amount that can be related to the ratio of formation permeability to fluid
kNMR, md 10-1
1
10
viscosity. Given a viscosity for the borehole fluid, in well-controlled circumstances such as laboratory measurements or boreholes with no mudcake, the permeability can then be estimated.26 The DSI tool generates Stoneley waves with a special monopole transmitter at frequencies of 600 Hz to 5 kHz, ideal for tube-wave logging and a quantum leap ahead of previous technology equipped with transmitters operating in the 10 to 20 kHz range. Recent estimation of permeability using Stoneley-wave velocity as obtained from the DSI tool shows impressive agreement with core permeability measurements in an Abu Dhabi carbonate reservoir (next page, top right ).27 The method of obtaining permeability using Stoneley-wave velocity requires knowing the formation’s density and shear velocity. A second method establishes permeability from the Stoneley wave without other data. This method is based on observing how permeability attenuates Stoneleywave energy by directly comparing signals from near and far receivers.28 Attenuation is
Porosity, p.u. 102
XX50
20
10
0 6.6 p.u. 0.07md
Core Depth, ft
from other logs. Following visual analysis of the cores, a second cutoff at 750 msec was selected to isolate vugs from intergranular porosity. This was also applied to spectra measured downhole, providing a log of vuggy porosity (below ). Recent laboratory work on core samples from the carbonate Mubarraz field in Abu Dhabi, UAE, confirms the potential of NMR measurements.25 A challenge in this area is to distinguish small micropores in the micrite matrix from the much larger productive intergranular pores. Analyzing 20 samples from two wells, a team of Schlumberger and Abu Dhabi Oil Company geoscientists found that micropores were correctly identified using a relaxation time cutoff of 190 msec on laboratory-measured T2 spectra. Furthermore, permeable grainstone facies could be distinguished from lower-permeability packstones and mudstones with a cutoff of 225 msec. Finally, the NMR data could be interpreted to give more accurate permeability estimates than those obtained from conventional porosity logs. The CMR logging tool is currently being tested in Abu Dhabi, and expectations are high that similarly impressive results will be obtained in boreholes. Another logging tool, the DSI imager, gains direct entry to permeability by physically moving fluid through the formation. This is achieved when low-frequency tube waves—called Stoneley waves—propagate
Vug porosity
11.9 p.u. 13 md
X100
Vug
8.7 p.u. 0.7md
X150
10.9 p.u. 0.2 md
1 X200
100
10,000
T2, msec
nT2 spectra for vuggy carbonates showing dual-porosity system, and a log of vuggy porosity obtained from evaluating the right end of the NMR spectra only, above a cutoff of 750 msec. 52
Oilfield Review
greater at higher frequencies, so the comparison is more sensitive if measured at the high end of the Stoneley-wave frequency spectrum. Excellent agreement has been observed in Middle East carbonate reservoirs between permeability estimates obtained using this second method and production logs and core data (bottom ). Research continues into improving Stoneley-wave permeability, for example in
XX900
Permeability index from Stoneley velocity
Permeability index from Stoneley attenuation
nPermeability evaluated from Stoneley-wave velocity and attenuation compared with core measurements in a Middle East carbonate reservoir. Evaluation using velocity requires formation density and shearvelocity measurements.
X1000
Depth, ft
X1100
Core permeability
X1200
X1300 0.26 mm
nFilled fenestral porosity in a blue-green algal biolithite. Porosity may be due to air spaces in crinkled mat sediments. Upper Silurian Limestone, Pennsylvania, USA. Fluid Analysis
100
50
Permeability index from Stoneley velocity 0 Low
XX50
X100
Residual oil Moved oil
X1400
Permeability
Permeability index from Stoneley attenuation
Cumulative Stoneley Perm. Spinner High 0
10
20 Low
High
nPermeability estimated from Stoneley-wave attenuation without recourse to other log data, compared with permeability from Stoneley-wave velocity, in a Middle East carbonate reservoir. Integrated permeability shows an excellent match with a flowmeter production log.
Depth, ft
X150
X200
X250
Water X300
Caliper X350
January 1995
53
Large-Scale Features
Mapping reservoirs at the large scale and understanding their complex petrophysics at the small scale are all part of the challenge facing reservoir geologists and engineers. But in carbonates, additional care must be taken to recognize and evaluate two types of medium- to large-scale features that are caused by overburden and tectonic stresses. Either can dramatically affect reservoir performance, creating heterogeneous or anisotropic behavior where none might otherwise be suspected. These two features are stylolites and fractures.29 Stylolites occur in any sedimentary formation, but are particularly common in carbonates—picture the thin, sawtooth “veins” visible on polished marble tiles and floors. Stylolites are easily recognized on outcrops and cores as irregular planes of discontinuity between rock units. Formed during compaction, probably through the mechanism of pressure solution, stylolites concentrate fine-grained insoluble residue along their irregular seams. They are usually assumed to act as permeability barriers, but some core measurement results confirm that stylolites can develop permeability. Identifying them and evaluating their imact on permeability are therefore top priorities for the reservoir engineer. Borehole imaging has greatly facilitated the identification of stylolites downhole (right and next page). Viewed with the FMI
Depth, ft
accounting for the presence of mudcake, which almost certainly interferes with the tube wave’s ability to move formation fluids.
XX32
XX33
XX34
XX35
nThree types of stylolites identifiable on FMI images. Dark colored stylolites (left), probably filled with clay; stylolites associated with a light band (below), are probably resistive calcite. Some stylolites are associated with extensional fractures (next page, left).
Depth, ft
XX80
0.29 mm
nFilled shelter porosity beneath a large
XX81
mollusk fragment. Pliocene and Pleistocene Marl, Florida.
XX82
54
Oilfield Review
tool, they appear in three common varieties. First, some stylolites exhibit undulating but slightly irregular surfaces and are filled with dark, therefore conductive material, probably clay. A second group of stylolites seems to have an associated band of light color, most likely resistive calcite. The third type of stylolite clearly shows associated extensional fractures caused by excessive overburden stress. The question remains: Which stylolites form permeability barriers and which do not? Until recently, there has been no sure way of deciding. Now, answers are obtainable from a third-generation wireline testing tool, the MDT Modular Formation Dynamics Tester tool. Unlike previous wireline testers, this tool permits testing between probes set as far apart as 8 ft [2.4 m], a large
Depth, ft
XX88
XX89
XX90
0.22 mm
nReduced fracture porosity (black). Upper Jurassic Limestone, Germany. Cross-polarized light photograph. enough interval to comfortably straddle a stylolite. In such tests recently performed in the Middle East, MDT measurements indicated that stylolites previously assumed to be completely impermeable may in fact be
partly conductive to fluid flow. If stylolites generally impede flow, fractures almost always enhance it. Indeed, some reservoirs, particularly carbonate ones, rely exclusively on fractures to achieve commercial levels of production. Before the advent of wireline imaging techniques, detecting fractures was difficult and characterizing anything about them was almost impossible. That bleak outlook changed dramatically with the introduction of the FMI and DSI tools. The recently introduced ARI Azimuthal Resistivity Imager tool also makes an important contribution in fracture detection. Briefly, all three tools contribute to fracture interpretation, but each alone may not provide a complete picture. 30 On FMI images, open fractures filled with invading water-base mud of high conductivity are recognizable as dark and usually fragmented sinusoid traces. With the help of interactive FracView image processing, the interpreter can reliably pinpoint fractures, calculate their dip and azimuth, and estimate spatial density at the borehole. Additional analysis of image resistivity near the fracture can also lead to an estimation of fracture aperture.31 With simple models of fracture geometry, the combined log information may provide an effective fracture permeability. This can then be integrated with permeability estimates for the unfractured part of the rock to yield a permeability for the whole rock. In 29. For general reference: Nelson RA: “Analysis of Anisotropic Reservoirs,” in Geologic Analysis of Naturally Fractured Reservoirs. Houston, Texas, USA: Gulf Publishing Company, 1985. 30. Cheung PS-Y and Heliot D: “Workstation-Based Fracture Evaluation Using Borehole Images,” paper SPE 20573, presented at the 65th SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, September 23-26, 1990. 31. Luthi SM and Souhaité P: “Fracture Apertures from Electrical Borehole Scans,” Geophysics 55 (July 1990): 821-833.
XX91
January 1995
55
0.001
Mean aperture mm
10.00
Depth, ft
XX00
XX10
nFracture evaluation and estimation of fracture aperture from an FMI image in a Rocky Mountain carbonate reservoir. This information was used to derive an effective permeability that matched well test data. Squares and triangles signify a level of confidence in fracture identification, with squares having a higher level than triangles.
0.26 mm
nFilled fracture porosity, with conjugate set of large fracture veins. Middle Ordovician Limestone, Pennsylvania.
56
0.22 mm
nEnlarged moldic porosity (black). Note large mollusk mold with upper edge enlarged. Other grains include mainly miliolid foraminifera. Upper Oligocene Limestone, Florida. Cross-polarized light photograph.
the Rocky Mountains, where a low-porosity carbonate reservoir depends on fractures for production, such a combined permeability has been successfully compared to permeability obtained from drill-stem tests (left ). There are a few caveats, however, to fracture interpretation using FMI resistivity images. First, the calculated fracture aperture seems to be influenced by the fluid originally filling the fracture—fractures in water zones appear systematically wider than nearby fractures in hydrocarbon zones. It is suspected that invasion fails to remove all hydrocarbon from the walls of the fracture, thereby making the fracture look thinner to electrical imaging techniques.32 In a depleted carbonate field being exploited for additional oil using horizontal wells, this phenomenon has been put to good use in identifying fractures that are likely to allow water breakthrough (next page ). Second, the FMI tool is a relatively shallow measurement, and this limits the tool’s ability to distinguish natural fractures that contribute to reservoir performance from drilling-induced fractures that do not. Certain types of drilling-induced fractures are easily recognized by their geometry—for example, vertical fractures oriented perpendicular to the least horizontal stress and therefore intersecting a vertical borehole over a lengthy interval. Nonvertical drillinginduced fractures, however, are harder to distinguish and may be easily confused with the natural variety. Fracture identification in highly deviated and horizontal wells becomes harder still. The ARI tool provides some added depth of investigation but a poorer along-borehole resolution, and as a result, fewer fractures are detected. However, ARI image processing provides some clue to fracture depth as well as aperture, although neither is unambiguously determined.33 The two parameters are genetically linked, so the tool response to a fracture enables an estimate of one of the parameters once a value for the other is taken. Greater depth of investigation, up to several meters, is provided by the DSI tool that detects open fractures in the same way that it senses a permeable formation—by employing the Stoneley wave to physically pulse mud into and out of the fracture.34 However, there is a commensurate deterioration in resolution along the borehole axis, to about 1.5 m [4.9 ft], showing closely spaced fractures as a single fracture. Like the FMI measurement, the Stoneley wave permits the evaluation of fracture aperture, though again this may actually represent the
Oilfield Review
Mean aperture, mm
400
m
500
nFracture apertures determined from FMI images in two subhorizontal wells in a fractured and partly depleted reservoir. The apertures in the top example show four zones where aperture is higher than elsewhere. This was interpreted to indicate the fractures are water-filled and liable to water out early. After completion, the top well produced with 50% water cut; the bottom well produced no water.
600
10
Potential water entry
.001
Oil/water contact
Mean aperture, mm
10
.001
m
300
400
500
Oil/water contact
cumulative apertures of several neighboring fractures. Comparisons between fracture aperture estimated from the two techniques have shown good agreement in metamorphic volcanics at a UK waste disposal site. On the down side, the Stoneley wave yields no information on fracture dip or azimuth. There may be further to go in fracture interpretation, but a comparison between the techniques of ten years ago and those of today reveals the extraordinary advances achieved by novel wireline technology and a spirited community of interpreters. It is a level of improvement common to all areas of carbonate interpretation. Today’s exploitation in increasingly complex and difficult reservoirs has gained from a veritable revolution in formation evaluation. Long may it continue. —HE, LS
January 1995
32. Standen E, Nurmi R, ElWazeer F and Ozkanli M: “Quantitative Applications of Wellbore Images to Reservoir Analysis,” Transactions of the SPWLA 34th Annual Logging Symposium, Calgary, Alberta, Canada, June 13-16, 1993, paper EEE. 33. Faivre O: “Fracture Evaluation from Quantitative Azimuthal Resistivities,” paper SPE 26434, presented at the 68th SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, October 3-6, 1993. 34. Hornby BE, Johnson DL, Winkler KW and Plumb RA: “Fracture Evaluation Using Reflected StoneleyWave Arrivals,” Geophysics 54 (October 1989): 1274-1288.
57
