Pressure Gradients in Formation - Schlumberger

The Pressures of Drilling and Production

Yves Barriol Karen Sullivan Glaser Julian Pop Sugar Land, Texas, USA

Pressure measurement is essential to optimized hydrocarbon recovery. Now,

Bob Bartman Devon Energy Houston, Texas

production, accurate, cost-effective pressure data can be acquired to help reduce

accurate formation pressures can be determined at almost any time in a well’s life cycle. Whether while drilling, when the well is at total depth, or years after initial

risk and improve recovery.

Ramona Corbiell New Orleans, Louisiana, USA Kåre Otto Eriksen Harald Laastad Statoil Stavanger, Norway James Laidlaw Aberdeen, Scotland Yves Manin Clamart, France Kerr Morrison BP Exploration and Production Aberdeen Colin M. Sayers Houston, Texas Martin Terrazas Romero Petróleos Mexicanos (PEMEX) Poza Rica, Mexico

Many everyday pressure effects go unnoticed. We seldom consider why water flows from a tap or how an airplane flies. And certainly, while fueling the family automobile, the nature of geopressures that drive hydrocarbons to the surface is far from our minds. Our world relies on pressure in many ways, not the least of which is the production of oil and gas. The story of geopressure is rooted in the Earth’s beginning. As the molten outer core of the Earth cooled, plate-tectonic movements driven by convection within the Earth generated stresses in the crust. Movement, warping and buckling of these stressed crustal plates caused

mountains and basins to form. Volcanic eruptions associated with plate-tectonic forces spewed material from within the Earth, forming the atmosphere and oceans. As plate-tectonic activity continued to influence subsurface pressure conditions, weather patterns formed. Cycles of oceanic evaporation, atmospheric saturation, condensation and inland precipitation fueled rivers that run to the oceans, carrying with them large amounts of eroded rock and terrigenous and marine organic material. As the transport velocity slowed, these materials settled in depositional basins (below). Later, during continued burial and compaction, these

Condensation

Evaporation

Precipitation

Yakov Volokitin Shell E & P Americas New Orleans, Louisiana

Surface runoff

Ocean

Lake Flow

For help in preparation of this article, thanks to Jeff Cordera and Aaron Jacobson, Clamart, France; Roger Goobie, Houston, Texas; Jose de Jesus Gutierrez, Mexico City, Mexico; Martin Isaacs, Frederik Majkut and Lorne Simmons, Sugar Land, Texas; and Paula Turner, independent consultant, Houston. adnVISION675, AIT (Array Induction Imager Tool), arcVISION675, CHDT (Cased Hole Dynamics Tester), CQG (Crystal Quartz Gauge), FloWatcher, MDT (Modular Formation Dynamics Tester), Platform Express, PowerDrive Xtra, PressureXpress, proVISION675, Sapphire, Smart Pretest, StethoScope, TeleScope, USI (UltraSonic Imager), VISION and WellWatcher are marks of Schlumberger.

22

Water table

Particle transport Sedimentation

Bedding

> The hydrologic cycle. Evaporation of water from the ocean forms clouds. These clouds drift over the land and produce rain that flows along rivers back to the ocean, carrying with it rock and organic debris that accumulates in basins. The cycle repeats, depositing massive layers of material.

Oilfield Review

Autumn 2005

23

materials were transformed by heat, pressure and organic activity into the many hydrocarbon compounds we know as petroleum. Thus begins the story of geopressure, hydrocarbons and production. In this article, we first review the development of geopressured systems, and then discuss the effects of formation pressure on drilling, evaluation, production and recovery of hydrocarbons. Case histories from the Gulf of Mexico, Mexico and the North Sea show how drillers, engineers and geoscientists are using advanced techniques to

predict, measure and manage pressure, allowing wells to be drilled more safely, boreholes to be placed more accurately, and reservoir contents to be evaluated and managed for maximum oil and gas recovery. Development of Geopressured Systems The Earth’s outer crust hosts a complex system of stresses and geopressures that constantly seek equilibrium. Although the subsurface comprises many geologic features in different pressure and stress regimes, one of the most commonly

Congo fan free-air gravity

Seep clusters Bathymetry contours

> Seepage identified offshore Angola, West Africa. About 75% of the world’s petroliferous basins contain surface seeps. Knowing where oil and gas seeps are emerging is helpful in locating the sources of subsurface oil and gas accumulations. Scientists use satellite imagery to help identify potential hydrocarbon reservoirs. In this image, free-air gravity values derived from European Remote-Sensing Satellite (ERS) data identify areas of high gravity resulting from sediment emitted from the Congo River, known as the Congo Fan. The data are also used to help identify areas of hydrocarbon seepage shown as red-outlined contours. The subsea seep source is typically located using sonar or shallow seismic reflection. The hydrocarbons can then be sampled, helping to identify oil type and field maturity and to correlate with other subsea seeps. (Image courtesy of NPA Group; Archive block outlines courtesy of IHS Energy.)

24

studied subsurface pressure distributions occurs in relatively shallow sediments, laid down in deltaic depositional environments. Rivers wash large amounts of sand, silt and clay into offshore basins where they accumulate, are lithified over millions of years and form primarily sandstone, siltstone and shale. Initially, the sediments deposited at the mouths of rivers are unconsolidated and uncompacted, with relatively high porosity and permeability that allow remnant seawater, or connate water, in the pores to remain in full hydraulic communication with the ocean above. With time and compaction as more sediment is deposited, water is squeezed out of pore spaces and grain-to-grain contact supports more and more of the depositional load. Provided there is a conduit for the water to escape, pressure equilibrium is maintained in the pore spaces. Once formed, oil and gas migrate upward to zones of lower pressure, possibly reaching the surface to form seeps if there is no mechanical blockage along the way. Geological and archeological evidence shows that hydrocarbon seeps have occurred naturally in various parts of the world for thousands of years. In some cases, subsurface pressures drive large volumes of hydrocarbons to the surface. Along the California coast near Goleta Point, USA, commercial volumes of natural gas continue to escape from natural fractures in the Earth’s crust. Here engineers designed a unique underwater gasrecovery system, capturing more than 4,000 million ft3 [113 million m3] of natural gas since 1982. This is enough natural gas to supply the annual needs of more than 25,000 typical California residential consumers.1 Seeps generally occur where erosion exposes hydrocarbon-bearing rocks at the Earth’s surface, or where a fault or fracture allows pressure-driven hydrocarbons to migrate to the surface. Historical records indicate that surface seepage led to the discovery of many petroleum deposits.2 Today, aerial and satellite imagery helps geologists detect natural oil and gas seeps migrating from great ocean depths, offering the promise of yet undiscovered hydrocarbon reserves (left). Fortunately, most subsurface hydrocarbons do not leak to the surface. As oil and gas migrate upward, they typically become trapped beneath low-permeability layers, or seals. These seals may consist of many rock types, including shales, calcareous shales, well-cemented sandstone, lithified volcanic ash, anhydrite and salt. Hydrocarbon traps are frequently grouped according to the geologic processes that cause them, such as folding, faulting and structural

Oilfield Review

Gas Oil Salt water

> Structural traps. The weight of overlying sediments causes salt layers to plastically deform, creating diapirs. As diapirs evolve, sediments onlap their margins, forming traps that typically accommodate hydrocarbons (left). Where the strata have deformed to form an anticline (center), oil (green) and gas (red) may become trapped under a seal. Faulting may also trap hydrocarbons (right) by sealing the updip margin of a reservoir.

changes caused by plate-tectonic activity or the plastic deformation of salts or shales (above). Many hydrocarbon traps involve combinations of structural and stratigraphic features, but once trapped beneath a seal, reservoir fluids have no hydraulic communication with the surface. Given time and the right circumstances, pressure increases in the rock pore space (see “Causes of Abnormal Pressure,” page 26). Early Oil and the Uncertainty of Pressure Sometime before 200 BCE, the Chinese relied on geopressure to help produce the first gas wells.3 Other records indicate that as early as 1594, near Baku, Azerbaijan, shallow pits or wells were dug by hand to depths of 35 m [115 ft], making this area the first actual oil field.4 In the USA, the history of drilling prior to the 19th Century is unclear, although seep-oil use is noted in numerous early historical accounts. In 1821, drillers completed the first well in the USA designed specifically to produce natural gas. This well, in Fredonia, New York, USA, reached a depth of 27 ft [8.2 m] and produced enough gas, driven by natural pressure, to light dozens of burners at a nearby inn. Then, Edwin L. Drake drilled an exploratory well in 1859 near Titusville, Pennsylvania, USA, to locate the source of an oil seep.5 At a depth of 69.5 ft [21 m], drillers pulled their tools from the well. Within

24 hours, geopressure effects forced oil up the borehole. Fortunately for Drake, oil seeps in the area precluded abnormal pressure buildup. Using a manual pump, drillers produced about 25 bbl/d [3.9 m3/d] of oil. Although production soon dropped to around 10 bbl/d [1.5 m3/d], the well is said to have continued producing for a year or more.6 By the early 1900s, drillers, geoscientists and engineers recognized the importance of geopressure in oil and gas production. The Spindletop discovery, which blew out during drilling near a salt dome at 1,020 ft [311 m], produced around 800,000 bbl [127,120 m3] of oil in eight days and provided scientists with insights into salt dome-related abnormal geopressure effects.7 As drilling activity increased, exploration reached into new and uncharted territories. Remembering the uncontrolled oil gushers of the past, drillers were constantly on the lookout for abnormal geopressures. Engineers and scientists began to seek new ways to predict abnormal pressures during the search for oil. At around the same time that Drake drilled his first well, seismic instruments were being developed and used to record and measure Earth movements during earthquakes. Researchers developed the technologies that form the basis of reflection seismology. In reflection seismology,

subsurface formations are mapped by measuring the time it takes for acoustic pulses transmitted into the Earth to return to the surface after being reflected by geological formations with varying physical properties.8 Over time, seismic technology moved into the oil field, providing geophysicists, geologists and drilling engineers with the tools to evaluate reservoirs and pressure regimes long before drilling a well. Although early geopressure estimations from seismic image analysis were crude, drillers needed predrilling pressure estimates for mudweight selection, casing design and well-cost estimation, among other uses. Engineers found early pressure estimates too uncertain, especially in complex oil and gas reservoirs. To more easily understand and visualize the geopressure environment, geoscientists now use sophisticated seismic data acquisition and processing techniques, mechanical earth models and pore-pressure cubes to study, evaluate and visualize pressure environments within a given basin or area. Engineers use reflection tomography that yields higher spatial resolution than conventional seismic techniques to accurately predict pore pressure from seismic data. This high resolution also helps differentiate variations in pore pressure from variations in lithology and fluid content.9 (continued on page 28)

1. Natural Oil and Gas Seepage in the Coastal Areas of California; U.S. Department of the Interior, Minerals Management Service. http://www.mms.gov/omm/pacific/ enviro/seeps1.htm (accessed October 8, 2005). 2. For more on seeps and oil exploration: http://www.npagroup.co.uk/oilandmineral/offshore/oil_ exploration/ (accessed October 8, 2005). 3. Brufatto C, Cochran J, Conn L, Power D, El-Zeghaty SZAA, Fraboulet B, Griffin T, James S, Munk T, Justus F, Levine JR, Montgomery C, Murphy D, Pfeiffer J, Pornpoch T and Rishmani L: “From Mud to Cement—

Autumn 2005

Building Gas Wells,” Oilfield Review 15, no. 3 (Autumn 2003): 62–76. 4. For more on the chronology of oil events: http://www. sjgs.com/history.html#ancient_to_present (accessed October 8, 2005). 5. For a chronology of oil- and gas-well drilling in Pennsylvania: http://www.dep.state.pa.us/dep/ deputate/minres/reclaiMPa/interestingfacts/ chronlogyofoilandgas (accessed October 8, 2005). 6. Yergin D: The Prize. New York City: Simon & Schuster, 1992.

7. For more on the history of the Spindletop oil field: http://www.tsha.utexas.edu/handbook/online/articles/SS/ dos3.html (accessed October 8, 2005). 8. For more on the evolution of seismic technology: http://www.spe.org/spe/jsp/basic/0,1104_1714_ 1004089,00.html (accessed October 8, 2005). 9. Sayers CM, Woodward MJ and Bartman RC: “Seismic Pore-Pressure Prediction Using Reflection Tomography and 4-C Seismic Data,” The Leading Edge 21, no. 2 (February 2002): 188–192.

25

Causes of Abnormal Pressure

Normally pressured formations generally have pore pressure that equals the hydrostatic pressure of the pore water. In sedimentary basins, pore water usually has a density of 8.95 lbm/galUS [1,073 kg/m3], which establishes a normal pressure gradient of 0.465 psi/ft [10.5 kPa/m]. Significant deviation from this normal hydrostatic pressure is referred to as abnormal pressure. Abnormal geopressures, above or below the normal gradient, are found in many hydrocarbon-producing reservoirs. While the origin of these pressures is not understood completely, abnormal pressure development is usually attributed to the effects of compaction, diagenetic activity, differential density and fluid migration.1 Abnormal pressure involves both physical and chemical actions within the Earth. Pressures above or below the normal gradient may be detrimental to the drilling process. Subnormal pressures, or those below the normal gradient, can cause lost-circulation problems in wells drilled with liquid drilling mud. Subnormal pressure conditions frequently occur when the surface elevation of a well is much higher than the subsurface water table or sea level. This is seen when drilling in hilly or mountainous locations, but it may also occur in arid regions where the water table may be more than 1,000 ft [305 m] deep. Abnormally low pressures are also frequently found in depleted reservoirs. These are reservoirs whose original pressure has been reduced by production or leakage. Depletion is not unusual in mature reservoirs from which significant volumes of oil and gas have been produced without waterflooding or pressure maintenance. By contrast, abnormally high pressures are typical in most oil-producing regions. Abnormal overpressures always involve a particular zone becoming sealed or isolated. The amount of overpressure depends on the structure, the depositional environment and the processes and rate of deposition.

26

Well 1

Well 2

Well 3

Well 4

Well 5

Pressured sand

Pressured sand

Pressured sand

> Pressure isolation by fault displacement. In zones of faulting, pressure-bearing zones (brown) can be displaced along a fault plane. If adequately sealed, the displaced zone maintains its abnormal pressure. Although the top of an abnormally pressured zone may be defined in a given area or structure, faulting can cause significant changes in formation depth only a short distance away. For the driller, this not only is confusing but also causes an increased drilling risk.

1. Bourgoyne AT, Millheim KK, Chenevert ME and Young FS: Applied Drilling Engineering, 1st ed. Richardson, Texas: Society of Petroleum Engineers, 1986.

2. For more on faulting: Cerveny K, Davies R, Dudley G, Fox R, Kaufman P, Knipe R and Krantz B: “Reducing Uncertainty with Fault-Seal Analysis,” Oilfield Review 16, no. 4 (Winter 2004): 38–51.

Oilfield Review

One of the most common mechanisms generating abnormally high pressures is trapping of pore water during deposition. If a seal forms before pore water is displaced, grain-to-grain contact between solids is not established. Over time and with increases in compaction due to overburden pressure, the water in the pore space becomes compressed, causing abnormally high pore pressure. Another cause of abnormally high geopressure is geologic uplifting and displacement of a formation, which physically relocates a higher pressured formation from one depth to another (previous page). When a previously normal pressure zone at great depth is displaced by tectonic activity to a shallower depth with the seals remaining intact, the resulting pressure will be abnormally high. Undercompaction during deposition is another mechanism for generating high pore pressure. In the Gulf of Mexico and other depositional basins, compaction disequilibrium is believed to be the most important cause of overpressure. For sediment to compact, pore water must be expelled. However, if sedimentation is rapid compared to the time required for fluid to be expelled from the pore space, or if seals that prevent dewatering and compaction form during burial, the pore fluid becomes overpressured and supports part of the overburden load. Artesian systems are a unique source of abnormally high pressures. In these systems, the surface elevation of the well is below sea level or the water table, a condition that might exist if drilling in a valley between mountains (above right). The same principle also applies to structural situations in which highly dipping permeable formations allow pressure transmission from a higher-pressured deep zone to a shallower depth. Abnormal pressures caused by structural effects are common adjacent to salt domes, where the rising, migrating salt has uplifted the surrounding formations, tilting them and sealing the permeable formations.

Autumn 2005

Rain

Rain Well elevation below water table Artesian well

Ground level Permeable sand

Seal

> Artesian pressure system. In these systems, the surface elevation of the well is below sea level or below the water table. This commonly occurs when drilling in a valley or basin surrounded by hills or mountains—locations where a connected water table is charged by water from higher locations.

Overpressures can also occur in shallow sands if higher-pressured fluids migrate from lower formations as the result of faulting or through a seal in a network of microfractures (below right).2 In addition, man-made actions can charge upper sands. Poorly cemented casings, lost circulation, hydraulic fracturing and underground blowouts can cause otherwise normally pressured zones to become abnormally pressured. Another cause of overpressure is chemical activity. If massive deposition of organic material becomes sealed with time and exposed to higher temperatures, this organic matter generates methane and other hydrocarbons that charge the formation. Increasing depth, temperature and pressure may cause gypsum to convert to anhydrite, releasing water that may charge a formation. Conversely, anhydrite that is exposed to water may form gypsum, resulting in as much as a 40% increase in volume, thereby increasing zonal pressures. Pore pressure may also be increased by the transformation of smectite to illite at increasing temperature and depth. As water is expelled from the clay crystal lattice, pore pressure increases.

Charged sand

Zone of higher pressure

> Fracture migration. Fault planes may allow pressure transmission from a zone of higher pressure to a more shallow, lower-pressured zone. This results in an abnormally pressured, or charged, sand. These effects are common in tectonically stressed environments and adjacent to salt domes.

27

Pore Pressure from Stacking Velocities

0.5

13.0

1.0

12.5

1.5

12.0

2.0

11.5

2.5

11.0

3.0

10.5 10.0

3.5 16

lbm/galUS

Depth, km

14.0 13.5

9.5 14

9.0 12 10 y, k 8 m 6

6

8

12 10 km , x

14

16

Pore Pressure from Tomographic Velocities

0.5

15

1.0

14

1.5

13

2.0

12

2.5

11

3.0

10

3.5 16

lbm/galUS

Depth, km

16

9 14

12 10 y, k 8 m 6

6

8

12 10 km x,

14

16

8

> Seismic tomography. In previous methods, interpreters stacked seismic velocities to improve resolution; from this they generated a pore-pressure cube representing pore pressures across a given area (top). Now, tomographic techniques dramatically improve pore-pressure resolution, reducing uncertainty and increasing accuracy in well planning (bottom).

Reflection tomography offers significant advantages compared with conventional seismic data. Conventional seismic data processing smoothes out velocity fluctuations, and velocity interval picks are often too coarse for accurate pore-pressure prediction. Reflection tomography replaces the low-resolution, conventional velocity analysis with a completely general raytrace modeling-based approach. Although an interpretable image might be obtained using a relatively poor but smooth conventional seismicvelocity field, the resolution is sometimes too low

28

to accurately predict pore pressure for wellplanning purposes. By contrast, the tomographically refined velocity model leads to better understanding of the magnitude and spatial distribution of pore pressure, reducing the uncertainty of pore-pressure predictions (above). Reducing Uncertainty In areas where the geology is unknown and where few, if any, wells have been drilled, seismic geopressure prediction may be the only planning tool available to the engineer. However, data from multiple sources, especially drilling, can be used

in conjunction with seismic tomography to refine models, and reduce risk and cost, while improving drilling efficiency. Once wells have been drilled, drillers and planning engineers have access to additional data, including mud records, mud logging information, formation samples, wireline and while-drilling logs, and formation test data. Tools such as the MDT Modular Formation Dynamics Tester sample formation fluids and provide accurate reservoir pressures.10 Pressures in shale sections above a reservoir can be estimated based on offset-well mud weights. Daily drilling reports of problems such as kicks, lost circulation, differential sticking and other drilling problems also may indicate the presence of abnormal pressures. Planning engineers generally use offset-well data cautiously. Using mud weights to estimate formation pressure may be misleading, particularly when the data come from older wells. Most wells are drilled in an overbalanced condition, with mud weights that exceed actual formation pressure by 1 lbm/galUS [120 kg/m3] or more. Drillers frequently increase mud weights to control troublesome, or sloughing, shales. A more detailed evaluation of geopressure can be obtained by combining drilling data with offset-well electric, acoustic and density logs. To predict pore pressures from wireline or whiledrilling logs, analysts often correlate changes in shale porosity with the potential presence of abnormal pressure. This is possible because shales generally compact with increasing depth in a uniform manner. Because of this compaction, the porosity and electrical conductivity decrease at a uniform rate with increasing depth and overburden pressure. However, if a seal is present, higher than normal levels of conductive connate water may be present, increasing conductivity and indicating abnormal pressure (next page, top). Although conductivity is a good indicator, many variables such as connate-water salinity, mineralogy, temperature and drillingmud filtrate may also affect electric log response. Acoustic velocity derived from sonic logs provides another tool for determining pore pressure that is less affected by wellbore conditions. Acoustic tools measure the time it takes for sound to travel a specified distance. As formation characteristics change, so do the velocity and the interval transit time. Shales with porosities approaching zero may transmit sound at velocities on the order of 16,000 ft/s [4.88 km/s] and transit times of 62.5 µs/ft [205 µs/m].11 Shales with higher porosity have more pore space filled with

Oilfield Review

400 600 1,000 2,000

Normal pressure trend line

16

14 13 12 11 10 14

12

10 y, k 8 m

Autumn 2005

6

10

8

6

12

14

x, km

16

9 8

Top of abnormal pressure

> Electric log analysis to reduce the uncertainty of seismic-based pore-pressure predictions. In normally compacted sediment, electrical conductivity will decrease with depth as water is squeezed from pore spaces. A deflection in conductivity from the normal trend (dashed circle, left and middle) may indicate a change in pore-water concentration, hence the potential for abnormal pressure. Using both seismic and electric log data, computer processing refines the data and generates threedimensional predictive models that help engineers and drillers visualize pore-pressure trends (right).

Change in interval transient time, µs/ft 200 300 50 70 100

Conductivity, mS 200 0

400 600 1,000 2,000 Normal pressure trend line

1 2 3

Top of abnormal pressure

4 16

5 Depth, km Depth, 1,000 ft

10. For more on the MDT tool: Crombie A, Halford F, Hashem M, McNeil R, Thomas EC, Melbourne G and Mullins OC: “Innovations in Wireline Fluid Sampling,” Oilfield Review 10, no. 3 (Autumn 1998): 26–41. 11. The unit µs means microsecond, or one millionth of a second. 12. Sayers CM, Hooyman PJ, Smirnov N, Fiume G, Prince A, de Leon Mojarro JC, Romero MT and Gonzales OM: “Pore Pressure Prediction for the Cocuite Field, Veracruz Basin,” paper SPE 77360, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, September 29–October 2, 2002. 13. Drillers often perform downhole seismic measurements to provide data for correlation of surface seismic data to actual downhole conditions. A checkshot measures the seismic traveltime from the surface to a known depth in the well. Compressional, or P-wave, velocity of the formations encountered in a wellbore can be measured directly by lowering a geophone to each formation of interest, sending out a source of energy from the surface of the Earth, and recording the resultant signal. The data are then correlated to prewell surface seismic data by correcting the sonic log and generating a synthetic seismogram to confirm or modify seismic interpretations. Mechanical earth models and pore-pressure predictions can then be updated.

lbm/galUS

Resistivity

15

0.5 1.0 1.5 2.0 2.5 3.0 3.5

Depth

Spontaneous Potential (SP)

Depth, km

15

6 7 8 9

10 11 12 13 14

0.5 1.0 1.5 2.0 2.5 3.0 3.5

14 13 12 11

lbm/galUS

Improving Pore-Pressure Predictions in the Veracruz Basin Inaccuracies in pore-pressure prediction may lead to well-control problems, exposing an operator to undue risk and excessive cost. Drilling problems in the Veracruz basin, offshore Mexico, led the operator, Petróleos Mexicanos (PEMEX), to reevaluate pore-pressure predictions.12 Engineers at PEMEX and Schlumberger found that predicted mud weights in the Cocuite field were higher than required, resulting in loss of circulation and significant cost overruns. To improve drilling efficiency and reduce risk, engineers and geoscientists used previously acquired three-dimensional (3D) surface seismic data along with sonic logs, mud weights, checkshot surveys and pressure tests from offset wells to improve pore-pressure predictions.13

Conductivity, mS 200

Depth

formation water, hydrocarbons, or both. At 30% porosity, the velocity drops to 12,700 ft/s [3.87 km/s], and the interval transit time increases to about 103 µs/ft [338 µs/m]. Normally pressured shales exhibit decreasing interval transit times with depth. However, as increased pore pressure is encountered, the trend will reverse (below right). Density logging tools also help engineers predict geopressures. The tool irradiates a formation with gamma rays that interact with the electrons surrounding the borehole. The intensity of the backscattered gamma rays varies with bulk density. Since the bulk density of abnormally pressured shale is less than the density of normally pressured shale, engineers can combine predictions made by density, electric and acoustic measurements with surface seismic data to better refine models and reservoir pressure profiles.

10 14

12

10 y (k 8 m)

6

6

8

10

12

m) x (k

14

16

9 8

15

> Acoustic logs for pore-pressure prediction. Sound waves slow when encountering rock with higher pore-water concentrations. The top of an abnormally pressured zone can be predicted based on the change in interval transit time (dashed circle, right), then correlated to changes in conductivity (left). Both measurements can then be used to reduce the uncertainty of the seismic pore-pressure cube (center).

29

Before pore pressure can be estimated from seismic velocities, local knowledge of the total vertical stress must be obtained. In the area covered by the Cocuite 3D seismic survey, the only density log available was from Well Cocuite 402, covering a depth range of 643 to 7,690 ft [196 to 2,344 m]. To estimate the overburden stress to the required depth of more than 13,000 ft [3,962 m], density data from the Cocuite 402 density log were combined with other density information from the Veracruz basin into a composite density log. This information was then used to calculate a general overburden stress gradient for the area. Calculated formation velocities were verified by comparing them to sonic logs upscaled to seismic wavelengths and to seismic interval velocities obtained by inverting traveltime depth pairs from checkshots. Although reasonable agreement was found over the intervals for which sonic logs and checkshots were available, variations in the

velocity field from location to location were observed (below). Similar variations were seen for the other wells in the study area. To mitigate these small-scale variations, geoscientists smoothed the velocities laterally before converting the seismic interval velocities to pore pressure. This technique results in 3D models with a high data density that are less uncertain than those acquired with conventional techniques. Using seismic velocities from the Cocuite 3D survey and a velocity-to-pore-pressure transform, engineers optimized drilling operations by adjusting mud weights. Engineers believe that further refinement in this pore-pressure prediction is possible using reflection tomography to enhance lateral resolution of the seismic velocities.14

Adjusting Pore-Pressure Predictions While Drilling The progression from conventional seismic porepressure prediction to reflection-tomographic techniques significantly reduced uncertainty and improved the accuracy of pore-pressure estimates. However, drilling deep into the Earth continued to be fraught with uncertainty. During well construction, drillers strive to balance mud weight and formation pressure, often based solely on indirect measurements or indicators. Real-time drilling parameters are closely monitored for changes in penetration rate, gas shows and the condition of cuttings returning to the surface, along with signals transmitted from measurements-while-drilling (MWD) and logging-while-drilling (LWD) tools. Schlumberger geophysicists developed a technique for updating uncertainties in the predicted velocities and pore pressures while drilling.15 This technique evaluates uncertainties in the predicted pore pressure on the basis of

0 1 2

U S A 4 101 102

15

12

10

Cuatas #1

4.0

3

402 405 403 13 6

Depth, km

3

1 5

1

9

2

10 1

2

3 Velocity, km/s

4

5

3

15

10 y, k

m 5

0 0

5

10 x, km

e f M Gulf o

xic

o

2.0

0

1.8

1 2

1.6

3 M E X I C O

1.4

4 600 Cocuite field

Inli 400 ne num 200 ber

600 400 200

n ine

sl Cros

Pore-pressure gradient, g/cm3

4 15

Depth, km

Depth, km

Velocity, km/s

2.0

6

8

3.5

2.5

5

7

0

3.0

4

1.2

er umb

> Comparison of P-wave interval velocities. Data (top right) obtained by upscaling the sonic log (green curve) and by inverting traveltime-depth pairs from the checkshot (red curve) in Cocuite Well 101 are compared with the seismic interval velocities (blue dots) for all locations surveyed in the Cocuite field study. From this information, engineers generated a seismic interval velocity 3D cube (left) and a pore-pressure gradient cube (bottom right) showing a transition zone around 3 km [9,843 ft]. This cube helped to define lower and upper formation-pressure limits.

30

Oilfield Review

2

500

500

1,000

1,000

1,500

1,500

Depth, m

Depth, m

1

2,000

2,000 Sonic

2,500

2,500

3,000

3,000

1,500

2,000 2,500 Vp, m/s

3,000

1,500

10 15 20 Pore-pressure gradient, lbm/galUS

3,000

10 15 20 Pore-pressure gradient, lbm/galUS

4

3

500

500 Mud weights

1,000

1,500

2,000

1,500

2,000

Sonic 2,500

3,000

3,000

2,000 2,500 Vp, m/s

Porepressure data

Sonic

2,500

1,500

Mud weights

1,000

Depth, m

Depth, m

2,000 2,500 Vp, m/s

3,000

10 15 20 Pore-pressure gradient, lbm/galUS

1,500

2,000 2,500 Vp, m/s

3,000

10 15 20 Pore-pressure gradient, lbm/galUS

> Reducing uncertainty. The degree of uncertainty in a pore-pressure gradient is exemplified by the width and low resolution of the compressional-velocity (Vp) and pore-pressure gradient curves (1). Vp data from sonic logs are added to the model, somewhat reducing pore-pressure uncertainty (2). Adding mud weights from drilling reports (3) and physical pore-pressure measurements (4) refines estimates and dramatically improves pore-pressure resolution.

borehole-seismic, well-logging and pressure measurements acquired while drilling. The technique was evaluated on two wells in the Gulf of Mexico, USA. 14. Sayers et al, reference 12. 15. Malinverno A, Sayers CM, Woodward MJ and Bartman RC: “Integrating Diverse Measurements to Predict Pore Pressures with Uncertainties While Drilling,” paper SPE 90001, presented at the SPE Annual Technical Conference and Exhibition, Houston, September 26–29, 2004. Bryant I, Malinverno A, Prange M, Gonfalini M, Moffat J, Swager D, Theys P and Verga F: “Understanding Uncertainty,” Oilfield Review 14, no. 3 (Autumn 2002): 2–15.

Autumn 2005

The process involved establishing baseline uncertainties in the coefficients of the velocity to pore-pressure relationship from compressionalwave velocity and density. When drilling began, uncertainties were completely defined by baseline values (above). As drilling progressed on the first evaluation well, a checkshot survey provided data for velocity structure calibration, allowing geophysicists to refine baseline projections and decrease the uncertainty in velocity and pore-pressure

predictions. A relatively small decrease in velocity uncertainty resulted, due to the small size of the checkshot dataset, which comprised traveltime measurements acquired at variable intervals of 50 to 200 m [164 to 656 ft]. After initial logging, engineers incorporated sonic log data to further refine the pressure profile. This additional information markedly reduced velocity uncertainty and gave a correspondingly more detailed pore-pressure prediction. The improved pore-pressure

31

> The StethoScope 675 tool. The tool is 31-ft [9.1-m] long; it has a 6.75-in. collar with an 8.25-in. stabilizer or optional 9.25-in. stabilizer. The stabilizer is made up of a four-blade spiral section at the bottom and two straight blades at the top. The packer and probe are mounted in the blade of the stabilizer (black). The stabilizer blade rests, or is pressed, against the formation by gravity or by force applied by the drillable setting piston (not shown), negating the need to orient in holes up to 10.5 in. for the 8.25-in. stabilizer tool. The probe can extend out of the blade by 3⁄4 in., but normally only moves out level with the surface of the blade and is compressed against the formation to make the seal. The probe is then opened to the formation to take a pressure measurement. The retainer ring (Q-shaped piece around black packer) minimizes deformation of the packer during a test, helping to maintain an effective seal (inset).

prediction continued to have a level of uncertainty that could be reduced only by incorporating measured pore-pressure data. In the absence of direct pore-pressure measurements, mud weight was used to represent pore-pressure boundaries. On the second test well, relatively low velocities were inferred from surface seismic data below 1,500 to 2,000 m [4,921 to 6,562 ft], corresponding to predicted overpressure. Geophysicists incorporated sonic log data to reduce uncertainty. Although pore-pressure predictions improved, the inclusion of mud weights and direct pore-pressure measurements calibrated the coefficients in the velocity to porepressure relationship and placed an upper boundary on the predicted pore pressures. Initially, the pore-pressure gradient from 1,500 to 2,000 m was estimated to be above 13 lbm/galUS [1,560 kg/m3], using porepressure predictions based solely on surface seismic data, checkshot values and sonic logs. With the inclusion of MDT pore-pressure measurements, the calibrated pore-pressure prediction constrained the equivalent pore 16. Pop J, Laastad H, Eriksen KO, O’Keefe M, Follini J-M and Dahle T: “Operational Aspects of Formation Pressure Measurements While Drilling,” paper SPE/IADC 92494, presented at the SPE/IADC Drilling Conference, Amsterdam, February 23–25, 2005.

32

pressure to less than 13 lbm/galUS. Uncertainty was reduced, allowing drillers to better control mud weights, define casing points and improve overall drilling efficiency. Measuring Reservoir Pressures After drilling, concerns regarding pressure usually switch to reservoir management and production operations. Understanding pressures in the reservoir ultimately impacts production and payout, and today may even provide guidance to place additional wellbores for optimized production. Operational demands dictate how and when pressure measurements are performed, with many methods and tools being available to measure and monitor reservoir pressures at almost any time during a well’s life cycle. As described above, the understanding of pressure begins with predrill estimates from seismic data and offset wells, and is further refined during drilling. Reservoir and production engineers make additional measurements using logging tools or permanent sensors in the well or at surface.

Some of the many ways reservoir engineers use precise pressure measurements are for fluid identification and typing, defining fluid contacts and assessing reservoir continuity. Obtaining the required measurement precision involves use of services such as the MDT tool, the PressureXpress reservoir pressure while logging service or while-drilling formation-pressure tools. In these services, high quality data are obtained by allowing time for pressure stabilization before the measurement, by allowing sufficient time for the pressure within the tool to equilibrate with the pressure in the formation and by obtaining a large number of pretests to establish fluid gradients. Later, in mature reservoir environments where substantial production has taken place, formation-pressure measurements are used to quantify depletion, to assess pressure support or to further analyze reservoir continuity. Even though the accuracy requirements of pressure measurements may not be as strict in mature reservoirs, the ability to measure pressures over a wide range of formation permeabilities may be critical for increasing hydrocarbon recovery. While-Drilling Pressure Measurement in Norway Although borehole seismic techniques have brought the driller closer to understanding and predicting pore pressures in real time, scientists and engineers continue to develop tools for obtaining direct pressure measurements while drilling. As LWD technology advanced, engineers adapted the CQG Crystal Quartz Gauge and strain-gauge pressure sensor technologies used in other pressure tools, such as the MDT system, to real-time, while-drilling, pressure-sensing tools (see “Quartz Pressure Sensors,” next page]. Engineers at Statoil and Schlumberger field tested the new StethoScope formation pressurewhile-drilling service in 2004 in several fields located offshore Norway.16 The goal of field testing was to establish whether a while-drilling formation-pressure measurement could be of comparable quality to wireline MDT tester measurements given the range of permeablities, well conditions and mud properties encountered in these fields. All formation tester tools measure pore pressure at the interface between the external filtercake and the wellbore wall, or sandface. Whether or not the pressure at the sandface is a good estimate of the true, far-field formation pressure depends not only on the properties of the mud, the filtercake and the formation, but also on the drilling-fluid circulation-rate history.

Oilfield Review

If the filtercake is totally ineffective in sealing between the formation and test probe, then wellbore pressure will be measured; if the filtercake provides a perfect seal, given sufficient time, the tester should measure the true formation pressure. In most drilling situations, filtercakes are neither perfect nor uniform in composition. During the course of normal drilling operations, the filtercake is eroded by mud circulation, scraped away during trips, and then rebuilt at the borehole face. Laboratory experiments with both water-base and oil-base muds indicate that dynamic wellbore conditions influence the mudfiltration rate into the formation and, by implication, the pressure measured at the sandface. A leaking filtercake is frequently a problem and may result in significant differences between measured and true formation pressures. When the difference between the measured sandface pressure and the true formation pressure is significant, the formation is usually said to be supercharged. This situation may occur in both while-drilling and conventional wireline-conveyed methods of measurement, but may be more common in a while-drilling method due to the dynamic environment. To improve confidence in pressure measurements, the StethoScope tool was designed with a pressure-measurement probe embedded in a stabilizer blade surrounded by an elastomeric sealing element, or packer (previous page). The stabilizer design maximizes the flow area at the cross section of the probe, diverts the flow away from the probe-to-formation interface, and minimizes the mud velocity in the vicinity of the probe, thereby helping to reduce filtercake erosion and filtrate leakage into the formation while testing. A drillable setting piston is employed to push the stabilizer containing the probe against the formation face. The tool draws power from a downhole MWD turbine. Additional power is provided by a battery, capable of fully operating the formationpressure-while-drilling tool, for example during tests, when the pumps are off. Sandface pressures are measured by two drilling-qualified pressure gauges: a proprietary, ruggedized CQG pressure sensor and a strain-gauge sensor. A second strain-gauge pressure sensor, located near the probe, measures wellbore pressure continuously. All data acquired during formation tests are stored in tool memory, including pressures, temperatures, actual pretest volumes and drawdown rates, and tool-related state and

Autumn 2005

Quartz Pressure Sensors

Quartz is one of several minerals that display piezoelectric properties. When pressure is applied to a quartz crystal, a positive electrical charge is created at one end of the crystal and a negative charge at the other. Quartz crystals are also strongly pyroelectric; temperature changes cause the development of positive and negative charges within the crystal. A correctly cut quartz crystal has a resonant frequency of vibration, similar to a tuning fork. As the quartz vibrates, there is a detectable sinusoidal variation in electrical charge on its surface. Pressure-induced stress applied to the crystal causes the sinewave frequency to vary in a predictable and precise manner. These properties make quartz valuable in many electronics and sensing applications, including oilfield pressure sensors. Researchers at Schlumberger-Doll Research in Ridgefield, Connecticut, USA, began work on a highly sensitive pressure gauge based on the unique properties of quartz crystals in 1980, and proposed the dual-mode oscillation concept that became fundamental to the development of the CQG Crystal Quartz Gauge sensor (above right).1 The project was transferred to SchlumbergerFlopetrol, Melun, France, in 1982. The development team was supported by researchers at Ecole Nationale Supérieure de Mécanique et des Microtechniques, Besançon, France. Pressure sensors are sensitive to temperature and pressure variations, and must be corrected for temperature fluctuations. The CQG gauge improved on 1. Two mechanical oscillation modes are excited and maintained by the electronics in the CQG resonator plate. One is more sensitive to lateral stresses caused by pressure applied on the sensor; the other is more sensitive to temperature variations. These two resonance frequencies provide simultaneous information on pressure and temperature and allow computation of a temperaturecorrected pressure measurement.

> Sensor for making temperaturecompensated pressure measurements. The crystal blade CQG sensor (gold) is a dualmode resonator with one mode dependent on applied pressure and a second dependent on the applied temperature. The pressure and temperature measurements are taken at exactly the same time.

previous crystal pressure transducers by providing both temperature and pressure measurements from a single sensitive element, eliminating problems associated with thermal lag between separate pressure and temperature sensors. This sensor produces a small peak error induced by transient conditions. Transient errors are further minimized by a dynamic temperature compensation algorithm based on a simple model of the sensor. CQG sensors operate efficiently at pressures ranging from 14.5 to 15,000 psi [0.1 to 103.4 MPa] and in a temperature range of 77 to 300°F [25 to 150°C]. In 1989, the CQG sensor was optimized for commercial fabrication and applied to numerous oilfield pressure-sensing applications, including the MDT tool. More recently, the CQG sensor was ruggedized for LWD and MWD applications, and today it is the primary pressure sensor in both the StethoScope tester and the PressureXpress tool.

33

Pretest 2

Pretest 1 Mobility (mD/cP) A: ≥ 0.1

Rate, cm3/s

Volume, cm3

Buildup time, s

Rate, cm3/s

Volume, cm3

Total time, s

0.2

2

450

0.2

0.5

900

B:

≥1

0.3

5

100

0.3

6.0

300

C:

≥ 10

0.5

10

100

1.0

15.0

300

1.0

10

60

2.0

15.0

180

D: ≥ 100 1,800

Wellbore pressure 1,600 Pretest 1

Pretest 2

Pressure, psi

1,400 1,200 1,000 800 Pumps off Pumps on

600 100

200

300

400 500 Elapsed time, s

600

700

800

> Fixed-mode pretests, A through D, with two drawdown-buildup pairs. Parameters are chosen to cover a wide formation-fluid mobility range (top). The parameters specify the volumes employed and the duration of the buildups for Pretest 1 (the “investigation” pretest) and Pretest 2 (the “measurement” pretest). The graph (bottom) demonstrates the StethoScope tool response when testing a 1.5-mD/cP limestone formation using a pretest sequence similar to that of fixed-mode Type B. During this test, the second buildup was extended, allowing engineers to observe the pressure stabilization time across a longer than normal test sequence. There is a variance in the sandface pressure measurement when performing the measurement with pumps on (red) at a rate of 1,363 L/min [360 galUS/min], and with pumps off (blue).

2,000 First estimate of formation pressure

1 Pressure, psi

3

5

Pressure data

Final estimate of formation pressure

1,000

2 Investigationphase pretest Measurementphase pretest

4

0 0

100

200

300

400

500

Time, s

> Pressure data in real time. In this field-test example, pressure data, presented as open triangles, are displayed in real time at surface during a 5-minute time-limited pretest performed with the mud pumps on, circulating at a rate of approximately 600 galUS/min [2,271 L/min]. The telemetry rate for this test was 6 bits/s. The colored circles represent the principal event markers identified as the data are acquired. The first event marker (1) identifies the well pressure prior to the test; the second marker (2) indicates the beginning of buildup for the investigation phase; the third marker (3) shows the investigation-phase, sandface-pressure estimate; the fourth marker (4) identifies the beginning of buildup for the measurement phase; and the fifth marker (5) represents the sandface pressure as determined during the measurement phase. The formation fluid mobility was determined to be approximately 1.4 mD/cP.

34

operating information. The tools have sufficient memory to store more than 80 five-minute pressure tests. When acquiring formation pressures and fluid mobilities, engineers may choose between two different modes of operating the pretest: an optimized pretest sequence or fixed-mode pretest sequence.17 An optimized, or timelimited, pretest consists of a small initial pretest during which the formation is tested to determine its dynamic properties, followed by one or more, preferably larger-volume, optimized pretests. The optimized pretests are designed downhole by the tool’s logic systems using information obtained from previous tests so that at the end of a prescribed test time, multiple stabilized sandface pressures have been achieved. Only as many tests are performed as will deliver stabilized pressures in the prescribed time; for formations with low mobilities, this may result in a single drawdown. In the field-testing stage, offshore Norway, fixed-mode pretests were employed. Four fixedmode pretest sequences that utilize different preset test parameters are available in the StethoScope tool (left). Each fixed-mode pretest sequence comprises two drawdown and buildup pairs designed to deliver two stabilized sandface pressures within a specified time frame, generally 5 minutes. When consistent, these two independent pressure measurements per test location, or station, together with an estimate of the formation fluid mobility, give confidence in the final pressure result. Comparison of the two pressures obtained in conjunction with the computed mobility can reveal the effects of a static or dynamic pressure environment. An order of magnitude estimate of the formation fluid mobility is helpful in deciding which particular fixed-mode sequence to employ in any given situation, but there is sufficient overlap in their ranges of application that this decision is not critical. Communication to and from the tool is by means of the TeleScope high-speed telemetrywhile-drilling service, specifically designed to provide increased data rate and bandwidth for data delivery. A special telemetry protocol for use with the Telescope telemetry system allows a single device, such as the StethoScope tool, to monopolize data transmission when it has a large amount of data to transmit over a short time interval. The combination of the TeleScope system and on-demand data transmission allows StethoScope data to be visualized at surface in real time (left).

Oilfield Review

Autumn 2005

600 While-drilling tool Test 6: 1,850 to 2,250 L/min

550 500

Start of change in circulation

450 Pressure, bar

400 350 300

While-drilling tool Test 5: 2,262 L/min

While-drilling tool Test 7: No circulation

250 MDT test: 20 cm3 at 20 cm3/min 200 0

100

200

300

400

500

Time, s 407 While-drilling tool Test 6: 1,850 to 2,250 L/min 406

While-drilling tool Test 5: 2,262 L/min

405 Pressure, bar

During field testing, the performance of the tool in both low-mobility (less than 0.2 mD/cP) and high-mobility (more than 350 mD/cP) formations was evaluated, with most data acquired being compared with wireline MDT pressure data and cores. Tests were conducted in a vertical well, highly deviated wells (up to 75°) and a horizontal well, with circulation rates ranging from pumps off to 2,300 L/min [600 galUS/min]. To assess the effects of time elapsed since drilling, pressure measurements were taken from one hour to 43 hours after the bit penetrated the test depth. While-drilling pressures were compared with those obtained with a MDT tester up to 24 days after the whiledrilling measurements. Field tests in Norway established that realtime pressure measurements made by the StethoScope tool are comparable to those made by wireline MDT testers under similar permeability, mud type and borehole conditions. In general, the most accurate pressure measurements were obtained in formations with the highest mobility, with pumps off, or when using a circulation rate that was as low and as constant as possible, and while tripping out of hole (right). Measurements made during the drilling process should be repeated at selected stations while tripping out to confirm the pressure values obtained, especially if supercharging is suspected. Engineers determined that for formations having fluid mobilities below 5 mD/cP, there is a distinct advantage to acquiring sandface pressures with pumps off. The degree of supercharging as a function of circulation rate depends directly on the amount of time since the filtercake was mechanically disturbed. High circulation rates can promote the erosion of an established filtercake, yielding supercharged sandface pressures even after a lengthy period between drilling and the pressure test. It is not always safe to assume that the supercharging effect decreases with time after drilling. Timelapse data are important in identifying dynamic supercharging in formations with low mobilities. The field test conducted by Statoil and Schlumberger in the North Sea yielded positive results. The StethoScope tool demonstrated its ability to accurately measure formation pressures in real time without needing to orient the tool or incurring excessive nonproductive time. In formations where the mobility is sufficiently high, 5 mD/cP or more, the StethoScope measurements are of the same quality as those acquired with the MDT tool. Today, both tools are helping engineers,

404

403

Start of change in circulation

While-drilling tool Test 7: No circulation

402 MDT test 401 150

200

250

300

350

400

450

Time, s

> Measuring pressure in a low-mobility formation. Three StethoScope test sequences (top) were conducted at the same depth under different mud-circulation rates in a 3-mD/cP formation: a rate of 2,262 L/min [597 galUS/min] (Test 5); a changing rate from 1,850 to 2,250 L/min [489 to 594 galUS/min] (Test 6); and a test with pumps off (Test 7). All of the while-drilling tests are of pretest Type C. For comparison, a single-pretest MDT tester conducted some 24 days after the StethoScope tests is displayed. The graph (bottom) shows details of the final buildup. Here the effects of the dynamic wellbore can be clearly seen by comparing the while-drilling tests with pumps on and pumps off with the MDT tool test, in which the filtercake should have reached maximum resistance.

geologists and drillers quickly make decisions, reduce drilling uncertainty and save time and money. While-Drilling Reservoir Pressure in the Gulf of Mexico In deepwater drilling and production environments, operators strive to reduce risk, uncertainty and cost. An example is the Ram Powell Production Unit, operated by Shell Offshore. Covering eight blocks in the Viosca Knoll area, eastern Gulf of Mexico, USA, wells are located in

2,000- to 4,000-ft [609- to 1,219-m] water depths, about 125 miles [200 km] east-southeast of New Orleans. Production began in September 1997, making it one of the most mature deepwater oil fields in the Gulf of Mexico.18 17. Pop J, Follini J-M and Chang Y: “Optimized Test Sequences for Formation Tester Operations,” paper SPE 97283, presented at Offshore Europe 2005, Aberdeen, September 6–9, 2005. 18. Volokitin Y, Stachowiak J and Bourgeois T: “Value of Specialized While Drilling Measurements: Recent Experience in Ram/Powell, GOM,” Transactions of the SPWLA 46th Annual Logging Symposium, New Orleans, June 26–29, 2005, paper C.

35

Five commercial sands between 5,500- and 13,500-ft [1,676- and 4,114-m] subsea true vertical depth (TVD) are responsible for most of the Ram Powell production. Geologists and engineers reevaluated the field between 2001 and 2003, including time-lapse seismic surveys that identified potentially undrained infill-drilling opportunities. In January 2004, Shell initiated redevelopment activities. Engineers found a high degree of risk and uncertainty in the new drilling projects. New well targets required drilling complicated directional wells. Production had also depleted several pay sands, rendering them unstable and difficult to drill. Although these circumstances made formation evaluation more difficult, the added uncertainty increased the need for formation evaluation while drilling. To reduce cost and improve efficiency, Shell and Schlumberger engineers planned to use LWD and MWD technologies to evaluate the reservoir and drilling environment in real time on Well 2 of the redevelopment campaign. Engineers selected a bottomhole assembly (BHA) consisting of a PowerDrive Xtra rotary steerable system, a suite of 6.75-in. VISION Formation Evaluation and Imaging While Drilling tools, and the StethoScope tool components—all positioned below a hole opener. The VISION suite included an arcVISION675 6 3⁄4-in. drill collar resistivity tool, an adnVISION675 6.75-in. Azimuthal Density Neutron tool and a proVISION675 6.75-in. nuclear magnetic resonance (NMR) tool. The TeleScope telemetry service provided real-time data transmission and control. Engineers planned to use the pressure data acquired by the StethoScope tool for completion design and to verify dynamic reservoir models. Obtaining formation pressure measurements while drilling reduced both rig cost and borehole exposure times, and allowed reservoir engineers and geologists to make timely wellplacement decisions. After the driller set 11 3⁄4-in. casing at 10,474-ft [3,192-m] measured depth (MD), the initial 10 5⁄8-in. hole section was drilled at an inclination of about 45° from 10,514- to 15,688-ft [3,205- to 4,782-m] MD. Shell’s petrophysicist selected the formation pressure-measurement points using data from the density neutron tool to determine the location of target sands. When on station, the StethoScope tool automatically began a drawdown-wait-retract sequence. After each pressure measurement, the probe was retracted and the tool was moved to the next station. On-station time averaged 10 minutes or less per measurement.

36

Real-time formation pressure measurements showed good pressure support within the reservoir and confirmed that the low-resistivity zone at the bottom of the target sand was swept oil, indicating a higher than expected oil/water

contact. Using this real-time data, Shell engineers decided to sidetrack the well. The new hole was placed higher in the reservoir by drilling updip from 11,501- to 16,952-ft [3,506- to 5,167-m] MD at 58° inclination. The pressure measurements Sidetrack

Original Well

40-in. Blended Attenuation Resistivity

40-in. Blended Attenuation Resistivity

0

10 0

10

adnVISION Bottom Quadrant Density

adnVISION Bottom Quadrant Density

g/cm3

g/cm3

1.65

2.65 1.65

Photoelectric Factor 0

b/e

20 0

Gamma Ray Hydrostatic Pressure 30 psi per division

Formation Pressure 200 psi per division

4 3 5 6 7 8

11

150 0

Neutron Porosity 0.6

%

20

gAPI

150

Neutron Porosity 0 0.6

%

0

4 5 6 7 8

9 12

11

12

XX,800 9 1110

13

1

1

2

2 5

XX,900

5

6 7

6 7

11

11

10

13

gAPI

b/e Gamma Ray

3

13

9

0 Depth

2.65

Photoelectric Factor

10 9

8

XY,000

13

Sidetrack Original well

> Using adnVISION data to help pick pressure stations. Density data from the adnVISION675 logging tool (Tracks 3 and 4 - red) helped select locations, or pressure stations, to be tested with the StethoScope pressure tool. Tracks 1 and 2 show formation and hydrostatic pressures, respectively, from the original hole (red circles) and the sidetrack (green circles). Zones with lower density have higher porosity and fluid mobility.

Oilfield Review

> Wireline pressure-measurement tool. The PressureXpress tool diameter and profile are designed to reduce the risk of sticking. Shown are the sampling seal system (black) and the backup anchoring pistons (top image – lower side) used to push and hold the tool in position at the formation face.

confirmed good connectivity within the reservoir, so casing was run to total depth. The proVISION675 NMR data helped engineers calibrate net-sand calculations and improve the petrophysical evaluation of laminated sands. Combining while-drilling data from NMR with data from other LWD tools provided important information on rock texture, permeability, grain size and net sand. NMR fluid-property data were used with formation pressure and fluid mobilities to estimate formation permeability. In total, 26 pressure measurements were made—13 while drilling the initial section and 13 while drilling the sidetrack. Pressure measurements provided critical information for making sidetrack and completion-design decisions. Pressures were successfully acquired in both massive and laminated sands (previous page). Advances in MWD technologies now provide pressure and fluid-mobility measurements that were previously available only with wireline logs. Engineers can perform complex evaluations based on MWD data alone, significantly reducing the risk, cost and uncertainty of drilling deepwater development wells. Engineers estimate that formation pressure while drilling and associated measurements saved more than US$1 million by eliminating the

Autumn 2005

need for two conventional drillpipe-conveyed pressure-measurement runs. In addition, whiledrilling NMR data provided important information regarding fluid viscosity and rock texture for net-sand calculation as well as permeability and grain-size estimations, which were used in the completion design. Shell plans to continue using pressure-while-drilling and other real-time technologies to improve efficiency and reduce risk, particularly on challenging deepwater projects. Quick and Precise Reservoir Pressures Wireline-conveyed formation-testing measurements have long been recognized as key to collecting essential information to help identify fluids in place, pressure regimes and dynamic properties of a reservoir. Although they measured formation pressures accurately, previous techniques required that a wireline tool be stationary for relatively long periods of time while testing the formation. This is particularly true in low-mobility zones where longer evaluation times increase the cost and risk of tool sticking.19 Now, reservoir engineers have options that provide fast and accurate pressure measurements. Tools such as the PressureXpress

service quickly make multiple, highly accurate pressure measurements. Engineers at the Schlumberger Riboud Product Center in Clamart, France, integrated advanced versions of the CQG gauge and the Sapphire pressure gauge into the PressureXpress tool. These pressure sensors provide highresolution pressure measurements, dynamically compensated for temperature (above). Earlier formation testing tools relied on hydraulically driven pretest systems that were monitored and controlled from surface. The lag time between surface commands and changes in the downhole hydraulic-sampling actuator limited pretest volume control. The system was redesigned, replacing the hydraulic system with an electromechanical motor coupled to a planetary roller-screw mechanism and highreduction gearbox, greatly enhancing the stability and accuracy of both the pretest rate and volume. Transferring the controls and commands from the surface to the downhole electronic cartridge improved the response time, making it possible to achieve pretest volumes as low as 0.1 cm3 [0.006 in.3]. 19. Manin Y, Jacobson A and Cordera J: “A New Generation of Wireline Formation Tester,” Transactions of the SPWLA 46th Annual Logging Symposium, New Orleans, June 26–29, 2005, paper M.

37

AIT H 30-in. Investigation Test is Not Recommended 0.2

ohm.m

2,000

AIT H 60-in. Investigation Tight 0.2

Permeability Volume of Clay 0

—

Slow Pretest

Computed Micro Normal

0.2

Gamma Ray 0

gAPI

ohm.m

20

150 0

Env. Corr. Thermal Neutron Porosity (TNPH)

ohm.m 10

Computed Micro Inverse ohm.m 10

Gas Effect From DPHZ to TNPH

2,000

AIT H 20-in. Investigation 0

in.

2,000

AIT H 10-in. Investigation 1

Caliper 0

ohm.m

Normal Pretest 0.2 AIT H Water Saturation

ohm.m

2,000

0.6

AIT H 90-in. Investigation

%

0

Std. Res. Density Porosity (DPHZ) Fast Pretest

0.2

ohm.m

2,000

0.6

%

0

> Locating pressure-test stations while logging. During the running of the combination PlatformExpress integrated wireline logging tool and the PressureXpress tool, a real-time estimate of formation fluid mobility is output using the k-lambda correlation, a permeability estimator for siliciclastics. Engineers use this information to help select zones for pressure testing (Track 5). The data output also helps select an appropriate pretest flow rate and volume that minimize time on station during the subsequent formation pressure-test operations.

20. For more on time-lapse seismic monitoring: Alsos T, Eide A, Astratti D, Pickering S, Benabentos M, Dutta N, Mallick S, Schultz G, den Boer L, Livingstone M, Nickel M, Sønneland L, Schlaf J, Schoepfer P, Sigismondi M, Soldo JC and Strønen LK: “Seismic Applications Throughout the Life of the Reservoir,” Oilfield Review 14, no. 2 (Summer 2002): 48–65; and Aronsen HA, Osdal B, Dahl T, Eiken O, Goto R,

38

Khazanehdari J, Pickering S and Smith P: “Time Will Tell: New Insights from Time-Lapse Seismic Data,” Oilfield Review 16, no. 2 (Summer 2004): 6–15. 21. Burgess K, Fields T, Harrigan E, Golich GM, MacDougall T, Reeves R, Smith S, Thornsberry K, Ritchie B, Rivero R and Siegfried R: “Formation Testing and Sampling Through Casing,” Oilfield Review 14, no. 1 (Spring 2002): 46–57.

Formation testers have traditionally been run alone or on the bottom of a wireline toolstring, because of their inability to pass the telemetry of other wireline tools. Implementation of new through-wiring hardware and a new software telemetry system now permits combinations with all other wireline tools, which can be run anywhere above or below the new tool. Through 2004, the PressureXpress service was field-tested on a total of 57 jobs, and more than 1,300 pretests were performed, in a wide variety of environments, including sandstone and carbonate formations (left). These involved formation fluids varying from gas to heavy oil under steam recovery. Bottomhole temperatures ranged from approximately 100 to 310°F [38 to 154°C], at hydrostatic pressures of 0 to 13,000 psi [0 to 90 MPa]. Engineers incorporated the Smart Pretest dynamically controlled intelligent pressure testing system to automatically find the best possible compromise between the formationproduced volume and the pressure-buildup time. In formations with fluid mobilities greater than about 1 mD/cP, the new tool can perform a pressure and mobility test in less than one minute: this represents an improvement of as much as four to five minutes over other testers. In tight formations, the pretest system can select fluid volumes as low as 0.1 cm3, allowing minimal test times (next page, top). In some areas of Texas, tight gas-bearing sandstone reservoirs may have permeabilities ranging from a few microdarcies to tens of millidarcies. In these fields, gas production relies on hydraulic fracturing to provide the conduit for reservoir flow. Many of these areas are mature and partially depleted, resulting in large pressure differences between reservoir layers. Determining these pressures accurately is key to optimizing hydraulic-fracturing programs. Failed attempts to measure pressures with conventional tools led to fracturing the entire reservoir thickness, including depleted zones, resulting in unnecessary completion expenses and lost production. For one well, Schlumberger engineers used the PressureXpress service; 58 pretests were attempted and 56 formation pressures were measured in less than seven hours. Data obtained from the testing program identified zones in the mid-reservoir section depleted by as much as 4,000 psi [27.6 MPa], while the last 500 ft [152 m] of pay was still at a

Oilfield Review

relatively high pressure (below right). Engineers designed a four-stage hydraulic-fracturing stimulation procedure. Six fewer stages were pumped than on previous wells, yet production increased by more than 50%, saving the operator more than US$ 400,000 in fracturing costs.

4,400

Mud pressure before test, psi: Mud pressure after test, psi: Last buildup pressure, psi: Drawdown mobility, mD/cP:

4,200 Pressure, psi

4,436.22 4,428.86 3,809.33 0.01

4,000 3,800 3,600 3,400 3,200

0

100

200

300

400 Time, s

500

600

700

800

> An electromechanical pretest system to reduce time on station. At time 0 s, the PressureXpress tool is on station and because it is not yet set, the flowline pressure is reading the wellbore mud pressure, approximately 4,430 psi [30.5 MPa] (black square, left). The tool is then hydraulically set, corresponding to an increase in the hydraulic pump velocity (green). The pressure curve (black) measures a pretest drawdown in a 0.01-mD/cP formation at about 45 s, followed by a gradual pressure buildup. After about 200 s, the tool initiated a second pressure drawdown (red triangles), extracting a volume of 0.1 cm3 of fluid from the formation. From 280 s to 680 s, reservoir pressure stabilized, then the tool was hydraulically retracted from the formation (green curve), and the flowline pressure increased to the wellbore mud pressure. By using an electromechanical motor, the pressure-testing tool accurately controls the pretest volume and rate at low values (0.1 cm3), effectively reducing buildup duration and time on station.

8,500 Mud gradient, 0.493 psi/ft

9,000

9,500 Depth, ft

Managing Reservoir Pressures Geopressures drive oil from a reservoir to a producing wellbore. As production starts, a pressure drop in the formation around the wellbore causes oil to flow through pore networks in the reservoir toward the point of withdrawal. With oil withdrawal and subsequent pressure drop in the reservoir, the oil, water and rock expand. Changes in pressure, expansion and movement of all these materials influence oil production. Once a reservoir is producing, engineers and geophysicists use various techniques to monitor the movement of fluids and changes in pressure. Recent advances in seismic imaging allow acquisition of 3D surveys over time, known as time-lapse or four-dimensional (4D) seismic surveys.20 Understanding the movement of fluids and changes in reservoir pressures allows engineers to better model reservoir behavior and improve recovery efficiency. From a purely mechanical point of view, geopressure data are relatively simple to obtain during or shortly after drilling. As discussed earlier, real-time, while-drilling pressure tools, such as the StethoScope system, are providing engineers with valuable geosteering and reservoir data for completion design, while wirelineconveyed tools, such as the PressureXpress service, provide accurate pressure and mobility data shortly after the well is drilled. But, there is a problem: once casing is run into the well, these tools cannot access the formation where there are no perforations. Thus with time, known pressures become unknown, and production decisions become more uncertain. Using sensors similar to those mounted in while-drilling and wireline pressure tools, Schlumberger engineers designed the CHDT Cased Hole Dynamics Tester, capable of measuring pressure and extracting fluid samples from behind a cased wellbore.21

4,600

10,000

10,500

11,000 0

4,000

2,000

6,000

Pressure, psi

> Identifying depleted zones. Formation pressures (red) are compared with the mud gradient (green), identifying zones depleted by as much as 4,000 psi [27.5 MPa] in the middle part of the reservoir, while the last 500 ft of pay is still at a relatively high pressure.

Autumn 2005

39

> Cased-hole sampling tool. Engineers use the CHDT tool to make multiple pressure measurements and collect fluid samples from behind a cased wellbore. Powered by a wireline cable, the backpressure pads (lower side of tool) push the tool into an eccentric position (tool shown eccentered against blue casing frame), the probe (shown tool top side) seals against the casing, and then drills a hole and measures pressure, samples fluids and plugs the hole. As the drill bit penetrates the target, the onboard integrated instrument package monitors pressure, fluid resistivity and drilling parameters.

The CHDT tool can drill through casing and cement and then into the formation, take multiple pressure measurements, recover fluid samples and plug the holes made in the casing, all in a single descent (above). The ability to reseal the drilled holes makes the tester uniquely suitable for several reservoir and production applications: for example, locating bypassed hydrocarbons, evaluating unknown pay zones, producing or injecting through a few holes, and determining formation-evaluation parameters when no openhole logs are available. Engineers can then optimize recompletion plans, enhance old or incomplete log data, assess unknown pay zones and evaluate wells for economic potential. The tool reduces rig cost by eliminating the costs of conventional plugsetting and cement-squeeze operations. An operator in south Texas requested an evaluation of a well drilled in 1941. Cased-hole logging tools identified multiple zones with potential hydrocarbons. Engineers used a USI UltraSonic Imager device to evaluate casing

40

condition and cement quality, and then the CHDT tool to measure reservoir pressure and confirm fluid type. During the test, seven formation pressures were acquired. Four samples confirmed hydrocarbons. The CHDT tool successfully plugged all holes. Based on test data, the operator was able to plan a recovery program for the bypassed hydrocarbons. Even though engineers can evaluate reservoir pressures behind casing long after production has begun, running tools in the hole is a costly, invasive procedure. Well problems are addressed most efficiently when acted on quickly. Developments in downhole telemetry, pressure sensors and advanced completion systems offer the reservoir engineer the flexibility to make production decisions in real time. Permanent downhole pressure sensors and monitoring tools, such as the WellWatcher realtime production monitoring and surveillance system, provide a continuous source of downhole

pressure measurement throughout the life of a well.22 Most often placed in the wellbore along with completion hardware, permanently installed sensors constantly monitor production pressures (next page). When these sensors are used with other real-time monitoring hardware such as fiber-optic temperature sensors, engineers can constantly update reservoir models and optimize the complete reservoir system. Before the introduction of these systems, data acquired through well intervention provided only a snapshot of well performance at the time the parameters were measured. Now, highly reliable downhole monitoring systems are sustainable in most downhole environments. The WellWatcher system has been in place on 15 BP North Sea assets for more than eight years.23 From 1995 through 2003, BP installed 75 of the monitoring systems in platform and subsea projects, all based on permanent pressure sensors. BP reports that during this period, only four of the systems have failed, translating to a success rate of 95%.24

Oilfield Review

In the BP Madoes field, a satellite of the Eastern Trough Area Project (ETAP) in the central North Sea, three producing wells are tied back 12.5 miles [20 km] to the primary platform. During completion, field specialists installed a FloWatcher integrated permanent production monitor system in each well to monitor pressure, temperature, absolute flow rate and fluid density in oil-water flows. Real-time data from the monitoring system allowed BP production engineers to quickly react, slowing water production by reducing production rate. If they had relied only on surface measurements, the problem would have been much worse. Engineers involved in the Foinaven deepwater subsea development, West of Shetlands, North Sea, are using real-time pressure sensors to better understand reservoir flow dynamics, adjust gas lift rates and maximize production. Prior to real-time downhole pressure measure-

Centralizers and electrodes

Gravel pack

Shrouded valve (WRFC-E)

Sand screens

Production packer

ments, engineers relied on modeling software to help optimize gas lift rates and hence, overall production rate. These models required data from costly and time-consuming well tests and generally provided results with limited accuracy. Real-time downhole pressure monitoring systems now allow BP engineers to adjust gas lift rates to achieve minimum flowing bottomhole pressure, and hence maximum production rates. Additional value is derived from the ability to track well performance based on establishing baseline parameters early in well life and then taking periodic time-lapse measurements of permeability height and skin factors from analysis of buildup pressure transients. This has provided early identification of well-performance problems, allowed more detailed evaluation and permitted intervention approaches to be optimized, with associated time and cost savings.

Electrical line to valves

Resistivity cable

External casing packer (ECP)

Hydraulic line to ECP

Sand screens ECP

ECP

Oil Water

Zone 1

Zone 2

Zone 3

The Foinaven team has also used the data from permanent downhole sensors to significantly improve understanding of reservoir connectivity, and then optimize various water injection and associated voidage replacement strategies. Engineers involved in the project estimate that over a three-year period, beginning in 2000, the combined benefit of these monitoring systems accounted for 1 to 3% of incremental production. Data from continuous real-time sensors provide engineers with the necessary information to optimize reservoir performance and recovery by detecting problems early and defining timely, preemptive reservoir-management solutions. Managing the Pressure System Maintaining reservoir pressure and optimizing recovery of oil and gas have become parts of a global challenge. The measurement of pressure throughout a reservoir’s life cycle is key to reservoir management. Accurate and efficient pressure measurement helps engineers and geophysicists manage subsidence, improve sweep efficiency in secondary-recovery operations and improve asset performance. Case histories demonstrate that engineers can refine predrilling seismic pore-pressure models using downhole data, then adjust models with real-time data allowing wells to be drilled more quickly with reduced cost, improved borehole placement and better management of the reservoir. As efforts continue to define the next energy revolution, reservoir engineers, geologists and geophysicists are combining the current developments in pressure-measurement tools with advances in seismic interpretation and modeling to optimize recovery and extend the life of known hydrocarbon reserves. — DW

> Highly complex sensor and control installation. As downhole control-system technology evolves, the complexity of completion, monitoring and control systems continues to increase. Multizone completion systems comprising packers, sand-control screens, flow-control valves (WRFC-E) and sensor packages for temperature, pressure, resistivity and more are not uncommon.

22. For more on permanent downhole pressure sensors: Bates R, Cosad C, Fielder L, Kosmala A, Hudson S, Romero G and Shanmugam V: “Taking the Pulse of Producing Wells—ESP Surveillance,” Oilfield Review 16, no. 2 (Summer 2004): 16–25. 23. “Keeping Watch on Production, Deepwater,” A Supplement to Hart’s E&P, Hart Energy Publishing, August 2004.

Autumn 2005

24. For more on pressure gauges and downhole monitoring: Eck J, Ewherido U, Mohammed J, Ogunlowo R, Ford J, Fry L, Hiron S, Osugo L, Simonian S, Oyewole T and Veneruso T: “Downhole Monitoring: The Story So Far,” Oilfield Review 11, no. 4 (Winter 1999/2000): 20–33.

41