Autumn 2013
Acidizing Advances
Monitoring Casing Corrosion
Geomagnetic Referencing
Solar Storms
ew pp
n
he ad a
t
ing w iP
c
n
e
ou evi
n
An eld R
lfi
i
O
Apple,
iPad, and iTunes are marks of Apple Inc., registered in the U.S. and other countries.
13-OR-0004
Schlumberger
Oilfield Review
www.slb.com/oilfieldreview
Executive Editor
Lisa Stewart
Senior Editors
Tony Smithson
Matt Varhaug
Rick von Flatern
Editor
Richard Nolen-Hoeksema
Contributing Editors
H. David Leslie
Ted Moon
Parijat Mukerji
Erik Nelson
Ginger Oppenheimer
Rana Rottenberg
Design/Production
Herring Design
Mike Messinger
Illustration
Chris Lockwood
Mike Messinger
George Stewart
Printing
RR DonnelleyWetmore Plant
Curtis Weeks
On the cover:
The aurora borealis appears as shimmering curtains of colored light in the
Arctic regions of the Earths northern
hemisphere. Auroras, which may occur
in both of the Earths polar regions, are
created when emissions from solar
ares and coronal mass ejections interact with the Earths magnetic eld. A
large loop of plasma, referred to as a
prominence, emanates from the Suns
surface (inset). Such a mass of plasma
ejected in the direction of the Earth
would create space weather events
that could disrupt modern electromagnetic-related technologies, including
well guidance methods that depend on
magnetic measurements.
Autumn 2013
Volume 25
Number 3
ISSN 0923-1730
Advisory Panel
Hani Elshahawi
Shell Exploration and Production
Houston, Texas, USA
Gretchen M. Gillis
Aramco Services Company
Houston, Texas
Roland Hamp
Woodside Energy Ltd.
Perth, Australia
Dilip M. Kale
ONGC Energy Centre
Delhi, India
George King
Apache Corporation
Houston, Texas
Andrew Lodge
Premier Oil plc
London, England
61 Contributors
63 Defining Production Logging:
Principles of Production Logging
This is the eleventh in a series of introductory articles
describing basic concepts of the E&P industry.
Editorial correspondence
Oilfield Review
5599 San Felipe
Houston, TX 77056
United States
(1) 713-513-1194
Fax: (1) 713-513-2057
E-mail: editorOilfieldReview@slb.com
Subscriptions
Customer subscriptions can be obtained
through any Schlumberger sales office.
Paid subscriptions are available from
Oilfield Review Services
Pear Tree Cottage, Kelsall Road
Ashton Hayes, Chester CH3 8BH
United Kingdom
E-mail: subscriptions@oilfieldreview.com
Distribution inquiries
Matt Varhaug
Oilfield Review
5599 San Felipe
Houston, TX 77056
United States
(1) 713-513-2634
E-mail: DistributionOR@slb.com
uids contact the largest possible reservoir surface area. Engineers and chemists
Bruno Lecerf
Alejandro Pea
Sugar Land, Texas, USA
Tim Lesko
Conway, Arkansas, USA
treatment uids tend to enter the fractures and avoid less permeable regions.
Effective uid diversion techniques are usually necessary to ensure that stimulation
Fred Mueller
College Station, Texas
Alexandre Z. I. Pereira
Petrobras
Rio de Janeiro, Brazil
Fernanda Tellez Cisneros
Villahermosa, Mexico
Oileld Review Autumn 2013: 25, no. 3.
Copyright 2013 Schlumberger.
For help in preparation of this article, thanks to
Charles-Edouard Cohen, Rio de Janeiro;
Victor Ariel Exler, Maca, Brazil; Luis Daniel Gigena,
Mexico City; Daniel Kalinin, Al-Khobar, Saudi Arabia;
and Svetlana Pavlova, Novosibirsk, Russia.
ACTive, MaxCO3 Acid, POD, SXE and VDA are marks
of Schlumberger.
1. Crowe C, Masmonteil J, Touboul E and Thomas R:
Trends in Matrix Acidizing, Oileld Review 4, no. 4
(October 1992): 2440.
2. Robert JA and Rossen WR: Fluid Placement and
Pumping Strategy, in Economides MJ and Nolte KG
(eds): Reservoir Stimulation, 3rd ed. Chichester,
West Sussex, England: John Wiley & Sons, Ltd (2000):
19-219-3.
Oileld Review
Since the dawn of the oil and gas industry, operators have endeavored to maximize well productivity, employing a variety of techniques to do so. For
example, as early as the 19th century, engineers
began pumping acid in wells to improve production. Acidizing treatments dissolve and remove
formation damage resulting from drilling and
completion operations, create new production
pathways in producing formations or both.
Acidizing treatments fall into two categories.
Matrix acidizing consists of pumping uid into
the formation at rates and pressures that will not
fracture the reservoir. The resulting treatment
stimulates a region extending up to about 1 m
[3 ft] around the wellbore. Fracture acidizing is a
hydraulic fracturing treatment that pumps acid
during at least one uid stage. The stimulation
distance may extend one or two orders of magnitude farther into the formation than that
achieved by matrix acidizing.
The composition of acidizing uids depends
on the type of formation to be stimulated.
Carbonate formations, composed mainly of limestone (calcium carbonate [CaCO3]) or dolomite
(calcium magnesium carbonate [CaMg(CO3)2]),
are treated with hydrochloric acid [HCl], various
organic acids or combinations thereof. Sandstone
formations typically consist of quartz [SiO2] or
feldspar [KAlSi3O8NaAlSi3O8CaAl2Si2O6] particles bound together by carbonate or clay minerals. Silicate minerals do not react with HCl; they
respond instead to stimulation uids that contain
hydrouoric acid [HF] or uoboric acid [HBF4].1
Despite the uid chemistry differences, the engineering aspects of carbonate and sandstone
acidizing are largely similar. However, this article
concentrates on recent advances that are particularly relevant to carbonate acidizing.
Carbonate Acidizing Fundamentals
Limestone and dolomite rapidly dissolve in
HCl, forming water-soluble reaction products
mainly calcium and magnesium chloridesand
liberating carbon dioxide. The dissolution rate
is limited by the speed at which acid can be
delivered to the rock surface. This dissolution
process results in rapid formation of irregularly
shaped channels called wormholes (above right).
Wormholes radiate outward in a dendritic pattern from points where acid leaves the well and
enters the formation. Once formed, they become
the most permeable pathways into the formation
and carry virtually all of the uid ow during production. For efcient stimulation, the wormhole
network should penetrate deeply and uniformly
throughout the producing interval.
Autumn 2013
to divert acidizing uids away from high-permeability intervals and into less permeable zones.
Engineers accomplish diversion by employing
mechanical or chemical means or both.2
Mechanical diversion of treatment uids may be
achieved using drillpipe or coiled tubingconveyed tools equipped with mechanical packers
that isolate and direct uid into low-permeability
zones. Alternatively, ow can be blocked at individual perforations by dropping ball sealers into
Ball Sealers
Straddle Packers
> Mechanical diversion methods. Ball sealers (green spheres) are pumped down the well during the
stimulation treatment (left). The balls provide mechanical diversion because they preferentially block
the perforations that take the highest volume of treatment uid. Straddle packers may also be deployed
on coiled tubing to isolate the preferred treatment interval (right). In this example, engineers have
already stimulated the bottom zone and moved the packers up in preparation for stimulating the next zone.
CaCO3 + 2HCl
diverting acid system. VDA uids have been particularly successful in both matrix and fracture
acidizing applications around the world.4
The surfactant molecule in the VDA system,
derived from a long-chain fatty acid, is zwitterionica neutral molecule that carries a positive
and a negative charge at separate positions.5
While being pumped down a well, VDA uida
blend of HCl, VES and common acid-treatment
additivesmaintains a low viscosity. As the acid
is consumed in the formation, the surfactant molecules begin to aggregate into elongated
micelles.6 The micelles become entangled and
cause the uid viscosity to increase (below). The
higher-viscosity uid forms a temporary barrier
that forces fresh acid to ow elsewhere. In addition to providing diversion, the viscosity decreases
the rate at which the acid reacts with the formation, thereby allowing more time for the creation
of deeper and more intricate wormholes.
When production begins, VDA uid is exposed
to hydrocarbons, which alters the ionic environment and causes the micelles to become spherical. Entanglement ceases, the micelles roam
freely, and the uid viscosity decreases dramatically, enabling efcient poststimulation cleanup.
Unlike polymer-base uids, VESs leave virtually
no damaging residue behind that may interfere
with well productivity.
Naturally fractured reservoirs are the most
challenging environments for carbonate acidizing because they can present extreme permeability contrasts. The fractured regions may be
several orders of magnitude more permeable
than the unfractured layers. Until recently, the
industrys considerable portfolio of diversion
technologies has been inefcient in this environment. Even when using self-diverting uids such
as the VDA formulation, engineers struggled to
block the fractures and treat the rest of the formation. Consequently, operators were forced to
pump large volumes of uid to achieve stimulation, leading to higher treatment costs and less
than optimal results.
However, Schlumberger engineers and chemists discovered that signicant diversion improvements could be achieved by adding degradable
bers to VDA uid. As ber-laden diversion uid
enters a fracture, the bers congregate, entangle
and form structures that limit uid entry. The
new product, MaxCO3 Acid degradable diversion
acid system, has been used successfully and efciently to stimulate notoriously difcult carbonate reservoirs around the world.
Spent acid
Hydrocarbon
Surfactant
molecules
Elongated micelles
Spherical micelles
> Viscoelastic surfactant (VES) uid behavior during an acidizing treatment. Initially, when the surfactant is dispersed in acid, each molecule moves
independently throughout the uid (left). As the acid reacts with the carbonate minerals, the surfactant molecules assemble and create elongated micelles
(center). The micelles entangle and hinder uid ow, resulting in higher uid viscosity. When hydrocarbon production begins after the treatment, the
elongated micelles transform into spheres (right), resulting in a dramatic decrease in uid viscosity and facilitating efcient cleanup.
Oileld Review
Openhole Acidizing
Wormhole
Perforation
Wormhole
Wellbore
wall
Casing
Filtercake
Filtercake
Treatment fluid
Treatment fluid
Filtercake
Perforation
Filtercake
Well
Well
Casing
> Fiber deposition and diversion scenarios. During openhole acidizing (top and bottom left), bers form
a ltercake that covers the entire wellbore wall. During cased hole acidizing (top and bottom right),
bers form ltercakes in the perforation tunnels.
Autumn 2013
Openhole Simulation
Pressure
Acid and
fibers
Filtercake
Pressure cell
Core
Cased Hole Simulation
Backpressure
regulator
Pump
Filtrate
Pressure sensor
Wormhole Geometry
Balance
Orifice
Piston
1 to 2 mm
20 mm
25.75 mm
Fluid flow
142 cm
Acid
and fibers
2 to 6 mm
ID 21 mm
Filtercake
65 mm
75 mm
Orifice
> Laboratory-scale equipment for testing leakoff behavior and ltercake deposition. Engineers used a conventional ltration cell to simulate an openhole
stimulation (top). Technicians rst placed a carbonate core at the bottom of the cell and then poured in ber-laden acid. After sealing the cell, they applied
differential pressure across the core and used a balance to measure the amount of ltrate passing though the core. For the cased hole simulation (bottom),
engineers used a bridging apparatus. The apparatus consisted mainly of a 300-mL tube tted with a piston, a high-performance liquid chromatography
(HPLC) pump and an orice (left). The orice could be circular to simulate a wormhole (top right) or rectangular to mimic a fracture (bottom right).
Technicians installed a piston at the top of the tube, which contained ber-laden acid. Acid exiting the tube passed through the orice, and the technicians
assessed the diversion capability of bers by measuring the ltrate volume, the ber ltercake volume and the pumping pressure at various ow rates.
The engineers then began performing experiments with laboratory-scale equipment for
simulating uid leakoff and ber deposition
(above). The principal simulator was a bridging
apparatus that accommodated a variety of orices through which ber-laden acid could pass
at various ow rates. Circular orices, with
diameters between 1 and 2 mm [0.04 and
0.08 in.], simulated wormholes. Rectangular orices with widths between 2 and 6 mm [0.08 and
0.24 in.] were analogous to fractures. Engineers
observed ber plug formation and recorded the
corresponding system pressure as ber-laden
acid passed through an orice.
Oileld Review
150
16.4
32.8
49.2
65.6
82.0
98.4
Fluid inflow
60
50
40
Pressure, psi
30
20
Bridging region
100
Nonbridging region
10
50
0
10
20
30
40
50
60
70
80
1,000
Apparent permeability, mD
10
15
20
25
30
ing ber deposition, the measured permeabilities varied between 400 and 2,400 mD. These
data led engineers to conclude that bers would
provide the most efcient diversion in zones
with permeabilities exceeding 100 mD (left).7
The data acquired during the simulator experiments also allowed scientists to develop a mathematical model for predicting the behavior of
ber-laden acids under openhole and cased hole
conditions; the model may be used to optimize
treatment designs.8 They performed 340 ne-scale
3D simulations that evaluated typical perforation
schemes, brous ltercake permeabilities and
formation permeabilities. The resulting model
allows scientists to track the movement of the uids and bers through the wellbore and into the
reservoir and track the propagation of wormholes
generated as the acid reacts with carbonate rock.
10,000
100
10
0.1
0.1
10
100
1,000
10,000
Core permeability, mD
Autumn 2013
Time, s
10,000
1,000
Layer permeability
30 D
10 D
3D
1D
Flow rate
100
10
0.1
0.1
10
100
Reservoir permeability, mD
1,000
10,000
Time
> Diversion predictions from the MaxCO3 Acid simulator. During ber deposition experiments in the perforation simulator, the permeabilities of the resulting
ber plugs varied between about 400 and 2,400 mD (left). The simulator predicts how the ber plugs decrease the apparent permeabilities of reservoirs and
promote diversion. Lower-permeability ber plugs are more efcient diverters. Modeling studies also demonstrated that brous ltercakes provide uid
diversion by equalizing the permeabilities of layers in the treated interval. For example, if the interval contains four layers with various permeabilities, the
uid ow rate into the more permeable layers decreases and the uid ow rate into the less permeable layers increases. Eventually, the ow rates
converge to a single ow rate, and the interval behaves as if it has a single permeability (right). Flow rate convergence occurs more quickly in a cased hole
with perforations because the ltercake surface area is lower.
> MaxCO3 Acid uid batch mixing. The degradable bers (top left) are light and nely divided, presenting a mixing challenge. Traditional equipment for
batch mixing of acidizing uids was inefcient. Engineers discovered that equipment for batch mixing cement slurries (bottom left) could disperse the bers
in VDA uid. The VDA uid ows into an 8,000-L [50-bbl] paddle mixer (top right). To avoid the formation of clumps, eld personnel manually add bers to the
uid. After the bers have been added, the tank is lled with more VDA uid, and agitation continues until the mixture reaches a uniform consistency
(bottom right). During the job, engineers maintain the agitation to preserve uid uniformity.
10
Oileld Review
120
100
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
16-h
shut-in
16-h
shut-in
10
9
8
Autumn 2013
Permeability, mD
K0
K6
K1
K7
N2
K4
K2
K5
K3
3
2
1
0
0
10
15
20
25
30
35
40
45
50
55
> Behavior of degradable bers. Engineers performed static bottle tests during which degradable
bers were immersed in partially spent HCl uids. The data show that the rate of ber dissolution
decreases as the HCl becomes neutralized. Nevertheless, complete ber dissolution occurs within a
few days (top). Core testing demonstrated that the acidic ber degradation products may further
stimulate the formation (bottom). Using a standard core testing apparatus at 115C [239F], engineers
pumped 2% KCl solution into a limestone core rst in the injection direction and then in the reverse, or
production, direction (K0 and K1). Technicians recorded the pressure across the core and, applying
Darcys law, determined that the initial core permeability was 5.1 mD. Next, they injected a partially
spent 20% HCl uid (pH = 6.5) containing degradable bers (N2). Subsequent pumping of 2% KCl in both
directions revealed that the core permeability had fallen to 3.5 mD (K2 and K3). Following a 16-h shut-in
period, the bers had begun to degrade, and the core permeability rose to about 4.8 mD (K4 and K5).
After another 16-h shut-in period, complete ber degradation had occurred, and the core permeability
rose to 5.5 mD (K6 and K7)an 8% improvement over the initial permeability of 5.1 mD.
11
IRAN
IRAN
SAUDI
ARABIA
South
Pars
BAHRAIN
North
Field
Fiber feeder
SAUDI
ARABIA
QATAR
0
0
> Continuous mixing of MaxCO3 Acid uid. A POD blender is outtted with a
special ber delivery feeder (top right) that has no restrictions or bends,
thus ensuring smooth metering. Field workers place a berm (top left) under
the blender to guard against uid spills. A plastic sidewall around the mixing
tubs (bottom) further shields the mixing process.
MEXICO
Jujo-Tecominoacn
Field
50
Tabasco
State
Villahermosa
km
50
miles
50
> Jujo-Tecominoacn eld. This region is among the most prolic oil and gas producing areas in
southern Mexico. The reservoirs are naturally fractured and difcult to stimulate uniformly.
12
50
mi
> Qatar North eld. Discovered in the 1970s, this accumulation is the largest
gas eld in the world, with estimated reserves as high as 25.5 trillion m3
[900 Tcf]. The reservoir is called the South Pars eld on the Iranian side of
the maritime border (dashed black line). The producing formation is
characterized by large interzonal permeability contrastsup to a ratio of
100:1. The reservoir depth is about 3,000 m [9,800 ft] below the seabed, and
the elevated hydrostatic pressure tends to favor stimulation of bottom
zones at the expense of upper reservoir layers, further increasing the
difculty of achieving uniform stimulation in one treatment.
UNITED STATES
0
0
km
Oileld Review
Autumn 2013
12,200
12,300
12,400
12,500
Measured depth, ft
12,600
12,700
12,800
12,900
13,000
13,100
13,200
0.1
10
100
1,000
Permeability, mD
7,500
40
35
8,000
30
BHP, psi
7,000
25
20
6,500
15
6,000
10
5,500
5,000
80
5
0
100
Fluid at perforations
MaxCO3 Acid fluid
Gas
Water
HCI
VDA acid
120
140
160
Time, min
Measured BHP
Simulated BHP
Pump rate
> Simulated and measured pressures from a eld test in the Qatar North eld. Engineers pumped four
stages of 28% HCl and MaxCO3 Acid uid. A VDA uid spacer preceded and followed each MaxCO3
Acid stage to preserve ber suspension uniformity. The excellent agreement between the measured
(blue curve) and simulated (black) bottomhole pressures (BHP) helped conrm the validity of the
MaxCO3 Acid placement model.
13
Fluid Name
Stage Fluid
Volume, m3
Preflush
Aromatic solvent
10
Acid
20
Diverter
Spacer
3% NH4Cl brine
Preflush
Aromatic solvent
10
Acid
20
Diverter
Spacer
3% NH4Cl brine
Preflush
Aromatic solvent
10
Acid
20
Flush
Nitrogen
Stage Name
Nitrogen Pump
Rate, m3/min
80
80
150
> Pumping schedule for a matrix acidizing treatment in the JujoTecominoacn eld. During the 11-stage treatment, engineers pumped an
aromatic solvent to clean up perforations, an HClformic acid blend,
MaxCO3 Acid uid and an ammonium chloride brine spacer. The nal stage
contained nitrogen [N2] to enhance well cleanup.
3,500
3,000
Oil production
2,500
2,000
1,500
Apr 2009
July 2009
Oct 2009
Jan 2010
Apr 2010
Date
> Production history in a PEMEX well in the Jujo-Tecominoacn eld. Initial oil production was
1,278 bbl/d [203 m3/d]. Subsequent matrix acidizing treatments employing conventional techniques
failed to achieve sustained production improvements. After a MaxCO3 Acid treatment in December
2009, oil production increased to 3,000 bbl/d and stabilized at 1,600 bbl/d, exceeding the original
production rate.
14
Oileld Review
Autumn 2013
IRAN
IRAN
EGYPT
SAUDI
ARABIA
BAHRAIN
South Ghawar
Field
QATAR
UNITED ARAB
EMIRATES
0
0
Oil
km
SAUDI ARABIA
100
mi
100
Gas
> South Ghawar eld in eastern Saudi Arabia. The producing reservoirs, in the Khuff Formation, are
composed of heterogeneous carbonates. The permeability and porosity vary widely within 100 to 200 ft
[30 to 60 m] of formation thickness, presenting difcult uid diversion challenges.
Treatment Schedule
Stage Name
Pump Rate,
bbl/min [m3/min]
Fluid Name
Stage Fluid
Volume, galUS [m3]
Acid
Concentration, %
Pad
20 [3.2]
9,000 [34]
Acid 1
20 [3.2]
9,000 [34]
28
Pad
30 [4.8]
3,000 [11]
Diverter 1
30 [4.8]
3,000 [11]
15
Pad
30 [4.8]
9,000 [34]
Acid 2
30 [4.8]
9,000 [34]
28
Pad
35 [5.6]
3,000 [11]
Diverter 2
35 [5.6]
3,000 [11]
15
Pad
40 [6.4]
9,000 [34]
Acid 3
40 [6.4]
9,000 [34]
28
Pad
40 [6.4]
3,000 [11]
Diverter 3
40 [6.4]
3,000 [11]
15
Pad
40 [6.4]
10,000 [38]
Acid 3
40 [6.4]
9,000 [34]
28
Overflush 1
40 [6.4]
Overflush
7,000 [26]
Diverter 4
10 [1.6]
3,000 [11]
15
Acid 4
10 [1.6]
28% HCl
7,000 [26]
28
Overflush 2
10 [1.6]
Overflush
5,000 [19]
Flush
10 [1.6]
Water
11,200 [42]
> Pumping schedule for an acid fracturing treatment in Saudi Arabia. The total uid volume was
124,200 galUS [2,960 bbl, 470 m3], allowing simultaneous stimulation of three zones without the need for
mechanical diversion techniques. Such treatment simplicity saved several days of rig time, resulting in
signicant operational cost savings.
15
15,000
13,600
12,200
10
100
10,800
85
9,400
Fracturing pressure
70
8,000
55
6,600
Rate, bbl/min
Pressure, psi
1115
40
5,200
25
3,800
2,400
10
1,000
10
30
50
70
90
110
130
150
170
> Pressure and temperature data. During a Saudi Aramco acid fracturing treatment, the pumping rate
(blue line) varied from 10 to 40 bbl/min [1.6 to 6.4 m3/min], and the bottomhole treating pressure (red
line) exceeded the formation fracturing pressure (dashed black line) throughout most of the treatment.
The vertical blue bars denote periods during which MaxCO3 Acid uid entered the perforations.
SOUTH
AMERICA
0
1,000
2,000
Overburden formations
BRAZIL
Espirito Santo
Basin
Depth, m
3,000
4,000
Salt
5,000
Rio de Janeiro
So Paulo
Campos Basin
6,000
Curitiba
Presalt
oil
7,000
Santos Basin
8,000
0
0
km
500
mi
500
9,000
> The presalt reservoirs of Brazil. The main producing elds are located primarily offshore (left). The reservoirs are in carbonate formations that lie
underneath a thick layer of evaporite minerals (right). The reservoir depth is between 4,500 and 6,500 m [14,800 and 21,300 ft].
16
Oileld Review
8,000
40
8,000
36
7,000
7,500
32
6,000
7,000
4,000
3,000
6,500
24
20
6,000
16
5,500
5,000
28
12
2,000
5,000
8
1,000
4,500
4
0
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
4,000
10,000
Time, s
> Matrix acidizing treatment. In a presalt well offshore Brazil, engineers pumped 13 uid stages
consisting of alternating portions of 15% HCl, VDA diverter and MaxCO3 Acid uid at various pump rates
(blue curve). A mixture of 15% HCl and a mutual solvent preceded and followed the treatment. As the
treatment progressed, the rig pressure (red curve) and bottomhole pressure (green curve) rose,
indicating that the bers were effectively diverting treatment uid to zones with lower permeability.
Autumn 2013
17
Dalia Abdallah
Mohamed Fahim
Abu Dhabi Company for Onshore Oil
Operations
Abu Dhabi, UAE
Corrosion challenges are not new to the oil and gas industry, and producers are
continually seeking new ways to keep corrosion at bay. Experts have made advances
in corrosion monitoring along several fronts. The implementation of these technologies may help operators optimize infrastructure utilization, maximize production and
Khaled Al-Hendi
Mohannad Al-Muhailan
Ram Jawale
Kuwait Oil Company
Ahmadi, Kuwait
Adel Abdulla Al-Khalaf
Qatar Petroleum
Doha, Qatar
Oil and gas companies typically serve two masters. On the one hand, protability dictates that
producers maximize long-term production while
minimizing operating expenditures. On the other
Zaid Al-Kindi
Abu Dhabi, UAE
Abdulmohsen S. Al-Kuait
Hassan B. Al-Qahtani
Karam S. Al-Yateem
Saudi Aramco
Dhahran, Saudi Arabia
Nausha Asrar
Sugar Land, Texas, USA
Syed Aamir Aziz
J.J. Kohring
Dhahran, Saudi Arabia
Abderrahmane Benslimani
Ahmadi, Kuwait
M. Aiman Fituri
Doha, Qatar
Mahmut Sengul
Houston, Texas
Oileld Review Autumn 2013: 25, no. 3.
Copyright 2013 Schlumberger.
For help in preparation of this article, thanks to Ram Sunder
Kalyanaraman, Clamart, France.
Avocet, EM Pipe Scanner, FloView, Petrel, PipeView,
PS Platform, Techlog, UCI and USI are marks of
Schlumberger.
> Typical rening-corrosion life cycle for metals. Energy is stored in a metal as it is rened from its
naturally occurring state (such as iron ore) to an alloy. Corrosion takes place spontaneously and
releases the stored energy, which returns the metal back to a lower energy state. That process can be
slowed by the application of one or more eld-based mitigation measures.
18
Oileld Review
Autumn 2013
Problem
Cause of Corrosion
Control Methods
Monitoring
Oxygen corrosion
Oxygenated water
Internal attack
External attack
Resistant materials
Oxygen scavengers
Oxygen stripping
Improved seal design
Coatings
Cathodic protection
Hydrogen sulfide
corrosion pitting
Water from
production aquifer
or other deep aquifer
Water contaminated
by stripping or lift gas
Degassing at
low pressures
Control of
contaminated gas
Use of resistant
materials
Probes
Iron counts
Wall thickness surveys
Sulfate-reducing
bacteria (SRB)
Anaerobic fluids
Stagnant fluids
Conditions under scales
or other deposits
Biocides
Chlorination
Anaerobic bacteria
counts
Chlorine residuals
measurements
Carbon dioxide
corrosion
Water from
production aquifer
or other deep aquifer
Water contaminated
by stripping or lift gas
Degassing at
low pressures
Control of
contaminated gas
Use of resistant
materials
Probes
Iron counts
Wall thickness surveys
Hydrogen sulfide
stress corrosion
cracking
Hydrogen-induced
cracking
Produced fluids
containing hydrogen
sulfide
Anaerobic systems
contaminated with SRB
Suitable materials
Acid corrosion
Stimulation and
cleaning acids
Acid inhibitors
Galvanic (bimetallic)
corrosion
Electrical isolation
of metals
(cathodic coating)
Improved design
Design reviews
Immersion
Inert surface films
Materials selection
Equipment inspections
Subdeposit
corrosion
Pigging
Biocides
Improved sealing and
design
Minimum velocity design
Equipment inspections
Bacteria counts
Crevice corrosion
Poor design
Imperfections in metal
Improved design
Materials selection
Equipment disassembly
and inspections
Leak detections
Chloride corrosion
(rapid cracking on
exposure to hot
chloride media)
Salt solution
Oxygen and heat
Materials selection
Equipment inspections
Oxygen analyses
Fatigue
Rotating equipment
Wave-, wind- or
current-induced loading
Vibration design
Equipment inspections
> Summary of corrosion problems and solutions. In the oil eld, corrosion is pervasive and takes many
forms. By properly identifying the source of corrosive attack, an operator can implement a suitable
corrosion monitoring and control program.
19
Drilling and
Completions
Production
Well
Decline
Decommissioning
Production
Well
Design
Perform
reservoir
modeling.
Perform
core
analysis.
Select
suitable
drilling mud.
Select
suitable alloys
for pipe work
and equipment.
Perform
materials Select suitable
selection. oxygen and
sulfide
Perform
scavengers.
risk
analysis.
Use corrosion
monitoring tools
and services.
Implement more-stringent
and expansive asset integrity evaluation.
Implement or expand
oil and water separation operations.
technologies (corrosion
inhibitors, sand control systems
and oxygen scavengers).
Evaluate infrastructure condition
and track corrosion rates.
Implement repairs and
replacement strategies
as needed.
Ensure long-term
containment of
abandoned well.
Ensure
compliance with
environmental
regulations.
> Corrosion considerations at each stage of the asset life cycle. During each stage of a wells life, engineers must consider
operational factors to keep corrosion at bay and minimize the threat of production uid leaks into the surrounding environment.
20
the Middle East demonstrate how corrosion monitoring tools and mitigation technologies have
helped operators identify the location and severity of corrosion in the subsurface infrastructure,
which informed each companys choice of mitigation solution.
Corrosion and the Life Cycle
Corrosion is a major concern throughout the life
of a well, and specic considerations and mitigation strategies are required at each stage. Asset
personnel usually begin making corrosion mitigation decisions for a well before drilling.
During the well design stage, the operator conducts comprehensive reservoir studies, which
include reservoir simulation modeling, core
studies and uid analysis from offset well data.
Engineers use the information obtained from
these studies to develop risk assessments for
corrosion threats in subsequent stages of the
well. Engineers then develop and implement
mitigation strategies that include appropriate
materials selection, optimal production rates,
monitoring programs and corrosion inhibitor
treatments (above).
During the drilling process, operators focus
corrosion mitigation strategies on extending the
working life of drillpipe, which is exposed to high
operational stresses as well as potentially corrosive drilling muds and formation uids. The drillpipe may undergo one of several types of corrosion
mechanisms, including localized pitting, in which
H2S, chloride salts or oxygen in water-base drilling muds cause a corrosion rate that exceeds
25 cm [9.8 in.] per year.8 Other corrosion sources
Oileld Review
Autumn 2013
Operators realize a prot during a wells production stage, which may last from only a few
years to several decades. During this stage, corrosion mitigation efforts are generally focused
on keeping corrosion rates low and preventing
leaks (below). The operator must continually
Corrosion-induced
cracks
Water sand
Packer
Perforations
Oil sand
> Corrosions impact on casing integrity. Casing leaks typically arise from
excessive corrosion in the production system. These leaks, which can prove
costly and environmentally damaging, may allow additional formation water
and sand to enter the production string of the well (blue arrow). Alternatively,
crossows (green arrows) may result, which can be difcult to characterize
and treat, and in severe cases, the operator may have to pull and replace the
entire casing string.
21
Casing
Ultrasonic signal
Amplitude
Transducer
Time
Amplitude
Radius
Thickness
> Basic principles of the UCI ultrasonic corrosion imager. The UCI tool uses
a 2-MHz focused transducer to improve the resolution of the ultrasonic
measurement. The transducer also acts as a receiver of the reected signal
and records its amplitude and time of arrival. This signal is emitted (or
pulsed) through the well uid and into the casing (top). As this signal
encounters a discontinuity, such as the inner or outer wall of the casing
(center), the signal is reected back. Most of the energy is reected in the
initial echo at the inner casing wall because of the large impedance contrast
between the mud and the steel; the remaining energy transmitted into the
casing is again reected at the outer wall. The signal reected back at the
inner wall can be used to evaluate the casing condition and radius. The time
difference between the rst two echoes can be used to determine the
thickness of the casing (bottom). In comparison, the USI tool is more
commonly used for ultrasonic pipe inspection and employs a 200- to 700-kHz
unfocused ultrasonic transducer to induce a casing resonance. In the USI
measurement, thickness is determined from the resonance frequency.
(Adapted from Hayman et al, reference 15.)
22
coupon technique is advantageous because coupons can be fabricated from the same alloy that
makes up the system under study, the corrosion
rate can be easily calculated from the coupons
weight loss over the time of exposure and the
technique allows visual verication of corrosion
deposits or localized corrosion. However, if a corrosion event such as a leak occurred while the
coupon was in the system, the operator could not
use the coupon alone to accurately pinpoint its
time of occurrence. In addition, the coupon technique is applicable only in system locations that
provide easy or practical access for placing and
extracting the coupon.
This second limitation makes coupon monitoring, or any visual inspection technique, essentially impossible for the wells downhole tubulars
and casing strings. The remaining options are
indirect measurement techniques that incorpo-
Oileld Review
Autumn 2013
Tool outer
diameter
2D discrimination
Discriminator
transmitter, TH
TH
RP
Pipe
Pad
receiver, RP
RP
RP
TL
RLL
2D thickness
TH
Average thickness
RLL
RP
RLS
RLS
TL
RLS
RLL
TL
RLS
RLL
Z properties
RZ
RZ
RZ
TZ
RZ
TZ
1 1
0 ID
23
1.5
1,000
2,000
in.
1.5
1,000
1,000
2,000
2,000
3,000
in.
1.5
in.
1.5
1,000
2,000
3,000
3,000
3,000
in.
4,000
4,000
4,000
4,000
9 5/8-in. shoe
9 5/8-in. shoe
9 5/8-in. shoe
9 5/8-in. shoe
5,000
5,000
5,000
> EM Pipe Scanner logs. The logs for four Saudi Aramco wells showed varying degrees of metal loss
(red), remaining thickness (gray) and total measured thickness (green) with respect to depth. A distinct
pattern correlation, as well as a similar decrease in total thickness with depth, existed among the
wellbores. All wells showed metal losses in the range of 62% to 65% of the outer double casings at a
depth of approximately 2,500 ft. The operator used this information to anticipate similar metal loss
patterns and expected a comparable level of severity of corrosion in adjacent wells not yet logged.
24
Operating companies can obtain these measurements without having to pull the completion
tubing out of the hole, which saves rig time and
intervention expense. While the engineer lowers
the EM Pipe Scanner tool into the well on wireline, tractor or coiled tubing, the tool conducts an
initial high-speed reconnaissance run to ag
areas of interest for detailed diagnostic scans to
be performed as the tool is retrieved to the surface. The tool records a continuous log of both
the average casing inner diameter and total
metal thickness and provides corrosion estimates. The tool responds to overall metal thickness, allowing corrosion of the outer casing or
tubing to be detected. Measurements of the inner
casing metal radius are valid in the presence of
most kinds of scale. Its 2 1/8-in. diameter affords
access through tight restrictions. The tool can
operate in gas or liquid environments.
Forewarned Is Forearmed
In 2011, using the EM Pipe Scanner tool, Saudi
Aramco conducted a well-casing corrosion monitoring campaign in a eld containing both
onshore and offshore wells. Initial scans of seven
onshore wells indicated relatively little metal
loss and conrmed that the existing ICP system
was working satisfactorily. Because of the lack of
a sufciently large power supply, the offshore
wells had limited ICP, which raised the possibility of higher corrosion rates.
The EM Pipe Scanner tool was deployed to
determine the extent of metal loss from well casings in the offshore portion of the eld and to
help the operator geographically map wells
exhibiting the most severe metal loss. In one
campaign, in four adjacent wells that were originally completed in 1976, Saudi Aramco checked
to determine whether any of these well had concentric casings that might soon leak.19 If engineers observed metal loss, they planned to
analyze the loss prole for the purpose of mapping and anticipating the likelihood of casing corrosion in nearby, nonlogged wells.
The EM Pipe Scanner logs showed varying
degrees of metal loss in each of the four subject
wells, although the logs indicated a distinct depth
correlation among them. One noticeable correlation occurred between 2,500 and 2,800 ft [760 and
850 m], where the four wells had casing metal
losses ranging from 62% to 65% (left). The operator
concluded that other wells in this geographic
vicinity were susceptible to signicant metal loss
and at risk of casing leaks in this depth interval.
This conclusion may guide completion decisions
for future wells drilled in the area, which could
include landing the outermost casing string
Oileld Review
typically 13 3/8-in. casingdeeper than in the previous wells. The original landing depth of 700 ft
[213 m] could be extended to a depth of 3,000 ft
[914 m] to provide an additional layer of corrosion
protection to the inner string. Another solution
could be to add a further level of protection by running chrome alloy or coated 13 3/8-in. casing from
1,000 ft [300 m] to 3,000 ft.
The metal loss proles from these wells also
may inuence the operators decision to implement more cost-effective and efcient workovers
for repairing leaks. For example, the operator
could reduce workover costs by running a cement
squeeze limited to the depth of signicant metal
loss rather than incurring additional costs of a
liner, casing patch or scab liner, which might be
normally recommended if massive metal loss covered a long interval.20
In addition to the acoustic and electromagnetic monitoring techniques discussed, mechanical methods are also helpful. A multinger
mechanical caliper tool uses a fundamentally different approach. Caliper tools rely on direct
physical contact with the pipe wall to make measurements and to detect small changes in the
tubular wall such as deformations arising from
the buildup of scale or metal losses from corrosion. While they are well established for evaluating internal problems, caliper tools provide no
data regarding the condition of the external wall.
The Schlumberger PipeView multinger caliper tool for PS Platform toolstring has been
deployed to investigate corrosion in many types
of wells but particularly in those with excessive
scale and corrosion in which acoustic-based
tools cannot be run. The tool can be deployed
with 24, 40 or 60 ngers and used in casing
diameters ranging from 13/4 in. to 14 in. It provides a mechanical image of the internal tubular corrosion using 3D analysis and visualization
software (right).
Measurements over Time
Abu Dhabi Company for Onshore Oil Operations
(ADCO) deployed the PipeView tool to measure
corrosion over time in a well within a mature and
prolic eld. The well was originally drilled in
1969 and has been worked over many times.
During the most recent workover in 2006, a 7-in.
19. Because the production uids in these wells were
known to be noncorrosive and the tubing-casing
annulus contained diesel and corrosion inhibitor, any
measured metal loss was assumed to be external only.
20. A cement squeeze is a remedial operation designed to
force cement into leak paths in wellbore tubulars and
casing strings. Squeeze cementing operations are
performed to repair poor primary cement jobs, isolate
perforations or repair damaged casings or liners.
Autumn 2013
Motorized
centralizers
Calipers
Motorized
centralizers
PMIT-24 Fingers
PMIT-40 Fingers
> Multinger caliper tools. Multinger caliper tools measure the internal
diameter in casings. Numerous calipers, or ngers, pressed against the wall
of the pipe detect small changes in the pipe inner diameter that may be
interpreted as wear or corrosion. In general, multinger calipers come with
varying numbers of ngers; a higher number of ngers is required for larger
internal pipe diameters. The PipeView PS Platform new-generation
production services multinger imaging tool, the PMIT-24 ngers tool (left),
requires mechanical centralizers (not shown). The PipeView PMIT-40 ngers
tool (right) incorporates motorized centralizers. A third version, not shown,
has 60 ngers.
25
X,500
X,550
Depth, ft
50
40
30
20
10
0
Depth
1,000 ft
Y,000
Y,050
Depth, ft
50
40
30
20
10
0
500 ft
Depth
> Corrosion logs obtained from a caliper tool. PipeView service data (top left) and average metal loss versus depth (top right) were recorded in 2009, 2010
and 2011 for the tubing across the point of gas injection. The logs in Track 1 (top and bottom left) include a measurement for nominal internal radius (dashed
black line), nominal outer radius (dashed green line), eccentricity (dashed red line), minimum internal radius (solid blue line), maximum internal radius (solid
red line) and average internal radius (solid black line) above and below the point of gas injection (top and bottom left, respectively). Track 2 is a trace of
each caliper. Track 3 is an image log of thickness loss in the casing. Dark blue indicates the presence of scale, blue to white indicates 0% to 20% thickness
loss, white to pink is 20% to 40% loss and orange to red indicates 40% to 80% loss. Pure red (not shown) would indicate 100% loss and a hole in the casing.
The average metal loss above the point of gas injection (top right) did not change signicantly during the three years, suggesting that the presence of the
gas had a mitigating effect on corrosion. A similar plot for the tubing below the point of gas injection (bottom right) showed greater metal loss, which
increased over the three-year period, suggesting more-aggressive corrosion.
tieback liner was run and cemented to the surface to cover a corroded section of 9 5/8-in. casing.
The operator then drilled a single 5 7/8-in. horizontal well into a previously bypassed carbonate
formation. This lateral was completed as a gas
lift oil producer.21
Company engineers used naturally produced
gas with no corrosion inhibitor treatment as the
injection gas, which entered the system through a
gas lift side pocket mandrel. Concerned with the
corrosion potential posed by the injection gas,
ADCO engineers elected to run time-lapse monitoring surveys with the multinger imaging tool to
identify, quantify and track the growth of internal
corrosion in the tubing and estimate a corrosion
rate and time-to-failure. ADCO conducted surveys
over a three-year period2009 to 2011using a
111/16-in., 24-nger version of the tool.
26
drel did not contain lift gas, that section experienced a higher corrosion rate.
ADCO engineers are still speculating about
the exact inhibitor mechanism; one plausible
theory holds that the injected gas adds turbulence to the production ow and alters the ow
regime, which reduces water holdup and water
contact with the tubings internal surface. This
same phenomenon of less corrosion above the gas
injection point has been observed in other gas lift
wells in which caliper surveys were acquired. A
caliper log in a similar well, in combination with
a FloView holdup measurement, corroborates the
theory that gas injection may be reducing water
contact with the tubing (next page). The operator plans to use these results to rene the design
of future gas lift well completions to take advantage of this effect.
Oileld Review
Depth, ft
Gamma Ray,
2011
gAPI 100 0
Gamma Ray,
2010
gAPI 100 0
Gamma Ray,
2009
gAPI 100 0
100
100
100 0
Image View
Water Holdup,
Flow Volume
Corrected
0.9 0.4
Flow
regime
Cross-sectional
holdup
distribution
D,500
E,000
E,500
F,000
Above the
side pocket
mandrel,
gas breaks the
water/metal
contact.
F,500
G,000
Gas
injection
G,500
H,000
H,500
Below the
side pocket
mandrel, the
water/metal
contact is
stable.
I,000
I,500
J,000
> Changes to the water holdup prole. A caliper log run in combination with the FloView water holdup probes in an ADCO well
shows increasing corrosion over time (Track 2) below the point of gas injection and very little corrosion above the gas
injection point. This phenomenon is attributed to a decrease in water holdup above the gas entry point. Analysts believe there
is increased gas within the ow regime (right, red dots), which also includes signicant water (blue) and oil (green dots).
Water holdup, corrected for owline volume (Track 4), is reduced in the upper section; the corrosion rate is less in the upper
section than in the lower section, in which less gas is present. Water holdup is imaged (Track 3); blue represents water and
red represents oil and gas.
Autumn 2013
completed in September 2001 as a single producerengineers discovered a leak in the wellbore. To locate the leak zone by quantifying the
21. Gas lift is an articial lift completion method in which
gas is injected into the production tubing to reduce the
hydrostatic pressure of the uid column in the well and
the bottomhole pressure. This method allows reservoir
liquids to enter the wellbore at a higher ow rate.
27
in.
1.9
Minimum
Internal Radius
1.4
in.
1.9
Depth, ft
Well Schematic
Casing Collar
Locator Depth
8
in.
1.9
in.
1.9
1 1.4
in.
1.9 0.8
Tubing Pseudothickness
0.649 in.
in.
0.4
0.8
0.254 in.
degree
Double Coil B
Long-Spacing Phase
0.2
in.
360
in.
X,000
100%
metal loss
Y,000
Z,000
> Side-by-side comparison. Logs from the PipeView multinger caliper tool and the EM Pipe Scanner casing inspection tool run through the interval that
contained 3 1/2-in. and 9 5/8-in. casing strings showed damage and holes in the 3 1/2-in. tubing and showed metal loss (Tracks 4, 5 and 6, green shading) on the
9 5/8-in. casing, including one section that suffered 100% metal loss.
28
Based on these ndings, KOC pulled the tubing to conrm the damage. The processed caliper
log and a photograph of the damage from the tubing show a direct correlation between the corrosion measurements and the location of the
damage (next page, left). The results of this survey gave KOC condence that it could accurately
measure corrosion and identify a leaking interval
behind the tubing in wells in the future without
having to pull the tubing out of the hole.
Qatar Petroleum also implemented a combined corrosion measurement strategy in a well
in an offshore eld. The well, which was drilled in
Oileld Review
Eccentricity
0
in.
in.
20-in.
casing shoe
in.
in.
Excentralization
0
in.
in.
13 3/8-in.
casing shoe
Relative Bearing
degree
360
3 1/2-in. tubing
Collar
9 5/8-in. casing
shoe
7-in. liner shoe
Internal Radius
Minus Average PMIT-A
in.
0.08
in.
0.08
1.9 0.16
in.
0.44
Caliper, in.
1.500 1.625
1.750
> Processed caliper log. The eld logs from the PipeView multinger caliper tool of the interval with
3 1/2-in. and 9 5/8-in. casing (top) correlated precisely with the visual damage observed in the retrieved
tubing (bottom right). The caliper log (top left) includes measurements for eccentricity (dashed red
line), average internal radius (solid black line), maximum internal radius (solid red line), minimum
internal radius (solid blue line), excentralization (dashed black line) and nominal outer radius (dashed
green line). The caliper log (top center) is composed of three traces that indicate casing collars used
for depth correlation (horizontal red line). The image log (top right) in the casing indicates thickness
losses. Dark blue indicates scale, blue to white indicates 0% to 20% metal loss, white to pink is 20% to
40% loss and orange to red indicates 40% to 80% loss. Red (circled) indicates 100% loss and a hole
in the casing. A 3D processing image (bottom left) based on multinger caliper data also indicates
strong correlation with the damage observed in the retrieved tubing, as do the processed logs
(bottom center).
Autumn 2013
29
Water
Water
Annulus
Annulus
Ultrasonic Azimuth
0
degree
0.5
Wave Amplitude
Minus Maximum
Cable Speed
Casing
Casing
Maximum Internal
Radius
Maximum Internal
Radius
Bonded
Microdebonding
ft/h
1,000 6
Motor Speed
6
RPM
8 0
Depth, ft
Amplitude of
Eccentricity
0
in.
0.5 0
Casing Collar
Locator, Ultrasonic
20
in.
dB
0.5
Wave Maximum
Amplitude
20 0
dB
5
Maximum Internal
Radius
100 4.2
in.
4.7 5
Wave Minimum
Amplitude
Average Internal
Radius
dB
in.
100 4.2
Minimum Internal
Radius
dB
in.
3.5 5
in.
3.5 5
in.
3.5
in.
in.
in.
3.5
Average External
Radius
Average External
Radius
in.
in.
4.7 5
3.5 5
Microdebonding Ratio
Minimum Thickness
3.5
Average Internal
Radius
3.5 5
Liquid
Minimum Internal
Radius
Average Internal
Radius
4.7 5
Wave Average
Amplitude
100 4.2
in.
Minimum Internal
Radius
0.1
Internal Radii
Minus Average
in.
0.6
Maximum Thickness
0.1
in.
0.6
Average
Casing Thickness
Cement Acoustic
Impedance
0.08 0.1
in.
0.6 0.08
Microdebonding
Gas
Measurements/Total
Average Thickness
3.5 0.08
0
Cement
Measurements/Total
0.08 0
Mrayl
8.0 1
0 2.0
Mrayl
8.0
300
310
320
330
340
350
360
No severe metal
loss detected
Top of cement
370
380
390
400
410
420
430
440
> Top of cement. Engineers used logs from the USI tool to accurately locate the top of the cement behind the 9 5/8-in. casing (Tracks 10 and 11); standard USI
tool data indicated that the 7-in. liner and the 9 5/8-in. casing were free of signicant corrosion.
retrieval (above). They could then directly evaluate the 13 3/8-in. casing for corrosion defects.
Engineers next deployed the EM Pipe Scanner
tool to evaluate the external casing strings.
Despite the fact that measurements were made
outside of the recommended specications, the
tool identied an anomaly at a depth above the
seabed; the amplitude level across the anomaly
was high, and the phase level was low, both of
which suggested that less metal was present at
the anomaly than would be expected under normal circumstances. This information reinforced
the interpretation of the annular pressure test
30
Qatar Petroleum had performed several workovers on another offshore well in the eld and is
using the well as a dump ooder, in which produced water is injected into another formation.
Because the injected water is untreated, the production casing regularly experiences corrosion.
The well was originally cased with three sets
of steel casing: a 20-in. surface casing, a 13 3/8-in.
intermediate casing and a 9 5/8-in. production casing. After corrosion problems were detected in
2002, engineers overlapped the production casing with 7-in. casing. The well is perforated in one
formation from 6,290 to 6,320 ft [1,918 to 1,926 m]
Oileld Review
Autumn 2013
Caliper, in.
6.250 6.359 6.469 6.578 6.688
74
Metal Loss
78
Double Coil B
Long-Spacing Phase
20-in. casing
40
13 3/8-in. casing
degree
82
400
86
Double Coil B
Long-Spacing Amplitude
Depth, ft 60
dB
90
94
Depth, ft
9 5/8-in. casing
90
98
102
106
110
6.688
100
Caliper, in.
6.578
114
6.469
118
6.359
6.250
122
> Signicant metal loss. Even though the EM Pipe Scanner tool was run outside of its specied range
for amplitude and phase, it detected signicant metal loss across the three casing strings (dashed red
box, top left). The 3D (top center) and 2D (top right) visualizations produced from the PipeView
multinger caliper tool log run in the 13 3/8-in. casing indicate the casing had corroded to the point
that it had parted (bottom left and right) at the depth where the EM Pipe Scanner tool had detected
metal loss.
data recorded from various monitoring techniques and analyzes such data for their impact
on production. The software ags those areas
with higher corrosion rates or a history of corrosion-related events, and as a result, the operator
can prioritize corrosion mitigation efforts and
implement suitable preventive measures.
As the industry moves into more-aggressive
corrosion environments and technically demanding production regions, corrosion monitoring
advances such as these must continue to expand
and evolve if operators are to remain both protable and environmentally responsible.
TM
31
32
Plumb bob
Disk
Oileld Review
Historical Perspective
Traditionally, wellbores were drilled vertically
and were widely spaced. Well spacing decreased
as elds matured, regulations tightened and reservoirs were targeted in remote areas. Over time,
drilling multiple horizontal wells from a single
pad became common practice. Today, more than
a dozen wells may fan out into the reservoir from
a single offshore platform or onshore drilling pad.
Pad drillinggrouping wellheads together at
one surface locationnecessitates fewer rig
moves, requires less surface area disturbance and
Autumn 2013
33
West
P
X
Geogra
north phic
D
I
Y
East
M
nor agneti
th
c
F
c fi
neti
Mag
or
vect
eld
Z
Down
> Magnetic eld orientation. At any point P, the magnetic eld vector (red) is
commonly described in terms of its direction, its total magnitude, F, in that
direction and H and Z, the local horizontal and vertical components of F. The
angles D and I describe the orientation of the magnetic eld vector. The
declination, D, is the angle in the horizontal plane between H and
geographic north. The inclination, I, is the angle between the magnetic eld
vector and the horizontal plane containing H. Of these measurements, D and
I are required to convert the compass orientation of a wellbore to its
geographic orientation. The absolute magnitudes of F, Z or H are used for
quality control and calibration.
34
Oileld Review
Autumn 2013
1,000 ft
1,000 ft
200 ft
X
Y
35
10
10
10
10
20
10
10
20
30
40
20
20
30
> Values of declination along lines of equal declination (isogonic lines) of the Earths magnetic eld. In
the areas surrounded by red lines, or the lines of equal positive declination, a compass points to the
east of true north. Lines of equal negative declination, for which the compass points to the west of
true north, are blue. Along the green, agonic lines, for which declination equals zero, the directions to
magnetic north and true north are identical. The eld shown is the International Geomagnetic
Reference Field for the year 2010. [Adapted from Historical Main Field Change and Declination,
CIRES Geomagnetism, http://geomag.org/info/declination.html (accessed June 24, 2013).]
36
Oileld Review
400
150
90
70
Alaska
50
P a c
i f i
c
C A N A D A
30
20
10
0
10
20
30
40
60
80
Year 2010
125
175
300
> Geomagnetic crustal eld. Airborne measurements of the strength of the magnetic eld provide data
that are used to determine the anomalous contribution from earth crustal materials. The total intensity
anomaly (TIA) is the difference between the magnitude of the total eld and that of the main magnetic
eld. The TIA eld over western Canada; Alaska, USA; and the northwest continental US varies from
300 nT (blue) to +400 nT (pink). The mean total eld strength is about 55,000 nT in this region. The
crustal eld shows local intensity ridges, with variation on a much ner spatial scale than that of the
main magnetic eld. [Adapted from Magnetic Anomaly Map of North America, USGS, http://mrdata.
usgs.gov/geophysics/aeromag-na.html (accessed July 23, 2013).]
Magnetopause
Magnetosheath
Bow shock
Autumn 2013
Solar wind
Earth
Van Allen
radiation belts
Magnetotail
> Distortion of the Earths magnetosphere from the solar wind. The sun emits a ux of particles, called
the solar wind, which consists of electrons, protons, helium [He] nuclei and heavier elements. The
Earths magnetic eld is conned by the low-density plasma of the solar wind and the interplanetary
magnetic eld (IMF) that accompanies it. These distort the Earths magnetic eld away from its dipolar
shape in the magnetosphere, the extensive region of space bounding the Earth. The eld becomes
compacted on the sunward side and elongated on the opposite side. The solar wind produces a
variety of effects, including the magnetopause, radiation belts and the magnetotail. Time-varying
interactions of the magnetosphere with the solar wind produce magnetic storms and the external
disturbance eld.
37
Bm
Bc
Bobserved
B int
B int
> Contributions to the total observed magnetic eld. During periods of solar
quiet, the discrepancy between the observed eld, Bobserved (red), and the
main magnetic eld, Bm (green), is largely due to the local crustal eld Bc
(blue) and the drillstring interference, Bint (yellow). At other times, the
external disturbance eld also makes a contribution. (Adapted from
Poedjono et al, reference 30.)
The external magnetic eld exhibits variations on several time scales, which may affect the
applicability of magnetic reference models.17 Very
long-period variations are related to the solar
cycle of about 11 years. Short-term variations
arise from daily sunlight variation, atmospheric
tides and diurnal conductivity variations.
Irregular time variations are inuenced by the
solar wind. Perturbed magnetic states, called
Model
Organization
Order
Resolution, km
Update Interval
WMM
12
3,334
5 years
IGRF
IAGA
13
3,077
5 years
BGGM
BGS
50
800
1 year
720
56
> Magnetic eld reference models. Several groups and organizations have
developed reference models of differing resolution; the models are updated
at various intervals. In the Order column, order increases with the
complexity of the model and in this case refers to spherical harmonic
models. These models construct the global magnetic eld as a sum of terms
of varying order and degree. Terms of order n have a total of n circular
nodal lines on the sphere at which the magnetic eld contribution is zero.
The orientation of the lines depends on the combination of order and
degree. Resolution corresponds to the wavelength of the highest order term.
38
Oileld Review
The World Magnetic Model (WMM) characterizes the long-wavelength portion of the magnetic
eld that is generated in the Earths core; it does
not represent the portions that arise either in the
crust and upper mantle or from the disturbance
eld generated in the ionosphere and magnetosphere.21 Consequently, magnetic measurements
may show discrepancies when referenced to the
WMM alone. Local and regional magnetic declination anomalies occasionally exceed 10, and declination anomalies on the order of 4 are not
uncommon but are usually of small spatial extent.
To account for secular variation, the WMM is
updated every ve years. An international task force
formed by the IAGA has released International
Geomagnetic Reference Field IGRF-11, a series of
mathematical models of the Earths main magnetic
eld and its rate of change. These models have resolution that is comparable to that of the WMM.22
Directional drilling requires higher resolution models than WMM or IGRF alone. The BGS
Global Geomagnetic Model (BGGM), widely used
in the drilling industry, provides the main magnetic eld at 800-km [500-mi] resolution and is
updated annually.23 The Enhanced Magnetic
Model (EMM) improves greatly on this spatial
resolution. The EMM and a successor, the HighDenition Geomagnetic Model (HDGM), resolve
anomalies down to 56 km [35 mi], an order of
magnitude improvement over previous models.
By accounting for a larger waveband of the geomagnetic spectrum, the HDGM improves the
accuracy of the reference eld, which in turn
improves the reliability of wellbore azimuth
determination and enables high-accuracy drillstring interference correction.24
Autumn 2013
39
> Plan view of wellbore trajectories, looking down. PGE used a multiwell
pad design for 14 wells drilled into the Marcellus Shale from a single pad.
The plan shows initial uncertainty disks at true vertical depths of 2,500 ft
(red) and 5,000 ft (yellow). As expected, uncertainty grows larger with
increasing distance from the surface location and can impact the drilling
program. None of the red disks intersect each other, nor do the yellow
disks, indicating that the wellbores (blue) are clear of each other at those
depths. (Copyright 2010, SPE Eastern Regional Meeting. Reproduced with
permission of SPE. Further reproduction prohibited without permission.)
able to incorporate disturbances caused by diurnal solar activity and magnetic storms into survey
data processing.
The technique of ineld referencing (IFR)
makes use of data from local magnetic surveys
near a wellsite to characterize the crustal magnetic eld. Service companies have developed
> Pad design and well trajectories. PGE drilled 14 wells into two reservoirs during Phases 1 (magenta)
and 2 (blue) of the drilling campaign. The graphical size of each wellbore corresponds to the size of
the EOUs as dened in the survey program. The drilling team conrmed the anticollision condition. At
the reservoir entry point, each well needed to have a minimum 200-ft [60-m] separation from its
counterpart drilled in the opposite direction. (Copyright 2010, SPE Eastern Regional Meeting.
Reproduced with permission of SPE. Further reproduction prohibited without permission.)
40
Oileld Review
Autumn 2013
Y-axis
0m
0m
400 m
400 m
800 m
1,200 m
800 m
1,600 m
1,200 m
2,000 m
2,400 m
1,600 m
2,800 m
2,000 m
3,200 m
Z-axis
2,400 m
3,600 m
2,800 m
6,000 m
5,000 m
X-axis
3,200 m
4,000 m
3,600 m
3,000 m
6,000 m
5,000 m
4,000 m
3,000 m
2,000 m
Y-axis
1,000 m
0m
> Hitting distant targets with an extended-reach well in the Jeanne dArc basin, offshore Canada. This
well trajectory (center) extends approximately 7,000 m [23,000 ft] before dropping to hit two targets
(red) at about 4,000 m [13,000 ft]. Insets (top and bottom) show close-up views of the targets and the
ellipsoids of uncertainty (EOUs) for two survey methods. The positional uncertainty (green) of the
magnetic surveys without GRS (top) is so large that the well may be outside the targets. With GRS
(bottom), the positional uncertainty (blue) is well within the size of the targets. (Adapted from Poedjono
et al, reference 27. The images in this gure are copyright 2010, IADC/SPE Drilling Conference and
Exhibition. Reproduced with permission of SPE. Further reproduction prohibited without permission.)
For a successful drilling program, the operator required an accurate description of positional
uncertainty and a small error ellipsoid. The GRS
guided drilling program met these requirements
and provided extended drillability, reduced drilling time and improved chances of hitting the geologic target (above).
High Precision in High Latitudes
Geomagnetic referencing brings signicant
advantages but encounters its greatest challenge
when applied at high latitude, where the magnitude of geomagnetic disturbance eld variations
41
Start
Geomagnetic
referencing
processing
DED observatory
adjusted data
QA/QC by USGS
Pass
QA/QC?
Real-time GRS
azimuth correction
to drill ahead
QA/QC
remove data with
external interference
No
Nearby well?
End of
bit run?
Yes
No
Yes
No
QA/QC
calibration?
failed sensor?
Yes
Final GRS
definitive surveys
final GRS report
Total depth
reached?
Yes
No
Sectional GRS
definitive surveys
Stop
> Geomagnetic referencing workow. The workow starts with raw MWD and magnetic observatory
data streams (shown here as from the DED observatory) and combines them with crustal magnetic eld
data then progresses through geomagnetic processing, data adjustment and quality control.
Processing continuously generates directional drilling corrections and provides denitive surveys at
the end of bit runs. (Adapted from Poedjono et al, reference 32.)
Data
Reference
58,239
58,039
nT
57,839
57,639
57,439
57,239
57,039
10,500
11,776
13,052
14,328
15,605
16,881
18,157
19,433
20,710
Depth, ft
Data
Reference
58,826
58,479
nT
58,132
57,785
57,438
57,091
56,744
10,500
11,776
13,052
14,328
15,605
16,881
18,157
19,433
20,710
Depth, ft
> Time-varying reference data. Raw magnetic MWD survey data (top, blue) initially exceeded the data
quality acceptance limits (red) at several depths, but the data passed when referenced to DED
observatory data (bottom). Initial acceptance limits were based on a static reference value (top, green)
for the local magnetic eld strength, whereas the DED data provided actual time-varying values
(bottom, green) to which the limits could be referenced. (Adapted from Poedjono et al, reference 32.)
42
Oileld Review
Autumn 2013
Williston Basin
1 mi
1 mi
95
0f
> Field development plan. In a eld in Montana and North Dakota, USA, operators started eld
development with one well per 1-mi2 [640-acre, 2.6-km2] parcel. Alternate rows of injector (blue) and
producer wells (gray) show planned down-spacing to an interwell spacing of 950 ft [290 m] (red box).
Positional uncertainty needs to be minimized to keep the well trajectories parallel and reduce the risk of
premature breakthrough of water from the injectors. (Adapted from Landry et al, reference 37.)
43
Plan View
Well-to-well separation
at surface location
Surface location
Well 3
Surface location
Well 1
Surface location
Well 2
Survey Program B
provides separation at TD.
Separation at
measured depth
Survey Program A
does not provide
separation at TD.
> Strategies for ensuring optimal spacing to prevent water breakthrough. Survey Program B (pink)
delivers higher accuracy than Survey Program A (blue). Had Wells 1 and 2 been drilled from adjacent
surface locations using Survey Program A, the wells may have collided at TD. Survey Program B, with
compensation for magnetic interference, ensures noncollision and allows the wells to be extended to
planned total depth. By staggering one wellhead to the surface location of Well 3, the operator could
increase well separation at total depth, drill wells with the desired orientation and spacing and prevent
early water breakthrough. The operator chose to use both Survey Program B and wellhead staggering.
(Adapted from Landry et al, reference 37.)
23.8
23.4
2048
23.4
22
23.4
23.2
2136
22.2
Latitude
Field
23
2200
23.2
2224
23
23.2
23.6
2112
23
23.2
2248
24
4000
3936
4000
3936
4000
3936
Longitude
> Magnetic eld declination maps offshore Brazil. The standard model (left) shows smooth, largescale variations in magnetic eld declination in the vicinity of the hydrocarbon eld (red polygon). The
higher resolution HDGM (center) includes more detail. The combined HDGM and aeromagnetic survey
model (right) contains the highest resolution information of all three models. All maps show declination
at mean sea level. Differences of nearly 1 in declination are observed between the standard and
highest resolution models near the eld. (Adapted from Poedjono et al, reference 38.)
44
Crustal Variations
In some situations, the main concern is not the
time-varying eld but the crustal correction.
Such was the case for one operator in a deepwater heavy-oil eld offshore Brazil.38 The project
lies in 1,100 m [3,600 ft] of water in the northern
Campos basin. The operator had drilled several
wells using MWD and had observed discrepancies
between downhole tool readings and those
expected from the BGGM. To improve magnetic
surveying here, it was necessary to develop a better model of the local magnetic eld so that wellbore trajectories would attain their targets. The
company needed to employ a highly accurate geomagnetic model to avoid eld acceptance criteria
failures in real-time drilling. Such failures may
lead to unnecessary tool retrieval operations
because of suspected tool failure.
To resolve the survey discrepancies, a
research team composed of representatives from
the operator, Schlumberger, other contractors
and academia developed a method for mapping
the magnetic variations using the High-Denition
Geomagnetic Model (HDGM2011), which had
recently been developed at the US NGDC. The
team integrated this large-scale magnetic eld
model with data from a local aeromagnetic survey to extend the spatial spectrum of the magnetic eld from regional scales down to the
kilometer scale (left).
The team used two independent methods to
analyze the crustal magnetic model.39 Method 1
combined the BGGM with aeromagnetic survey
data and employed an equivalent source method
for downward continuation of the eld to reservoir depth. Method 2 combined the aeromagnetic
Oileld Review
Autumn 2013
2.0
1.8
1.6
0.4
0.2
0.6
1.4
10,000 m
0.8
1.2
1.0
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
10,000 m
Crustal Contribution at 5,000 m
2.0
0.8
1.8
0.6
0.4
1.6
1.2
1.4
1.4
0.2
0.2
10,000 m
1.2
1.0
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.6
10,000 m
> Crustal contribution to the magnetic eld declination at two depths in the vicinity of a eld
offshore Brazil. The crustal eld contribution to the magnetic declination is shown in plan
view at mean sea level (top) and at a depth of 5,000 m (bottom). Values were calculated using
a method that combined an aeromagnetic survey with a long-wavelength crustal eld model
provided by the German CHAMP satellite survey; the method then created a 3D magnetic
model for the lease area. The 3D magnetic eld changes with depth, in large part because of
the magnetic properties of the Earths crust underlying the sediments offshore Brazil.
(Adapted from Poedjono et al, reference 38.)
45
Units
Survey Data
Meters
Geomagnetic Reference
GeoMag Field
Declination
4.5900 deg
90.00 deg
Grid Convergence
0.2100 deg
Total lBl
4.3800
Region
Geographic Region
Feet
Inclination
Azimuth Reference
Check Reference
True North
Grid North
49895 nT Dip
61.75 deg
25
D2 100.51 ft
Medium (6.75 and all Medium sizes)
L2
Large (8 OD or More)
BHA Conguration
52.78
ft
D1
36.02 ft
ft
S2
ft
L1
ft
S1
ft
Type
D2
MP
D1
Steel
Results
Azimuth Error:
MP
NMR
4.22 deg
Interfering Field:
1746 nT
Calculate
33 nT
Report
0.07 deg
NonMag
0.0 ft
Add ..
Below MWD
0.0 ft
Undo Last
Open
Save
Save As ..
Clear
Exit
> Quantifying magnetic measurement sensitivity to toolstring interference. Modeling codes are used
to simulate the extent of magnetic interference for various survey orientations and BHA designs.
This simulation, taken from the Schlumberger Drilling & Measurements Survey Tool Box, shows the
large azimuthal error (red) that would occur at this particular wellbore grid azimuth of 270 and
inclination of 90 if the driller did not add nonmagnetic spacing material to the BHA in addition to
that included in the initial design (blue). Drilling engineers use these simulations to determine the
length of nonmagnetic material above or below the MWD measure point necessary to reduce the
error sufciently.
Deepwater Success
Accurate real-time magnetic surveys allow directional drillers to stay on path and to reduce the
number of required conrmatory gyro surveys.
Tullow Ghana Ltd. used geomagnetic referencing
to achieve its objectives to hit distant geologic
targets accurately and within budget while developing the Jubilee eld offshore Ghana.40
The operator wanted to drill all wells safely
and successfully in the shortest possible time
because rig spread costs are exceptionally high in
this area. To enable accurate GRS, Schlumberger
surveying experts conducted numerical
simulations, which quantied the sensitivity of
46
Oileld Review
512,000 m
511,800 m
X-axis
X-axis
515,000 m
514,000 m
513,000
Y-axis
0m
512,000511,600 m
400 m
400
800 m
800
1,200 m
1,200
1,600 m
512,000 m
Z-axis
1,600
2,000 m
X-axis
511,800 m
2,000
511,600 m
2,400 m
2,400
2,800 m
2,800 m
3,200 m
3,200 m
515,000 m
514,000 m
Y-axis
513,000 m
512,000 m
X-axis
> An extended-reach well in the Jubilee eld offshore Ghana. The Tullow Ghana Ltd. Well 4 has a long step-out and tangent
prole to hit the target (red). The EOU from standard MWD (top left, green) is larger than the rectangular geologic target.
Because of the smaller EOU from GRS (center left, blue), the operator was able to drill the well with high condence that the
wellbore would penetrate the target. (Adapted from Poedjono et al, reference 40. The images in this gure are copyright
2011, SPE EUROPEC/EAGE Annual Conference and Exhibition. Reproduced with permission of SPE. Further reproduction
prohibited without permission.)
For the initial wells in the Jubilee eld, standard MWD surveys yielded small enough EOUs
to hit the geologic targets with condence.
These initial well paths had relatively shallow
inclination angles. For more-distant targets
with higher inclination angles and longer tangent sections, the uncertainty associated with
standard MWD surveys was unacceptably large.
However, uncertainty was considerably smaller
for GRS processed magnetic data, and drillers
reached their objectives with high condence.
Using GRS, the operator was able to drill the
well with guaranteed placement of the wellbore
inside the target (above).
Autumn 2013
aid processing for drillstring interference compensation and enhance measurement quality
control by employing customized acceptance
criteria. Geomagnetic referencing improves
well placement accuracy, reduces positional
uncertainty and mitigates the danger of collision with existing wellbores. When used in realtime wellbore navigation, GRS saves rig time,
reduces drilling costs and helps drillers reach
their targets.
HDL
47
Anatoly Arsentiev
Irkutsk, Russia
David H. Hathaway
National Aeronautics and Space Administration
(NASA) Marshall Space Flight Center
Huntsville, Alabama, USA
Rodney W. Lessard
Houston, Texas, USA
Oileld Review Autumn 2013: 25, no. 3.
Copyright 2013 Schlumberger.
For help in preparation of this article, thanks to
Don Williamson.
1. Cliver EW: The 1859 Space Weather Event: Then and
Now, Advances in Space Research 38, no. 2 (2006):
119129.
2. Boteler DH: The Super Storms of August/September
1859 and Their Effects on the Telegraph System,
Advances in Space Research 38, no. 2 (2006): 159172.
3. Stephens DL, Townsend LW and Hoff JL: Interplanetary
Crew Dose Estimates for Worst Case Solar Particle
Events Based on Historical Data for the Carrington Flare
of 1859, Acta Astronautica 56, no. 912 (MayJune 2005):
969974.
48
Those of us in the energy industry owe our livelihoods to the Sun. The hydrocarbons we search for
and produce were formed from organic matter
that stored ancient energy that originated within
the Sun. In the not too distant past, the Sun was
an object of reverence because of its control over
our lives. Today, familiarity with and understanding of the Sun has removed much of our sense of
veneration; however, we understand that our very
existence is based on a relationship to the seemingly unchanging presence of the solar systems
shining star.
On occasion, however, the Suns apparent stability is interrupted by powerful displays of its
dynamism. One such example occurred on the
morning of September 1, 1859. From his private
observatory, amateur astronomer Richard
Carrington observed a cluster of large spots on
the surface of the Sun. Suddenly, a brilliant ash
of white lighta solar areerupted from the
area of the spots.1 This particular are was the
harbinger of a gigantic coronal mass ejection
(CME), which spewed solar plasma into interplanetary space.
This massive cloud of charged particles
arrived at Earth in less than 18 hours. It proceeded to disrupt the most advanced technology
of the daythe telegraph.2 The interaction
between the CME and the Earths magnetic eld
induced electrical currents in exposed telegraph
wires. Current raced through the wires, causing
some of them to overheat, fall to the ground and
set off res. Telegraph machines were hit by pow-
Oileld Review
Autumn 2013
49
250
Quebec blackout
Carrington event
Sunspot number
200
150
100
50
0
1750
2
1770
4
1790
6
1810
1830
9
1850
10
11
1870
12
13
1890
Date
14
15
1910
16
1930
17
18
1950
19
20
1970
21
22
1990
23
24
2010
> Sunspot cycles. Scientists have systematically recorded the number of sunspots and numbered the sunspot peaks dating from the 1700s. In several
recent cycles, sunspot counts approached or exceeded 200; the current cycle average count is less than 100.
50
Oileld Review
Autumn 2013
Convection zone
Radiative zone
Core
Photosphere
Prominence
Sunspots
Flare
Coronal hole
Chromosphere
Corona
> The Suns structure. Fusion reactions take place in the Suns central core. The pull of gravity
accelerates hydrogen nuclei inward, toward the Suns center, where they fuse and form helium;
the reaction releases energy. The energyin the form of photons and other elementary particle
by-productsrises through the Suns radiative and convection zones and then exits from the photosphere.
The corona is the Suns outer atmosphere, a layer of plasma surrounding the chromosphere. Features
displayed on the Suns surface seen here include a prominence, solar ares, sunspots and a coronal
hole. [Illustration courtesy of the US National Aeronautics and Space Administration (NASA).]
> The Suns magnetic eld lines. Convoluted magnetic eld lines (green) may
extend thousands of kilometers out from the surface of the Sun. (Image
courtesy of the NASA Goddard Space Flight Center Scientic Visualization
Studio.)
51
Magnetotail
Plasmasphere
Plasma sheet
Solar wind
Van Allen
radiation belts
> Earths magnetosphere. The magnetosphere, the area of space around the Earth created by Earths
magnetic eld, is a dynamic structure that responds to variations in solar activity and space weather.
Solar wind, which compresses the sunward side of the magnetosphere, determines its shape. A
supersonic shock wavethe bow shockforms on the sunward side of Earth. Most of the solar wind
particles are slowed at the bow shock and directed around the Earth in the magnetosheath. The solar
wind pulls at the magnetosphere on the Earths night side, extending the length of the magnetosphere
up to 1,000 Earth radii, creating what is known as the magnetotail. The outer boundary of Earths
conned geomagnetic eld is called the magnetopause. Trapped charged particlesthe Van Allen
radiation belts, the plasmasphere and the plasma sheetreside within the magnetosphere. (Adapted
from an image courtesy of Aaron Kaase, NASA Goddard Space Flight Center.)
Penumbra
Penumbra
Umbra
Umbra
> Sunspots. Regions on the Sun that appear darker than the rest of the disk, sunspots are formed by
concentrated magnetic elds that project through the hot gases of the photosphere out to the Suns
surface. These magnetic elds create cooler, darker regions called sunspots. The dark center of a
sunspot is called the umbra; the light area around the umbra is the penumbra. Sunspots occur in
groups and frequently in pairs. The two spots in a pair have opposite magnetic polarities.
(Photographs courtesy of NASA.)
52
Oileld Review
Prominence
> Solar prominence photographed on September 23, 1999. The space-based Solar and Heliospheric
Observatory (SOHO) captured this image of an eruptive prominence using extreme ultraviolet
frequencies. The release of energy from twisted magnetic eld lines ings plasma above the Suns
surface. [Photograph courtesy of the SOHO Extreme Ultraviolet Imaging Telescope (EIT) consortium.]
> Solar are. The NASA Solar Dynamics Observatory (SDO) captured this image of a solar are on
May 22, 2013. The image captures light in the 13.1-nm wavelength, which highlights material heated to
intense temperatures during a are. The teal coloration is typical of images using this wavelength.
(Photograph courtesy of the NASA SDO.)
Autumn 2013
53
> Auroras at high latitudes. Charged particles from solar wind and geomagnetic storms follow the Earths magnetic eld lines and can ionize gases in Earths
upper atmosphere. Ionized oxygen molecules emit green to brownish-red light; ionized nitrogen emissions are blue or red. The aurora borealis (left) was
photographed from the International Space Station over the Midwest US on January 25, 2012. The photograph of the aurora australis (right) captured by the
NASA IMAGE satellite on September 11, 2005, was taken four days after a solar are. The aurora encircles the South Pole and would appear as a curtain of
light if observed from ground level. (Photographs courtesy of the NASA International Space Station and IMAGE Science Center.)
8 minutes. A day or two later, high-energy particles may also arrive at the Earth, producing auroraslights in the polar night skiesand
affecting radio communications (above).15
Suns
diameter
> CME image captured from space on October 22, 2011. The Large Angle and Spectrometric
Coronagraph (LASCO), on board the NASA SOHO satellite, captured this image in which plasma was
hurled in the direction of Mars. The Sun is obscured by a disk that allows the instruments sensor to
focus on the emissions from the Suns surface, which enhances the observation of the corona by
blocking direct light from the Sun. The white circle on the disk represents the size and location of the
Suns surface. (Photograph courtesy of the SOHO EIT consortium.)
54
During some solar ares, a more violent reaction may occura coronal mass ejection (below
left). When the twisted magnetic eld lines cross,
their stored energy explodes outward with tremendous force. A CME occurs when the force of
the released energy ings a mass of superheated
plasma from the Suns surface into space.
CMEs vary in intensity and magnitude. A large
CME can contain 9 1012 kg [20 1012 lbm] of
matter that may be accelerated into space at several million kilometers per hour. The speed at
which the plasma travels depends on the original
energy release. A high-energy CME can arrive at
the Earth in as little as 16 hours, but lower-energy
releases may take days to make the journey.
Upon impact by a CME, the Earths magnetosphere temporarily deforms, and the Earths
magnetic eld is distorted. During these disruptions, Earth-orbiting satellites are exposed to
ionized particles, compass needles can behave
erratically and electrical currents may be
induced in the Earth itself. These eventsgeomagnetic stormscan disrupt technical infrastructure on a global scale. Because of the risks
associated with solar storms and CMEs, scientists constantly monitor space weather.
At a solar minimum, the estimated occurrence of a CME is about one event every ve days
compared with about 3.5 per day at a solar maximum. Although this may appear to put the planet
in frequent jeopardy, the probability that a CME
will be directed toward Earth is small. In comparison to the Sun and the expanse of the solar
Oileld Review
L4
Moon
Earth
L3
L1
L2
Sun
L5
> Lagrange points. Scientists have identied ve points (L1 through L5) associated with Earths orbit of
the Sun where satellites can maintain stable orbits. These locations, called Lagrange points (green),
are shown here with the gravitational potential lines (gray lines) established by the Sun-Earth system.
These positions in space correspond to regions where the gravitational forces of attraction (red
arrows) and repulsion (blue arrows) are in balance. The Wilkinson Microwave Anisotropy Probe
(WMAP) is located around position L2, which is about 1.5 million km [930,000 mi] from the Earth.
The WMAP spacecraft aligns with the Sun-Earth axis, similar to a geostationary orbit, but course
corrections are required to maintain its relative position. The illustration is not to scale. (Illustration
courtesy of the NASA WMAP Science Team.)
Autumn 2013
55
> The NASA Advanced Composition Explorer (ACE). Launched on August 25, 1997, the ACE satellite, a
crucial component of the NASA space weather monitoring eet, is stationed at Lagrange point L1.
From this position, the satellite records radiation emitted from the Sun, the solar system and the
galaxy. When bursts of solar material stream toward Earth, instruments on board ACE record the
increase in the number of particles and transmit this information to scientists on Earth who use these
data to warn of impending space weather events. Alerts and warnings are issued to relevant
organizations and posted online by the NOAA SWPC. (Illustration courtesy of NASA.)
Relative size
of Earth
> Space weather monitoring by SOHO. The SOHO satellite (right) was launched in December 1995.
SOHO is a joint project between the European Space Administration (ESA) and NASA to study the Sun
from its deep core to the outer corona and the solar wind. The satellite weighs about 17.8 kN [2 tonUS],
and its solar panels extend about 7.6 m [25 ft]. This solar eruption (left), which lasted four hours, was
photographed on December 31, 2012, by the Extreme Ultraviolet Imaging Telescope (EIT) in 30.4-nm
emission. Most of the plasma fell back to the Suns surface. The Earth is shown for scale. (Solar
photograph courtesy of the SOHO EIT consortium; satellite image courtesy of Alex Lutkus.)
56
Oileld Review
Filament
> The Solar Dynamics Observatory (SDO). The SDO satellite (left) was launched in February 2010 as part of the NASA Living with a Star Program, which
studies solar variability and potential impacts on Earth and space. By examining the solar atmosphere on small scales and capturing emissions at many
wavelengths simultaneously, the study hopes to determine how the Suns magnetic eld is generated and structured and how stored magnetic energy is
converted and released into the heliosphere and space. This image of the Suns magnetic eld lines (right), captured on June 4, 2013, was taken in extreme
ultraviolet light and highlights the bright coils of magnetic eld lines rising up in the background above an active region. A lament, which appears as a
darker region on the Suns surface, can also be seen. (Photograph and image courtesy of the NASA SDO.)
Autumn 2013
57
Typical auroral
zone location
Region
conductivity, S/m
CANADA
1 to 10
101 to 1
102 to 101
103 to 102
104 to 103
CANADA
MEXICO
UNITED
STATES
Highest risk
Medium risk
Connected power grids
MEXICO
> Power system susceptibility. Power systems in areas with the lowest ground conductivity (left, red and darkest yellow) are the most vulnerable to the
effects of intense geomagnetic activity. The high ground resistance beneath these areas facilitates the ow of geomagnetically induced currents (GICs) in
power transmission lines. Auroral zones for North America are susceptible to GICs because of their proximity to polar regions. (Data from the American
Geophysical Union and the Geological Survey of Canada.) For the US, scientists produced a map based on scenarios for existing power systems to
determine their vulnerability to geomagnetic storms (right). Should a storm 10 times larger than the 1989 storm that disrupted power systems in Quebec
arrive at Earth, the systems most at risk have been identied (red). The blue lines encircle the largest population centers served by at-risk systems.
(Adapted from the National Research Council of the National Academies, reference 16.)
58
in the New England region of the US was interrupted before power companies could bring other
reserves online.22
Damage caused by energized particles emitted from the Sun is not limited to terrestrial systems. Satellites, space exploration vehicles and
manned space missions can be affected by solar
emissions, some of which are too weak to enter
Earths magnetic eld. For instance, weak solar
ares and CMEs may produce solar proton events
(SPEs) that are mostly unnoticed on the surface
Oileld Review
Autumn 2013
264
263
Azimuth, degree
If the solar arrays that provide power to satellites are struck by high-energy protons from SPEs
and CMEs, the silicon atoms in the solar cell
matrix may shift positions, which increases the
internal resistance of the solar cells and reduces
electrical output. A single solar storm event can
decrease panel life expectancy by years. If attitude control systems on satellites used to correct
their orientation and position are damaged by
high-energy particle events, a satellite can lose
its orbital control, which may result in an
unplanned and premature reentry into Earths
atmosphere.23 Satellites play such a crucial role
in communications that a loss could affect television, cable programming, radio service, weather
data, cell phone service, automated banking services, commercial airline systems and GPS and
navigation services. Routine losses as a result of
satellite malfunction and premature asset failure
caused by solar storms are estimated in the billions of US dollars.
Consequences of solar storms may not be
limited to electrical damage. The July 1989
solar storm caused compression of the Earths
magnetosphere, reducing its typical depth of
more than 54,000 km [33,500 mi] to less than
30,000 km [18,640 mi], well inside the Earths
geosynchronous region where satellites orbit.
As the Earths atmosphere was bombarded by
energetic particles and compressed by the solar
wind, the density of the upper atmosphere
increased by a factor of 5 to 10. The increased
drag on low-Earth orbit satellites caused orbital
decaythe U.S. Air Force Space Command
reported losing track of more than 1,300 orbiting objects that fell to lower altitudes.24 In a
separate event, on March 13, 1989, NOAA
reported the loss of the GOES-7 weather satellite. Circuit problems caused by a shower of
energized particles rendered most of its systems
useless. Critical solar power arrays on GOES-7
lost 50% of their efciency. Engineers with
NASA reported many other satellites experienced electrical failures that temporarily shut
down onboard computers.25 The storm disrupted
communications on the Earth and between
ground controllers and orbital satellites.
Oil and gas pipeline and distribution systems
are also vulnerable. In the event of a geomagnetic
storm, operators may immediately lose supervisory control and data acquisition (SCADA) systems. Operators must also consider the long-term
effects associated with increased pipeline corrosion rates. Cathodic protection systems used on
pipelines to minimize corrosion maintain a negative potential with respect to the ground. During
solar storms, GIC events in a pipeline reduce the
Magnetic
storm
262
261
Drilling azimuth
Corrected azimuth
260
259
3,600
3,700
3,800
3,900
4,000
4,100
4,200
Depth, ft
4,300
4,400
4,500
4,600
4,700
> Geomagnetic storms and directional drilling. Directional drillers use MWD tools to determine drillbit
orientation and position; these measurements depend on data derived from magnetometers and
accelerometers. During geomagnetic storms, magnetometers may provide erroneous readings. A solar
storm occurred while an operator drilled a North Sea well, and the MWD drilling azimuth measurement
(blue) was affected by the geomagnetic storm. Engineers corrected the data using a technique
developed by the British Geological Survey that adjusts for space weather. The results provided a more
accurate well location (green). (Adapted from Clark and Clarke, reference 28.)
59
Sunspot numbers
250
Temperature
200
Dalton
Minimum
150
100
Modern
maximum
Maunder
Minimum
50
0
1600
1650
1700
1750
1800 1850
Date
1900
1800
1900
1950
2000
Mean temperature
1000
1100
1200
1300
1400
1500
1600
1700
2000
Date
> Sunspot cycles and terrestrial weather. Scientists have not reached consensus regarding the effects of solar activity on the Earths climate and weather.
Most, however, would agree that the Sun is the primary heat source for the Earth, thus the major driver of climate. Some scientists have tried to draw
a correlation between the absence of sunspots during the Maunder Minimum (top)a 70-year period in the 17th centuryand the Little Ice Age that
affected much of the Earth, especially Europe (bottom). The Dalton Minimum, another period of low sunspot occurrences around 1800, corresponded to
lower than average global temperatures, as well. The rise in total average number of sunspots (black) beginning in the 1900s appear to correspond to
increases in global temperatures. Although a close examination of the data points to other factors producing temperature variations, such as volcanic
eruptions and changes in CO2 levels, some observers propose solar activity as a major component in climate and temperature uctuations. The activity of
Solar Cycle 24 is comparable to that in the cycles around 1800 rather than those of the 20th century. A century from now, scientists may be able to look back
and debunk or validate the causal relationship of sunspots to climate change.
60
Oileld Review
Contributors
Dalia Abdallah is a Senior Production Chemist for Abu
Dhabi Company for Onshore Oil Operations (ADCO)
in Abu Dhabi, UAE. She joined the company in 2008
and has focused on mitigation strategies for scales and
asphaltenes, corrosion issues, produced water treatment and effective stimulation strategies. Previously,
she worked as a Schlumberger uid analysis engineer
in Abu Dhabi. Dalia, who holds two patents and is the
author of several articles, has a PhD degree in chemistry
from Queens University, Kingston, Ontario, Canada.
Khaled Al-Hendi is a Drilling and Workover Supervisor
for Kuwait Oil Company (KOC) in Ahmadi, Kuwait.
He joined KOC in 2006 as a company representative
overseeing the drilling and workover of wells affected
by the Iraqi invasion of Kuwait. Khaled holds a degree
in petroleum engineering from Kuwait University,
Kuwait City.
Adel Abdulla Al-Khalaf is a Senior Petroleum
Engineer and Petrophysicist with Qatar Petroleum
in Doha, Qatar, working on well integrity in offshore
elds. He previously worked for the company as assistant petroleum engineer and then as reservoir engineer in the Dukhan onshore eld. Adel earned a BS
degree in petroleum and natural gas engineering from
West Virginia University, Morgantown, USA.
Zaid Al-Kindi, who is a Well Integrity Domain
Champion for Schlumberger in Abu Dhabi, UAE,
provides support and training for pipe integrity and
zonal isolation projects in Egypt, Oman, Pakistan and
the UAE. With the company since 1994, he has held
positions in service quality and account management.
Previously, he worked in technical sales for Galadari
Heavy Equipment Company and as a project engineer
in the UK. Zaid received a BS degree in mechanical
engineering from Kings College London.
Abdulmohsen S. Al-Kuait is a General Supervisor of
the Saudi Aramco Safaniya production engineering
division in Dhahran, Saudi Arabia. During more than
25 years at Saudi Aramco, he has worked on numerous
projects with a focus on production engineering and
producing operations. Abdulmohsen obtained a BS
degree from King Fahd University of Petroleum and
Minerals, Dhahran, Saudi Arabia.
Mohannad Al-Muhailan is Team Leader Deep Drilling
Engineering for Kuwait Oil Company in Ahmadi,
Kuwait. He has 15 years of experience in conventional
drilling and high-pressure, high-temperature drilling,
workover operations and drilling management and
nances. Mohannad has also worked in underbalanced, managed pressure and cluster drilling operations. He has a degree in petroleum engineering from
Kuwait University in Kuwait City.
Hassan B. Al-Qahtani is a Supervisor of the Saudi
Aramco Safaniya production engineering division in
Dhahran, Saudi Arabia. In more than 17 years as a
petroleum engineer for Saudi Aramco, he has worked
on reservoir management best practices, production
engineering and eld development. Hassan holds a
BS degree from King Fahd University of Petroleum
and Minerals, Dhahran, and an MS degree from The
University of Texas at Austin, USA. He is a graduate of the Saudi Aramco Petroleum Engineering
Technologist Development Program with a specialty
in improved oil recovery.
Autumn 2013
61
62
Oileld Review
Spinner Flowmeter
Caliper, water holdup,
bubble count,
relative bearing, centralizer
Autumn 2013
63
Perforations
Depth, ft
Cable
Speed
60 ft/min
Gamma
Ray
Spinner
Fluid
Fluid
Well
Density Temperature Pressure
Water
Holdup
Cumulative
Production Production
90 ft/min
Oil Water
1,200 ft/min
120
120 0
1.0 0
bbl/d 3,000 0
Oil
Water
bbl/d 5,000
X,900
Y,000
> Production log. This log shows data typically provided by downhole
sensors in a production logging survey. Two intervals have open
perforations (Track 1, red). Engineers make multiple passes at different
logging speeds (Track 2); negative cable speeds represent down passes
and positive logging speeds are up passes. Color-coding based on
logging speed helps differentiate datasets. Gamma ray data (Track 3)
provide correlation with openhole logs. From spinner data (Track 4),
engineers identify changes in fluid velocity associated with inflow from
production, outflow from thief zones or outflow from injection. The fluid
density data (Track 5) indicate water (1.0 g/cm3) below the perforations
(the sump), which then transitions to mainly oil (0.75 g/cm3). Temperature
data (Track 6) indicate heating or cooling effects caused by inflow of
fluids. Steady well pressure (Track 7) is a characteristic of stable flow
during acquisition. Holdup data (Track 8) indicate water and oil fraction
within the wellbore. The software computes incremental and cumulative
flow rates (Tracks 9 and 10). The two intervals are producing oil; the
lower interval is also producing a trace amount of water (Track 9).
64
Near-Vertical Well
Deviated Well
Near-Horizontal Well
> Fluid flow. Theoretical work and flow loop experiments have helped
engineers understand multiphase flow in vertical, deviated and
horizontal wellbores. The parameters of interest include pipe diameter
and inclination and fluid density, viscosity and flow rate. Each case
shows the variation in water and oil holdup based on well deviation.
oil, and both uids are traveling at essentially the same velocity. This simplied example becomes complex in deviated wellbores or when uids of various phases enter the wellbore from multiple zones. Modeling programs
attempt to resolve these complexities using production log data.
Reservoir Surveillance and Management
Production logs help engineers diagnose problems that occur during the life
of a single well and are also used for management and surveillance of multiple wells or of the entire reservoir. A common challenge faced by operators
in mature elds is identifying zones that produce excessive amounts of
water. Produced water impacts surface handling operations because water
must be separated from the ow stream for disposal. Produced water may
also affect well performance by limiting the volume of hydrocarbons being
produced. After identifying water-producing zones, production engineers
may perform remediation to reduce or eliminate water production.
Operators may also use production logs to calibrate reservoir simulation
models. During full-eld reservoir model history-matching, engineers can
use production log data to help identify or match zonal contributions, netune parameters and align the model with the empirical performance data.
Lifetime Logging
Production logs provide reservoir and production engineers with a diagnostic aid for understanding the downhole wellbore environment. These
in situ measurements acquired under dynamic conditions are a snapshot
of the existing situation. But that snapshot captures the situation only
for that moment in time. Whether by fluid extraction or injection, oil
and gas production changes reservoir conditions. Production logs help
operators understand well and reservoir dynamics over the life of a well
and create a roadmap for future remediation, production enhancements
and reservoir development programs.
Oileld Review
AUTUMN 2013
VOLUME 25 NUMBER 3