Journal of the Geological Society, London, Vol. 164, 2007, pp. 755–769. Printed in Great Britain.

Deformation bands in sandstone: a review

H A A KO N F O S S E N 1 , R I C H A R D A . S C H U LT Z 2 , Z O E K . S H I P TO N 3 & K A R E N M A I R 4
1
Centre of Integrated Petroleum Research, University of Bergen, Allégaten 41, N-5007 Bergen, Norway
(e-mail: [email protected])
2
Geomechanics–Rock Fracture Group, Department of Geological Sciences/172, Mackay School of Earth Sciences and
Engineering, University of Nevada, Reno NV 89557, USA
3
Department of Geographical and Earth Sciences, Gregory Building, University of Glasgow, Glasgow G12 8QQ, UK
4
Physics of Geological Processes, University of Oslo, PO 1048 Blindern, 0316 Oslo, Norway

Abstract: Deformation bands are the most common strain localization feature found in deformed porous
sandstones and sediments, including Quaternary deposits, soft gravity slides and tectonically affected
sandstones in hydrocarbon reservoirs and aquifers. They occur as various types of tabular deformation zones
where grain reorganization occurs by grain sliding, rotation and/or fracture during overall dilation, shearing,
and/or compaction. Deformation bands with a component of shear are most common and typically
accommodate shear offsets of millimetres to centimetres. They can occur as single structures or cluster zones,
and are the main deformation element of fault damage zones in porous rocks. Factors such as porosity,
mineralogy, grain size and shape, lithification, state of stress and burial depth control the type of deformation
band formed. Of the different types, phyllosilicate bands and most notably cataclastic deformation bands show
the largest reduction in permeability, and thus have the greatest potential to influence fluid flow.
Disaggregation bands, where non-cataclastic, granular flow is the dominant mechanism, show little influence
on fluid flow unless assisted by chemical compaction or cementation.

Deformation of stiff, low-porosity rock in the uppermost few Antonellini & Aydin 1994, 1995; Beach et al. 1997; Knipe et al.
kilometres of the Earth’s crust occurs primarily by fracturing. 1997; Gibson 1998; Antonellini et al. 1999; Heynekamp et al.
This can result in extensional fractures, such as joints and veins, 1999; Hesthammer & Fossen 2000; Taylor & Pollard 2000; Lothe
or shear fractures such as slip surfaces, which generally form the et al. 2002; Shipton et al. 2002, 2005; Sample et al. 2006) and
primary deformation elements of faults in low-porosity rocks. because they commonly indicate proximity to a larger offset
The process of fault formation and propagation in brittle low- fault. From an academic point of view, deformation bands
porosity rocks has been described in terms of linking of deserve attention because they provide important information on
microfractures and the reactivation or linking of mesoscopic the unique way that faults form in porous sandstones (e.g. Aydin
joints (e.g. Pollard & Fletcher 2005). The key element in a fault & Johnson 1978; Johnson 1995; Davis 1999) and on progressive
is the slip surface, where the majority of offset has accumulated. deformation in porous rocks in general (e.g. Wong et al. 2004;
Surrounding fractures constitute an enveloping damage zone Schultz & Siddharthan 2005). In this paper we review the
(Caine et al. 1996). Slip surfaces and extension fractures, existing literature on deformation bands, present a classification
structures that will be referred to in this paper as ordinary of deformation bands based on deformation mechanism and
fractures, typically represent mechanically weak structures that discuss how the distinctive characteristics of deformation bands
are prone to reactivation and continued slip during subsequent relate to burial depth, lithology and fluid flow.
stress build-up.
Strain in highly porous rocks and sediments is not initially
accommodated by extensional fractures or slip surfaces. Instead,
strain localization occurs by the formation of deformation
Characteristics of deformation bands
structures commonly referred to as deformation bands. Localized The term deformation band has long been used in different ways
(higher offset) faults subsequently form by the failure of in fields such as material science (e.g. Brown et al. 1968) and
deformation band zones. crystal–plastic deformation of rock (e.g. Passchier & Trouw
Deformation bands in porous rocks are low-displacement 1996); however, it was first applied in the context of sandstone
deformation zones of millimetres to centimetres thickness (Fig. deformation by Aydin and co-workers (Aydin 1978; Aydin &
1) that tend to have enhanced cohesion and reduced permeability Johnson 1978, 1983). Since then, the term has gradually been
compared with ordinary fractures. Quaternary geologists find adopted to encompass terms such as microfaults (Jamison &
them in glacially or gravitationally deformed sand, where they Stearns 1982), cataclastic faults (Fisher & Knipe 2001), faults
may reveal information on the local glacial history. Sedimentol- (Manzocchi et al. 1998; Fisher et al. 2003), (micro)fractures
ogists frequently encounter them in sandstones, where they may (Borg et al. 1960; Dunn et al. 1973; Gabrielsen & Koestler
be generated during soft-sediment deformation or post-burial 1987), shear bands (Menéndez et al. 1996), deformation-band
faulting. Petroleum geologists and hydrogeologists (should) look shear zones (Davis 1999), Lüders’ bands (Friedman & Logan
for them in cores from clastic reservoirs and aquifers because of 1973; Olsson 2000), cataclastic slip bands (Fowles & Burley
their potential role as barriers or baffles to fluid flow (Pittman 1994), and granulation seams (Pittman 1981; Beach et al. 1999;
1981; Jamison & Stearns 1982; Gabrielsen & Koestler 1987; Du Bernard et al. 2002b). The most important characteristics of

755
756 H. FOSSEN ET AL.

Fig. 1. (a) Disaggregation bands (centre,

locally invisible) cut by cataclastic
deformation bands (white) in the Navajo
Sandstone, Utah. (b) Rapid variation from
phyllosilicate band to disaggregation band
in sandstone in Jurassic sandstone (Gullfaks
Field, North Sea). Mica-rich layers are local
sources of phyllosilicate minerals. (c)
Phyllosilicate band, Brent Group, Gullfaks
Field. The positive relief (increase of
cohesion) and loss of porosity in the band
should be noted. (d) Photomicrograph of a
single cataclastic deformation band,
showing a low-porosity cataclastic core
mantled by a zone of compaction. Blue
indicates pore space.

deformation bands (in the context of porous rock and sediment than a few centimetres even when the bands themselves are
deformation) are summarized as follows. 100 m long. Localized higher-offset faulting in porous rocks
(1) Deformation bands are restricted to porous granular media, commonly occurs by the failure of existing deformation band
notably porous sands and sandstones. The formation and evolu- zones along a slip surface.
tion of a deformation band involves a significant amount of grain (5) Deformation bands are found in many upper-crustal
rotation and translation, and this process, whether it includes tectonic and non-tectonic regimes (Fig. 2).
grain crushing or merely rotation and frictional sliding along There are several important characteristics that distinguish
grain boundaries, requires a certain amount of porosity. If deformation bands from ordinary fractures (such as slip surfaces
porosity is too low, then tension fractures, stylolites and/or slip or extension fractures). First, they are thicker and exhibit smaller
surfaces will preferentially form. offsets than classical slip surfaces of comparable length. Also,
(2) A deformation band does not represent a slip surface. Slip whereas cohesion is lost or reduced across ordinary fractures,
surfaces can, however, form within bands or, more commonly, most deformation bands maintain or even increase cohesion.
along or within zones of deformation bands, but this represents a Furthermore, deformation bands often exhibit a reduction in
more mature stage in the development of deformation band porosity and permeability, whereas both slip surfaces and tension
faults. fractures are typically associated with a permeability increase.
(3) Deformation bands occur hierarchically as individual Strain hardening behaviour, commonly associated with deforma-
bands, as zones of bands, or within zones associated with slip tion band formation, also contrasts to the strain softening
surfaces (also known as faulted deformation bands). associated with classical fractures. These differences in mechani-
(4) Individual deformation bands rarely host offsets greater cal evolution and structural expression may significantly influ-
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 757

dilation bands, shear bands, compaction bands or hybrids of

these types (e.g. Aydin et al. 2006). The majority of deformation
bands described in the geological literature are shear bands with
attendant compaction (compactional shear bands) caused by
grain reorganization with or without cataclasis. This compaction
contributes to strain hardening and the creation of a localized
band network or zone that precedes faulting (e.g. Schultz &
Balasko 2003; Shipton & Cowie 2003). Early stages of shear
band formation may also involve a component of dilation, and,
although rare, dilational shear bands have been observed in
experiments and in the field (Antonellini et al. 1994; Bésuelle
2001; Borja & Aydin 2004; Okubo & Schultz 2005). Pure
compaction bands have been described in experiments and theory
(Olsson 1999; Olsson & Holcomb 2000; Issen & Rudnicki 2001;
Wong et al. 2001; Baud et al. 2004) and have been recognized in
the field (Hill 1989; Mollema & Antonellini 1996; Sternlof et al.
2005). They are favoured in high-porosity (20–30%) coarse sand
and sandstone (Mollema & Antonellini 1996) and have been
reported to occur in the contractional (leading) quadrants of
faults (e.g. Mollema & Antonellini 1996; Du Bernard et al.
2002a).

Classification of deformation bands by mechanisms

Fig. 2. Some different settings where deformation bands commonly Although a kinematics-based classification (Fig. 3) is logical, it
develop: vertical uplifts and related monoclinal drape folds (Jamison &
is also useful to classify deformation bands in terms of the
Stearns 1982); rift settings (Fisher & Knipe 2001); around salt structures
dominant deformation mechanism operating during their forma-
(Antonellini et al. 1994); above shale diapirs, around thrusts and reverse
tion (Fig. 4). Deformation mechanisms depend on internal and
faults (Cashman & Cashman 2000); glaciotectonic settings (Hooke &
Iverson 1995); areas of gravity-driven collapse (Hesthammer & Fossen
external conditions such as mineralogy, grain size, shape, sorting,
1999). cementation, porosity and stress state. Different mechanisms
produce bands with different petrophysical properties. Thus, such
a classification is particularly useful where permeability and fluid
ence fluid flow and therefore have direct implications for the flow are an issue. The dominant deformation mechanisms are:
management of the porous hydrocarbon and groundwater reser- (1) granular flow (grain boundary sliding and grain rotation); (2)
voirs in which they are very likely to occur. cataclasis (grain fracturing and grinding or abrasion); (3) phyllo-
Kinematically, deformation bands can be classified (Fig. 3) as silicate smearing; (4) dissolution and cementation.

Fig. 3. Kinematic classification of

deformation bands.
758 H. FOSSEN ET AL.

Fig. 5. Deformation band thickness v. displacement, plotted for different

lithologies. Cataclastic deformation bands are from the Entrada
Sandstone, San Rafael Desert, Utah. Data from disaggregation–
phyllosilicate bands are from Jurassic sandstones in the Gullfaks Field
reservoir, northern North Sea. It should be noted that fine-grained bands
are thinner than coarse-grained bands.

Macroscopically, disaggregation bands are effectively ductile

shear zones in the sense that sand laminae can typically be traced
continuously through the band. The amount of shearing and
Fig. 4. The principal types of deformation bands, based on deformation
compaction (pore-space collapse) that actually occurs along
mechanism.
disaggregation bands depends on the nature and properties of the
sandstone. Most pure and well-sorted quartz sand deposits are
already compacted to the extent that the initial stages of shearing
involves some dilation (Antonellini & Pollard 1995; Lothe et al.
Disaggregation bands
2002), although continued shear-related grain reorganization may
Disaggregation bands (Figs 1a and 4a) develop by shear-related reduce the porosity at a later point.
disaggregation of grains by means of grain rolling, grain
boundary sliding and breaking of grain bonding cements, a
process referred to as granular flow (e.g. Twiss & Moores 1992)
Phyllosilicate bands
or particulate flow (e.g. Rawling & Goodwin 2003). They are Phyllosilicate bands (framework phyllosilicate bands of Knipe et
commonly found in sands and poorly consolidated sandstones al. 1997) form in sand(stone) where the content of platy minerals
(Mandl et al. 1977; Du Bernard et al. 2002a; Bense et al. 2003), exceeds 10–15%. They can be considered as a particular type of
and form the ‘faults’ produced in sandbox experiments (e.g. disaggregation band where platy minerals promote frictional
McClay & Ellis 1987). Disaggregation bands can be almost grain boundary sliding (Fig. 1b) rather than grain fracturing
invisible in homogeneous quartz sand(stone)s, but may be (cataclasis).
detected where they cross and offset laminae (Fig. 1a). Their true Where clay is the dominant platy mineral, the clay minerals
shear offsets are typically some centimetres, their lengths less tend to mix with other mineral grains by a process referred to as
than a few tens of metres, and their thicknesses vary with grain deformation-induced mixing (Gibson 1998). The resulting bands
size of the host (Fig. 5). Fine-grained sand(stone)s develop bands are fine-grained, low-porosity zones (Fig. 1c) called deformation
c. 1 mm thick, whereas coarser-grained sand(stone)s host single bands with clay smearing by Antonellini et al. (1994). Coarser
bands that may be at least 5 mm thick. phyllosilicate grains align to form a local fabric within the bands
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 759

as a result of shear-induced rotation (Fig. 1b). Such phyllosilicate

bands tend to show rotation of mica-rich laminae into the band.
In general, phyllosilicate bands can accumulate greater offsets
than other types of deformation bands. This is due to the
smearing of the platy minerals along phyllosilicate bands that
counteracts strain hardening from interlocking of grains. They
are easily detected, as the aligned phyllosilicates give the band a
distinctive colour or fabric. An ordinary disaggregation band
may transform into a phyllosilicate band where the phyllosilicate
content of the rock increases (Fig. 1b).
If the clay content of the host rock is high enough (.40%
according to Fisher & Knipe 2001), the structure becomes a
clay smear. A clay smear is a continuous surface or thin zone of
clay that forms by reorientation, flow, and/or extrusion of clay
minerals. Striations seen on many clay smears indicate that they
act (and should be classified) as slip surfaces rather than
deformation bands. Field examples of cataclastic deformation
bands becoming phyllosilicate bands or clay smears as they cross
sandstone–siltstone boundaries are relatively common (Johansen
& Fossen 2007).
Fig. 6. Displacement–length relationship for cataclastic deformation
Cataclastic bands bands and disaggregation bands. The two categories of deformation
bands occupy different fields of the diagram, and define different semi-
The classic cataclastic deformation bands described by Aydin
linear trends; cataclastic bands are much longer than disaggregation
(1978), Aydin & Johnson (1983) and Davis (1999) occur when bands with respect to offset.
mechanical grain fracture is a significant deformation mechan-
ism. These bands consist of a central cataclastic core, commonly
within a volume of compacted rock (Fig. 1d). The core is
characterized by a wide grain-size distribution and high matrix formed in unconsolidated marine sand buried no deeper than
content because of grain-size reduction, angular grains and a 50 m exhibit grain crushing.
distinct absence of pore space. The surrounding volume is
typically characterized by compaction (as a result of granular
Dissolution, cementation and diagenesis
flow) and gentle fracture of grains. As pointed out by Aydin
(1978), the crushing of grains during cataclasis results in Dissolution and cementation may occur preferentially along a
extensive grain interlocking, promoting strain hardening. Strain deformation band during or, more commonly, after deformation.
hardening may explain the somewhat smaller displacements If solution, also referred to as chemical compaction or pressure
observed on cataclastic deformation bands (,3–4 cm), compared solution, is significant the term ‘solution band’ is warranted.
with disaggregation bands of similar lengths (Fig. 6). Solution bands (Gibson 1998) typically consist of tightly packed
Cataclastic bands are found in porous sandstones throughout grains smaller in size than the matrix, but showing little evidence
the world; for example, in the Suez rift (Beach et al. 1999; Du of cataclasis. Although quartz dissolution accelerates at .90 8C
Bernard et al. 2002b), France (Wibberley et al. 2000), the UK (Walderhaug 1996; i.e. depths greater than c. 3 km), dissolution
(Underhill & Woodcock 1987; Beach et al. 1997; Knott 1993), is a common feature of deformation bands formed at shallower
Ordovician sandstones of Oklahoma (Pittman 1981), Permian depths. Whereas dissolution is promoted by clay minerals on
sandstones of the southern North Sea (Fisher & Knipe 2001) and grain boundaries, cementation in deformation bands is promoted
southeastern Norway (Lothe et al. 2002), and the Jurassic by fresh and highly reactive surfaces formed during grain
sandstones of southwestern USA (e.g. Aydin 1978; Jamison & crushing and/or grain boundary sliding. Cementation is particu-
Stearns 1982; Davis 1999). They are mostly observed in rocks larly pronounced in deformation bands where undeformed host
that have been buried to depths of 1.5–2.5 km; hence it is sand grains are coated by diagenetic minerals such as chlorite
assumed that most cataclastic bands form at such depths; that is, (Ehrenberg 1993) and illite (Storvoll et al. 2002). The coating
after lithification but prior to uplift. Interestingly, cataclastic prevents cementation except in the deformation bands, where the
bands are also observed in unconsolidated or poorly consolidated coating is broken by fracturing and sliding to expose fresh quartz
sands in accretionary prism sediments (Lucas & Moore 1986; surfaces (Leveille et al. 1997; Hesthammer et al. 2002). Cemen-
Karig & Lundberg 1990; Ujiie et al. 2004), Californian marine tation may also be promoted by localized tensile fracture in the
terrace sand (Cashman & Cashman 2000) and loose sandstones centre of a deformation band (Gabrielsen & Koestler 1987;
of the Rio Grande Rift (Heynekamp et al. 1999; Rawling & Leveille et al. 1997) and subsequent precipitation of minerals
Goodwin 2003). Cataclastic bands have also been observed in such as calcite, anhydrite, salt, hydroxides and quartz. Fisher &
non-welded ignimbrites and tuffs (Wilson et al. 2003), Rhine Knipe (2001) suggested such cementation to be discontinuous
Graben and Roer Valley loess (Bense et al. 2003) and subglacial and restricted to extensional jogs in deformation band samples
till (Hooke & Iverson 1995; van der Meer et al. 2003). Rawling from the southern North Sea.
& Goodwin (2003) suggested that cataclastic bands formed in
sediments at shallow depths are characterized by grain spalling
Petrophysical properties
and flaking whereas the deeper cataclastic bands commonly
exhibit transgranular fracturing and grain crushing. However, Permeability measurements across deformation bands (Fig. 7)
Cashman & Cashman (2000) showed that cataclastic bands have led many workers to conclude that deformation bands
760 H. FOSSEN ET AL.

Fig. 7. Permeability data (plug measurements) for deformation bands.

Data from Fowles & Burley (1994), Sigda et al. (1999), Fisher & Knipe
(2001) and Lothe et al. (2002).

reduce transmissibility in a reservoir. However, there are other

cases where deformation bands appear to be conduits for fluids
(e.g. Parry et al. 2004; Sample et al. 2006). The influence of
deformation bands on fluid flow depends on their internal
permeability relative to the surrounding rock. Two extreme cases
are shown in Figure 8, the first representing a deformation band
with a higher porosity and permeability than the host rock (Fig.
8a), the second with considerably lower porosity and permeabil-
ity than the host (Fig. 8b). In these two examples the mineralogy
is almost identical, and the difference lies in deformation
mechanism: disaggregation with little grain fracture in the first
case, compared with intense cataclasis in the second (Fig. 8b).
We now highlight the influence of specific deformation mechan-
isms on the resulting petrophysical properties of deformation
bands.

Disaggregation bands
Disaggregation bands can result in an enhancement or reduction
of porosity depending on whether they have a dilational or Fig. 8. Two deformation bands in the same layer of the Nubian
compactional component. Du Bernard et al. (2002a) reported Sandstone, Tayiba Red Beds, Sinai. (a) Disaggregation (dilation) band
that their pure dilation bands represent an increase of porosity of where the porosity is higher than that of the host rock. (b) Cataclastic
7%, although the pore space has later been filled with clay-rich band showing considerable porosity collapse. Whereas the cataclastic
cement in this case. Antonellini et al. (1994) found a similar deformation band represents a low-permeability structure in the
figure of 8% porosity increase in dilatant shear bands in Arches sandstone, the disaggregation band represents a conduit for fluids.
National Park. Mollema & Antonellini (1996) reported that
compaction bands reduced the porosity from 25% in the host
rock to less than a few per cent in the compaction band. Du
Bernard et al. (2002a) suggested that the increased porosity of ever, most of these porosity and permeability contrasts are
dilation bands should be transient, because of the increased relatively low, and disaggregation bands generally have little
infiltration of clays into the enhanced pore network. These influence on the permeability of sandstone reservoirs (Fig. 7).
observations agree with field evidence for preferred fluid flow
along dilation bands, as reported by Bense et al. (2003) and
Phyllosilicate bands
Sample et al. (2006). Other disaggregation bands may be less
porous and permeable than the host rock. A permeability Phyllosilicate bands typically reduce permeability by an amount
reduction of up to one order of magnitude has been observed in depending on phyllosilicate abundance, phyllosilicate type, phyl-
phyllosilicate-bearing sandstones (Fisher & Knipe 2001). How- losilicate distribution, displacement along the band, and grain
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 761

size (Knipe 1992). On average, the reduction in permeability for been described. Fisher & Knipe (2001) cautioned, however, that
North Sea reservoirs is around two orders of magnitude but can the discontinuous nature of many such cements in the North Sea
be up to five orders of magnitude where the phyllosilicate grains reservoirs makes them a less significant influence on fluid flow
are small (,0–5 ìm; Fisher & Knipe 2001). The reduction is than may be expected from thin-section or hand-sample investi-
caused mainly by mixing and alignment of platy minerals, and gations. An additional effect of quartz cementation is to lower
depends on the specific arrangement of platy minerals and thus porosity and increase the strength of the host rock such that
the shear strain. Typically, these factors, and therefore also subsequent deformation may lead to the development of ordinary
permeability, vary along the deformation bands, depending on fractures that actually represent fluid-flow conduits.
the local source of phyllosilicates. Hence, the effective influence
of phyllosilicate bands on fluid flow is controlled by the points of
Effect on fluid flow
lowest and highest permeability. Estimates of permeability reduc-
tion associated with phyllosilicate bands from core plugs may It has been shown that the majority of deformation bands show
therefore incorrectly reflect their effective influence on fluid flow some reduction in permeability, some by as much as several
during production of a hydrocarbon reservoir. orders of magnitude. However, their practical effect on fluid flow
is not clear. For single-phase flow (i.e. water flowing in a water-
saturated rock or oil flowing in an oil-saturated rock), the
Cataclastic bands
thickness and permeability of the deformation band zone are the
The majority of published studies of petrophysical properties in controlling factors on fluid flow (Darcy flow). Simple numerical
deformation bands have focused on cataclastic deformation analyses demonstrate that the number of deformation bands (i.e.
bands dominated by shear deformation with or without additional thickness of the zone) and/or the permeability reduction must be
compaction. The porosity of cataclastic deformation bands is significant for deformation bands to seriously effect fluid flow
reduced by up to an order of magnitude by grain crushing and (Matthai et al. 1998; Walsh et al. 1998). Nevertheless, complex
resulting change in grain-size distribution. The reduction of zones of deformation bands have been blamed for reduced
porosity produces a corresponding decrease in permeability of productivity in some oil wells (e.g. Harper & Moftah 1984).
two to three, and locally as much as six, orders of magnitude For two-phase flow (i.e. oil flowing through a water-saturated
with respect to the host rock (Pittman 1981; Jamison & Stearns rock, or groundwater flowing through the vadose zone), capillary
1982; Harper & Moftah 1985; Knott 1993; Antonellini & Aydin pressure becomes relevant. In hydrocarbon reservoirs the capil-
1994; Gibson 1994, 1998; Knipe et al. 1997; Crawford 1998; lary threshold pressure of the fault rock determines how much
Antonellini et al. 1999; Fisher & Knipe 2001; Jourde et al. 2002; oil can accumulate on one side of the fault before across-fault
Shipton et al. 2002). The very low (,1%) porosity core of some migration occurs. Calculations predict that deformation bands
well-developed cataclastic shear bands results in permeabilities cannot hold much more than a 20 m (Harper & Lundin 1997) or
as low as 0.001 mD (Freeman 1990; Antonellini & Aydin 1994; perhaps up to 75 m (Gibson 1998) high column of hydrocarbons.
Knipe et al. 1997; Fisher & Knipe 2001; Shipton et al. 2002). Regardless of whether one- or two-phase flow is considered,
Cataclastic compaction bands produced experimentally by Hol- the practical consequence of deformation bands depends on other
comb & Olsson (2003) showed a reduction in permeability of factors than permeability contrasts. In particular, their continuity
around two orders of magnitude. or variation in thickness and permeability in three dimensions is
critical. Field observations of deformation bands indicate that
their thickness and porosity change significantly even along
The effect of dissolution and cementation
single bands. The same is the case with deformation band
Cementation and dissolution in deformation bands may signifi- clusters. Clearly, the weakest point of the deformation band
cantly increase the reduction of porosity and permeability caused network influences its effect on flow. The physical connectivity
by mechanical crushing and reorganization of grains (Ngwenya of bands is a related factor, and they both undermine the effect
et al. 2000; Ogilvie & Glover 2001). A transient increase in of deformation bands as sealing and flow-reducing structures.
permeability occurred in the experiments of Main et al. (2000), However, the presence of deformation bands and deformation
associated with initial dilation (e.g. Mandl et al. 1977; Bernabe band zones may still change the flow pattern if they have a
& Brace 1990). This provides a way for fluids to enter the preferred orientation. Sigda et al. (1999) observed that low-
deformation band, and the entrance of reducing fluids at this porosity deformation bands can act as preferential groundwater
stage offers an explanation of the bleaching of deformation bands flow paths through the vadose zone. Similar channelization can
(Parry et al. 2004). A similar mechanism may explain cementa- be visualized during production of a petroleum reservoir. During
tion within low-porosity cataclastic deformation bands (Fowles oil production stimulated by water injection, pockets of residual
& Burley 1994; Labaume & Moretti 2001; Parnell et al. 2004; oil may also remain in ‘shadow zones’ as a result of capillary
Sample et al. 2006). trapping (Manzocchi et al. 2002). This effect should be consid-
Cementation probably occurs after, rather than during, the ered during planning of wells and simulation of oilfields where
formation of deformation bands, and the solution and precipita- low-permeability deformation bands are a concern.
tion of quartz accelerates after burial and heating to above c.
90 8C (Walderhaug 1996). Fisher & Knipe (2001) reported a
Formation conditions of deformation bands
general decrease of permeability with depth for cataclastic
deformation bands that have experienced post-deformational Given the range of deformation band characteristics and their
burial in the southern North Sea. As a result of the enhanced influence on fluid flow, considerable attention has been devoted
chemical reactivity of fresh broken or abraded grain surfaces in to understanding the conditions that control their formation. A
the cataclastic bands, precipitates probably include efficient number of factors are important, including confining pressure
permeability-reducing clay minerals in addition to quartz. Pre- (burial depth), deviatoric stress (tectonic environment), pore fluid
cipitation of secondary minerals such as carbonates and anhydrite pressure and host rock properties, such as degree of lithification,
along fractures during deformation band reactivation has also mineralogy, grain size, sorting, and grain shape.
762 H. FOSSEN ET AL.

Some of these intrinsic host rock properties are approximately

constant for a given sedimentary rock layer. However, they may
vary dramatically from one layer to another, resulting in rapid
changes in deformation band style across lithological boundaries.
Factors such as porosity, permeability, confining pressure,
stress state and cementation are likely to change with time; hence
deformation bands may record a temporal evolution associated
with, for instance, increasing burial depths. The temporal
sequence of deformation structures in a given rock is an
important geological signature that reflects the physical changes
experienced during burial, lithification and uplift.

Temporal sequence of deformation in sandstones

The earliest forming deformation bands in sandstones are
typically disaggregation bands (Fig. 9). These structures form at
low confining pressures (i.e. shallow burial) when forces acting
across grain contact surfaces are low and grain bindings are weak
(Fig. 10). Early disaggregation bands are often related to local,
non-tectonic gravity-controlled deformation, such as local shale
diapirism, underlying salt movement, gravitational sliding and
glaciotectonics (e.g. Antonellini et al. 1994).
Cataclastic deformation bands can form in poorly consolidated Fig. 10. Schematic illustration of how the different deformation band
sands at ,1 km burial depths (e.g. Lucas & Moore 1986; types relate to phyllosilicate content and depth. Many other factors
influence on the boundaries outlined in this diagram, and we do not know
Cashman & Cashman 2000; Rawling & Goodwin 2003). Shallow
in any great detail how they interact. In addition, the transitions are
cataclasis is promoted where well-sorted and well-rounded grains
gradual, so the boundaries drawn should be considered as uncertain.
lead to high grain-contact stress, and in feldspar and other
minerals that have well-developed crystallographic cleavage (e.g.
Zhang et al. 1990). In general, shallowly formed cataclastic
bands show less intense cataclasis than those formed at greater Once a rock becomes a cohesive lithology with reduced
(1–3 km) depths. Abundant examples of cataclastic deformation porosity, deformation tends to occur by crack propagation instead
bands found in the Jurassic sandstones of the Colorado Plateau of pore space collapse, and slip surfaces, joints and mineral-filled
(Fig. 1d) highlight a clear temporal evolution from early fractures can form. Slip surfaces associated with deformation
disaggregation bands to later cataclastic bands (e.g. Antonellini bands probably form as a result of significant porosity reduction
et al. 1994). Phyllosilicate bands may form at a variety of depths within many deformation band zones. Joints and veins can also
if enough (c. 15% or more) phyllosilicates are present. form in sandstones that have lost porosity because of lithification
and quartz cementation. Thus, slip surfaces, joints and veins
almost invariably postdate both disaggregation bands and cata-
clastic bands in sandstones. Because quartz cementation-related
porosity reduction may vary locally, deformation bands and
joints may develop simultaneously in different parts of a
sandstone, but locally the temporal sequence is: (1) deformation
bands; (2) faulted deformation bands (i.e. slip surface); (3)
jointing; (4) reactivation of joints as faults (e.g. Johansen et al.
2005).
Lithification and loss of porosity are not the only reasons why
extension fractures occur as late structures in most exposed
porous sandstones. Such sandstone sequences tend to portray
regional joint sets influenced by removal of overburden and
related cooling during regional uplift. Although clearly impor-
tant, such features are unlikely to be developed in subsurface
petroleum reservoirs unless they have been significantly uplifted.
Thus, knowing the burial and uplift history of a basin in relation
to the timing of deformation events is very useful when consid-
ering the type of small-scale structures present in, say, a
sandstone reservoir. Conversely, examination of the type of
deformation structure present also gives information about
deformation depth and other conditions at the time of deforma-
tion.

Fig. 9. Theoretical stress history for a simple burial and uplift history of Sensitivity to lithological variations
sandstones (Engelder 1993) in relation to structural development. In
contrast to shales, sandstones enter the tensile regime during uplift, and Field observations of deformation bands crossing lithological
tension fractures (joints) form. contacts in layered sedimentary sequences provide important
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 763

information on lithology control on deformation band style. We fluvial to shallow marine North Sea sands deformed at ,1 km
assume that a single deformation band that crosses adjacent depth typically exhibit 10–15 deformation bands on either side
layers formed at approximately the same geological time, depth of the slip surface (Hesthammer & Fossen 2001), whereas small
and stress conditions in all layers. Field observations clearly faults in aeolian sandstones deformed at c. 2 km depth may have
reveal that the transformation of a single structure from a 50–100 bands or more (Aydin 1978). This indicates that more
cataclastic or disaggregation band into a phyllosilicate band substantial fault damage zones form at greater burial depths
coincides with the transition from well-sorted sandstone to poorly (Mair et al. 2002a). The influence of host rock lithologies is
sorted sandstone with a higher percentage of phyllosilicates. demonstrated in the Moab area of Utah, where the Navajo
Another common observation is that cataclastic bands change Sandstone develops considerably more deformation bands than
into disaggregation bands as they enter more fine-grained and the Entrada Sandstone for a given strain. This relationship is also
poorly sorted sandstones. Such transition clearly depends on seen on bed-scale, where deformation band frequency may vary
factors such as porosity, mineralogy and grain size. If the dramatically from bed to bed (Fig. 12), depending on the
contrast is high enough, the deformation band may actually lithological factors discussed above. Preliminary field data sug-
terminate at the contact between two layers or transform into a gest that high-porosity, well-sorted sandstones develop the widest
slip surface (Schultz & Fossen 2002). damage zones around minor faults.
These observations are consistent with experimental work The length of the deformation band process zone ahead of a
indicating that a decrease in porosity and grain size would inhibit fault tip also varies depending on lithology. This zone is most
cataclasis in sandstone (e.g. Chuhan et al. 2002). Laboratory and extensive (up to .100 m) in well-sorted and highly porous
field observations suggest that the presence of silica cement sandstones such as the Entrada and Navajo Sandstone (Shipton
promotes microfracturing and finer-grained cataclastic deforma- & Cowie 2001; Rotevatn et al. 2007). Such deformation band
tion bands (Johansen et al. 2005). However, the presence of a process zones (Fig. 11) may therefore influence fluid flow in
hematite grain coating (Main et al. 2001) encourages grain regions ahead of seismically mapped fault tips. Given that fault
boundary sliding, thus favouring disaggregation bands rather than offsets less than 10–20 m are not resolved in commercial seismic
cataclastic deformation bands. Primary grain mineralogy will surveys, it is common to use displacement–length scaling
also play a role, and cataclasis is observed to be more intense in relations to extend seismically resolvable fault tips (Pickering et
lithic or feldspathic sand than in quartz sand (e.g. Chuhan et al. al. 1997). The presence of deformation band process zones
2002; Rawling & Goodwin 2003). Flodin et al. (2003) argued should also be included in this type of analysis (although see
that porosity is a primary control on deformation band structure, below for a discussion of displacement–length scaling).
where increased porosity leads to high grain contact stress, and Once a continuous slip surface has formed, strain accumulates
thus favours the formation of cataclastic deformation bands. predominantly by frictional sliding. If subsequent fault growth
For a subsurface petroleum or groundwater reservoir, the and strain accommodation was dominated by strain softening,
ability to make an accurate prediction of deformation structures then damage zone thickness should be independent of fault
and their permeability characteristics at various stratigraphic displacement. However, in many cases, large faults appear to
levels from a basic input of lithology and burial history is highly have wider damage zones than small faults (Fig. 13), suggesting
desirable. Although important advances have been made in this that damage zones are still active during localized fault slip
direction, more experimental and field-based work is required to (Shipton & Cowie 2003). This may be caused by fault locking as
properly understand the coupling between the many factors that a result of non-planar or interfering slip surfaces (Rykkelid &
control deformation in sandstones and other porous media. Fossen 2002). Therefore, structural elements in damage zones
around deformation band faults may be both remnants from the
pre-faulting stage as well as syn-faulting damage (Schultz &
The connection between deformation bands, faults and Siddharthan 2005).
damage zones The orientations of deformation bands in damage zones will
Field data show that deformation bands occur as isolated clearly influence the permeability structure. Conjugate sets of
structures, linked systems, complex zones of multiple, intercon- deformation bands are common in places such as the Colorado
nected deformation bands, and in fault damage zones (Fig. 11) Plateau (e.g. Berg & Skar 2005; Fossen et al. 2005), the Permian
(e.g. Aydin & Johnson 1983; Hesthammer & Fossen 2001). basins of the southern North Sea–UK area (e.g. Fowles & Burley
Laboratory observations (Mair et al. 2000) have confirmed field- 1994) and North Sea Middle Jurassic reservoirs (Hesthammer et
based predictions that the number of distinct deformation bands al. 2000), with one set subparallel to the main slip plane and the
increase with increasing strain. Detailed mapping of outcrops of other dipping in the opposite direction. Mutual cross-cutting
faulted cataclastic deformation bands shows that slip surfaces relationships show that these conjugate sets form contempora-
tend to nucleate in small patches in deformation band zones that neously (Zhao & Johnson 1991; Olsson et al. 2004). It is
propagate, link up, and ultimately form through-going slip anticipated that fluid flow parallel to the strike of these conjugate
surfaces with accumulated strain (Shipton & Cowie 2001). bands would be easier than flow across damage zones. In detail,
Mature, through-going slip surfaces are commonly associated complications in damage zone structure often increase at fault
with a thin (millimetre thick) core of ultracataclasite (e.g. Aydin branch points or stepovers (Antonellini & Aydin 1995; Tindall &
& Johnson 1978; Shipton & Cowie 2001). Intense localized grain Davis 1999; Johansen et al. 2005). The complex variation of
crushing also occurs within zones of deformation bands prior to deformation band geometry in damage zones has the potential to
slip-surface development (Shipton & Cowie 2001, fig. 11; influence flow in a complicated manner.
Johansen et al. 2005, fig. 9), suggesting that grain crushing is an
incipient stage in the formation of slip surfaces.
Deformation band mechanics
The number of deformation bands formed locally at the time
of slip-surface formation is probably sensitive to several factors, A considerable amount of theoretical work has been carried out
including porosity, grain size, cement, mineralogy and over- on the development of deformation bands in rocks (e.g. Rudnicki
burden stress (i.e. depth). Small-scale (5–20 m throw) faults in & Rice 1975; Rudnicki 1977; Aydin & Johnson 1983; Issen &
764 H. FOSSEN ET AL.

Fig. 11. Fault tip in porous sandstone.

Deformation bands are formed in the
‘process zone’ ahead of the fault tip. The
length of the zone varies from lithology to
lithology, but can be up to several hundred
metres and thus should be considered when
evaluating fluid flow in many faulted
hydrocarbon reservoirs.

Fig. 12. Characterization of the damage

zone for a minor fault in layered sand–
siltstones of the Entrada Sandstone (San
Rafael Desert). Band frequency varies from
sandstone layer to sandstone layer, and is
highest in the well-sorted, aeolian layers (1
and 4).

Rudnicki 2000; Borja & Aydin 2004), providing a firm founda- Siddharthan 2005; Aydin et al. 2006). This approach provides a
tion for understanding these structures. An approach called the consistent framework for understanding the development of
‘Cam cap’ model of yielding and band formation is now widely deformation bands, damage zones, and attendant faulting.
used (Wong et al. 1992, 2004; Borja & Aydin 2004; Schultz & The model is best described using a q–p stress diagram (Fig.
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 765

Fig. 13. Half-width of damage zone plotted against displacement in log–

log space for faults, together with thickness–displacement data for
deformation band zones and individual deformation bands in some
porous sandstones. Gullfaks data from Fossen & Hesthammer (2000),
Entrada data from Fossen (unpubl.), Navajo data from Shipton & Cowie
(2001) and Fossen (unpubl.), and Sinai data from Beach et al. (1999) and
Rotevatn & Fossen (unpubl.)

14), where the coordinate axes q and p represent the differential

Fig. 14. The q–p diagram (differential stress–mean stress) applied to
stress and the mean stress, respectively. The yield surface (Fig. porous rocks (after Schultz & Siddharthan 2005). (a) Inelastic yielding of
14) separates the elastic (recoverable strain) from the inelastic the host rock (bold curve) produces dilatant bands (1), dilatant shear
(deformation-band-forming) regimes. Its shape depends on the bands (2), shear bands (3), compactional shear bands (4), and compaction
physical properties of the deforming rock (e.g. Wong et al. 1992; bands (5). The yield surface or cap depends on porosity and grain size as
Issen & Rudnicki 2000; Mair et al. 2002a; Borja 2004; Borja & indicated. (b) Localization of deformation bands for a compressive
Aydin 2004). The type of deformation band that forms will remote stress state. Stress paths A–C are for dilatant shear bands, and
depend on the state of stress at the moment of inelastic yielding; D–F for compactional (i.e. cataclastic) shear bands. Dashed grey lines in
that is, on the point of intersection between the loading path and (b) depict moving yield surfaces in the direction of the frictional sliding
the yield surface. For example, dilatant shear bands are formed line, leading to faulted deformation bands.
at relatively low confining pressures (segment 2 of Fig. 14a)
whereas compactional shear bands are formed at higher confining
pressures (segment 4). The critical pressure, P* (point 5 in Fig.
14a), is the pressure at which compaction occurs in the absence yield cap and upward until the stress state associated with
of shearing. This value scales approximately with the product of frictional sliding along some sections of the band array is
grain size and porosity (Zhang et al. 1990; Wong et al. 1997) achieved (Fig. 14b, path D–E). At this point, slip surfaces
such that as grain size and/or porosity increase, the critical nucleate and grow throughout the band network (Aydin &
pressure and the yield surface decrease (Fig. 14a). Thus unconso- Johnson 1978; Shipton & Cowie 2001, 2003), and the band
lidated sand can form compaction bands at relatively shallow ‘fails’ unstably (Aydin & Johnson 1983), forming a faulted
depth whereas consolidated sandstone requires much higher compactional shear band array (Schultz & Siddharthan 2005).
confining pressures (Mair et al. 2002a).
As an illustration, compactional shear bands (path D–E in Fig.
14b) form at relatively high confining pressure when differential
Scaling relationships
stress increases and the rock begins to compact and shear. As
Displacement–length relationships
this occurs, the grain-to-grain contacts experience a much larger
compressive stress, eventually promoting grain crushing and Deformation band displacement profiles are qualitatively similar
fracturing (Zhang et al. 1990). Grain crushing leads to (1) to those of faults; for example, both develop an along-strike
reduced average grain size within a growing band, (2) a tighter displacement profile with a central maximum (e.g. Fossen &
packing geometry, (3) increased grain angularity, and conse- Hesthammer 1997, 1998). However, some distinct differences
quently increased shear resistance (Mair et al. 2002b). These between faults and deformation bands are worth examining.
factors inhibit shearing displacements within a cataclastic band, Faults with length in excess of c. 100 m commonly exhibit a
resulting in strain hardening. However, because grain crushing power law relationship between length (L) and maximum
depends on several factors including mineralogy, grain size, displacement (D) that can be expressed as (e.g. Cowie & Scholz
packing geometry, grain composition, cementation (lithification) 1992; Clark & Cox 1996)
and shape (Wong et al. 1997; Wong & Baud 1999), the same D ¼ cL n
stress state may produce compaction with or without cataclasis
in different sandstones. Strain hardening moves the rock off the where c is a constant. The exponent (n) has a value of
766 H. FOSSEN ET AL.

approximately one across nine orders of magnitude for faults as Porosity seems to be an important factor controlling whether
small as tens of centimetres long (Clark & Cox 1996; Schlische deformation bands or slip surfaces form as the first mesoscopic
et al. 1996) but was found to be c. 0.5 for a population of structures during rock failure. Few experimental or field data are
cataclastic deformation bands in Utah (Fig. 6). Wibberley et al. available to pinpoint such a critical porosity limit, but it may be
(2000) also reported unusually small D/L ratios for cataclastic of order 10–15% for many sandstones (Wong et al. 1997). Other
deformation bands in porous Cretaceous sandstone in Orange, rock properties (cementation, clay content, grain size, grain
France, indicating that cataclastic deformation bands have less shape), burial depth and fluid pressure will contribute to this
displacement per unit length than faults. limit, as may the local and remote state of stress. Field studies
Wibberley et al. (2000) explained their D/L observations by indicate that unusually thin deformation bands (a half to a
invoking the high frictional strength of cataclastic deformation quarter the size of normal bands) may form in the borderland
bands in porous sandstones; they estimated that porous sand- between the two regimes of classical fracturing and deformation
stones have lower shear moduli than other coarse-grained banding. Johansen et al. (2005) related a sequence of deforma-
sedimentary rocks. Although this can certainly account for the tion band (first), thin deformation bands and jointing–shear
smaller values of D/L, it cannot alone account for a reduction in fractures (last) to progressive quartz cementation and the corre-
slope from 1.0 to 0.5 observed for cataclastic deformation bands. sponding decrease in porosity. Clearly, detailed field observations
Schultz & Fossen (2002) attributed the anomalously low D/L coupled with laboratory tests and physical and numerical model-
ratios for Entrada Sandstone deformation bands to lithological ling will yield a better understanding of these structures and a
layering, and suggested that the deformation bands that nucleate better prediction of their effect on fluid flow in hydrocarbon
in the sandstone layer have difficulties propagating into the reservoirs and groundwater aquifers.
adjacent low-porosity, silty layers. They suggested that once a
growing deformation band spans the thickness of the sandstone
layer, it keeps growing horizontally until a through-going slip References
surface forms and cuts through the stratigraphy. The abundant
Antonellini, M. & Aydin, A. 1994. Effect of faulting on fluid flow in porous
field evidence indicating that deformation bands are sensitive to sandstones: petrophysical properties. AAPG Bulletin, 78, 355–377.
lithology and grow selectively in layers of high porosity supports Antonellini, M. & Aydin, A. 1995. Effect of faulting on fluid flow in porous
this hypothesis. sandstones: geometry and spatial distribution. AAPG Bulletin, 79, 642–671.
Disaggregation bands exhibit D/L scaling similar to ordinary Antonellini, M. & Pollard, D.D. 1995. Distinct element modeling of deforma-
tion bands in sandstone. Journal of Structural Geology, 17, 1165–1182.
faults and slip surfaces (Fig. 6). Wibberley et al. (1999) related Antonellini, M.A., Aydin, A. & Pollard, D.D. 1994. Microstructure of
this observation to ‘slow’ tip propagation that exploits the deformation bands in porous sandstones at Arches National Park, Utah.
reduced shear modulus in unlithified sand compared with Journal of Structural Geology, 16, 941–959.
sandstone. In this hypothesis, ‘slow’ implies that strain hardening Antonellini, M.A., Aydin, A. & Orr, L. 1999. Outcrop aided characterization of
a faulted hydrocarbon reservoir: Arroyo Grande oil field, California, USA. In:
in cataclastic bands may promote tip propagation rather than
Haneberg, W.C., Mozley, P.S., Moore, C.J. & Goodwin, L.B. (eds) Faults
continued accumulation of offset in the central portion of the and Subsurface Fluid Flow. Geophysical Monograph, American Geophysical
band. The result would be long cataclastic bands with small D/L Union, 113, 7–26.
ratios. For unlithified sand, strain hardening appears to be Aydin, A. 1978. Small faults formed as deformation bands in sandstone. Pure and
subordinate or absent, and strain accumulates more readily Applied Geophysics, 116, 913–930.
Aydin, A. & Johnson, A.M. 1978. Development of faults as zones of deformation
through continued shearing along the band (large D/L ratios). bands and as slip surfaces in sandstones. Pure and Applied Geophysics, 116,
931–942.
Aydin, A. & Johnson, A.M. 1983. Analysis of faulting in porous sandstones.
Conclusions Journal of Structural Geology, 5, 19–31.
Aydin, A., Borja, R.I. & Eichhubl, P. 2006. Geological and mathematical
The distinct style of deformation band faulting observed in framework for failure modes in granular rock. Journal of Structural Geology,
porous rocks, where faulting is preceded by a history of 28, 83–98.
deformation band formation and accumulation, has several Baud, P., Klein, E. & Wong, T.F. 2004. Compaction localization in porous
implications. First, the resulting fault is contained in a volume of sandstones: spatial evolution of damage and acoustic emission activity.
Journal of Structural Geology, 26, 603–624.
rock containing deformation bands (the damage zone) that add to Beach, A., Brown, J.L., Welbon, A.I., McCallum, J.E., Brockbank, P. &
the flow-reducing property of the fault. The thickness of this Knott, S. 1997. Characteristics of fault zones in sandstones from NW
deformation band zone, and therefore their effect on fluid flow, England: application to fault transmissibility. In: Meadows, N.S., True-
scales positively with fault displacement. Second, because the blood, S.P., Hardman, M. & Cowan, G. (eds) Petroleum Geology of the
Irish Sea and Adjacent Areas. Geological Society, London, Special Publica-
damage zone extends beyond the tip line of the fault by up to tions, 124, 315–324.
several hundred metres, this potentially lengthens the region that Beach, A., Welborn, A.I., Brockbank, P. & McCallum, J.E. 1999. Reservoir
may perturb fluid flow. A third implication is that fault orienta- damage around faults: outcrop examples from the Suez rift. Petroleum
tion (strike and dip) is influenced by the orientation of the Geoscience, 5, 109–116.
deformation band zone in which it grows (Johnson 1995; Bense, V.F., Van den Berg, E.H. & Van Balen, R.T. 2003. Deformation
mechanisms and hydraulic properties of fault zones in unconsolidated
Johansen et al. 2005; Okubo & Schultz 2006). sediments; the Roer Valley Rift System, The Netherlands. Hydrogeology
The effectiveness of deformation bands as fluid flow barriers Journal, 11, 319–332.
or baffles depends only partly on how their internal petrophysical Berg, S.S. & Skar, T. 2005. Controls on damage zone asymmetry of a normal
properties are altered relative to the surrounding rock, which is fault zone: outcrop analyses of a segment of the Moab fault, SE Utah.
Journal of Structural Geology, 27, 1803–1822.
primarily dictated by deformation mechanisms and mineralogy. Bernabe, Y. & Brace, W.F. 1990. Deformation and fracture of Berea sandstone.
Other important factors are the number of bands (collective In: Duba, A.G., Durham, W.B., Handin, J.W. & Wang, H.F. (eds) The
thickness), their orientation and their continuity, and the variation Brittle–Ductile Transition in Rocks. Geophysical Monograph, American
in permeability and porosity along strike and dip. Geophysical Union, 56, 91–101.
Bésuelle, P. 2001. Evolution of strain localisation with stress in a sandstone:
The specific properties of deformation bands compared with brittle and semi-brittle regimes. Physics and Chemistry of the Earth, A26,
ordinary fractures (joints and slip surfaces) makes it important to 101–106.
investigate the controls on when deformation bands will form. Borg, I., Friedman, M., Handin, J. & Higgs, D.V. 1960. Experimental
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 767

deformation of St. Peter Sand: a study of cataclastic flow. In: Griggs, D. & Gibson, R.G. 1998. Physical character and fluid-flow properties of sandstone-
Handin, J. (eds) Rock Deformation. Geological Society of America, derived fault zones. In: Coward, M.P., Johnson, H. & Daltaban, T.S.
Memoirs, 79, 133–191. (eds) Structural Geology in Reservoir Characterization. Geological Society,
Borja, R.I. 2004. Computational modeling of deformation bands in granular media. London, Special Publications, 127, 83–97.
II. Numerical simulations. Computer Methods in Applied Mechanical Harper, T.R. & Lundin, E.R. 1997. Fault seal analysis: reducing our dependence
Engineering, 193, 2699–2718. on empiricism. In: Möller-Pedersen, P. & Koestler, A.G. (eds) Hydro-
Borja, R.I. & Aydin, A. 2004. Computational modeling of deformation bands in carbon Seals—Importance for Exploration and Production. Norwegian
granular media. I. Geological and mathematical framework. Computer Meth- Petroleum Society Special Publication, 7, 149–165.
ods in Applied Mechanical Engineering, 193, 2667–2698. Harper, T. & Moftah, I. 1985. Skin effect and completion options in the Ras
Brown, N., Duckett, R.A. & Ward, I.M. 1968. Deformation bands in Budran Reservoir. In: Society of Petroleum Engineers Middle East Oil
polyethylene terephthalate. British Journal of Applied Physics, 1, 1369–1379. Technical Conference and Exhibition. SPE, 13708, 211–226.
Caine, J.S., Evans, J.P. & Forster, C.B. 1996. Fault zone architecture and Hesthammer, J. & Fossen, H. 1999. Evolution and geometries of gravitational
permeability structure. Geology, 24, 1025–1028. collapse structures with examples from the Statfjord Field, northern North
Cashman, S. & Cashman, K. 2000. Cataclasis and deformation-band formation in Sea. Marine and Petroleum Geology, 16, 259–281.
unconsolidated marine terrace sand, Humboldt County, California. Geology, Hesthammer, J. & Fossen, H. 2000. Uncertainties associated with fault sealing
28, 111–114. analysis. Petroleum Geoscience, 6, 37–45.
Chuhan, F.A., Kjeldstad, A., Bjørlykke, K. & Høeg, K. 2002. Experimental Hesthammer, J. & Fossen, H. 2001. Structural core analysis from the Gullfaks
compression of loose sands simulating porosity reduction in petroleum area, northern North Sea. Marine and Petroleum Geology, 18, 411–439.
reservoirs during burial. Canadian Geotechnical Journal, 40, 995–1011. Hesthammer, J., Johansen, T.E.S. & Watts, L. 2000. Spatial relations within
Clark, R.M. & Cox, S.J.D. 1996. A modern regression approach to determining fault damage zones in sandstone. Marine and Petroleum Geology, 17, 873–
fault displacement scaling relationships. Journal of Structural Geology, 18, 893.
147–152. Hesthammer, J., Bjørkum, P.A. & Watts, L.I. 2002. The effect of temperature
Cowie, P.A. & Scholz, C.H. 1992. Displacement–length scaling relationship for on sealing capacity of faults in sandstone reservoirs: examples from the
faults: data synthesis and analysis. Journal of Structural Geology, 14, 1149– Gullfaks and Gullfaks Sør fields, North Sea. AAPG Bulletin, 86, 1733–1751.
1156. Heynekamp, M.R., Goodwin, L.B., Mozley, P.S. & Haneberg, W.C. 1999.
Crawford, B.R. 1998. Experimental fault sealing: shear band permeability Controls on fault zone architecture in poorly lithified sediments, Rio Grande
dependency on cataclastic fault gouge characteristics. In: Coward, M.P., Rift, New Mexico: implications for fault zone permeability and fluid flow. In:
Johnson, H. & Daltaban, T.S. (eds) Structural Geology in Reservoir Haneberg, W.C., Mozley, P.S., Moore, C.J. & Goodwin, L.B. (eds) Faults
Characterization. Geological Society, London, Special Publications, 127, and Subsurface Fluid Flow. Geophysical Monograph, American Geophysical
83–97. Union, 113, 27–49.
Davis, G.H. 1999. Structural geology of the Colorado Plateau Region of southern Hill, R.E. 1989. Analysis of deformation bands in the Aztec Sandstone, Valley of
Utah. Geological Society of America, Special Papers, 342. Fire, Nevada. MS thesis, University of Nevada, Reno.
Du Bernard, X., Eichhubl, P. & Aydin, A. 2002a. Dilation bands: a new form Holcomb, D.J. & Olsson, W.A. 2003. Compaction localization and fluid flow.
of localized failure in granular media. Geophysical Research Letters, 29, Journal of Geophysical Research, 108(B6), 2290.
2176–2179. Hooke, R. & Iverson, N. 1995. Grain-size distribution in deforming sub-glacial
Du Bernard, X., Labaume, P., Darcel, C., Davy, P. & Bour, O. 2002b. tills: role of grain fracture. Geology, 23, 57–60.
Cataclastic slip band distribution in normal fault damage zones, Nubian Issen, K.A. & Rudnicki, J.W. 2000. Conditions for compaction bands in porous
sandstones, Suez rift. Journal of Geophysical Research, 107, 2141. rock. Journal of Geophysical Research, 105, 21529–21536.
Dunn, D.E., LaFountain, L.J. & Jackson, R.E. 1973. Porosity dependence and Issen, K.A. & Rudnicki, J.W. 2001. Theory of compaction bands in porous rock.
mechanism of brittle fracture in sandstones. Journal of Geophysical Research, Physics and Chemistry of the Earth, A26, 95–100.
78, 2403–2417. Jamison, W.R. & Stearns, D.W. 1982. Tectonic deformation of Wingate
Ehrenberg, S.N. 1993. Preservation of anomalously high porosity in deeply buried Sandstone, Colorado National Monument. AAPG Bulletin, 66, 2584–2608.
sandstones by grain-coating chlorite: examples from the Norwegian continen- Johansen, S.E. & Fossen, H. 2008. Internal geometry of fault damage zones in
tal shelf. AAPG Bulletin, 77, 1260–1286. siliclastic rocks. In: Kurz, W., Wibberley, C.A.J., Imber, J., Collettini,
Engelder, J.T. 1993. Stress Regimes in the Lithosphere. Princeton University C. & Holdsworth, R.E. (eds) The Internal Structure of Fault Zones: Fluid
Press, Princeton, NJ. Flow and Mechanical Properties. Geological Society, London, Special
Fisher, Q.J. & Knipe, R.J. 2001. The permeability of faults within siliciclastic Publications, in press.
petroleum reservoirs of the North Sea and Norwegian Continental Shelf. Johansen, T.E.S., Fossen, H. & Kluge, R. 2005. The impact of syn-kinematic
Marine and Petroleum Geology, 18, 1063–1081. porosity reduction on damage zone architecture in porous sandstone; an
Fisher, Q.J., Casey, M., Harris, S.D. & Knipe, R.J. 2003. Fluid-flow properties of outcrop example from the Moab Fault, Utah. Journal of Structural Geology,
faults in sandstone: the importance of temperature history. Geology, 31, 965– 27, 1469–1485.
968. Johnson, A.M. 1995. Orientations of faults determined by premonitory shear
Flodin, E., Prasad, M. & Aydin, A. 2003. Petrophysical constraints on zones. Tectonophysics, 247, 161–238.
deformation styles in Aztec Sandstone, southern Nevada, USA. Pure and Jourde, H., Flodin, E.A., Aydin, A., Durlofsky, L.J. & Wen, X. 2002.
Applied Geophysics, 160, 1589–1610. Computing permeabilities of fault zones in eolian sandstone from outcrop
Fossen, H. & Hesthammer, J. 1997. Geometric analysis and scaling relations of measurements. AAPG Bulletin, 86, 1187–1200.
deformation bands in porous sandstone. Journal of Structural Geology, 19, Karig, D.E. & Lundberg, N. 1990. Deformation bands from the toe of the Nankai
1479–1493. accretionary prism. Journal of Geophysical Research, 95, 9099–9109.
Fossen, H. & Hesthammer, J. 1998. Deformation bands and their significance in Knipe, R.J. 1992. Faulting processes and fault seal. In: Larsen, R.M., Brekke,
porous sandstone reservoirs. First Break, 16, 21–25. B.T., Larsen, B.T. & Talleraas, E. (eds) Structural and Tectonic Modelling
Fossen, H. & Hesthammer, J. 2000. Possible absence of small faults in the and its Application to Petroleum Geology. Norwegian Petroleum Society
Gullfaks Field, northern North Sea: implications for downscaling of faults in Special Publication, 1, 325–342.
some porous sandstones. Journal of Structural Geology, 22, 851–863. Knipe, R.J., Fisher, Q.J. & Clennell, M.R. et al. 1997. Fault seal analysis:
Fossen, H., Johansen, S.E., Hesthammer, J. & Rotevatn, A. 2005. Fault successful methodologies, application and future directions. In: Møller-
interaction in porous sandstone and implications for reservoir management; Pedersen, P. & Koestler, A.G. (eds) Hydrocarbon Seals: Importance for
examples from southern Utah. AAPG Bulletin, 89, 1593–1606. Exploration and Production. Norwegian Petroleum Society Special Publica-
Fowles, J. & Burley, S. 1994. Textural and permeability characteristics of faulted, tion, 7, 15–40.
high porosity sandstones. Marine and Petroleum Geology, 11, 608–623. Knott, S.D. 1993. Fault seal analysis in the North Sea. AAPG Bulletin, 77, 778–
Freeman, D. H. 1990. Permeability effects of deformation bands in porous 792.
sandstones. MS thesis, University of Oklahoma, Oklahoma City. Krantz, R.W. 1988. Multiple fault sets and three-dimensional strain: theory and
Friedman, M. & Logan, J.M. 1973. Lüders’ bands in experimentally deformed application. Journal of Structural Geology, 10, 225–237.
sandstone and limestone. Geological Society of America Bulletin, 84, 1465– Labaume, P. & Moretti, I. 2001. Diagenesis-dependence of cataclastic thrust fault
1476. zone sealing in sandstones. Example from the Bolivian Sub-Andean Zone.
Gabrielsen, R.H. & Koestler, A.G. 1987. Description and structural implications Journal of Structural Geology, 23, 1659–1675.
of fractures in late Jurassic sandstones of the Troll Field, northern North Sea. Leveille, G.P., Knipe, R. & More, C. et al. 1997. Compartmentalization of
Norsk Geologisk Tidsskrift, 67, 371–381. Rotliegendes gas reservoirs by sealing faults, Jupiter Fields area, southern
Gibson, R.G. 1994. Fault-zone seals in siliciclastic strata of the Columbus Basin, North Sea. In: Ziegler, K., Turner, P. & Daines, S.R. (eds) Petroleum
offshore Trinidad. AAPG Bulletin, 78, 1372–1385. Geology of the Southern North Sea; Future Potential. Geological Society,
768 H. FOSSEN ET AL.

London, Special Publications, 123, 87–104. Bulletin, 81, 82–99.

Lothe, A.E., Gabrielsen, R.H., Bjørnevoll-Hagen, N. & Larsen, B.T. 2002. Pittman, E.D. 1981. Effect of fault-related granulation on porosity and permeabil-
An experimental study of the texture of deformation bands; effects on the ity of quartz sandstones, Simpson Group (Ordovician) Oklahoma. AAPG
porosity and permeability of sandstones. Petroleum Geoscience, 8, 195–207. Bulletin, 65, 2381–2387.
Lucas, S.E. & Moore, J.C. 1986. Cataclastic deformation in accretionary wedges: Pollard, D.D. & Fletcher, R. 2005. Fundamentals of Structural Geology.
Deep Sea Drilling Project Leg 66, southern Mexico, and on-land examples Cambridge University Press, Cambridge.
from Barbados and Kodiak Islands. In: Moore, C.D. (ed.) Structural fabrics Rawling, G.C. & Goodwin, L.B. 2003. Cataclasis and particulate flow in faulted,
in Deep Sea Drilling Project cores from forearc region. Geological Society of poorly lithified sediments. Journal of Structural Geology, 25, 317–331.
America, Memoirs, 166, 89–103. Rotevatn, A., Fossen, H., Hesthammer, J., Aas, T.E. & Howell, J.A. 2007.
Main, I.G., Kwon, O., Ngwenya, B.T. & Elphick, S.C. 2000. Fault sealing Are relay ramps conduits for fluid flow? Structural analysis of a relay ramp in
during deformation-band growth in porous sandstone. Geology, 28, 1131– Arches National Park, Utah. In: Lonergan, L., Jolly, R.J.H., Rawnsley,
1134. K. & Sanderson, D.J. (eds) Fractured Reservoirs. Geological Society,
Main, I., Mair, K., Kwon, O., Elphick, S. & Ngwenya, B. 2001. Recent London, Special Publications, 270, 55–71.
experimental constraints on the mechanical and hydraulic properties of Rudnicki, J.W. 1977. The inception of faulting in a rock mass with a weakened
deformation bands in porous sandstones: a review. In: Holdsworth, R.E., zone. Journal of Geophysical Research, 82, 844–854.
Strachan, R.A., Magloughlin, J.F. & Knipe, R.J. (eds) The Nature and Rudnicki, J.W. & Rice, J.R. 1975. Conditions for the localization of deformation
Significance of Fault Zone Weakening. Geological Society, London, Special in pressure-sensitive dilatant materials. Journal of the Mechanics and Physics
Publications, 186, 43–63. of Solids, 23, 371–394.
Mair, K., Main, I. & Elphick, S. 2000. Sequential growth of deformation bands Rykkelid, E. & Fossen, H. 2002. Layer rotation around vertical fault overlap
in the laboratory. Journal of Structural Geology, 22, 25–42. zones: observations from seismic data, field examples, and physical experi-
Mair, K., Elphick, S.C. & Main, I.G. 2002a. Influence of confining pressure on ment. Marine and Petroleum Geology, 19, 181–192.
the mechanical and structural evolution of laboratory deformation bands. Sample, J.C., Woods, S., Bender, E. & Loveall, M. 2006. Relationship between
Geophysical Research Letters, 29(10), doi:10.1029/2001GL013964. deformation bands and petroleum migration in an exhumed reservoir rock,
Mair, K., Frye, K.M. & Marone, C. 2002b. Influence of grain characteristics on Los Angeles Basin, California, USA. Geofluids, 6, 105–112.
the friction of granular shear zones. Journal of Geophysical Research, Schlische, R.W., Young, S.S., Ackermann, R.V. & Gupta, A. 1996. Geometry
107(B10), 2219, doi:10.1029/2001JB000516. and scaling relations of a population of very small rift-related normal faults.
Mandl, G., DeJong, L.N.J. & Maltha, A. 1977. Shear zones in granular material. Geology, 24, 683–686.
Rock Mechanics, 9, 95–144. Schultz, R.A. & Balasko, C.M. 2003. Growth of deformation bands into echelon
Manzocchi, T., Ringrose, P.S. & Underhill, J.R. 1998. Flow through fault and ladder geometries. Geophysical Research Letters, 30, doi:10.1029/
systems in high-porosity sandstones. In: Coward, M.P., Johnson, H. & 2003GL018449.
Daltaban, T.S. (eds) Structural Geology in Reservoir Characterization. Schultz, R.A. & Fossen, H. 2002. Displacement–length scaling in three
Geological Society, London, Special Publications, 127, 65–82. dimensions: the importance of aspect ratio and application to deformation
Manzocchi, T., Heath, A.E., Walsh, J.J. & Childs, C. 2002. The representation bands. Journal of Structural Geology, 24, 1389–1411.
of two phase fault-rock properties in flow simulation models. Petroleum Schultz, R.A. & Siddharthan, R. 2005. A general framework for the occurrence
Geoscience, 8, 119–132. and faulting of deformation bands in porous granular rocks. Tectonophysics,
Matthai, S.K., Aydin, A., Pollard, D.D. & Roberts, S.G. 1998. Numerical 411, 1–18.
simulation of departures from radial drawdown in faulted sandstone reservoir Shipton, Z.K. & Cowie, P.A. 2001. Analysis of three-dimensional damage zone
with joints and deformation bands. In: Jones, G., Fisher, Q.J. & Knipe, R.J. development over ìm to km scale range in the high-porosity Navajo
(eds) Faulting, Fault Sealing, and Fluid flow in Hydrocarbon Reservoirs. sandstone, Utah. Journal of Structural Geology, 23, 1825–1844.
Geological Society, London, Special Publications, 147, 157–192. Shipton, Z.K. & Cowie, P.A. 2003. A conceptual model for the origin of fault
McClay, K.R. & Ellis, P.G. 1987. Geometries of extensional fault systems damage zone structures in high-porosity sandstone. Journal of Structural
developed in model experiments. Geology, 15, 341–344. Geology, 25, 333–345.
Menéndez, B., Zhu, W. & Wong, T.-F. 1996. Micromechanics of brittle faulting Shipton, Z.K., Evans, J.P., Robeson, K., Forster, C.B. & Snelgrove, S. 2002.
and cataclastic flow in Berea sandstone. Journal of Sturctural Geology, 18, Structural heterogeneity and permeability in faulted eolian sandstone:
1–16. implications for subsurface modelling of faults. AAPG Bulletin, 86, 863–883.
Mollema, P.N. & Antonellini, M.A. 1996. Compaction bands: a structural Shipton, Z.K., Evans, J.P. & Thompson, L.B. 2005. The geometry and thickness
analog for anti-mode I cracks in aeolian sandstone. Tectonophysics, 267, 209– of deformation-band fault core and its influence on sealing characteristics of
228. deformation-band fault zones. In: Sorkhabi, R. & Tusuji, Y. (eds) Faults,
Ngwenya, B.T., Elphick, S.C., Main, I.G. & Shimmield, G.B. 2000. Experi- fluid flow and petroleum traps. American Association of Petroleum Geolo-
mental constraints on the diagenetic self-sealing capacity of faults in high gists, Memoirs, 85, 181–195.
porosity rocks. Earth and Planetary Science Letters, 183, 187–199. Sigda, J.M., Goodwin, L.B., Mozley, P.S. & Wilson, J.L. 1999. Permeability
Ogilvie, S.R. & Glover, P.W.J. 2001. The petrophysical properties of deformation alteration in small-displacement faults in poorly lithified sediments: Rio
bands in relation to their microstructure. Earth and Planetary Science Letters, Grande Rift, central New Mexico. In: Haneberg, W.C., Mozley, P.S.,
193, 129–142. Moore, C.J. & Goodwin, L.B. (eds) Faults and Subsurface Fluid Flow.
Okubo, C.H. & Schultz, R.A. 2005. Evolution of damage zone geometry and Geophysical Monograph, American Geophysical Union, 113, 51–68.
intensity in porous sandstone: insight gained from strain energy density. Sternlof, K.R., Rudnicki, J.W. & Pollard, D.D. 2005. Anticrack inclusion
Journal of the Geological Society, London, 162, 939–949. model for compaction bands in sandstone. Journal of Geophysical Research,
Okubo, C.H. & Schultz, R.A. 2006. Near-tip stress rotation and the development 110, B11403, doi:10.1029/2005JB003764.
of deformation band stepover geometries in mode II. Geological Society of Storvoll, V., Bjørlykke, K.O., Karlsen, D. & Saigal, G. 2002. Porosity
America Bulletin, 118, 343–348. preservation in reservoir sandstones due to grain-coating illite: a study of the
Olsson, W.A. 1999. Theoretical and experimental investigation of compaction Jurassic Garn Formation from the Kristin and Lavrans fields, offshore Mid-
bands in porous rock. Journal of Geophysical Research, 104, 7219–7228. Norway. Marine and Petroleum Geology, 19, 767–781.
Olsson, W.A. 2000. Origin of Lüders’ bands in deformed rock. Journal of Taylor, W.L. & Pollard, D.D. 2000. Estimation of in situ permeability of
Geophysical Research, 105, 5931–5938. deformation bands in porous sandstone, Valley of Fire, Nevada. Water
Olsson, W.A. & Holcomb, D.J. 2000. Compaction localization in porous rock. Resources Research, 36, 2595–2606.
Geophysical Research Letters, 27, 3537–3540. Twiss, R.J. & Moores, E.M. 1992. Structural Geology. Freeman, San Francisco,
Olsson, W.A., Lorenz, J.C. & Cooper, S.P. 2004. A mechanical model for CA.
multiply-oriented conjugate deformation bands. Journal of Structural Geol- Ujiie, K., Maltman, A.J. & Sánchez-Gómez, M. 2004. Origin of deformation
ogy, 26, 325–338. bands in argillaceous sediments at the toe of the Nankai accretionary prism,
Parnell, J., Watt, G.R., Middleton, D., Kelly, J. & Baron, M. 2004. Southwest Japan. Journal of Structural Geology, 26, 221–231.
Deformation band control on hydrocarbon migration. Journal of Sedimentary Underhill, J.R. & Woodcock, N.H. 1987. Faulting mechanisms in high-porosity
Research, 74, 552–560. sandstones: New Red Sandstone, Arran, Scotland. In: Jones, M.E. &
Parry, W.T., Chan, M.A. & Beitler, B. 2004. Chemical bleaching indicates Preston, R.M.F. (eds) Deformation of Sediments and Sedimentary. Geologi-
episodes of fluid flow in deformation bands in sandstone. AAPG Bulletin, 88, cal Society, London, Special Publications, 29, 91–105.
175–191. van der Meer, J.J.M., Menzies, J. & Rose, J. 2003. Subglacial till: the deforming
Passchier, C.W. & Trouw, R.A.J. 1996. Microtectonics. Springer, Berlin. glacier bed. Quaternary Science Reviews, 22, 1659–1685.
Pickering, G., Peacock, D.C.P., Sanderson, D.J. & Bull, J.M. 1997. Modelling Walderhaug, O. 1996. Kinetic modeling of quartz cementation and porosity loss
tip zones to predict the throw and length characteristics of faults. AAPG in deeply buried sandstone reservoirs. AAPG Bulletin, 80, 731–745.
 D E F O R M AT I O N BA N D S I N S A N D S TO N E 769

Walsh, J.J., Watterson, J., Heath, A.E. & Childs, C. 1998. Representation and Modelling of Quasi-Brittle Materials Behavior. Applied Mechanics Reviews,
scaling of faults in fluid flow models. Petroleum Geoscience, 4, 241–251. 45, 281–293.
Wibberley, C.A.J., Petit, J.-P. & Rives, T. 1999. Mechanics of high displacement Wong, T.-f., David, C. & Zhu, Q. 1997. The transition from brittle faulting to
gradient faulting prior to lithification. Journal of Structural Geology, 21, 251– cataclastic flow in porous sandstones: mechanical deformation. Journal of
257. Geophysical Research, 102, 3009–3025.
Wibberley, C.A.J., Petit, J.-P. & Rives, T. 2000. Mechanics of cataclastic Wong, T.-f., Baud, P. & Klein, E. 2001. Localized failure modes in a compactant
‘deformation band’ faulting in high-porosity sandstone, Provence. Comptes porous rock. Journal of Geophysical Research, 28, 2521–2524.
Rendus de l’Académie des Sciences, Série IIA, 331, 419–425. Wong, T.-f., David, C. & Menéndez, B. 2004. Mechanical compaction. In:
Wilson, J.E., Goodwin, L.B. & Lewis, C.J. 2003. Deformation bands in non- Guéguen, Y. & Boutéca, M. (eds) Mechanics of Fluid-saturated Rocks.
welded ignimbrites: petrophysical controls on fault zone deformation and Elsevier, Amsterdam, 55–114.
evidence of preferential fluid flow. Geology, 31, 837–840. Zhang, J., Wong, T.-f. & Davis, D.M. 1990. Micromechanics of pressure-induced
Wong, T.-f. & Baud, P. 1999. Mechanical compaction of porous sandstone. Oil grain crushing in porous rocks. Journal of Geophysical Research, 95, 341–
and Gas Science and Technolology&Reviews IFP, 54, 715–727. 352.
Wong, T.-f., Szeto, H. & Zhang, J. 1992. Effect of loading path and porosity on Zhao, G. & Johnson, A.M. 1991. Sequential and incremental formation of
the failure mode of porous rocks. In: Li, V.C. (ed.) Micromechanical conjugate sets of faults. Journal of Structural Geology, 13, 887–895.

Received 28 March 2006; revised typescript accepted 4 December 2006.

Scientific editing by Maarten Krabbendam