Journal of the Geological Society, London, Vol. 159, 2002, pp. 605–617. Printed in Great Britain.
Mechanisms and controls on the formation of sand intrusions
1,2
1
1
R I C H A R D J. H . J O L LY & L I D I A L O N E R G A N
Department of Earth Science and Engineering, Imperial College, Royal School of Mines, Prince Consort Road, London
SW7 2BP, UK (e-mail: [email protected]. uk)
2
Present address: BP Exploration, Chertsey Road, Sunbury-on-Thames TW16 7LN, UK
Abstract: Sandstone intrusions are found in all sedimentary environments but have been reported most
commonly from deep-water settings. They also appear to be more frequently developed in tectonically active
settings where applied tectonic stresses facilitate development of high fluid pressures within the sediments. A
variety of mechanisms have been cited as triggers for clastic intrusions. These include seismicity induced
liquefaction, application of tectonic stresses, excess pore fluid pressures generated by deposition-related
processes and the influx of an overpressured fluid from deeper within the basin into a shallow sand body. The
formation of sandstone dykes and sills is investigated here by considering them as natural hydraulic fractures.
When the seal on an unconsolidated, overpressured sand body fails the resulting steep hydraulic gradient may
cause the sand to fluidize. The fluidized slurry can then inject along pre-existing or new fractures to form
clastic intrusions. The scale and the geometry of an intrusive complex is governed by the stress state, depth
and pre-existing joints or faults within the sedimentary succession, as well as the nature of the host sediments.
For the simplest tectonic setting, where the maximum stress in a basin is vertical (gravitational loading), small
irregular intrusions commonly result in the formation of sills at shallow depths within a few metres of the
surface, whereas at greater depth dykes and sills forming clastic intrusion networks are more typical. A simple
relationship is derived to calculate the maximum burial depth at which a dyke–sill complex forms as a
function of the source-bed to sill height, the bulk density of the surrounding sediments, and the ratio of the
vertical to horizontal effective stresses, K0 . When applied to three examples of large-scale dyke–sill
complexes developed within Paleocene and Eocene deep-water reservoir sand bodies of the North Sea,
maximum burial depths in a range of 375 to c. 500 m, 450–700 m and 550–850 m are estimated for intrusion
of each of the three complexes.
Keywords: soft sediment deformation mechanism, sandstone dykes, sandstone sills, sand volcanoes.
The forceful intrusion of remobilized clastic sediment, and its
emplacement into the surrounding strata, forms sheets of sediment that are discordant to bedding (dykes) or concordant with
bedding (sills) (Fig. 1). Dykes that form by the passive infilling
of a fracture or karst feature that is open at the surface, or the
sea bed, are generally referred to as Neptunean dykes. Such
features are not considered further in this paper, because we
restrict our discussion to intrusions that have formed by injection.
Clastic dykes do not exclusively intrude sediments; they also
intrude granitic rock and mafic sills and are associated with lava
flows in volcanic environments (Walton & O’Sullivan 1950;
Harms 1965; Sturt & Furnes 1976). More complex mechanisms
are likely to be involved in the formation of these clastic
intrusions because of the presence of hydrothermal fluids and the
action of thermal stresses. These igneous-related clastic intrusions are also excluded from the following discussion.
Although sandstone intrusions within sedimentary successions
have long been known and documented (e.g. Diller 1889; Jenkins
1930; Peterson 1966; Truswell 1972; Archer 1984; among many
others), in the past these structures have tended to be viewed as
relatively localized phenomena indicative of sediment deformation during shallow burial. This view is now changing, particularly
in areas such as the North Sea basin where the sediment
deformation and intrusion processes directly affect the identification, definition and understanding of important hydrocarbon
reservoirs in deep-water successions (Lonergan et al. 2000). The
recent identification of volumetrically important (cubic kilometres
in scale) intrusions in Tertiary sequences of the North Sea (e.g.
Forth–Harding, Alba, Balder and Gryphon fields; Jenssen et al.
1993; Dixon et al. 1995; Lawrence et al. 1999; Lonergan &
Cartwright 1999; Lonergan et al. 2000) has shown that sandstone
intrusion may be a more common and widespread phenomenon
that previously thought. There is therefore a growing need for a
more rigorous understanding of the mechanisms of clastic intrusion and the controls on sandstone intrusion geometries if we are
to predict and understand the nature and degree of modification of
primary (depositional) sand body geometries. In this paper we
review the conditions necessary for a deposited sand body to
fluidize and intrude. We consider the influence of mechanics on
the geometry of the intrusive system, and introduce a new
relationship, which allows the maximum burial depth at which a
dyke–sill complex formed to be estimated, for simple settings
where the maximum stress in a basin is vertical. Our results are
illustrated with field and subsurface examples. We conclude with
a brief review of sandstone intrusion trigger mechanisms.
Geological settings in which clastic intrusions are found
Clastic dykes are found in a wide range of sedimentary environments. They have been documented in all sedimentary environments, in many basins world-wide: glacial (Johnston 1993);
lacustrine (Aspler & Donaldson 1986; Martel & Gibling 1993);
fluvial (Oomkens 1966; Plint 1985); deltaic, tidal and coastal
(Gill & Kuenen 1958; Reimnitz & Marshall 1965; Dionne 1976;
Hardie 1999); and offshore shallow marine (Johnson 1977) to
deep-water marine fans and turbiditic successions (Truswell
1972; Hiscott 1979; Archer 1984; Parize 1988) (Fig. 2). The
extensive literature documenting sandstone intrusion indicates
605
606
R . J. H . J O L LY & L . L O N E R G A N
Fig. 2. Sedimentary environments in which clastic intrusions occur. Data
sources used for this figure can be obtained from the Society Library or
the British Library Document Supply Centre, Boston Spa, Wetherby,
West Yorkshire LS23 7BQ, UK as Supplementary Publication No. SUP
18171 (7 pages). Only clastic intrusions that are well described in
published papers are cited. Other occurrences of clastic intrusions
documented in conference abstracts, or observed by us or others in the
field are not included.
Fig. 1. Schematic diagram illustrating principal features of clastic sills
and dykes. (a) Intrusion in the subsurface; (b) intrusion associated with
fissures and sand volcanoes. In (a) W is dyke width, L is dyke length
(strike length if exposed in outcrop) and T is sill thickness. Dykes
associated with surface fissures and sand volcanoes that form as a result
of earthquake shaking can have strike lengths up to several hundreds of
metres long; sand volcanoes typically have a small height to diameter
ratio, and maximum diameters reported of 40 m.
that intrusive sand is more commonly encountered in deep-water
marine settings than in any other sedimentary environment (Fig.
2). From a tectonic perspective large and extensive sandstone
intrusion networks are more commonly reported in tectonically
active environments with high sedimentation rates, mud-dominated sedimentary systems and where applied tectonic stresses
facilitate development of high fluid pressures within the sediments. Examples of these include the extensive swarms of dykes
and sill complexes described from fold and thrust belt settings
(e.g. Dzulynski & Radomski 1957; Winslow 1983), and some of
the largest known field examples of clastic intrusions, found in
strike-slip basins along the San Andreas Fault system (Newsom
1903; Thompson et al. 1999).
A variety of water-escape structures, which include dish and
pillar structures, convolute lamination, load structures and mud
flames, form in sediments during or very soon after deposition
(Collinson 1994). The formation of these structures is attributed
to both the rearrangement of grains by escaping fluids and the
action of shear or gravitational stresses on fluidized or liquefied
sediment (Anketell et al. 1970; Lowe 1975; Owen 1987;
Collinson 1994). As clay-rich sediments become buried they
consolidate and become more cohesive, with a corresponding
increase in tensile strength. It is within this regime that clastic
dykes and sills are able to intrude. A sandstone dyke can form at
any depth provided there is a source of non-cemented sand that
can be fluidized. If there is a sufficient reservoir of fluid and the
fluidized sediment breaches the surface, sand volcanoes (also
known as blows or vents) and craters form (e.g. Fuller 1912; Gill
& Kuenen 1958; Obermeier 1996, and references therein). Sand
volcanoes and craters can be fed by cylindrical pipes but are
more commonly fed by largely tabular dyke swarms (Tuttle &
Barstow 1996; Obermeier 1998) (Fig. 1b).
Forceful injection of unconsolidated sand can occur in all
directions. In the glacial environment dykes are typically injected
downwards, as a result of the loading force exerted on poorly
consolidated sediment beneath the glacier (Von Brunn & Talbot
1986). Within other sedimentary environments dykes are generally assumed to intrude upwards or laterally, following the
hydraulic gradient. Often the location of the source sediment,
which would confirm the intrusion direction, is unknown. This
difficulty in identifying a clear and unequivocal source bed often
remains one of the more puzzling features associated with some
otherwise well-exposed classic clastic intrusions (Diller 1889;
Peterson 1966; Parize 1988).
Mechanics of clastic intrusion
A clastic sill or dyke can be considered an example of a natural
hydraulic fracture (Lorenz et al. 1991; Cosgrove 2001). Injection
of a high-pressure fluid (with entrained sand grains) into the
surrounding sediments requires a sustained pressure differential
between the fluid in the propagating fracture or intrusion and the
fluid in the pores of the sediment or sedimentary rock, so that
C L A S T I C I N T RU S I O N S
the fracture dilates and the sand–fluid mixture can flow through
the fracture (Lorenz et al. 1991). Once the excess pressure
dissipates, fracture propagation terminates, and the intrusion
‘freezes’. It is helpful to examine this process in three steps: (1)
a build-up of an excess fluid pressure in a sand body; (2) failure
of the seal; (3) subsequent fluidization of the unconsolidated
sand and injection into host sediments.
Overpressure
An excess fluid pressure (known as overpressure) is generated by
having a sealed bed or pocket of sediment in which the pore
fluid pressure is greater than hydrostatic pressure (Maltman
1994). Overpressured sand bodies commonly occur in a depositional environment where they are enclosed in low-permeability
mudstones. The mudstones immediately surrounding the sand
body prevent the expulsion of pore fluids, and as a result, the
overburden load is partially supported by the excess pore fluid
pressure, causing the sand body to be underconsolidated. The
remaining component of the overburden load is that sustained by
the sediment grains and is called the effective stress, ó9 (Terzaghi
1936):
ó 9 ¼ ó Lith Pf
(1)
where óLith is the overburden or lithostatic stress and Pf is the
fluid pressure. The muds adjacent to the sand body are able to
continue consolidating relative to the sand body, and so form a
low-permeability seal.
Excess fluid pressure generation is thought to occur most
commonly as a result of a process referred to as disequilibrium
compaction. During rapid burial of low-permeability sediments,
the burial rate exceeds the ability of the fluid to escape from
within the pore spaces between the sediment grains, and excess
pore-fluid pressures build up. As a sand body is buried the pore
fluid pressure initially follows the hydrostatic gradient when the
system is unsealed (Fig. 3). When the sand body becomes sealed
or partially sealed, the pore fluid pressure gradient deviates away
from the hydrostatic gradient, and an excess fluid pressure begins
607
to build up. Pore fluids draining into the sand body from the
surrounding compacting mud may increase the fluid pressure
gradient (Magara 1981). Once any excess fluid has drained into
the sand body and becomes trapped the overpressure gradient
remains parallel to the lithostatic stress gradient with increasing
burial (Fig. 3). In tectonically active areas horizontal stresses
may also contribute to the generation of overpressure, and thus
complicate the simple model of disequilibrium compaction
(Berry 1973). The fluid responsible for a build-up of overpressure need not just be ‘local’ fluids; migrating hydrocarbons
or fluids released during geochemical reactions may also cause
an increase in pore pressure in porous sandstones if the sand
bodies are sealed by lower-permeability lithologies (see
Swarbrick & Osborne (1998) for a review of these processes).
Seal failure
A propagating dyke (whether of igneous or clastic origin) is an
example of an opening mode (mode I) fracture that propagates
as a tensile crack in a plane normal to the least compressive
stress direction (Stevens 1911; Delaney et al. 1986). For mode I
failure, the host sediment requires some tensile strength. Cohesion of the host sediment is, therefore, critical if clastic intrusions
are to form. Close to the surface, muds have low cohesion, and
failure in this regime is dominated by shear failure, plastic
deformation and density inversions leading to local soft-sediment
mixing (Lowe 1975; Nichols et al. 1994; Nichols 1995). Porosity
reduction with increasing burial allows a closer packing of the
grains, leading to a greater strength (mainly as a result of an
increase in electrostatic charges between mineral grains). The
depth at which the host sediments acquire enough tensile
cohesive strength (T) to allow a fracture to form is a function of
both porosity reduction and clay chemistry. Where the sediments
do not have sufficient cohesive strength, injection of fluidized
sand manifests itself in the form of pipes (with or without sand
volcanoes at the sediment surface) rather than planar dykes or
sills, and fluidized mixing between the overpressured sandy fluid
and the fluid-rich surrounding muds and silts (Nichols 1995).
Fluidization of the clastic horizon
Fig. 3. Pressure v. depth for a sedimentary succession; relationship
between overburden load (lithostatic pressure, óv ), pore fluid pressure
(Pf ), overpressure and effective stress (ó9) shown. An idealized depth–
pore fluid pressure path for a sand body that has become overpressured
as a result of disequilibrium compaction is indicated (continuous fine
black line).
A fluid-saturated sand bed can liquefy in response to a sudden
shock (e.g. earthquake, or loading by a slump), but such an event
does not necessarily fluidize the bed. A simple increase of the
pore pressure within a sealed system does not remobilize the
sediment grains. For there to be movement of the grains, and
hence flow, there must be a differential pressure gradient across
the bed. The upward movement of the pore fluid in response to
the pressure gradient imposes a drag force on the grains
(Davidson & Harrison 1971; Nichols 1995). Grains become
entrained in the flow when the drag forces exceed the effective
weight of the grains. The velocity at which the upward drag
force just balances the downward forces acting on the grains is
known as the minimum fluidization velocity (Richardson 1971).
Within a sealed unit there has to be a pressure release mechanism that imposes a pressure differential across the bed and allows
loss of fluid from the body thereby generating fluid flow within
the bed. This event must be relatively sudden: diffuse fluid
seepage at low velocities from a sand body does not generate a
large pressure gradient across the body, and the flow velocities
never reach the minimum velocities required to entrain sand
grains (Lowe 1975).
In the absence of intergranular cohesive forces, Richardson
(1971) has shown that the minimum fluidization velocity (Umf )
608
R . J. H . J O L LY & L . L O N E R G A N
of a particulate source bed of uniform spheres is dependent on
the grain diameter (d), the densities of the fluid (rf ) and the
grains (rs ), acceleration due to gravity (g), and the dynamic
viscosity of the fluid (ì), and can be approximated by the
equation
Umf ¼
0:00059d 2 (rs rf ) g
:
ì
(2)
In general, low-density sediments with a small grain size and
size range, together with a spherical shape, are easiest to fluidize
(Richardson 1971). As the grain size becomes smaller and clay
content increases intergranular cohesive forces become stronger
and sediments with grain sizes less than about 0.05 mm and high
mud content are thus more resistant to fluidization (Lowe 1975).
Well-sorted, fine-grained sand beds should therefore be most
prone to fluidization. Typical minimum fluidization velocities for
sand grains in this size range are 0.3–1 cm s1 where water is
the fluidizing medium (Lowe 1975; Eichhubl & Boles 2000).
Contrary to the predictions from theory, which suggest that
fine-grained sand beds should be fluidized most easily, some
natural examples of clastic intrusions show preferential remobilization of the coarsest grain-sized beds (coarse to granule-sized
sand) present in the sedimentary succession (e.g. Ordovician
clastic intrusions, in western Ireland; Archer 1984). It seems
probable therefore that there are other factors associated with
deposition and fluidization that may influence fluidization potential in sandy sediments. Alternatively, the coarser grain-sized
intrusions seen at outcrop may be testimony to high flow
velocities during fluidization and injection, with any finer-grained
sand or silt having been elutriated from the system. Fluidization
experiments (Nichols et al. 1994) show that as flow velocities
increase small grains can be carried through fluidizing beds
containing a mixture of coarse and fine grains.
Determining the geometry of an intrusive system
Brittle failure and dyke orientation
We can, in some circumstances, predict the orientation of a
clastic intrusion by considering the state of stress in the
sedimentary section and the response of the sediments to brittle
failure. We consider four ways in which the host sediment might
fail, as follows.
(1) When the pore pressure (Pf ) within the sand bodies
exceeds the minimum principal stress (ó3 ) and the tensile
strength (T) of the host material, brittle failure occurs (Fig. 4a):
Pf . ó 3 þ T :
(3)
For a sedimentary basin without imposed tectonic stresses, the
maximum principal stress is typically vertical (as a result of
gravitational loading) and the minimum principal stress is in the
horizontal plane. If such a sedimentary basin was filled with
isotropic, homogeneous sediments then vertical dykes would
form (Fig. 4a).
(2) When there are faults or fractures within the host sediment
with little or no tensile strength, the pore fluid pressure (Pf ) need
only exceed the resolved normal stress (ón ) across the older
fracture for dilation to occur (Fig. 3b):
Pf . ó n :
(4)
As the fluid pressure increases a larger range in fracture
orientations are able to dilate and be intruded by the injecting
sand and fluid mixture (Delaney et al. 1986; Baer et al. 1994;
Fig. 4. Mechanisms by which a seal fails, allowing unconsolidated sand
to fluidize and then be injected into overlying sediments. (a) No preexisting fractures; dykes intrude in a plane perpendicular to the least
principal stress. (b) Pre-existing fractures exist; for dilation to occur, the
fluid pressure (Pf ) must exceed the normal stress on the fracture. (c) The
stress fields surrounding an active fault and a sealed sand body
(approximated as an elliptical crack exerting an outward fluid pressure)
interact mechanically if within a distance equal to their dimension of
each other. At favourable orientations (such as that shown) the slipping
fault deflects towards the higher-pressure area, and intersects the seal as
an opening mode fracture, because locally the minimum principal stress
(ómin ) is orientated parallel to the boundary between the sediments and
the sealed sand body.
Jolly & Sanderson 1995, 1997). The photograph in Fig. 5a
illustrates a steeply dipping segmented dyke intruding into the
Miocene Santa Cruz siliceous mudstones. During its formation
two orthogonal fracture set orientations were utilized. The dyke
C L A S T I C I N T RU S I O N S
609
Fig. 5. Sandstone dykes and sills in the
Miocene Santa Cruz mudstone, north of
Santa Cruz, California. (a) Photograph of
cliff section in flat-lying mudstones
illustrating a steeply dipping segmented
dyke (in section view). The fluidized sand
was injected along a number of subparallel
fractures (labelled A; dashed line) and used
a second fracture set (B; continuous lines)
to sidestep between segments. (b)
Photograph of road section. The high fluid
pressure of the injecting sand–fluid mixture
caused the four linked fractures illustrated
here to slip, and at the tip or overlap of the
fractures, dilational jogs formed that were
filled with short sills subparallel to bedding.
intrudes along a fracture from one fracture set (less well
developed in the photograph) and then sidesteps using a fracture
from the orthogonal set of fractures, thus allowing the overall
orientation of the dyke to be vertical. This process of sidestepping is repeated a number of times, resulting in a zigzag
nature to the geometry of the dyke.
(3) An increase in fluid pressure can cause pre-existing
fractures to shear (Phillips 1972; Barton et al. 1995). The
geometry of the pre-existing fractures can be such that areas of
overlap between shearing fractures are dilated, forming jogs that
are then infilled by the fluidized sand. This is illustrated in Fig.
5b, where four fracture segments have slipped with a normal
sense of shear and in the dilational jogs, parallel or subparallel to
the bedding, injected sandstone sills can be seen. It is likely that
this mechanism for host sediment failure would occur in
conjunction with the dilation of other fractures as described in
(2).
(4) The seal on an overpressured sand body can be breached
by a fault that intersects the seal. If faults are propagating in the
vicinity of the overpressured sand body, at distances of the order
of the dimensions of the fault slip plane, or size of the
overpressured sand body, then the stress fields of two structures
interact mechanically (Pollard & Segall 1987). With favourable
fault orientations a normal fault may be deflected such that it
breaches the seal. In the example illustrated in Fig. 4c, the fault
tends to be deflected towards the higher-pressure area, and forms
610
R . J. H . J O L LY & L . L O N E R G A N
an opening mode crack, because locally the minimum principal
stress (ómin ) is orientated parallel to the boundary between the
sediments and the overpressured sand body (e.g. Segall & Pollard
1983).
Sills or dykes and the effect of burial depth and bedding
anisotropy
In this section, using some simple mechanical principles, we
illustrate that it is possible to predict the conditions that might
favour the formation of dykes or sills within a compacting
sedimentary section and gain some qualitative insights into a
geologically complex problem. The stress state in the sediments
depends on the material properties of the sediments and the
boundary conditions. We consider the relatively simple boundary
condition of sediments in a tectonically relaxed basin, that is,
one in which the maximum principal stress is vertical and is due
to overburden load (e.g. post-rift infill in an extensional basin–
passive margin or intracratonic basin) and make the following
simplifying assumptions: (1) because of the anisotropy of
sediment layering and bedding, the sediments have different
tensile strengths in orientations parallel and perpendicular to the
bedding; in a well-laminated mud-dominated siliciclastic sequence the tensile strength in the direction normal to bedding
(Tv ) is likely to be less than that parallel to it (Th ) (Cosgrove
1995); (2) the bedding is horizontal; (3) we neglect any viscous
drag effects that tend to decrease the flow velocities at the
margins of dykes or sills; (4) we consider only the first kilometre
or so of burial before the onset of significant cementation.
The fluid pressure (Pf ) must exceed the horizontal stress (óh )
and the tensile strength of the host sediment parallel to the
bedding (Th ) for a dyke to propagate (Delaney et al. 1986;
Chapter 3 of Price & Cosgrove 1990):
Pf . ó h þ Th :
(5)
For a sill to form the fluid pressure must exceed the vertical
stress (óv ) and the tensile strength perpendicular to the bedding
(Tv ) (Price & Cosgrove 1990):
Pf . ó v þ Tv
(6)
In most sedimentary basins the vertical stress is larger than the
minimum horizontal stress and pressure data such as leak-off
tests from boreholes and reservoirs show that sedimentary rocks
fail before the fluid pressures exceed the lithostatic or overburden
pressure (in this example óv ) (Gaarenstroom et al. 1993; Hillis
2001). Hence clastic intrusions will tend to be vertical, forming
perpendicular to the direction of the minimum horizontal compressive stress, óh . At significant differential stresses, dykes tend
to be preferentially formed because the differential stress is
greater than the differential tensile strength (Cosgrove 1995):
ó h þ Th , ó v þ TV
and therefore
ó v ó h . Th TV :
(7a)
Here failure occurs according to equation (5), before equation
(6).
Where the maximum principal compressive stress is vertical,
dykes that form are vertical, or inclined if the excess fluid
pressures cause existing fractures and faults to dilate or shear as
discussed previously. Dyke strike is controlled by the difference
in magnitude and orientation of the two horizontal stresses,
maximum (óH ) and minimum (óh ). When significantly different,
aligned suites of subparallel dykes form, as seen in the Cretaceous sequence in the Sacramento basin in California (Peterson
1966; Jolly et al. 1998). Another example of aligned dyke
swarms is the Tertiary clastic dykes of the Chilean fold and
thrust belt (Winslow 1983). They crop out along 150 km strike
length of the southern Andean fold and thrust belt. They are
concentrated near the toe of major thrust sheets and are oriented
predominantly perpendicular to the strike of folds and thrusts
(Winslow 1983). According to Winslow (1983), the structural
features of the dykes, and their geometrical relationships to other
structures, suggest rapid, single-phase injection, which she
attributed to the arrival of the front edge of the thrust sheet on an
already overpressured sedimentary sequence. The thrust sheet
loading caused a near-instantaneous increase in overburden
pressure, and a significant increase in the intermediate stress (ó2 )
in the direction of thrust propagation, leading to mobilization of
the underlying overpressured sediments into a slurry and injection forming a swarm of subparallel dykes aligned perpendicular
to the thrust front. In the case of isotropic principal horizontal
stresses dykes formed have random strikes.
When there is a faster rate of reduction of the vertical stress
than fluid pressure for an upward propagating dyke a point is
reached where the fluid pressure exceeds the vertical stress. At
this point equation (6) is satisfied and the dyke can turn into a
sill. There has to be a significant initial fluid pressure gradient
(or large enough reservoir of fluid) to sustain the flow. If not, and
the fluid pressure dissipates rapidly, the intrusion ‘freezes’ as a
dyke before it ever becomes shallow enough to form a sill.
At very shallow burial depths, within a few metres of the
surface, differential stresses are typically low. In this case, the
differential stress may not be less than the differential tensile
strength and the bedding or fabric anisotropy favours the
formation of sills, as it is more difficult for a dyke to propagate
against the anisotropy,
ó h þ Th . ó v þ Tv :
(7b)
Bedding or layering forms a closed planar discontinuity along
which there is lower tensile and shear strength, making it easier
for material to intrude along the discontinuity. Pollard (1973)
conducted some analogue experiments of this process using
grease and gelatin to show how a dyke turns into a sill when it
encounters a discontinuity. He also investigated the effect of
changing the elastic moduli for different layers of gelatin in his
experiments to simulate the different mechanical properties of
interbedded lithologies, and found that the change of material
property at a layer interface also forced a dyke to turn into a sill.
Burial depth and intrusion scale
Making the assumption that the source of the fluidizing fluid in
the sand body is the locally derived original pore fluid, then a
sealed sand body that is breached at very shallow depths is likely
to form a compact sill and dyke complex. The initial pressure
differential is small and the dyke propagates only a small
distance vertically before the fluid pressure equals the vertical
pressure and a sill develops (see path a, a9, a0 in Fig. 6). The
limiting case is where the source sand body is small with only a
small amount of sealed fluid, restricting the amount of sand
grains that can be mobilized before the pressure dissipates and
fluid pressure drops.
Next, let us consider the case where a sand body is sealed
early on but the seal is not breached until greater burial depths
are reached (path b, b9 in Fig. 6). In this instance a greater
pressure differential is established, higher flow rates ensue, and a
larger volume of sediment can be fluidized. Obviously, thicker
and larger volume clastic intrusions can form. The dyke
C L A S T I C I N T RU S I O N S
611
Fig. 6. A simple model for how the scale
of clastic sill and dyke complexes is a
function of the depth of seal onset and
depth at which the seal ruptures, valid for
the case of a basin where the maximum
principal stress is vertical (gravitational
loading). At very shallow depths the
differential stresses (óv óh ) are small, and
the bedding anisotropy favours the
formation of sills in preference to dykes. If
a sand body seals a little deeper (e.g. point
a), the fluid pressure deviates from the
hydrostatic gradient. In the simplest case
where no further pore fluid enters the sand
body, the fluid pressure increases with
continued burial following the lithostatic
gradient. When the fluid pressure reaches
the minimum horizontal stress (óh ) seal
failure ensues and a dyke forms (point a9).
Assuming there is a large enough reservoir
of overpressured fluid the dyke will
propagate upwards until the fluid pressure
exceeds the lithostatic pressure (óv ) and a
sill forms (point a0). If the sand body is
sealed at deeper depths (e.g. b) the distance
to seal failure and sill formation (b9 and b0,
respectively) are larger. Longer dykes must
form between the source bed and sill
formation depth.
propagates further before the fluid pressure exceeds the maximum overburden pressure and a sill forms because of the greater
depth of initiation (path b9– b0, Fig. 6). The depth of failure,
therefore, produces distinctive styles of intrusion.
The influence of depth and differential stress on intrusion
scale: Rosroe Peninsula, Ireland, and Panoche Hills,
California
Two examples are now used to illustrate the control of burial
depth and/or differential stress on geometry, scale and style of
intrusion as discussed above. A dyke and sill complex exposed in
the Panoche Hills in the San Joaquin Valley, northern California
(Smyers & Peterson 1971) represents one end of the depth–
differential stress spectrum as might be represented by a
trajectory between those labelled a–a9– a0 and b–b9– b0 in Fig.
6, whereas the sill-dominated clastic intrusions of the Rosroe
Peninsula, western Ireland represents a near-surface, sill-dominated example (Fig. 7).
Sill-dominated clastic intrusion complexes occur within a
well-exposed, Ordovician deep-marine, sand-dominated, channelized turbiditic succession on the Rosroe Peninsula, at the
westernmost end of Killary harbour, western Ireland (Archer
1984). The Rosroe Formation is made up of two main facies
associations: (1) medium- to granule-sized arkosic sandstones
forming thick-bedded and channelized units, deposited by turbidity currents and debris flows; (2) laminated mudstones and
siltstones interbedded with thin, fine- to medium-grained sandstones, forming packages up to a maximum of 7 m thick. The
sill-dominated intrusive complexes occur exclusively within this
latter facies. Source-bed to sill distances are never more than
50 cm and dykes are volumetrically a very minor component of
the intrusion complexes.
In the Panoche Hills, on the western side of the San Joaquin
Valley, the intrusive complex is best viewed on the west-facing
slopes (Fig. 7b). The host rock for the intrusion is a Cretaceous
shale (Moreno Shale Formation) that dips gently to the east. The
underlying Dosados sandstone member, 80 m below, is the source
bed for the intrusions (Smyers & Peterson 1971). The clastic
dykes are interpreted as intruding along a conjugated joint set,
giving rise to two, distinct, steeply dipping clastic dyke orientations striking NE–SW and approximately north–south. Along
with the dyke intrusion utilizing the joints, sills were also formed
(Fig. 7b). The cliff illustrated in Fig. 7b is c. 80 m high and the
scale of the intrusive complex is an order of magnitude larger
than those found in the Rosroe Peninsula. The main sill is at
least 80 m above the source bed, again much larger than the
source-bed to sill distances observed in the Rosroe examples.
The lack of ptygmatic folding suggests that significant compaction had already occurred before dyke intrusion. The intrusion
occurred at depths greater than a minimum of 120 m (a
representative conservative estimate of the original decompacted
section present today), and probably even deeper because of the
lack of compaction observed in the dykes. These data suggest
that there was a greater differential stress present at the time of
intrusion at Panoche Hills than when the sill-dominated intrusions were formed at Rosroe. These two examples illustrate the
influence of depth, and more specifically the influence of
differential stress, on the scale of intrusions.
Estimating intrusion depth
With a knowledge of the state of stress in the sedimentary
succession, and the appropriate lithostatic and hydrostatic gradients, we can quantify the distances a–a9 and a9– a0 in Fig. 6 and
make estimates of the depth of seal failure and at what distance
612
R . J. H . J O L LY & L . L O N E R G A N
Fig. 7. (a) Photograph and sketch interpretation of a hillside in the
Panoche Hills, western margin of the Great Valley, northern California.
Cliff is c. 80 m high. Both sills and dykes can be seen intruding the
Moreno Shale Formation. The source bed is below the base of the cliff
within the Dosados Member. (b) An example of a sill-dominated
intrusion formed at shallow burial depths, Rosroe Peninsula, western
Ireland.
above the source bed a dyke can become a sill. The problem is
formulated in Fig. 8.
During continuing sedimentation and burial the maximum
applied stress can be considered vertical (as a result of uniform
gravitational loading) and the minimum principal stress horizon-
tal. If the boundary conditions on the basin are such as to prevent
lateral strain in the sediments, then the sediments are constrained
to compact in one dimension, or uniaxially. This stress path has
been investigated in soil mechanics and is known as the state of
‘earth at rest’, or K0 condition (Lambe & Whitman 1979; Jones
1994), defined as the ratio between the principal horizontal and
vertical effective stresses. According to Jones (1994), in all
sedimentary basins without significant surface relief, the sediments are likely to be under stress conditions close to K0 at
shallow burial depths. With increasing depth, the presence of
non-gravitational tectonic stresses moves the sediment away from
the K0 stress path. Data from laboratory sediment consolidation
experiments show that under K0 conditions the ratio of the
horizontal and vertical stresses remains constant. The magnitude
of K0 depends on the type of sediment. Generally the finer
grained and more clay rich the sediment the larger the K0 value
(Jones 1994). For mud and clays, K0 is typically c. 0.7 or higher,
but it can be as low as 0.4 for sands (Lambe & Whitman 1979).
Consolidation experiments conducted by Karig & Hou (1992) to
stresses higher than those routinely achieved in soil mechanics
tests showed that K0 remained remarkably constant at a value of
0.63 for a silty clay up to effective vertical stresses of 35 MPa
(equivalent to depth of several kilometres). As a first-order
approximation, we can assume that at the burial depths of
interest for the formation of clastic intrusions in sedimentary
sequences of a single lithology, the ratio of the vertical to
horizontal effective stresses remains constant, and for a succession dominated by muddy sediment K0 is c. 0.6–0.7. Further
discussion of the meaning of K0 in terms of poroelasticity, or the
coefficient of friction, and alternative ways of estimating it, have
been given by Hillis (2001).
If we know an appropriate K0 value for the sedimentary
succession and óv (the lithostatic pressure), using equation (1)
we can calculate the principal horizontal stress, óh , as follows:
K0 ¼
ó h 9 ó h Phydr
¼
ó v 9 ó v Phydr
and therefore
ó h ¼ K 0 (ó v Phydr ) þ Phydr
(8)
where Phydr is the fluid pressure (hydrostatic gradient in this
case). We define K as the ratio of the principal vertical and
horizontal stresses:
Fig. 8. Method to determine depth of seal
failure, distance from source bed to sill
formation and initiation of injection as a
function of original sealing depth of an
isolated sand body, for a sedimentary
succession of principally a single lithology
(e.g. all muds) and assuming a constant
lithostatic gradient (i.e. neglecting the
changes in bulk sediment density as a result
of porosity reduction over burial depth of
interest).
C L A S T I C I N T RU S I O N S
K¼
óh
:
óv
P2 ¼ rw H 1 g þ rs H 2 g
Thus using equation (8) and knowing óv we can calculate K for a
sedimentary section where the pore fluid is predominantly at
hydrostatic pressures. At K 0 ¼ 0:6, K ¼ 0:8; and at K 0 ¼ 0:8,
K ¼ 0:9 this brackets the range considered reasonable for silty
clays and clays.
Let H be the distance from the source bed to the point where a
dyke becomes a sill, and z be the depth to seal failure (Fig. 8a).
The pressure at point P1, where a dyke turns into a sill, can be
calculated
P1 ¼ rs (z H) g
and
P1 ¼ Krs zg rw Hg
therefore
z¼ H
rs rw
(1 K)rs
or
H¼z
613
(1 K)rs
rs rw
(9)
where rs is the bulk density of the sediments and rw is the
density of the pore fluid (water).
In the case where the sedimentary section consists of muds,
with a K0 value of 0.7 (equivalent to K ¼ 0:85) and assuming a
bulk sediment density of 2000 kg m3 , the depth of failure, z, is
3.33H, the complex height. Or a dyke becomes a sill at a
distance above the source bed of 0.3z. This relationship is
illustrated graphically in Fig. 9b, for a range of K0 and rs values
considered likely for a muddy sedimentary succession at burial
depths up to c. 1.0 km.
Assuming that pore fluid at a hydrostatic pressure is sealed
into the sand body at a depth H1 and no other fluid enters the
sand body before it fails, the depth at which the seal ruptures
with respect to the seal formation depth (H2 ) can be calculated
as a function of the seal formation depth (H1 ) (Fig. 8). At point
P2, when the seal fails, the pressure can be estimated
and
P2 ¼ Krs ( H 1 þ H 2 ) g
and therefore
H2 ¼ H1
(Kr s r w )
:
(1 K)r s
(10)
For a shale section, with a K0 value of 0.7 (K ¼ 0:85) and an
average bulk sediment density of 2000 kg m3 , the seal ruptures
at a depth of H1 þ 2.33H1 .
Equation (9) (Fig. 9b) is likely to be a more useful relationship
than equation (10), because it is independent of how the overpressure in the sealed sand body is achieved. It should be valid
even if the sand body is overpressured by fluid migrating from
greater depths within the basin and being trapped by a top seal in
the sand body. Equation (10) assumes that the overpressure is
entirely generated by disequilibrium compaction, and that there
is no contribution to excess fluid pressures in the source sand
body from diagenetic reactions or hydrocarbon generation, nor
inflow of fluid from elsewhere in the basin. Overpressure
development as a result of disequilibrium compaction is unlikely
to build up in as simple and predictable a fashion as we have
shown in the scenarios illustrated in Figs. 6 and 8 (where the
pore fluids seal at one depth and then the fluid pressure increases,
with a gradient parallel to the lithostatic gradient). However, with
the approach outlined above, some simple, order of magnitude
constraints can be placed on scaling relationships for the
formation of clastic sill and dyke complexes.
The above approach is, however, a considerable oversimplification, because we do not consider the effect of coupling fluid
pressure and stress (Lorenz et al. 1991; Hillis 2001). When the
stress state and fluid pressures are considered in a coupled
system, an increase in fluid pressure is accompanied by a
corresponding decrease in effective differential stress. The decrease in effective differential stress means that the height above
Fig. 9. (a) Line drawings from seismic data of cross-sections through low-angle clastic dykes feeding a shallower sill, Tertiary sequence, North Sea. The
examples are from the Harding field reservoir, quadrant 9, UK sector; Block 24/9 of the Norwegian sector (both in the Viking Graben) and Alba field
reservoir, Block 16/29, UK Central Graben. (b) Source-bed to sill distance plotted v. depth of seal failure for an isolated, overpressured sand body in a
hydrostatically pressured sedimentary succession. Solutions shown for bulk sediment densities of 2200–1500 kg m3 and K 0 ¼ 0:65–0:8. These values are
representative of a clay-dominated succession. This simple relationship can be considered appropriate only for conditions of shallow burial in a basin (e.g.
,1.0 km), where the sediments undergo 1D consolidation with no lateral strain, and no other compaction processes (such as cementation or pressure
solution) operate. Intrusion depth ranges for clastic intrusions illustrated in (a) are also plotted.
614
R . J. H . J O L LY & L . L O N E R G A N
the source bed, H, where a dyke changes to a sill will be reduced.
The equations derived here are still of some use because they
give the upper limit, or maximum dimension that an intrusive
complex might reach.
Estimating intrusion depth in Tertiary deep-water deposits
in the North Sea
As an illustration of the practical use of the relationships
derived in the previous section we now apply equation (9) to
three large-scale clastic intrusions occurring within Paleocene–
Mid-Eocene deep-water sediments in the Central and Viking
grabens in the North Sea basin. The three examples, shown in
Fig. 9a, are from the Harding field in Quadrant 9, UK northern
North Sea, Block 24/9 in the Norwegian sector of the northern
North Sea and from the Alba field in Block 16/26 in the UK
Central Graben. Fluidized sand has intruded up low-angle dykes
(probably along faults) feeding sills that are now located
between 130 and 200 m above the source bed. Their geological
context and setting have been discussed further by Dixon et al.
(1995), Lonergan et al. (2000) and Molyneux (2001). During
Paleogene time the North Sea basin was in its post-rift infill
phase. The absence of inversion or extensional structures of this
age in the central and northern North Sea implies that there
were no significant horizontal stresses acting on the basin at this
time and it would appear reasonable to conclude that the
maximum principal stress (owing to the overburden) acted
vertically on shallowly buried mud-dominated sediments in the
Alba, Harding and 24/9 areas. Cuttings and well-log data from
wells drilled through these sand bodies demonstrate that the
overlying Eocene and Oligocene successions were mud dominated. The sand bodies were deposited in flat-lying, base-ofslope or basin-floor settings. This relatively simple geological
setting and the inferred stress boundary conditions are consistent
with the assumptions made at the start of this discussion
suggesting that it is appropriate to apply equation (9) to obtain
an upper limit for the depth at which the source sand bed seal
failed and the intrusions formed. Assuming a K0 value of 0.7
(consistent with the mud-dominated section overlying the source
sand bed) suggests burial depths in a range of 375–500 m, 450–
700 m and 550–850 m for intrusion of the Alba, Harding and
24/9 dyke–sill complexes, respectively (Fig. 9b). We cite a
range of plausible intrusion depths because the bulk host
sediment density at the time of intrusion is difficult to estimate
as it depends on the rate of porosity loss during burial. Using
an exponential decay porosity–depth relationship for mudstone,
its density varies from 1500 kg m3 at the depositional surface
to c. 2200 kg m3 at 1 km burial.
Intrusion triggers
Lastly we briefly consider clastic intrusion triggering mechanisms. An understanding of sand intrusion triggers has a bearing
on how the mechanical relationships derived in the previous
sections might be applied to other studies. A century of research
has identified four principal triggers for the fluidization and
injection of unconsolidated sand to form sandstone dykes and
sills: (1) seismicity-induced liquefaction (e.g. Fuller 1912;
Reimnitz & Marshall 1965; Hesse & Reading 1978; Obermeier
1989; Saucier 1989; Sims & Garvin 1995; Galli 2000); (2)
tectonic stress (Peterson 1966; Winslow 1983; Huang 1988); (3)
localized excess pore fluid pressures generated by depositionrelated processes such as slumping (Truswell 1972), the passage
of storm waves (Allen 1985; Martel & Gibling 1993) or channel
switching (Hiscott 1979); (4) the influx of an overpressured fluid
from deeper within the basin into a shallow sand body (Jenkins
1930; Brooke et al. 1995). A trend that emerges from previous
work is that seismicity and localized excess pore fluid pressures
resulting from depositional processes are the most commonly
cited triggers for clastic intrusion (Fig. 10). In general, the
smaller, centimetre- to metre-scale clastic intrusions that appear
to have formed close to the depositional surface are attributed to
depositional triggers (Truswell 1972; Hiscott 1979; Archer 1984;
Collinson 1994).
It is useful to consider the relationship between earthquakeinduced liquefaction and sandstone dykes and sills a little further,
particularly because seismicity is so often considered a trigger
for the generation of clastic intrusions and the presence of clastic
intrusions is often used to infer palaeoseismicity (Obermeier
1996). Studies of ground deformation associated with historical
earthquakes have shown that near-surface, water-saturated sands
become liquefied as a result of cyclical shear stresses during
earthquake shaking and produce dykes, sills, volcanoes and
craters at distances up to several hundred kilometres from the
epicentre (Allen 1985; Obermeier 1996; Galli 2000). Worldwide
data on historical shallow earthquakes show that features having
a liquefaction origin are confined to shallow, saturated, Holocene
sediments, poorly compacted artificial sand fills and, more rarely,
Pleistocene sands (Ambraseys 1988). These data also show that
liquefaction structures form for earthquakes with magnitudes as
low as five (Kuribayashi & Tasuoka 1975; Ambraseys 1988; Tani
& Nakashima 1999; Galli 2000) but that a magnitude of 5.5–6 is
the lower limit at which liquefaction becomes common (Fig. 11).
If one then considers the global distribution and magnitude of
recent earthquakes to be representative of the geological record,
only basins in very active tectonic settings or at plate boundaries
are likely to have experienced earthquakes of magnitude greater
than five.
Trenching across Holocene and latest Pleistocene sand volcanoes and their associated clastic intrusion structures shows that
dykes travel a maximum of c. 10 m vertically from their source
bed during earthquake-induced liquefaction (Obermeier 1996;
Tuttle & Barstow 1996). At greater depths it is increasingly
difficult to liquefy sediment by the application of cyclical
shearing stresses because the higher overburden pressure greatly
increases the resistance of sediment to shearing. Earthquake
liquefaction dykes can be up to 2 m thick (although most appear
to be ,30 cm thick) and several hundreds of metres long in plan
view; sills are typically several centimetres thick, but after severe
earthquakes thicknesses up to 0.5 m have been documented
(Obermeier 1996). When appealing to seismicity as a trigger for
clastic intrusion or remobilization it is important to consider (1)
the depth and scale of the intrusion (only near-surface watersaturated unconsolidated sands have been observed to deform by
liquefaction) and (2) the likelihood that earthquakes greater than
magnitude five may have occurred in the past, at the time of
intrusion. However, it is important to acknowledge a significant
observational bias in the datasets used to construct the empirical
relationships between earthquake magnitudes and liquefaction
such as that in Fig. 11. The sedimentary environments in which
recent or historic earthquake-induced clastic intrusions have been
observed are exclusively terrestrial and coastal (which includes
the emergent parts of deltas; such as the study of the effects of
the 1964 Alaskan earthquake on the Copper River delta by
Reimnitz & Marshall (1965)). There have been few, if any,
published studies of liquefaction effects in offshore parts of
deltas and deep marine slope and basinal settings.
More rarely considered as a possible trigger for clastic
C L A S T I C I N T RU S I O N S
615
Fig. 10. Literature citations of intrusion
trigger mechanisms. Points not located
within the cluster at apices of the triangle
either specify more than one mechanism
operating or indicate the possibility that
more than one mechanism was responsible
for the clastic intrusion in question,
according to the original workers. Examples
that specifically mention earthquakeinduced liquefaction are identified with an
asterisk. References for this figure not cited
elsewhere in the text are listed in the
Supplementary Publication.
intrusions is the additional influx of a remote fluid such as oil or
gas at higher pressures migrating up dip from deeper within a
basin. Jenkins (1930) recognized that the migration of hydrocarbon fluids may have played an important role in the formation
of the large number of dykes known in the oil-producing basins
of California. More recently, Brooke et al. (1995) have identified
gas-bearing circular mounded sand bodies at 500 m depth in the
Plio-Pleistocene sequences of the North Sea, associated with gas
conduits. Those workers attributed the formation of the gas
mounds to upward migration of gas (along faults?) causing
liquefaction of the sand body. It is possible that the migration of
overpressured hydrocarbons from deeper within the North Sea
basin has also played an important role in the formation of the
large-scale lower Tertiary sandstone intrusions such as those
illustrated in Fig. 9 as postulated by Lonergan et al. (2000). The
migration of overpressured hydrocarbon-charged fluid into shallow sand bodies would certainly be an effective way of
remobilizing large volumes of sand grains, with enough reservoir
of fluid to generate extensive, large-scale clastic dyke and sill
networks.
Conclusions
Previous work has tended to examine clastic intrusions in
isolation with little consideration of the mechanics involved in
their formation. Many factors have a role to play in the formation
of clastic intrusions: these include the presence of fine-grained
facies within the depositional environment to form a seal,
processes that may contribute to the build-up of overpressures
ranging from shear-liquefaction in the depositional realm to
seismic activity in tectonic settings, cementation processes within
the source bed, and any significant fluid fluxes within the basin
under consideration. Using simple mechanical concepts and
drawing on the existing, more rigorous analysis of the formation
of tabular igneous intrusions (e.g. Delaney & Pollard 1981;
Delaney et al. 1986; Pollard & Segall 1987) useful insights can
be gained about the formation of clastic intrusions. This paper
has used theory to propose the following key factors that
contribute to the formation and geometry of clastic intrusions,
and illustrated these results with natural examples.
(1) Burial depth and the ratio of maximum to minimum
616
R . J. H . J O L LY & L . L O N E R G A N
Fig. 11. Earthquake moment magnitudes for shallow focus earthquakes
v. maximum distance from epicentre at which surface liquefaction has
been observed. Data from Ambraseys (1988). Liquefaction occurs only
for earthquake events larger than magnitude five.
principal stresses, plus their respective orientations, play a
fundamental role in controlling intrusion geometry and scale.
(2) In cases of a horizontal minimum principal stress within a
basin, dykes should form in preference to sills. However, at
shallow levels (10–20 m?) bedding anisotropy and very low
differential stresses are predicted to favour the formation of sills.
(3) For a mechanically simple system of a tectonically
unstressed basin, a relationship has been derived that allows the
depth of intrusion to be calculated as a function of K0 and
sediment density, when the height between the source bed and a
sill is known.
We thank T. Bai, J. Cosgrove, D. Dewhurst, P. Eichhubl, H. D. Johnson,
N. Lee, D. Pollard and J. Wilkinson for fruitful discussions. Valuable
suggestions were made by H. Reading, J. Lorenz and K. Pogue on an
early version of the text. R. Hillis and A. Hurst are thanked for their
constructive reviews of the manuscript. R.J.H.J. was funded by an NERC
Ropa award. L.L. thanks the Royal Society for supporting her research,
and the Geomechanics Group at Stanford for hospitality and a stimulating
environment in which to write the first draft of this paper.
References
Allen, J.R. 1985. Principles of Physical Sedimentology. George, Allen and Unwin,
London.
Ambraseys, N.N. 1988. Engineering seismology. Earthquake Engineering and
Structural Dynamics, 17, 1–105.
Anketell, J.M., Cegla, J. & Dzulinski, S. 1970. On the deformational structures
in systems with reverse density gradients. Annales de la Société Géologique
de Pologne, 40, 3–29.
Archer, J.B. 1984. Clastic intrusions in deep-sea fan deposits of the Rosroe
formation, Lower Ordovician, Western Ireland. Journal of Sedimentary
Petrology, 54, 1197–1205.
Aspler, L.B. & Donaldson, J.A. 1986. Penecontemporaneous sandstone dykes,
Nonacho Basin (early Proterozoic, NW Territories): horizontal injection in
vertical, tabular fissure. Canadian Journal of Earth Sciences, 23, 827–838.
Baer, G., Beyth, M. & Reches, Z. 1994. Dikes emplaced into fractured basement,
Timna Igneous Complex, Israel. Journal of Geophysical Research, 99,
24039–24051.
Barton, C.A., Zoback, M.D. & Moos, D. 1995. Fluid flow along potentially
active faults in crystalline rock. Geology, 23, 683–686.
Berry, F.A.F. 1973. High fluid potentials in California Coastal Ranges and their
tectonic significance. AAPG Bulletin, 57, 1219–1245.
Brooke, C.M., Trimble, T.J. & Mackay, T.A. 1995. Mounded shallow gas sands
from the Quaternary of the North Sea: analogues for the formation of sand
mounds in deep water Tertiary sediments? In: Hartley, A.J. & Prosser,
D.J. (eds) Characterisation of Deep Marine Clastic Systems. Geological
Society, London, Special Publications, 94, 95–101.
Collinson, J. 1994. Sedimentary deformational structures. In: Maltman, A. The
Geological Deformation of Sediments. Chapman and Hall, London, 95–125.
Cosgrove, J.W. 1995. The expression of hydraulic fracturing in rocks and
sediments. In: Ameen, M.S. Fractography: Fracture Topography as a Tool in
Fracture Mechanics and Stress Analysis. Geological Society, London, Special
Publications, 92, 187–196.
Cosgrove, J.W. 2001. Hydraulic fracturing during the formation and deformation
of a basin: a factor in the dewatering of low-permeability sediments. AAPG
Bulletin, 85, 737–748.
Davidson, J. F. & Harrison, D. (eds) 1971. Fluidization. Academic Press,
London.
Delaney, P.T. & Pollard, D.D. 1981. Deformation of Host Rocks and Flow of
Magma during Growth of Minette Dikes and Breccia-bearing Intrusions near
Ship Rock, New Mexico. US Geological Survey Professional Paper, 1202.
Delaney, P.T., Pollard, D.D., Ziony, J.I. & McKeE, E.H. 1986. Field relations
between dikes and joints: emplacement processes and paleostress analysis.
Journal of Geophysical Research, 91, 4920–4938.
Diller, J.S. 1889. Sandstone dikes. Geological Society of America Bulletin, 1,
411–442.
Dionne, J.C. 1976. Miniature mud volcanoes and other injection features in tidal
flats, James Bay, Quebec. Canadian Journal of Earth Sciences, 13, 422–428.
Dixon, R.J., Schofield, K., Anderton, R., Reynolds, A.D., Alexander,
R.W.S., Williams, M.C. & Davies, K.G. 1995. Sandstone diapirism and
clastic intrusion in the Tertiary sub-marine fans of the Bruce–Beryl
Embayment, Quadrant 9, UKCS. In: Hartley, A.J. & Prosser, D.J. (eds)
Characterization of Deep Marine Clastic Systems. Geological Society,
London, Special Publications, 94, 77–94.
Dzulynski, S. & Radomski, A. 1957. Clastic dikes in the Carpathian Flysch.
Annales de la Société Géologique de Pologne, 26, 225–264.
Eichhubl, P. & Boles, J.R. 2000. Rates of fluid flow in fault systems—evidence
for episodic rapid fluid flow in the Miocene Monterey Formation, Coastal
California. American Journal of Science, 300, 571–600.
Fuller, M.L. 1912. The New Madrid Earthquake. US Geological Survey Bulletin,
494.
Gaarenstroom, L., Tromp, R.A.J., De Jong, M.C. & Brandenberg, A.M. 1993.
Overpressures in the Central North Sea: implications for trap integrity and
drilling safety. In: Parker, J.R. (ed.) Petroleum Geology of Northwest
Europe. Proceedings of the 4th Conference. Geological Society, London,
1305–1313.
Galli, P. 2000. New empirical relationships between magnitude and distance for
liquefaction. Tectonophysics, 324, 169–187.
Gill, W.D. & Kuenen, P.H. 1958. Sand volcanoes on slumps in the Carboniferous
of County Clare, Ireland. Quarterly Journal of the Geological Society,
London, 113, 441–460.
Hardie, J.K. 1999. Clastic Dikes Intruding Cretaceous Coals of Western Colorado.
Bulletin of the Colorado Geological Survey, 53.
Harms, J.C. 1965. Sandstone dikes in relation to Laramide faults and stress
distribution in the Southern Front Range, Colorado. Geological Society of
America Bulletin, 76, 981–1002.
Hesse, R. & Reading, H.G. 1978. Subaqueous clastic fissure eruptions and other
examples of sedimentary transposition in the lacustrine Horton Bluff
Formation (Mississippian), Nova Scotia, Canada. In: Matter, A. & Tucker,
M. Modern and Ancient Lake Sediments. International Association of
Sedimentologists Special Publications, 2, 241–257.
Hillis, R.R. 2001. Coupled changes in pore pressure and stress in oil fields and
sedimentary basins. Petroleum Geoscience, 7, 419–425.
Hiscott, R.N. 1979. Clastic sills and dykes associated with deep-water sandstones,
Tourelle formation, Ordovician Quebec. Journal of Sedimentary Petrology,
49, 1–10.
Huang, Q. 1988. Geometry and tectonic significance of the Albian sedimentary
dykes in the Sisteron area, SE France. Journal of Structural Geology, 10,
453–462.
Jenkins, O.P. 1930. Sandstone dikes as conduits for oil migration through shales.
AAPG Bulletin, 14, 411–421.
Jenssen, A.I., Bergslien, D., Rye-Larsen, M. & Lindholm, R.M. 1993. Origin
of complex mound geometry of Paleocene submarine-fan reservoirs, Balder
Field, Norway. In: Parker, J.R. (ed.) Petroleum Geology of Northwest
Europe. Proceedings of the 4th Conference. Geological Society, London,
123–133.
C L A S T I C I N T RU S I O N S
Johnson, H.D. 1977. Sedimentation and water escape structures in some late
Precambrian shallow marine sandstones from Finnmark, North Norway.
Sedimentology, 24, 389–411.
Johnston, J.D. 1993. Ice wedge casts in the Dalradian of South Donegal; evidence
of subaerial exposure of the Boulder Bed. Irish Journal of Earth Sciences,
12, 13–26.
Jolly, R.J.H. & Sanderson, D.J. 1995. Variation in form and distribution of the
dykes of the Mull Swarm, Scotland. Journal of Structural Geology, 17,
1543–1557.
Jolly, R.J.H. & Sanderson, D.J. 1997. A Mohr-circle construction for the
opening of a pre-existing fracture. Journal of Structural Geology, 19,
887–892.
Jolly, R.J.H., Cosgrove, J.W. & Dewhurst, D.N. 1998. Thickness and spatial
distributions of clastic dykes, Northwest Sacramento Valley, California.
Journal of Structural Geology, 20, 1663–1672.
Jones, M.E. 1994. Mechanical principles of sediment deformation. In: Maltman,
A. The Geological Deformation of Sediments. Chapman and Hall, London,
36–71.
Karig, D.E. & Hou, G. 1992. High-stress consolidation experiments and their
geologic implications. Journal of Geophysical Research, 97, 289–300.
Kuribayashi, E. & Tasuoka, F. 1975. Brief review of liquefaction during
earthquakes in Japan. Soils Foundations, 15, 81–92.
Lambe, W.T. & Whitman, R.V. 1979. Soil Mechanics. Wiley, New York.
Lawrence, D.A., Sancar, B. & Molyneux, S. 1999. Large-scale clastic
intrusion in the Tertiary of Block 29/4, Norwegian North Sea: Origin, timing
and implications for reservoir continuity. AAPG Bulletin, 83, 1324.
Lonergan, L. & Cartwright, J.A. 1999. Polygonal faults and their influence on
deepwater sandstone reservoir geometries, Alba Field, UK Central North Sea.
AAPG Bulletin, 83, 410–432.
Lonergan, L., Lee, N., Johnson, H.D., Jolly, R.J.H. & Cartwright, J.A. 2000.
Remobilization and injection in deepwater depositional systems: implications
for reservoir architecture and prediction. In: Wiemer, P., Slatt, R. M.,
Coleman, J. & 5 others (eds) Deep-water Reservoirs of the World.
GCSSEPM Foundation 20th Annual Bob F. Perkins Research Conference,
515–532.
Lorenz, J.C., Teufel, L.W. & Warpinski, N.R. 1991. Regional fractures 1: a
mechanism for the formation of regional fractures at depth in flat-lying
reservoirs. AAPG Bulletin, 75, 1714–1737.
Lowe, D.R. 1975. Water escape structures in coarse grained sediments. Sedimentology, 23, 157–204.
Magara, K. 1981. Mechanisms of natural fracturing in a sedimentary basin. AAPG
Bulletin, 65, 123–132.
Maltman, A. (ed.) 1994. The Geological Deformation of Sediments. Chapman and
Hall, London.
Martel, A.T. & Gibling, M.R. 1993. Clastic dykes of the Devono-Carboniferous
Horton Bluff Formation, Nova Scotia; storm-related structures in shallow
lakes. Sedimentary Geology, 87, 103–119.
Molyneux, S. 2001. Sandstone remobilization in the Eocene to Miocene of the
Central and Northern North Sea. PhD thesis, University of London.
Newsom, J.F. 1903. Clastic dykes. Geological Society of America Bulletin, 14,
227–268.
Nichols, R.J. 1995. The liquification and remobilization of sandy sediments. In:
Hartley, A.J. & Prosser, D.J. (eds) Characterisation of Deep Marine
Clastic Systems. Geological Society, London, Special Publications, 94, 63–
76.
Nichols, R.J., Sparks, R.S. & Wilson, C.J.N. 1994. Experimental studies of the
fluidization of layered sediments and the formation of fluid escape structures.
Sedimentology, 41, 233–253.
Obermeier, S.F. 1989. The New Madrid Earthquakes: an Engineering-Geologic
Interpretation of Relict Liquefaction Features. US Geological Survey Professional Paper, 1336-B.
Obermeier, S.F. 1996. Use of liquefaction induced features for paleoseismic
analysis. Engineering Geology, 44, 1–76.
Obermeier, S.F. 1998. Liquefaction evidence for strong earthquakes of Holocene
and latest Pleistocene ages in the states of Indiana and Illinois, USA.
Engineering Geology, 50, 227–254.
Oomkens, E. 1966. Environmental significance of sand dikes. Sedimentology, 7,
145–148.
Owen, G. 1987. Deformation processes in unconsolidated sands. In: Jones, M.E.
& Preston, M.F. (eds) Deformation of Sediments and Sedimentary Rocks.
617
Geological Society, London, Special Publications, 29, 11–24.
Parize, O. 1988. Sills et Dykes Gréseux sédimentaries: paléomorphologie, fracturation précoce, injection et compaction. PhD thesis, École des Mines de Paris.
Peterson, G.L. 1966. Structural interpretation of Sandstone dikes, Northwest
Sacramento Valley, California. Geological Society of America Bulletin, 77,
833–842.
Phillips, W.J. 1972. Hydraulic fracturing and mineralization. Journal of the
Geological Society, London, 128, 337–359.
Plint, A.G. 1985. Possible earthquake-induced soft-sediment faulting and remobilization in Pennsylvanian alluvial strata, southern New Brunswick, Canada.
Canadian Journal of Earth Sciences, 22, 907–912.
Pollard, D.D. 1973. Derivation and evaluation of a mechanical model for sheet
intrusions. Tectonophysics, 19, 233–269.
Pollard, D.D. & Segall, P. 1987. Theoretical displacements and stresses near
fractures in rock: with applications to faults, joints, veins, dikes, and solution
surfaces. In: Atkinson, B.K. Fracture Mechanics of Rocks. Academic Press,
London, 277–346.
Price, N.J. & Cosgrove, J.W. 1990. Analysis of Geological Structures. Cambridge
University Press, Cambridge.
Reimnitz, E. & Marshall, N.F. 1965. Effects of the Alaska earthquake and
tsunami on recent deltaic sediments. Journal of Geophysical Research, 70,
2363–2376.
Richardson, J.F. 1971. Incipient fluidization and particulate systems. In:
Davidson, J.F. & Harrison, D. Fluidization. Academic Press, London,
25–64.
Saucier, R.T. 1989. Evidence for episodic sand-blow activity during the 1811–
1812 New Madrid (Missouri) earthquake series. Geology, 17, 103–106.
Segall, P. & Pollard, D.D. 1983. Mechanics of discontinuous faults. Journal of
Geophysical Research, 85, 4337–4350.
Sims, J.D. & Garvin, C.D. 1995. Recurrent liquefaction induced by the 1989 Loma
Prieta earthquake and 1990 and 1991 aftershocks: implications for paleoseismicity studies. Bulletin of the Seismological Society of America, 85, 51–65.
Smyers, N.B. & Peterson, G.L. 1971. Sandstone dikes and sills in the Moreno
Shale, Panoche Hills, California. Geological Society of America Bulletin, 82,
3201–3208.
Stevens, B. 1911. The laws of intrusion. Bulletin of the American Institute of
Mining Engineers, 49, 1–23.
Sturt, B.A. & Furnes, H. 1976. Spatial and temporal relationships of intrusive
limestones from Las Palmas, Canary Islands. Journal of Sedimentary
Petrology, 46, 555–562.
Swarbrick, R.E. & Osborne, M.J. 1998. Mechanisms that generate abnormal
pressures; an overview. In: Law, B.E., Ulmishck, G.F. & Slavin, V.I. (eds)
Abnormal Pressures in Hydrocarbon Environments. American Association of
Petroleum Geologists, Memoirs, 70, 13–34.
Tani, S. & Nakashima, M. 1999. Earthquake damage to earth dams in Japan—
maximum epicentral distance to cause damage as a function of magnitude.
Soil Dynamics and Earthquake Engineering, 18, 593–602.
Terzaghi, K. 1936. The shearing resistance of saturated soils. In: Proceedings 1st
International Conference on Soil Mechanics and Foundation Engineering,
Cambridge, MA, 1, 54–56.
Thompson, B.J., Garrison, R.E. & Moore, C.J. 1999. A late Cenozoic sandstone
intrusion west of Santa Cruz, California: fluidized flow of water and
hydrocarbon-saturated sediments. In: Garrison, R.E., Aiello, I.W. &
Moore, J.C. Late Cenozoic Fluid Seeps and Tectonics along the San
Gregorio Fault Zone in the Monterey Bay Region, California. American
Association of Petroleum Geologists, Pacific Section, Volume and Guide
Book, GB-76, 53–74.
Truswell, J.F. 1972. Sandstone sheets and related intrusions from Coffee Bay,
South Africa. Journal of Sedimentary Petrology, 42, 578–583.
Tuttle, M. & Barstow, N. 1996. Liquefaction-related ground failure: a case
study in the New Madrid seismic zone, Central United States. Bulletin of the
Seismological Society of America, 86, 636–645.
Von Brunn, V. & Talbot, C.J. 1986. Formation and deformation of subglacial
intrusive clastic sheets in the Dwyka Formation of northern Natal, South
Africa. Journal of Sedimentary Petrology, 56, 35–44.
Walton, M.S. & O’Sullivan, R.B. 1950. Mechanics of intrusion of a clastic
dyke. American Journal of Science, 248, 1–21.
Winslow, M.A. 1983. Clastic dike swarm and the structural evolution of the
foreland and thrust belt of the Southern Andes. Geological Society of America
Bulletin, 94, 1073–1080.
Received 21 February 2002; revised typescript accepted 15 March 2002.
Scientific editing by Alex Maltman