Earth and Planetary Science Letters 246 (2006) 217 – 230
www.elsevier.com/locate/epsl
Emplacement of shallow dikes and sills beneath a small basaltic
volcanic center – The role of pre-existing structure
(Paiute Ridge, southern Nevada, USA)
Greg A. Valentine ⁎, Karen E.C. Krogh
Earth and Environmental Sciences Division, Mail Stop D462, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
Received 28 January 2006; received in revised form 13 April 2006; accepted 19 April 2006
Editor: C.P. Jaupart
Abstract
Late Miocene sills and dikes in the Paiute Ridge area of southern Nevada were emplaced in an extensional setting beneath a
small volume, alkali basaltic volcanic center. Dikes (400–5000 m long, 1.2–9 m wide) mostly occupy pre-existing E-dipping
normal faults. Elastic deformation of the wall rocks alone cannot explain dike dimensions; inelastic deformation, wall rock erosion
by flowing magma, and syn-emplacement extension of the host structural system also contributed to dike widths. After primarily
subvertical emplacement, flow focused toward the southern end of one of the dikes to form a volcanic conduit. This dike and a
fault-hosted radial dike subsequently were subject to high pressures due to transient volcanic processes. Three small sills
(extending laterally up to ∼ 500 m, and 20–46 m thick) and two larger sills (each having lateral dimensions ∼ 1 km) locally branch
off some dikes within ∼ 250 m of the paleosurface. Individual small sills extend only into the hanging wall blocks of the faults that
host their parent dikes, and are connected to the dikes by stems that are only a few tens of meters wide; elsewhere along their strikes
the parent dikes extend above the sills. This mode of sill emplacement was caused by local rotation of principal stresses related to
the intersection of the dike-hosting fault planes with the complex contact between relatively strong Paleozoic carbonates and weak
Tertiary tuffs. Orientation of bedding planes in the tuffs controlled the direction of sill propagation. The three most areally extensive
sills formed lopoliths with sagging roofs, indicating interaction with the free surface.
© 2006 Elsevier B.V. All rights reserved.
Keywords: dike; sill; basalt
1. Introduction
The most common mechanism by which basaltic
magmas move through the earth's crust is that of magmafilled, self-propagating fractures referred to as dikes [1–
3]. Dikes intrude normal to the regional direction of least
principal stress, or to the local direction of least principal
⁎ Corresponding author. Fax: +1 505 665 3285.
E-mail address: [email protected] (G.A. Valentine).
0012-821X/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.epsl.2006.04.031
stress in the vicinity of large structures such as volcanic
centers [4]. Dikes can also propagate by dilation of preexisting fracture surfaces that are suitably oriented with
respect to the least principal stress [5–7]. Existing
theoretical models treat the ratio of dike width to dike
length as a function of the pressure of the magma and the
elastic deformation of wall rocks in response to that
pressure.
In many cases, subvertical dikes feed subhorizontal
sheet intrusions (loosely called sills in this paper,
218
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
although the strict definition of a sill requires that it be
concordant with bedding or foliation of the host rock).
Previous research has identified three main factors that
determine whether a dike will feed a sill. (1) If dike
propagation is driven by magma buoyancy relative to
local wall rocks, then when a dike reaches a height in the
lithosphere where its magma is no longer buoyant (the
level of neutral buoyancy, or LNB), further upward propagation will cease. Further influx of magma is accommodated by lateral flow at the LNB. Depending on
whether σ3 (least principal compressive stress) is horizontal or vertical at the LNB, the dike will become a
bladed (laterally propagating) dike or a sill, respectively
[1]. (2) The emplacement of a dike changes the stress
field because a dike exerts pressure on its walls. Under
some conditions [8] the vertical stress could become the
least principal stress and sills could form by continued
magma influx. This process is enhanced where dikes
cross zones of ductile rocks where ambient differential
stresses are low, as discussed by Parsons et al. [9]. (3) If
the host rock can behave as a fluid, then fracture propagation ceases and the dike magma simply spreads out
laterally at its LNB as a density current. An example of
this is where a propagating dike encounters another intrusive body that is still partially molten [9,10].
Here we describe the basaltic intrusive complex at
Paiute Ridge, Nevada, which contains both dikes and sills
emplaced at shallow depths in association with formation
of a small basaltic volcanic center. This area provides a
favorable setting in which to document and test models of
dike and sill emplacement mechanisms because the contact relations are generally well exposed. The dikes intruded pre-existing normal faults in a small horst and
graben system, and we suggest that inelastic deformation,
erosion of dike walls, and syn-emplacement extension of
the structural system contributed significantly to their
dimensions. Unlike sills produced by the mechanisms
given above, the sills in this complex were strongly controlled by complex pre-existing structures and by interaction with the free (earth's) surface. This paper is part of a
series aimed at constraining the shallow plumbing, ascent
and eruption mechanisms, and geomorphic evolution of
small-volume continental basalt centers [11–19]. Although partly motivated by volcanic risk assessment for
the proposed permanent underground radioactive waste
Table 1
Representative major element oxide composition of Paiute Ridge
alkali basalta, determined by X-ray fluorescence analysis
Oxide
wt.%
SiO2
TiO2
Al2O3
FeO
MgO
CaO
Na2O
K2O
P2O5
Total
48.92
2.84
17.35
10.95
5.28
8.12
4.17
1.70
0.58
99.92
a
Sample B-13, reported in Appendix A of Crowe et al. [20].
repository at Yucca Mountain, Nevada (USA), these
studies also have broader implications for our understanding of this common type of volcanism.
2. Geologic Setting
The Paiute Ridge intrusive complex is composed of
dikes, sills and small remnants of scoria cones and lava
flows of alkali basalt (hawaiite) composition (Table 1;
Fig. 1). The complex was originally mapped by Byers
and Barnes [21] and has briefly been described by
Crowe et al. [11] and Valentine et al. [22,23] and in more
detail by Carter Krogh and Valentine [24]. Paleomagnetic evidence and thermal modeling [15,25,26],
radiometric age determinations [21,26], and petrologic
studies (F.V. Perry, unpublished data) indicate that the
basalts were emplaced during a single, brief magmatic
pulse that had a total volume of a few tenths of km3. No
cross-cutting relationships have been found between
intrusions. Most of the basaltic rocks in the Paiute Ridge
area are intrusive (either dikes or sills), but there is a
small remnant of a lava flow on the eastern side of the
complex that is underlain by scoria fallout deposits, as
well as patches of variably welded agglutinate in downdropped blocks in the central part of the complex, all of
which indicate that the magma erupted onto the surface.
A neck-like body at the southern end of E dike (labelled
N on Fig. 1) is inferred to be the feeder conduit for at
least some of the eruptions. The 40Ar/39Ar age of this
Fig. 1. Simplified geologic map of the Paiute Ridge area, emphasizing basaltic rocks. Modified from Byers and Barnes [21]. Cross-section lines in
Fig. 2 are indicated. Pz – Paleozoic basement rocks (mainly carbonates and shales) undivided; Tp – undivided bedded and massive Tertiary tuffs
including units belonging to Tunnel Formation, Indian Trail Formation, Paintbrush group (composed of fallout, pyroclastic flow deposits, and
reworked non-welded tuffs with varying degrees of zeolitic alteration). Tm – Timber Mountain Tuffs (mainly ignimbrites of the Rainier Mesa and
Ammonia Tanks members); Qac – Quaternary alluvium and colluvium. PR – Paiute Ridge fault and dike; W – W sill; M – M dike and fault; E – E
dike and fault; ES – ES sill along E dike near volcanic neck N; EN – EN sill along E dike.
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
219
220
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
event is 8.59 ± 0.07 (2σ) Ma; it is part of a group of alkali
basalts that erupted within the south-central Great Basin
in the past ∼ 9 Ma, coinciding with the waning of silicic
volcanism in the region [27,28].
The structural setting at the time of intrusion was that
of a small horst and graben system that consisted dominantly of N- to NNW-trending normal faults. These faults,
which formed during approximately WSW- directed
extension during the early Tertiary [29], displace a sequence of middle Miocene silicic tuffs (from the South-
western Nevada Volcanic Field located 25 km to the
west [30,31]) that had been deposited on rolling hills of
Paleozoic carbonate rocks (see Fig. 2).
3. Dikes
We focus on three dikes (labeled PR, M, and E on Fig. 1)
that are best exposed and have the clearest relationships
to adjacent sills. Many smaller en echelon dike sets occur
within the complex (Fig. 1) but are described elsewhere
Fig. 2. Cross-sections through Paiute Ridge area (A–A′ and B–B′ in Fig. 1) showing present day topography and geology, with topography (dashed)
at the time of basaltic activity (inferred by projection of erosional remnants of the original terrain). Letters are the same as in Fig. 1. Note that the sills
(W sill in B–B′ and ES sill in A–A′) intrude Tertiary tuffs just above the Tertiary–Cambrian unconformity.
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
[24]. In addition, there are many smaller examples of faulthosted dikes that share characteristics of PR, M, and E
dikes, but are relatively localized in their occurrence.
3.1. Description of PR, M, and E dikes
Basaltic dikes in the study area have intruded dominantly within a 2-km-wide graben between the Carbonate and Paiute Ridges (Fig. 1). The longest dikes occur
along pre-existing, NNW-trending, E-dipping normal
fault surfaces that cut Paleozoic sedimentary rocks (carbonates and shales) and overlying Tertiary tuffs (Figs. 1
and 2). There are many other faults in the area, including: (1) those with similar (NNW) strikes that either dip
more shallowly east or dip toward the west and; (2)
those with E to NE strikes. At the current depth of
exposure, however, none of these faults are occupied by
dikes. Additionally, there are a few small dikes that
radiate out from the volcanic neck (Fig. 1) and do not
appear to follow pre-existing structures.
The westernmost, or Paiute Ridge (PR), dike varies in
strike from N to NNW and intrudes Cambrian carbonates
near its southern end and non-welded, bedded Tertiary
tuffs (Fig. 1) toward the north. The dike visibly occupies
the steeply E-dipping Paiute Ridge fault at Slanted Buttes
(Figs. 1 and 2) and along the trend of the fault at its
southern tip. Along the central segment, the dike does not
reside along a mappable fault surface; however, it connects the fault-entrenched tips and most likely occupies a
fault surface as well. The dike nowhere appears displaced
by the Paiute Ridge fault, which has a throw of less than
15 m across the Rainier Mesa member of the Timber
Mountain Tuff (Tm on Fig. 1) at Slanted Buttes.
The PR dike is composed of a main continuous dike
and at least 24 dike-parallel segments. The main dike is
4600 m long; its width is highly variable as it pinches and
swells along strike, with a typical width in its central part
being 5 m but locally swelling to 9 m wide. PR dike is
widest where it feeds the sill at its southern end (Fig. 1).
Dike-parallel segments vary from 4 to 100 m in length
with widths of a few decimeters to 1.2 m. These segments
occur within tens of meters of, but do not visibly connect
with, the parent dike. In several places along strike, two
segments diverge from the parent dike and enclose lenses
of host tuff. In the northern half of the PR dike, where its
shallowest intrusion level is preserved at Slanted Buttes
(Fig. 3), it splits into two parallel dikes with contorted
foliation patterns and encloses fault breccia. The breccia is
in smooth contact with the dike and contains vesiculated
basalt and fused tuff in a type of peperite. Also at this tip,
elongated vesicles in the dike are folded into an open fold
with a nearly horizontal axis.
221
PR dike contains ubiquitous near-vertical joints that
result in a pervasive platy texture with plates parallel to
the dike–host contact. Conversely, with the exception of
local cooling joints in fused wall rock (extending 10–
20 cm into the wall rock, perpendicular to the dike
margin), joints are nowhere visible in the host rock
along the length of the dike. The contact between basalt
and the tuff host rock is consistently smooth and shows
no brecciation. Along strike at this contact, the tuff host
rock is fused or welded to varying degrees [15]. In
places the tuff is densely welded and forms black vitrophyre that grades rapidly away from the contact, over a
distance of ∼ 20–100 cm, into non-welded tuff that is
apparently unaffected by the dike. In other places the
tuff is only partially welded at the contact and black
fiamme are flattened parallel to the contact. We infer that
this “contact welding” is the combined result of heat
from the dike and compressive stress exerted by the
magma on the wall rocks [14]. Welded host rock commonly contains vesicles that are elongated vertically and
parallel to the margin. In some places, welded tuff coats
the basalt and displays rills or elongate smooth ridges
(flutes). Most rills plunge nearly vertically, however, a
sub-horizontal rill is present in the central part of the
dike (Fig. 4).
The two eastern dikes, M and E (Fig. 1), show
geometries and textures similar to those of PR; however,
M dike is much shorter and does not feed a sill, and E
dike feeds two sills. M dike radiates out from the necklike body mentioned above (Fig. 1) and visibly occupies
a normal fault, oriented 25°N–40°W, 61°E, that displaces the Tertiary tuff. M dike varies in width between
2 and 4 m, and has an exposed length of 400 m. Its host
rocks are only Tertiary tuffs that show no brecciation or
jointing near the dike contact. Near the neck M dike is a
single continuous dike, whereas farther from the neck it
is composed of several segments. The north end of the
dike veers west and narrows before terminating. Dike
texture is characterized by a vertical platy fabric that
parallels the dike margins.
E dike is the easternmost dike studied within the
graben. Like the M dike, it extends from the neck and is
emplaced into Tertiary tuffs. Near the neck the dike
visibly occupies a NNW-trending, E-dipping normal
fault that displaces bedded tuffs 3.5 m and does not cut
the dike. The exposed dike length is 2020 m and the
width varies from 3 m near the northern tip, to 6 m near
the southern (ES) sill. A zone comprising about 30 m
along strike forms a pod with about 20 m width, presumably representing localized wall rock erosion, between the neck and ES sill. E dike is less segmented
than PR dike, and is composed of fewer than a dozen
222
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
Fig. 3. (A) Photograph of PR dike, looking northward toward Slanted Buttes along the dike's strike. Dashed lines indicate margins of the dike, dotted
line indicates PR fault where it is not occupied by basalt. Photo by D. Krier. (B) Tip region of PR dike, looking up the slope of Slanted Buttes. Here the
dike forms segments with variable width, enclosing fault breccia and with local peperite zones. Person is ∼ 1.9 m tall. Dotted line indicates PR fault
above the dike tip.
segments. In at least three places near the contact with
the Timber Mountain Tuff (Fig. 1) and one location near
the neck, the dike locally diverges to form two segments
that engulf an 8-m-wide lens of the tuff host rock.
The texture of E dike is characterized by the pervasive
vertical platy fabric common to M and PR dikes. Adjacent
host tuffs are neither jointed nor brecciated, except for
local vertical jointing of the Rainier Mesa member of the
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
223
nation shows vertical rills. At its northern termination, the
dike ends in three left-stepping en echelon segments. The
margins of these segments contain densely welded tuff host
rock (black vitrophyre) and occasional patches of peperite.
Flow direction indicators (Fig. 4) such as flutes and rills
in PR dike are dominantly vertical and therefore indicate
simple vertical flow. Similar features in E dike suggest that
flow in the northern part of the dike was dominantly
vertical, while toward its southern end flow was lateral.
We suggest that the dike was emplaced first by flowing up
along a fault plane, but flow gradually became focused at
the southern end of the dike to form the volcanic neck. As
this conduit became established, magma flowing up the
southern end of the dike was diverted toward the conduit,
resulting in lateral flow indicators. As the conduit was
established, it exerted pressure on its walls. Because of the
conduit's roughly cylindrical shape, this pressure reoriented the stress field around the conduit, such that
maximum horizontal compression was in a radial direction
and minimum horizontal compression was circumferential, resulting in the formation of radial dikes (Fig. 1), a
process that is especially well documented for stratovolcanoes [32–34]. One of these radial dikes was “captured”
by a preexisting fault as it propagated away from the neck
to form M dike. This sequence of events, where magma
rises in a dike but gradually becomes focused into a central
conduit, is very common for basalts [4,35,36].
3.2. Dike emplacement mechanics
Fig. 4. Lower hemisphere stereogram of structures at Paiute Ridge.
Great circles show surfaces and kinematic indicators: fm = fault surface
confining m dike and slickenside; de = E dike with flow indicator;
dp = PR dike with flow indicator, Sen = layering in EN sill, Sw = layering in W sill. Isolated dots represent poles to dike orientations: p = PR
dike, e = E dike, m = M dike; and + indicate layering: w = layering in W
dike, b = layering in tuff adjacent to intrusive bodies, e = layering in E
dike. px is an anomalous fabric within PR dike that is perpendicular to
its strike. Fabric elsewhere along dikes is coplanar with the dike
margin.
Timber Mountain Tuff (related to post-emplacement
cooling of that ignimbrite) that is intruded by the dike at
its shallowest level. The contact of the dike and host tuff is
well exposed in places and varies from partially to densely
welded in the same manner as described above for the PR
dike. Where dense welding has occurred, vesicles in the
host tuff are vertical and parallel the dike margin. Contact
welding of the host tuffs formed oblate fiamme that
parallel the dike–host contact. Visible thermal effects on
the wall rocks disappear within 1 m of the dike margin.
The southern termination of E dike, near the volcanic
neck, shows sub-horizontal rills on the near vertical dike–
host contact, while the same contact at the northern termi-
Decades of observations of dike geometries have documented that, in the absence of local stress perturbations,
dikes typically form normal to the direction of the least
compressive stress, σ3. To propagate a vertical dike, σ3
must be horizontal and the magma pressure (Pm) within a
dike must exceed σ3 [37]. In order to inject into a fracture
dipping less than 90° but still perpendicular to σ3, magma
pressure in the crack must exceed σ3 plus a component of
the vertical stress, σv.
If the strike of a pre-existing fracture plane is not
normal to the direction of σ3, magma pressures must be
greater than σ3 in order to dilate the fracture. Delaney et al.
[5] determined that if Pm is only slightly greater than σ3,
or if differential stresses are high [37], only cracks nearly
perpendicular to σ3 can be dilated. If, on the other hand,
Pm is greater than the greatest horizontal compressive
stress, a crack of any orientation can be dilated. The normal
and normal–oblique faults in our study area that were
dilated by dikes formed between 11.45 Ma (the age of the
Ammonia Tanks member of the Timber Mountain Tuff,
the youngest of the Miocene tuffs deposited at Paiute
Ridge and subsequently displaced by the faults) and
224
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
8.5 Ma under a stress field characterized by a SW- to
WSW-directed σ3 [29,31]. Subsequent to this faulting
event and into present times, the regional stress field underwent a gradual clockwise rotation of 65°, resulting in a
NW-directed σ3 [29,38,39]. The dikes in the Paiute Ridge
complex were probably emplaced prior to the stress field
rotation, based on the following information: (1) The age
of the Paiute Ridge basalts is 8.59 ± 0.07 Ma, which is
within (although near the end of) the period of WSWdirected σ3. (2) Dikes were emplaced exclusively along
NNW-striking faults, which would probably have required
unusually high magma pressures if they were emplaced
during the later NW-directed σ3 unless the horizontal stress
differential was very low. Therefore we assume that the
faults that are occupied by the dikes were approximately
perpendicular to σ3 at the time of magma emplacement.
The dikes occupy only steep (N60°) east-dipping fault
planes. For example, PR dike occupies a fault with a near
vertical dip of 88°, but near its southern end the fault
shallows to a dip of 50° and contains no basalt. Other
preexisting fault surfaces with shallower or west dips were
not intruded. Although dips vary along the dikes we
describe here, representative values of 88° (PR dike), 61°
(M dike), and 71° (E dike) yield fault-normal vertical
stress components of 0.2 MPa, 2.1 MPa, and 1.4 MPa,
respectively, using an average density of 1770 kg/m3 for
the Tertiary tuffs (silicic tuffs with ∼25% porosity).
For a quasistatic dike in an elastic medium the ratio of
maximum width (w) to dike length (l) of a dike is related
to the pressure available to deform the dike walls and
the resistance of the host rock to deformation [3,37], as
expressed by:
w
Pd
¼
l
l=ð1−mÞ
ð1Þ
[7,40], where Pd = Pm − σ3 is the driving pressure, μ is the
elastic shear modulus, and ν is Poisson's ratio. Estimated
driving pressures for the PR, M, and E dikes, based upon
Eq. (1) and accounting for a component of vertical stress
due to dips of the host fault planes and for the material
properties of the tuff host rock (the three dikes only are
exposed where hosted by tuffs), are given in Table 2.
These estimated driving pressures are minimum values
because the country rock might have relaxed inward due
to waning magma pressure as the intrusive event came to
a close (i.e., the preserved thickness may be less than the
active thickness during magma emplacement).
Dike driving pressures are commonly estimated to be
on the order of a few MPa (e.g., Delaney and Gartner [7]
estimate Pd ≈ 5 MPa for basaltic dikes of the San Rafael
Swell, UT); the values for PR, M, and E dikes exceed
this value. It is unlikely that dikes can sustain pressures
larger than the host rock tensile strength because the
pressure would simply be accommodated by continued
upward dike propagation as the host rock fails at the
dike tip [42] (note that tensile strength likely represents a
maximum pressure criterion as used here, since tensile
stresses at a dike tip are very high even at low values of
Pd [3]). Host tuff tensile strengths are expected to be on
the order of a few MPa, based on comparison with similar
tuffs at Yucca Mountain, Nevada [43]. The lowest value
of Pd among the three dikes is 7.7 MPa (PR dike), which
may be feasible but is likely close to or above the tensile
strength of the host tuffs, but the calculated values of Pd
for M and E dikes are clearly unrealistically high, and this
directly reflects the w/l values that we measured.
There are four possible reasons why the three dikes
depart from the assumptions embodied in Eq. (1) (quasistatic fluid-filled crack in an elastic medium).
(1) Inelastic deformation. The contact welding described above, where a zone of tuff immediately
adjacent to the dike–tuff contact was heated and
compressed by the dike such that its pore space was
irrecoverably closed (see also WoldeGabriel et al.
Table 2
Calculated magmatic driving pressures (Pd) for PR, M, and E dikes
accounting for elastic deformation only
Dike
Length
(m)
Width
(m)
Dip
(deg)
Fault-normal
component σa
(MPa)
Estimated Pd in
Tertiary rocksb
(MPa)
PR
M
E
4600
400
2020
5
4c
6
88
61
71
0.2
2.1
1.4
7.7
72
22
a
Calculated for a depth of 250 m beneath Tertiary tuffs with average
density of 1770 kg/m3.
b
Calculated using Eq. (1) plus a fault-normal component of σv.
Values of elastic shear modulus (μ) and Poisson's ratio (ν) were
calculated from longitudinal and shear wave velocities measured on
samples of Tertiary tuffs from wells Ue4g#2 and Ue4b (located
∼ 6.5 km west of Slanted Buttes), as reported in van Burkirk et al. [41].
Average values of μ = 4.3 GPa and ν = 0.38 were used to calculate Pd.
Values of these elastic constants, as calculated from data on individual
samples [41], range from μ = 2.8 GPa, ν = 0.41 to μ = 6.8 GPa, ν = 0.34.
The conclusion reached in the text that elastic deformation alone
cannot produce the observed dike width / length ratios is true for the
full range of elastic constants (with the exception of the PR dike if the
lowest value of μ is assumed). Because the tuff host rocks are “soft”
due to relatively high matrix porosity and weak to moderate induration,
the laboratory-scale data are thought to be representative of the larger
rock masses in the field, differently from crystalline rocks where
elastic properties derived on small-scale (ca. tens of cm) samples result
in overestimates of the shear modulus of field-scale rock masses that
include multiple fractures [3]. The values reported here supersede
those in Carter Krogh and Valentine [24].
c
Maximum thickness of M dike measured near volcanic neck.
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
[14]), indicates that inelastic deformation did occur.
The widest zone of contact welding observed is
∼1 m wide, and represents a gradation from completely welded, dense vitrophyre within 2–3 dm of
the contact, to unaffected tuff. Assuming that such a
zone experienced a 25% volume reduction due to
pore collapse, on both sides of a dike, this inelastic
deformation could account for ∼0.5 m of the dike
width.
(2) Erosion of wall rocks by flowing magma in the
dikes or by thermal spalling and subsequent
entrainment. Local pinching and swelling of dike
width, particularly where it occurs on only one side
of a dike, could be an indicator of this process.
Other dikes in the Paiute Ridge complex clearly
show “arrested” wall rock erosion processes [23].
Valentine and Groves [13] estimated dike widening
by erosion and entrainment for alkali basaltic eruptions in the Lucero Volcanic Field, and found that
the erosional component of dike widening in very
competent host rocks is, on average, b 10 cm (note
that this widening was probably quite variable on a
local scale) at depths of a few hundred meters. It
seems reasonable to assume that the relatively weak
tuff host rocks of the PR, M, and E dikes at the
Paiute Ridge complex may have been eroded an
average of a few decimeters. This erosion would
likely have occurred during initial dike emplacement while the host rocks in contact with magma
were still brittle, rather than after heating which
would result in more plastic behavior of the tuffs at
the contact. Meltback and fluid entrainment of
melted wall rock [36] might also have contributed
to this erosional component of dike widening.
3. Transient eruptive phenomena might have produced short-lived pressurization events in the neck
that were transmitted into the connected M and E
dikes. The neck might have been a feeder conduit
to a small basaltic volcano, most likely a scoria
cone, that produced the pyroclastic deposits and
lava flow remnants that are preserved just to the
east–northeast of the neck (Fig. 1). Scoria cone
vents commonly experience temporary blockage
due to avalanching of pyroclastic material [44,45]
or temporary solidification of a magma plug [46],
which can result in overpressures of 5–10 MPa in
the plumbing system prior to failure and ejection
of the blockage [47,48]. Since this pressurization
phenomenon would only have affected dikes
connected to the neck, this may partially, but not
completely, explain the high calculated values of
Pd for M and E dikes compared to PR dike.
225
4. Structural extension during magma emplacement
might have provided additional space for magma.
Emplacement of magma might have relaxed the
friction across the pre-existing host fault planes
(lubrication), allowing slip and structural extension
during the magmatic event. Evidence for co-magmatic fault slip would not be preserved within the
dikes themselves because they would have been
fluid at the time, except at their upper tips which
might have been cooler (viscous or plastic) and
where the dike might have interacted with the upward (but magma-free) continuation of the fault.
Only the PR dike has part of its upper tip preserved,
where the dike extends into Slanted Buttes (Fig. 1).
At this location, as described above, the dike tip
consists of two parallel segments with contorted
internal foliation and with fault breccia mixed with
basalt clasts between them. These features may be
due to contemporaneous fault displacement and
dike emplacement.
Subtracting estimated contributions of the above
processes to the dike width, we can then estimate the
driving pressure needed to achieve the remaining component of width by elastic deformation, embodied in Eq.
(1). Pd for PR dike is reduced to ∼6 MPa if we assume
that about 1 m of the average dike width is due to a
combination of inelastic deformation and mechanical
erosion of wall rocks. If an additional 1 m is assumed to
result from syn-emplacement extension (perpendicular to
the fault), Pd ≈ 4.7 MPa, which is very reasonable with
respect to expected tensile strength values of the host tuff.
Assuming similar inelastic and erosion components of
dike width for M and E dikes, about 2.5–3 m of synemplacement extension on each host fault would reduce
Pd to ∼8–10 MPa. Such a range in Pd is reasonable to
expect from temporary plugging of the volcanic neck to
which they are connected, as discussed above. Thus, if the
hypothesis that structural extension accompanied dike
emplacement is true, the three dikes imply a total of ∼6–
7 m of extension in the Paiute Ridge horst and graben
system during the magmatic event. Note that other studies
of extension and normal fault activation in rift systems,
such as at Hawaii or Iceland [40,49,50], focus on the
effects of the transient stress field around a rising vertical
dike, a somewhat different process than the magma
lubrication considered here. Our hypothesis also departs
from that of Parsons and Thompson [8], who argue that,
on a regional scale, emplacement of dikes (which they
assume to be dominantly vertically oriented) can accommodate regional extension, thereby suppressing normal
faulting and associated seismicity. The three dikes we
226
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
have described at Paiute Ridge suggest that if a dike can
occupy a pre-existing normal fault, the reduction in friction may actually induce slip and substantial extension. It
will be interesting to further test this hypothesis at additional sites and a range of scales.
4. Sills
4.1. Description of W, ES, and EN sills
Subhorizontal basaltic sills that are roughly planar to
dish-shaped were emplaced into non-welded bedded
tuffs (Fig. 1). Sills occur consistently as extensions off
fault-hosted dikes into the hanging wall block of the
NNW-trending normal faults. The sills were injected
exclusively into tuffs at positions near the contact between the Paleozoic sedimentary rocks and overlying
tuffs.
The westernmost sill, W, forms east of PR dike in the
hanging wall of the Paiute Ridge fault in an area where
the contact between Tertiary tuffs and Paleozoic rocks
forms a swale on the side of a larger carbonate paleohill
that rises to the south. It is fed by PR dike, which widens
near the sill and can be traced to within about 3 m of the
sill's western margin. Based on reconstructions (Fig. 2),
W sill was emplaced in non-welded bedded tuffs within
tens of meters above the presently buried Paleozoic–
Tertiary contact. The minimum thickness of W sill is
difficult to determine because of erosion of the uppemost part of the sill and overlying tuffs, but is estimated
at about 3 m. Its maximum thickness appears to be
less than 20 m. The sill is roughly circular (diameter
∼ 500 m) in plan view and dish shaped in cross
section. The depth of sill intrusion was at most 250 m
below the earth's surface (Fig. 2), and ∼ 160 m below
the highest level reached by the parent PR dike (as
currently exposed about 1.5 km north of the sill).
The lower contact of W sill is generally conformable
with bedding of the host tuffs that mantle the abovementioned paleoswale, as is the upper sill–tuff contact
where it is preserved. Textures within the sill are highly
variable. At the western margin, the base of the sill has
a pervasive platy fabric in which the plates (a few cm
thick) are parallel to the subhorizontal, S-dipping tuff–
sill contact. Where this contact is irregular the basalt is
either not jointed or has joints that follow the shape of
the contact. Above the basal contact, the platy fabric
alternates with thicker (N 10 cm) subhorizontal zones of
massive basalt. Just north or south of the point where the
dike and sill connect, the subhorizontal platy fabric is
absent; in two places, 1.5 m and 10 m above the basal
contact, the basalt is either irregularly vertically jointed,
or has a chaotic, jumbled appearance. On the north side of
the sill, where the sill and tuff dip on average 44° toward
the south, rills plunging toward 235° are preserved on the
under side of the sill. The margins of the sill dip moderately inward on all sides. Thin, discontinuous sills,
b1 m thick and b 10 m long, are locally present beneath
the main W sill; these parallel the main sill and are
separated from it by 2–3 m of country rock.
Two sills, ES and EN, were injected into the hanging
wall of the fault utilized by E-dike. The sills are fed by E
dike, which widens near the sill contacts and can be
traced directly into the sills. Based on reconstructions
(Fig. 2), the southernmost sill, ES, was injected into
bedded tuffs (Fig. 1) within ∼20 m of the presently
buried Paleozoic–Tertiary contact. ES is approximately
450 m in the long dimension and is up to 24 m thick
(Fig. 5). The depth of intrusion was probably ∼ 170 m
below the earth's surface, only 60 m below the shallowest level attained by the parent dike elsewhere along
its strike.
Along its lower contact with the non-welded bedded
tuffs, ES sill is parallel with the trend of bedding in some
places, and is disconformable elsewhere. Tuff host rock
underlying the ES sill contains fractures normal to the
sill–tuff contact. The fractures typically have sharp lower
tips and extend into the host tuff less than 2.5 m. Through
the center of the sill, a set of steeply E-dipping faults,
comprising a N-trending normal fault zone, displaces ES
sill approximately 4 m down to the east.
EN, the northernmost sill attached to E dike (Fig. 1),
is emplaced in non-welded bedded tuffs just above the
Paleozoic–Tertiary contact and continues laterally
toward the east to intersect a horst of fractured Ordovician carbonate rocks. EN is approximately 380 m long
and as much as 46 m thick. The depth of intrusion was
probably less than 100 m below the earth's surface, at
approximately the same height as that attained by the
parent dike at its northern tip. EN sill appears generally
discordant with bedding of the underlying tuff. Rills
exposed on the underside of the basalt in contact with
tuff plunge toward 227° on an EW-striking, 68°S
dipping surface (Fig. 4).
Three additional sills are only briefly described here
because their relationships to feeder dikes are not clear.
One of these is a dish-shaped body intruded into tuff,
with a plan-view diameter of ∼ 1 km, which occupies
the southeastern part of the Paiute Ridge intrusive complex (Fig. 1). This sill is up to 150 m thick [15], and
locally differentiated to form coarse pods of syenite. Its
tuff roof rocks are highly altered but tend to dip inward
toward the center of the sill area. A similar, but thinner,
dish-shaped sill occupies the northeastern part of the
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
227
Fig. 5. Photograph of ES sill, taken from the volcanic neck looking northward along the strike of E dike. Photo by G. Keating.
intrusive complex (this sill has been referred to as the
Papoose Lake sill by Matyskiela [51]). A third sill was
emplaced just beneath the lava flow remnant in the eastcentral part of the complex. All three of these sills and
the lava flow remnant occupy local structurally low
areas with respect to the Paleozoic–Tertiary contact,
near the axis of the horst and graben system.
4.2. Sill emplacement mechanics
Sill geometries and spatial distributions in the Paiute
Ridge area are not adequately explained by commonly
invoked mechanisms for sill formation, such as lateral
diversion of magma at the level of neutral buoyancy or
LNB [1] or at the point of dike induced stress reorientation [8,9]. The buoyancy force in an LNB mechanism
depends on the integrated density of vertical columns
through the dike and through adjacent wall rock to the
surface, not on the local density contrast between
magma and immediately adjacent wall rock. With this
in mind, it seems unlikely that an LNB mechanism
could have acted along such localized segments of the
main dikes, although Crowe et al. [11] did invoke this
mechanism to explain the relationships at Paiute Ridge.
In addition, along the PR and E dikes' lengths, there is
evidence that magma continued to rise to a level significantly higher than the sills, and, the sills propagated
from only one side of the dikes. Similar arguments can
be made against the mechanism of dike-induced stress
reorientation [8,9] for sills at Paiute Ridge. This mechanism would most likely have caused sills to form along
most of the length of a dike, barring any externally (i.e.,
non-dike related) induced major stress heterogeneities
along the length of the dike.
Two requirements must be met for sill injection.
First, the driving pressure must have exceeded either the
strength of the rock or the tensile strength of a subhorizontal pre-existing plane of weakness, such as a bedding plane, plus the vertical stresses at the level of sill
injection (∼ 100–250 m depth). As mentioned above,
tensile strengths for the tuff host rocks are expected to
be on the order of a few MPa. Pre-existing weaknesses
(bedding planes) would tend to have lower tensile strengths
than intact rock. We assume that the shallow sill depths
allowed for interaction with the earth's free surface (also
discussed below) and that if the driving pressure was
sufficiently high, the magmas were able to mechanically
lift the roof rocks, rather than creating space primarily by
elastic deformation. Vertical (overburden) stresses at this
depth can be estimated, using σv = ρgh, at roughly 1.7–
4.3 MPa in the Tertiary tuffs. The driving pressures estimated above for the sill-feeding dikes suggest that this
requirement could easily have been met.
The second requirement for sill emplacement is that
the least compressive stress (σ3) must have a subvertical
orientation. The W, ES, and EN sills occur only locally
228
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
along their feeder dikes. Since the feeder dikes require a
subhorizontally oriented σ3, the sills imply very localized
rotation of the stress field. We suggest that this may result
from the fact that the dike-hosting faults cut a complex
contact between Paleozoic carbonates and Tertiary tuffs,
which have very different mechanical properties. As described above and illustrated in Fig. 2, the carbonates
formed a hilly terrain upon which the tuffs were deposited,
partly mantling and partly ponding in lows and smoothing
the terrain. The normal faults intersect that contact, so that
at a given depth, the rocks that contact each other across
the fault vary along the fault's strike. In addition, the dips
of the faults likely vary depending upon the host rock
properties (e.g., the southern end of the fault that hosts PR
dike shallows substantially as it extends from tuffs into a
carbonate paleohill). Such dip variations could result in
local asperities or waviness of the fault plane, for example
where a stronger rock (carbonate) on the footwall extends
into the weaker rock (tuff) of the hanging wall. The
combination of varying footwall and hanging wall rock
contrasts along strike at a given depth, and asperities on
the fault plane, could result in local rotation of the
principal stresses (for example, a small area along the fault
plane may have higher friction than adjacent areas).
Chester and Chester [52] present stress field solutions
around a wavy fault surface and show that such stress
rotations are expected. Their results indicate that the
maximum stress rotates to an orientation that is oblique to
the fault near an asperity, and this is enhanced by decreasing fault friction. Therefore, this effect might be
exacerbated during magma emplacement if slip is actively
occurring along the faults due to reduction of friction (see
dike emplacement discussion above), as parts of the fault
are “hung up” while other parts are dilating and slipping.
Such a process, occurring locally along a fault, may result
in lateral diversion of magma.
While local rotation of stresses determined whether
magma flowed upward or laterally into sills, the consistency with which sills were emplaced into the Tertiary
tuffs within the hanging walls of faults (Fig. 2), just above
the contact with Cambrian carbonates, may be related to
the orientation of bedding planes. Adjacent to PR dike,
bedding planes are generally W-dipping in both the Cambrian and Tertiary rocks. The dip direction of these potential planes of weakness would favor the injection of
sills upwards toward the east, rather than downwards
towards the west (Fig. 2) similar to the fluid-filled cracks
that seek the earth's surface analyzed by Weertman [53].
Based on an analogous structural setting, it is likely that
the same mechanical configuration controlled the injection of the two sills fed by E dike. Thus, the consistent
occurrence of sills in particular geometric situations that
are highly spatially localized suggests that their emplacement is controlled by the small scale (tens of m along dike
strike) variations in ambient stresses and by rock structure.
Three of the sills in the Paiute Ridge complex dip
inward around their margins to form dish-shaped bodies
(lopoliths) and their roof rocks also dip inward. The sagging roofs of these bodies suggest that the magma interacted with the overlying free surface. This geometry is not
shared by the two smaller sills (ES and EN) fed by E dike.
We suggest, following results of the laccolith study by
Pollard and Holzhausen [54], that the size of the sills is an
important factor in the formation of the lopolithic geometry at Paiute Ridge. Pollard and Holzhausen [54]
calculate that the transition from a sill to a laccolith occurs
at a ratio of half-length to depth (L/d) of approximately
1.6. L/d of W sill is between 1.0 and 1.7 (depending on the
estimated depth); the L/d of the largest sill is probably N 3.
EN and ES sills are not lopoliths and have L/d ratios
between 1 and 2, but these sills have a much smaller area
than the lopoliths and are elongate in plan view rather than
the more equant lopoliths. Therefore, the areal extent and
shape of the sill, in addition to simply the L/d ratio, must
also be important factors in whether it becomes a lopolith.
There are three possible reasons why the roofs of Paiute
Ridge intrusions sag inward (forming lopoliths) rather than
bow upward. (1) The magma may have been vesicular at
these shallow depths during intrusion, so that while
magma initially pushed the roof up, as the volatiles leaked
out and the intrusion cooled the roof gradually sagged
downward. (2) The host rocks were originally non-welded
silicic tuffs, but the heat and pressure of the intrusions
caused welding, pore collapse, and attendant decrease in
rock volume adjacent to the intrusions. This accounts for
inward sagging of both the roofs and floors of the lopoliths.
(3) The lopoliths formed where tuffs mantled paleotopographic lows, such that bedding surfaces, which formed
weak zones through which magma intruded and could
move toward the free surface, had a bowl-like geometry.
As described above, this appears to be the case for W sill.
5. Conclusion
The location and geometry of Miocene basaltic
intrusive bodies in the Paiute Ridge area were strongly
controlled by pre-existing structures and local stresses.
The longest dikes intruded normal to σ3, along preexisting, NNW-striking and E-dipping faults. Preserved
dike widths resulted from a combination of inelastic
deformation and erosion of wall rocks, and syn-intrusive
extension, as well as elastic wall rock deformation. Two
of the dikes are connected to a basalt neck that probably
represents a volcanic conduit, and they may have
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
experienced transient high pressures due to periods of
blockage of the overlying eruptive vent. Sills, which
occur locally along the dikes within 250 m of the paleosurface, consistently occur in the hanging wall block
of the dike-injected faults (see Fig. 2). The sills reflect
local rotation of the stress field due to roughness or
asperities on the planes of the dike-hosting faults. Dips
of bedding surfaces determined the direction of sill
injection: magma flowed upward towards the east rather
than downward toward the west. The three largest sills
with roughly equal lateral dimensions formed lopoliths
indicating that the shallow level and large areal extent of
the sills allowed interaction with the free surface.
The strong and highly localized control by preexisting structure on shallow dike and sill emplacement
at the Paiute Ridge complex is a departure from more
common models that invoke the level of neutral buoyancy or dike-induced stress rotations as determining
factors of dike versus sill formation. The latter mechanisms are more likely to be important on regional scales
and at the scale of the thickness of the earth's crust,
while the processes recorded at Paiute Ridge are most
likely in the uppermost few hundred meters of an existing extensional system with heterogeneous materials.
We speculate that the processes that governed dike
propagation and local shallow sill formation at Paiute
Ridge may operate in other shallow rift settings.
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Acknowledgments
[15]
We thank Claudia Lewis, Gordon Keating, Don Krier,
Branco Damjanac, and Ed Gaffney for very helpful
reviews of the manuscript. This work was funded by the
U.S. Department of Energy. KECK was supported in part
by an appointment to the U.S. Department of Energy
Distinguished Postdoctoral Research Program sponsored
by the U.S. Department of Energy, Office of Science
Education Programs, and administered by the Oak Ridge
Institute for Science and Education. Artwork in the paper
is by Andrea Kron.
References
[1] J.R. Lister, R.C. Kerr, Fluid–mechanical models of crack
propagation and their application to magma transport in dykes,
J. Geophys. Res. 96 (1991) 10,049–10,077.
[2] A.M. Rubin, Tensile fracture of rock at high confining pressure:
implications for dike propagation, J. Geophys. Res. 98 (1993)
15919–15935.
[3] A.M. Rubin, Propagation of magma-filled cracks, Annu. Rev.
Earth Planet. Sci. 23 (1995) 287–336.
[4] P.T. Delaney, D.D. Pollard, Deformation of host rocks and flow
of magma during growth of minette dikes and breccia-bearing
[16]
[17]
[18]
[19]
[20]
[21]
[22]
229
intrusions near Shiprock, New Mexico, U.S. Geol. Surv. Prog.
Pap. 1202 (1981) 61.
P.T. Delaney, D.D. Pollard, J.I. Ziony, E.H. McKee, Field
relations between dikes and joints: emplacement processes and
paleostress analysis, J. Geophys. Res. 91 (1986) 4920–4938.
G. Baer, M. Beyth, Z. Reches, Dikes emplaced into fractured
basement, Timna Igneous Complex, Israel, J. Geophys. Res. 99
(1994) 24039–24050.
P.T. Delaney, A.E. Gartner, Physical processes of shallow mafic
dike emplacement near the San Rafael Swell, Utah, Geol. Soc.
Amer. Bull. 109 (1997) 1177–1192.
T. Parsons, G.A. Thompson, The role of magma overpressure in
suppressing earthquakes and topography: worldwide examples,
Science 253 (1991) 1399–1402.
T. Parsons, N.H. Sleep, G.A. Thompson, Host rock rheology
controls on the emplacement of tabular intrusions: implications
for the underplating of extending crust, Tectonics 11 (1992)
1348–1356.
G.A. Valentine, Note on the distribution of basaltic volcanism
associated with large silicic centers, J. Volcanol. Geotherm. Res.
56 (1993) 167–170.
B.M. Crowe, S. Self, D. Vaniman, R. Amos, F. Perry, Aspects of
potential magmatic disruption of a high-level radioactive waste
repository in southern Nevada, J. Geol. 91 (1983) 259–276.
S.G. Wells, L.D. McFadden, C.E. Renault, B.M. Crowe,
Geomorphic assessment of late Quaternary volcanism in the
Yucca Mountain area, southern Nevada: implications for the
proposed high-level radioactive waste repository, Geology 18
(1990) 549–553.
G.A. Valentine, K.R. Groves, Entrainment of country rock during
basaltic eruptions of the Lucero volcanic field, New Mexico,
J. Geol. 104 (1996) 71–90.
G. WoldeGabriel, G.N. Keating, G.A. Valentine, Effects of shallow
basaltic intrusion into pyroclastic deposits, Grants Ridge, New
Mexico, USA, J. Volcanol. Geotherm. Res. 92 (1999) 389–411.
G.N. Keating, J.W. Geissman, G.A. Zyvoloski, Multiphase modeling
of contact metamorphic systems and application to transitional
geomagnetic fields, Earth Planet, Sci. Lett. 198 (2002) 429–448.
G.A. Valentine, D. Krier, F.V. Perry, G. Heiken, Scoria cone
construction mechanisms, Lathrop Wells Volcano, southern
Nevada, USA, Geology 33 (2005) 629–632.
G.A. Valentine, F.V. Perry, D. Krier, G.N. Keating, R.E. Kelley,
A.H. Cogbill, Small-volume basaltic volcanoes: eruptive products and processes, and post-eruptive geomorphic evolution in
Crater Flat (Pleistocene), southern Nevada, Geol. Soc. Am.
Bull., in press.
F.V. Perry, A.H. Cogbill, R.E. Kelley, Uncovering buried
volcanoes at Yucca Mountain: new data for volcanic hazard
assessment, Eos, Trans. - Am. Geophys. 86 (2005) 485–488.
G.A. Valentine, C.D. Harrington, R.E. Kelley, Clast size controls
and longevity of Pleistocene desert pavements at Lathrop Wells
and Red Cone volcanoes, southern Nevada, Geology 34 (2006)
533–536.
B.M. Crowe, K.H. Wohletz, D.T. Vaniman, E. Gladney, N.
Bower, Status of Volcanic Hazard Studies for the Nevada Nuclear
Waste Storage Investigations, Los Alamos National Laboratory
Report, LA-9325-MS, vol. II, 1986.
F.M. Byers, H. Barnes, Geologic Map of the Paiute Ridge
Quadrangle, Nye and Lincoln Counties, Nevada, U.S. Geol.
Surv. Quad. Map, vol. GQ-577, 1967.
G.A. Valentine, B.M. Crowe, F.V. Perry, Physical processes and
effects of magmatism in the Yucca Mountain region, Proc. 1992
230
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
G.A. Valentine, K.E.C. Krogh / Earth and Planetary Science Letters 246 (2006) 217–230
Conf. High Level Rad, Waste Management, American Nuclear
Society, 1992, pp. 2014–2024.
G.A. Valentine, K.R. Groves, C.W. Gable, F.V. Perry, B.M.
Crowe, Effects of magmatic processes on the potential Yucca
Mountain repository: field and computational studies, Proc.
Focus '93 Site Characterization Model Validation, American
Nuclear Society, 1993, pp. 167–173.
K.E. Carter Krogh, G.A. Valentine, Structural Control on
Basaltic Dike and Sill Emplacement, Paiute Ridge Mafic Intrusion Complex, Southern Nevada, Los Alamos National Laboratory Report, vol. LA-13157-MS, 1996, 22 pp.
C.D. Ratcliff, Paleomagnetic and rock magnetic study of the
Paiute Ridge subvolcanic center, southern Nevada, M.S. Thesis,
University of New Mexico, Albuquerque, NM, 1993, 64 pp.
C.D. Ratcliff, J.W. Geissman, F.V. Perry, B.M. Crowe, P.K.
Zeitler, Paleomagnetic record of a geomagnetic-field reversal
from late Miocene mafic intrusions, southern Nevada, Science
266 (1994) 412–416.
D.T. Vaniman, B.M. Crowe, E.S. Gladney, Petrology and
geochemistry of hawaiite lavas from Crater Flat, Nevada,
Contrib. Mineral. Petrol. 80 (1982) 341–357.
F.V. Perry, B.M. Crowe, G.A. Valentine, L.M. Bowker,
Volcanism Studies: Final Report for the Yucca Mountain Project,
Los Alamos National Laboratory Report, vol. LA-13478-MS,
1998, 554 pp.
S.A. Minor, Superposed local and regional paleostresses: fault–
slip analysis of Neogene extensional faulting near coeval caldera
complexes, Yucca Flat, Nevada, J. Geophys. Res. 100 (1995)
10507–10528.
T.A. Vogel, F.M. Byers, Introduction to the special section on the
southwestern Nevada volcanic field, J. Geophys. Res. 94 (1989)
5907.
D.A. Sawyer, R.J. Fleck, M.A. Lanphere, R.G. Warren, D.E.
Broxton, M.R. Hudson, Episodic caldera volcanism in the
Miocene southwestern Nevada volcanic field: revised stratigraphic framework, 40Ar/39Ar geochronology, and implications
for magmatism and extension, Geol. Soc. Amer. Bull. 106 (1994)
1304–1318.
K. Nakamura, Volcanoes as possible indicators of tectonic stress
orientation; principle and proposal, J. Volcanol. Geotherm. Res. 2
(1977) 1–16.
F.V. Perry, G.A. Valentine, E. Desmarais, G. Woldegabriel,
Probabilistic assessment of volcanic hazard to radioactive waste
repositories in Japan: intersection by a dike from a nearby
composite volcano, Geology 29 (2001) 255–258.
M.P. Poland, J.H. Fink, L. Tauxe, Patterns of magma flow in
segmented silicic dikes at Summer Coon Volcano, Colorado:
AMS and thin section analysis, Earth Planet. Sci. Lett. 219
(2004) 155–169.
D.H. Richter, J.P. Eaton, K.J. Murata, W.U. Ault, H.L. Krivoy,
Chronological narrative of the 1959–1960 eruption of Kilauea
Volcano, Hawaii, U. S. Geol. Surv. Prof. Pap. 537-E (1970)
E1–E73.
P.M. Bruce, H.E. Huppert, Solidification and melting along
dykes by the laminar flow of basaltic magma, in: M.P. Ryan
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
(Ed.), Magma Transport and Storage, John Wiley and Sons Ltd.,
New York, 1990, pp. 87–101.
D.D. Pollard, Elementary fracture mechanics applied to the
structural interpretation of dykes, in: H.C. Halls, W.H. Fahrig
(Eds.), Mafic Dyke Swarms, Geol. Assoc. Can. Spec. Pap., vol.
34, 1987, pp. 5–24.
B.C. Haimson, J. Lacomb, A.H. Jones, S.J. Green, Deep Stress
Measurements in Tuff at the Nevada Test Site, Advances in Rock
Mechanics, vol. IIa, National Academy of Sciences, Washington
D.C., 1974, pp. 557–561.
A. McGarr, N.C. Gay, State of stress in the earth's crust, Annu.
Rev. Earth Planet. Sci. 6 (1978) 405–436.
D.D. Pollard, P.T. Delaney, W.A. Duffield, E.T. Endo, A.T.
Okamura, Surface deformation in volcanic rift zones, Tectonophysics 94 (1983) 541–584.
R.G. van Burkirk, D.S. Gardiner, S.W. Butters, Characterization
of Yucca Flat materials, Terra Tek Report, vol. 78-67, Terra Tek
Inc., Salt Lake City, Utah, 1978.
Y.A. Fialko, A.M. Rubin, Thermal and mechanical aspects of
magma emplacement in giant dike swarms, J. Geophys. Res. 104
(1999) 23,033–23,049.
Bechtel SAIC Company, Magma dynamics at Yucca Mountain,
Nevada, Las Vegas, Nevada, Bechtel SAIC Company Report
ANL-MGR-GS-000005 Rev 00, Las Vegas, NV, 2005.
T.R. McGetchin, M. Settle, B.A. Chouet, Cinder cone growth
modeled after Northeast Crater, Mount Etna, Sicily, J. Geophys.
Res. 79 (1974) 3257–3272.
G. Heiken, Characteristics of tephra from Cinder Cone, Lassen
Volcanic National Part, California, Bull. Volcanol. 41 (1978)
1–12.
J. Taddeucci, O. Spieler, B. Kennedy, M. Pompilio, D.B.
Dingwell, P. Scarlato, Experimental and analytical modeling of
basaltic ash explosions at Mount Etna, Italy, 2001, J. Geophys.
Res. 109 (2004), doi:10.1029/2003JB002952.
E.S. Gaffney, B. Damjanac, D. Krier, G. Valentine, Deformation
of scoria cone by conduit pressurization, American Geophysical
Union Fall Meeting, San Francisco, 2005, abstract V53B-1570.
B. Damjanac, Z. Radakovic-Guzina, E.S. Gaffney, Conditions
leading to sudden release of magma pressure, American
Geophysical Union Fall Meeting, San Francisco, 2005, Abstract
V33A-0664.
A.M. Rubin, Dike-induced faulting and graben subsidence in
volcanic rift zones, J. Geophys. Res. 97 (1992) 1839–1858.
A.M. Rubin, D.D. Pollard, Dike-induced faulting in rift zones of
Iceland and Afar, Geology 16 (1988) 413–417.
W. Matyskiela, Silica redistribution and hydrologic changes in
heated fractured tuff, Geology 25 (1997) 1115–1118.
F.M. Chester, J.S. Chester, Stress and deformation along wavy
frictional faults, J. Geophys. Res. 105 (2002) 23421–23430.
J. Weertman, The stopping of a rising, liquid-filled crack in the
earth's crust by a freely slipping horizontal joint, J. Geophys.
Res. 85 (1980) 967–976.
D.D. Pollard, G. Holzhausen, On the mechanical interaction
between a fluid-filled fracture and the earth's surface, Tectonophys 53 (1979) 27–57.