RESEARCH ARTICLE
‘Snake River (SR)-type’ volcanism at the Yellowstone
hotspot track: distinctive products from unusual,
high-temperature silicic super-eruptions
M. J. Branney & B. Bonnichsen & G. D. M. Andrews &
B. Ellis & T. L. Barry & M. McCurry
Received: 26 April 2005 /Accepted: 8 March 2007 / Published online: 20 June 2007
# Springer-Verlag 2007
Abstract A new category of large-scale volcanism, here
termed Snake River (SR)-type volcanism, is defined with
reference to a distinctive volcanic facies association
displayed by Miocene rocks in the central Snake River
Plain area of southern Idaho and northern Nevada, USA.
The facies association contrasts with those typical of silicic
volcanism elsewhere and records unusual, voluminous and
particularly environmentally devastating styles of eruption
that remain poorly understood. It includes: (1) large-
volume, lithic-poor rhyolitic ignimbrites with scarce pumice
lapilli; (2) extensive, parallel-laminated, medium to coarse-
grained ashfall deposits with large cuspate shards, crystals
and a paucity of pumice lapilli; many are fused to black
vitrophyre; (3) unusually extensive, large-volume rhyolite
lavas; (4) unusually intense welding, rheomorphism, and
widespread development of lava-like facies in the ignim-
brites; (5) extensive, fines-rich ash deposits with abundant
ash aggregates (pellets and accretionary lapilli); (6) the
ashfall layers and ignimbrites contain abundant clasts of
dense obsidian and vitrophyre; (7) a bimodal association
between the rhyolitic rocks and numerous, coalescing low-
profile basalt lava shields; and (8) widespread evidence of
emplacement in lacustrine-alluvial environments, as
revealed by intercalated lake sediments, ignimbrite peper-
ites, rhyolitic and basaltic hyaloclastites, basalt pillow-lava
deltas, rhyolitic and basaltic phreatomagmatic tuffs, alluvial
sands and palaeosols. Many rhyolitic eruptions were high
mass-flux, large volume and explosive (VEI 6–8), and
involved H2O-poor, low-δ
18O, metaluminous rhyolite mag-
mas with unusually low viscosities, partly due to high
magmatic temperatures (900–1,050°C). SR-type volcanism
contrasts with silicic volcanism at many other volcanic
fields, where the fall deposits are typically Plinian with
pumice lapilli, the ignimbrites are low to medium grade
(non-welded to eutaxitic) with abundant pumice lapilli or
fiamme, and the rhyolite extrusions are small volume silicic
domes and coulées. SR-type volcanism seems to have
occurred at numerous times in Earth history, because
elements of the facies association occur within some other
volcanic fields, including Trans-Pecos Texas, Etendeka-
Paraná, Lebombo, the English Lake District, the Protero-
zoic Keewanawan volcanics of Minnesota and the Yardea
Dacite of Australia.
Bull Volcanol (2008) 70:293–314
DOI 10.1007/s00445-007-0140-7
This paper constitutes part of a special issue dedicated to Bill
Bonnichsen on the petrogenesis and volcanology of anorogenic
rhyolites.
Editorial responsibility: W Leeman
Electronic supplementary material The online version of this article
(doi:10.1007/s00445-007-0140-7) contains supplementary material,
which is available to authorized users.
M. J. Branney (*) :G. D. M. Andrews : B. Ellis
Department of Geology, University of Leicester,
University Road,
Leicester LE1 7RH England, UK
e-mail: mjb26@le.ac.uk
B. Bonnichsen
Idaho Geological Survey, University of Idaho,
Moscow, ID 83844-3014, USA
T. L. Barry
Volcano Dynamics Group, Open University,
Milton Keynes,
MK7 6AA England, UK
M. McCurry
Department of Geosciences, Idaho State University,
Pocatello, ID 83209-8072, USA
G. D. M. Andrews
Volcanology Lab. & MDRU, EOS,
University of British Columbia,
Vancouver, BC, Canada

Keywords Snake River . Yellowstone . Intraplate .
Hot-spot . Ignimbrite .Welded tuff . Rheomorphic .
Super-eruption
Introduction
In this paper we define a new category of large-scale silicic
volcanism that is distinct in character to the more well-
known pumice-rich, Plinian/ignimbrite type of silicic
volcanism that is described extensively in the literature
(e.g. Cas and Wright 1987). The distinctive category of
volcanism has occurred at several times in Earth history, but
is under-represented in the literature, possibly because the
unusual, and particularly environmentally devastating,
eruption styles are not well understood. To highlight its
character, we describe the best-preserved example known to
us, which serves as a defining ‘type’ example, and where
further study would be richly rewarded. We then contrast
SR-type volcanism with silicic volcanism elsewhere,
discuss the enigmatic eruption styles and, finally, list other
volcanic provinces where SR-type volcanism may be
represented.
Large-volume silicic eruptions involving 10s to
1,000s km3 of magma are the most catastrophic form of
volcanism and can cause abrupt regional obliteration and
climatic perturbation (Sparks et al. 2005). Typically, they
involve (1) Plinian explosive eruptions that produce fallout
layers of pumice lapilli with subordinate lithic clasts,
accompanied by (2) emplacement of voluminous pumice
lapilli-bearing ignimbrites, followed by (3) extrusion of
relatively small-volume domes and coulées of degassed
silicic lava. However, the silicic eruption products in some
volcanic fields differ significantly from this general pattern.
The youngest, and best-preserved, example of this is found
in the central part of the Snake River Plain in southern
central Idaho and northernmost Nevada, north-west USA.
There, Miocene rhyolitic volcanism was large-scale and
catastrophic but the assemblage of volcanic deposits is
markedly different to the familiar pumice-rich assemblage
conventionally associated with large Plinian and ignimbrite
eruptions. Unusual characteristics include a paucity of
pumice lapilli, a predominance of particularly intensely-
welded rheomorphic ignimbrites, layers of parallel-strati-
fied coarse ash in which large cuspate shards are commonly
clearly visible in the field, and the presence of exceptionally
large-volume, low-aspect ratio rhyolite lavas. We recognise
that this association of facies represents a distinctive
category of volcanism, and propose the name ‘Snake
River-type’ (SR-type) volcanism, after the c.13 Ma–
c.8 Ma succession in the central part of the Snake River
Plain (Fig. 1), where it is best represented. Elements of the
same facies association occur in other parts of the world
and suggest that SR-type volcanism has punctuated Earth
Fig. 1 The bimodal Columbia
River–Yellowstone volcanic
province of northwest USA
showing the type area of SR-type
volcanism as defined in this paper
(rectangle) and the location of
Miocene–Pleistocene Lake Idaho
(vertical hachure; SE limits yet to
be defined). Northern Nevada
rhyolites after Pierce et al. (2002).
Rhyolitic calderas: Y, Yellowstone
caldera field; N, Newberry; B,
Burns; M, McDermitt. Approxi-
mate outlines of inferred, largely
concealed eruptive centres: OH,
Owyhee-Humboldt; BJ, Bruneau-
Jarbidge; TF, Twin Falls; P, Pic-
abo; H, Heise, after Bonnichsen
et al. (1989) and Morgan and
McIntosh (2005); ages given in
Ma. Locations of outflow ignim-
brite–ashfall successions: R, Rog-
erson Graben; C, Cassia Hills;
BH, Bennett Hills; T, Trapper
Creek–Goose Creek Basin. Dark
grey: extent of Columbia River–
Oregon plateau basalts (including
Steens Mountain, Malheur Gorge
and Picture Gorge basalts). MR,
Magic Reservoir centre
294 Bull Volcanol (2008) 70:293–314

history; during such events, the eruption of large volumes
of silicic ash at extremely high temperatures and at high
eruptive mass flux rates is likely to have had a profound
environmental impact.
Regional setting
Miocene silicic rocks of south central Idaho and northern-
most Nevada (Fig. 1) form part of the bimodal (basalt–
rhyolite) Columbia River–Yellowstone volcanic province of
northwest USA (Fig. 1). The volcanic activity generally
youngs north-eastwards towards the most recent activity at
Yellowstone (Fig. 1) and is thought to relate to the passage
of the North American continent over a stationary hot-spot
(e.g. refs in Pierce et al. 2002). An overall northeast-
younging of silicic eruptive centres in southern Idaho is
thought to reflect diachronous injection of basaltic magma
into the lithosphere, causing the generation of lithospheric
melts (e.g. Leeman 1982a) and/or the partial re-melting of
earlier injected material (refs in Christiansen and McCurry
2007). The chemistry of the rhyolitic rocks across the
province broadly reflects the nature of the lithosphere from
which the magma derives, i.e. peralkaline to metaluminous
older rhyolites in the west where the lithosphere comprises
accreted oceanic terrane material, to younger metaluminous
rhyolites of the central and eastern parts of the Snake River
Plain, where lithosphere is part of the North American
craton (e.g. Wright et al. 2002). This transition in
lithospheric affinity is reflected in the Sr, Nd, Pb and O
isotopes (e.g. Bennett and DePaolo 1987; Wooden and
Mueller 1988; Leeman et al. 1992). The voluminous central
Snake River Plain rhyolitic rocks that are the subject of this
paper formed during a major regional ignimbrite ‘flare-up’
that heralded continental rifting, with formation of the western
Snake River rift (Fig. 1). They are typical of intracontinental
A-type granitic melts; metaluminous and relatively anhy-
drous, with elevated HFSE concentrations and high halogen
concentrations (e.g. Christiansen and McCurry 2007).
The Columbia River–Yellowstone volcanic province
records diverse eruptive styles, ranging from the effusion
of extensive basaltic lava fields, low-profile basaltic shields,
to silicic extrusion and magmatic and phreatomagmatic
explosivity. Features characteristic of SR-type volcanism
occur in several parts of the province, but are best
developed in the Miocene succession in the type area
(depicted in Fig. 1). This location is the original type area
for the ‘Idavada Volcanic Group’ of Malde and Powers
(1962) and it is where the term ‘super-eruptions’ was first
coined (Bonnichsen 2000 BBC Horizon); super-eruptions
are catastrophic environmentally devastating eruptions
>300 km3 (Sparks et al. 2005). The rhyolitic (70–76%
SiO2) eruptions are inferred to have occurred from broad
eruptive centres (e.g. Bruneau-Jarbidge and Twin Falls
centres; Fig. 1), which are largely concealed beneath
Pliocene–Pleistocene basalt lavas, so much of our under-
standing derives from well exposed and deeply dissected
outflow successions in massifs to the north and south of the
Snake River Plain, such as around Jarbidge, the Cassia
Mountains and the Bennett Hills (Fig. 1; e.g. Bonnichsen
and Citron 1982; Honjo et al. 1992; McCurry et al. 1996;
Andrews et al. 2007; Bonnichsen et al. 2007). The
stratigraphy and petrology of the individual volcanic
successions are described elsewhere (previous refs), and
the present paper presents a synthesis of the physical
volcanology of the area. Field locations are given as grid
references (prefixed GR) using 11 T zone UTM co-
ordinates.
Pliocene silicic volcanic rocks, 2–5 m.y. younger than
those described in this paper, at the Magic Reservoir
eruptive centre (MR on Fig. 1) to the north of the hotspot
track in the central Snake River Plain are not entirely SR-
type in that they exhibit features such as rhyolite domes and
Plinian pumice fallout layers (Honjo and Leeman 1987;
Leeman 1982b; Leeman 2004).
Snake River-type eruptive activity has never been
observed and so must be inferred from the deposit record.
Therefore, we will describe the volcanic facies that make up
the distinctive SR-type volcanism facies-association. Some
of the features are unusual or have unusual facets, whilst
others are fairly common in silicic volcanism elsewhere.
The distinctiveness of SR-type volcanism lies in the relative
proportions of the facies; for example, in SR-type ignim-
brites lava-like facies predominate, yet in many other
volcanic fields lava-like facies are, typically subordinate,
if present at all.
SR-type ignimbrites
We estimate that there are ≥40 extensive welded rhyolitic
ignimbrite sheets in the central Snake River Plain area.
Ignimbrite successions dominate the SR-type volcanism
landscape and give rise to widespread trap-topography and
terracing of multiple cliffs along the walls of river canyons
and escarpments (Appendix Fig. 1). Individual escarpments
expose as many as ten ignimbrites, and bases of many of these
successions are not exposed (e.g. Bonnichsen and Citron
1982; Honjo et al. 1992; Perkins et al. 1995; McCurry et al.
1996; Cathey and Nash 2004; Andrews et al. 2007).
Size and morphology
Ignimbrite outflow sheets are up to 200 m thick and include
both simple and compound cooling units (e.g. the Grey’s
Landing ignimbrite and Cougar Point Tuff VII, respective-
Bull Volcanol (2008) 70:293–314 295

ly). Several, but not all, of the ignimbrites have low aspect-
ratios (Fig. 2): they trace laterally for tens of kilometres,
thickening locally into basins (e.g. Rogerson Graben) and
tapering to less than 3 m thick across palaeotopographic
highs (Fig. 3). This makes estimation of eruption volumes
difficult: in addition, mapping of individual outflow sheets
remains poorly advanced, and thicknesses and extents of
proximal ignimbrite within eruptive centres in the axis of
the Snake River Plain are obscured by younger basalt lavas.
Current understanding is that the ignimbrite eruptions were
similar in scale to those elsewhere in the volcanic province,
with eruption volumes spanning three orders of magnitude,
from tens to thousands of km3 (VEI 6–8). For example, the
Sand Springs ignimbrite (Andrews et al. 2007) has an
estimated volume of 10 km3, whereas the ignimbrites of
Jackpot, Big Bluff and Cougar Point Tuff XIII may exceed
1,000 km3 (Bonnichsen and Citron 1982; Andrews et al.
2007).
Pyroclast populations
Snake River-type ignimbrites are commonly true tuffs (e.g.
the Backwaters Member; Andrews et al. 2007). This is in
contrast to ignimbrites elsewhere, which are commonly
composed mostly of massive lapilli-tuff (or lapilli-ash).
Pumice lapilli are mostly absent (Fig. 5), as are fiamme
formed from welded pumice lapilli. A few ignimbrites
contain pumice lapilli locally. Sieving of rare, non-welded
parts of massive ignimbrites reveals grainsizes of φ>2, and
sorting values of σ
φ
<1.5 (Fig. 4), expressing significantly
better sorting than is typical for massive ignimbrites
elsewhere (σ
φ
2–5; Branney and Kokelaar 2002). Where
granulometric analysis is precluded by welding, the fine
grain-size and the absence of pumice lapilli is evident from
the absence of fiamme at field exposures and in thin section.
Another peculiarity of SR-type ignimbrites is that most
contain no, or very few, lithic lapilli or lithic blocks. In
other volcanic provinces, ignimbrite sheets commonly
contain extensive layers of coarse lithic breccia; these can,
for example, be used to indicate that caldera collapse
occurred during the climactic phase of an eruption (Druitt
and Sparks 1982; Branney and Kokelaar 2002). However,
such layers have not been reported in SR-type ignimbrites.
SR-type ignimbrites instead characteristically contain small,
sub-angular grains, <3 mm in size, of non-vesicular
obsidian, perlite and vitrophyre (e.g. Fig. 5 and Tuff of
McMullen Creek in the Cassia Mountains). These reach
1 cm in size in a few of units (Fig. 8a). Some may be
2area covered by unit A (km )
-210
-110
010
th
ick
ne
ss
o
f u
nit
t (
km
)
1
1:1
0 2
1:1
0
3
1:1
0
4
1:1
0
5
1:1
0
asp
ect
rat
io
0
t/d
= 1
:10
1
3
10 km
-1
3
10 km
-2
3
10 km
3
3
10 km
4
3
10 km
-4
3
10 km
5
3
10 km
-310 310210110010-110
diameter of unit d (km)
610510410310210110010-110-210-310
SRP rhyolite lavas
(n = 22)
other rhyolite ‘long-
lavas’ (n = 2)
SRP
ignimbrites (n = 19)
welded
Key
Big Southern Butte -
ESRP dome
Basalt lavas
n = 479
2
3
10 km
eruption
volume
SR-type lavas
Rhyolite lavas
n = 176
-3
3
10 km
Intermediate lavas
n = 239
0
3
10 km
low
as
pec
t-ra
tio
ign
imb
rite
s
hig
h a
spe
ct-r
atio
ign
imb
rite
s
SR-type
ignimbrites
domes lava flows
sub
-so
lidu
s e
xtru
sio
n
sup
ra-
sol
idu
s e
ffus
ion
rhyolitic
Fig. 2 Dimensions and aspect-ratios of SR-type ignimbrites and SR-
type rhyolite lavas compared with those from other volcanic fields
(data for non-SR-type volcanics adapted from Walker 1983; Cas and
Wright 1987). SR-type ignimbrites (black spots) span the range from
high to low aspect-ratio type. The field of SR-type rhyolite lavas
(dashed line) is quite distinct from that of non-SR-type rhyolite lavas:
SR-type lavas have similar aspect ratios to many basalt lavas (not
extensive flood basalts, which are not indicated), and they occupy a
field that overlaps with that of high aspect-ratio ignimbrites. The two
scales on the x-axis give the unit’s area (A) and the diameter of a circle
with that area (d). Positively sloping lines represent aspect ratios (t/d)
and negatively sloping dotted lines represent volumes (A.t). SR-type
ignimbrites field includes the Gomez Tuff and Bracks Rhyolite of
Trans-Pecos Texas (Henry et al. 1989) and the Yardea Dacite (Creaser
and White 1991)
296 Bull Volcanol (2008) 70:293–314

juvenile, others may be derived from underlying welded
ignimbrites or lavas.
Most SR-type ignimbrites are massive, and some show
some vertical zoning with respect to phenocryst abundance
and/or composition (Wright et al. 2002; Andrews et al.
2007). These characteristics indicate deposition from
sustained density currents whose lower parts were a
granular fluid (‘granular fluid-based pyroclastic density
currents’ of Branney and Kokelaar 2002). Some contain
moulds of trees (e.g. Greys Landing Ignimbrite at Monu-
autobreccia
silicic coulee
5 km
100 m
silicic dome
autobreccia
100 m
2 km
proximal
tephrafeeder
dyke
basal
vitrophyre
autobreccia
columnar jointing
sheet
joints flow
lobe
large gas cavities
flow interior flow margin
~ 30 m
~ 300 m
30 km
upper
vitrophyre
SR-type rhyolite lava
lithoidal
rhyolite
~ 100 m
basal
vitrophyre
autobreccia
upper
vitrophyre
medium-scale
rheomorphic folds
sub-horizontal
welding and small-scale
rheomorphic fabric
feather-edge vitrophyre
columnar
jointing
5 m
100 km
lithoidal
rhyolite
~ 100 m
pumice-rich lapilli-tuff
eutaxitic lapilli-tuff
10 km
5 m
‘Typical’ non-SR-type rhyolite lava (e.g. Italy, Jemez Mtns, La Primavera)
SR-type ignimbrite
‘Typical’ non-SR-type ignimbrite (e.g. Italy, Jemez Mtns, Pinatubo)
local
peperites
splay-and-fade
stratification
Fig. 3 Characteristic features of
SR-type rhyolite lavas (top) and
SR-type ignimbrites (bottom)
contrasted with those typical of
other volcanic fields. SR-type
lavas (note vertical exaggera-
tion) are much more volumi-
nous, more extensive, and have
lower aspect ratios than non-SR-
type silicic lavas. SR-type
ignimbrites are extremely high-
grade, commonly lack pumice
lapilli (or fiamme), with intense
welding, rheomorphism and de-
velopment of lava-like facies.
They are distinguishable from
lavas by their tapering margins,
extensive zones containing per-
vasive vitroclastic textures, and
the absence of extensive basal
autobreccia and blunt, lobate
terminations
Bull Volcanol (2008) 70:293–314 297

ment Canyon, Twin Falls County). A few non-welded
zones also include horizons, traceable for several kilo-
metres, of well-developed tractional cross-stratification,
indicative of deposition from currents in which clast
concentrations were too low to cause significant grain
interactions or dampen turbulence even within basal parts
(‘fully dilute’ pyroclastic density currents of Branney and
Kokelaar 2002), and they contain accretionary lapilli with
concentric laminations as well as cored accretionary lapilli
in which the concentric laminations enclose a central vitric
fragment (Fig. 8a and b).
Vitroclastic textures reveal that most of the juvenile shards
are relatively thick, platy and cuspate, bubble-wall types
(Fig. 6a). In addition there are rare occurrences of sub-
spherical, globule-shaped shards that appear to be more
mafic than the surrounding cuspate shards (Fig. 6b). These
are non-flattened despite the eutaxitic nature of the surround-
ing cuspate vitroclastic matrix. This suggests that, as with
achneliths in basaltic fire-fountaining eruptions, the globule-
shaped shards were liquid at the time of fragmentation, yet
solidified prior to deposition, possibly as a result of higher
solidification temperatures than those of the surrounding
welded rhyolitic shards. In many SR-type ignimbrites,
however, welding and rheomorphism has pervasively trans-
posed and obliterated the vitroclastic textures (Fig. 6c).
Unusually high welding intensities, and contact-thermal
effects
Most Miocene ignimbrites in the central Snake River Plain
range from high-grade (sensu Walker 1983) to extremely
high-grade (sensu Branney and Kokelaar 1992). Welding is
intense—typically off the top of the welding intensity scale
proposed by Quane and Russell (2005)—and it extends
right to the base of many of the ignimbrites. Contact-
thermal effects extend to as much as 5 m beneath some
ignimbrites: underlying stratified ash layers are fused to
black vitrophyre (Fig. 7a) and substrate palaeosols are
intensely baked to red-brown terracotta and have developed
columnar joints by cooling-contraction (Fig. 7a). Another
unusual characteristic of the welding in SR-type ignimbrites
is that it sometimes extends to, or close to, the preserved
Fig. 5 Non-welded facies of SR-type ignimbrites, showing the
absence of pumice lapilli and lithic lapilli, and the presence of
abundant small vitric chips supported in a massive, fine ash matrix. a
Typical SR-type non-welded massive ignimbrite with small vitric
grains; Sand Springs ignimbrite, US Highway 93. b Jackpot 6
ignimbrite also contains coarser vitric fragments; from US Highway
93 south of Jackpot, Nevada. Rule shows centimetres
SR-type ignimbrites of the central SRP
Eastern SRP ignimbrites
(finer grained)
Sand Springs Ignimbrite
Grey's Landing Ignimbrite
Dry Gulch 1 Ignimbrite
Steer Basin ignimbrite
Cougar Point Tuff 15
Cougar Point Tuff 11b
Cougar Point Tuff 11a
Tuff of Kilgore
Tuff of Wolverine Creek
99%
96%
-10 -8 -6 -4 -2 0 2 4 6
(no grainsize
variation)
(less well
sorted)
median
so
rti
ng
(8
4%
-16
%)
/ 2
(p
hi)
diameter (phi)
6
5
4
3
2
1
0
Fig. 4 Grain size and sorting
characteristics of SR-type
ignimbrites, compared to those
from other volcanic provinces.
Shaded fields from Walker
(1971); percentages cited are
from 300 ignimbrites; sorting
is described in terms of a
standard deviation,
σ
φ
¼ φ84
φ16ð Þ=2.
SR-type ignimbrites are
generally better sorted than
ignimbrites from other volcanic
provinces
298 Bull Volcanol (2008) 70:293–314

tops of units, with fusing of overlying stratified ash layers
(e.g. layer of fused accretionary lapilli - bearing stratified
tuff overlying Jackpot 5; Fig. 8b).
Pervasive and intense rheomorphism
The intense welding of SR-type ignimbrites is commonly
associated with intense rheomorphic deformation. Perva-
Fig. 7 Field characteristics of SR-type ignimbrites. Columnar-jointed
baked palaeosol (behind notebook) and fused parallel-laminated
ashfall, beneath the massive basal vitrophyre (top) of the Greys
Landing ignimbrite. Arrow indicates base of ignimbrite. Scale is
10 cm. Backwaters, Idaho, GR:685154 4659673. b Isoclinal folding in
flow-laminated, lava-like facies of the Greys Landing Ignimbrite
(location and scale as in a). c Open rheomorphic fold with
subhorizontal fold axis in upper part of ignimbrite folds steep isoclinal
folds (similar to those in b). Steer basin ignimbrite. Cassia Mountains.
Scale 5 m high
Fig. 6 Typical microscopic textures of Snake River-type ignimbrites.
a SEM image of non-welded bi- and tri-cuspate bubble-wall type
shards. Base of Cougar Point Tuff Member XV, Murphy Hot Springs.
b Welded cuspate shards form a eutaxitic fabric, with rare, non-
deformed mafic globule-shaped shards (white arrows). Ignimbrite
from the Mount Bennett Hills (Tuff of Fir Grove GR: 663660 478290;
photomicrogram of thin section in PPL; scale as in A). c Intense
welding and rheomorphism has transposed and attenuated former
vitroclastic textures, producing a flow-lamination similar to that seen
in silicic lavas (the flow lamination grades into vitroclastic tuff near
the base and top of the ignimbrite: not shown). Isoclinal fold pair
indicates top-to-left rheomorphic shear. Basal vitrophyre of the Grey’s
Landing ignimbrite (Rogerson Formation); Grey’s Landing, Twin
Falls County, Idaho
Bull Volcanol (2008) 70:293–314 299

sive development of elongation lineations, including
stretched vesicles associated with pervasive flow-lamina-
tion, and open to isoclinal small- to medium-scale folds
(Fig. 7b, c), including oblique folds and sheath folds,
indicate that the ignimbrites underwent agglutination and
rapid ductile shear whilst still hot (Branney et al. 2004). In
ignimbrites elsewhere, welding is commonly thought to
post-date ignimbrite emplacement (‘load welding’ Freundt
1999), but in SR-type ignimbrites, both the welding and the
onset of rheomorphic deformation are thought to have
occurred very rapidly, during deposition (Branney and
Kokelaar 1992). Shear directions recorded by fold-axes
and elongation lineations in some SR-type ignimbrites vary
with height through an individual unit, and are thought to
indicate changing shear directions within a diachronous
rheomorphic shear-zone that ascended through the aggrad-
ing agglutinate while deposition occurred from the base of
an over-riding, sustained and shifting pyroclastic density
current (Branney et al. 2004; Andrews and Branney 2005;
Andrews 2006). Top surfaces of several of the ignimbrites
have been folded into tight ogive-like medium-scale folds,
commonly associated with high-angle thrusts: this indicates
late-stage rheomorphic flow, similar to that seen elsewhere
in silicic lavas and strongly peralkaline ignimbrites (e.g.
Sumner and Branney 2002). Intense welding and rheo-
morphism in some of the ignimbrites persists even where
the ignimbrite thins to less than 5 m.
Lava-like lithofacies
An unusual characterisitc of SR-type ignimbrites is the
widespread development of lava-like facies. We use ‘lava-
like’ as a non-genetic term (after Branney and Kokelaar
1992) for a lithofacies that looks like lava; it may be
massive or flow-banded (e.g. Fig. 6c) but does not exhibit
vitroclastic or eutaxitic texture (e.g. Fig. 6a, b). In most
volcanic fields lava-like facies constitute a rare or restricted
facies of rheomorphic ignimbrites. However, in SR-type
volcanism lava-like facies dominate entire ignimbrite sheets
(e.g. House Creek, Grey’s Landing, Big Bluff, Jackpot 1–5
and Castleford Crossing ignimbrites and Cougar Point
Tuffs 11 and 13; Bonnichsen et al. 2007). Lava-like facies
of SR-type ignimbrites are indistinguishable from true lava
in both hand specimen and in thin section, and their
pyroclastic origin can be inferred only from the field
relations: for example, rheomorphic ignimbrites generally
do not have widespread basal autobreccias, whereas basal
autobreccias characterise most rhyolite lavas; also, the lava-
like facies in some ignimbrites grade laterally and/or
vertically into less intensely welded, unequivocal vitroclas-
tic tuff (Bonnichsen and Kauffman 1987; Branney et al.
1992; Henry and Wolff 1992). A characteristic problem of
SR-type volcanism is that in cases where critical field
relations (e.g. basal contacts and distal feather-edges) are
not exposed, it can be almost impossible to determine
whether some predominantly lava-like units are rheomor-
phic ignimbrites or true lavas. Although lava-like facies
occur within ignimbrites of some strongly peralkaline
volcanic fields (Sumner and Branney 2002), they are a
Fig. 8 Ash aggregates in SR-type tephras. a Concentric-laminated
accretionary lapilli from a non-welded, cross-stratified lapilli-tuff,
deposited from a long-runout, fully dilute pyroclastic density current.
Note abundant angular vitric fragments (black), some coated in fine
ash. Jackpot Member 6; US Highway 93, south of Jackpot, Nevada. b
Welded accretionary lapilli, some with vitric cores (black). The
accretionary lapilli are framework supported in some layers, and
supported in fine tuff vitrophyre matrix (dark, bottom) in others.
Jackpot 6, Nevada. c Framework-supported ash coated pellets, some
with vitric cores (left) in ashfall deposits. Lower part of Cougar Point
Tuff XV at Murphy Hot Springs, Idaho. Scale in cm
300 Bull Volcanol (2008) 70:293–314

relatively subordinate facies of ignimbrites within most
metaluminous rhyolitic provinces, with the exception of
some thick, caldera-filling ignimbrites (e.g. Batchelor
caldera, Lipman 1984; Scafell caldera, Branney et al. 1992).
Other characteristics of the ignimbrites
SR-type ignimbrites form cooling units with upper and
lower vitrophyres, each as much as 5 m thick, enclosing a
lithoidal (microcrystalline) central zone, which may be as
thick as 150 m (Bonnichsen and Citron 1982). Where the
ignimbrites thin over topography, the vitrophyres merge to
form a single vitrophyre. They characteristically contain
laterally extensive zones of spherulites and lithophysae,
ranging from a few millimetres up to 50 cm in diameter,
close to the contact with the central lithoidal zone. The
spherulites are typically spheroidal, but are oblate where
their growth was influenced by the presence of a pro-
nounced welding foliation, and are prolate where they grew
in vitrophyre with a strong rheomorphic elongation linea-
tion. Primary porosity has disappeared in intensely welded
facies, but upper parts of many of the ignimbrites variously
contain stretched and spherical vesicles that indicate
continued vesiculation during and after the final stages of
rheomorphic deformation. Some of the ignimbrites have
developed scoriaceous or coarsely pumiceous zones in
upper parts (e.g. in Cougar Point Tuff 11; Bonnichsen and
Citron 1982: Tuff of Dry Gulch at Rock Creek, GR:720159
4698838), in which the walls of the elongated vesicles have
become so attenuated that the normally black vitrophyre
takes on a golden brown colour, similar to that seen in
basaltic reticulite. Clearly, in SR-type ignimbrites density
cannot be used as a proxy for welding intensity. However,
most SR-type rheomorphic ignimbrites are denser and
much less vesicular than is typical of rheomorphic
ignimbrites elsewhere, such as the eutaxitic Wall Mountain
Tuff of Colorado (Chapin and Lowell 1979) and the Green
Tuff of Pantelleria (Mahood 1984).
Flow banding, or lamination, is common in the ignim-
brite vitrophyres and lithoidal zones. Some of it may have
originated from healed fractures, former vesicular zones
and other textural heterogeneities that have become
transposed and attenuated during ductile shear, as in flow-
banded lavas, except that in the case of SR-type ignimbrites
it must have developed at the surface by rheomorphism
after deposition and agglutination, rather than within a
magma-filled eruption conduit. Close-spaced (1–2 cm)
straight or curved ‘sheeting joints’ (Bonnichsen and Citron
1982) are abundant within the lithoidal zones. Their
orientation is usually developed sub-parallel to the flow
banding, apart from at tight fold hinges, and they are
probably developed during static devitrification, mimetic
after the anisotropy of the flow banding.
SR-type volcanism is characterised by a general paucity
of low-grade (non-welded) and moderate-grade (partly
welded) ignimbrites with extensive zones of non-welded
lapilli-ash, sillar, and non-rheomorphic eutaxitic lapilli-tuff.
This is in marked contrast with many other ignimbrite
fields, such as the Jemez Mountains of New Mexico, Taupo
Volcanic Zone in New Zealand, and the Central Mexican
Volcanic Belt, where those facies dominate. Non-welded
ignimbrites do occur in the central Snake River Plain, such
as the ash-pellet—bearing ignimbrites in the Cassia Hills
(Fig. 1; see later section), but are not widely reported: this
may in part be due to burial by talus and slope-wash from
overlying, cliff-forming welded units, and partly because of
bioturbation and reworking (e.g. unit below the Three
Creek Rhyolite of Bonnichsen and Citron 1982). Loose,
non-welded ignimbrites would be more prone to erosion and
reworking, and may have contributed to the numerous layers
of rhyolitic volcaniclastic sands (see later section). However
the paucity of eutaxitic and moderately welded facies cannot
be so easily explained in this way. Several other common
features typical of ignimbrites elsewhere, such as coarse-tail
grading, internal layering and diffuse stratification, and
lithic-rich and pumice-rich lenses (Branney and Kokelaar
2002) also have not been reported in the region.
Extensive layers of parallel-stratified rhyolitic ash
Thin-bedded to laminated, grey to white rhyolitic ash is a
common facies of SR-type volcanism (Fig. 9a) and forms
layers ≤5 m thick. The stratification is formed of sharp and
gradational changes in grainsize; millimetre-scale normal
and reverse, vertical distribution grading is common, with
no obvious overall asymmetry or cyclicity. Individual beds,
<10 cm thick, are well-sorted (Fig. 10 and Appendix Fig. 2)
and predominantly of medium to coarse rhyolitic ash, with
abundant large platy, bicuspate and tricuspate bubble-wall
type shards (Fig. 10) and a small proportion (<10%) of
crystals, mostly quartz, sanidine and/or plagioclase. A
characteristic of SR-type volcanism is that many of the
cuspate shards are sufficiently large to be readily distin-
guished in the field without a hand lens (Fig. 9c). This
indicates that during fragmentation, a significant compo-
nent of the vesicularity was megascopic. Small fragments
of coarsely tubular pumice also occur, including ‘striated’
shards, derived from fragmented thick-walled tubular
pumice (Perkins et al. 1995). Similar shard morphologies
occur in the ignimbrites. Blocky shards are rare. As with
the ignimbrites, many of the ashfall layers contain angular,
black or grey glassy grains, of obsidian, perlite and
vitrophyre, <1 mm in size. Some ashfall layers contain
framework-supported, typically openwork ash-coated pel-
lets, 2–6 mm in diameter (Fig. 8c).
Bull Volcanol (2008) 70:293–314 301

Pumice fall deposits of lapilli-sized clasts, such as
characterise many silicic volcanic fields elsewhere, are
relatively rare: we have found thin (<5 cm) beds of
framework-supported angular pumice lapilli within ashfall
layers at just three locations: Nat-Soo-Pah Hot Springs
(near Hollister) and locally beneath the Grey’s Landing
ignimbrite and Cougar Point Tuff 15.
Many of the stratified ash layers are overlain either by a
welded ignimbrite (e.g. Tuff of Wooden Shoe Butte in Rock
Creek Canyon; McCurry et al. 1997) or rhyolite lava (e.g.
Triguero Homestead Rhyolite at Bruneau Canyon). Those
beneath welded ignimbrites have been wholly or partially
fused to black vitrophyre (e.g. Figs 7a and 9b). Some of
these fused facies contain fused, vitrophyric ash pellets
(Fig. 9b). Some stratified ash layers are partly bioturbated
and soilified (e.g. variously bioturbated stratified and
massive ash deposits in Goose Creek Basin, Fig. 1; and
below the Three Creek Rhyolite, Bonnichsen 1982).
A fallout origin is inferred for much of the parallel-
stratified ash on the basis of the lateral consistency of
individual laminations across tens of metres, together with
the granulometry, the predominance of juvenile compo-
nents, and the absence of low-angle truncations. The layers
are laterally extensive and probably represent large erupted
volumes: widespread atmospheric dispersal is suggested by
ashfall layers correlated thousands of kilometres from the
Snake River Plain (Kimmel 1982; Swirydczuk et al. 1982;
Perkins et al. 1995; Perkins and Nash 2002), e.g. in the
Ogallala Formation of Nebraska, c. 1,400 km to the east
(Perkins et al. 1998; Rose et al. 2003) where the ash, which
is reworked, is surprisingly coarse (75–200 μm diameter
shards) given the great distance from source, with irregular,
platy shard morphologies. No isopach maps have been
constructed to determine eruptive volumes, and the eruption
styles and mass-flux rates have not been investigated. Some
layers may be from vent-derived eruption columns, others
may be co-ignimbrite ashes.
Large-volume, low aspect-ratio rhyolite lavas
Rhyolite lavas in the central Snake River Plain are
considerably more extensive (tens of km) and more
voluminous (>10 km3) than rhyolite domes and coulées
typically associated with explosive silicic caldera volcanoes
elsewhere (Figs. 2 and 3). They are well documented (e.g.
Fig. 9 Typical SR-type ashfall
deposits. a Parallel-laminated
and thin-bedded fine to coarse
ash, showing laterally persistent
stratification. Nat-Soo-Pah,
Twin Falls County, Idaho. Metre
ruler. b Fused, largely parallel-
stratified tuff of dominantly
ashfall origin, with layers of
abundant fused framework-sup-
ported small (<2 mm in diame-
ter) ash pellets with massive ash
cores and fine ash rims (pale).
Tuff of Wooden Shoe Butte,
Rock Creek, Idaho. c Field shot
of ashfall deposit showing un-
usually coarse cuspate shards
(dark) supported in fine ash
matrix (white). Tuff of Dry
Gulch, Rock Creek, Idaho. Scale
shows cm
302 Bull Volcanol (2008) 70:293–314

Bonnichsen 1982; Ekren et al. 1984; Bonnichsen and
Kauffman 1987; Bonnichsen et al. 1989; Henry and Wolff
1992) and we summarise only their salient features.
At least eight rhyolite lavas from the Bruneau-Jarbidge
eruptive centre exceed 10 km3 each: the Dorsey Creek
Rhyolite (Appendix Fig. 3) exceeds 75 km3 and is over
40 km long, and the Sheep Creek Rhyolite exceeds 200 km3
(Bonnichsen 1982). Large rhyolite lavas are present also at
the inferred Twin Falls eruptive centre (e.g. Shoshone Falls
Rhyolite, Balanced Rock Rhyolite, Bonnichsen et al. 1989),
and in the Juniper Mountains of SW Idaho (e.g. the Badlands
Rhyolite, Manley 1996; Manley and McIntosh 2002). Small
rhyolite domes and coulées are not characteristic of SR-type
volcanism, although the possibility that some lie concealed
within subsided eruptive centres cannot be excluded.
Aspect ratios of SR-type rhyolite lavas are unlike those of
viscous rhyolite lavas elsewhere (Fig. 2). They are sufficient-
ly low to coincide with those of many basalt lavas (Fig. 2).
This is reflected in their long distance run-outs from their
inferred source locations. The unusual dimensions led to
early interpretations that they were ignimbrites (Ekren et al.
1984), but they are now thought to be lavas on the basis of
several criteria, including the presence of widespread basal
autobreccias and abrupt, thick, stubby lobate terminations
with thick talus aprons (e.g. see Bonnichsen and Kauffman
1987; Henry and Wolff 1992; Manley and McIntosh 2002).
In contrast, basal autobreccias in rheomorphic ignimbrites
are rare and restricted to locations where rheomorphic flow
has carried the hot agglutinate beyond the original extent of
pyroclastic deposition (e.g. Sumner and Branney 2002).
The lavas are blocky with thick (>5 m), variously vitric
and coarsely pumiceous lower, marginal and upper cara-
pace autobreccias (Bonnichsen 1982). Hot-state shear of
basal, marginal, and internal coarsely pumiceous autobrec-
cia during flowage has caused localised fusing and ductile
shear, with the development of vitroclastic textures that can
superficially resemble welded pyroclastic textures (Manley
1996) as is common in viscous blocky lavas elsewhere (e.g.
Iddings 1889; Pichler 1981; Sparks et al. 1993). Thick,
central zones of the lavas (Appendix Fig. 3) are lithoidal
(microcrystalline), massive or flow-banded, and dominated
by steep columnar jointing, and low-angle close-spaced
sheeting joints that in places form intersecting sets,
producing ‘pencil-type’ jointing (Bonnichsen 1982). Spher-
ulites and lithophysae are common near the base of the
lithoidal zone. Some of the lavas, particularly to the west of
the central Snake River Plain, exhibit agglutinate textures
with spatter rags that indicate a clastogenic origin by some
form of rhyolitic fire fountaining (e.g. Juniper Mountains;
Manley and McIntosh 2002), although the majority of the
rhyolite lavas are not visibly clastogenic.
Lacustrine-alluvial facies
The facies association that characterises SR-type volcanism
in south central Idaho and northern Nevada includes
widespread evidence for the presence of surface water at
the time of the eruptions.
Aqueously deposited volcaniclastic sediments
Numerous thin layers of aqueously deposited volcaniclastic
sediments are intercalated with ignimbrites across the central
and western Snake River Plain (Appendix Fig. 4a, b). Alluvial
and lacustrine facies are represented, with graded beds,
parallel-laminated silts, ripple cross-lamination, scours, and
abundant soft-sediment deformation, loading and dewatering
structures. Many of the parallel-stratified ash layers appear to
have been deposited directly into shallow standing water as
there are local gradations into facies with minor truncations
and scours that indicate reworking by gentle aqueous currents
or wind, and into rippled sands of the same composition (e.g.
Perkins et al. 1995). Lacustrine sediments occur between the
Fig. 10 Typical vitroclastic textures of SR-type ashfall deposits. a
SEM image of cuspate, bubble-wall shards. Basal ashfall of Cougar
Point Tuff Member XV, Murphy Hot Springs, Owyhee County, Idaho.
b SEM image of bubble-wall shards from an ashfall fall deposit
overlying the Grey’s Landing ignimbrite, at Salmon Dam, Twin Falls
County, Idaho
Bull Volcanol (2008) 70:293–314 303

Cougar Point Tuffs around Jarbidge; within the Rogerson
Formation near Jackpot and in the Rogerson Graben; and
between ignimbrites dated at 13.7 to 8.6 Ma in the Shoshone
Basin, Cassia Hills and Goose Creek Basin (Fig. 1; e.g.
Hildebrand and Newman 1985; Perkins et al. 1995).
Some of the lacustrine intervals within the Miocene
ignimbrite-dominated successions north and south of the
central Snake River Plain (Appendix Fig. 4a, b) may reflect
somewhat isolated and/or short-lived bodies of water, as
they form thin (<10 m thick) layers within successions that
also include palaeosols with rootlet horizons, ≤2 m thick.
The lakes may have been connected, however, because the
distribution of lacustrine sediments and their fossils along
the axis of the Snake River Plain reveals the former
presence of a large lake, Lake Idaho (Hearst 1999; Kimmel
1982; Smith et al. 1982; Swirydczuk et al. 1982; Link et al.
2002; Wood and Clemens 2002). Lake Idaho extended
across the Western Snake River rift to cover a significant
part of the central Snake River Plain (Fig. 1; extent to the
SE yet to be defined). Its presence reflects low topography
partly due to opening of the NW-trending Western Snake
River rift in response to SW–NE crustal extension parallel
to the Yellowstone hotspot track. At its maximum high-
stand of about 1,150 m, the lake may have covered nearly
20,000 km2, about the same area as modern Lake Ontario in
eastern North America. Lake Idaho’s initiation has been
inferred by some workers to post-date the rhyolitic volcanism
in the central Snake River Plain (Wood and Clemens 2002)
but there is increasing evidence that it existed during the
emplacement of many western Snake River rhyolitic units in
the 11.7–10.0 Ma interval (McCurry et al. 1997; Bonnichsen
et al. 2007; Bonnichsen and Godchaux 2002; Godchaux and
Bonnichsen 2002), and it may have been a continuation of
older Miocene lakes such as those in the Oregon–Idaho
graben (Cummings et al. 2000) and Goose Creek Basin
(Fig. 1). The lake(s) waxed and waned in surface elevation
and area, and finally disappeared between 2.5 and 1.5 Ma
due to incision by the Snake River (Wheeler and Cook 1954;
McCurry et al. 1997; Wood and Clemens 2002).
Peperitic bases of welded ignimbrites
Peperites formed by the emplacement of hot outflow
ignimbrites onto wet substrate occur on both north and south
sides of the central Snake River Plain (Fig. 11). The chilled
basal vitrophyres show hydroclastic brecciation, with perva-
sive injection of disaggregated sediment into fractures.
Clouds of the angular vitrophyre hydroclasts are locally
loaded into, and admixed with disaggregated sediment
(Fig. 11). We attribute the sediment disaggregation and
mobilization to steam-fluidisation; and the lack of contact
baking or fusing of the substrate sediments to the insulating
effect of a steam carapace generated by boiling of pore water
in the sediment: both are common features of peperites
(Kokelaar 1982). Peperites formed where hot ignimbrite was
emplaced onto wet ash or sediment are known elsewhere,
such as in southern Korea (Fig. 6.15 of Branney and
Kokelaar 2002), Scafell caldera, England (the lava-like
Oxendale Tuff; Kokelaar and Branney 1999), and the
rheomorphic Pitts Head and Garth ignimbrites of Wales
(Kokelaar 1982; Kokelaar and Königer 2000). The Idaho
examples (Fig. 11) lie just outside the Snake River Plain,
where the water depth may have been as little as a couple of
metres, as indicated by small wave ripples and by field
relations; e.g. near the margin of the plain the base of an
ignimbrite is peperitic but higher levels in the ignimbrite
show no evidence of interaction with water and were
probably emergent, and a few tens of metres farther away
from the plain, the base of the same ignimbrite was
deposited onto dry subaerial ashfall beds with no formation
of peperite and with intense fusing of the ashfall layers (Tuff
of Wooden Shoe Butte; Fig. 9b and Appendix Fig. 4a). It is
possible that water depth increased towards the axis of the
Snake River Plain, but axial sites are not exposed. Non-
explosive types of interaction between ignimbrites and
ponded water are also indicated by localised silicified zones
with veins of jasper and chalcedony where ignimbrites are
intercalated with fine-grained lacustrine sediments.
Evidence of rhyolite lava–water interactions
Miocene rhyolite lavas at relatively low elevations (below
c.1,200 m) in the interior of the Snake River Plain locally
show abundant evidence of interaction with lake water
(Appendix Fig. 4c; Godchaux and Bonnichsen 2002), e.g.
the Dorsey Creek, Bruneau Jasper, and Shoshone Falls
lavas (Bonnichsen and Kauffman 1987). The Jump Creek,
Reynolds Creek, Horse Basin, and the Cerro Otono
rhyolites in the Western Snake River Plain show similar
evidence. Marginal peperites occur where rhyolite extruded
through the wet lake-floor sediments; sediments beneath the
lavas show abundant chaotic soft-sediment disruption,
where large tilted rhyolite blocks have foundered into them.
Rhyolitic phreatomagmatic tuffs at the tops of some of the
lavas are thought to record rootless explosions from the
interiors or bases of submerged lavas. Tops and margins of
some of the water affected lavas have developed particu-
larly thick (>100 m) masses of dominantly vitrophyric,
clast-supported breccia, and chalcedony, jasper and opal
locally fills fractures and other cavities in the lavas, with
extensive groundmass silicification to their deep interiors.
Basaltic pillow lava–hyaloclastite deltas
Basalt lavas in central southern Idaho include hydrated
basalts (water-affected basalts or ‘WAB’ of Bonnichsen and
304 Bull Volcanol (2008) 70:293–314

Godchaux 2002), pillow lavas and hyaloclastites, including
prograding pillow-hyaloclastite deltas (Appendix Fig. 4d;
Shervais et al. 2005). Evidence for the interaction of the
basalt lavas with lake water, streams and associated ground
water has been widely documented (Jenks and Bonnichsen
1989, refs in Bonnichsen and Godchaux 2002).
Phreatomagmatic tuffs
Several basaltic tuff cones and tuff rings dating back to
Miocene times occur within the region of former Lake
Idaho (Godchaux and Bonnichsen 2002). The tuffs contain
abundant ash pellets and accretionary lapilli, exhibit wet,
soft-state deformation and are associated with aqueously
reworked facies. Both Surtseyan and Taalian (maar-form-
ing; Kokelaar 1986) eruption styles are represented and
indicate that the erupting magma interacted explosively
with, respectively, lake water and groundwater. There are
few equivalent examples of rhyolitic phreatomagmatic
centres because the rhyolitic source vents are buried in the
interior of the central Snake River Plain. However, the
numerous layers of fine vitric ash containing abundant ash
aggregates that occur within the ignimbrite successions may
derive from water-enhanced explosivity. For example,
white, fine ash - rich ashfall layers and associated non-
welded ignimbrites containing abundant coated ash pellets
occur in eruption units in the Cassia Hills and Trapper
Creek (Fig. 1). The subaerial vent area of the Wilson Creek
ignimbrite is exposed in the western Snake River Plain, and
proximal water-altered phreatomagmatic tuffs contain
abundant angular chips of older rock types indicating
explosive fragmentation of near-surface, water-bearing
rocks by rising rhyolite magma prior to the main ignimbrite
eruption into Lake Idaho (Ekren et al. 1984; Godchaux and
Bonnichsen 2002; Bonnichsen et al. 2007).
Associated basaltic volcanism
The rhyolitic rocks are associated with basalts, in the form
of scattered tuff rings and coalescing low-profile shield
volcanoes with gentle concave slopes (1–2° flanks steep-
ening to 5° near summits) and sometimes spatter ramparts
surrounding Hawaiian-type summit collapse craters (e.g.
Shervais et al. 2005). Some of the basalt lavas are fed by
fissures along rift-zones. The basaltic volcanism is transi-
tional between typical (i.e. steeper) shield volcanoes and
true flood basalts, and has been termed ‘basaltic plains style
volcanism’ to distinguish it from those end-members
(Greeley 1977, 1982). Most of the visible basalt forms a
cap overlying the Miocene ignimbrites and lavas across the
width of the plain. Because of limited incision, it not known
when basaltic volcanism began in the central Snake River
Plain and whether any early basalt effusion occurred
associated with emplacement of the inferred mid-crustal
mafic intrusion (Rogers et al. 2002).
Discussion
This section considers the eruptive processes, the nature of
the eruptive centres, and other examples worldwide where
SR-type volcanism may have occurred.
Fig. 11 Ignimbrite peperites from the central Snake River Plain,
Idaho. a Peperite at the base of an ignimbrite (‘Picabo Tuff’), just
north of the plain. Basal vitrophyre (dark) is finely brecciated and
loaded into wet silt (pale). b Ignimbrite vitrophyre peperite at the base
of the Wooden Shoe Tuff, Rock Creek Canyon, Cassia Mountains.
Note pervasive injection of disaggregated silt (pale) into fractures in
the glassy welded ignimbrite. c Detail of peperite showing clouds of
admixed angular hyaloclasts and silt. Location as for b
Bull Volcanol (2008) 70:293–314 305

SR-type eruptive processes
Snake River-type eruptions were voluminous and extraor-
dinarily environmentally devastating, with large-scale
explosivity (VEI 6–8) during which vast glassy ignimbrites
were fused across the landscape, and accompanied by
widespread atmospheric dispersal of vitric ash. The
extensive, predominantly massive and intensely welded
nature of the ignimbrites is best reconciled with high-
temperature, sustained, granular fluid-based pyroclastic
density currents with run-out distances exceeding 100 km
that were fed from high mass-flux pyroclastic fountaining
eruptions (e.g. Bursik and Woods 1996; Branney and
Kokelaar 2002). High eruptive mass flux, possibly via
fissures associated with caldera subsidence, would have
minimised cooling during transport (e.g. Branney et al.
1992; Freundt 1999). Non-welded bases of a few of the
ignimbrites may be the legacy of entrainment and mixing of
atmospheric air into leading parts of the currents during
early waxing stages of eruptions while the currents initially
advanced across the landscape. Such flow-front cooling
seems to have been minor in cases where the ignimbrites
are welded right to their bases. However, it is doubtful that
such ignimbrites were ever initially isothermal as is
commonly assumed in welding models (refs in Russell and
Quane 2005): phenocryst geothermometry indicates that
initial thermal gradients may have existed in the ignimbrites
even before they started cooling (Andrews et al. 2007).
Gradual vertical changes in the trend of rheomorphic
structures (elongation lineations and sheathfold hinges)
within some of the most intensely welded ignimbrites
indicate that pyroclasts in these examples had viscosities
sufficiently low to enable them to agglutinate and shear
even during deposition (Branney et al. 2004). The abun-
dance of lava-like facies indicates that coalescence of hot
pyroclasts, with obliteration of the clast outlines (Branney
and Kokelaar 1992) was common: such rapid welding
requires unusually low viscosities compared to those
required for post-depositional welding (‘load welding’ of
Freundt 1999) which is thought to characterise ignimbrites
elsewhere (e.g. Ross and Smith 1961). The hot deposits
retained a sufficiently low viscosity to spread gravitation-
ally across low (≤5°) topographic slopes. Exsolution of
volatiles continued, with the growth and rheomorphic
attenuation of ellipsoidal vesicles and in some cases
development of golden, thick-walled coarse pumiceous
zones in upper parts of ignimbrites; such frothy lava-like
facies are generally rare in ignimbrites elsewhere.
Low viscosity of pyroclasts in some rheomorphic
ignimbrites elsewhere is attributable to strongly peralkaline
chemistries in which alkalis act as network-modifiers
disrupting silicate polymerisation (Mahood 1984). The
globular shards discovered in the (metaluminous) SR-type
ignimbrites are found also in some strongly peralkaline
rheomorphic ignimbrites (e.g. Johnson 1968; Sumner and
Branney 2002) and indicate pyroclast viscosities sufficient-
ly low for clast shapes to be controlled by surface-tension
prior to chilling during transport (Branney and Kokelaar
1992), just as in droplet-shaped pyroclasts from Hawaiian
fire-fountaining eruptions (Pele’s tears or ‘achneliths’ of
Walker and Croasdale 1972). However, the SR-type
eruptions are not peralkaline and the low viscosities must
result from high emplacement temperatures and possibly
retention of some dissolved volatiles. This is consistent
with high magmatic temperatures of 830–1,050°C estimat-
ed from pyroxenes, feldspars and oxide phases in SR-
ignimbrites, and estimates that at least some of the rhyolite
magmas had high fluorine contents (e.g. Honjo et al. 1992;
Cathey and Nash 2004; Christiansen and McCurry 2007).
Elsewhere, unusually low rhyolite viscosities indicated by
rocks thought to record rhyolitic fountaining and aggluti-
nation have also been attributed to high fluorine contents
(Duffield 1990). The high mass-flux eruptions may have
minimised both cooling and the exsolution of certain
volatile species during eruption and emplacement.
The absences in SR-type ignimbrites of features like lithic
and pumice dunes, bedding and pumice concentration zones,
may be because agglutination of sticky particles limited
granular segregation. However, absence of such features also
characterises the rare non-welded facies, and so may simply
reflect the paucity of pumice and lithic lapilli. Paucity of
lithic lapilli is a characteristic of extremely high-grade
ignimbrites elsewhere (e.g. Branney et al. 1992) and may
reflect minimal erosion and entrainment from the vent
margins and from the land surface by the pyroclastic current
which, instead, tended to plaster surfaces with hot aggluti-
nate. The general absence of pumice lapilli and fiamme is
difficult to explain (fiamme are common in many rheomor-
phic ignimbrites elsewhere) and points to some facet of the
fragmentation process, in which cuspate shards were
generated from the vesiculating magma more readily than
were pumiceous blocks and lapilli. Such a fragmentation
mechanism is not understood. The morphology and large
size of the cuspate shards in both the ignimbrites and the
ashfall layers indicates that bubble sizes were larger than is
typical of rhyolitic micro-vesicular pumice elsewhere; this
may be the result of more rapid diffusion rates of volatiles
within the rhyolite magmas due to their unusually high
temperature and low viscosity. Fiamme present locally in
some of the rheomorphic ignimbrites (e.g. Cougar Point Tuff
XI) appear to be lenticular zones of late-vesiculated welded
tuff matrix rather than former pumice clasts.
The stratified ashfall deposits with their widespread
distal correlatives have great value in long-distance,
terrestrial stratigraphy and tephrochronology (Perkins and
Nash 2002) and record widespread atmospheric dispersal of
306 Bull Volcanol (2008) 70:293–314

large volumes of vitric ash, and is associated with major
mammalian death assemblages, c.1,400 km to the west of
the eruptive sources (Voorhies and Thomasson 1979; Rose
et al. 2003). The widespread dispersal would have been
favoured by the combination of high eruptive temperatures
with high eruptive mass-fluxes. It may have occurred via
vent-derived convective ash plumes, and/or from phoenix
clouds (co-ignimbrite ash plumes) that lofted from the
large, hot pyroclastic density currents. Many of the ash
layers are coarser-grained than is typical of co-ignimbrite
ashes elsewhere (cf. Walker 1981; Smith and Houghton
1995), even distally (see Rose et al. 2003). This could
reflect the unusually large shard-sizes produced by SR-type
explosivity coupled with enhanced elutriation and lofting as
a result of the large thermal input by the exceptionally hot
pyroclastic currents. However, although some fine vitric
ash layers overlie several of the ignimbrites, a significant
proportion of the stratified ash occurs beneath ignimbrites
of the same eruption-unit; such ashes could only be of co-
ignimbrite origin if the leading edges of the pyroclastic
density currents advanced rather slowly across a landscape
that was already being mantled in co-ignimbrite ash from
earlier stages of the eruption. The stratified ash layers
beneath the ignimbrites could, alternatively, derive from
Plinian explosivity. However, they are finer grained than is
typical for Plinian deposits of a similar thickness: Plinian
deposits thicker than a metre or so are typically composed
of predominantly lapilli-sized pumice clasts (e.g. at Vesu-
vius, Valles, Santorini); because they tend to thin out as
they become more fine grained distally, a Plinian layer
more than a couple of metres thick of only ash sized
particles would represent an unusually large eruption.
Alternatively, the lack of pumice lapilli may be a facet of
the fragmentation mechanism. One possibility is that the
explosive fragmentation of the vesiculating magma was
enhanced by interaction with meteoric water. This would be
consistent with the widespread evidence for lacustrine
conditions: Snake River Plain volcanic rocks are important
aquifers, and one might expect that water occupied
fractures beneath lakewater, for example in areas of
proximal volcanic subsidence. The abundance of ash pellets
and accretionary lapilli in the rhyolitic ash also are
consistent with the involvement of external water. Such
ash aggregates form commonly (but not exclusively) during
phreatomagmatic eruptions, or where there is abundant
moisture in ash plumes such as is generated by surface
evaporation where a hot density current flows across
standing water. The parallel bedding and lamination
suggests unsteady, pulsatory fallout, generated either at
the vent or by convective instabilities within an umbrella
cloud (e.g. Carey et al. 1988; Branney 1991). Pulsatory
explosivity is also a characteristic of some phreatomag-
matic eruptions, and can arise from intermittent access of
water to the erupting magma. Some of the ash-pellet-
bearing ash deposits are very fine grained and have a
distinctly phreatomagmatic character (e.g. units beneath the
Tuff of Wooden Shoe Butte, Cassia Mountains).
However, the granulometry of many of the stratified ash
layers in the Snake River Plain is not typical of phreato-
magmatic ash. The ash layers are relatively well-sorted and
medium to coarse-grained, whereas phreatoplinian ashes
(e.g. Hatepe and Rotongiao ashes of New Zealand; Walker
1981) tend to be finer grained and less well-sorted. The
cuspate shape of the shards indicates that the magma was
coarsely vesicular at the time of fragmentation, and so the
explosive fragmentation was probably driven primarily by
volatile exsolution. Given that the absence of pumice
blocks and lapilli characterises the ignimbrites as well as
the fallout deposits, reconciling the involvement of large
volumes of meteoric water with the very high welding
intensity of the ignimbrites presents something of a
paradox. However, intensely welded pyroclastic rocks do
occur in association with non-welded, ash-aggregate bear-
ing deposits at some flooded caldera volcanoes, like Taal in
the Philippines (Torres et al. 1995) and Scafell in England
(Branney and Kokelaar 1995).
The abundant obsidian and vitrophyre fragments in the
ignimbrites and ashfall deposits may have been entrained at
walls of eruption conduits cutting dense proximal obsidian
and vitrophyre. The absence of other lithic lithologies (e.g.
basalts, basement) suggests that the explosivity occurred at
shallow-levels in eruptive centres that at the time were
characterised by thick accumulations of glassy rhyolitic
products, possibly in a lacustrine setting. Incorporation and
fragmentation of these lithologies may have been facilitated
by groundwater within fractures in the glassy rocks flashing
to steam. Similar angular glass chips are reported from
Plinian (Brown and Branney 2004) and phreatoplinian
(Smith and Houghton 1995) fallout deposits elsewhere, and
in the latter case a shallow obsidian source at the floor of a
caldera lake was invoked.
A feature of the SR-type rhyolite lavas and ignimbrites
of the central part of the Snake River Plain is that they have
remarkably low δ18O values (−1.4 to 3.8‰; Boroughs et al.
2005) compared to non-SR-type rhyolites elsewhere in the
province. The low values have been ascribed to large-scale,
shallow melting of Idaho batholith basement that had been
extensively hydrothermally altered during the Eocene
(Boroughs et al. 2005). Intriguingly, however, the low
δ
18O values coincide with the SR-type volcanism, with its
association with former lakes: further exploration of the
possible role of sub-lacustrine groundwater or hydrothermal
waters in fractures associated with rifting and/or caldera
formation might be illuminating.
The SR-type rhyolite lavas were emplaced onto low
slopes. Emplacement predominantly by ductile effusion
Bull Volcanol (2008) 70:293–314 307

rather than brittle and semi-brittle extrusion is indicated by
the complex refolded folds and the large ratio of non-
brecciated to brecciated lava. Their large sizes and runnout
distances, and low aspect-ratios (Fig. 2) indicate that the
lava flows were erupted at far higher mass-flux rates, and
with significantly lower viscosities than is typical for
rhyolite spines, domes or coulées. As with other examples
(e.g. Bracks Rhyolite of Trans-Pecos Texas; Henry et al.
1990) the aspect ratios of the silicic lavas are consistent
with estimated high magmatic temperatures and anhydrous
compositions. However, it is not clear what drove such
large volumes of magma so rapidly to the surface.
Caldera subsidence in the Snake River plain
Explosive eruptions larger than a few km3 generally
produce calderas, and it is likely that the larger SR-type
eruptions were caldera-forming. Silicic calderas are best
identified by thick intra-caldera ignimbrite together with
caldera-collapse breccias, and overlying caldera lake sedi-
ments, rhyolite domes, and associated shallow intrusions,
hydrothermal alteration, and contemporary faulting (e.g.
Lipman 1984; Branney 1995). Where these are not
exposed, the presence of a caldera may be inferred from
an extant topographic rim, together with proximally-coars-
ening lithic breccias in outflow ignimbrites (e.g. Druitt and
Sparks 1982) and thick, coarse pumice fall layers. Most of
these features have not been recorded in the central Snake
River Plain, so the location and size of any calderas in the
region is tentative. For these reasons, Bonnichsen (1982)
inferred the presence of broad ‘eruptive centres’ (Fig. 1) that
were considered to have been discrete topographic depres-
sions 1,000s km2 containing lakes and extensive rhyolite
lavas, prior to burial by younger basalt lavas.
Source locations of most of the ignimbrites are impre-
cisely known and cannot be inferred from thickness
variations of outflow sheets. Spatial distributions of
azimuth orientations of rheomorphic lineations, folds and
kinematic indicators can help locate an eruptive source
(Branney et al. 2004). For example, lineations in the upper
tuff of McMullen Creek, Cassia Hills (Fig. 1) form a fan-
shaped pattern that may indicate northerly source (McCurry
et al. 1996). However, rheomorphic transport directions are
likely to be influenced by local topographic slopes rather
than an overall transport direction from source (e.g. lower
units of the tuffs of McMullen Creek; McCurry et al. 1996).
It has been proposed that local successions of outflow
ignimbrites derive from a common eruptive centre, for
example that all nine Cougar Point Tuffs and associated
lavas derive from the Bruneau-Jarbidge eruptive centre, and
that the ignimbrites of the Cassia Mountains derive from
the inferred Twin Falls eruptive centre (Fig. 1). In this
scenario, a small number of distinct and spaced eruptive
centres (Fig. 1) each underwent numerous large eruptions.
It is equally possible, however, that a larger number of
subsidence structures formed, each accompanied by em-
placement of just one large ignimbrite. In this scenario, the
large number of ignimbrites would suggest that the central
and eastern Snake River Plain is a complex of numerous
partly overlapping calderas, rather like the Olympic rings,
in which the locus of subsidence migrated overall north-
eastwards. This is consistent with our understanding of the
younger calderas around Yellowstone, which are shingled,
and where each caldera subsided during a single ignimbrite
eruption (Morgan and McIntosh 2005). The diameters of
some of these calderas (Blacktail, Kilgore, Huckleberry
Ridge, and Yellowstone) are thought to be similar to the
width of the Snake River Plain. In detail, the temporal trend
is likely to have been rather more complex than a simple
north-eastward migration (e.g. a large ignimbrite flare-up
occurred 11.7–10.0 Ma from a broad region of the central
Snake River Plain, and temporal overlap of eruptions in
widely separate areas occurred during the subsequent ∼4
million years; Bonnichsen et al. 2007).
The low topography of the Snake River Plain relative to
the adjacent massifs is associated with marked down-
warping of ignimbrites and basement towards the axis of
the plain. This structure has been interpreted variously as
due to SW–NE extension, thermal contraction following the
volcanism, or loading by large mid-crustal mafic intrusions
(discussion in Rogers et al. 2002). The structurally-defined
subsidence exceeds 4.5 km and, in some locations, 8.5 km
(Rogers et al. 2002). We propose that this includes a
component of caldera subsidence; the central and eastern
Snake River Plain may thus be regarded as a volcano-
tectonic depression formed by successive, partly over-
lapping caldera subsidence events in addition to later
thermal adjustments and flexural loading by a mid-crustal
mafic intrusion (Rogers et al. 2002). The inward dip of the
basement and ignimbrite sheets towards the axis of the
plain may include a component of caldera-related down-
warping (‘downsag’ of Walker 1984; Branney 1995)
peripheral strata that steepen with proximity to the caldera
margins is a common feature of caldera subsidence
(Branney and Kokelaar 1992; Branney 1995; Roche et al.
2000; Kokelaar and Moore 2006), e.g. the inward tilting
around the 18 km diameter Namibian Messum Complex
and Goboboseb mountains, interpreted as due to caldera
subsidence associated with very large-volume silicic erup-
tions (Ewart et al. 1998).
Given the thickness of the outflow ignimbrite succes-
sions, we expect proximal ignimbrite accumulations >2 km
thick under the axis of the central Snake River Plain. These
thicknesses are not known (base not seen) but in the eastern
Snake River Plain seismic refraction data suggest rhyolitic
rocks are c. 2.5 km thick (Sparlin et al. 1982) and borehole
308 Bull Volcanol (2008) 70:293–314

INEL-1 penetrated 2.39 km of rhyolitic ignimbrite without
reaching its base; this may represent a caldera fill (Doherty
et al. 1979).
Other possible occurrences of SR-type silicic volcanism
Features of SR-type volcanism also occur elsewhere in the
Proterozoic and Phanerozoic record. Most, though not all,
examples involve large-volume eruptions of intracontinen-
tal, high-temperature and relatively anhydrous, potassic and
high-Fe, metaluminous rhyolites and dacites (or latites, e.g.
Marsh et al. 2001) in a bimodal association with volumi-
nous basaltic magmatism. In many cases, the physical
volcanology of the deposits, particularly of the non-welded
tephras, is poorly understood.
Elsewhere along the Yellowstone hot-spot track
Styles of volcanism vary along the Yellowstone hot-spot
track. The variation reflects temporal changes in both
chemistry and magmatic temperatures of the magmas
generated as the hot-spot encountered differences within
the overriding continental lithosphere (Perkins and Nash
2002). Amongst the resultant great diversity of facies
within the province, some successions outside the area
described in this paper (Fig. 1) also exhibit SR-type
characteristics. In the eastern Snake River Plain, several
rhyolitic ignimbrites are high to extremely high-grade,
rheomorphic, locally lava-like, and associated with paral-
lel-stratified ashes and aqueously deposited volcaniclastic
sands silts (e.g. Walcott Tuff, Blacktail Creek Tuff, Tuff of
Kilgore, and associated units; Morgan and McIntosh
2005). Other ignimbrites, however, are more transitional
in character and exhibit some but not all of the
characteristic features of SR-type volcanism (e.g. the of
Wolverine Creek; Morgan and McIntosh 2005). Well-
known ignimbrites from the Yellowstone area have both
welded and non-welded parts, eutaxitic fabrics, and restricted
rheomorphic, lava-like facies. They contain abundant pumice
clasts and were erupted at lower temperatures than typical SR-
type ignimbrites. Some Quaternary rhyolites in the volcanic
province (Magic Reservoir centre, Leeman 2004; Big Butte,
Cedar Butte and related rhyolite domes further east,
McCurry et al. 1999) contrast with the Miocene volcanic
rocks described in this paper, and are not of SR-type. The
13.8–12 Ma Owyhee-Humboldt eruptive centre (Fig. 1)
produced widely dispersed ash, intensely rheomorphic
ignimbrites some with lava-like facies, sheet-like rhyolite
lavas, rhyolite lavas inferred to have been fountain-fed
(clastogenic) and ascribed to eruptions of low-viscosity, hot
and relatively anhydrous rhyolite magma (Ekren et al. 1982;
Bonnichsen and Kauffman 1987; Manley 1995; Manley and
McIntosh 2002).
Eocene–Oligocene rhyolites of Trans-Pecos Texas
Associations of extensive, low-aspect ratio silicic lavas with
extremely high-grade rheomorphic ignimbrites, including
some lava-like units of equivocal origin occur in Trans-
Pecos Texas (Henry et al. 1988, 1989; Henry and Wolff
1992) and have characteristics of SR-type volcanism. The
38 Ma–32 Ma mafic to silicic volcanic rocks are related to
subduction beneath the North American continent, and
include caldera-related, large volume, predominantly
alkalic low-silica rhyolites. Intense welding and rheo-
morphism of ignimbrites (e.g. Buckshot ignimbrite,
Gomez Tuff and in the Barrel Springs and Sleeping Lion
formations, Henry et al. 1989), and the low aspect-ratios
of lavas (Fig. 2), such as the 25–120 m thick Bracks
Rhyolite, which covers 1,000 km2 and flowed ≤35 km
from source, have been ascribed to eruptive temperatures
≥900°C (Henry et al. 1990). Other ignimbrites exhibit
fiamme and lithic clasts, and are less typically SR-type.
Cretaceous Etendeka-Paraná volcanic province, Namibia,
Angola and South America
Vast silicic sheets of the early Cretaceous Etendeka
volcanic provinces of Namibia and Angola, and the
southern Paraná of South America (Marsh et al. 2001;
Garland et al. 1995; Milner et al. 1995; Kirstein et al. 2001)
have features of SR-type volcanism. Individual sheets in
Namibia are 70–250 m thick, exceed 25,000 km2 (Milner et
al. 1995; Jerram 2002), and have estimated magmatic
temperatures exceeding 1,000°C. They have been inter-
preted as predominantly lava-like ignimbrites, although few
unequivocal pyroclastic features are preserved and some
have basal breccias (Milner et al. 1992; Ewart et al. 1998).
Some units contain non-flattened globules, some larger than
those in the Snake River ignimbrites. Little has been
published on the fallout tephras, and few of the eruptive
centres are well exposed, with the exception of the Messum
Complex (Ewart et al. 1998).
Middle Proterozoic Keweenawan volcanics, Minnesota
Low-aspect ratio rhyolite lavas and extremely high-grade
rheomorphic ignimbrites of SR-type have been described in
the Middle Proterozoic Keweenawan plateau volcanics of
Minnesota (Green 1989; Green and Fitz 1993). They reach
250 m in thickness and trace for 40 km, with volumes of
100–400 km3. Some are lava-like and their pyroclastic
versus effusive origin is equivocal. Less is known about the
non-welded pyroclastic units, although one 20 cm-thick
layer of pumice lapilli is reported beneath a lava. Former
surface water is indicated by thin units of fluvio-deltaic
sediments and basaltic pillow lavas. Temperatures of c.
Bull Volcanol (2008) 70:293–314 309

900°C have been invoked for the rhyolites, which are
thought to have derived from crustal melting during
continental rifting (Vervoort and Green 1997).
Jurassic Karoo volcanic province, southern Africa
Rhyolite sheets with some characteristics of SR-type
volcanism occur within the Lebombo monocline of south-
ern Africa. The rhyolitic eruptions were large, with a
combined volume of 35,000 km3 (Cleverly et al. 1984).
Individual rhyolite sheets are ≤60 km, ≤200m thick, and have
been interpreted as extremely high-grade, largely lava-like
rheomorphic ignimbrites (Cleverly 1979). They have low
δ
18O values (c. 5.6 ‰) ascribed to deep penetration and
extensive circulation of meteoric water facilitated by brittle
fracturing accompanying continental rifting (Harris and
Erlank 1992).
Middle Proterozoic Yardea Dacite, south Australia
The Yardea Dacite (67% SiO2) of south Australia is as much
as 250 m thick and covers an area of 12,000 km2 (Creaser
and White 1991). It is not known whether or not it is a single
eruptive unit, but it shows mostly lava-like characteristics
and may be either rheomorphic ignimbrite or unusually
extensive lava. It is thought to have been erupted from a hot
(>1,000°C) felsic magma that was relatively poor in H2O.
Unlike most SR-type rhyolites, however, it locally contains
abundant lithic fragments, and the magmas are slightly more
mafic and crystal rich. Little is published on the volcanology
of associated facies (Creaser and White 1991).
Ordovician Scafell caldera volcano England, UK
Associations of intensely rheomorphic ignimbrites, parallel-
stratified ashes and lacustrine volcaniclastic sediments occur
in Ordovician extensional continental arc successions of the
English Lake District (Millward 2004; Branney 2006). Some
of the rheomorphic ignimbrites, such as the Bad Step Tuff in
Scafell caldera are of SR-type, with extensive lava-like
facies, rheomorphic folding, sheeting joints, zones of
lithophysae zones, and upper autobreccias (Branney et al.
1992). Intercalated thin layers of stratified tuffs lack pumice
lapilli. Some have ash pellets, trace for long distances, and
resemble ashfall layers of the central Snake River Plain.
Some are interpreted as recording very large-volume
phreatoplinian ashfalls that blanketed a subaerial landscape
and, proximally, fell into subsiding syn-eruptive lakes
(Branney 1991). The rheomorphic ignimbrite eruptions are
thought to have been characterized by high temperature, high
mass-flux pyroclastic fountaining generating hot density
currents, and the intercalated non-welded stratified units are
thought to record intermittent explosive interaction with
ground and lake water. Differences with SR-type volcanism
include the presence of lava domes and coulees, an
intracaldera setting, and ignimbrites that contain pumice
and fiamme (Branney and Kokelaar 1995).
Silurian Glencoe caldera volcano, Scotland, UK
The Lower, Middle and Upper Etive Rhyolites of Glencoe
caldera volcano, Scotland, are each 100–150 m thick,
predominantly lava-like, intensely rheomorphic ignimbrites
with contorted flow lamination and upper autobreccias, and
they occur interstratified with fluvial and lacustrine sedi-
ments and thin layers of stratified rhyolitic ash containing
ash aggregates (Kokelaar and Moore 2006). They are
overlain by more conventional eutaxitic ignimbrites. They
are high-K rhyolites with 74–77% silica and thought to
represent crustal melts erupted in a regional volcanic flare-
up in a transtensional continental setting.
Most examples of volcanic successions that exhibit SR-
type features record large-volume, high mass-flux eruptions
of hot, relatively anhydrous, high-K, metaluminous rhyo-
lite, at continental large igneous provinces (LIPs). With
some exceptions, relatively little is known about their
physical volcanology; for example, the nature and dispersal
characteristics of non-welded tephras—such studies have
been hindered by the vast scale of the phenomena. Clearly,
not all silicic volcanism at continental LIPs is of SR-type.
For example, large-volume Oligocene silicic pyroclastic
eruptions in Yemen–Ethiopia bimodal LIP produced low- to
moderate-grade welded ignimbrites in which pumice lapilli
and fiamme are abundant (e.g. Peate et al. 2005).
Conclusion: a summary of the characteristic features
of SR-type volcanism
The characteristic features of SR-type volcanism as
exhibited by the type example, the Miocene rocks of
south-central Idaho and northernmost Nevada, are listed (1–
18 and Appendix Table 1). (1) A bimodal association of
basalts (‘basaltic plains-style’ volcanism) with H2O-poor,
metaluminous rhyolites (70–77% SiO2) that have elevated
contents of HFSE and halogens. (2) Large, regionally
devastating explosive rhyolitic eruptions (VEI 6–8). (3)
Voluminous, high to low aspect-ratio rhyolitic ignimbrites.
(4) Numerous ashfall layers with parallel thin-bedding and
lamination; extensive (to 1,000s of km) atmospheric
dispersal of ash. (5) Unusually large-volume, extensive
rhyolite lavas with low aspect-ratios. Domes and coulées
are not characteristic. (6) A paucity of pumice lapilli in the
ignimbrites (or, in welded ignimbrites, of fiamme after
pumice lapilli) and an absence of pumice lenses and
pumice-concentration zones. (7) The ignimbrites are gen-
310 Bull Volcanol (2008) 70:293–314

erally lithic-poor and lack proximal lithic breccias. (8)
Small angular glassy fragments are abundant both in many
ashfall layers and ignimbrites. (9) The ignimbrites are true
tuffs rather than lapilli-tuffs, and they are unusually well
sorted (σ
φ
<2.5) for ignimbrites. (10) Most are extremely
high-grade, with few low- to medium-grade ignimbrites,
such as sillar and moderately welded eutaxitic ignimbrite.
(11) Rheomorphism is typically intense and widespread,
even in some thin outflow sheets, and in some cases
welding (agglutination and coalescence) and rheomorphism
started while the ignimbrite was still being deposited. (12)
Lava-like facies are common, and dominate many of the
ignimbrite sheets. (13) The ignimbrites have well-devel-
oped lower and upper vitrophyres with extensive spherulitic
zones, some with large (≤30 cm diameter) lithophysae. (14)
Plinian pumice fall layers with framework-supported
angular pumice lapilli are relatively rare. (15) Fusing of
ashfall deposits is common, sometimes associated with
baked palaeosols. (16) Ashes and tuffs are composed
predominantly of platy, bicuspate and tricuspate shards,
with rare globule-shaped shards. Large, megascopic cuspate
vitric shards are commonly visible in the field. (17) Ash
aggregates (pellets, accretionary lapilli, and cored accre-
tionary lapilli) are common, and some of the fine-grained
silicic ashes may have a phreatomagmatic origin. (18)
Evidence for surface water is widespread (rippled silts and
sands, reworked ash, ignimbrite peperites, rhyolite hyalo-
clastites, basalt pillow lava and hyaloclastite deltas, tuff
cones, and tuff rings), and there are also alluvial sands,
palaeosols and bioturbation in non-welded layers.
The eruption mechanisms that give rise to this associa-
tion must differ significantly from pumice-rich explosive
eruptions typical of many other volcanic provinces (see
Appendix Table 1). They involve sustained, high mass-flux,
volatile-exsolution driven explosivity of very hot, and
unusually low-viscosity, relatively anhydrous rhyolitic
magma. It is possible, but not established, that volatile-
driven fragmentation was enhanced by interaction with
meteoric water beneath lakes ponded in topographic lows
related to continental rifting and caldera subsidence. Cool-
ing during transport was minimised in sustained, high
mass-flux granular-fluid based pyroclastic density currents,
and some volatile species (possibly halogens) remained
dissolved within the hot pyroclasts long enough to maintain
low viscosities during transport, welding, rheomorphism
and cause late-stage re-vesiculation of the welded rheo-
morphic ignimbrite. We clearly have much to learn about
these particularly awesome, enigmatic eruptions.
Acknowledgements GDMA and BSE acknowledge NERC PhD
studentships NER/S/A/2001/06292 and NER/S/A/2004/12340 held at
the University of Leicester, UK. Many thanks to Martha Godchaux,
Bill Leeman, and Dougal Jerram for discussion in the field and
improvements to the manuscript.
References
Andrews GDM (2006) The emplacement and deformation of high-
temperature tuffs: a structural analysis of the Grey’s Landing
ignimbrite, Snake River Plain, Idaho. PhD thesis, University of
Leicester, UK
Andrews GDM, Branney MJ (2005) Folds, fabrics, and kinematic
criteria in rheomorphic ignimbrites of the Snake River Plain,
Idaho: insights into emplacement and flow. Geol Soc Am Field
Guide 6:311–328
Andrews GDM, Branney MJ, Bonnichsen B (2007) Volcanic
stratigraphy and eruption history of the voluminous high-grade
ignimbrites and tephra deposits of the Rogerson Graben, Snake
River Plain, Idaho. Bull Volcanol (this issue)
Bennett VC, DePaolo DJ (1987) Proterozoic crustal history of the
western United States as determined by neodymium isotopic
mapping. Geol Soc Amer Bull 99:674–685
Bonnichsen B (1982) Rhyolite lava flows in the Bruneau-Jarbidge Eruptive
Centre, southwestern Idaho. In: Bonnichsen B, Breckenridge RM
(eds) Cenozoic Geology of Idaho. Idaho Bur Mines Geol Bull
26:283–320
Bonnichsen B, Christiansen RL, Morgan LA, Moye FJ, Hackett WR,
Leeman WP, Honjo N, Jenks MD, Godchaux MM (1989)
Excursion 4A: Silicic volcanic rocks in the Snake River Plain
—Yellowstone Plateau province. In: Chapin CE, Zidek J (eds)
Field excursions to volcanic terranes in the western United
States; Volume II, Cascades and Intermountain West. NM Bur
Mines Mineral Resourc Mem 47:135–182
Bonnichsen B, Citron GP (1982) The Cougar Point Tuff, southwestern
Idaho. In: Bonnichsen B, Breckenridge RM (eds) Cenozoic
Geology of Idaho. Idaho Bur Mines Geol Bull 26:255–281
Bonnichsen B, Kauffman DF (1987) Physical features of rhyolite lava
flows in the Snake River Plain volcanic province, Southwestern
Idaho. Geol Soc Am 212:119–145 (Special Paper)
Bonnichsen B, Godchaux MM (2002) Late Miocene, Pliocene and
Pleistocene geology of southwestern Idaho with emphasis of
basalts in the Bruneau-Jarbidge, Twin Falls, and Western Snake
River Plain regions. In: Bonnichsen B, White CM, McCurry M
(eds) Tectonic and magmatic evolution of the Snake River Plain
Volcanic Province. Idaho Geol Surv Bull 30:387–434
Bonnichsen B, Leeman WP, Honjo, N, McIntosh WC, Godchaux MM
(2007) Miocene silicic volcanism in southwestern Idaho: geo-
chronology, geochemistry, and evolution of the central Snake
River Plain. Bull Volcanol (this issue)
Boroughs S, Wolff J, Bonnichsen B, Godchaux M, Larson P (2005)
Large-volume, low-δ18O rhyolites of the central Snake River Plain,
Idaho, USA. Geology 33(10):821–824; DOI 10.1130/G21723.1
Branney MJ (1991) Eruption and depositional facies of the Whorney-
side Tuff: an exceptionally large-magnitude phreatoplinian
eruption. Bull Geol Soc Am 203:886–897
Branney MJ (1995) Downsag and extension at calderas: new
perspectives on collapse geometries from ice-melt, mining, and
volcanic subsidence. Bull Volcanol 57:303–318
Branney MJ (2006) The Borrowdale Volcanic Group and Ordovician
continental arc volcanism in northern England In: Brenchley PJ,
Rawson PF (eds) The Geology of England and Wales, 2nd ed. Geol
Soc London pp 113–122
Branney MJ, Kokelaar BP (1992) A reappraisal of ignimbrite
emplacement: changes from particulate to non-particulate flow
during progressive aggradation of high-grade ignimbrite. Bull
Volcanol 54:504–520
Branney MJ, Kokelaar BP (1995) Volcanotectonic faulting, soft-state
deformation, and rheomorphism of tuffs during development of a
piecemeal caldera, English Lake District. Geol Soc Amer Bull
106:507–530
Bull Volcanol (2008) 70:293–314 311

Branney MJ, Kokelaar BP, McConnell BJ (1992) The Bad Step Tuff:
a lava-like ignimbrite in a calc-alkaline piecemeal caldera,
English Lake District. Bull Volcanol 54:187–199
Branney MJ, Kokelaar BP (2002) Pyroclastic density currents and the
sedimentation of ignimbrites. Geol Soc Lond, Memoirs 27:1–152
Branney MJ, Barry TL, Godchaux M (2004) Sheathfolds in
rheomorphic ignimbrites. Bull Volcanol 66:485–491
Brown RJ, Branney MJ (2004) Event-stratigraphy of a caldera-
forming ignimbrite eruption on Tenerife: the 273 ka Poris
Formation. Bull Volcanol 66:392–416
Bursik MI, Woods AW (1996) The dynamics and thermodynamics of
large ash flows. Bull Volcanol 38:175–193
Carey SN, Sigurdsson H, Sparks RSJ (1988) Experimental studies of
particle-laden plumes. J Geophys Res 93: 15314–15328
Cas RAF, Wright JV (1987) Volcanic successions: modern and
ancient. Alan & Unwin, London, pp 1–528
Cathey HE, Nash BP (2004) The Cougar Point Tuff: implications for
thermochemical zonation and longevity of high-temperature,
large volume silicic magmas of the Miocene Yellowstone
hotspot. J Petrol 45:27–58
Chapin CE, Lowell GR (1979) Primary and secondary flow structures
in ash-flow tuffs of the Gribbles Run Palaeovalley, central
Colorado. In: Chapin CE, Elston WE (eds) Ash-flow tuffs. Geol
Soc Am Spec Pap 180:137–154
Christiansen EH, McCurry M (2007) Contrasting origins of Cenozoic
silicic volcanic rocks from the western Cordillera of the United
States. Bull Volcanol (this issue)
Cleverly RW (1979) The volcanic geology of the Lebombo monocline
in Swaziland. Tran Geol Soc S Africa 82:227–230
Cleverly RW, Betton PJ, Bristow JW (1984) Geochemistry and
petrogenesis of the Lebombo rhyolites. Geol Soc S Africa Spec
Pub 13:171–195
Creaser RA, White AJ (1991) The Yardea Dacite—large-volume,
high-temperature felsic volcanism from the Middle Proterozoic
of South Australia. Geology 19:48–51
Cummings ML, Evans JG, Ferns ML, Lees KR (2000) Stratigraphy
and structural evolution of the middle Miocene synvolcanic
Oregon-Idaho graben. Geol Soc Amer Bull 112:668–682
Doherty DJ, McBroome LA, Kuntz MA (1979) Preliminary geolog-
ical interpretation and lithologic log of the exploratory geother-
mal test well (INEL-1), Idaho National Engineering Laboratory,
eastern Snake River Plain, Idaho. US Geol Surv Open File Rep
79–1248 pp 10
Druitt TH, Sparks RSJ (1982) A proximal ignimbrite breccia facies on
Santorini volcano, Greece. J Volcanol Geotherm Res 13:141–171
Duffield WA (1990) Eruptive fountains of silicic magma and their
possible effects on the tin content of fountain-fed lavas, Taylor
Creek Rhyolite, New Mexico. In: Stein HJ, Hannah JL (eds) Ore-
bearing granite systems: petrogenesis and mineralising processes.
Geol Soc Am Spec Pap 246:251–261
Ekren EB, McIntyre DH, Bennett EH, Marvin RF (1982) Cenozoic
stratigraphy of western Owyhee County, Idaho. In: Bonnichsen
B, Breckenridge RM (eds) Cenozoic Geology of Idaho. Idaho
Bur Mines Geol Bull 26:215–235
Ekren EB, McIntyre DH, Bennett EH (1984) High-temperature, large-
volume, lava like ash-flow tuffs without calderas in southwestern
Idaho. US Geol Surv Prof Pap 1272
Ewart A, Milner SC, Armstrong RA, Duncan AR (1998) Etendeka
volcanism of the Gobobseb Mountains and Messum Igneous
Complex, Namibia, Parts I and II. J Petrol 39:191–253
Freundt A (1999) Formation of high-grade ignimbrites. Part II. A
pyroclastic suspension current model with implications also for
low-grade ignimbrites. Bull Volcanol 60:545–576
Garland F, Hawksworth CJ, Mantovani MSM (1995) Description and
petrogenesis of the Paraná rhyolites, Southern Brazil. J Petrol
36:1193–1227
Godchaux MM, Bonnichsen B (2002) Syneruptive magma–water and
posteruptive lava–water interactions in the Western Snake River
Plain, Idaho, during the past 12 million years. In: Bonnichsen B,
White CM, McCurry M (eds) Tectonic and magmatic evolution
of the Snake River Plain Volcanic Province. Idaho Geol Surv
Bull 30:387–434
Greeley R (1977) Basaltic “plains” volcanism. In: Greeley R, King JS
(eds) Volcanism of the eastern Snake River Plain, Idaho: a
comparitive planetary guidebook.Washington DC, NASA, pp 23–44
Greeley R (1982) The Snake River Plain, Idaho: representative of a
new category of volcanism. J Geophys Res 2705–2712
Green JC (1989) Physical volcanology of mid-Proterozoic plateau
lavas: the Keweenawan North Shore Volcanic Group, Minnesota.
Geol Soc Amer Bull 101:486–500
Green JC, Fitz TJ III (1993) Extensive felsic lavas and rheoignim-
brites in the Keweenawan Midcontinent Rift plateau volcanics,
Minnesota: petrographic and field recognition. J Volcanol Geo-
therm Res 54:177–196
Harris C, Erlank AJ (1992) The production of large-volume, low-δ18O
rhyolites during the rifting of Africa and Antarctica: the
Lebombo Monocline, southern Africa. Geochem Cosmochim
Acta 56:3561–3570
Hearst J (1999) Depositional environments of the Birch Creek local
fauna (Pliocene: Blancan), Owyhee County, Idaho. In: Akersten
WA, McDonald HG, Meldrum DJ, Flint MET (eds) Papers on the
Vertebrate Paleontology of Idaho Honoring John A White, 1:
Idaho Mus Nat Hist Occ Pap 36:56–93
Henry CD, Wolff JA (1992) Distinguishing strongly rheomorphic tuffs
from extensive silicic lavas. Bull Volcanol 54:171–186
Henry CD, Price JG, Rubin JN, Parker DF, Wolff JA, Self S, Franklin
R, Barker DS (1988) Widespread, lava-like silicic volcanic rocks
of Trans-Pecos Texas. Geology 16:509–512
Henry CD, Price JG, Parker DF, Wolff JA (1989) Mid-Tertiary silicic
alkaline magmatism of Trans-Pecos Texas: rheomorphic tuffs and
extensive silicic lavas. Excursion 9A: Silicic volcanic rocks in
the Snake River Plain—Yellowstone Plateau province. In:
Chapin CE, Zidek J (eds) Field excursions to volcanic terranes
in the western United States; Volume I, Southern Rocky
Mountains Region. NM Bur Mines Mineral Resourc Mem
47:231–274
Henry CD, Price JG, Rubin JN, Laubach SE (1990) Case study of an
extensive silicic lava: the Bracks Rhyolite, Trans-Pecos Texas. J
Volcanol Geotherm Res 43:113–132
Hildebrand RT, Newman KR (1985) Miocene sedimentation in the
Goose Creek basin, south-central Idaho, north eastern Nevada
and northwestern Utah. In: Flores RM et al. (eds) Cenozoic
paleogeography of the western US: Rocky Mountains Section,
Soc Econ Paleont Mineral 55–70
Honjo N, Leeman WP (1987) Origin of hybrid ferrolatite lavas from
Magic Reservoir eruptive centre, Snake River Plain, Idaho.
Contrib Mineral Petrol 96:163–177
Honjo N, Bonnichsen B, Leeman WP, Stormer Jr JC (1992)
Mineralogy and geothermometry of high-temperature rhyolites
from the central and western Snake River Plain. Bull Volcanol
54:220–237
Iddings JP (1889) The rhyolites. In: Geology of the Yellowstone
National Park. US Geol Surv Monogr 32(Part II):356–432
Jenks MD, Bonnichsen B (1989) Subaquesous basalt eruptions into
Pliocene Lake Idaho, Snake River Plain, Idaho. In: Chamberlain
VE, Breckenridge RM, Bonnichsen B (eds) Guidebook to the
geology of Northern and Western Idaho and surrounding area.
Idaho Geol Surv Bull 28:17–34
Jerram DA (2002) Volcanology and facies architecture of flood
basalts. In: Menzies MA, Klemperer SL, Ebinger CJ and Baker
J (eds) Volcanic Rifted Margins. Geol Soc Amer Spec Pap
362:119–132
312 Bull Volcanol (2008) 70:293–314

Johnson RW (1968) Volcanic globule rock from Mount Suswa,
Kenya. Geol Soc Amer Bull 79:647–651
Kimmel PG (1982) Stratigraphy, age, and tectonic setting of the
Miocene–Pliocene lacustrine sediments of the Western Snake
River Plain, Oregon and Idaho. In: Bonnichsen B, Breckenridge
RM (eds) Cenozoic geology of Idaho. Idaho Bur Mines Geol
Bull 26: 559–578
Kirstein LA, Hawksworth CJ, Garland FG (2001) Felsic lavas or
rheomorphic ignimbrites: is there a chemical distinction? Contrib
Mineral Petrol 142:309–322
Kokelaar BP (1982) Fluidization of wet sediment during the
emplacement and cooling of various igneous bodies. J Geol
Soc Lond 139:21–33
Kokelaar BP (1986) Magma–water interactions in subaqueous and
emergent basaltic volcanism. Bull Volcanol 48:275–289
Kokelaar BP, BranneyMJ (1999) Inside silicic calderas (Snowdon, Scafell
and Glencoe, UK): interactions of caldera development, tectonism
and hydrovolcanism. Field Guide to the IAVCEI Commission on
Explosive Volcanism Field Workshop 7–18 July, 1999, pp 154
Kokelaar BP, Königer A (2000) Marine emplacement of welded
ignimbrite: the Ordovician Pitts Head Tuff, North Wales. J Geol
Soc Lond 157:517–536
Kokelaar BP, Moore ID (2006) Classic areas of British geology:
Glencoe caldera volcano, Scotland. Keyworth, Nottingham,
British Geol Surv, pp 127
Leeman, WP (1982a) Rhyolites of the Snake River Plain—Yellowstone
Plateau Province, Idaho and Wyoming: a summary of petrogenetic
models. In: Bonnichsen B, Breckenridge RM (eds) Cenozoic
geology of Idaho. Idaho Bur Mines Geol Bull 26:203–212
LeemanWP (1982b) Geology of the Magic Reservoir area, Snake River
Plain, Idaho. In: Bonnichsen B, Breckenridge RM (eds) Cenozoic
geology of Idaho. Idaho Bur Mines Geol Bull 26:369–376
Leeman WP (2004) Evolution of Snake River Plain (SRP) silicic
magmas—the Magic Reservoir eruptive center. Geol Soc Am.
Abst with Prog 36: 24
Leeman WP, Oldow JS, Hart WK (1992) Lithosphere-scale thrusting in
the western U.S. Cordillera as constrained by Sr and Nd isotopic
transitions in Neogene volcanic rocks. Geology 20:63–66
Link PK, McDonald HG, Fanning CM, Godfrey AE (2002) Detrital
zircon evidence for Pleistocene drainage reversal at Hagerman
Fossil Beds National Monument, central Snake River Plain,
Idaho. In: Bonnichsen B, White CM, McCurry M (eds) Tectonic
and magmatic evolution of the Snake River Plain Volcanic
Province: Idaho Geol Surv Bull 30:105–119
Lipman PW (1984) Roots of ash-flow calderas in western North
America: windows into the tops of granitic batholiths. J Geophys
Res 89:8801–8841
Mahood GA (1984) Pyroclastic rocks and calderas associated with
strongly peralkaline rocks. J Geophys Res B89:8540–8552
Manley CR (1995) How voluminous rhyolitic lavas mimic rheo-
morphic ignimbrites: eruptive style, emplacement conditions, and
formation of tuff-like textures. Geology 23:349–352
Manley CR (1996) In situ formation of welded tuff-like textures in the
carapace of a voluminous silicic lava flow, Owyhee County, SW
Idaho. Bull Volcanol 57:672–686
Millward D (2004) The Caradoc volcanoes of the English Lake
District. Proc Yorks Geol Soc 55:145–177
Malde HE, Powers HA (1962) Upper Cenozoic stratigraphy of western
Snake River Plain, Idaho: Geol Soc Amer Bull 73:1197–1220
Manley CR, McIntosh WC (2002) The Juniper Mountain Volcanic
Center, Owyhee County, Southwestern Idaho: age relations and
physical volcanology. In: Bonnichsen B, White CM, McCurry M
(eds) Tectonic and magmatic evolution of the Snake River Plain
Volcanic Province. Idaho Geol Surv Bull 30:205–227
Marsh JS, Ewart A, Milner SC, Duncan AR, Miller RMcG (2001)
The Etendeka Igneous Province: magma types and their
stratigraphic distribution with implications for the evolution
of the Paraná-Etendeka Flood Basalt Province. Bull Volcanol
62:464–486
McCurry M (eds) Tectonic and magmatic evolution of the Snake
River Plain Volcanic Province. Idaho Geol Surv Bull 30:233–312
Morgan LA, McIntosh WC (2005) Timing and development of Heise
volcanic field, Snake River Plain, Idaho, western USA. Geol Soc
Amer Bull 117:288–306
McCurry M, Bonnichsen B, White C, Godchaux MM, Hughes SS
(1997) Bimodal basalt-rhyolite magmatism in the central and
western Snake River Plain, Idaho and Oregon. In: Link PK,
Kowallis BJ (eds) Proterozoic to Recent stratigraphy, tectonics,
and volcanology, Utah, Nevada, southern Idaho, and central
Mexico. Brigham Young Univ Geol Stud 42:381–422
McCurry M, Watkins AM, Parker JL, Wright K, Hughes SS (1996)
Preliminary volcanological constraints for sources of high-grade,
rheomorphic ignimbrites of the Cassia Mountains, Idaho:
implications for the evolution of the Twin Falls Volcanic Centre.
Northwest Geol 26:81–91
McCurry M, Hackett WR, Hayden K (1999) Cedar Butte and
cogenetic rhyolite domes of the Eastern Snake River Plain. In:
Hughes SS, Thackray GD (eds) Guidebook to the Geology of
Eastern Idaho: Idaho Mus Nat Hist: 169–179
Milner SC, Duncan AR, Ewart A (1992) Quartz latite rheoignimbrite
flows of the Etendeka Formation, north-western Namibia. Bull
Volcanol 54:200–219
Milner SC, Duncan AR, Whittingham AM, Ewart A (1995) Trans-
Atlantic correlation of eruptive sequences and individual silicic
volcanic units within the Paraná-Etendeka igneous province. J
Volcano Geotherm Res 69:137–157
Pichler H (1981) Italienische Vulkan-Gebiete III: Lipari, Vulcano,
Stromboli. Samml Geol Führ 69:233
Peate IU, Baker JA, Al-Kadasi M, Al-Subbary A, Knight KB,
Riisager P, Thirlwall MF, Peate DW, Renne PR, Menzies MA
(2005) Volcanic stratigraphy of large-volume silicic pyroclastic
eruptions during Oligocene Afro-Arabiam flood volcanism in
Yemen. Bull Volcanol 68: 135–156
Perkins ME, Nash BP (2002) Explosive silicic volcanism of the
Yellowstone hotspot: the ash fall tuff record. Geol Soc Amer Bull
114:367–381
Perkins ME, Nash WP, Brown FH, Fleck RJ (1995) Fallout tuffs of
Trapper Creek Idaho—a record of Miocene explosive volcanism
in the Snake River Plains volcanic province. Geol Soc Amer Bull
107:1484–1506
Perkins ME, Brown FH, Nash WP, McIntosh W, Williams SK (1998)
Sequence, age, and source of silicic fallout tuffs in middle to late
Miocene basins of the northern Basin and Range province. Geol
Soc Am Bull 110:344–360
Pierce KL, Morgan LA, Saltus RW (2002) Yellowstone plume head:
postulated tectonic relations to the Vancouver slab, continental
boundaries, and climate. In: Bonnichsen B, White CM, McCurry
M (eds) Tectonic and magmatic evolution of the Snake River
Plain Volcanic Province. Idaho Geol Surv Bull 30:5–53
Quane SL, Russell JK (2005) Ranking welding intensity in pyroclastic
deposits. Bull Volcanol 67:129–143
Ross CS, Smith RL (1961) Ash-flow tuffs, their origin, geological
relations and identification. US Geol Surv Prof Pap 366:1–77
Russell JK, Quane SL (2005) Rheology of welding: inversion of field
constraints. J Volcanol Geotherm Res 142:173–191
Roche O, Druitt OH, Merle O (2000) Experimental study of caldera
formation. J Geophys Res 105:395–416
Rogers DW, Ore HT, Bobo RT, McQuarrie N, Zentner N (2002)
Extension and subsidence of the eastern Snake River Plain,
Idaho. In: Bonnichsen B, White CM, McCurry M (eds) Tectonic
and magmatic evolution of the Snake River Volcanic Province.
Idaho Geol Surv Bull 30:121–155
Bull Volcanol (2008) 70:293–314 313

Rose WI, Riley CM, Dartevelle S (2003) Sizes and shapes of 10-Ma
distal fall pyroclasts in the Ogallala Group, Nebraska. J Geol
111:115–124
Shervais JW, Kauffman JD, Gillerman VS, Othberg KL, Vetter SK,
Hobson VR, Zarnetsske M, Cooke F, Matthews SH, Hanan BB
(2005) Basaltic volcanism of the central and western Snake River
Plain: a guide to field relations between Twin Falls and Mountain
Home, Idaho. In: Pederson J, Dehler CM (eds) GSA Field Guide
Interior Western US 6:27–52
Smith KR, Swirydczuk K, Kimmel PG, Wilkinson BH (1982) Fish
biostratigraphy of late Miocene to Pleistocene sediments of the
western Snake River Plain, Idaho. In: Bonnichsen B, Breckenridge
RM (eds) Cenozoic geology of Idaho. Idaho Geol Surv Bull
26:519–541
Smith RT, Houghton BF (1995) Vent migration and changing eruptive
style during the 1800a Taupo eruption: new evidence from the
Hatepe and Rotongaio phreatoplinian ashes. Bull Volcanol
57:432–439
Sparks RSJ, Stasuik MV, Gardeweg M, Swanson DA (1993) Welded
breccias in andesite lavas. J Geol Soc Lond 150:897–902
Sparks RSJ, Self S, working group (2005). Super-eruptions: global
effects and future threats. Rep Geol Soc London Working Group
(2nd ed) Geol Soc, London, UK, p 24
Sparlin MA, Braile LW, Smith RB (1982) Crustal structure of the
Eastern Snake River Plain determined from ray trace modelling
of seismic refraction data. J Geophys Res 87(B4):2619–2633
Sumner JM, Branney MJ (2002) The emplacement of a remarkable
heterogeneous, chemically zoned and locally lava-like rheo-
morphic ignimbrite: ‘TL’ on Gran Canaria. J Volcanol Geotherm
Res 115:109–138
Swirydczuk K, Larson GP, Smith GR (1982) Volcanic ash beds as
stratigraphic markers in the Glenns Ferry and Chalk Hills
formations from Adrian, Oregon, to Bruneau, Idaho. In:
Bonnichsen B, Breckenridge RM (eds) Cenozoic geology of
Idaho. Idaho Geol Surv Bull 26:543–558
Torres RC, Self S, Punongbayan RS (1995) Attention focuses on Taal:
decade volcano of the Philippines. EOS Trans Am Geophys
Union 76:241–247
Voorhies MR, Thomasson JR (1979) Fossil grass anthoecia within
Miocene Rhinoceras Skeletons: diet in an extinct species. Science
206:331–333
Vervoort JD, Green JC (1997) Origin of evolved magmas in the
Midcontinent Rift System, NE Minnesota: Nd isotope evidence
for melting of Archean crust. Can J Earth Sci 34:521–535
Walker GPL (1971) Grain-size characteristics of pyroclastic deposits.
J Geol 79:696–714
Walker GPL (1981) Two phreatoplininan deposits and their water-
flushed origin. J Volcanol Geotherm Res 9:395–407
Walker GPL (1983) Ignimbrite types and ignimbrite problems. J
Volcanol Geotherm Res 17:65–88
Walker GPL (1984) Downsag calderas, ring faults, caldera sizes, and
incremental caldera growth. J Geophys Res 89(B10):395–407
Walker GPL, Croasdale R (1972) Characteristic of some basaltic
pyroclasts. Bull Volcanol 35:303–317
Wheeler HE, Cook EF (1954) Structural and stratigraphic significance
of the Snake River capture, Idaho–Oregon. J Geol 62:525–536
Wood SH, Clemens DM (2002) Geologic and tectonic history of the
Western Snake River Plain, Idaho and Oregon. In: Bonnichsen B,
White CM, McCurry M (eds) Tectonic and magmatic evolution
of the Snake River Plain Volcanic Province. Idaho Geol Surv
Bull 30:69–104
Wooden JL, Mueller PA (1988) Pb, Sr, and Nd isotopic compositions
of a suite of Late Archean igneous rocks, eastern Beartooth
Mountains: Implications for crust–mantle evolution. Earth Planet
Sci Lett 87:59–72
Wright KE, McCurry M, Hughes SS (2002) Petrology and geochem-
istry of the Miocenetuff of McMullen Creek, central Snake River
Plain. In: Bonnichsen B, McCurry M, White CM (eds) Tectonic
and Magmatic Evolution of the Snake River Plain Province.
Idaho Geol Surv Bull 30:177–194
314 Bull Volcanol (2008) 70:293–314