Bulletin of Volcanology

, 75:739 | Cite as

Unbedded diatreme deposits reveal maar-diatreme-forming eruptive processes: Standing Rocks West, Hopi Buttes, Navajo Nation, USA

  • N. S. Lefebvre
  • J. D. L. White
  • B. A. Kjarsgaard
Research Article


Diatremes provide partial records of how dyke-delivered magma, periodically interacting with water, produces the largest known cylindrical conduit structures of any volcano type. We address how pre-eruptive country rock is disrupted and redistributed to form the diatreme structure during an eruption by establishing the internal architecture of the unbedded diatreme at Standing Rocks West, and the volumes and sources of wall-rock within the diatreme and in a complementary tephra ring at Teshim, both in the Hopi Buttes volcanic field. The unbedded diatreme is dominantly exposed as a massif comprising multiple subvertical columns, interpreted as ephemeral-conduit deposits of well-mixed, poorly sorted, composite and juvenile pyroclast-rich deposits that truncate marginal layered deposits, plus a peripheral heterolithic country rock breccia. Wall-rock clasts are scarce in the massif and were mainly sourced from ~75 to 245 m depth below the paleosurface. By contrast, the tephra ring has abundant wall-rock fragments, dominantly from the upper 75 m of the pre-eruptive sedimentary sequence. The diatreme structure is interpreted to have been formed by many small-volume explosions fed by a low flux of basaltic magma. Late stage explosive activity was rooted mostly within pyroclastic debris at shallow to intermediate depths within the diatreme structure. This resulted in damped and shifting intra-diatreme explosions and jets that facilitated gradual mixing, recycling and remobilization of debris in the diatreme, with incremental addition of juvenile material and possibly a local rise in the crater floor.


Diatreme Country rock breccia Monogenetic Fissure Hopi buttes volcanic field 


Maar-diatreme volcanoes, unlike other small volcanoes, are formed by eruptions that excavate an open crater into the substrate. Most maar-diatreme-forming activity takes place below the land surface, producing a significant sub-conical sub-crater structure filled by below-ground deposits. These deposits occupy compound conduit structures, the largest found among all volcano types relative to their surface edifice. Based on common differences in geometry and internal architecture at different depths, maar-diatreme volcanoes are divided into three main volcano-structural levels: feeder dyke, diatreme structure, and tephra ring (Clement 1982; White and Ross 2011). The feeder dyke carries magma from its source toward the surface. The diatreme structure is specifically formed by fragmentation and mobilization of magma and country rock, and the primary deposits in it can be divided into those of a root zone and the diatreme (cf. White and Ross 2011). The tephra ring is the site of deposition for pyroclastic material ejected out of the maar crater onto the surrounding land surface.

Although maar-diatreme volcanoes worldwide of different magma compositions show generally very similar characteristics (White and Ross 2011), there is no consensus on what drives country rock excavation and how this volcano type evolves over the duration of an eruption to form a subterranean structure filled with well-mixed pyroclastic debris (e.g. Lorenz 1986; Sparks et al. 2006; Lorenz and Kurszlaukis 2007; Cas et al. 2008; Valentine 2012; Valentine and White 2012). It is important to better understand subvolcanic processes and the development of shallow plumbing systems in monogenetic volcanic fields, which are directly linked to surface activity including the inception of hazardous eruptions.

To build a robust data-set that would establish a rigorous framework we would ideally investigate good outcrops of a fully exposed, well-developed diatreme structure and its ejecta ring, then interpret our findings in light of observations from historical eruptions. Presently, we know of nowhere that all volcano-structural levels of a single substantial diatreme are exposed (though a small diatreme is inaccessibly exposed in full section at Suoana crater, Miyakejima Volcano, Japan, Geshi et al. 2011) and no historical maar-forming eruptions have been fully observed with their deposits comprehensively investigated. Taking an alternative approach, we report here an investigation by detailed mapping and petrographic analysis of an exemplary exposure of an unbedded diatreme remnant (White 1991), the Standing Rocks West volcano (SRW) in the Hopi Buttes Volcanic field (HBVF), Arizona (Fig. 1), and then explore the relationship between SRW diatreme rocks and a complementary tephra ring deposit (Teshim; White 1991), which is typical of many exposed in the Hopi Buttes. Both basaltic volcanoes are of similar Miocene age (ca. ~7 Ma) and cross-cut the same identifiable, flat-lying Cenozoic, Mesozoic and Paleozoic sedimentary rocks, of consistent and known thickness and depth, within the tectonically stable Colorado Plateau (Gilbert et al. 2007; Billingsley et al. 2013). This allows us to estimate the volumes of different stratigraphic units represented as wall-rock clasts within the diatreme and tephra ring. Using this information we can infer volumes of country rock disrupted and redistributed from specific depths during formation of the volcano, and how the diatreme structure evolved through the eruption.
Fig. 1

a Satellite image of Standing Rocks West diatreme (blue dashed line—diatreme outline; red line—vertical massif faces; yellow lines—10 m contours) within TRcp (Petrified Forest Member, Chinle Formation). UTM coordinates: Datum NAD83, Zone 12N. b Location of Hopi Buttes Volcanic Field (HBVF) in Arizona (AZ), USA. c Photograph of massif and flat-lying country rock strata from eruption surface to current depth of erosion (labelled after Billingsley et al. 2013). Direction of view to NNW; Ti lava; Tb Miocene–Pliocene Bidahochi Formation poorly consolidated claystone and siltstone; Jm Lower Jurassic Moenave Formation competent sandstone; TRco Upper Triassic Owl Rock Member, Chinle Formation cherty nodular, clastic limestone interbedded with calcareous claystone, siltstone and mudstone; TRcp Upper Triassic Petrified Forest Member, Chinle Formation mudstone interbedded with lenticular sandstone


The data presented in this study were acquired by detailed volcanological mapping of vertical faces at metre-scale, with systematic lithological descriptions and quantitative clast counts of the individual mapped units, and by petrographic investigations of 42 representative samples from geological units identified in the diatreme massif. Clast counts, to determine component abundance, were acquired by 29 line counts (1 to 2 per unit) for 18 lapilli tuff units (see Electronic Supplement Materials (ESM) 1–3 for further details). They are supported by multi-scale image analysis and functional stereology on six samples that provide volumetric data from 0.5 mm to 150 cm by combining information from outcrop and thin-section images (after Jutzeler et al. 2012; see ESM 4–6 for further details). Volumes of the maar-diatreme volcano-structural levels, disrupted wall-rock, and wall-rock distribution within the volcano-structural levels were estimated using simple geometry calculations for an inverted cone. Information fed into the calculations comes from measurements of the diatreme at the current erosional levels, inferences based on other published maar-diatreme dimensions, and the relative proportions of wall-rock to juvenile material determined from clast counting (ESM 7 gives details).

Standing Rocks West volcanic architecture

Diatreme remnant geometry

The unbedded diatreme remnant is exposed ~300 m below the eruptive palaeosurface in varicolored mudrocks of the upper Petrified Forest Member (TRcp) and mudrock with pebbly horizons and thin beds of freshwater marl in the lower Owl Rock Member (TRco), both within the Chinle Formation (Fig. 1c). The diatreme outline is irregularly subcircular, ~150 m wide and ~10,000 m2 in area including the N-trending arm. The diatreme at mapped level can be divided into two general deposit types: (1) country rock breccia and (2) a diatreme massif consisting dominantly of unbedded lapilli tuff, occupying 30 % and 70 % of the diatreme area, respectively (Fig. 2). Access to most of the massif is limited by steep, subvertical walls extending ~50 m above the current land surface (Figs. 1 and 2).
Fig. 2

Map of the Standing Rocks West diatreme showing the division into the massif and country rock breccia. The massif comprises two main deposit types: (1) layered lapilli tuff/tuff breccia (LT/TBl) and country rock-rich lapilli tuff/tuff breccia (CRLT/TBl); and (2) lapilli tuff/tuff breccia columns (LT/TBc; individual units distinguished by different grey shading)

Heterolithic country rock breccia

The country rock breccia is the oldest unit preserved in the diatreme structure (Fig. 2), and is cross-cut by the massif rocks and a NW–SE trending, poorly preserved, moderately to highly welded spatter dyke (cf. Lefebvre et al. 2012). The breccia is eroded to the present land surface and partly covered by scree. The breccia characteristically has a mixture of metre-sized blocks of Moenave Formation sandstone and Owl Rock Member wall-rocks sourced from shallower depths. No pyroclasts are apparent within the breccia.

Diatreme massif structure and componentry

The diatreme massif consists of two main types of pyroclastic rock: (1) subvertical lapilli tuff and tuff breccia columns that comprise most of the massif’s volume; and (2) layered lapilli tuff and tuff breccia (Fig. 2). Based on the subvertical perimeter of the diatreme massif, and visible contacts spanning ~10 m of topography, we infer that the external massif contacts are sharp, curvi-planar to slightly undulating, and subvertical to possibly upward flaring (~ 80°).

Subvertical lapilli tuff and tuff breccia columns

A minimum of 35 individual pyroclastic columns comprise the dominant infill of the massif (Fig. 2; see ESM 8–11 for further details). The columns are distinguished based on their steep, planar to irregular, and sharp to diffuse (widths up to 1 m) contacts (Fig. 3) and differences in component abundance (Fig. 4). Neighbouring units within any given part of the massif typically have similar clast-type proportions (Fig. 4a,b), including the interstitial matrix (see below), but each individual unit has its own subtle signature based on the relative abundance of wall-rock clasts and pyroclast types (Fig. 4c). Where distinct, the columns generally extend either the entire height of the massif (Fig. 3), or are rarely overlain by a different infill type (Fig. 5b). Individual columns span a few metres to 20 m wide and cover areas of up to 400 m2 in plan view. They range from irregular subcircular, tabular to sheet-like in map view and sometimes occur as thin screens (decimetres- to metres-thick) along the diatreme massif margin (Figs. 2, 3).
Fig. 3

Photographs and corresponding cross-sections of Standing Rocks West massif east and west faces. a West face. b West-face map outlining eleven subvertical lapilli tuff/tuff breccia columns (LT/TBc). c East face. d East-face map outlining six subvertical lapilli tuff/tuff breccia columns and thin remnant of intercalated pyroclast-rich lapilli tuff and country rock-rich tuff breccia (LTl/CRTBl) cross-cut by LTc columns. Note the crude steep wall-rock clast alignment. e Map of massif (Fig. 2) showing location and orientation of face-maps XX′ (west face (b)) and YY′ (east face (d))

Fig. 4

Map of massif (see Fig. 2 for legend of massif infill patterns) showing relative proportions of component types for individual mapped geological units. a Pyroclasts, wall-rock clasts and interstitial material (<4 mm) proportions determined using line method (ESM 1). A33a,b coarse and fine layered lapilli tuff, respectively; proportions estimated visually. b Pyroclasts, wall-rock clasts and interstitial material (<0.5 mm) proportions determined using image analysis and stereology method after Jutzeler et al. (2012). c Proportions of wall-rock clasts by source and pyroclast types >4 mm for lapilli tuff/tuff breccia columns determined using line method. SP simple juvenile; C-LP composite-loaded juvenile (pyroclast contains two or more dispersed juvenile pyroclasts); C-LW composite-loaded wall-rock (pyroclast contains two or more dispersed juvenile and wall-rock clasts); C-C composite-cored; C-S composite sediment-bearing; Tb Bidahochi Formation; Jm Moenave Formation; TRco Owl Rock Formation; U unknown wall-rock type

Fig. 5

a Photograph of north face of Standing Rocks West massif; close-ups of c, d and e described below. b Corresponding massif north-face map showing truncated layered pyroclast-rich lapilli tuff (LTl) and country rock-rich lapilli tuff/tuff breccia vestiges (CRLT/TBl). CRTB is country rock-rich tuff breccia; Va is a volcaniclastic dike. c Sharp, subvertical contact of CRLT/TBl (unit A33) and LT/TBl (unit A35) where unit A33 is a thin screen in front of unit A35. d Crudely layered CRLT/TBl cross-cut by weakly outward-flaring (~80°) lapilli tuff column (LTc). A33 layers are wall-rock richer (25–45 %) and coarser (average size 3–6 cm, maximum size 2 m) vs. wall-rock poorer (5–10 %) and finer (average size 1–3 cm, maximum size 60 cm). e Thin screen of crudely layered country rock-rich tuff breccia (CRTBl) and pyroclast-rich lapilli tuff (LTl). Bottom CRTBl layer is dominantly fresh, competent Moenave Formation sandstone plus Owl Rock Member clastic, cherty limestone; both clast types sourced up section. f Plastically deformed bomb at base of Moenave Formation block (white arrow) indicating downward movement

The lapilli tuff column deposits are considered typical of unbedded diatreme deposits (“lower diatreme” in White and Ross 2011) comprising inhomogeneously distributed and generally poorly sorted wall-rock clasts and pyroclasts. Individual units here are structureless to domainal, and matrix-supported to locally framework-supported. The units mostly lack fabric but some units have rare, crude and often steep alignment of elongate clasts. Decimetre- to metre-scale domains defined by differences in component abundance and grain size occur in several units, including local concentrations of blocks and bombs. Rare units contain isolated, discontinuous, millimetre- to decimetre-scale, internally massive, graded and cross-laminated layers, defined by a different grain size and/or component abundance compared to the host unit. These layered domains are often convolute and steeply dipping (~50–85°). In addition, some columns are cross-cut by narrower, discontinuous, irregular, subvertical lapilli tuff to tuff breccia domains that are distinguished by their framework-supported texture, coarser grain size, highly domainal grain size distribution, low matrix and high cement contents, better sorting, and relative wall-rock and juvenile abundance.

Layered lapilli tuff and tuff breccia

Four layered lapilli tuff units, cross-cut by subvertical lapilli tuff columns, occur as thin screens (less than a few metres thick) along the massif’s northern and eastern margins (Figs. 2 and 5). The largest preserved vestige (unit A33; ~125 m perimeter) occurs as several 10 to 50 m long remnants confined by country rock breccia or wall-rock. It is a layered, matrix-supported, closely packed, well-mixed, wall-rock-rich lapilli tuff and tuff breccia with a pink clastic matrix interpreted as Bidahochi Formation mudstone (Fig. 5d). The metre-scale layers, shallowly dipping predominantly towards the massif centre, are crudely defined by wall-rock clast abundance and size and have diffuse contacts (Figs. 4a and 5d). The wall-rock clast population differs from the other diatreme infill types in that Owl Rock and upper Petrified Forest Member clasts are relatively abundant. Rare country rock-rich beds exhibit normal grading (Fig. 5d).

Two layered, juvenile-rich, matrix-supported, closely packed lapilli tuff and tuff breccia units have similar component type and abundance to the lapilli tuff columns. Unit A35 truncates unit A33, and unit A38 overlies the lapilli tuff columns (Fig. 5a–c). The crude, discontinuous, metre-scale thick pyroclast-rich layers are marked by the abundance of blocks and bombs, and diffuse contacts. They dip at shallow angles toward the massif centre and have planar to irregular lenticular geometries (Fig. 5a–c).

By contrast, unit A14, a thin screen on the southeast massif margin (Fig. 2, 3c,d), comprises country rock-rich, matrix- to framework-supported tuff breccia, intercalated with matrix-supported, pyroclast-rich lapilli tuff (Fig. 5e). Layers are metre-scale in thickness and irregular, with lenticular geometries. The tuff breccia layers dominantly comprise blocks (~65–85 %; average size 20 cm, maximum 2 m), with each layer derived mostly from a single stratigraphic unit up section (i.e. the Moenave Formation or Owl Rock Member; Fig. 5e); there are rare flattened bombs moulded to their bases (Fig. 5f). The blocks are set in a finer-grained and juvenile-rich matrix similar to the juvenile-rich lapilli tuff layers and neighbouring lapilli tuff columns.

Wall-rock fragments

Wall-rock clast abundance within the massif is generally <10 % (>4 mm) with only rare wall-rock-rich columnar and layered units (Fig. 4a,b). Weathered outcrops at the top of the massif reveal little detail, but do show that these rocks are comparable in appearance to the juvenile-rich outer wall and southeast inner region deposits described previously, with only a sparse population of wall-rock clasts. Sources of the well-mixed wall-rock clast population of the massif, ranging from ash- to block-size, span from the eruptive surface down to ~400 m below the eruption surface (~100 m below the present exposure level of the massif) but are dominantly from up section, ~75–245 m below the eruption surface. Moenave Formation sandstone and Owl Rock Member cherty limestone are present in all units (Figs. 4c and 6a), but clasts from the underlying Petrified Forest Member mudstone are less common. Clasts from below the present exposure level, and from the Tertiary Bidahochi Formation (typically mudstone domains that contain dispersed pyroclasts) are rare or absent at all sites examined. Unaltered wall-rock clasts are predominant, but several lapilli tuff columns contain weakly to strongly discoloured, thermally altered ones (Fig. 7a).
Fig. 6

Plan view map of massif (see Fig. 2 for legend of massif infill patterns) showing distribution of: a Wall-rock clast types within the massif according to their stratigraphic country rock levels per unit determined from the relative proportions of wall-rock clast types from line method data. Inset: Schematic stratigraphy (Billingsley et al. 2013); b Medium lapilli to bomb-sized pyroclast types within the massif determined from the relative proportions of pyroclast types excluding simple juvenile from line method data and visually. C-LP composite-loaded juvenile (pyroclast contains two or more dispersed juvenile pyroclasts); C-LW composite-loaded wall-rock (pyroclast contains two or more dispersed juvenile and wall-rock clasts); C-C composite-cored; C-S composite sediment-bearing; SX single crystal

Fig. 7

a Composite-cored bomb; Moenave Formation core with thermally altered margins. b Irregular, smooth ridged surface of an elongate bomb. c Loaded-wall-rock bomb with dispersed pyroclasts and thermally altered wall-rock clasts. Inset shows close-up. d Loaded-wall-rock bomb with cauliflower surface texture

Juvenile and composite pyroclasts

All units of the massif contain juvenile and composite pyroclasts, ranging from extremely fine ash to bombs, with the majority of fine lapilli-size or smaller. The term “pyroclast” here is used for all primary volcaniclasts, formed by the fragmentation of magma by any volcanic event, including explosions that produced a crater or/and debris-jet activity, rockfall, and dyke injections. We justify this usage because the final emplacement of these primary clasts within the massif occurred from jets or fallback (White and Houghton 2006). These events took place during the single eruption inferred to have formed this volcano probably over days to years as observed for historical monogenetic eruptions (White and Ross 2011; calculated by Lorenz and Kurszlaukis 2007; other small-volume basaltic volcanoes, Valentine and Gregg 2008). We consider that deposits within a diatreme structure are primary unless demonstrably reworked by sedimentary processes, and apply the primary volcaniclastic terminology recommended by White and Houghton (2006).

Juvenile and composite pyroclasts are classified into five categories (Table 1). We can distinguish primary grains from composite grains that indicate recycling, but not juvenile ‘reused’ ones (Table 2), so “juvenile” is used in the general sense for grains newly produced from magma during the eruption, but not definitely by the specific event that emplaced them in their current positions. The dominant type is simple juvenile, but composite pyroclasts are also abundant and are mostly of the composite-loaded variety (Fig. 4c, 6b). There is wide variability in groundmass crystallinity, vesicularity, vesicle populations, included wall-rock clast types and morphology (i.e., circularity, shape, and surface texture; Figs. 78) in pyroclasts within each geological unit, even on a millimetre- to centimetre-scale. This makes it impossible to identify juvenile populations produced by single eruption events. All juvenile clasts are hypocrystalline and contain microlites, but with highly varied groundmass crystallinity ranging from microlitic sideromelane (mostly palagonitized), transitional (translucent brown), tachylite (brown, virtually opaque), and rarely holocrystalline (Fig. 8c,d). They consist of Ti-augite and less common serpentinized olivine phenocrysts set in a groundmass of acicular clinopyroxene, and euhedral, cubic Fe-Ti spinel. Pyroclasts range from non- to extremely vesicular, but most are non- to poorly vesicular (after Houghton and Wilson 1989; Fig. 8) with vesicle shapes ranging from spherical to elliptical to contorted.
Table 1

Definitions of the five types of juvenile and composite pyroclasts classified based on internal structure and componentry

Pyroclast category

Structure type



Simple, juvenile

Homogeneous, massive particle consisting of groundmass ±phenocrysts


Complex, juvenile

Heterogeneous juvenile particle with fluidal domains of different groundmass crystallinity and vesicularity; the most obvious are those of mingled microlitic sideromelane with either transitional or tachylite

Cored (cf. Fisher and Schmincke 1984)


Single proportionately large lithic clast or juvenile clast enclosed in one or more partial to complete rim(s) of juvenile material

Loaded (cf. Rosseel et al. 2006)


Proportionately small particles (lithic, juvenile) within juvenile pyroclast; pyroclast may be domainal, show internal welding



Fluidal clast with domains of interstitial matrix from the host deposit; includes fragments of peperite, i.e. from a zone of magma-host mingling on a millimetre- to centimetre scale (cf. McClintock and White 2006), but can also include clasts with sediment within vesicles or as domains between agglutinated clasts

Table 2

Recycled juvenile and composite pyroclasts

Clast name




Generic term for grains with evidence for repeated ejection of particles

Primary volcaniclasts did not escape vent or conduit at time they formed


= “recycled” of Houghton and Smith (1993); clasts have fallen into vent and been ejected again. Clasts may be abraded, broken, or coated.

Particles fell back into debris within a vent and were later entrained in a second eruptive jet or plume; cycle may repeat





A composite clast containing one (cored) or more (loaded) “recycled juvenile” clasts from the same eruption

Primary or reused pyroclasts that fell back into a vent and were incorporated into coherent magma subsequently fragmented to form new clasts (Rosseel et al. 2006 “recapitulated”; Lefebvre et al. 2012)



Fig. 8

a Polished slab comprising dominantly variably coloured simple and composite-loaded wall-rock pyroclasts mostly with sub-irregular shapes. Plane-light thin-section photographs: b Pyroclasts with overall low vesicularity and smooth to cuspate margins. c Simple (sp), composite-cored (C-C) and composite-loaded (C-L) pyroclasts with varying groundmass crystallinities (palagonitized microlitic sideromelane—honey and brown coloured; tachylite—brownish black). d Multiple-cored composite clast with a sideromelane innermost clast, transitional middle layer and tachylitic outer rim. e Cored juvenile-loaded clast (C-LP, outlined in blue) with another loaded pyroclast as a core; dispersed simple pyroclasts (sp) (clasts, outlined in yellow). f Back-scattered electron image of pyroclast shards down to submicron scales (outlined in yellow)

Interstitial material

Between framework grains are clastic matrix (grains < 0.5 mm), cement and pore space. Generally, the massif deposits include varied amounts of cement and open pore space is rare. The matrix of the massif deposits mostly consists of juvenile pyroclasts, which are locally well-resolved to sizes smaller than 30 μm (Fig. 8f) and subordinate sandstone, siltstone and mudstone clasts, detrital quartz grains, single crystals of clinopyroxene, olivine and spinel, and unidentified, opaque microcrystalline material. By contrast, rare wall-rock-rich units contain a pale pink matrix similar to clayey Bidahochi Formation or Petrified Forest Member mud, and have a higher detrital quartz content. The cement in all exposed units consists of multiple generations of carbonate, phyllosillicate, and zeolite, which vary across millimetre- to several metre-scale domains throughout the massif irrespective of infill type. In many places these cements demonstrably replaced the clastic matrix.

Teshim tephra ring as a tephra ring proxy for SRW

The Teshim tephra ring, located in the northeastern part of the HBVF, was chosen as a proxy for the now eroded SRW tephra ring because it was previously studied by White (1991), and in general seems to be representative of tephra rings within the HBVF. Unlike some maars in the field, Teshim is a relatively simple, circular maar and its tephra ring strata are completely preserved where overlain by lava from a neighbouring volcano. It comprises a low, broad apron with a maximum height of 15 m proximal to the 850 m diameter maar crater and extends ~1 km away where the distal deposits terminate. White (1991) observed that the tephra ring consists of decimetre- to metre-thick lapilli tuff deposits, both cross-laminated and unbedded and interpreted as formed by pyroclastic density currents (ESM 12; White 1991, Fig. 2, p. 241). These deposits show an overall decrease in wall-rock to juvenile clast ratio up section with the lower deposits dominant in Bidahochi Formation mud and the upper deposits containing a comparatively high abundance of Moenave Formation sandstone clasts.

Interpretation and discussion

Significance of lapilli tuff columns

Detailed mapping substantiates the observation of White (1991) that the massif consists predominantly of multiple, commonly cross-cutting, subvertical lapilli tuff columns (Figs. 2 and 3), also recognized at many other unbedded diatreme localities worldwide (e.g. several South African kimberlites, Clement 1982; Coombs Hills, Antarctica, Ross and White 2006; Maegok volcano, Korea, Kwon and Sohn 2008). The SRW columns are interpreted as superimposed ephemeral-conduit deposits that formed from many individual active “debris-jets” (dispersed pyroclasts and wall-rock clasts, gases ± liquid water droplets). The jets were sourced from multiple discrete explosion sites, and propagated as expanding cavities through pre-existing pyroclastic debris in the diatreme; the later events left the strongest imprints (as inferred elsewhere by McClintock and White 2006 and Ross et al. 2008a, b). We interpret their irregular distribution to indicate shifting fragmentation sites that were not all located along a feeder dyke whose position was confined by country rock, and instead shifted both laterally and vertically within pyroclastic debris of the diatreme from earlier phases of eruption (e.g. Houghton and Nairn 1991; Valentine 2012). Intrusion of dykes into diatreme infill is consistent with observations elsewhere of dykes that have intruded into unconsolidated pyroclastic material with irregular morphology, small offshoots, and marginal peperite (e.g. Hooten and Ort 2002; McClintock and White 2006; Befus et al. 2009).

Gradual mixing, recycling and remobilization of pyroclastic debris within the diatreme

Non-bedded deposits, grain-by-grain mixture of wall-rock clasts from several different country rock stratigraphic levels, and many generations of juvenile pyroclasts within individual lapilli tuff columns are features shared by some other diatremes (e.g. Clement 1982; White and Ross 2011). The mixing process was not a sorting one, because fines remain prominent throughout the diatreme deposits except where obscured or replaced during diagenesis. Based on the componentry data for the lapilli tuff columns (Fig. 4) and the internal architecture of the massif, mixing must have been gradual and facilitated by multiple, small-volume explosions that repeatedly transported and emplaced small volumes of pyroclastic debris vertically and laterally including debris from earlier explosions (Valentine and White 2012).

The interpretation of multiple discrete fragmentation events (e.g. Ross and White 2006), rather than deposition from the waning of an eruption that had been dominated by sustained discharge of material (e.g. Sparks et al. 2006; Gernon et al. 2008; Porritt et al. 2008) is based on the following observations in combination. (1) The unbedded diatreme margins and the inner southeast region consist of multiple, randomly distributed, subvertical columns and planar domains that are inferred to make up most of the unbedded massif, as also seen in plan view at Coombs Hills (Ross and White 2006). Column boundaries are not mappable on the weathered upper surface of the massif, but deposit component mixtures and grain size appear indistinguishable from those in identified columns elsewhere in the massif. (2) Individual lapilli tuff columnar units were emplaced at different times, recorded by cross-cutting relationships where one column is partly enclosed by another, and contacts that truncate local clast alignments. (3) Individual columnar units have small volumes relative to that of the massif. (4) The lapilli tuff columns contain common recycled material. Further support that the maar-diatreme-forming eruptions involved repeated small-volume events, typical of other small-volume volcanoes (as shown by Valentine et al. 2012), includes the multiple decimetre- to metre-thick lapilli tuff layers comprising the tephra rings within the field (White 1991; Vazquez and Ort 2006; Newkirk 2009) and observations of such activity at rare historical eruptions over days to weeks (e.g. summarised in White and Ross 2011).

Small-scale heterogeneities seen at SRW have also been documented in other unbedded diatreme deposits and interpreted to indicate incomplete homogenization of within-structure deposits during an eruption (e.g. Clement 1982; Ross and White 2006; Brown et al. 2008; Kwon and Sohn 2008). Low-angle bedding and cross-lamination must have originally formed on a depositional surface open to the sky, such as the syn-eruptive crater floor or tephra ring; thus the randomly oriented, steeply dipping beds surrounded by unbedded pyroclastic deposits are not interpreted as in situ deposits. Some isolated layers may have been incorporated during closure of debris-jet cavities (cf. Ross et al. 2008a), which may have also caused local collapse of layered diatreme deposits.

The complexity of juvenile clasts observed at SRW suggests substantial remobilization, mixing and recycling of materials within the diatreme. Here, we consider recycled grains (Table 2) to include “reused” ones that have simply dropped back into the vent debris and then moved again (“recycled” clasts of Houghton and Smith 1993), and “reborn” ones that involved capture of an earlier-formed pyroclast into coherent magma, which was then fragmented again with the old pyroclast inside it (“recapitulated” clasts of Rosseel et al. 2006).

The origins of “recycled reborn” composite grains are incompletely understood, but involve delivery of original pyroclasts back into sites where they become incorporated into coherent, subsequently fragmented, magma. Such movement and mixing of debris within the diatreme can occur: (1) at and near the explosion site when debris driven into the opening jet-cavity sediments into it; (2) stratigraphically above the explosion site, when overlying debris is pushed up from below during cavity growth then flows inward and downwards during cavity collapse to capture mixed tephra during closure of the cavity walls (Lorenz and Kurszlaukis 2007; Ross et al. 2008b). To form the composite fragments, either coherent magma of a dike or pond, or juvenile pyroclasts sufficiently hot to weld back into coherent magma must be available in the mixing zone. This debris can include collapse material derived from crater floor or maar-rim ejecta (Lorenz 1986; Houghton and Smith 1993; McClintock and White 2006; Lorenz and Kurszlaukis 2007; Valentine 2012). Along with pyroclastic debris, condensed water from deflating debris-jet cavities, or from lofted debris that falls back into the syn-eruptive crater, may also be recycled back into the diatreme; this provides accessible water not directly coupled to wall-rock aquifer transmissivity or saturation (White and McClintock 2001; Valentine and White 2012).

All massif deposits at SRW have evidence for single- or multi-stage recycling of pyroclastic debris, as follows. (1) There are common ‘loaded’ composite pyroclasts containing dispersed older pyroclasts and/or wall-rock clasts from different stratigraphic levels. (2) Composite pyroclasts are abundant, and include kernels that are themselves ‘loaded’ pyroclasts, as well as multiply rimmed varieties. (3) There are “hypabyssal” holocrystalline lithic clasts that were broken from the feeder dyke or/and from dikes that extended from it into the diatreme fill. (4) Convolute domains comprising Bidahochi Formation mud with dispersed pyroclasts are sparsely present. (5) Some deposit domains have several different pyroclast types that were mixed on a grain-to-grain scale, with varied levels of grain palagonitization. (6) Wall-rock clasts, typically millimetre- to centimetre-size, from the Moenave and upper Chinle Formations are now hosted within the massif deposits tens to ~200 m below their original stratigraphic positions.

Many of the wall-rock clasts dispersed within discrete units throughout the diatreme originated from higher stratigraphic levels and are variably (but weakly) thermally altered, indicating that they experienced diverse events during emplacement. Wall-rock failure results in deposition either onto the syn-eruptive crater floor, or within an ephemeral cavity (Kurszlaukis and Barnett 2003; Brown et al. 2009). Wall-rock clasts within the diatreme structure can subside during debris-jet activity (e.g. White and Ross 2011) or be transported upward during emplacement of debris-jet deposits (Ross and White 2006).

Evidence for low magma flux during diatreme formation

Relatively common composite pyroclasts within the juvenile-rich lapilli tuff columns comprising the SRW massif suggest low magma flux during diatreme formation because in a high-flux eruption most clasts are carried high in a plume or dispersed in lateral currents, thus becoming unavailable for any form of recycling in the vent. The paucity of wall-rock clasts (<10 %, mostly from higher stratigraphic levels) implies regions of considerable “substitution” of wall-rock material during diatreme formation, and also of repeated magma fragmentation away from the walls of the diatreme structure (Ross and White 2006; Valentine and White 2012). The fact that composite, mostly loaded, pyroclasts are common demonstrates that the repeated explosions were accompanied by significant recycling. However, determining the volumes of recycled pyroclasts versus primary juvenile pyroclasts involved in an explosion is exceptionally difficult, because different simple pyroclast generations can be chemically indistinguishable, particle reuse leaves little or no signature in lithified deposits, and textural inhomogeneities are common in conduits (Houghton and Gonnermann 2008). At other localities where differentiation is possible, the proportion of juvenile pyroclasts added during a fragmentation event appears to be low when there is recycling in the form of reuse, about 10–30 % (Houghton and Smith 1993; Lorenz et al. 2002; Ross and White 2006). Multiple small-volume eruptive bursts are also reflected by the multiple thin tephra ring layers (White 1991; Vazquez and Ort 2006), and are consistent with expected flux of magma through thin (<1 m) feeder dykes (Hooten and Ort 2002; Lorenz and Kurszlaukis 2007). Thus, it is likely that each event produced a small volume of pyroclasts, implying a low rate of magma supply, which is consistent with the general behavior of small-volume monogenetic mafic volcanoes with magma fluxes on the order of a cubic metre per second (Valentine and Perry 2006; Valentine and White 2012).

Fragmentation sites appear to have lain mainly within the middle parts of the western portion of the diatreme, below the current exposure level, during the formation of massif deposits. This is inferred because there were no feeder dykes observed within the massif, there is little wall-rock material from below the massif, and the individual lapilli tuff columns comprising the massif are interpreted as conduit remnants that extend below the current surface. The massif deposits are interpreted to have formed in the main part of the eruption represented here, after emplacement of the almost juvenile-free country rock breccia. Each eruptive burst from closely spaced sites within earlier-formed diatreme debris would have added an increment of primary juvenile pyroclast material and helped recycle older clastic material, leading to high concentrations of juvenile material in the massif. Such enrichment, particularly in recycled pyroclasts, is not a predicted result of explosion sites that progressively excavate downward into wall-rock (e.g. Lorenz and Kurszlaukis 2007).

Distribution of wall-rock debris within a maar-diatreme volcano

Surface and diatreme dimensions, and volume calculations for Standing Rocks West volcano

Dimensions and volumes of the SRW volcano-structural levels are estimated using a simplified cone-shaped geometry with steep-sided (80°) external contacts (after Hawthorne 1975) for the competent Mesozoic wall-rock and shallower contacts (60° after Sparks et al. 2006; Son et al. 2011) for the upper poorly consolidated Tertiary Bidahochi Formation, and applying the average 150 m diameter of the diatreme at the current erosional surface of ~300 m below the pre-eruptive surface (Fig. 9). The tephra ring volume was estimated using dimensions from White (1991) of the Teshim tephra ring, i.e. a height of 15 m (maximum height of the Teshim tephra ring) and a very shallow slope (2°) and the angle of repose (33° after Lorenz 2003) for the tephra ring dipping away from and towards the crater, respectively (Fig. 9). We estimate that at the end of the eruption SRW would have had: (1) a crater ~300 m in diameter at the paleosurface and ~75 m deep (similar to the Bidahochi Formation thickness and a crater height/width ratio of 1:5 as shown by Büchel and Lorenz 1993; White and Ross 2011); (2) a diatreme structure that extended downwards ~700 m from the paleosurface to the feeder dike; (3) a tephra ring diameter of ~1,100 m, not including the thinner distal fall deposits. Volumes of ~1 × 107 m3 and ~8 × 106 m3 are calculated for the diatreme structure and the tephra ring, respectively (Fig. 9; Table 3). These dimensions and related volumes reveal that SRW was a relatively small maar-diatreme, within the lower range of maars in the HBVF that have average crater diameters ranging from 350 to 2,000 m and inferred depths from 40 to >120 m, respectively, tephra ring radial lengths of 1,000 to 2,300 m for the larger maars, and heights of 5 to 15 m (White 1991; Vazquez 1998; Newkirk 2009). Dimensions are comparable to those of maar-diatremes elsewhere, with average crater diameters and depths ranging from 100 m to >2,000 m and 40 to 300 m, respectively (Lorenz 2003; White and Ross 2011). The inferred SRW crater depth to diameter ratio is similar to that of Quaternary maars (White and Ross 2011), which have tephra ring heights generally <30 m (White and Ross 2011), crater diameters up to 5,000 m (Lorenz 2006), and diatremes extending to >2 km depth (Clement 1982; Brown and Valentine 2013). In addition, SRW volumes are reasonable when compared to other maar-diatreme localities with volumes of ejected tephra deposits of 106 m3 for a similar-sized maar (see Sottili et al. 2012 for comparison) and ~107 to 108 m3 for larger maars (Sottili et al. 2012; Valentine 2012), and diatreme volumes that typically range from 106 to 108 m3 (Brown and Valentine 2013). The SRW volumes also coincide with those of other small-volume volcanoes, which have volumes of 105 to 109 m3 (White and Ross 2011). The calculated SRW volumes seem reasonable, but are inevitably a strong function of the contact angle assigned for the diatreme structure walls, the inferred depth of the crater below the paleosurface, and the maximum height, diameter and dip angles of the unpreserved tephra ring.
Fig. 9

a Schematic of dimensions used for maar-diatreme volume calculations. Note the Petrified Forest Member, Shinarump Member, Moenkopi Formation, and Palaeozoic sediments have been grouped together. b Plot of the volume of individual wall-rock units distributed within the total disrupted diatreme structure, diatreme (massif and country rock breccia), tephra ring and not accounted for

Table 3

Calculated volumes of the Standing Rocks West maar-diatreme and distribution of disrupted wall-rock in the volcano-structural levels based on the dimensions in Fig.  9 (ESM 7 gives details of calculations and assumptions)


Volume(×105  m3)

Volume of maar-diatreme volcano-structural levels

  Diatreme structure (total volume of disrupted wall-rock)






  Country rock breccia


  Tephra ring




Volume of disrupted individual wall-rock units of the maar-diatreme structure

  Bidahochi Fm


  Moenave Fm


  Owl Rock Mbr


  Petrified Forest and Shinarump Mbr


  Moenkopi Fm


  Palaeozoic sediments


Volume of wall-rock within the maar-diatreme pyroclastic deposits:





  Country rock breccia


  Tephra ring




Wall-rock volume budget

  Total wall-rock accounted for


  Total wall-rock not accounted for (distal deposits)


Magma volume budget



  Tephra ring


  Downwind dispersal




The total volume of wall-rock disrupted by formation of the diatreme structure should equal the sum of wall-rock still in the diatreme plus that in the tephra ring and distal fall deposits just after the eruption; the crater is empty by definition. We calculate (Table 3) that ~70 % of the total disrupted wall-rock volume was ejected from the diatreme structure with ~45 % in the tephra ring and ~30 % returned to or remaining within the diatreme; 25 % is unaccounted for and interpreted to have been dispersed downwind in unmapped and probably unpreserved fall deposits. Distal fall deposits from historical maar-diatreme eruptions extend kilometres to tens of kilometres from their sources (e.g. Houghton and White 2000; Lorenz and Kurszlaukis 2007); a 40 km × 10 km fall deposit with volume 31 × 105 m3 would average less than 1 cm thick and be easily overlooked even in young deposits.

Distribution in the diatreme and tephra ring of fragments from different wall-rock strata

Estimated volumes of fragments from individual wall-rock units in the tephra ring and diatreme are given in Table 4. The inferred downward tapering of the diatreme structure implies that larger volumes of shallow rocks would have been disrupted, compared to deeper country rocks. This is partly why lithic clasts in the Teshim tephra ring are dominantly from the Bidahochi Formation, followed by the Moenave Formation; but, we also infer that these clasts were ejected by relatively shallow explosions, mostly in the upper ~175 m. The lower to middle tephra ring layers typically have abundant Bidahochi Formation material, giving way up section to more abundant Moenave clasts and juvenile material and White (1991) inferred from this that the earliest ejecta layers were sourced less than ~75 m below the surface, with subsequent explosion sites further down. Although tephra ring deposits have varied juvenile/wall-rock ratios, many are dominated by wall-rock clasts from shallow depths, and exhibit stratigraphic inversions of the shallow country rock as observed at other Hopi Butte localities (Vazquez and Ort 2006; Ort, pers. comm. 2012) and elsewhere (e.g. Büchel and Lorenz 1993; Pardo et al. 2008; Valentine 2012). In other words, earlier explosions are commonly shallow enough to eject large volumes of upper level disrupted wall-rock onto the surface (e.g. White and Ross 2011). This interpretation gains support from recent cratering experiments that show only shallow explosions are able to eject material onto the surface beyond the crater; deeper explosions and those subsequent to crater formation result in most ejected material falling back into the crater due to a weaker and vertically focused eruptive jet (Valentine et al. 2012; Taddeucci et al. 2013; Ross et al. 2013).
Table 4

Comparison of calculated disrupted wall-rock volumes with volumes of wall-rock in diatreme

Wall unitsa

Volume of disrupted wall-rock (×105 m3)

Volume of disrupted wall-rock units in the tephra ring (× 105 m3)

Volume of disrupted wall-rock units in the country rock breccia (×105 m3)

Volume of disrupted wall-rock units in the diatreme massif (×105 m3)

Volume of disrupted wall-rock units not accounted for (×105 m3)

Bidahochi Fm






Moenave Fm






Owl Rock Mbr






Wall-rock units below the Owl Rock Mbr












aThe volumes of rock from individual wall-rock units now in the diatreme and tephra ring were estimated using the relative proportions; ~1:8:4:1 (Bidahochi/Moenave/Owl Rock/Petrified Forest and units below; from visual estimates) and ~32:14:3:1 (Bidahochi/Moenave/Owl Rock/Petrified Forest and units below; calculated from White 1991), respectively. Note that White’s (1991) counts are for matrix (thin section) only, and hence under-represent Moenave that is locally abundant as large lapilli and blocks

Clasts in both the SRW massif and in the country rock breccia are mostly derived from parts of the country rock stratigraphy that originated above the present erosional surface (depth interval ~75–245 m), but from deeper levels than observed in the Teshim tephra ring (<75 m). Most massif and country rock breccia clasts are from the Moenave Formation (54 %) and the Owl Rock Member (31 %). The wall-rock-fragment population of the massif differs from that of the country rock breccia in containing both rare clasts from the Bidahochi Formation, and from strata originating below ~245 m. About 40 % of the disrupted Moenave Formation remained in the diatreme as wall-rock collapse deposits (country rock breccia), with less than 5 % incorporated into the pyroclast-rich massif. In addition, most Owl Rock Member debris remained within the diatreme at or below its original stratigraphic level. Some wall-rock material from depth may have been transported upwards beyond current exposure levels (Fig. 9) and deposited within the diatreme structure, but those deposits are not preserved, leaving the material unaccounted for.

Both the SRW diatreme massif, and the Teshim tephra ring contain mostly wall-rock material that originated from above Standing Rocks West’s present exposure level, with little from the deeper strata. This indicates that the majority of the wall-rock within the diatreme arrived at its present location by moving downward, and supports the interpretation that space generated at shallow depths facilitates wall-rock collapse. The scarcity of Bidahochi Formation within the diatreme suggests most was ejected early on and remained at or near the surface, though some could be ‘hidden’ as clay, sand and silt grains in clastic matrix of different SRW units.

The downward tapering diatreme shape does not, on its own, explain the observed paucity of disrupted wall-rock from below. Wall-rock clasts in the tephra ring and diatreme at the level of observation (~300 m depth) that were sourced from below ~245 m represent only ~10 % by volume of the diatreme structure below that depth. If conventional diatreme geometries are assumed, wall-rock was disrupted to depths below ~700 m (Fig. 9; Table 3), so about 400 m of deep stratigraphy appears under-represented. Possible explanations are that the diatreme structure rapidly narrows below the level of SRW exposure, which seems unlikely, or that material from depth was rarely transported significantly upwards. The latter is likely as large volumes from these depths are unknown to the authors at higher elevations elsewhere in the field. This implies that most of the diatreme below the current exposure level is dominated by locally derived country rock-rich breccia. If this is the case, there must have been several main sites of fragmentation with one, ejecting material for the tephra ring, lying above current exposure levels, a second producing pyroclast-rich material, lying somewhat below the current level and forming the massif, and another one deeper, generally below 300 m, breaking and mixing material that only occasionally reached shallower levels. Explosions generated below 300 m contributed to growth of the diatreme structure, but provided only scarce debris to higher levels and did not involve large volumes of magma.

The difference in wall-rock proportions between the diatreme and tephra ring, and the high proportion of pyroclasts in the diatreme massif, indicates that the tephra ring does not provide a full record of a maar-diatreme’s eruptive history. The tephra ring was built largely by early, shallow explosions, perhaps later added to by deeper ones with much higher energies (e.g. Ross et al. 2008b; Valentine and White 2012). By contrast, the diatreme massif is inferred to reflect later, deeper stages of ‘eruption’, when explosive energy was typically insufficient to eject material from depth onto the palaeosurface. The country rock breccia predates the massif, and comprises material almost entirely dropped down from higher levels. Taken as a whole, we infer that most products of SRW’s ‘eruptive’ activity remained within the diatreme structure. The overall estimated magma volume is low (~7 × 106 m3; Table 3), as at other small-volume volcanoes (e.g. Brown et al. 2012; Mattsson 2012; Valentine 2012).

Maar-diatreme emplacement

During the early stages of eruption, explosions were probably at shallow- to optimal-scaled depths (Valentine et al. 2012; Ross et al. 2013); shallow (<200 m) and strong enough to incrementally excavate the country rock, lower the syn-eruptive crater floor levels and deposit most of the tephra ring (Fig. 10a–d). Subsequent explosions are likely to have taken place at greater than optimum scaled depths within diatreme fill, and hence failed to eject material to the tephra ring despite expenditure of large amounts of energy to form the deep diatreme structure with wall-rock breccia infill. The diatreme massif deposits mostly reflect late stage intra-diatreme fragmentation at large scaled depths within the diatreme structure (Fig. 10e–g). Instead of a downward progression of explosion sites, fragmentation became confined at shallow to intermediate levels of the diatreme structure within loose debris. This facilitated local lateral and vertical shifting of the fragmentation sites marked by the random distribution of the lapilli tuff columns and their superimposed relationships, cross-cutting layered vestiges, and the common domainal nature of the columns due to incomplete homogenization and reincorporation of layered deposits into debris-jet activity. The prolonged explosive activity took place at deep-scaled depths within the diatreme fill, and probably enclosed by substantial crater walls, allowing little or no ejecta to escape the structure (Taddeucci et al. 2013). Recycling of debris captured within the diatreme structure, together with repeated addition of new juvenile material, caused a gradual increase in juvenile pyroclast abundance and decrease in wall-rock concentration with each subsequent explosion. In addition, the multiple fragmentation events, each affecting a small part of the diatreme structure, yielded abundant reused and reborn material. These events are un- or under-represented in the tephra ring, but may have caused a local rise of the crater floor. Minimal disrupted wall-rock from deeper than 200 m was deposited onto the paloeosurface, in part because the largest volume of wall-rock disrupted was at shallow levels (within 200 m below surface); most explosions were too deep and/or weak to transport material to intermediate diatreme levels, let alone eject material onto the tephra ring. The result was segregation of deposits accompanying fragmentation activity at different depths.
Fig. 10

Schematic of Standing Rock West diatreme emplacement sequence. Diagrams (eg) illustrate only below 75 m. a Shallow emplacement of narrow dyke propagating towards the surface by hydraulic fracturing causing localized mechanical weakening of competent wall-rock (e.g. Delaney and Pollard 1981), and local peperite and dyke bifurcation within the Bidahochi Formation (Hooten and Ort 2002). b Initial shallow cratering, ejection of Bidahochi Formation mud dispersed with juvenile material onto palaeosurface, and formation of tephra ring and broad maar crater. c Early stage of maar-diatreme formation; multiple, shallow, strong explosive events ejecting disrupted wall-rock and juvenile material onto the palaeosurface forming tephra ring (White 1991). d Further crater excavation and progressive deepening of the fragmentation site, widening of crater, and sporadic collapse of locally weakened maar collar (e.g. Kienle et al. 1980; Lorenz and Kurszlaukis 2007). In addition, formation of thin, distal deposits and wall-rock-rich pyroclastic debris locally intercalated with collapse deposits within the diatreme structure (White and Ross 2011). e Debris accumulation within the diatreme structure, eruptions are increasingly confined and continued gravitational collapse of brecciated country rock walls. Disrupted diatreme debris by subsequent shallow explosion and/or incorporation into an active debris-jet resulting in gradual mixing and recycling of diatreme material (Valentine and White 2012). fg Explosive activity mainly focused within shallow to intermediate diatreme levels resulted in incremental build-up of the crater floor. Magma intruding the relatively thick, anisotropic unconsolidated diatreme debris continued to be unrestrained facilitating several migratory and variably energetic intra-diatreme fragmentation episodes (Valentine and White 2012). Formation of an irregular crater floor due to on-going wall-rock collapse along diatreme margins and erupting debris-jets that punctured the upper diatreme deposits and formed local vents on the crater floor (e.g. Houghton and Nairn 1991; Calvari et al. 2005) that deposited new pyroclastic material. In addition, debris-jets that reached relatively shallow levels also disrupted the bedded diatreme deposits during jet collapse and closure


The well-exposed Standing Rocks West (SRW) diatreme, together with the tephra ring proxy (Teshim) offer valuable insight into the emplacement of well-developed, unbedded, juvenile-rich diatremes deposits. SRW was formed by multiple, small explosive events, mostly confined to the diatreme structure, fed by a low flux of magma. There was no direct link between the volcano-structural “levels” (i.e. root zone, diatreme and tephra ring) and the sites of fragmentation, transportation and deposition. In other words, the root zone, as the connection between the feeder dyke and diatreme, was not the main and only site of fragmentation, the diatreme did not solely act as a conduit and collapse structure, and the tephra ring is not the repository of deposits from all explosive events.



We gratefully acknowledge M. Jutzeler for his guidance and use of his image analysis Excel spreadsheets, and A. Proussevitch for processing the data using his statistical functional stereology software. We thank B. Lefebvre for invaluable field assistance, P-S Ross, R. Brown and S. Moss for their thoughtful comments on an earlier version of the manuscript, and G. Valentine, B. Brand and an anonymous reviewer for constructive feedback on this version. This work was supported by a Society of Economic Geologists Student Grant, Sigma Delta Epsilon Graduate Women in Science Fellowship, Commonwealth PhD Scholarship to N. Lefebvre and GNS Science funding to J.D.L. White. Field work on the Navajo Nation was conducted under a permit from the Navajo Nation Minerals Department. Any persons wishing to conduct geological investigations on the Navajo Nation must first apply for, and receive, a permit from the Navajo Nation Minerals Department, P.O. Box 1910, Window Rock, Arizona 86515, USA telephone: 01 (928) 871–6587.

Supplementary material

445_2013_739_MOESM1_ESM.pdf (11.2 mb)
ESM 1(PDF 11496 kb)


  1. Befus KS, Hanson RE, Miggins DP, Breyer JA, Busbey AB (2009) Nonexplosive and explosive magma/wet-sediment interaction during emplacement of Eocene intrusions into cretaceous to Eocene strata, Trans-Pecos igneous province, West Texas. J Volcanol Geotherm Res 181(3–4):155–172CrossRefGoogle Scholar
  2. Billingsley GH, Block D, Hiza-Redsteer M (2013) Geologic map of the Winslow 30′ × 60′ quadrangle, Coconino and Navajo Counties, northern Arizona. U.S. Geol Surv Sci Invest Map 3247, pamphlet 25 pp, 3 sheets, scale 1:50,000, GIS data, http://pubs.usgs.gov/sim/3247/
  3. Brown R, Valentine G (2013) Physical characteristics of kimberlite and basaltic intraplate volcanism, and implications of a biased kimberlite record. Geol Soc Am Bull. doi:10.1130/B30749.1
  4. Brown RJ, Gernon T, Stiefenhofer J, Field M (2008) Geological constraints on the eruption of the Jwaneng centre kimberlite pipe, Botswana. J Volcanol Geotherm Res 174(1–3):195–208CrossRefGoogle Scholar
  5. Brown R, Tait M, Field M, Sparks R (2009) Geology of a complex kimberlite pipe (K2 pipe, Venetia mine, South Africa): insights into conduit processes during explosive ultrabasic eruptions. Bull Volcanol 71(1):95–112CrossRefGoogle Scholar
  6. Brown R, Manya S, Buisman I, Fontana G, Field M, Niocaill C, Sparks R, Stuart F (2012) Eruption of kimberlite magmas: physical volcanology, geomorphology and age of the youngest kimberlitic volcanoes known on earth (the Upper Pleistocene/Holocene Igwisi Hills volcanoes, Tanzania). Bull Volcanol 74(7):1621–1643CrossRefGoogle Scholar
  7. Büchel G, Lorenz V (1993) Syn- and post-eruptive mechanism of the Alaskan Ukinrek Maars in 1977. In: Negendank J, Zolitschka B (eds) Paleolimnology of European maar lakes. Springer, Berlin, pp 15–60CrossRefGoogle Scholar
  8. Calvari S, Spampinato L, Lodato L, Harris AJL, Patrick MR, Dehn J, Burton MR, Andronico D (2005) Chronology and complex volcanic processes during the 2002–2003 flank eruption at Stromboli volcano (Italy) reconstructed from direct observations and surveys with a handheld thermal camera. J Geophys Res 110(B2), B02201CrossRefGoogle Scholar
  9. Cas RAF, Hayman P, Pittari A, Porritt L (2008) Some major problems with existing models and terminology associated with kimberlite pipes from a volcanological perspective, and some suggestions. J Volcanol Geotherm Res 174(1–3):209–225CrossRefGoogle Scholar
  10. Clement CR (1982) A comparative geological study of some major kimberlite pipes in northern Cape and Orange Free State. PhD Dissertation, University of Cape Town, 431 ppGoogle Scholar
  11. Delaney PT, Pollard DD (1981) Deformation of host rocks and flow of magma during growth of minette dikes and breccia-bearing intrusions near Ship Rock, New Mexico. US Geol Surv Prof Paper 1202:1–61Google Scholar
  12. Fisher RV, Schmincke HU (1984) Pyroclastic rocks. Springer-Verlag, BerlinGoogle Scholar
  13. Gernon TM, Gilbertson MA, Sparks RSJ, Field M (2008) Gas-fluidisation in an experimental tapered bed: Insights into processes in diverging volcanic conduits. J Volcanol Geotherm Res 174(1–3):49–56CrossRefGoogle Scholar
  14. Geshi N, Nemeth K, Oikawa T (2011) Growth of phreatomagmatic explosion craters: a model inferred from Suoana crater in Miyakejima Volcano, Japan. J Volcanol Geotherm Res 201:30–38CrossRefGoogle Scholar
  15. Gilbert H, Velasco AA, Zandt G (2007) Preservation of Proterozoic terrane boundaries within the Colorado Plateau and implications for its tectonic evolution. Earth Planet Sci Lett 258:237–248Google Scholar
  16. Hawthorne JB (1975) Model of a kimberlite pipe. Phys Chem Earth 9:1–15CrossRefGoogle Scholar
  17. Hooten JA, Ort MH (2002) Peperite as a record of early-stage phreatomagmatic fragmentation processes: an example from the Hopi Buttes volcanic field, Navajo Nation, Arizona, USA. J Volcanol Geotherm Res 114(1–2):95–106CrossRefGoogle Scholar
  18. Houghton BF, Gonnermann HM (2008) Basaltic explosive volcanism: constraints from deposits and models. Chemie Erde - Geochem 68(2):117–140Google Scholar
  19. Houghton BF, Nairn IA (1991) The 1976–1982 Strombolian and phreatomagmatic eruptions of White Island, New Zealand: eruptive and depositional mechanisms at a ‘wet’ volcano. Bull Volcanol 54(1):25–49CrossRefGoogle Scholar
  20. Houghton BF, Smith RT (1993) Recycling of magmatic clasts during explosive eruptions: estimating the true juvenile content of phreatomagmatic volcanic deposits. Bull Volcanol 55(6):414–420CrossRefGoogle Scholar
  21. Houghton BF, White JDL (2000) Surtseyan and related phreatomagmatic eruptions. In: Hazel Rymer JS, McNutt S, Sigurdsson H, Houghton B (eds) Encyclopedia of Volcanoes. Academic, San Diego, CA, pp 495–512Google Scholar
  22. Houghton BF, Wilson CJN (1989) A vesicularity index for pyroclastic deposits. Bull Volcanol 51(6):451–462CrossRefGoogle Scholar
  23. Jutzeler M, Proussevitch AA, Allen SR (2012) Grain-size distribution of volcaniclastic rocks 1: a new technique based on functional stereology. J Volcanol Geotherm Res 239–240:1–11CrossRefGoogle Scholar
  24. Kienle J, Kyle PR, Self S, Motyka RJ, Lorenz V (1980) Ukinrek Maars, Alaska, I. April 1977 eruption sequence, petrology and tectonic setting. J Volcanol Geotherm Res 7(1–2):11–37CrossRefGoogle Scholar
  25. Kurszlaukis S, Barnett WP (2003) Volcanological and structural aspects of the Venetia kimberlite cluster—a case study of South African kimberlite maar-diatreme volcanoes. S Afr J Geol 106(2–3):165–192CrossRefGoogle Scholar
  26. Kwon C, Sohn Y (2008) Tephra-filled volcanic neck (diatreme) of a mafic tuff ring at Maegok, Miocene Eoil Basin, SE Korea. Geosci J 12(4):317–329CrossRefGoogle Scholar
  27. Lefebvre NS, White JDL, Kjarsgaard BA (2012) Spatter-dike reveals subterranean magma diversions: Consequences for small multivent basaltic eruptions. Geology 40(5):423–426CrossRefGoogle Scholar
  28. Lorenz V (1986) On the growth of maars and diatremes and its relevance to the formation of tuff rings. Bull Volcanol 48(5):265–274CrossRefGoogle Scholar
  29. Lorenz V (2003) Maar-diatreme volcanoes, their formation, and their setting in hard-rock or soft-rock environments. Geolines 15:63–74Google Scholar
  30. Lorenz V (2006) Syn- and posteruptive hazards of maar-diatreme volcanoes. J Volcanol Geotherm Res 159(1–3):285–312Google Scholar
  31. Lorenz V, Kurszlaukis S (2007) Root zone processes in the phreatomagmatic pipe emplacement model and consequences for the evolution of maar-diatreme volcanoes. J Volcanol Geotherm Res 159(1–3):4–32CrossRefGoogle Scholar
  32. Lorenz V, Zimanowski B, Buettner R (2002) On the formation of deep-seated subterranean peperite-like magma-sediment mixtures. J Volcanol Geotherm Res 114(1–2):107–118CrossRefGoogle Scholar
  33. Mattsson HB (2012) Rapid magma ascent and short eruption durations in the Lake Natron–Engaruka monogenetic volcanic field (Tanzania): a case study of the olivine melilititic Pello Hill scoria cone. J Volcanol Geotherm Res 247–248:16–25CrossRefGoogle Scholar
  34. McClintock M, White J (2006) Large phreatomagmatic vent complex at Coombs Hills, Antarctica: Wet, explosive initiation of flood basalt volcanism in the Ferrar-Karoo LIP. Bull Volcanol 68(3):215–239CrossRefGoogle Scholar
  35. Newkirk TT (2009) Anisotropy of magnetic susceptibility of phreatomagmatic surge deposits, Hopi Buttes, Navajo Nation, Arizona, USA. Unpub MS thesis, Northern Arizona University, 89 ppGoogle Scholar
  36. Pardo N, Avellán DR, Macías JL, Scolamacchia T, Rodríguez D (2008) The 1245 yr BP Asososca maar: new advances on recent volcanic stratigraphy of Managua (Nicaragua) and hazard implications. J Volcanol Geotherm Res 176(4):493–512CrossRefGoogle Scholar
  37. Porritt LA, Cas RAF, Crawford BB (2008) In-vent column collapse as an alternative model for massive volcaniclastic kimberlite emplacement: an example from the fox kimberlite, Ekati diamond mine, NWT, Canada. J Volcanol Geotherm Res 174(1–3):90–102Google Scholar
  38. Ross P-S, White JDL (2006) Debris jets in continental phreatomagmatic volcanoes: a field study of their subterranean deposits in the Coombs Hills vent complex, Antarctica. J Volcanol Geotherm Res 149(1–2):62–84CrossRefGoogle Scholar
  39. Ross P-S, White J, Zimanowski B, Büttner R (2008a) Rapid injection of particles and gas into non-fluidized granular material, and some volcanological implications. Bull Volcanol 70(10):1151–1168CrossRefGoogle Scholar
  40. Ross P-S, White JDL, Zimanowski B, Büttner R (2008b) Multiphase flow above explosion sites in debris-filled volcanic vents: insights from analogue experiments. J Volcanol Geotherm Res 178(1):104–112CrossRefGoogle Scholar
  41. Ross P-S, White JDL, Valentine GA, Taddeucci J, Sonder I, Andrews RG (2013) Experimental birth of a maar-diatreme volcano. J Volcanol Geotherm Res. doi:10.1016/j.jvolgeores.2013.05.005 Google Scholar
  42. Rosseel J-B, White JDL, Houghton BF (2006) Complex bombs of phreatomagmatic eruptions: role of agglomeration and welding in vents of the 1886 Rotomahana eruption, Tarawera, New Zealand. J Geophys Res 111(B12), B12205CrossRefGoogle Scholar
  43. Son M, Kim JS, Jung S, Ki JS, Kim M-C, Sohn YK (2011) Tectonically controlled vent migration during maar-diatreme formation: an example from a Miocene half-graben basin in SE Korea. J Volcanol Geotherm Res 223–224:29–46Google Scholar
  44. Sottili G, Palladino D, Gaeta M, Masotta M (2012) Origins and energetics of maar volcanoes: examples from the ultrapotassic Sabatini Volcanic District (Roman Province, Central Italy). Bull Volcanol 74(1):163–186CrossRefGoogle Scholar
  45. Sparks RSJ, Baker L, Brown RJ, Field M, Schumacher J, Stripp G, Walters A (2006) Dynamical constraints on kimberlite volcanism. J Volcanol Geotherm Res 155(1–2):18–48CrossRefGoogle Scholar
  46. Taddeucci J, Valentine GA, Sonder I, White JDL, Ross P-S, Scarlato P (2013) The effect of pre-existing craters on the initial development of explosive volcanic eruptions: an experimental investigation. Geophys Res Lett 40(3):507–510CrossRefGoogle Scholar
  47. Valentine GA (2012) Shallow plumbing systems for small-volume basaltic volcanoes, 2: evidence from crustal xenoliths at scoria cones and maars. J Volcanol Geotherm Res 223–224:47–63CrossRefGoogle Scholar
  48. Valentine GA, Gregg TKP (2008) Continental basaltic volcanoes—processes and problems. J Volcanol Geotherm Res 177(4):857–873CrossRefGoogle Scholar
  49. Valentine GA, Perry FV (2006) Decreasing magmatic footprints of individual volcanoes in a waning basaltic field. Geophys Res Lett 33(14), L14305CrossRefGoogle Scholar
  50. Valentine GA, White JDL (2012) Revised conceptual model for maar-diatremes: subsurface processes, energetics, and eruptive products. Geology 40:1111–1114CrossRefGoogle Scholar
  51. Valentine GA, White JD, Ross P-S, Amin J, Taddeucci J, Sonder I, Johnson PJ (2012) Experimental craters formed by single and multiple buried explosions and implications for volcanic craters with emphasis on maars. Geophys Res Lett 39(20)Google Scholar
  52. Vazquez JA (1998) Maar volcanism in the Wood Chop Mesa area, Hopi Buttes Volcanic field, Navajo Nation, Arizona. Unpub MS thesis, Northern Arizona University, 209 ppGoogle Scholar
  53. Vazquez JA, Ort MH (2006) Facies variation of eruption units produced by the passage of single pyroclastic surge currents, Hopi Buttes volcanic field, USA. J Volcanol Geotherm Res 154(3–4):222–236CrossRefGoogle Scholar
  54. White JDL (1991) Maar-diatreme phreatomagmatism at Hopi Buttes, Navajo Nation (Arizona), USA. Bull Volcanol 53(4):239–258CrossRefGoogle Scholar
  55. White JDL, Houghton BF (2006) Primary volcaniclastic rocks. Geology 34(8):677–680CrossRefGoogle Scholar
  56. White JDL, McClintock MK (2001) Immense vent complex marks flood-basalt eruption in a wet, failed rift: Coombs Hills, Antarctica. Geology 29(10):935–938CrossRefGoogle Scholar
  57. White JDL, Ross PS (2011) Maar-diatreme volcanoes: a review. J Volcanol Geotherm Res 201(1–4):1–29CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • N. S. Lefebvre
    • 1
  • J. D. L. White
    • 1
  • B. A. Kjarsgaard
    • 2
  1. 1.Department of GeologyUniversity of OtagoDunedinNew Zealand
  2. 2.Geological Survey of CanadaOttawaCanada

Personalised recommendations