April 2016 Issue Table of Contents
Silicic volcanic systems provide timed snapshots at the Earth’s surface of the magmatic processes that also build complementary plutons in the crust. Links between these two realms are considered here using three Quaternary (<2.6 Ma) examples from New Zealand and the USA. In these systems, magmatic processes can be timed and the changes in magmatic conditions can be followed through the sequence of quenched volcanic eruption products. Before an eruption, magma accumulation processes can occur on timescales as short as decades, and whole magma systems can be rebuilt in millennia. Silicic volcanic processes, in general, act on timescales that are too rapid to be effectively measured in the exposed plutonic record.
Keywords: Caldera, rhyolite, magma chamber, Quaternary, silicic volcanism
The timescales and processes associated with silicic volcanic rocks are often compared and contrasted with the plutonic record and can be inferred from many types of information, from field-focussed studies to theoretical models. These sources of information can, however, deliver quite different messages, depending on what kind of evidence is used. Here, we consider what field-focussed studies tell us of the nature and behaviour of silicic volcanic systems. The simplest method of tracing volcanic timescales and processes is to date and study the products of successive eruptions from a particular volcano. In such cases, dating the eruptions and determining the compositions of the products help track the changing nature of the hidden subsurface magmatic system, including where magma is generated and where melt accumulation occurs.
Even given a timed sequence of eruptions and their deposits, however, linking and comparing volcanic records to the inferred complementary plutonic record is not simple. From the start, we draw attention to two aspects. First, most fine-scale knowledge of volcanic processes comes from studies of Quaternary rocks (produced during the last 2.6 My), whereas nearly all accessible in situ plutonic materials are typically tens of millions of years old, or older. The resulting contrasts in the precision with which volcanic and plutonic processes can be timed are one aspect of why contrasts in interpretation have arisen. Second, a key aspect of silicic volcanic rocks is that, with few exceptions, their isotopic compositions (particularly Sr, Pb, Nd and O) and trace element data indicate that they are the end product of large amounts of crystal fractionation and also, to a greater or lesser extent, the product of interaction with evolved (i.e. fractionated) crustal materials through melting and assimilation. For rhyolites, such as the early erupted parts of the Bishop Tuff (Rb/Sr ratios >100), there needs to have been a substantially greater volume of crystals grown and lodged at depth to form a pluton than the volume of the magma that was erupted at the surface. This concept is the basis behind the mush model for silicic magmatism, whereby substantial plutons are considered to be the required counterpart to silicic volcanism (Bachmann and Bergantz 2004; Hildreth 2004).
Here, we summarise evidence from three Quaternary silicic volcanic systems that bears on the timescales and processes leading to the generation of silicic magmas and their release in volcanic events. We consider how the volcanic record can illuminate aspects of the large-scale magmatic processes associated with modern batholith-scale pluton growth. We can only draw the reader’s attention to a fraction of the relevant papers, and have, thus, mostly focussed on recent publications as an entry point to the literature.
DATING THE VOLCANIC RECORD
There are two complementary approaches with which to measure the tempo of Quaternary (and older) volcanic records and their parental magmatic systems.
First, date the volcanic events and then use the changing eruptive compositions to follow the detailed evolution of the magma system through time. Young sequences (<40–50 ka) are generally dated through radiocarbon methods; deposits with overlapping and older ages are typically dated by 40Ar/39Ar techniques. This approach works well in areas like the Taupo Volcanic Zone (TVZ; New Zealand), or Long Valley (California, USA), where tens or more eruptions are available with which to follow magmatic processes. It is less easy to apply at localities like Yellowstone (western USA) where eruptions are widely spaced in time, or if the eruption of interest happens to be the first one in a sequence.
Second, date the minerals in eruption products to look at the evolution of the magma system as captured in a single eruption. All methods exploit as internal clocks the radioactive decay of certain isotopes, the ratios of which are considered to be unaffected by residence at high temperatures in the magma system. Early approaches, specifically in very evolved silicic systems, exploited the Rb–Sr system in feldspars. More recently, however, most dating of magmatic systems has been done through the analysis of U- and Th-bearing minerals (particularly zircon). We outline below the methods and results from zircon dating.
Zircon as a dating tool has been used for many decades (Hanchar and Hoskin 2003). All methods revolve around one or more of the decay chains of 238U (to 206Pb), 235U (to 207Pb) and 232Th (to 208Pb), and exploit the fact that zircon incorporates significant amounts of U and Th and effectively excludes Pb from its structure as it crystallises. Thus, ingrowth of ppb levels of radiogenic Pb can be measured in the crystals without it being swamped by the tens of ppm levels of ‘common’ Pb that are typically present in the host melt. For the Quaternary examples considered here, the most used tool is the 238U–206Pb system, in part because of the relative abundance of 238U and in part because of a characteristic of its decay chain. Within this chain, one intermediate product is 230Th, which has a half-life of ~75 ky. At secular equilibrium, the activity of 238U is equal to that of 230Th. As zircon grows, however, U is preferentially taken up over Th in the crystal (typically by a factor of five to ten), and the 238U/230Th activity shifts to values >1. Subsequent decay of 238U then serves to accumulate 230Th at a rate faster than the 230Th can decay until the decay and production rates are matched and a new secular equilibrium is achieved after about 350 ky (Schmitt 2011).
There are, thus, two ways of dating the zircons in young volcanic rocks. The first, using U/Th disequilibrium, can in principle be applied back to 350 ka (i.e. ~5 half-lives of 230Th), although the precisions of calculated ages decrease as the decay system approaches secular equilibrium. Once a data point falls within error of secular equilibrium, no meaningful age information can be derived, and the technique is therefore best suited to rocks that are <100 ka in age. The second method, applied to rocks that are >~150 ka, measures the 238U and 206Pb abundances in the crystals and so derives a direct age estimate, but this has to then incorporate a correction for the initial 230Th disequilibrium. Typically, this correction adds 60–100 ky to the age estimate, so it can be neglected for Neogene and older rocks, but is significant in Quaternary examples.
At present, measurements of isotopes for dating purposes in zircons revolve around three spectrometry techniques (Hanchar and Hoskin 2003; Schaltegger et al. 2015) (Fig. 1). Each technique has its advantages and disadvantages. Isotopic dilution thermal ionization mass spectrometry (ID-TIMS) is highly precise, but requires specialist chemical separation, clean laboratory techniques and is generally applied to single whole grains. Secondary ion mass spectrometry (SIMS) and laser ablation inductively coupled mass spectrometry (LA-ICP-MS) methods are less precise but have the advantages of the rapidity of analysis, the ability to analyse growth zones within crystals in order to place age determinations into a textural context, and not requiring specialist skills. In situ analysis of cross-sectioned grains allows for core–rim relationships to be explored, but the spot size (20–30 mm) does not permit the outermost growth zone to be analysed. Surface profiling can be used to analyse the outermost surfaces of grains but is then ‘blind’ with respect to textural information until after analysis.
CASE STUDIES IN THE VOLCANIC RECORD
Taupo Volcanic Zone (TVZ)
The central Taupo Volcanic Zone (New Zealand) has been an exceptionally active area of silicic volcanism for the past ~2 My (Wilson and Rowland 2016) and is comparable, in overall terms, to the modern Yellowstone system. Both are exceptionally dynamic areas that have produced multiple supereruptions, although associated with contrasting crustal settings: the TVZ with a rifting continental arc; Yellowstone with hotspot magmatism on thick continental crust. We use two scales here to highlight the volcanic–magmatic history of the TVZ: first, its shorter-term record (<60 ka), as expressed at the two highly active volcanoes of Taupo and Okataina; second, its overall 2 My record of waxing and waning.
Shorter-term Histories at Taupo and Okataina Volcanoes
Four key points arise about volcanic processes and timings from studies of young TVZ silicic volcanic rocks at Taupo and Okataina.
- Crystals in the eruption products are a mixture (just like in andesites). Some crystals (or their outer parts) are phenocrystic, grown in the evolving melt in which they were erupted; others are antecrystic, grown in forerunner parental melts; while others are xenocrystic, derived from unrelated sources. Age data and compositional and/or isotopic contrasts in zircon or other minerals can distinguish between the different crystal origins (Charlier et al. 2005). Mineral populations include variety at the whole-grain scale due to mixing (Storm et al. 2014), or a contrast between diverse, older cores extracted from the deeper magmatic roots and uniform rims grown in the melt-dominant body (a magmatic E Pluribus Unum) (Allan et al. 2013). Evidence for the existence of melt-dominant bodies may be muted or absent in the crystals that are left behind and that go to form any associated growing plutonic body.
- The (geo-)chemical processes involved in the development of a magma act on separate, longer timescales than the physical processes which assemble the erupted melt-dominant body. For example, the magma system for the 25.4 ka Oruanui eruption has a source history going back to ~100 ka, but the 530 km3 erupted magma body was assembled in <3,000 years (Allan et al. 2013), and the magma system was largely reset with respect to compositions and zircon age spectra in <5,000 years after the Oruanui eruption (Barker et al. 2014; Fig. 2). The TVZ rhyolites emphasise that eruptible melt-dominant bodies are ephemeral entities associated with longer-lived crystal-dominant plutonic sources.
- There may be multiple melt-dominant bodies present simultaneously in the shallow crust, each with independent crystallization histories (from zircon age spectra) and geochemical origins (e.g. Shane et al. 2008; Allan et al. 2012; Storm et al. 2014). Despite their eruptive vigour, neither Taupo nor Okataina volcanoes developed a single, all-encompassing magmatic system; instead, they have multiple heterogeneous magmatic domains developed over vertical and horizontal length scales of kilometres to tens of kilometres.
- There is strong evidence for external, tectonic controls on whether eruptions occur or not, and how they progress (Allan et al. 2012). Overall, there are complex interactions between magmatic and tectonic processes that (a) generate magma in crystal-rich crustal bodies, (b) cause melt-dominant magma bodies to accumulate, and (c) control whether such bodies erupt or cool back into crystal-rich mush (Rowland et al. 2010). The record at Taupo (Charlier et al. 2005; Barker et al. 2014) shows periods of enhanced zircon growth with cooling (and growth of other minerals) and a lack of eruptive activity, alternating with periods of disturbance (heating and/or rifting) with numerous eruptions (e.g. Fig. 2). These periods recur on timescales of 103–104 years, which is too short a scale to be easily seen in the Long Valley and Yellowstone systems where eruptions are spaced at 104–105 year intervals.
Medium- to Longer-Term Volcanic–Plutonic Record of the TVZ
On timescales of 105–106 years, the TVZ volcanic record is notably punctuated, with shifting caldera sources for large-scale volcanism. The zircon record implies that plutonism is equally episodic. Sometimes the two records are closely linked such that the peak of zircon crystallisation occurs just prior to eruption. However, in the larger systems (Oruanui, Whakamaru, Kidnappers, Ongatiti), there are peaks in the zircon age spectra sometimes >100 ky prior to eruption, suggesting that the plutonic and volcanic rhythms are not always synchronised on that timescale. In contrast, the deeper magmatic input into the TVZ, as reflected in the location and vigour (total ~4.2 gigawatt) of high-temperature (>250 °C) geothermal systems, is often inferred to have remained more uniform (Wilson and Rowland 2016).
Over the past 350 ky, within the central TVZ about 50% of the area has been occupied by calderas. If depths and volumes of the parental magma systems are similar to those of younger examples studied in detail, then there is a composite pluton being developed at depths of ~4–15 km below the surface (Figs. 3 and 4). Outside the caldera areas, the intense geothermal fluxes require that there are abundant and voluminous crustal intrusions, but these have reached levels accessible by drilling (~3.2 km) at only one location. If the activity from 2.0 Ma to 350 ka was similar in extent and intensity to that at <350 ka, then over ~6,000 km2 and a vertical depth of >10 km within the quartzo-feldspathic crust, the TVZ composite pluton has been constructed, and is continuing to grow as you read this article, at a modern rate of ~50 km3 per ky, with about 10–20% of its volume having been periodically spat out as eruptions.
This pluton is compositionally zoned, with only the shallower parts being quartz-bearing (Allan et al. 2013), and probably shows strong lateral compositional diversity at any given depth. Isotopically, it reflects a mixture of ~25% recycled metasedimentary crustal rocks and ~75% fractionates from mantle melts, which volumetrically means net crustal growth. Zircon age data from different portions of this pluton will show clustering into periods of enhanced crystallisation, but only longer-term periodicities would be discernible in a Mesozoic analogue. The TVZ pluton is only one of several composite silicic plutons developed as part of the New Zealand convergent plate boundary. Other plutons underlie the Coromandel area north of the TVZ, and their volcanic products are represented by voluminous silicic ash beds in deep-ocean cores (Carter et al. 2003); the earliest of the corresponding plutons – the 16.4 ± 0.1 Ma Paritu pluton (zircon U–Pb age: pers comm TR Ireland 2012) – has already been uplifted to the surface.
Long Valley, California
Long Valley and adjacent areas in eastern California are best known for the 0.77 Ma Bishop Tuff eruption, but the silicic volcanic record extends over ~3.5 My, up to as recent as ~1,350 CE (Hildreth 2004). Aspects of the volcanic record at Long Valley that are relevant here include the following.
- There have been six magmatic foci active since 3.5 Ma, with their surficial volcanic footprints migrating between adjacent areas, with minimal overlap and distinct compositional characteristics (Hildreth 2004). At depth, however, it is likely that all six foci are contributing towards the construction of a composite pluton (Figs. 3 and 4).
- Rhyolitic volcanic products of the Glass Mountain (~2.8–0.86 Ma) and Bishop systems have compositions indicative of extreme fractionation, consistent with co-generation of thousands of cubic kilometres of intermediate to silicic mush at depth (Hildreth 2004). Overall, the relative contributions of older pre-existing crustal lithologies have diminished with time as the proportions of mantle-derived material have increased (Simon et al. 2014). However, the Bishop Tuff zircon record shows that recycling of older Glass Mountain material was minimal: the Bishop magma body accumulated with its own zircon age signature over ~80 ky, as shown by age contrasts between zircon cores and rims (Chamberlain et al. 2014). The average of multiple rim ages from SIMS analyses replicates within uncertainty the average ID-TIMS age estimate from Bishop zircons and show that the ID-TIMS method masks the histories inherent in the grains. Any recycling of older plutonic material suggested from geochemical evidence (Simon et al. 2014) is concealed by the stripping out of older zircons and resetting of the magma chronological record (cf. Taupo: Barker et al. 2014).
- In the recent record there is a close link between volcanism, dike emplacement and tectonism (Bursik et al. 2003). Similar links are suspected to have occurred in the past, but are hard to demonstrate. The overall position and shape of the Bishop Tuff caldera reflects a pre-existing jog in the Sierra Nevada Range’s front (Riley et al. 2012). During the Bishop eruption, the initial vent area in the south-central part of the caldera lay along the line of the Laurel Creek Fault. The extension of this fault connects through the post-caldera resurgent dome structure to reach the northern caldera rim where activity first broke out in that sector. A tectonic connection, since exploited by the resurgent dome activity, is suspected to be present.
There is a still-growing ‘Long Valley pluton’ that extends (from the volcanic footprints) over ~1,500 km2, elongate parallel to the eastern Sierra front, and extending from ~6 km depth to the base of the crust (Figs. 3 and 4). It has been growing over ~3.5 My, with a record that can be investigated through zircon studies of ~2.8 My. As with the TVZ pluton, there is clear evidence for vertical and lateral heterogeneity (Hildreth 2004) and a long-term role for tectonic processes (Riley et al. 2012). In contrast with the TVZ, there is evidence for systematic changes in the evolution of the pluton (Simon et al. 2014). Note that the time gap between growth of the Long Valley pluton and the youngest of the adjacent Sierra Nevada plutons (i.e. 2.8 to 87 Ma) is less than the time gap between adjacent plutons within the Sierra Nevada batholith. If translated back into the Precambrian, would the Long Valley pluton be treated, for example, as another contributor to the greater Sierra Nevada batholith?
Yellowstone is the latest of at least seven large silicic caldera-related volcanic systems associated with migration of the North American plate over a deep-seated hotspot (Morgan et al. 2009). These systems have developed over 16.5 My, stretching for 700 km along the Snake River Plain and marking the volcanic expression of major silicic magmatism, although the earlier examples are now buried by voluminous basalts. The Yellowstone system itself dates back to 2.1 Ma, has generated three caldera-forming events and last erupted with voluminous rhyolite lava around 70 ka (Christiansen 2001). Zircon age (and other crystal-specific) data on the caldera-forming eruptions and their aftermaths paint diverse pictures of the Yellowstone system. Contrasting views are held regarding the timing and origins of the younger rhyolites: these views centre on the role of melting and recycling of crustal lithologies versus the extraction of melts from a long-lived mush system (Watts et al. 2012; Stelten et al. 2015; Wotzlaw et al. 2015).
At present, geophysical data outline the pluton-scale volumes of inferred partially molten material below the youngest caldera (Huang et al. 2015). There is an upper crustal (<20 km depth) volume of 10,000 km3 inferred to be rhyolitic with an average 9% melt, and a lower crustal (25–45 km depth) volume of 46,000 km3 inferred to be basaltic and with ~2% melt. These volumes of partially molten material are the drivers of exceptional fluxes of geothermal heat (about 5.3 GW) and of volatiles from magma and heated country rock (Hurwitz and Lowenstern 2014). None of the other silicic systems considered here has been geophysically imaged to this level of detail, or to such depths.
Yellowstone, when compared to the TVZ and Long Valley, shows some contrasts, despite the generally comparable timespans of activity.
- In the Snake River Plain, silicic volcanism and plutonism are followed by voluminous mafic volcanism, burying the silicic source calderas and modifying the crustal signature of the silicic magmatism. At Yellowstone, minor amounts of basalt predate and are interspersed with rhyolitic products of the older two calderas. Then, as the volcanic focus has migrated east, the western parts of the two earlier calderas have begun to be buried under basalt lavas. The appearance of abundant basalt volcanism has been taken as the sign (as at Long Valley) that the upper crustal magmatic system has waned and solidified and allowed basalt to pass through, rather than acting as a density trap (Hildreth 2004). This systematic progression contrasts with the great diversity of behaviour in the TVZ where the crustal magmatic systems wax and wane multiple times within a timescale equivalent to the lifespan of one caldera cycle at Yellowstone and where there is no overall progression to basaltic volcanism at the surface.
- The overall bimodal (basalt and rhyolite) nature of the Yellowstone eruptive products has fuelled much debate over the relative roles of crustal melting versus fractionation and the apparent absence of intermediate compositions (e.g. Streck 2014). Petrological contrasts between Yellowstone and areas like the TVZ are mirrored in the western USA in older Cenozoic silicic magmatism of the Basin-and-Range province, and attributed there to contrasts in the volatile abundances and temperatures of the magmatic systems (Christiansen and McCurry 2008).
- The plutonic bodies associated with the rhyolitic volcanism at Taupo and Yellowstone both appear to have distinct felsic upper parts (10–12 km thick at Taupo, ~15 km at Yellowstone) and mafic lower parts. At Taupo, such stratification cannot be distinguished geophysically from mantle lithologies. At Long Valley, the roots of the pluton merge upwards through the crust over a thickness of ~30 km, and a distinct ‘mafic versus felsic’ separation is not thought to be present.
This brief overview of three Quaternary silicic volcanic systems reveals several factors that should be considered when making any comparison with the plutonic record.
- Although large-scale crustal magmatism and growth of plutonic bodies need not be accompanied by large-scale volcanism (e.g. Bacon et al. 1981), large-scale silicic volcanism demands the presence of complementary greater volumes of intrusive material (cf. Streck 2014). There is a central role for mafic magmatism in providing heat, volatiles and differentiated melts, but these are expressed in different guises: as a sharply defined mafic underplate to the TVZ (Wilson and Rowland 2016); as diffuse ‘distributed intrusions’ at Long Valley (Hildreth 2004); and as a vast, vertically extensive, thermal root at Yellowstone (Huang et al. 2015) (Fig. 4).
- Zircon chronologies for silicic volcanic and plutonic rocks are open to different interpretations. Age estimates for single crystals or parts of crystals only date those crystals or parts of crystals – they do not, in themselves, provide direct information on the onset of any particular magmatic system or constrain the magma residence time. Context is paramount in interpreting age information. Timescales visible through zircon records are also limited by the associated precisions. Analytical precisions of ±0.2% at 95% confidence for ID-TIMS are readily achievable, but, even so, the uncertainties on ages of grains in Mesozoic Sierra Nevada batholiths cover the lifespan of entire caldera-forming supereruptive magma systems in the TVZ. The techniques associated with ID-TIMS age dating of zircons (in particular the use of chemical abrasion and the analysis of whole or half grains) serve to mask the very diversity of age data that give insights into the rhythm and tempo of Quaternary volcanic systems (e.g. Chamberlain et al. 2014). The demonstrable rapidity with which super-sized volcanic systems can grow, erupt, then change into new systems is invisible in the plutonic record, not only because of the prolonged cooling histories of plutons but also because the dating methodologies cannot keep up with the pace. A fundamental implication from studies of Quaternary systems is that large-scale volcanic processes can act on timescales that are too rapid for most plutonic records to see.
- Fluctuations in volcanic output do not always directly reflect magmatic input at depth. This is demonstrably the case on the ‘short’ timescales of thousands to tens-of-thousands of years in the TVZ, where periods of enhanced crystallization (i.e. magma cooling) recorded in the zircon age spectra occur during lulls in surface volcanism. The processes associated with physical accumulation of eruptible bodies of magma operate on timescales that may be wholly disconnected from the processes of assimilation and fractionation that control the chemical compositions of silicic volcanic rocks. As a result, the physical accumulation of super-sized melt-dominant bodies can occur much more rapidly than gradualistic models would suggest.
We thank Joe Wooden (Stanford University, USA) and Trevor Ireland (Australian National University) for their help and collaboration in our zircon ion-probe studies. We also thank Aidan Allan, Simon Barker, Katy Chamberlain, George Cooper and Wes Hildreth for many valuable discussions, and Craig Lundstrom, Axel Schmitt and Justin Simon for illuminating reviews.
Allan ASR, Wilson CJN, Millet M-A, Wysoczanski RJ (2012)
The invisible hand: tectonic triggering and modulation of a rhyolitic supereruption. Geology 40: 563-566
Allan ASR, Morgan DJ, Wilson CJN, Millet M-A (2013)
From mush to eruption in centuries: assembly of the super-sized Oruanui magma body. Contributions to Mineralogy and Petrology 166: 143-164
Bachmann O, Bergantz GW (2004)
On the origin of crystal-poor rhyolites: extracted from batholithic crystal mushes. Journal of Petrology 45: 1565-1582
Bacon CR, Macdonald R, Smith RL, Baedecker PA (1981)
Pleistocene high-silica rhyolites of the Coso volcanic field, Inyo County, California. Journal of Geophysical Research: Solid Earth 86: 10223-10241
Barker SJ and 6 coauthors (2014)
Post-supereruption magmatic reconstruction of Taupo volcano (New Zealand), as reflected in zircon ages and trace elements. Journal of Petrology 55: 1511-1533
Bursik MI, Renshaw C, McCalpin J, Berry M (2003)
A volcanotectonic cascade: activation of range front faulting and eruptions by dike intrusion, Mono Basin-Long Valley caldera, California. Journal of Geophysical Research: Solid Earth 108, doi: 10.1029/2002JB002032
Carter L and 5 coauthors (2003)
Demise of one volcanic zone and birth of another - a 12 m.y. marine record of major rhyolitic eruptions from New Zealand. Geology 31: 493-496
Chamberlain KJ, Wilson CJN, Wooden JL, Charlier BLA, Ireland TR (2014)
New perspectives on the Bishop Tuff from zircon textures, ages and trace elements. Journal of Petrology 55: 395-426
Chamberlain KJ, Wilson CJN, Wallace PJ, Millet M-A (2015)
Micro-analytical perspectives on the Bishop Tuff and its magma chamber. Journal of Petrology 56: 605-640
Charlier BLA and 5 coauthors (2005)
Magma generation at a large, hyperactive silicic volcano (Taupo, New Zealand) revealed by U-Th and U-Pb systematics in zircons. Journal of Petrology 46: 3-32
Christiansen EH, McCurry M (2008)
Contrasting origins of Cenozoic silicic volcanic rocks from the western Cordillera of the United States. Bulletin of Volcanology 70: 251-267
Christiansen RL (2001)
The Quaternary and Pliocene Yellowstone Plateau volcanic field of Wyoming, Idaho, and Montana. U.S. Geological Survey Professional Paper 729-G: 1-143
Hanchar JM, Hoskin PWO (eds) (2003)
Zircon. Reviews in Mineralogy and Geochemistry, Volume 53. Mineralogical Society of America, Chantilly, 500 pp
Hildreth W (2004)
Volcanological perspectives on Long Valley, Mammoth Mountain, and Mono Craters: several contiguous but discrete systems. Journal of Volcanology and Geothermal Research 136: 169-198
Huang H-H and 5 coauthors (2015)
The Yellowstone magmatic system from the mantle plume to the upper crust. Science 348: 773-776
Hurwitz S, Lowenstern JB (2014)
Dynamics of the Yellowstone hydrothermal system. Reviews of Geophysics 52: 375-411
Morgan LA, Cathey HE, Pierce KL (eds) (2009)
The track of the Yellowstone hotspot: multi-disciplinary perspectives on the origin of the Yellowstone-Snake River Plain volcanic province. Journal of Volcanology and Geothermal Research 188: 1-304
Riley P, Tikoff B, Hildreth W (2012)
Transtensional deformation and structural control of contiguous but independent magmatic systems: Mono-Inyo Craters, Mammoth Mountain, and Long Valley caldera, California. Geosphere 8: 740-751
Rowland JV, Wilson CJN, Gravley DM (2010)
Spatial and temporal variations in magma-assisted rifting, Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research 190: 89-108
Schaltegger U, Schmitt AK, Horstwood MSA (2015)
U–Th–Pb zircon geochronology by ID-TIMS, SIMS, and laser ablation ICP-MS: recipes, interpretations, and opportunities. Chemical Geology 402: 89-110
Schmitt AK (2011)
Uranium series accessory crystal dating of magmatic processes. Annual Review of Earth and Planetary Sciences 39: 321-349
Shane P and 5 coauthors (2008)
Silicic recharge of multiple rhyolite magmas by basaltic intrusion during the 22.6 ka Okareka eruption episode, New Zealand. Lithos 103: 527-549
Simon JI and 5 coauthors (2014)
Assimilation of preexisting Pleistocene intrusions at Long Valley by periodic magma recharge accelerates rhyolite generation: rethinking the melting model. Contributions to Mineralogy and Petrology 167, doi: 10.1007/s00410-013-0955-5
Stelten ME, Cooper KM, Vazquez JA, Calvert AT, Glessner JJG (2015)
Mechanisms and timescales of generating eruptible rhyolitic magmas at Yellowstone caldera from zircon and sanidine geochronology and geochemistry. Journal of Petrology 56: 1607-1642
Storm S, Schmitt AK, Shane P, Lindsay JM (2014)
Zircon trace element chemistry at sub-micrometer resolution for Tarawera volcano, New Zealand, and implications for rhyolite magma evolution. Contributions to Mineralogy and Petrology 167, doi: 10.1007/s00410-014-1000-z
Streck MJ (2014)
Evaluation of crystal mush extraction models to explain crystal-poor rhyolites. Journal of Volcanology and Geothermal Research 284: 79-94
Watts KE, Bindeman IN, Schmitt AK (2012)
Crystal scale anatomy of a dying supervolcano: an isotope and geochronology study of individual phenocrysts from voluminous rhyolites of the Yellowstone caldera. Contributions to Mineralogy and Petrology 164: 45-67
Wilson CJN, Rowland JV (2016)
The volcanic, magmatic and tectonic setting of the Taupo Volcanic Zone, New Zealand, reviewed from a geothermal perspective. Geothermics 59: 168-187
Wilson CJN, Blake S, Charlier BLA, Sutton AN (2006)
The 26.5 ka Oruanui eruption, Taupo volcano, New Zealand: development, characteristics and evacuation of a large rhyolitic magma body. Journal of Petrology 47: 35-69
Wotzlaw J-F, Bindeman IN, Stern RA, D’Abzac F-X, Schaltegger U (2015)
Rapid heterogeneous assembly of multiple magma reservoirs prior to Yellowstone supereruptions. Nature Scientific Reports 5, doi: 10.1038/srep14026