Wes Hildreth

Crustal contributions to arc magmatism in the Andes of Central Chile

Fifteen andesite-dacite stratovolcanoes on the volcanic front of a single segment of the Andean arc show along-arc changes in isotopic and elemental ratios that demonstrate large crustal contributions to magma genesis. All 15 centers lie 90 km above the Benioff zone and 280±20 km from the trench axis. Rate and geometry of subduction and composition and age of subducted sediments and seafloor are nearly constant along the segment. Nonetheless, from S to N along the volcanic front (at 57.5% SiO2) K2O rises from 1.1 to 2.4 wt %, Ba from 300 to 600 ppm, and Ce from 25 to 50 ppm, whereas FeO*/MgO declines from >2.5 to 1.4. Ce/Yb and Hf/Lu triple northward, in part reflecting suppression of HREE enrichment by deep-crustal garnet. Rb, Cs, Th, and U contents all rise markedly from S to N, but Rb/Cs values double northward — opposite to prediction were the regional alkali enrichment controlled by sediment subduction. K/Rb drops steeply and scatters greatly within many (biotite-free) andesitic suites. Wide diversity in Zr/Hf, Zr/Rb, Ba/Ta, and Ba/La within and among neighboring suites (which lack zircon and alkali feldspar) largely reflects local variability of intracrustal (not slab or mantle) contributions. Pb-isotope data define a limited range that straddles the Stacey-Kramers line, is bracketed by values of local basement rocks, in part plots above the field of Nazca plate sediment, and shows no indication of a steep (mantle+sedimentary) Pb mixing trend. 87Sr/86Sr values rise northward from 0.7036 to 0.7057, and 143Nd/144Nd values drop from 0.5129 to 0.5125. A northward climb in basal elevation of volcanic-front edifices from 1350 m to 4500 m elevation coincides with a Bougueranomaly gradient from −95 to −295 mgal, interpreted to indicate thickening of the crust from 30–35 km to 50–60 km. Complementary to the thickening crust, the mantle wedge beneath the front thins northward from about 60 km to 30–40 km (as slab depth is constant). The thick northern crust contains an abundance of Paleozoic and Triassic rocks, whereas the proportion of younger arc-intrusive basement increases southward. Primitive basalts are unknown anywhere along the arc. Base-level isotopic and chemical values for each volcano are established by blending of subcrustal and deep-crustal magmas in zones of melting, assimilation, storage and homogenization (MASH) at the mantle-crust transition. Scavenging of mid-to upper-crustal silicic-alkalic melts and intracrustal AFC (prominent at the largest center) can subsequently modify ascending magmas, but the base-level geochemical signature at each center reflects the depth of its MASH zone and the age, composition, and proportional contribution of the lowermost crust.

Large partition coefficients for trace elements in high-silica rhyolites

Abstract The partitioning of 25 trace elements between high-silica rhyolitic glass and unzoned phenocrysts of potassic and sodic sanidine, biotite, augite, ferrohedenbergite, hypersthene, fayalite, titanomagnetite, ilmenite, zircon, and allanite has been determined by INAA on suites of samples from the mildly peralkaline lavas and tuff of the Sierra La Primavera, Mexico, and the metaluminous, compo. sitionally zoned, Bishop Tuff, California. The partition coefficients are much larger than published values for less silicic compositions; the range of values among Primavera samples that differ only slightly in temperature or bulk composition approaches that previously reported from basalts to rhyodacites. Intrinsic temperature dependence of the crystal/liquid partitioning is apparently small. The high values of partition coefficients reflect principally the strongly polymerized nature of the alkali-aluminosilicate liquid, whereas the marked variability of values for partition coefficients is attributed to differences in the concentrations of complexing ligands and/or different degrees of melt polymerization. Great variation in the values of partition coefficients that are potentially applicable to early stages in the partial melting of crustal rocks complicates assessment of 1. (1) source regions for granitic melts and 2. (2) contributions by crustal-melt increments to andesites.

The compositionally zoned eruption of 1912 in the Valley of Ten Thousand Smokes, Katmai National Park, Alaska

On June 6–8, 1912, ∼ 15 km3 of magma erupted from the Novarupta caldera at the head of the Valley of Ten Thousand Smokes (VTTS), producing ∼ 20 km3 of air-fall tephra and 11–15 km3 of ash-flow tuff within ∼ 60 hours. Three discrete periods of ash-fall at Kodiak correlate, respectively, with Plinian tephra layers designated A, CD, and FG by Curtis (1968) in the VTTS. The ash-flow sequence overlapped with but outlasted pumice fall A, terminating within 20 hours of the initial outbreak and prior to pumice fall C. Layers E and H consist mostly of vitric dust that settled during lulls, and Layer B is the feather edge of the ash flow. The fall units filled and obscured the caldera, but arcuate and radial fissures outline a 6-km2 depression. The Novarupta lava dome and its ejecta ring were emplaced later within the depression. At Mt. Katmai, 10 km east of the 1912 vent, a 600-m-deep caldera of similar area also collapsed at about this time, probably owing to hydraulic connection with the venting magma system; but all known ejecta are thought to have erupted at Novarupta. Mingling of three distinctive magmas during the eruption produced an abundance of banded pumice, and mechanical mixing of chilled ejecta resulted in deposits with a wide range of bulk composition. Pumice in the initial fall unit (A) is 100% rhyolite, but fall units atop the ash flow are > 98% dacite; black andesitic scoria is common only in the ash flows and in near-vent air-fall tephra. Pumice counts show the first half of the ash-flow deposit to be 91–98% rhyolite, but progressive increases of dacite and andesite eventually reduced the rhyolitic component to < 2%. The later, rhyolite-poor flows were hotter, less mobile, and widely produced partially welded tuff and vapor-indurated sillar. The main ash flow was too deflated and sluggish 16 km from the vent to surmount a 25-m-high moraine in its path but was diverted around it and continued 5 km down-valley, engulfing and charring trees but not toppling all of them. Thin ash-flow veneers feather 30–40 m up the enclosing valley walls but only where a constriction in the central VTTS locally raised the flow level. In the upper VTTS, the “high sand mark” is not a veneer but a marginal bench formed in thick tuff by differential compaction. Flooding from adjacent glaciers led to phreatic explosions that ejected blocks of tuff more welded than any yet exposed. A cluster of phreatic craters dammed a lake atop the tuff, the breaching of which caused a flood that scoured the ash-flow surface in the central VTTS, transported 50-cm blocks of welded tuff > 20 km to the lowermost VTTS, and deposited 1–8 m of debris there. Rhyolitic ejecta contain only 1–2% phenocrysts but andesite and dacite have 30–45%. Quartz is present and augite absent only in the rhyolite, but all ejecta contain plagioclase, orthopyroxene, titanomagnetite, ilmenite, apatite, and pyrrhotite; rare olivine occurs in the andesite. The zoning ranges of phenocrysts in the rhyolitic and intermediate ejecta do not overlap. New chemical data show the bulk SiO2 range to be: rhyolite 77 ± 0.6, dacite 66-64.5, and andesite 61.5–58.5%. The dacitic and andesitic ejecta contrast in color and density, and it is not certain whether they form a compositional continuum. Analyses reported by Fenner within the 66–76% SiO2 range were of banded pumice and lava and of bulk tephra that mechanically fractionated and mixed during flight. Despite the gap of 10% SiO2, Fe-Ti-oxide temperatures show a continuous range from rhyolite (805–850°C) through dacite (855–955°C) to andesite (955–990°C). Thermal continuity and isotopic and trace-element data suggest that all were derived from a single magmatic system, whether or not they were physically contiguous before eruption. If the rhyolitic liquid separated from dacitic magma, extraction was so efficient that no dacitic phenocrysts were retained and no bulk compositions in the range 66–76% SiO2 were created; if it were a partial melt of roof rocks atop an intermediate magma body, then such rocks had no O- or Sr-isotopic contrast with the andesite-dacite magma and clearly did not include the Jurassic arkosic or granitic basement. The presence of Holocene domes of pre-1912 glassy dacite adjacent to the 1912 vent suggest that the 7 km3 (or more) of high-silica rhyolitic magma (a composition rare in the Aleutian arc) was generated in less than a few thousand years. The 1912 vent is semi-encircled by several andesitic stratocones and is as close to Mageik, Trident, and Griggs volcanoes as it is to Mt. Katmai. The erupted magma probably occupied only shallow levels of an extensive system of injection and storage under a cluster of several stratovolcanoes. Although Quaternary basalt is not known to have erupted here, the intrusion of basaltic magma probably sustains the greater-VTTS magmatic system.

The Bishop Tuff: New Insights From Eruptive Stratigraphy

The 0.76 Ma Bishop Tuff, from Long Valley caldera in eastern California, consists of a widespread fall deposit and voluminous partly welded ignimbrite. The fall deposit (F), exposed over an easterly sector below and adjacent to the ignimbrite, is divided into nine units (F1‐F9), with no significant time breaks, except possibly between F8 and F9. Maximum clast sizes are compared with other deposits where accumulation rates are known or inferred to estimate an accumulation time for F1‐F8 as ca. 90 hrs. The ignimbrite (Ig) is divided into chronologically and/or geographically distinct packages of material. Earlier packages (Ig1) were emplaced mostly eastward, are wholly intraplinian (coeval with fall units F2‐F8), Lack phenocrystic pyroxenes, and contain few or no Glass Mountain‐derived rhyolite lithic fragments. Earlier packages (Ig2) were erupted mostly to the north and east, are at least partly intraplinian (interbedded with fall unit F9 to the east), contain pyroxenes, and have lithic fractions rich in Glass Mountain‐derived rhyolite or other lithologies exposed on the northern caldera rim. Recognition of the intraplinian nature of Ig1 east of the caldera and use of the fall deposit chronometry yields accumulation estimates of ca. 25 hrs for an earlier, less‐welded subpackage and ca. 36 hrs for a later, mostly welded subpackage. Average accumulation rates range up to ≥1 mm/s of dense‐welded massive ignimbrite, equivalent to ≥2.5 mm/s of non‐welded material. Comparisons of internal stratification in Ig1 and northern Ig2 lobes suggest the thickest northern ignimbrite accumulated in ≥35 hrs. Identifiable vent positions migrated from an initial site previously proposed in the south‐central part of the caldera (F1‐8, Ig1) in complex fashion; one vent set (for eastern Ig2) migrated east and north toward Glass Mountain, while another set (for northern Ig2) opened from west to east across the northern caldera margin. Vent locations for Ig1 and Ig2 southwest of the caldera have not been identified. The new stratigraphic framework shows that much of the Bishop ignimbrite is intraplinian in nature, and that fall deposits and ignimbrite units previously inferred to be sequential are largely or wholly coeval. Fundamental reassessment is therefore required of all existing models for the eruption dynamics and the nature and causes of pre‐eruptive zonations in trace elements, volatiles, and isotopes in the parental magma chamber.

Pre-eruption recharge of the Bishop magma system

The 650 km 3 rhyolitic Bishop Tuff (eastern California, USA), which is stratigraphically zoned with respect to temperatures of mineral equilibration, refl ects a corresponding thermal gradient in the source magma chamber. Consistent with previous work, application of the new TitaniQ (Ti-in-quartz) thermometer to quartz phenocryst rims documents an ~100 °C temperature increase with chamber depth at the time of eruption. Application of TitaniQ to quartz phenocryst cores, however, reveals lower temperatures and an earlier gradient that was less steep, with temperature increasing with depth by only ~30 °C. In many late-erupted crystals, sharp boundaries that separate low-temperature cores from high-temperature rims cut internal cathodoluminescent growth zoning, indicating partial phenocryst dissolution prior to crystallization of the high-temperature rims. Rimward jumps in Ti concentration across these boundaries are too abrupt (e.g., 40 ppm across a distance of <10 µm) to have survived magmatic temperatures for more than ~100 yr. We interpret these observations to indicate heating-induced partial dissolution of quartz, followed by growth of high-temperature rims (made possible by lowering of water activity due to addition of CO 2 ) within 100 yr of the climactic 760 ka eruption. Hot mafi c melts injected into deeper parts of the magma system were the likely source of heat and CO 2 , raising the possibility that eruption and caldera collapse owe their origin to a recharge event.

The plinian eruptions of 1912 at Novarupta, Katmai National Park, Alaska

The three-day eruption at Novarupta in 1912 consisted of three discrete episodes. Episode I began with plinian dispersal of rhyolitic fallout (Layer A) and contemporaneous emplacement of rhyolitic ignimbrites and associated proximal veneers. The plinian column was sustained throughout most of the interval of ash flow generation, in spite of progressive increases in the proportions of dacitic and andesitic ejecta at the expense of rhyolite. Accordingly, plinian Layer B, which fell in unbroken continuity with purely rhyolitic Layer A, is zoned from >99% to ∼15% rhyolite and accumulated synchronously with emplacement of the correspondingly zoned ash flow sequence in Mageik Creek and the Valley of Ten Thousand Smokes (VTTS). Only the andesiterichest flow units that cap the flow sequence lack a widespread fallout equivalent, indicating that ignimbrite emplacement barely outlasted the plinian phase. On near-vent ridges, the passing ash flows left proximal ignimbrite veneers that share the compositional zonation of their valley-filling equivalents but exhibit evidence for turbulent deposition and recurrent scour. Episode II began after a break of a few hours and was dominated by plinian dispersal of dacitic Layers C and D, punctuated by minor proximal intraplinian flows and surges. After another break, dacitic Layers F and G resulted from a third plinian episode (III); intercalated with these proximally are thin intraplinian ignimbrites and several andesite-rich fall/flow layers. Both CD and FG were ejected from an inner vent <400 m wide (nested within that of Episode I), into which the rhyolitic lava dome (Novarupta) was still later extruded. Two finer-grained ash layers settled from composite regional dust clouds: Layer E, which accumulated during the D-F hiatus, includes a contribution from small contemporaneous ash flows; and Layer H settled after the main eruption was over. Both are distinct layers in and near the VTTS, but distally they merge with CD and FG, respectively; they are largely dacitic but include rhyolitic shards that erupted during Episode I and were kept aloft by atmospheric turbulence. Published models yield column heights of 23–26 km for A, 22–25 km for CD, and 17–23 km for FG; and peak mass eruption rates of 0.7–1x108, 0.6–2x108, and 0.2–0.4x108 kg s-1, respectively. Fallout volumes, adjusted to reflect calculated redistribution of rhyolitic glass shards, are 8.8 km3, 4.8 km3, and 3.4 km3 for Episodes I, II, and III. Microprobe analyses of glass show that as much as 0.4 km3 of rhyolitic glass shards from eruptive Episode I fell with CDE and 1.1 km3 with FGH. Most of the rhyolitic ash in the dacitic fallout layers fell far downwind (SE of the vent); near the rhyolite-dominated ignimbrite, however, nearly all of Layers E and H are dacitic, showing that the downwind rhyolitic ash is of ‘co-plinian’ rather than co-ignimbrite origin.

Volcán Quizapu, Chilean Andes

Quizapu is a flank vent of the basalt-to-rhyodacite Holocene stratocone, Cerro Azul, and lies at the focus of a complex Quaternary volcanic field on the Andean volcanic front. The Quizapu vent originated in 1846 when 5 km3 of hornblende-dacite magma erupted effusively with little accompanying tephra. Between ∼ 1907 and 1932, phreatic and strombolian activity reamed out a deep crater, from which 4 km3 of dacite magma identical to that of 1846 fed the great plinian event of 10–11 April 1932. Although a total of >9 km3 of magma was thus released in 86 years, there is no discernible subsidence. As the pre-plinian crater was lined by massive lavas, 1932 enlargement was limited and the total plinian deposit contains only ∼ 0.4 wt % lithics. Areas of 5-cm and 1-cm isopachs for compacted 1932 fallout are about half of those estimated in the 1930s, yielding a revised ejecta volume of ∼9.5 km3. A strong inflection near the 10-cm isopach (downwind ∼110 km) on a plot of log Thickness vs Area1/2 reflects slow settling of fine plinian ash — not of coignimbrite ash, as the volume of pyroclastic flows was trivial (<0.01 km3). About 17 vol.% of the fallout lies beyond the 1-cm isopach, and ∼ 82 wt% of the ejecta are finer than 1 mm. A least 18 hours of steady plinian activity produced an exceptionally uniform fall deposit. Observed column height (27–30 km) and average mass eruption rate (1.5x108 kg/s) compare well with values for height and peak intensity calculated from published eruption models. The progressive “aeolian fractionation” of downwind ash (for which Quizapu is widely cited) is complicated by the large compositional range of 1932 juvenile pumice (52–70% SiO2). The eruption began with andesitic scoria and ended with basaltic scoria, but >95% of the ejecta are dacitic pumice (67–68% SiO2); minor andesitic scoria and frothier rhyodacite pumice (70% SiO2) accompanied the dominant dacite. Phenocrysts (pl>hb∼opx>mt>ilm∼cpx) are similar in both abundance and composition in the 1846 (effusive) and 1932 (plinian) dacites. Despite the contrast in mode of eruption, bulk compositions are also indistinguishable. The only difference so far identified is a lower range of δ D values for 1846 hornblende, consistent with pre-eruptive degassing of the effusive batch.

Geology of the peralkaline volcano at Pantelleria, Strait of Sicily

Situated in a submerged continental rift, Pantelleria is a volcanic island with a subaerial eruptive history longer than 300 Ka. Its eruptive behavior, edifice morphologies, and complex, multiunit geologic history are representative of strongly peralkaline centers. It is dominated by the 6-km-wide Cinque Denti caldera, which formed ca. 45 Ka ago during eruption of the Green Tuff, a strongly rheomorphic unit zoned from pantellerite to trachyte and consisting of falls, surges, and pyroclastic flows. Soon after collapse, trachyte lava flows from an intracaldera central vent built a broad cone that compensated isostatically for the volume of the caldera and nearly filled it. Progressive chemical evolution of the chamber between 45 and 18 Ka ago is recorded in the increasing peralkalinity of the youngest lava of the intracaldera trachyte cone and the few lavas erupted northwest of the caldera. Beginning about 18 Ka ago, inflation of the chamber opened old ring fractures and new radial fractures, along which recently differentiated pantellerite constructed more than 25 pumice cones and shields. Continued uplift raised the northwest half of the intracaldera trachyte cone 275 m, creating the islands present summit, Montagna Grande, by trapdoor uplift. Pantellerite erupted along the trapdoor faults and their hingeline, forming numerous pumice cones and agglutinate sheets as well as five lava domes. Degassing and drawdown of the upper pantelleritic part of a compositionally and thermally stratified magma chamber during this 18-3-Ka episode led to entrainment of subjacent, crystal-rich, pantelleritic trachyte magma as crenulate inclusions. Progressive mixing between host and inclusions resulted in a secular decrease in the degree of evolution of the 0.82 km3 of magma erupted during the episode.The 45-Ka-old caldera is nested within the La Vecchia caldera, which is thought to have formed around 114 Ka ago. This older caldera was filled by three widespread welded units erupted 106, 94, and 79 Ka ago. Reactivation of the ring fracture ca. 67 Ka ago is indicated by venting of a large pantellerite centero and a chain of small shields along the ring fault. For each of the two nested calderas, the onset of postcaldera ring-fracture volcanism coincides with a low stand of sea level.Rates of chemical regeneration within the chamber are rapid, the 3% crystallization/Ka of the post-Green Tuff period being typical. Highly evolved pantellerites are rare, however, because intervals between major eruptions (averaging 13−6 Ka during the last 190 Ka) are short. Benmoreites and mugearites are entirely lacking. Fe-Ti-rich alkalic basalts have erupted peripherally along NW-trending lineaments parallel to the enclosing rift but not within the nested calderas, suggesting that felsic magma persists beneath them. The most recent basaltic eruption (in 1891) took place 4 km northwest of Pantelleria, manifesting the long-term northwestward migration of the volcanic focus. These strongly differentiated basalts reflect low-pressure fractional crystallization of partial melts of garnet peridotite that coalesce in small magma reservoirs replenished only infrequently in this continental rift environment.

Ring-fracture eruption of the Bishop Tuff

Lithic fragments of precaldera basement rocks in the Plinian fallout deposit of the Bishop Tuff indicate that the eruption began in what is now the south-central part of Long Valley caldera, along or adjacent to the Hilton Creek fault. The earliest ash flows originated there as well, but contrasting lithic contents in several later ash flows indicate ring-fault propagation and incorporation of fragments of locally distinctive, caldera-margin lithologies into successive outflow sheets. The change from a single-vent Plinian mode of eruption to multiple vents along the ring fault took place after emplacement of as little as 20% of the total Bishop ejecta. Approximately two-thirds of the total eruptive volume was trapped within the subsiding caldera.

Katmai volcanic cluster and the great eruption of 1912

In June 1912, the world9s largest twentieth century eruption broke out through flat-lying sedimentary rocks of Jurassic age near the base of Trident volcano on the Alaska Peninsula. The 60 h ash-flow and Plinian eruptive sequence excavated and subsequently backfilled with ejecta a flaring funnel-shaped vent since called Novarupta. The vent is adjacent to a cluster of late Quaternary stratocones and domes that have released about 140 km 3 of magma in the past 150 k.y. Although the 1912 vent is closest to the Trident group and is also close to Mageik and Griggs volcanoes, it was the summit of Mount Katmai, 10 km east of Novarupta, that collapsed during the eruption to form a 5.5 km 3 caldera. Many earthquakes, including 14 in the range M 6−7, took place during and after the eruption, releasing 250 times more seismic energy than the 1991 caldera-forming eruption of the Philippine volcano, Pinatubo. The contrast in seismic behavior may reflect the absence of older caldera faults at Mount Katmai, lack of upward (subsidence opposing) magma flow owing to lateral magma withdrawal in 1912, and the horizontally stratified structure of the thick shale-rich Mesozoic basement. The Katmai caldera compensates for only 40% of the 13 km 3 of 1912 magma erupted, which included 7–8 km 3 of slightly zoned high-silica rhyolite and 4.5 km 3 of crystal-rich dacite that grades continuously into 1 km 3 of crystal-rich andesite. We have now mapped, sampled, and studied the products of all 20 components of the Katmai volcanic cluster. Pyroxene dacite and silicic andesite predominate at all of them, and olivine andesite is also common at Griggs, Katmai, and Trident volcanoes, but basalt and rhyodacite have erupted only at Mount Katmai. Rhyolite erupted only in 1912 and is otherwise absent among Quaternary products of the cluster. Pleistocene products of Mageik and Trident and all products of Griggs are compositionally distinguishable from those of 1912 at Novarupta. Holocene products of Mount Martin and Trident are closer in composition to the andesite-dacite array of 1912, but they reveal consistent differences. The affinity of the 1912 suite is closest with the array of products erupted by the Southwest Katmai cone, the edifice that had produced the only pre-1912 rhyodacite as well as the largest prehistoric Plinian eruption in the cluster. It is doubtful that any 1912 magma had been stored beneath Novarupta or Trident, and there is no evidence that more than one magma chamber erupted in 1912. Despite a compositional gap separating the aphyric rhyolite from the very crystal-rich andesite-dacite continuum, isotopic and chemical affinities linking all the 1912 ejecta and the continuity of all those ejecta in magmatic temperature and oxygen fugacity suggest that the rhyolite originated principally by incremental upward expulsion of interstitial melt from subjacent andesite-dacite mush. A large reservoir of such hot crystal mush is required both as the residue of rhyolitic melt separation and as a proximate heat source to thermally sustain the nearly aphyric condition of the overlying rhyolite. A model is presented for a unitary zoned chamber beneath Mount Katmai.