Thomas P. Miller

Late Quaternary caldera-forming eruptions in the eastern Aleutian arc, Alaska

Late Quaternary calderas have been identified at 12 of 40 volcanic centers in the eastern Aleutian arc, and sufficient radiocarbon dates and geologic information have now been obtained to either date or constrain the timing of the climactic caldera-forming eruptions. At least eight major caldera-forming events, each characterized by estimated eruption volumes of more than 10 km 3 , occurred at seven different volcanic centers in the Holocene, and as many as six of these had estimated eruption volumes of more than 50 km 3 . Eruptions of similar magnitude formed two other calderas in Wisconsin time. The dating of these hitherto little-known events adds significantly to the previously existing chronology of large prehistoric eruptions. This refined chronology is important in understanding eruption-induced climate changes, in assessing volcanic hazards, and in developing a tephrochronology for northwestern North America.

Dome growth and destruction during the 1989-1990 eruption of Redoubt Volcano

Abstract Much of the six-month-long 1989–1990 eruption of Redoubt Volcano consisted of a dome-growth and -destructive phase in which 14 short-lived viscous silicic andesite domes were emplaced and 13 subsequently destroyed. The life span of an individual dome ranged from 3 to 21 days and volumes are estimated at 1 × 10 6 to 30 × 10 6 m 3 . Magma supply rates to the vent area averaged about 5 × 10 5 m 3 / day for most of the dome-building phase and ranged from a high of 2.2 × 10 6 m 3 per day initially to a low of 1.8 × 10 5 m 3 per day at the waning stages of the eruption. The total volume of all domes is estimated to be about 90 × 10 6 m 3 and may represent as much as 60–70% of the volume for the entire eruption. The site of 1989–1990 dome emplacement, like that in 1966, was on the margin of a north-facing amphitheatre-like summit crater. The domes were confined on the east and west by steep cliffs of pre-eruption cone-building volcanic rocks and thus were constrained to grow vertically. Rapid upward growth in a precarious site caused each dome to spread preferentially to the north, resulting in eventual gravitational collapse. As long as the present conduit remains active at Redoubt Volcano, any dome formed in a new eruption will be confined to a narrow steeply-sloping gorge, leading to rapid vertical growth and a tendency to collapse gravitationally. Repetitive cycles of dome formation and failure similar to those seen in 1989–1990 are probably the norm and must be considered in future hazard analyses of Redoubt Volcano.

Geochemistry of the 1989-1990 eruption of redoubt volcano: Part I. Whole-rock major- and trace-element chemistry

Abstract The 1989–1990 eruption of Redoubt Volcano produced medium-K calc-alkaline andesite and dacite of limited compositional range (58.2–63.4% SiO 2 ) and entrained quenched andesitic inclusions (55% SiO 2 ) which bear chemical similarities to the rest of the ejecta. The earliest (December 15) magmas are pumiceous, often compositionally banded, and the majority is relatively mafic ( 2 ). The most silicic magmas of the eruption are the late December to early January domes (up to 63.4% SiO 2 ). Subsequent magmas formed domes and rare pumices which converge on 60% SiO 2 . Chemical variations among ejecta comprise tight, linear, two-component arrays for all elements for which the analytical uncertainty is much less than the compositional range. The two-component arrays are interpreted as mixing arrays between unrelated magmas because several of the arrays are at steep angles to the normal liquid line of descent. Additionally, the felsic endmember cannot be easily related to the mafic endmember by normal high-temperature igneous processes (e.g., the silicic endmember has higher Zr yet lower Hf than the mafic endmember). Also relative enrichments of highly incompatible elements are dramatically different across the arrays. The mixing event must have preceded eruption by a significant, yet unspecified amount of time because groundmass glass compositions are homogeneous for all post-December samples (Swanson et al., 1994-this volume), in spite of the whole-rock chemical diversity. This implies time for additional crystallization after the mixing event. Swanson et al. (1994-this volume) discuss evidence for a potentially different mixing event recorded only in December 15 magmas. Cognate cumulate xenoliths composed of pl+cpx+opx+hb+mt+melt were recovered from January and April deposits. These blocks differ from local batholithic country rock in their low concentrations of incompatible elements (e.g., Rb vs 20–90 ppm, Ba vs 300–2000 ppm) and low SiO 2 ( 60 wt.%). They have Mg, Cr, Ni, Sc, and V contents higher than the andesites, but lower than Redoubt basalts and basaltic andesites. Thus, they may be crystallization products of andesites, but do not represent the cumulate residue of basalt fractionation. The xenoliths were probably derived from a shallow or intermediate crustal chamber.

The 1989–1990 eruptions of Redoubt Volcano: an introduction

Abstract Redoubt Volcano, located on the west side of Cook Inlet in south-central Alaska, erupted explosively on over 20 separate occasions between December 14, 1989 and April 21, 1990. Fourteen lava domes were emplaced in the summit area, thirteen of which were subsequently destroyed. The eruption caused economic losses estimated at over

Potassium-Rich Alkaline Intrusive Rocks of Western Alaska

160,000,000 making this the second most costly eruption in U.S. history. This economic impact provided the impetus for a integrated comprehensive account of an erupting volcano using both modern and classical research and modern techniques which in turn led to advances in eruption monitoring and interpretation. Research on such topics as dome formation and collapse and the resulting pyroclastic flows, elutriated ash, lightning, tephra, and flooding was blended with the rapid communication of associated hazards to a large user group. The seismology successes in predicting and monitoring eruption dynamics were due in part to (1) the recognition of long-period seismic events as indicators of the readiness of the volcano to erupt, and (2) to the development of new tools that allowed the seismicity to be assessed instantaneously. Integrated studies of the petrology of erupted products and volatile content over time gave clues as to the progress of the eruption towards completion.

Geochemistry of the 1989–1990 eruption of redoubt volcano: Part II. Evidence from mineral and glass chemistry

Alkaline intrusive rocks occur in a 300-km-long belt of small plutons and intrusive complexes that extends from west-central Alaska to the Bering Sea. The occurrence of a massif of similar rocks on the easternmost tip of the Chukotsk Peninsula suggests the belt may extend into Siberia. These are highly potassic and subsilicic rocks that were emplaced at about mid-Cretaceous time (Albian and Cenomanian). The belt of alkaline potassic rocks cuts across two different geologic provinces. The eastern half of the belt is in the Yukon-Koyukuk volcanogenic province of Mesozoic age; the western half is in the Seward Peninsula province, which is comprised mainly of metamorphic and sedimentary rocks ranging in age from Precambrian to Paleozoic. The Yukon-Koyukuk province is an unusual setting for alkaline rocks, commonly thought to be typical of stable continental platform or shield areas. The possibility that the volcanic rocks of the Yukon-Koyukuk province rest directly on oceanic crust together with the regional extent of the potassium-rich alkaline rocks suggests that the potassic magma may have originated in the mantle. Confinement of the alkaline rocks to the general boundary area between the two geologic provinces suggests this highly faulted area was a zone of structural weakness along which deep-seated alkaline magma was emplaced. This magma may have influenced the composition of other plutonic rocks that are closely associated in time and space with the alkaline potassic rocks.

Holocene tephrochronology of the Cold Bay area, southwest Alaska Peninsula ☆

Abstract Early stages (December 1989) of the 1989–1990 eruption of Redoubt Volcano produced two distinct lavas. Both lavas are high-silica andesites with a narrow range of bulk composition (58–64 wt.%) and similar mineralogies (phenocrysts of plagioclase, hornblende, augite, hypersthene and FeTi oxides in a groundmass of the same phases plus glass). The two lavas are distinguished by groundmass glass compositions, one is dacitic and the other rhyolitic. Sharp boundaries between the two glasses in compositionally banded pumices, lack of extensive coronas on hornblende phenocrysts, and seismic data suggest that a magma-mixing event immediately preceeded the eruption in December 1989. Textural disequilibrium in the phenocrysts suggests both magmas (dacitic and rhyolitic glasses) had a mixing history prior to their interaction and eruption in 1989. Sievey plagioclase and overgrowths of magnetite on ilmenite are textures that are at least consistent with magma mixing. The presence of two hornblende compositions (one a high-Al pargasitic hornblende and one a low-Al magnesiohornblende) in both the dacitic and rhyolitic groundmasses indicates a mixing event to yield these two amphibole populations prior to the magma mixing in December 1989. The pargasitic hornblende and the presence of Ca-rich overgrowths in the sievey zones of the plagioclase together indicate at least one component of this earlier mixing event was a mafic magma, either a basalt or a basaltic andesite. Eruptions in 1990 produced only andesite with a rhyolitic groundmass glass. Glass compositions in the 1990 andesite are identical to the rhyolitic glass in the 1989 andesite. Cognate xenoliths from the magma chamber (or conduit) are also found in the 1990 lavas. Magma mixing probably triggered the eruption in 1989. The eruption ended when this rather viscous (rhyolitic groundmass glass, magma capable of entraining sidewall xenoliths) magma stabalized within the conduit.

Geological and seismological evidence of increased explosivity during the 1986 eruptions of Pavlof volcano, Alaska

Abstract The major-element glass geochemistry of 92 tephra samples from the southwest Alaska Peninsula provides the basis for establishing a Holocene tephrochronology for the region. Electron microprobe analysis has been combined with field descriptions of samples, stratigraphic relationships between tephra samples and sample localities, and glass shard micro-morphology to correlate these sampled distal tephra units throughout the area of Cold Bay and adjacent Morzhovoi Bay. Radiocarbon dating provides age constraints on correlated horizons. Previous research had clearly delineated only one horizon in the region, the so-called ‘Funk/Fisher’ ash, dating to between 8425±350 and 9130±140 14 C yr BP. In addition to constraining the bimodal andesitic and dacitic glass chemistry of that horizon, this study has recognized six additional tephra layers in the area. Two horizons pre-date the Funk/Fisher ash and four are younger than it. A tephra containing dacitic and andesitic components was identified in the vicinity of Morzhovoi Bay, with a minimum age of 9300±80 14 C yr BP and a maximum age of 10,200±75 14 C yr BP. A rhyolitic horizon composed of cm-sized, rounded pumice clasts was identified in the vicinity of Cold Bay; it has been correlated to the ca 9500 BP eruption of Roundtop volcano on Unimak Island. The four younger tephra beds date to between 6070±340 and 3600±140 14 C yr BP. The oldest of the four is rhyodacitic, followed by a mixed rhyodacitic–andesitic horizon, another rhyodacitic horizon, and finally an andesitic layer. Comparison of all the correlated horizons to proximal samples collected on Unimak Island provides conclusive geochemical evidence that the ca 9100 BP Caldera-forming eruption of Fisher volcano is the source of the Funk/Fisher ash. Correlation between the rhyodacitic tephra horizons and proximal samples from Fisher volcano suggests that Fisher Caldera is the source of one of the rhyodacitic tephra horizons that post-dates the Funk/Fisher ash. Additional tephra samples from the southwest Alaska Peninsula and Unimak Island that were collected prior to this study correlate to the tephra horizons identified in the Cold Bay area and identify one additional horizon.

Eruptive history and petrology of Mount Drum volcano, Wrangell Mountains, Alaska

We present results of study of the best-documented eruptions of Pavlof volcano in historic time. The 1986 eruptions were mostly Strombolian in character; a strong initial phase may have been Vulcanian. The 1986 activity erupted at least 8×106 m3 of feldspar-phyric basaltic andesite lava (SiO2=53–54%), and a comparable volume of wind-borne tephra. During the course of the eruption, 5300 explosion earthquakes occurred, the largest of which was equivalent to an ML=2.5 earthquake. Volcanic tremor was recorded for 2600 hours, and the strongest tremor was recorded out to a distance of 160 km and had an amplitude of at least 54 cm2 reduced displacement. The 1986 eruptions modified the structure of the vent area for the first time in over two decades. A possible pyroclastic flow was observed on 19 June 1986, the first time such a phenomenon has been observed at the volcano. Overall, the 1986 eruptions were the strongest and longest duration eruptions in historic time, and changed a temporal pattern of activity that had persisted from 1973–1984.

Double Glacier Volcano, a 'new' Quaternary volcano in the eastern Aleutian volcanic arc

Mount Drum is one of the youngest volcanoes in the subduction-related Wrangell volcanic field (80x200 km) of southcentral Alaska. It lies at the northwest end of a series of large, andesite-dominated shield volcanoes that show a northwesterly progression of age from 26 Ma near the Alaska-Yukon border to about 0.2 Ma at Mount Drum. The volcano was constructed between 750 and 250 ka during at least two cycles of cone building and ring-dome emplacement and was partially destroyed by violent explosive activity probably after 250 ka. Cone lavas range from basaltic andesite to dacite in composition; ring-domes are dacite to rhyolite. The last constructional activity occured in the vicinity of Snider Peak, on the south flank of the volcano, where extensive dacite flows and a dacite dome erupted at about 250 ka. The climactic explosive eruption, that destroyed the top and a part of the south flank of the volcano, produced more than 7 km3 of proximal hot and cold avalanche deposits and distal mudflows. The Mount Drum rocks have medium-K, calc-alkaline affinities and are generally plagioclase phyric. Silica contents range from 55.8 to 74.0 wt%, with a compositional gap between 66.8 and 72.8 wt%. All the rocks are enriched in alkali elements and depleted in Ta relative to the LREE, typical of volcanic arc rocks, but have higher MgO contents at a given SiO2, than typical orogenic medium-K andesites. Strontium-isotope ratios vary from 0.70292 to 0.70353. The compositional range of Mount Drum lavas is best explained by a combination of diverse parental magmas, magma mixing, and fractionation. The small, but significant, range in 87Sr/86Sr ratios in the basaltic andesites and the wide range of incompatible-element ratios exhibited by the basaltic andesites and andesites suggests the presence of compositionally diverse parent magmas. The lavas show abundant petrographic evidence of magma mixing, such as bimodal phenocryst size, resorbed phenocrysts, reaction rims, and disequilibrium mineral assemblages. In addition, some dacites and andesites contain Mg and Ni-rich olivines and/or have high MgO, Cr, Ni, Co, and Sc contents that are not in equilibrium with the host rock and indicate mixing between basalt or cumulate material and more evolved magmas. Incompatible element variations suggest that fractionation is responsible for some of the compositional range between basaltic andesite and dacite, but the rhyolites have K, Ba, Th, and Rb contents that are too low for the magmas to be generated by fractionation of the intermediate rocks. Limited Sr-isotope data support the possibility that the rhyolites may be partial melts of underlying volcanic rocks.