Unzen Research Articles

Juvenile volcanic glass erupted before the appearance of the 1991 lava dome, Unzen volcano, Kyushu, Japan

Fugen-dake, the main peak of Unzen volcano, began to erupt on 17 November 1990 after 198 years of dormancy. Following a period of steam-dominated tephra eruptions, a new lava dome appeared at the bottom of Jigokuato Crater in the summit region on 20 May 1991. Non-hydrated vesiculated glass was detected in the precursory tephra eruptions, beginning 12 February 1991. Determination of whether glass pyroclasts are hydrated or not appears to be an effective means of ascertaining whether the tephra contains juvenile material from ascending magma. The detection of juvenile pyroclasts at Fugen-dake more than 3 months before the appearance of the lava dome suggests that monitoring of early tephra emissions is an effective way to help forecast impending magmatic eruptions.

Co-ignimbrite ash-fall deposits of the 1991 eruptions of Fugen-dake, Unzen Volcano, Japan

Fugen-dake, the main peak of Unzen Volcano, began a new eruption sequence on November 17, 1990. On May 20, 1991, a new lava dome appeared near the eastern edge of the Fugen-dake summit. Small-scale, 10 4–10 6 m 3 in volume, Merapi-type block and ash flows were frequently generated from the growing lava dome during May–June, 1991. These pyroclastic flows were accompanied by co-ignimbrite ash plumes that deposited ash-fall deposits downwind of the volcano. Three examples of co-ignimbrite ash-fall deposits from Unzen pyroclastic flows are described. The volume of fall deposits was estimated to be about 30% by volume of the collapsed portions of the dome that formed pyroclastic flows. This proportion is smaller than that described for other larger co-ignimbrite ash-fall deposits from other volcanoes. Grain size distributions of the Unzen co-ignimbrite ash-fall deposits are bi-modal or tri-modal. Most ashes are finer than 4 phi and two modes were observed at around 4–7 phi and 9 phi. They are composed mainly of groundmass fragments. Fractions of another mode at around 2 phi are rich in crystals derived from dome lava. Some of the fine ash component fell as accretionary lapilli from the co-ignimbrite ash cloud indicating either moisture or electrostatic aggregation. We believe that the co-ignimbrite ash of Unzen block and ash flows were formed by the mechanical fracturing of the cooling lava blocks as they collapsed and moved down the slope. These ashes were entrained into the convective plumes generated off the tops of the moving flows.

Overview of the 1990–1995 eruption at Unzen Volcano

Following 198 years of dormancy, a small phreatic eruption started at the summit of Unzen Volcano (Mt. Fugen) in November 1990. A swarm of volcano-tectonic (VT) earthquakes had begun below the western flank of the volcano a year before this eruption, and isolated tremor occurred below the summit shortly before it. The focus of VT events had migrated eastward to the summit and became shallower. Following a period of phreatic activity, phreatomagmatic eruptions began in February 1991, became larger with time, and developed into a dacite dome eruption in May 1991 that lasted approximately 4 years. The emergence of the dome followed inflation, demagnetization and a swarm of high-frequency (HF) earthquakes in the crater area. After the dome appeared, activity of the VT earthquakes and the summit HF events was replaced largely by low-frequency (LF) earthquakes. Magma was discharged nearly continuously through the period of dome growth, and the rate decreased roughly with time. The lava dome grew in an unstable form on the shoulder of Mt. Fugen, with repeating partial collapses. The growth was exogenous when the lava effusion rate was high, and endogenous when low. A total of 13 lobes grew as a result of exogenous growth. Vigorous swarms of LF earthquakes occurred just prior to each lobe extrusion. Endogenous growth was accompanied by strong deformation of the crater floor and HF and LF earthquakes. By repeated exogenous and endogenous growth, a large dome was formed over the crater. Pyroclastic flows frequently descended to the northeast, east, and southeast, and their deposits extensively covered the eastern slope and flank of Mt. Fugen. Major pyroclastic flows took place when the lava effusion rate was high. Small vulcanian explosions were limited in the initial stage of dome growth. One of them occurred following collapse of the dome. The total volume of magma erupted was 2.1×10 8 m 3 (dense-rock-equivalent); about a half of this volume remained as a lava dome at the summit (1.2 km long, 0.8 km wide and 230–540 m high). The eruption finished with extrusion of a spine at the endogenous dome top. Several monitoring results convinced us that the eruption had come to an end: the minimal levels of both seismicity and rockfalls, no discharge of magma, the minimal SO 2 flux, and cessation of subsidence of the western flank of the volcano. The dome started slow deformation and cooling after the halt of magma effusion in February 1995.

Crustal deformation in the mid-May 1991 crisis preceding the extrusion of a dacite lava dome at Unzen volcano, Japan

Tilt observations at FG1 station located about 680 m west of the central crater of Unzen volcano, southwestern Japan, revealed details about the process of magma ascent during the mid-May 1991 crisis, just before the appearance of a lava dome. Crustal deformations measured during this period can be interpreted by a combination of upward growth of a buried magma column and lateral intrusion of a dike, modeled by a vertical line source and a planer tensile dislocation in an elastic half space. Marked change in tilt was observed at first in the EW direction at FG1 during May 11–14, suggesting the ascent of a magma column with a diameter of about 40 m from about 300 m to 160 m in depth beneath the central crater. After the interruption of the rise of the magma column, magma began to intrude laterally as a dike on May 15 toward approximately the N80°W direction. This resulted in a large tilt change essentially in the NS component at FG1. The horizontal extent of the dike might have exceeded 400 m in a few days with a thickness of about 6 m. On May 17, the magma column started to rise again, probably because the impediment to magma ascent was fractured by the growth of the dike. The top of the magma column finally reached to the crater bottom probably on May 19, and a new lava dome was actually observed on May 20. After which, the remarkable upward tilt to the south at FG1 due to the dike intrusion decreased rapidly. The deformation after May 20 can be interpreted by a gradual increase in the thickness of the dike up to 13 m by the end of May 1991. Although large errors are inevitable in estimated values because of a lack of sufficient data, the present results confirm the importance of making continuous observations of crustal deformation near the summit areas of active volcanoes in order to clarify the underground processes of magma movement leading to eruptions.

Magma ascent beneath Unzen Volcano, SW Japan, deduced from the electrical resistivity structure

The resistivity structure of Unzen Volcano has been revealed by extensive magnetotelluric surveys since the first eruption on November 17, 1990. This structure comprises a highly resistive surface layer, a low-resistive second layer at several hundred meters depth, interpreted as a water-saturated layer, a resistive third layer, and a low-resistive fourth layer at 10 km depth, possibly related to the deep magmatic activity. The structure has influenced the volcanic activity of Unzen. This activity was characterized by a series of dramatic changes in eruption type: a minor phreatic eruption on November 17, 1990; phreatic eruptions after February 12, 1991, preceded by several weeks of volcanic tremor; phreatomagmatic eruptions after April 9, and dome effusion beginning May 19, 1991. This paper presents a hypothesis in which the top of the magma column rose about 20 m/day, reached the base of the water-saturated layer at the end of January, 1991, and approached the upper boundary of this layer on April 9. Thus, the temporal change of eruption type and associated phenomena are systematically explained by an interaction between magma and groundwater contained in the saturated layer.

Ground temperature change observed at Unzen Volcano associated with the 1990–1995 eruption

Ground temperature associated with eruptive activity of Unzen Volcano, southwestern Japan, was observed in a cave about 680 m west of the central vent. Based on the data during 1991–1996, eliminating the effects of seasonal change, the temperature rose to the highest level probably around the first half of 1992. Hereafter the temperature decreased gradually year by year at least up to the autumn of 1996, beyond the level recovered at the beginning of the lava extrusion in May 1991. Although it is not necessarily conclusive, the present observation suggests a possible rise in ground temperature prior to the commencement of the first phreatic eruption in November 1990. If so, a careful observation may help to predict future eruptions by detecting a precursory rise of the ground temperature in this volcano.

Petrology and Fe–Ti oxide reequilibration of the 1991 Mount Unzen mixed magma

A dacitic magma (64.5 wt.% SiO 2), a mixture of phenocryst-rich rhyodacite and an aphyric mafic magma, was erupted during the recent 1991–1995 Mount Unzen eruptive cycle. The experimental and analytical results of this study reveal additional details about conditions in the premixing and postmixing magmas, and the nature of the mixing process. The preeruption rhyodacitic magma was at a temperature of 790±20°C according to Fe–Ti oxide phenocryst cores, and at a depth of 6 to 7 km (160 MPa) according to Al-in-hornblende geobarometry. The mafic magma that mixed with the rhyodacite is found as andesitic (54 to 62 wt.% SiO 2) enclaves in the erupted magma and was essentially aphyric when intruded. Phase equilibria indicate that an aphyric andesite at 160 MPa is >1030°C (H 2O-saturated) and possibly as high as 1130°C (2 wt.% H 2O). The composition of the rhyodacite which was mixed with the andesite is estimated to lie between 67 and 69 wt.% SiO 2. Using these compositions and temperatures, the temperature of the Unzen magma after mixing is estimated to be at least 850° to 870°C. The groundmass Fe–Ti oxide microphenocrysts and those in pargasite-bearing reaction zones around biotite phenocrysts both give 890±20°C temperatures; the oxide–oxide contacts give temperatures of 910±20°C. The 900±30°C postmixing temperatures are consistent with phase-equilibria experiments which show that the magma was not above 930°C at 160 MPa. Our Fe–Ti oxide reequilibration experiments suggest that the mixing of the two magmas began within a few weeks of the eruption, which is a shorter time than is calculated using available diffusion data. There is also evidence that some mixing took place much closer to the time of extrusion based on the presence of unrimmed biotite phenocrysts in the magma.

Generation of block and ash flows during the 1990–1995 eruption of Unzen Volcano, Japan

Processes generating block and ash flows by gravitational dome collapse (Merapi-type pyroclastic flow) were observed in detail during the 1990–1995 eruption of Unzen volcano, Japan. Two different types were identified by analysis of video records and observations during helicopter flights. Most of the block and ash flows erupted during the 1991–1993 exogenous dome growth stage initially involved crack propagation due to cooling and flowage of the dome lava lobes. The mass around the crack became unstable, locally decreasing in tensile strength. Finally, a slab separated from the lobe front, fragmented progressively from the base to the top within a few seconds, and became a block and ash flow. Rock falls immediately followed, in response to local instability of the lobe front. Clasts in these rock falls fragmented and merged with the preceding flow. In contrast, block and ash flows during the endogenous dome growth stage in 1994 were generated due to local bulge of the dome. Unstable lava blocks collapsed and subsequently fragmented to produce block and ash flows.

Petrology of the 1991–1995 eruption at Unzen: effusion pulsation and groundmass crystallization

Effusive eruption of dacite magma (2.1×10 8 m 3) during 1991–1995 formed a lava dome at the summit of Unzen Volcano, Japan. The effusion rate was highest at the beginning, 4.0×10 5 m 3/day (4.6 m 3/s), and decreased roughly with time, to almost zero before this pattern was repeated with a second pulse of magma supply. The whole-rock chemistry of lavas shows significant variation attributable to variations in phenocryst abundance; the more mafic, the more abundant the phenocrysts. The pattern of chemical variation with time shows some difference from that of the effusion rate. All phenocrysts in dacite (plagioclase, hornblende, biotite, quartz and magnetite) show evidence of disequilibrium with melt. Although a glomerophyric aggregation of phenocrysts suggests coexistence with each other, phenocrysts are isotopically heterogeneous from species to species. The calculated initial melt composition was rhyodacite, and was nearly constant throughout the activity. In contrast, the bulk phenocryst population is andesite. A model explaining the textures and the isotopic heterogeneity is the capture of diorite fragments (or xenocrysts) by parental rhyodacite magma. It is suggested that, when effusion rate was high, less viscous crystal-poor magma exited from the reservoir. Groundmass glass and plagioclase microlite rims show temporal chemical variations correlating with the effusion rate; the higher the effusion rate, the more evolved the compositions. Groundmass crystallinity increased with decreasing effusion rate; from 33% to 50%. Textures in dome lavas suggest that groundmass crystallization had been mostly completed when magma reached the conduit top. The Fe–Ti oxide temperature (880–780°C) was low when the crystallinity was high. Micropumice erupted before dome growth provided a sample recording magmatic foam in the conduit. Porosity of dome lavas was lower at lower effusion rates. Collapse of foam magma and simultaneous escape of volatiles through the conduit top were probably responsible for the accompanying low-frequency earthquakes. Phenocrysts were broken and the breakdown rims on hornblende phenocrysts were torn off during collapse and successive compaction. When effusion waned, degassing and the resultant crystallization proceeded more completely, so that the magma became too viscous to flow in the conduit top and behaved as a plug, resulting in a temporary halt of effusion. In turn, groundmass crystallization in magma below the plug increased excess pressure in the upper parts of conduit due to slow cooling. The plug was scavenged when rising excess pressure overcame its effective strength. Then, the second pulse of magma supply began. Strong endogenous growth and extrusion of a lava spine in the later stage probably occurred for the same reason.

Water contents and hydrogen isotopic ratios of rocks and minerals from the 1991 eruption of Unzen volcano, Japan

Water contents and hydrogen isotopic ratios were determined for blocks from pyroclastic flow deposits, and bread-crust bombs and blocks from the 1991 Vulcanian eruptions of Unzen volcano, Japan. Groundmass water contents and δD values of samples were calculated by subtracting the contribution of major hydrous minerals (hornblende and biotite) from the bulk rock analyses, and range from 0.1 to 0.5 wt.% and −83 to −49‰, respectively. The samples do not show a systematic H 2O– δD relationship, although the block samples tend to have lower δD values than the bomb samples. The non-systematic H 2O– δD relationship is likely a result of near surface, kinetically-controlled gas loss. High viscosity of this magma would hinder attainment of hydrogen isotopic equilibrium between exsolved vapor and melt in the final degassing stage. The near surface degassing, however, was accompanied by kinetic fractionation resulting in enrichment of deuterium in the final products as exemplified by bread-crust bombs with high H 2O–low δD margins and low H 2O–high δD cores. Relatively high δD values of the blocks and bombs as well as high temperature volcanic gas (−30 to −35‰) suggest a closed system degassing of an initial water-rich magma (H 2O=6 wt.%) until its water content was reduced to 0.5 wt.%. The pre-eruptive δD value (−46‰) was estimated from the volcanic gas data and D/H analysis of hornblende phenocrysts coupled with assumed isotopic equilibration in the initial hydrous magma.

Groundmass pargasite in the 1991–1995 dacite of Unzen volcano: phase stability experiments and volcanological implications

Pargasite commonly occurs in the dacitic groundmass of the 1991–1995 eruption products of Unzen volcano. We described the occurrence and chemical compositions of amphibole in the dacite, and also carried out melting experiments to determine the low-pressure stability limit of amphibole in the dacite. The 1991–1995 ejecta of the Unzen volcano show petrographic evidence of magma mixing, such as reverse compositional zoning of plagioclase and amphibole phenocrysts, and we used a groundmass separate as a starting material for the experiments. Reversed experiments show that the maximum temperature for the crystallization of amphibole is 930°C at 196 MPa, 900°C at 98 MPa, and 820°C at 49 MPa. Compared with the experimental results on the Mount St. Helens dacite, present experiments on the Unzen dacitic groundmass show that amphibole is stable to pressures ca. 50 MPa lower at 850°C. Available Fe–Ti oxide thermometry indicates the crystallization temperature of the groundmass of the Unzen dacite to be 880±30°C, suggesting that the groundmass pargasite crystallized at >70 MPa, corresponding to a depth of more than 3 km in the conduit. The chlorine content of the groundmass pargasite is much lower than that of phenocrystic magnesiohornblende in the 1991–1995 dacite of Unzen volcano, indicating that vesiculation/degassing of magma took place before the crystallization of the groundmass pargasite. The present study shows that the magma was water oversaturated and that the degassing of magma along with magma mixing caused crystallization of the groundmass amphibole at depths of more than 3 km in the conduit.

Deposits associated with the 1990–1995 eruption of Unzen volcano, Japan

The 1990–1995 eruption of Unzen volcano in Japan was characterized by lava dome growth and block-and-ash flows triggered by dome collapse. The repeated dome collapse events produced block-and-ash flow, ash-cloud surge and fine-grained ash-fall deposits. Block-and-ash flow deposits, which are confined to valleys, are massive to reversely graded, very poorly sorted and contain non-vesiculated to poorly vesiculated blocks up to 10 m across. The deposits often display finer-grained basal parts. Coarse-clast-depleted flow units, which have intermediate grain size distribution between main flow and surge deposits, occur as marginal part of block-and-ash flow deposits. Ash-cloud surge deposits, which spread more widely, are finer grained and better sorted than block-and-ash flow deposits. Ash-fall deposits, which were derived from ash clouds of block-and-ash flows and were dispersed around the volcano, are very fine grained and are as well sorted as the surge deposits. Lahars resulting from remobilization of pyroclastic debris devastated the base of the volcano and their deposits are moderately fines-depleted and occasionally display two different lahar facies: debris flow and hyperconcentrated flow units.

火山岩類の熱ルミネッセンス年代測定の精度検証 雲仙火山火砕流堆積物のβ線量精密補正年代

Thermoluminescence (TL) age determination was carried out for 18 samples from four pyroclastic flow deposits(Mutsugi, Kureishibaru, Furue and Yugawachi) of Unzen volcano, SW Japan. The TL growth curve method was used for quartz phenocrysts to determine the ages. Major inaccuracy of TL dating is caused in annual dose evaluation, especially in determination of beta rays contributions. Therefore, the beta ray correction was done by measurement of quartz phenocryst diameter on thin sections. The TL ages for Mutsugi, Kureishibaru, Furue and Yugawachi pyroclastic flow deposits indicated 4.9±0.4 ka, 23±1 ka, 23±1 ka and 77±6 ka, respectively, which are average of 3 to 6 samples collected from same outcrop each. These ages consistent with 14C ages and stratigraphy identified by tephra.

A17 雲仙岳 1991-1995 年噴出物中の角閃石斑晶の累帯構造 : 特に塩素の挙動とマグマの脱ガスについて

A17 雲仙岳 1991-1995 年噴出物中の角閃石斑晶の累帯構造 : 特に塩素の挙動とマグマの脱ガスについて

Block size distributions on silicic lava flow surfaces: Implications for emplacement conditions

We determined block size distributions on the surfaces of Holocene silicic lava flows at the Inyo domes and the Medicine Lake volcano, and studied the development of blocks on the active Mount St. Helens and Mount Unzen lava domes to better understand the emplacement history of young viscous flows. We measured block chord lengths along perpendicular 25 m long transects within vent, jumbled, and ridged morphologic units. Vent regions generally contain the largest average block sizes and largest range of average blocks, whereas ridged areas tend to have the smallest average blocks. Observations at the active Mount St. Helens and Mount Unzen lava domes show that block size distributions reflect stress conditions during flow. High extrusion rates produce small primary blocks and lead to rapid fracturing of the flow surface, whereas low extrusion rates allow large slabs to form in the vent area and lead to less severe fragmentation. A dramatic increase in the size of blocks evident in active vent regions may indicate a significant decrease in eruption rate, and thus could signal the cessation of extrusion. However, if the extrusion rate is too high or the cooling rate too low, a rigid crust and accompanying blocks will not form on an eruptive time scale. Blocks may fracture through mechanical and thermal processes as they move downslope. Most silicic lava flows reach a steady state downslope, where the average block size at the surface remains in the 20–30 cm size range with increasing distance from the vent. Fines (blocks <12 cm) do not accumulate on the flow surface because they slip toward the flow interior through void spaces between surface blocks. We therefore expect long silicic lava flows to have blocky surfaces throughout their lengths, an important consideration for evaluation of planetary lava-flow emplacement.

Satellite thermal analyses of lava dome effusion rates at Unzen Volcano, Japan

This paper presents analyses of infrared remote sensing data of the 1992–1994 lava dome growth episode of Unzen Volcano, Japan. The primary aim is to determine whether such data can indicate trends in the rate of magma supply to the dome, since this is one of the principal factors controlling the frequency of pyroclastic flow. Spaceborne, low‐spatial‐resolution infrared radiance data correctly delimit the two discrete phases of magma supply to the dome, and during phase 1 the short‐wavelength data are found to be strongly correlated with ground‐based estimates of lava effusion rate. The correlation is, however, significantly weaker during phase 2. Analysis of high‐spatial‐resolution data from airborne and satellite based systems allows us to determine the principal source of shortwave infrared flux as fumarolically heated dome surfaces, these being significantly hotter than the actively growing lava lobes. Variations in the low‐spatial‐resolution shortwave infrared signal are therefore directly attributed to fluctuations in the size and temperature of the fumarolically heated area. These fumarolic variations are largely dependent upon the flux rate of magmatic gas, which in turn is partly dependent upon the rate of magma supply. Gas fluxes measured using correlation spectrometer (COSPEC) measurements of SO2 concentration indicate that the relationship between fumarolic discharge and the effusion rate of lava was relatively stable during phase 1 but highly variable during phase 2. We suggest these variations as an explanation for the differing correlations observed between the emitted shortwave infrared flux and the effusion rate of lava during phases 1 and 2. We conclude that appropriate analysis of high‐ and low‐spatial‐resolution shortwave infrared remote sensing data may assist in elucidating trends in lava supply at volcanoes exhibiting similar dome growth episodes to that experienced at Unzen.

An anomaly of ELF band vertical magnetic flux as a precursor of dome formation at Unzen volcano and its model analysis

A peculiar environmental vertical magnetic flux anomaly was observed at the frequency of 223 Hz, which was 40 and 75 days before the 10th and 11th dome formations on Unzen volcano. The small but clear anomaly of order of 0.5 pT/ Hz had appeared just on the vertical magnetic flux and never on the horizontal ones, and showed in many cases an intermittent oscillations of about a half- to 1-min period with almost the same size amplitudes. In this paper, we show a possibility that the vertical anomaly corresponds to the electromagnetic phenomena when magma passes the upper layer of aquifer under a volcano and give a simplified model which represents the phenomena and mechanism of generating the anomalous vertical magnetic flux. The anomaly lasted from the end of 1992 to September 10, 1994. After the termination of the appearance, the Unzen volcano weakened the activity and no events have been reported since then.

Effect of rainfall on dacitic lava dome collapse at Unzen volcano, Japan.

At Unzen volcano, rainfall caused lava dome collapses and pyroclastic flow in some cases. Heavier precipitation increases the probability of dome collapses and pyroclastic flows, and increased pyroclastic flows are correlated with precipitation for certain periods, but not others. Dome collapses and pyroclastic flows were clearly triggered on fresh, and not yet cooled lava, and presumable originates in the instability of the lava dome cracked due to rapid cooling by rainwater.

Responses of Surface Runoff and Sediment Discharge to Revegetation on the Slope of Pyroclastic-flow Deposits, Mount Unzen, Japan

Responses of Surface Runoff and Sediment Discharge to Revegetation on the Slope of Pyroclastic-flow Deposits, Mount Unzen, Japan

Researchers discuss Mt. Unzen, a decade volcano

In November 1989, a swarm of earthquakes deep beneath Tachibana Bay, Kyushu Island, Japan, heralded the inexorable rise of magma toward the summit of Unzen Volcano, some 15 km upward and 15 km eastward, on the Shimabara Peninsula. When the “magma head” emerged in Jigokuato Crater on May 20, 1991, a beautiful but tragic drama began. It started peacefully as a budding flower unfolding lava petals (Figure 1). But by the time lava stopped flowing in February 1995, it had cost the city of Shimabara and the surrounding towns over $2 billion in damage and 44 human lives. At its height, the crisis required the prolonged evacuation of 11,000 residents. Amid this tragedy, however, volcanologists were able to make unprecedented visual and geophysical observations of processes of magma ascent, dome growth, and dome‐fed pyroclastic flows.