This paper presents results of geologic investigation of the eruptive activity in the last 300 years of Shinmoedake, an active volcano in the Kirishima Volcano Group. The recent activity of this volcano is divided into four eruptive episodes : the 1716-1717, 1771-1772, 1822 and 1959 episodes. The most important activity occurred in 1716-1717. During the 1716-1717 eruption, fallout deposits, pyroclastic flows and mudflows were widely dispersed around the volcano. The products of this episode show that the eruption progressed with time from phreatic to magmatic. These field data are in good agreement with historic records of eruptive activity. According to the historic records, the eruptive activity lasted from 11 March, 1716 to 19 September, 1717. The 1771-1772 and 1822 activities produced base surges, pyroclastic flows, fallout tephra and mudflows that were confined to the slope and eastern base of the volcano, but historic records do not reveal the details of these eruptions. The field evidence shows the same phreatic to magmatic sequence as the 1716-1717 activity. However, the eruptions of both episodes were on a smaller scale than the 1716-1717 eruption. The 1959 activity was well described. This episode produced minor gray silty to sandy lithic fallout tephra indicating that only phreatic activity occurred. The fallout was distributed northeast of the vent. In conclusion, the field evidence and historical records show that each eruptive episode of the current activity of Shinmoedake progressed from phreatic to magmatic. The eruptions are frequently accompanied by pyroclastic flows and mudflows.

抄録全体を表示

High magnesian olivines of Quaternary basalts and Ichinomegata Iherzolites were analyzed to estimate the compositions of primary magmas and residual mantles under Northeast (NE) Japan arc. The olivine phenocrysts from the back-arc side basalts have a trend with higher Fo content (lOO × Mg/Mg + Fe) in Fo-NiO wt% relationship than those from the volcanic front side basalts whose trends appear to be derived from a field of the olivines of Ichinomegata Iherzolites. The residual mantle olivine of the back-arc side primary magma (Fo = 94) is higher in Fo than that of the volcanic front side one (Fo≦90). Initial mantle before melt extraction would also have different bulk compositions in the sources for both magmas. The initial composition for the back-arc side primary magma may be higher in Xmg value (100 × Mg/Mg + Fe) and lower in FeO content than that for the volcanic front side one. The latter, which is similar to the Ichinomegata Iherzolite, may be shallower-seated than the former. It is postulated that the across-arc variations of Xmg and FeO in basalts results from the compositionally-layered mantle wedge under NE Japan arc.

抄録全体を表示

Applications of lanthanum-cerium (La-Ce) isotope systematics in the geochemistry are reviewed in this article. Rare earth elements (REE) behave coherently in the geological phenomena and are strongly partitioned into the liquid phase relative to mafic minerals through the magmatic processes. The incompatibility of REE regularly increases from heavy to light REE in the igneous processes. Therefore, REE have been used extensively along with other incompatible elements to understand igneous petrogenesis. Based on the above characteristics of REE, time-integrated REE fractionation in the mantle and crust have been broadly discussed using Sm-Nd isotope systematics. In 1982, Tanaka and Masuda first observed the variation of ¹³⁸Ce/¹⁴²Ce ratio caused by β^--decay of ¹³⁸La in the terrestrial materials and established a La-Ce isotope systematics as a geochemical tracer and a geochronometer. The time-integrated light REE (LREE) fractionation had been interpreted by the extrapolation from the results of Sm-Nd systematics. The more detailed observation of LREE evolution has been enabled by the introduction of La-Ce systematics. However, the data of Ce isotope analyses have been limited because only a few scientists have been involved in this study. Therefore, further Ce isotope analyses of the terrestrial and meteoritic materials with Nd isotope analyses will provide more detailed information on the LREE evolution in the Earth and the solar system.

抄録全体を表示

Possible evidence for magmatic activity in the crust and uppermost mantle beneath the northeastern Japan arc has been obtained from microearthquake observations. Tomographic images of the 3-D P-wave velocity structure estimated from more than 20,000 arrival time data clearly delineate low-velocity zones, which are continuously distributed from the uppermost mantle to the upper crust beneath active volcanoes. Exceptionally deep (25〜40 km) microearthquakes are found at 9 locations around the low-velocity zones beneath volcanoes, and all these events have extremely low predominant frequencies (1.5〜3.5 Hz for P-waves), suggesting a close relation to the magmatic activity in this depth range. At shallower depths (10〜15 km) in or around the low-velocity zones, there exist distinct reflectors of S-waves from shallow events. Reflection point locations and the calculated reflection coefficients suggest that the reflectors depict the upper surfaces of magma bodies. A detailed seismicity study reveals that focal depths of shallow crustal microearthquakes in the inland of northeastern Japan are shallower than about 15km, with exceptions of a small number of deep (25〜40km) low-frequency events. Cut-off depth for this shallow seismicity, which is sharply delimited, can be understood as a transition from brittle to ductile deformation or a transition from stick-slip to stable sliding due to increasing temperature. The cut-off depth changes with the location and becomes shallow beneath actitve volcanoes. This depth decrease can be attributed to the local elevation of temperature caused by magma coming up from below. Under the tectonic stress field of horizontal compression beneath this volcanic arc, stress concentrations will arise around the regions where the base of the brittle seismogenic zone is locally elevated. This interpretation seems to be supported by the fact that large crustal earthquakes are apt to occur around the low-velocity zones of P-wave. These observations appear to shed some light on the state magma at depths and its relation to shallow seismic activity beneath northeastern Japan.

抄録全体を表示

Typical island-arc volcanoes (10-100 km³ in volume) have typical life span of 20-60 × 10⁴yrs. Both their eruption rates and magmatic temperature tend to decrease during their life span. In order to explain above features of typical arc volcanoes, a series of numerical calculations have been carried out on cooling of a partially molten mantle diapir that has intruded and settled beneath the Moho. It is assumed that length of the volcanic life span may correspond with the time in which underlying mantle diapir cools from 1300℃ to 1200℃ and solidifies. A non-steady 3-dimensional thermal conduction equation was used for the calculation, in which the effects of thermal convection and melt segregation were neglected. The effect of the latent heat (10% partial melt) was taken into account so as to increase the effective initial temperature of diapir by 50℃. According to the present calculation, diapirs with initial volume of 1000 and 5000 km³ are found to solidify within 20 × 10⁴ yrs and 60 × 10⁴ yrs, respectively. It is thus suggested that volumes of cooling mantle diapirs that are compatible with life span of typical island-arc volcanoes are considered to be in the range between 1000 and 5000 km³. This estimation is consistent with other geological and petrological observations on Japanese Quarternary volcanoes.

抄録全体を表示

The magma system of Moriyoshi volcano, northeastern Japan is discussed, according to geological, petrographical and geochemical data. More than 90% of the total eruptive products are andesite and dacite, accompanied by a small amount of basalt. Judging from major and trace element chemistry and Sr isotopic ratios, the basaltic magma cannot be parental to the andesite and dacite by a process of fractional crystallization. In addition, the lack of correlation between the Sr isotopic ratios and silica and the absence of basaltic andesite indicates that assimilation of basaltic magma with crustal material did not take place. The calculated magmatic temperature, abundance of incompatible elements (LIL, HFS and REE), and the LIL/HFS ratios of the andesite and dacite systematically change with time. The most plausible model which accounts for the observed results involves hydrous crustal anatexis, whereby the basaltic magma played an important role in supplying heat for crustal melting. During the initial stage, hydrous partial melts may have been generated by consumption of hydrous minerals within the source crustal materials. Following migration of the melt, the solidus temperature of the source would have increased. The variability in the solidus temperature and the hydrous state of the source region would control not only the magmatic temperature but also the stability of accessory phases, containing HFS and REE. Thus both abundance and ratio of those incompatible element wolud change.

抄録全体を表示

A petrologic model concerning generation and evolution processes of coexisting low-alkali tholeiitic and calc-alkaline magmas at Adatara volcano is proposed. For tholeiitic suite, major-element variations are reproduced by addition-subtraction models of phenocrystic phases. Also, the variations of trace- and rare earth-elements and Sr isotopic compositions are consistent with the fractionation hypothesis. There are mineralogical observations supporting this hypothesis including : (1) positive correlations of Fe/Mg (Ca/Na) ratios of mafic phenocrysts (plagioclase) with those of the host lavas ; (2) increase of Usp content in the core of magnetite phenocrysts with increasing SiO₂ of host lavas ; and (3) olivine disappearance in the lavas containing > 55% SiO₂. In contrast, similar fractionation models to those used for the tholeiitic suite revealed that compositional variations for calc-alkaline suite are not explained solely by fractional crystallization ; other processes such as (1) intermittent mixing of magnesian magmas, (2) assimilation, and/or (3) incorporation of liquid which has suffered fractionation of some heavy-REE-enriched mineral (s) may have operated as additional processes. Also, mineralogical data are compatible with this view. Tholeiitic parental magma is distinctively poor in SiO₂ and incompatible elements, low in light-REE/heavy-REE, Rb/Ba, Zr/Nb and high in ⁸⁷Sr/⁸⁶Sr ratios, relative to the calc-alkaline parental magma. Further, the low-alkali tholeiitic magma has evolved under conditions of higher temperature and lower oxygen fugacity than the calc-alkaline magma. Different degrees of melting in a common peridotitic source can successfully predict the differences in incompatible-element concentration levels and light-REE/heavy-REE ratios between the two types of parental magmas, but it is inconsistent with the differences in Rb/Ba, Zr/Nb, or ⁸⁷Sr/⁸⁶Sr. Since lower crust beneath Northeast Japan is believed to have amphibolitic or gabbroic composition with ⁸⁷Sr/⁸⁶Sr as low as 0.7030, anatexis of the lower crust and subsequent incorporation of partial melt into mantle-derived basaltic magma may be a possible process to explain high Rb/Ba, Zr/Nb and, possively, low ⁸⁷Sr/⁸⁶Sr in the calc-alkaline parental magma.

抄録全体を表示

Nasu volcanoes, the southern part of the volcanic front of Northeast Japan arc, comprises 10 small volcanic centers. Minamigassan is one of these centers and is located at the southern end of this volcanoes. A petrological model of magmatic proccess of Minamigassan volcano is presented based on the mineralogy and whole-rock chemistry. The eruption products of Minamigassan volcano can be divided into five units ; E-1, E-2, L-1, L-2, and L-3 units from lower to upper. The caldera collapse occured at the beginning of L-1 unit. The E-1 unit belongs to medium-K tholeiite, E-2 to low-K tholeiite, L-1 to medium-K calc-alkaline, L-2 to medium-K tholeiite, and L-3 to medium-K calc-alkaline categories, respectively. L-1 unit is petrologically similar to E-1 unit. Least squares subtraction calculation for major elements and Rayliegh fractionation model calculation for trace elements indicate that the chemical variation within E-1 and E-2 unit can be interpreted by fractional crystallization of phenocrystic minerals. On the other hand calc-alkaline suites, L-1 and L-3 units suffered magma mixing. These units formed by mixing between basic magma which resembles E-1 unit magma and felsic magma which is derived from early stage magma through crustal contamination or crustal melt. Distinct differences in LIL/HFS ratio among E-1 and E-2 units cannot be explained either by the difference in degrees of partial melting of a common source or by any fractional crystallization process. The difference between E-1 and E-2 units have been originated in chemical difference of source material.

抄録全体を表示

Climactic effusions (〜40km³) of tholeiite lavas of Younger Fuji Volcano (Izu-Mariana Arc) occurred between 13000-8000 yBP. The mode of occurrence of these effusions differs greatly from the volcanic phenomena characterizing the Fuji Volcano eruptive history (from 80,000 years ago up to present), mainly represented by small scoria eruptions. Less fractionated and larger amounts of lavas erupted in the beginning and were followed by more fractionated and smaller amounts of lavas. In the more highly fractionated magma, there is a wide variation (0-40 vol %) in the amount of phenocrysts. The bulk rock K₂ O concentration proportionally increases with the increse of the amount of groundmass. Plagioclase phenocrysts coexist with olivine phenocrysts in most cases, and are relatively enriched in comparison with mafic minerals. The maximum size of phenocrystic plagioclase reaches 5-10 mm and is 2 to 3 times larger than that reached in other stages. These facts are consistent with a model that crystal settling in a slowly cooling large magma chamber is the major fractionation mechanism for the climactic effusions. The tholeiite lavas belong to the new magma series, the parental magma of which is slightly richer in incompatible elements than the old magma series. The feeding of the new magma series into the new magma chamber is estimated to have started about 10,000 years prior to the climactic effusions. The delay of the eruptions is due to the difficulty to make a new conduit and allow the growth of the slowly cooling large magma chamber. At the same time, a small magma chamber for the old series magma was still alive and led to intermittent eruptions until about 3,000 yBP.

抄録全体を表示

Evolution of whole-rock chemistry of volcanic rocks in Fuji volcano is presented. Basaltic andesite is predominant during 66,000 to 80,000 y. B. P. ; they are poor in FeO^*, MgO, MnO, K₂O, TiO₂, P₂O₅, and abundant in Al₂O₃ and Na₂O, with slightly high K₂O/TiO₂, Rb/Y, and Zr/Y ratios. Basalts erupted during 22,000 to 66,000 y. B. P. show low FeO^*/MgO ratio (2.0＞), and are lacking in those with high K₂O, TiO₂, P₂O₅, Rb, Ba, and Zr, and low Al₂O₃ ; they are characterized by low K₂O/TiO₂, Rb/Y, and Zr/Y ratios. Eruption products during 10,000 to 22,000 y. B. P., especially in its later stage, are basalt with relatively high FeO^*/MgO, Rb/Y, and Zr/Y ratios, and K₂O, TiO₂, P₂O₅, Rb, Ba, and Zr contents, and low Al₂O₃ ; their chemistry resembles to those after 10,000 y. B. P.. Ejecta during 8,000 to 10,000 y. B. P. are basalt and high in FeO^*/MgO, Rb/Y, and Zr/Y ratios, and K₂O, TiO₂, P₂O₅, Rb, Ba, and Zr contents, and Rb/Y and Zr/Y ratios, and low in Al₂O₃. Volcanic rocks during 3,000 to 8,000 y. B. P. are basalt showing low K₂0/TiO₂, Rb/Y, Ba/Y, and Zr/Y ratios, and K₂O, TiO₂, P₂O₅, Rb, Ba, Y, and Zr contents, and high FeO^*/MgO and Al₂O₃ ; they are rather similar to those erupted during 22,000 to 66,000 y. P. B. in chemistry. Erupted materials during 2,000 to 3,000 y. B. P. are basalt with low K₂O, TiO₂, P₂O₅, Rb, Ba, and Zr contents, and Rb/Y and Zr/Y ratios, and high Al₂O₃ and FeO^*/MgO ; their chemistry are akin to those during 3,000 to 8,000 y. B. P., but slightly enriched in incompatible elements. Eruption products after 2,000 y. B. P. are abundant in K₂O, TiO₂, P₂O₅, Rb, Ba, and Zr, and poor in Al₂O₃, with high FeO^*/MgO, Rb/Y, and Zr/Y ratios ; they are almost similar to basalts erupted during 8,000 to 10,000 y. B. P.. The Fuji volcano is classified into three magma-series on the view point of whole-rock Rb/Y and Zr/Y ratios : Early Ko-Fuji (66,000 to 80,000 y. B. P.), Late Ko-Fuji (10,000 to 66,000 y. B. P.), and Shin-Fuji (present to 10,000 y. B. P.). Basaltic magmas of each series may have probably been derived successively from chemically heterogeneous upper mantle. The high rate of change of source rock chemistry may reflect dynamic process in the upper mantle beneath the arc volcano.

抄録全体を表示

A petrologic model of Izu Oshima volcano is proposed on the basis of petrological investigations of rock samples collected systematically from various stratigraphic levels. Izu Oshima is divided into three groups ; (1) the Senzu group, which is the oldest, (2) the Older Oshima group and (3) the younger Oshima group. All three groups consist of many lava flows and scoria fall deposits. Rock types of the lavas and scoria vary from olivine-basalt through pyroxene-basalt and aphyric basalt to subordinate aphyric or magnetite bearing andesite. SiO₂ contents range from 50 to 58 wt%, although most of the rocks have 50 to 54 wt% SiO₂. They are low-K tholeiite and show typical tholeiitic trend with strong Fe enrichment. The bulk rock Mg# generally tends to decrease from the oldest to the youngest and is concordant with the change of phenocryst assemblages. These variations can be explained by fractional crystallization of olivine, two pyroxenes and plagioclase, and is consistent with experiment at 1 atm under dry conditions. It shows that, in Izu Oshima volcano, the fractionation appears to have taken place in a shallow magma reservoir under almost dry conditions. The first supplied relatively primitive (Mg# = 50 ; modal plagioclase phenocryst is less than 10%) basalt of Izu Oshima volcano exhibit high Al₂O₃ content (～18 wt%). Because estimated H₂O content of the Izu Oshima magma is low(＞0.5 wt%), the high Al₂O₃ character of the magma may be attributed to the Pl-free high pressure fractionation. Between the Older and the Younger Oshima groups, during which time the present caldera was formed, bulk Mg# abruptly increased whereas K₂O increased. Phenocrysts in the rocks of the transitional period show reverse zoning, and eruption volume is large in the earlier stage of the Younger Oshima Group. From these observations, it is inferred that a new, relatively high-Mg primitive magma with higher K₂O was supplied to the magma reservoir of the Izu Oshima volcano during the time of caldera formation.

抄録全体を表示

In chemical composition historical lava flows in Sakurajima volcano have a streight-linear correlation, and progressively change with time from dacitic to andesitic. Plagioclase phenocrysts in them commonly show bimodal compositional distributions with two peaks at An₅₈ and An₅₈. These two peaks stay at the same Positions, independently of progressive change in chemical composition of the host lavas. The An-rich peak grows and the other peak declines with time, suggesting the mixing between dacitic and basaltic magmas and the progressive increase in proportion of the basaltic component. The periodically refilled magma chamber model was improved. The new model consists of upper and lower chambers and a set of cylinder and plug of frustum shape. The upper chamber lies in the middle of the crust and was originally occupied by the dacitic magma. The lower chamber lies near the boundary between the mantle and the crust, which is occupied by the volatile-rich basaltic magma. The cumulate plug subsides into the lower chamber, drives the basaltic magma upward and mixes it with the magma in the upper chamber with intermitten gigantic explosions.

抄録全体を表示