Geochronological and geochemical studies were performed on alkali volcanic rocks from Oki-Dogo Island, locating about 60 km off the coast of Southwest Japan inside the Japan Sea. There are big differences both in age and in geochemistry between the oldest alkali rocks and the younger rocks; the former having significant Nb and Ta-deletion and Ba and K₂O-enrichment erupted at 19–18 Ma when the opening of the Japan Sea was ongoing, while the latter with no Ta-depletion were active between 5.5–0.6 Ma considerably after the cessation of the rifting activity. Syn-extensional rocks should not have erupted in the active subduction zone but are probably originated from or at least contaminated by the lithosphere beneath the early Proterozoic crust. The lithosphere is considered to have suffered mantle metasomatism to form phlogopite selectively enriched in Ba and K₂O by fluids, which had been released from previous subductions at the Asian continental margin. Partial meting of such lithospheric mantle during the active rifting was essential for the generation of Ba and K₂O-enriched alkali basalt magmas. The younger basaltic rocks are probably originated either from the asthenosphere or the deeper plume mantle. A suite of volcanic rocks from mugearite through trachyte to alkali rhyolite gave concordant ages of 5.5–5.4 Ma. These ages suggest that a previously reported Rb-Sr whole rock age of 6.8 ± 0.3 Ma does not represent the age of eruption and is probably disturbed by magma mixing between basic and acidic end members.

抄録全体を表示

In the San'in district (back-arc side of the Southwest Japan arc), violent subalkalic magmatism took place related to the opening of the Japan Sea during the Miocene. In the northeastern part of Shimane Peninsula, thick middle Miocene submarine volcanigenic strata crop out. This middle Miocene volcanism took place along with intense subsidence probably related to rifting in a mature arc. In the late Miocene, this volcanism and subsidence were followed by eruption of subaerial alkalic basalt (Matsue basalt). Thirty seven samples of these igneous rocks have been analyzed by XRF and INAA. Middle Miocene igneous activity is divided into three stratigraphic stages, and each stage shows a wide variation in chemistry. Intra-stage chemical variations result from crystal differentiation, whereas variations between Stages II and III rocks are interpreted in terms of various degrees of melting from a common source mantle. Stage I rocks were generated from a more depleted mantle. Most middle Miocene rocks have an arc signature, which resulted from the melting of subarc lithospheric mantle which had been metasomatized by a subduction-related fluid. Associated felsic rocks were formed by magmatic fractionation of basaltic magma.

抄録全体を表示

On the basis of Sr and Nd isotopic data for Tertiary to Quaternary volcanic rocks, the isotopic characteristics of the upper mantle beneath the NE Japan arc and genesis of Middle Miocene bimodal volcanism are discussed. Basaltic rocks from the back-arc side, erupted from 35 Ma to present, can be divided into two different groups based on initial Sr isotopic ratios (Sri), initial εNd values, and ages. One group has higher Sri ratios (∼0.7040–∼0.7058) and lower εNd values (∼0–∼+3) than the other (∼0.7029–∼0.7040 and ∼+3–∼+8). Basaltic activity of the high Sri group took place continually from 30 to 10 Ma, whereas activity of the other group occurred mainly in two periods; i.e., at around 34–35 Ma and after –15 Ma. A similar pattern is also observed in basaltic rocks from the transitional zone and the trench side, where basaltic activity characterized by higher Sri ratios took place continuously during ca. 15–0 Ma, whereas eruption of lower Sri basalts occurred at around 15 Ma and after about 3 Ma. One interpretation of these features is that the sub-continental mantle beneath the NE Japan arc may be layered; comprising upper isotopically undepleted (Sri = ∼0.7040–∼0.7058, εNd = ∼0–∼+3) and lower depleted (Sri = ∼0.7029–∼0.7040, εNd = ∼+3–∼+8) parts. Basaltic magmas with an undepleted isotopic signature are considered to have formed from the upper layer, while those with a depleted isotopic signature are from the lower layer. The sharp increase in basaltic activity with a depleted isotopic character after –15 Ma in the back-arc side may be due to the upwelling of hot depleted layer from a deeper region in the mantle (asthenosphere) to the upper undepleted layer beneath the back-arc side of the NE Japan arc during the opening of the Japan Sea. This asthenospheric upwelling might have also caused melting of the lower crust to produce the Middle Miocene acidic volcanic rocks in the back-arc side and the transitional zone. Rhyolitic rocks from Tobishima Island and the Tsugawa-Tadami area in the back-arc side show significantly higher Sri ratios (0.7055–0.7072) and lower εNd values (∼0–∼-5) than the associated basaltic rocks, indicating a lower crustal origin for these rhyolitic rocks. On the other hand, the similarity in εNd values (∼+4) of dolerites and rhyolites from the Kakudate-Rokugo area of the transitional zone suggests derivation from a common mantle source with an isotopic signature between the upper undepleted and lower depleted layers.

抄録全体を表示

The transition from tholeiitic to alkali basalts, which is well established in the Hawaiian series, is exposed in numerous other oceanic islands. Two case studies are presented here, which illustrate the more general geochemical evolution with time from “subalkaline” to “alkaline” compositions: Fangataufa atoll (French Polynesia, Pacific Ocean) and Kerguelen archipelago (T.A.A.F., South Indian Ocean). At Fangataufa atoll, drillings by the French C.E.A. have reached the basaltic bedrock under coral cap rock formations. Two basalt types have been cored: tholeiites form the submarine volcanic sequence, while alkali basalts constitute the top of the submarine zone and the whole subaerial sequence. Tholeiites contain less than 0.9 wt% K₂O with Na₂O/K₂O ranging mostly between 4 and 8, while alkali basalts contain up to 1.8 wt% K₂O with Na₂O/K₂O ranging mostly between 2 and 4. Both suites underwent weak differentiation effects. In Kerguelen archipelago, two Miocene (26–6 Ma) magmatic episodes are recorded in the southeastern province (Ronarc'h and Jeanne d'Arc peninsulae). Lower Miocene mildly alkaline basalts (0.4–2.1 wt% K₂O and Na₂O/K₂O ranging between 2 and 4) are overlain by Upper Miocene highly alkaline basalts (1.4–3.7 wt% K₂O and Na₂O/K₂O between 1 and 2). Lower Miocene magma type evolved towards trachyte compositions, whereas trachy-phonolite and phonolite compositions constitute the Upper Miocene residual melts. The thin transition zone with interstratified tholeiitic and alkali basalts, which is well exposed at Hawaii, was not observed in both Fangataufa and Kerguelen. These two case studies confirm that alkali contents increase with time within oceanic island basalts. Present data indicate that one single source is evidenced in the short-lived Fangataufa oceanic island, while more than one source was tapped in the long-lived Kerguelen magmatic system.

抄録全体を表示

Harra El-Jabban is the Jordanian part of the large intra-continental basalt terrain of Harra Ash Shamah. It covers an area of about 11, 400 sq. km and lies in NE-Jordan. It consists of six successive thick basalt flows, dotted with tephra volcanoes of Oligocene-Holocene age. This Harra is one part of a large number of very poorly studied Cenozoic basalt fields, that are known to exist from Yemen through Saudi Arabia, Jordan, Syria, and Turkey. Detailed investigations of two volcanic cones, namely, Jebal Aritain and Jebal Fahem, aligned along a fissure system trending NNW-SSE in the middle part of the Harra, indicate that they consist of scoria air-fall deposits, dominated by lapilli. They are reasonably classified as “cinder cones” of Strombolian type of volcanicity. The petrographical and geochemical data show that they are of scoriaceous glassy olivinephyric basalts, with average modal olivine (∼23 vol%) derived from primary magmas giving magnesian chrysolite (Fo₈₉) with NiO contents of 0.33–0.47 wt% and CaO > 0.1 wt%. The chemical composition of 20 fresh pyroclastic samples taken from the two cones is similar in contents of the major oxides and trace elements with limited variations. Their chemical trend suggests that the parental magmas ascended rapidly, and underwent only limited fractional crystallization, without significant crustal contamination. The primary nature of this magma is reflected by the high MgO content (>8 wt%), the high Mg-number (0.69–0.75), the low silica content (<50 wt%) and the high Cr and Ni contents of 147–353 ppm and 180–341 ppm, respectively. This conclusion is further supported by the occurrence of mantle-derived ultramafic xenoliths. The similarity in major and trace element composition of the two suites suggests that the magmas were derived from a single homogeneous mantle source. It is estimated that the magmas were generated by low degrees of partial melting (3–8%) of garnet peridotite at >100 km depth.

抄録全体を表示