JAMSTEC > Institute For Research on Earth Evolution > Seminar Schedule > Past seminar information (IFREE 3c&4)

Institute For Research on Earth Evolution

Seminar Schedule

Past seminar information (IFREE 3c&4)

2012

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 29th, November(Thursday)
Speaker: Akira Ishikawa (Invited researcher, IFREE3C)

Title:
Reevaluation of highly siderophile element concentrations and 187Os isotope compositions in geological reference materials

Abstract:
Highly siderophile elements (HSEs: Re, Au, Ir, Os, Ru, Rh, Pt and Pd) and the Pt-Re-Os isotope systems embedded within these elements are recognized as important tracers for understanding the formation and evolution of the Earth, complementary to the more commonly used lithophile elements and isotopes. Recent advances in analytical techniques and equipments enabled us to routinely determine HSE concentrations and Os isotopes for a wide variety of geological samples. However, there are still considerable ambiguities particularly in the determination of HSE abundances, as evidenced by scatter in reported values for reference materials due to lack of a universal method.
In this talk, I will summarize the current problems owing to digestion methods and introduce a renewed analytical protocol that is suitable for small amount of samples (~1 g), with the aim of obtaining accurate 187Os/188Os ratios and HSE concentrations (except for mono-isotopic Rh and Au) from a single aliquot of geological samples by isotope dilution ICP-MS and N-TIMS. The optimum method adopted after intensive tests includes the regular inverse aqua regia attack in Carius tube, followed by a desilicification step with HF, in order to liberate HSEs contained in residual silicates. The importance of the desilicification step for different sample types is examined by comparing the results from replicate analyses of basaltic, ultramafic and sedimentary rock reference materials digested with and without HF. Resultant analytical values for the reference materials, which are compared to literature values, support the previous notion that HF treatment is necessary for some basaltic materials but not for ultramafic and sedimentary samples.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 8th,November(Thursday)
Speaker: Hiroshi Kawabata(IFREE4B)

Title:
Origin of enrich mantle-1: Constraints from Pitcairn alkali basalts

Abstract:
Enrich Mantle-1 (EM1) is one of mantle end-members, which is characterized by lower 143Nd/144Nd than bulk silicate Earth and 206Pb/204Pb as low as depleted MORB source mantle. The EM1 was originally explained by mantle peridotite involving small amount of ancient subducted pelagic sediments. After that other continental materials (lower crust) and sub-continental lithospheric mantle were also favored as potential precursor of EM1. However, involvement of these recycled continental materials was often claimed, because the EM1 ocean island basalts lack the positive Pb and the negative Nb-Ta anomaly as observed in continental crust. At this stage, there are no models to explain consistently both trace element and isotopic ratios of EM1 alkali basalts.
This presentation summarizes the geochemical characters of ocean island basalts with EM-1 isotopic signature, and then discusses the origin of EM-1 using the petrographical and geochemical data of alkali basalts from Pitcairn Island, South Pacific. Pitcairn alkali basalts show wide isotopic variations ranging from EM1 to FOZO-like signature. Arc-like trace element signature is more significantly observed in EM1 basalts compared to FOZO basalts, suggesting the involvement of continental materials in the source of EM1 basalts. However, simple mixing of recycled ancient continental materials with peridotite fails to explain consistently the key geochemical characters of Pitcairn EM1 basalts: 1) Ba and Th enrichment compared to U and Nb, 2) Nb/U and Ce/Pb ratios similar to modern MORB, and 3) low 143Nd/144Nd ~0.51250. I suggest that incipient melts from depleted MORB source mantle are also essential recycled materials together with ancient continental materials in order to explain the geochemical characters of EM1.

● IFREE MC seminar
Place: Seminar Room at 1F, YOKOSUKA HQ
Time: 16:30-18:00 on 25th, Octorber(Thursday)
Speaker: Yoshihiko Tamura(IFREE 4A)

Title:
Mission Immiscible: two subduction components result in two primary magmas of Pagan Volcano, Mariana arc

Abstract:
Pagan is an island volcano located in the volcanic front of the Central Island Province (18°10'N/145°45'E) and is one of the largest volcanoes in the Mariana arc, having a volume of 2,160 km3 (Bloomer et al., 1989). Pagan has a maximum elevation of 570 m (Mt. Pagan), but its submarine flanks descend to depths of 2,000-3,000 m, and therefore most of the volcano lies below sea level. Until recently, little was known about the submarine parts of the volcano. ROV Hyper-Dolphin dives and bathymetric surveys in the submarine flanks of Pagan were carried out during NT10-12 (R/V Natsushima) in July 2010.

There are no systematic differences between subaerial and submarine lavas at > 52 wt % SiO2, suggesting that both lavas came from the same Pagan magmatic system. The least fractionated 20 magnesian basalts (48.5-50 wt % SiO2) recovered from the NE flank (HPD1147) extend to higher MgO(10-11 wt %) and Mg# (66-70), than have ever been sampled from Pagan island lavas. Moreover, the Fo contents of olivine (up to Fo94) and Cr-number of spinels (up to 0.8) suggest that these magmas formed from high degrees of mantle melting.

There are two geochemical groups of magnesian basalt lavas, having > 10 wt % MgO. All magnesian lavas contain clinopyroxene and olivine phenocrysts, thus, these two groups are referred to as COB1 and COB2. TiO2, FeO, Na2O, K2O, and all incompatible trace elements are lower in COB1 than COB2, suggesting that COB1 formed from higher degrees of mantle melting, which is consistent with lower Nb/Yb of COB1 lavas. The COB2 have LREE-enriched REE patterns and higher Th/Nb but COB1 show contrasting LREE-depleted patterns and have lower Th/Nb. On the other hand, Ba/Nb and Ba/Th are higher in COB1 than those in COB2. The higher Ba/Th and lower Th/Nb of COB1 and the contrasting opposite of COB2 indicate that main subduction addition in COB1 is slab-derived fluid component, but that in COB2 is sediment melt components. Importantly, the subduction addition, which caused higher degrees of melting of the COB1 source, is mostly hydrous fluid. Sr-Nd-Pb-Hf isotopes are also consistent to this scenario.

In conclusion, COB1 is enriched in aqueous fluid component and COB2 is enriched in sediment melt component, both of which have erupted almost at the same time and only 500 m apart. Why and how do these different primary magmas coexist and be neighbored in the same volcano? Hydrous fluid could not be miscible with sediment melt in the depth of the subducting slab below the volcanic front, which might have resulted in two neighboring, but completely different primary magmas.

● IFREE MC seminar
Place: Seminar Room at 1F, YOKOSUKA HQ
Time: 17:00-18:30 on 11th, October (Thursday)
Speaker: Dr. Sylvie Demouchy (CNRS, Geosciences Montpellier,France)

Title:
Solubility and Diffusivity of Water in the Deep Earth : Lab, Field and Consequences

Abstract:
Rocks and minerals from Earth's mantle show that the deep interior of our planet remains a potentially significant reservoir of 'water'. Moreover, the most recent seismological studies show that a low velocity zone is located just above the discontinuity at 410‐km depth. This zone is sometime interpreted as linked to the presence of hydrous melt. If geophysical observations indeed suggest the presence of water in the deep Earth, the 'water' quantities are still poorly constrained. Moreover, since the mantle is depleted in volatiles by volcanism, it must be re‐hydrated by deep subduction in order to keep an hydrous deep Earth.
This presentation will provide (1) recent higthlights on the 'water' storage capacity ' of the deep Earth, (2) but also constrain on the mobility of such 'water' in mantle material and (3), results on how much 'water' is contained in mantle-derived peridotites, based on the most recent studies of upper mantle xenoliths from South African, Eifel (Germany), Ichino-megata and Oki-dogo (Japan), and Malaïta (Ontong-Java Plateau). Finally, implications for physical and chemical properties on mantle rock and geodynamics will be discussed and remaining important questions on the field will be tackled.

● IFREE MC seminar
Place: Seminar Room at 1F, YOKOSUKA HQ
Time: 16:30-18:00 on 4th, October (Thursday)
Speaker: Toshihiro Suzuki (IFREE 4C)

Title: Trace element partitioning between Fe-Ni Alloy and sulfide melt under high pressure

Abstract:
Knowledge of the partitioning behavior of elements between solid and liquid metal is fundamental for resolving the evolution of the metallic core of the terrestrial planets.
Hence we performed high pressure melting experiments of Fe-Ni-S system at 10 and 25 GPa, and measured the partitioning coefficient (D) of elements. We synthesized
Fe-Ni(95:5) alloy doped with 13 trace elements (Co, Cu, Ge, Mo, Ru, Rh, Pd, W, Re, Os Ir, Pt, Au) in approximately 150 ppm, by arc-melting method. A small chip of this alloy and a small amount of FeS powder were packed in the MgO capsule, and high pressure melting experiments were performed using Kawai-type multi-anvil press installed at Tokyo Institute of Technology. It is well known that D-values of this system are strongly depends on S-content of melt phase (e.g., Chabot & Jones, 2003). Hence we performed several experiments at the same pressure condition with different S contents, and checked the compositional dependence in D-values. Samples were polished and major element compositions were measured by EPMA. Trace element abundances were determined by fs-laser ablation system with sector-type ICP-MS installed at Kyoto University. Van Orman et al.
(2008) pointed out that the D-values of Os, Re and Pt will decrease with pressure, based on their high pressure melting experiments. However, we cannot found any apparent pressure dependences in the D-values of these elements. The present experimental results are suggesting that D-values of Ge, Pd and Au increase with pressure, and W decrease with pressure.
Obvious changes with pressure in D-values were not found in the other elements.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 27th, September(Thursday)
Speaker: Takashi Miyazaki(IFREE 4B)

Title: Barium stable isotope analysis by double-spike MC-ICP-MS: Preliminary Results

Abstract:
Barium has seven isotopes, 130Ba, 132Ba, 134Ba, 135Ba, 136Ba, 137Ba and 138Ba. Except for 138Ba, whose amount is increased by the electron capture of 138La (λ138,EC = 4.44×10-12 yr-1), all are stable isotopes, and their abundances are only affected by mass dependent isotope fractionation. Mineral precipitation, diffusion, sorption-desorption, biological interaction, etc. may induce mass dependent fractionation of stable barium isotopes. If incoming and/or outgoing materials in subduction-hotspot system have detectable barium isotope variation, it is expected to be a tool to constrain the mobile barium behavior in the subduction-hotspot system. However, barium isotope fractionation studies of terrestrial samples are scarce relative to the studies of meteorite samples. Only a few recent studies analyzed the barium isotopes of natural minerals and experimental barium precipitates. For example, δ137/134Ba values of these minerals and precipitates are -0.5~0 ‰ and -0.3~0 ‰, respectively (δ137/134Ba =((137Ba/134Ba)sample/(137Ba/134Ba)standard-1)*1000 relative to the barium standard solution; Allemen et al., 2010). We established a stable barium isotope analysis method using 130Ba-135Ba double spike MC-ICP-MS, in order to clarify the problem described above. Using our method, the δ137/134Ba value from repeated measurements of IAEA-CO-9 was 0.010 ± 0.067 ‰ (2sd, n=8). The reproducibility of δ137/134Ba value obtained in our method is small enough relative to the δ137/134Ba range of natural minerals and experimental barium precipitates analyzed by Allemen et al. (2010). In this seminar, I will talk about the analytical method using 130Ba-135Ba double spike MC-ICP-MS in detail.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 17:00-18:30 on 30th, August (Thursday)
Speaker: Tatsuo Nozaki(IFREE 3C)

Title: Os isotope stratigraphy of Fe-Mn crust: its growth rate and growth hiatus

Abstract:
We report the temporal and depth variations in the growth rate of Fe-Mn crusts based on the Os isotope dating method.
Our samples were collected systematically from the Takuyo Daigo Seamount (#5 Takuyo Smt) in the northwestern Pacific Ocean using ROV HPD3000.

The Os isotope compositions of three Fe-Mn crust samples collected from 1424, 1440 and 2987 mbsl exhibit a similar trend and their values are almost the same as the seawater Os isotope record from present to 12 Ma, demonstrating that the growth rates of Fe-Mn crusts from the #5 Takuyo Smt are constant regardless of sea depth. This is consistent with the growth rate determined by the Be-10 dating method.
However, in the older part (bottom part in the side of basement rock) of Fe-Mn crust collected from 1424 and 1440 mbsl, the Os isotope compositions have a large deviation from the seawater Os isotope record. Moreover, the Fe-Mn crust collected from 2987 mbsl totally lacks the section older than 12 Ma. These results suggest that (1) the existence of the growth hiatus older than 12 Ma, recognized in the previous studies (Klemm et al., 2005; Li et al., 2008; Meng et al., 2008) or (2) the bottom part of Fe-Mn crust older than 12 Ma at 2987 mbsl was simply eroded and disappeared by land slide of the #5 Takuyo Smt.

We also report the secular variation in the Os concentration and isotope ratio from the Middle Miocene to present using Fe-Mn crust samples collected from the #5 Takuyo Smt. Both the Os/Fe ratio and Os isotope ratio increased exponentially toward present. Based on the flux calculation using a simple box model, these increase tendencies can be almost explained by the increase of riverine Os flux to the ocean and the present seawater Os mass is estimated to be ca. 1.85 times the size of that in 10 Ma. Therefore, the seawater Os mass may not be constant through the Earth's history and a Fe-Mn crust is one of the most appropriate materials to unravel the chemical evolution of the paleo-seawater.


● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 23th, August (Thursday)
Speaker: Iona McIntosh (Durham University, UK)

Title: (Mis)understanding bubble growth in magma: Evidence from preserved volatile concentrations in glass

Abstract:
Volcanic eruptions are driven by the nucleation and growth of bubbles that are formed when volatile species (particularly water) exsolve from magma. The kinetics of bubble formation and growth exert a critical influence on eruption dynamics, particularly on the explosivity of the eruption. Bubble growth history is recorded in the concentration profile in the surrounding magma, which is preserved when quenched to glass. Using the new technique of Humphreys et al (2008) we use SIMS-calibrated backscatter SEM images to extract such water concentration profiles around bubbles in synthetic volcanic glasses. These data can be used to test current models of bubble growth, and our findings highlight the potential for significant misinterpretation of both experimental and natural samples.

Preliminary results from FTIR work currently being conducted at IFREE, JAMSTEC will also be presented. These data show variation in water speciation around bubbles in volcanic glass. Such water speciation data provide insights into mechanisms of bubble growth and are expected to provide a method by which bubble growth history can be correctly interpreted in both experimental and natural samples.

Humphreys et al (2008) EPSL 270, 25-40..

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 13th, July (Friday)
Speaker:Marco Brenna (Massey Univ. New Zealand)

Title: Development of an intraplate basaltic volcanic field, Jeju Island, South Korea

Abstract:
Dispersed volcanism in intraplate settings produces volcanic fields that may erupt over millions of years to produce hundreds to thousands of scoria and spatter cones, tuff cones and maars, as well as lava flows. Many aspects of this globally ubiquitous process are poorly known, ranging from the tectonic drivers to the mechanisms controlling magma accumulation and ascent. To investigate magma generation processes leading to a variety of individual eruption types at volcanic fields and to understand the spatio-temporal evolution of these whole-systems, knowledge of the geochemical and petrological properties of erupted products must be linked to the geologic and tectonic framework. This study was based on detailed stratigraphic sampling of small- (<0.01 km3) and large-volume (>1 km3) eruptive sequences in the Jeju Island Volcanic Field, Korea at both individual exposed eruption centres and from deep drill cores. Pyroclastic and lava samples were analysed for whole-rock major-, trace-elements and Sr-Nd-Pb isotopes.
The Jeju magmatic system started with small-volume alkali basaltic eruptions sourced at mantle depths equivalent to c. 2.5 GPa in partially hydrous peridotite. These magmas passed through the crust and erupted rapidly, with minor modification. Intrusions and eruptions accommodated regional tectonic strain, and excess melts became stalled to fractionate toward trachyte compositions in both the lower and upper crust. Trachyte erupted sporadically, with the first episode at c. 750 ka. After this, the system started to erupt with volumetric rates two orders of magnitude higher. This accelerated magma production involved alkali basalt melts derived from greater depths/pressures (3.5 GPa) than earlier, along with subalkali basalts derived from c. 2.5 GPa. Despite prevailing extensional tectonics in the Ryukyu Volcanic Arc and strain accommodation at Jeju, further magmas accumulated and evolved to trachyte compositions at lower crustal depths and erupted in a second episode c. 25 ka ago.
Small-volume eruptions of rapidly rising primitive alkali basalt also continued throughout the life of the field. The geochemical data collected revealed how seemingly simple monogenetic eruptions can be fed by complex and distinct magmatic entities. The same was valid for the entire field, where magma source and evolution conditions vary over time. The variety in volcanic activity is a function of magma types influenced by prior mantle modification events, as well as local and distal tectonic stresses and strain arrangements. Ultimately the site and spatial pattern of melting and melt-production rate determine the final surface morphology, elevation and spatial distribution of magma types in a volcanic field.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 21st, June (Thursday)
Speaker:Takeshi Hanyu(IFREE 4B)

Title: Pb isotope evolution of the HIMU reservoir; implications to recycling of U and Th in the mantle

Abstract:
Geochemical heterogeneity in ocean island basalts and mid-ocean ridge basalts documents the presence of several mantle reservoirs. HIMU is one such mantle reservoir defined by very radiogenic Pb isotopes, and it has been considered to be formed by subduction and accumulation of ancient oceanic crust in the deep mantle. The 'extreme' HIMU basalts occur in limited localities at St. Helena in the Atlantic and Cook-Austral Islands in the south Pacific. These lavas exhibit remarkably similar isotopic compositions in terms of Sr, Nd, and Hf isotopes, suggesting uniform geochemical compositions of the HIMU reservoir that exist at different places in the mantle. The exception is 207Pb/204Pb, in which significant difference is confirmed by isotope analyses with both whole-rock and clinopyroxene; the St. Helena lavas show systematically higher 207Pb/204Pb for a given 206Pb/204Pb than the Cook-Austral lavas. This is explained by various formation age of the reservoir. The Pb isotope evolution model demonstrates that portions of the HIMU reservoir for St. Helena and Australs were formed at approximately 2.2 Ga and 1.8 Ga, respectively.
The relationship between 206Pb/204Pb and 208Pb/204Pb reflects time-integrated Th/U (or kappa = 232Th/238U) of the source. Both St. Helena and Austral lavas demonstrate that time-integrated Th/U of the HIMU reservoir is approximately 3.7, which is close to the chondritic Th/U (3.9) and is much higher than Th/U of the present-day MORB and depleted mantle (2.6-2.9). This indicates that the ancient oceanic crust, that is the precursor of the HIMU reservoir, had different Th/U from the modern MORB. Indeed, sub-seafloor alteration and subduction dehydration would decrease and increase Th/U in oceanic crust, respectively, but the net effect would be reduction of Th/U (< 2) in the subducted oceanic crust.
Consequently, the depleted upper mantle at the time when the HIMU reservoir was formed (1.8-2.2 Ga) must have had higher Th/U than at present. This is consistent with the model in which the Archean and early Proterozoic depleted mantle had chondritic Th/U and then the value decreased to the present due to selective recycling of U, relative to Th, from continent back into the mantle.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 7th, June (Thursday)
Speaker: Eiichi Takahashi(Tokyo Institute of Technology)

Title: On the Concentration of Water in Arc Basalts: case study in Izu Oshima, Miyakejima, Fuji and some perspectives

Abstract:
Genesis, transportation mechanism and distribution of water in subduction system is very important to understand various geological phenomena in subduction zones. Origin of subduction zone magma is thought be deeply connected with dehydration of subducting slabs. Because water degassed from magma prior to or during volcanic eruption, concentration of water in arc magma is not well constrained. In this study, we show evidences that basalt magma in volcanic front generally contains several weight percent of water. This is in contrast with previous view on lateral variation of water in Japanese Quaternary volcanoes(Sakuyama, 1979; Aoki et al, 1981). It also contradicts with estimated geographical variation of fluid components in Japanese Quaternary volcanoes by Nakamura et al.(2008) and Nakamura & Iwamori (2010).

Based on high-pressure melting experiments on primitive basalt of Izu-Oshima volcano, Hamada & Fujii (2007) concluded that presence of 3 to 6wt% of water is necessary in order to explain very calcic (˜An90) plagioclase phenocryst. Phase relation of primitive basalt from Ofunato stage of Miyakejima volcano has been studied experimentally (Ushioda et al, 2011) and it is found that about 3wt% of water is necessary in order to explain its phenocryst assemblage (ol + pl) and the calcic plagioclase composition (An90-94). Moreover, Hamada et al. (2011) has established a new method to estimate pre-degassing water content of magma using hydrogen concentration in plagioclase phenocryst.

Using this new technique, water content in main magma chamber of Izu-Oshima volcano prior to 1986 eruption was estimated be ˜5wt%.

Presence of large amount of water in basalt magma is also supported from explosive volcanic eruption style of Fuji volcano (e.g. Hoei sub-plinian eruption of 1707). According to Machida (1977), amount of volcanic ash (tephra) from Fuji volcano may be equal or greater than its volcanic edifice.

This indicates that the explosive erption style of Fuji volcano continued through time and therefore high water content in its basalt magma is a continuous feature.

Basalt magma is less abundant in the volcanoes on Honshu Arc due to the extensive fractionation, magma mixing and crustal melting. Precise estimate of water content in their primitive basalt magma is therefore difficult. However, presence of very high modal amount of plagioclase phenocryst in basalt and basaltic andesite (usually 30˜40 vol%), is a good indication of the presence of large amount of water in these mafic magmas. This is because, degassing of hydrous basalt at shallow magma chamber invariably accompanies crystallization of large amount of plagioclase(e.g., Hamada & Fujii, 2008).

We therefore propose that basalt magma in volcanic front of Izu Mariana Arc (e.g., Fuji, Izu-Oshima, Miyakejima) and those of North Honshu Arc are all wet, may be typically containing 5wt% of water or even higher. This view strongly contradicts with previous works; 1) lateral variation of water content similar to potassium (Sakuyama, 1979; Aoki et al, 1981); 2) nearly anhydrous magma genesis model by Tatsumi et al.(1983) at the volcanic front, and 3) recent estimation of fluid component in magma based on systematics in Nd and Pb isotopes (e.g. Nakamura & Iwamori, 2010). Our model, however, is not inconsistent with a model proposed by Kimura et al.(2010). We will discuss origin of the discrepancy between our model and previous works. We will also emphasize the importance of the large water flux released by hydrous basalt magma at the volcanic front in considering circulation of water in subduction zone.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 10th, May (Thursday)
Speaker: Takashi Nakagawa(IFREE 2C)

Title: Influence of magmatism on mantle-core cooling, surface heat flow, Urey ratio and material circulation in Earth's mantle

Abstract:
Here we present the state-of-the-art numerical modeling on thermo-chemical evolution in Earth's mantle and core, focusing in particular on matching the present-day surface heat flow and 'Urey ratio'. Most of numerical mantle convection models have ignored magmatic heat transport to argue their estimates on heat budget across the Earth's mantle. Regarding with the surface heat flow, the magmatic heat transport could contribute an upper bound of 9% to Earth's present heat flow. The magmatic heat transport may be dominated in first 2.0 billion years for surface heat flow. The mantle cooling has the similar phase to the magmatic heat transport. Since the heat production rate in the mantle inferred from geochemical and geoneutrino analyses has a great uncertainty, the Urey ratio should also have a huge uncertainty. This means that the Urey ratio is a poor constraint for thermal evolution of Earth's mantle.

New synthesized image of mantle dynamics from early to present Earth ('BAM': Basal Melange of both primordial and recycled origins of dense material in the deep mantle at the present day) has been proposed [Tackley, 2012], which has two different compositionally-distinct material coming from early Earth differentiation such as basal magma ocean [Labrosse et al., 2007] and upside-down differentiation [Lee et al., 2010] related to the magmatism in the deep mantle.
Here we present the possibility on new synthesized image of mantle dynamics with multi-component and phase system of numerical mantle convection simulations including magmatism and core cooling. As a consequence of numerical simulations, two styles of BAM-like structure are found at the present age of the Earth. 1. The primordial material is stably existed beneath the recycled basaltic material. 2. Both primordial and recycled materials are completely mixed. They are dependent on the relationship of density contrasts between recycled basalt and primordial material.

In addition, we show global-scale numerical mantle convection simulations with a continental lithosphere that allows to generate the ultra mafic residual beneath edges of continental lithosphere. Incorporating experimental data on the ultra mafic residue as phase change parameters into numerical mantle convection simulations, such a residual material can be transported by cold subjecting slabs with forming compositionally-dense piles above the core-mantle boundary. This behavior of ultra mafic residue from continental lithosphere might be important for explaining large-scale heterogeneous structure in the deep mantle.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 26th, April (Thursday)
Speaker: Junichi Kimura(IFREE 4B)

Title: Origin of matrix effect in determination of major and trace elements using laser ablation inductively coupled plasma mass spectrometry: 27 years after the first LA-ICPMS

Abstract:
An ultraviolet (200-nm) femtosecond laser ablation (200FsLA) sector-field inductively coupled plasma mass spectrometry (SF-ICPMS) system has been installed, tested, and developed at IFREE for major and trace element analyses in anhydrous silicate glasses and minerals. Use of 200FsLA minimized the matrix effect by 50% compared to that by a 193-nm nanosecond excimer LA. The origin of this improvement was identified as the suppression of 'melting point (MP)-induced' element fractionation at the laser ablation site due to a decreased thermal effect by 200FsLA. Sensitivity enhancement in elements with high first ionization energy still remained in the basalt aerosols relative to silica-rich aerosols. This was interpreted to be due to the higher thermal conductivity of the basalt aerosols in the inductively coupled plasma enhancing ionization controlled basically by first ionization energy of an element. We successfully confirmed this by a simulation using Saha's plasma state equation with the analytical data after reduction of the MP-induced fractionation at the laser ablation site. Accurate determination of trace elements (within 5% relative difference from the accepted values in 71% of the data obtained) was achieved for glasses ranging from MPI-DING komatiite to rhyolite, using single basalt glass BHVO-2G as the calibration standard. This seminar lecture also includes an overall review on the instrumentation and the mechanisms of element fractionation in LA-ICPMS for users/consumers of the relevant facilities/data.

● IFREE MC&HQ seminar
Place: International Exchanging Room at 5th Floor, YOKOSUKA HQ
Time: 16:30-18:00 on 12th, April (Thursday)
Speaker: Yuka Masaki(IFREE3C)

Title: Hydrothermal activity in the mid-Okinawa trough and experiment of hydrothermal alteration of dacite

Abstract:
I will talk about two aspects of my research on hydrothermal activity in the mid-Okinawa Trough: 1) heat flow observations within the Iheya-North Hydrothermal Field; and 2) ongoing experimental work that I am conducting to constrain the rates of hydrothermal alteration of dacite in the Okinawa trough.

The Okinawa Trough is an active back-arc basin, located between the Ryukyu Arc-Trench system and the Asian continent. It is considered be forming as a result of rifting of continental lithosphere. The trough is covered with both hemipelagic and volcanic sediments, and a number of hydrothermal sites have been found. My study area, the Iheya-North Hydrothermal Field is, in the mid-Okinawa Trough and is surrounded by the Iheya-North Knoll. Between 2002 and 2008 we obtained 78 heat flow data from 2002 to 2008 in and around the knolls to clarify the spatial scale of the hydrothermal circulation system. In 2010, IODP Expedition 331 "Deep Hot Biosphere", to the region obtained new subseafloor temperature data around the hydrothermal field. I will show the evidence that these data provide for the recharge of the Iheya-North hydrothermal circulation system and the pattern of this recharge.
Now I am working on dacite samples, representative of the mid-Okinawa Trough, to experimentally constrain how the dacite is hydrothermally altered and the rate at which it is altered at 325 °C and 310 bars. I will outline the experimental set up and present the initial results.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 5th, April (Thursday)
Speaker: Heye Freymuth(Bristol Univ.)

Title: Molybdenum isotopes as a novel tracer for fluids in arc magmas

Abstract:
At least two different slab-derived components have been proposed to be present in arc volcanoes: a sediment component and a fluid component derived from the dehydration of altered oceanic crust and/or underlying serpentinites.
Despite a widely held belief of the involvement of these components in the genesis of arc volcanoes [e.g. 1,2], unambiguously attributing geochemical signatures to either component and discriminating them from residual phase mineralogy in the subducting slab remains difficult. I will show stable Mo isotope data for samples from the Mariana arc which are suggested as a novel tracer for the fluid component in arc magmas.

Molybdenum isotopes were measured on basalts from the Mariana arc as well as representative samples of subducting sediments from ODP sites 800, 801 and 802 using a double-spike technique [3]. The sediments are dominantly light in their isotopic composition with δ98/95Mosediments < 0 ‰, consistent with the incorporation of Mo into the sediment under oxic conditions [4,5]. The arc basalts are isotopically heavy and their δ98/95Mo exceeds the range of compositions found for ocean island basalts. They are also enriched in Mo relative to Pr, an element with similar degree of incompatibility during mantle melting. The Mo isotopes in the arc basalts correlate well with Pr/Mo and tracers for fluid addition with fluid-dominated samples having the highest δ98/95Mo, suggesting that not sediments but Mo bearing fluids are the dominant source of Mo in arc magmas and that Mo isotopes can be used as a tracer for their fluid component.

[1] Plank & Langmuir (1998) Chem. Geol. 145, 325–394. [2] Elliott et al. (1997) JGR, 102, 14,991-15,019. [3] Archer et al. (2008) Nature Geosc., 1, 597-600. [4] Anbar & Rouxel (2007), Ann. Rev. Earth Planet. Sci., 35, 717–746. [5] Siebert et al. (2003) EPSL 211, 159–171.

● IFREE MC seminar
Place: Seminar Room, Administration Bldg. 1F, Yokosuka HQ
Time: 16:30-18:00 on 29th, March (Thursday)
Speaker: Haruka Ozawa(IFREE 4C)

Title: Chemical analysis on recovered sample from Earth's core pressure in a laser-heated diamond-anvil cell

Abstract:
In 1952, Birch first proposed that the Earth's core contains light element(s) such as S, O, Si, C, or H in addition to iron and nickel evident from the density difference between the preliminary reference Earth model (PREM) and pure iron at the relevant conditions. Although considerable research attention has been paid to thermodynamic and physical properties of iron-light element alloys at high pressures to understand the composition of the Earth's core, the identification of light element(s) in the core still remains uncertain. The solid inner core has grown by crystallization of the liquid core, and thus the solid and liquid should be thermodynamically equilibrated at the inner core boundary (ICB). Because the fraction of light elements between solid and liquid iron must be consistent with the seismologically determined density jump at the ICB, the melting phase relations in iron-light element systems at the relevant pressure can provide essential information to identify light elements and constrain their amounts. I would like to examine the melting phase relations in iron-light element alloys at the ICB pressure (329GPa) by combining high-pressure and -temperature experiment with chemical analysis on the recovered sample by transmission electron microscope (TEM). The material synthesized at ICB pressure in a laser-heated diamond-anvil cell is extremely small (20 μm in diameter and <1μm in thickness), and therefore chemical analysis by TEM is necessary. Focused ion beam technique will be used for thin foil preparation for TEM observation. I will talk about these experimental techniques and the actual status.

● IFREE HQ&MC seminar
Place: Conference Room, Administration Bldg. 1F, Yokosuka HQ
Time: 15:30-17:30 on 22nd, March (Thursday)
Speaker: Dr. Sebastien Pilet (University of Lausanne, Switzerland)

Title: Which Source(s) for Intraplate Basalts?

Abstract:
Intraplate lavas from oceanic and continental settings range from highly nepheline (ne)-normative alkaline rock types such as nephelinites and basanites to hypersthene (hy)-normative tholeiites and are characterized by fractionated incompatible trace-element contents that are higher than those found in mid-ocean ridge basalts (MORB). Although aspects of the high trace element contents of intraplate basalts can be explained by low degrees of partial melting of a garnet-peridotite source, the TiO2 contents of OIBs suggest that the sources of these lavas have TiO2 concentrations higher than those estimated for primitive mantle [1], which in turn suggests that other incompatible minor and trace elements may also be enriched in these sources. Such inferences for magma sources with traceelement enrichments relative to the primitive mantle are consistent with the need for trace-element enriched components in the sources of OIB to explain the range of their radiogenic isotopic compositions.

The question that I will discuss in this seminar is how the sources of alkaline rocks acquire these enrichments. Recycled oceanic crust, with or without sediment, is often invoked as a source component of alkaline magmas to account for their trace-element and isotopic characteristics.
Nevertheless, melting behaviour predicted for such material embedded in an upwelling plume suggests that recycled oceanic crust is not sufficiently enriched in trace element to explain by melting the composition observed in alkaline rocks [2, 3].
Alternatively, the source enrichment can be attributed to metasomatic process in the lithospheric mantle. Melting experiments on natural amphibole-rich veins [4] demonstrate that moderate to high degrees of partial melting of metasomatic veins can reproduce key major- and trace-element features of alkaline magmas.
These results are consistent with metasomatized lithosphere as a viable source for alkaline basalts, and, emphasize the need to understand how such veins form.

Amphibole-rich veins have been interpreted as cumulates formed by crystallization of low-degree melts of the underlying asthenosphere as they ascend through the lithosphere [5]. To constrain this mechanism and clarify the nature and origin of the initial metasomatic agent, we performed a series of high-P experiments simulating differentiation of such low-degree melts within continental or mature oceanic lithosphere to determine whether the evolution of such melts would produce cumulates similar to the assemblages of metasomatic veins [6]. We also conducted Monte Carlo simulations of vein formation to evaluate the trace-element budgets of these cumulates [7]. These complementary studies show that, in terms of both major- and trace-elements, hydrous cumulates formed in this way are suitable source regions of alkaline magmas and that such cumulates can be produced by high-P fractional crystallisation of low-degree melts from "normal" mantle (i.e., the formation of such veins does not require that the initial near-solidus melt have unusual compositional characteristics).

Given these results, we propose the following model for the generation of alkaline magmas: (1) low-degree melts from the convecting mantle percolate and differentiate within the lithospheric mantle, producing metasomatic veins plus cryptic enrichments in the surrounding peridotite; (2) subsequent melting of these hydrous cumulates characterized by low solidus temperatures generates ne-normative alkaline liquids (e.g., nephelinites, basanites); the interaction of these alkaline melts with lithospheric peridotite causes them to evolve toward less ne-normative compositions (e.g., alkali olivine basalts) thus explaining the compositional range of alkaline rocks observed in intraplate volcanoes worldwide.

[1] Prytulak and Elliott (2007) Earth Planet. Sci. Lett. 263, 388-403.
[2] Ito and Mahoney (2005) Earth Planet. Sci. Lett. 230, 29-46.
[3] Stracke et al. (2003) G3 doi:10.1029/2001GC000223.
[4] Pilet et al. (2008) Science 320, 916-919.
[5] Harte et al. (1993). Phil. Trans. Royal Soc. London, A 342, 1-21.
[6] Pilet et al. (2010) Contrib. Mineral. Petrol. 159, 621-643.
[7] Pilet et al. (2011) J. of Petrol. 52, 1415-1442.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 8th, March (Thursday)
Speaker: Toshiro Takahashi(IFREE 4B)

Title: Petrological study of Quaternary Chokai volcano - generation and evolution process of TH and CA suites

Abstract:
Tholeiitic (TH) and Calc-alkaline (CA) series lavas is coexisting at Quaternary volcanos in NE Japan arc.
Generally, CA lavas have evidence of magma mixing, e.g.
diseqilibrium phenocrystic assemblage such as olivine and quartz. Therefore, many previous works discussed that the former is evolved from mantle-derived basalt magma by fractional crystallization, and the latter is generated by magma mixing between basic and acidic magmas, both derived from the common TH basalt through fractionation. However, based on Sr isotope micro-analysis of plagioclase phenocrysts in lavas from Zao and Azuma volcanos at central area of NE Japan arc, Tatsumi et al. (2008) and Takahashi et al. (submitted) argued that isotopically radiogenic TH basalt was formed by melting of the lower-crustal amphibolite and CA was formed by magma mixing of the unradiogenic mantle-derived basalt, the radiogenic crust-derived basalt and the differentiated magma which relates to basalt magmas. Additionally, although estimated primary CA basaltic melt composition at Zao and Azuma volcanos is in frontal-arc, it is medium-K basalt and closely similar in composition to back-arc side basalt at Chokai volcano. This result is not in agreement with traditional across-arc variation model for mantle-derived basalt magma in island-arc magmatism (e.g. Kuno, 1966), and need reconsideration of the mechanism for geochemical across-arc variation in island-arc volcanic lavas.
The Quaternary Chokai volcano is located at the rear-arc side of NE Japan arc, and this is typical of stratovolcano in Chokai volcanic zone. Chokai volcano activity is divided into Stage1 to Stage3 (Hayashi, 1984: Ban et al., 2001).
Stage 1 lavas has not disequilibrium texture or rarely has plagioclase phenocryst which has dimly dusty zone. Stage 2 is composed largely of olivine two-pyroxene andesite with a small amount of olivine two-pyroxene basalt. Most of them contain hornblende as phenocryst. Stage 3 is olivine two-pyroxene andesite and two-pyroxene andesite. The almost plagioclase phenocryst in Stage 2 and 3 lavas has dusty zone and sieve texture. An% of plagioclase phenocrysts core in Stage 1 lavas generally shows monomordal distribution
(basalt: An80-90, andesite: 50-60), whereas these in Stage 2 lavas have wide range or bimordal distribution (An50-80). Chokai lavas are plotted on boundary of high-K and medium-K on the SiO2 vs. K2O diagram. On the FeO*/MgO vs. SiO2 diagram, trend of Stage 1 and Stage 2 & 3 lavas show the TH and CA, respectively. Additionally, some Stage 1 lavas are plotted in Basaltic trachyandesite to trachyandesite area on the SiO2 vs. total alkali diagram (High alkali TH). The range of bulk Sr isotope ratio of TH and High alkali TH (Stage 1), CA (Stage 2 and 3) are very similar (TH: 0.70303 to 0.70341, CA: 0.70297 to 0.70342). Sr isotope ratio of TH and High alkali TH are constant, whereas CA is ascent with increasing SiO2. Pb isotope ratio of TH is also constant, but TH and High alkali TH trends are parallel to each other(e.g. TH: 208Pb/204Pb=38.33, High alkali TH: 38.37).
Whereas, Pb isotope ratio of CA is distinctly ascent with increasing SiO2 (38.26 to 38.33).
Petrographical and petrological feature of Chokai volcanic lavas indicate that TH and High alkali TH are produced by fractional crystallization from basic magma, but Pb isotope ratio of the High alkali TH primary basic magma is higher than TH primary magma. On the other hand, CA is formed by magma mixing between basic (unradiogenic Sr and Pb) and felsic magma (radiogenic Sr and Pb). Sr and Pb isotopic features indicate that CA primary basic magma is mantle-derived basaltic magma, whereas it is surmised that primary basic magma of TH and High alkali TH are subject to the contribution of crustal melt. High alkali TH have most enriched Pb isotopic feature and relatively high Zr/Hf, and this has a possibility that primary basic magma of High alkali TH strongly shows the geochemical feature of crust-derived magma by melting of lower crustal amphibolite.
Next we need to study the matter further by Sr isotope miro-analysis of plagioclase phenocryst and groundmass in Chokai lavas.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 9th, February (Thursday)
Speaker: Yuka Masaki(IFREE3C)

Title: Hydrothermal activity in the mid-Okinawa trough and experiment of hydrothermal alteration of dacite

Abstract:
I will talk about two aspects of my research on hydrothermal activity in the mid-Okinawa Trough: 1) heat flow observations within the Iheya-North Hydrothermal Field; and 2) ongoing experimental work that I am conducting to constrain the rates of hydrothermal alteration of dacite in the Okinawa trough.

 The Okinawa Trough is an active back-arc basin, located between the Ryukyu Arc-Trench system and the Asian continent. It is considered be forming as a result of rifting of continental lithosphere. The trough is covered with both hemipelagic and volcanic sediments, and a number of hydrothermal sites have been found. My study area, the Iheya-North Hydrothermal Field is, in the mid-Okinawa Trough and is surrounded by the Iheya-North Knoll. Between 2002 and 2008 we obtained 78 heat flow data from 2002 to 2008 in and around the knolls to clarify the spatial scale of the hydrothermal circulation system. In 2010, IODP Expedition 331 "Deep Hot Biosphere", to the region obtained new subseafloor temperature data around the hydrothermal field. I will show the evidence that these data provide for the recharge of the Iheya-North hydrothermal circulation system and the pattern of this recharge.

 Now I am working on dacite samples, representative of the mid-Okinawa Trough, to experimentally constrain how the dacite is hydrothermally altered and the rate at which it is altered at 325 °C and 310 bars. I will outline the experimental set up and present the initial results.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 26th, January(Thursday)
Speaker: Ryoko Senda(IFREE 4B)

Title: How to determine trace elements in bulk rocks~ Acid digestion vs. Alkali fusion

Abstract:
To understand geological environments and systems, we use a lot of information from elements and isotopes. In these days, chemical analysis methods are developed and it becomes rapidly with high sensitivity. However, to select suitable analytical method which includes dissolution, extraction and measurement of the elements is important to get highly reliable data for each samples and for your purpose.
There are mainly two options for dissolving bulk rock samples for trace element analysis in IFREE4. One is acid digestion and another is fusion with flux. Acid digestion method is dissolved the powdered rock samples using HF with other acids, such as HClO4, HNO3, HCl, etc. HF cuts the bond of Si-O of silicate minerals most of which compose rocks so it is efficient to dissolve many kinds of rocks. On the other hand, alkali fusion method is the method of melting the sample powder with alkaline reagent like Na2CO3 and B4Li2O7 at around 1000℃. This method is a valuable method for rocks with hardly acid-soluble minerals like Zirconium, Monazite, Chromite and so on. I will show the detailed methods of them and compare them from the data of the reference rocks.

● IFREE MC seminar
Place: Meeting Room at 3F, YOKOSUKA HQ
Time: 16:30-18:00 on 12nd, January(Thursday)
Speaker: Yuka Hirahara(IFREE 4B)

Title: Ce isotopic compositions from Quaternary volcanic rocks in northeastern Japan arc: Determination of Ce isotope ratio using the three-step chemical separation procedure and thermal ionization mass spectrometry (TIMS)

Abstract:
The Rare earth elements (REE) have similar chemical, physical properties and exhibit a gradual change in the ionic radius therefore they are the most useful tool for the studies of geological process. The REE have two long-lived decay systems, the 147Sm-143Nd and 138La-138Ce decay systems. A combination of Ce and Nd isotope (143Nd/144Nd) ratios has provide us information on ages, initial ratios, and enable estimation of the time-integrated LREE evolution of complex geological reservoirs in the mantle or crust.

However, the variation of Ce isotopic ratio due to secular radiogenic decay to 138Ce is small because of the low abundance of 138La and its long half-life. For example, the total variation in reported global oceanic samples show less variation than 3 units of epsilon Ce. Therefore, highly precise determination of Ce isotopic ratios is necessary to obtain valuable data for application of geochemical studies.

We have reported a price determination method of Ce isotope using the three-step chemical separation procedure and TIMS (TRITON TI®, Thermo Fisher Scientific Co., Germany) equipped with nine Faraday cups at IFREE, JAMSTEC. The analytical result of Ce isotopic ratio in JMC 304, were obtained 0.0225733 ± 2 (2σ, N=67). The Ce isotope measurements with normalization to JMC 304 in BCR-1 were 0.0225644 ± 4 (2σ, N=7), the results agree well with reported values (0.0225652 ± 4 (N=7), Makishima & Nakamura, 1991).

In northeastern Japan, there are numerous Quaternary volcanoes are distributed. The volcanic rocks have a spatial variation of Sr-Nd-Pb-Hf isotopic compositions (e.g. the rocks form volcanic front have a more enriched isotopic compositions than those of the rear arc side), and a wide range of isotopic compositions could be explained by contamination of heterogeneous crust materials. However, the geochemical information of crustal materials is poor. We estimated the contaminated crustal materials using Ce isotopic ratios in NEJ volcanic rocks.