Geochemistry at DSDP Hole 57-439

Dating of a hornblende concentrate by the 40Ar/39Ar method gives an age of 23.4±5.5 m.y. for a dacite boulder from conglomerate in Deep Sea Drilling Project Hole 439. The conglomerate clasts range up to 1 meter in diameter and are nearly monolithologic, suggesting that a nearby former volcano erupted the dacite. The dacite is only 90 km landward from the Japan Trench, whereas modern trench-related volcanoes lie at least 120 km from their trenches. The dacite locality is on strike with and is probably an extension of a magmatic arc on the island of Hokkaido that crosses the Kuril arc at an angle of 65° and which was active 16 to 36 m.y. ago. The part of the former arc landward from the Kuril arc argues against an origin from a leaking subduction zone or from subduction of an active spreading ridge. The part seaward both from the Kuril and Japan arcs weakens an explanation based on migration of a trench-trenchtrench triple junction. The magmatic rocks probably formed along a middle-Tertiary plate boundary that had stepped seaward from a more-landward Cretaceous position. Later, the boundary stepped farther seaward at the Kuril arc and landward again at the Japan arc. If so, the present Japan subduction zone must have consumed most of the strata that had accumulated between it and the earlier trench.

Composition and age of volcanic rocks from the Sinii Utes Depression, Southern Primoye

New data are reported on structure of sections, chemical composition, and age of volcano-sedimentary and volcanic rocks from the Sinii Utes Depression in the Southern Primorye region. The Sinii Utes Depression is filled with two sequences: the lower sequence composed of sedimentary-volcanogenic coaliferous rocks (the stratotype of the Sinii Utes Formation) and the upper sequence consisting of tephroid with overlying basalts. This work considers chemical composition and problems of K-Ar dating of basalts. The uppermost basaltic flow has K-Ar age 22.0±1.0 Ma. The dates obtained for the middle and upper parts of lava flows are underestimated. It is explained by their heating due to combustion of brown coals of the Sinii Utes Formation underlying the lava flow. Calculations show that argon could only partly have been removed from the basalts owing to conductive heat transfer and was lost largely due to infiltration of hot gases in heterogeneous fissured medium. Basaltic volcanism on continental margins of the southern Primorye region and the adjacent Korean and Chinese areas at the Oligocene-Miocene boundary preceded Early-Middle Miocene spreading and formation of the Sea of Japan basin. Undifferentiated moderately alkaline basalts of intraplate affinity developed in the Amba Depression and some other structures of the southern Primorye region and intraplate alkali basalts of the Phohang Graben in the Korean Peninsula serve as indicators of incipient spreading regime in the Sea of Japan. Potassic basalt-trachybasalt eruptions occurred locally in riftogenic depressions and shield volcanoes. In some structures this volcanism was terminated by eruptions of intermediate and acid lavas. Such evolution of volcanism is explained by selective contamination of basaltic melts during their interaction with crustal acid material and generation of acid anatectic melts.

Trace-element geochemistry of tholeiitic basalts from DSDP Hole 55-433C

Hole 433C, a multiple re-entry hole drilled in 1862 meters of water on Suiko Seamount in the central Emperor Seamounts, penetrated 387.5 meters of lava flows overlain by 163.0 meters of sediments. The recovered volcanic rocks consist of three flow units (1-3) of alkalic basalt underlain by more than 105 flows or flow lobes (Flow Units 4-67) of tholeiitic basalt. This study reports trace-element, including rare-earth element (REE), data for 25 samples from 24 of the least altered tholeiitic flows. These data are used to evaluate the origin and evolution of tholeiitic basalts from Suiko Seamount and to evaluate changes in the mantle source between the time when Suiko Seamount formed, 64.7 ± 1.1 m.y. ago (see Dalrymple et al., 1980), and the present day. Stearns (1946), Macdonald and Katsura (1964) and Macdonald (1968) have established that chemically distinct lavas erupt during four eruptive stages of development of a Hawaiian volcano. These stages, from initial to final, are shield-building, caldera-filling, post-caldera, and post-erosional. The lavas of the shield-building stage are tholeiitic basalts, which erupt rapidly and in great volume. The shield-building stage is quickly followed by caldera collapse and by the caldera-filling stage, during which the caldera is filled by tholeiitic and alkalic lavas. During the post-caldera stage, a relatively thin veneer of alkalic basalts and associated differentiated lavas are erupted, sometimes accompanied by minor eruptions of tholeiitic lava. After a period of volcanic quiescence and erosion, lavas of the nephelinitic suite, which include both alkalic basalts and strongly SiO2-undersaturated nephelinitic basalts, may erupt from satellite vents during the post-erosional stage. Many Hawaiian volcanoes develop through all four stages; but individual volcanoes have become extinct before the cycle is complete. We interpret the tholeiitic lavas drilled on Suiko Seamount to have erupted during either the shield-building or the caldera-filling stage, and the overlying alkalic flows to have erupted during either the caldera-filling or the post-caldera stage (see Kirkpatrick et al., 1980).

Geochemistry of Ninetyeast Ridge basaltic glasses

The Ninetyeast Ridge (NER), a north-south striking, 5,000 km long, 77 to 43 Ma chain of basaltic submarine volcanoes in the eastern Indian Ocean formed as a hotspot track created by rapid northward migration of the Indian Plate over the Kerguelen hotspot. Based on the major and trace element contents of unaltered basaltic glasses from six locations along the NER, we show that the NER was constructed by basaltic magma derived from at least three geochemically distinct mantle sources: (1) a source enriched in highly incompatible elements relative to primitive mantle like the source of the 29-24 Ma flood basalts in the Kerguelen Archipelago; (2) an incompatible element-depleted source similar to the source of Mid-Ocean Ridge Basalt (MORB) erupted along the currently active Southeast Indian Ridge (SEIR); and (3) an incompatible element-depleted source that is compositionally and mineralogically distinct from the source of SEIR MORB. Specifically, this depleted mantle source was garnet-bearing and had higher Y/Dy and Nb/Zr, but lower Zr/Sm, than the SEIR MORB source. We infer that this third source formed as a garnet-bearing residue created during a previous melting event, perhaps an initial partial melting of the mantle hotspot. Subsequently, this residue partially melted over a large pressure range, from slightly over 3 GPa to less than 1 GPa, and to a high extent (~ 30%) thereby creating relatively high SiO2 and FeO contents in some NER basalts relative to SEIR MORB.

Major and trace element composition of melts from the Gorely volcanic center, southern Kamchatka

Melt inclusions in olivine and plagioclase phenocrysts from rocks (magnesian basalt, basaltic andesite, andesite, ignimbrite, and dacite) of various age from the Gorely volcanic center, southern Kamchatka, were studied by means of their homogenization and by analyzing the glasses in 100 melt inclusions on an electron microprobe and 24 inclusions on an ion probe. The SiO2 concentrations of the melts vary within a broad range of 45-74 wt%, as also are the concentrations of other major components. According to their SiO2, Na2O, K2O, TiO2, and P2O5 concentrations, the melts are classified into seven groups. The mafic melts (45-53 wt% SiO2) comprise the following varieties: potassic (on average 4.2 wt% K2O, 1.7 wt% Na2O, 1.0 wt% TiO2, and 0.20 wt% P2O5), sodic (3.2% Na2O, 1.1% K2O, 1.1% TiO2, and 0.40% P2O5), and titaniferous with high P2O5 concentrations (2.2% TiO2, 1.1% P2O5, 3.8% Na2O, and 3.0% K2O). The melts of intermediate composition (53-64% SiO2) also include potassic (5.6% K2O, 3.4% Na2O, 1.0% TiO2, and 0.4% P2O5) and sodic (4.3% Na2O, 2.8% K2O, 1.3% TiO2, and 0.4% P2O5) varieties. The acid melts (64-74% SiO2) are either potassic (4.5% K2O, 3.6% Na2O, 0.7% TiO2, and 0.15% P2O5) or sodic (4.5% Na2O, 3.1% K2O, 0.7% TiO2, and 0.13% P2O5). A distinctive feature of the Gorely volcanic center is the pervasive occurrence of K-rich compositions throughout the whole compositional range (silicity) of the melts. Melt inclusions of various types were sometimes found not only in a single sample but also in the same phenocrysts. The sodic and potassic types of the melts contain different Cl and F concentrations: the sodic melts are richer in Cl, whereas the potassic melts are enriched in F. We are the first to discover potassic melts with very high F concentrations (up to 2.7 wt%, 1.19 wt% on average, 17 analyses) in the Kuriles and Kamchatka. The average F concentration in the sodic melts is 0.16 wt% (37 analyses). The melts are distinguished for their richness in various groups of trace elements: LILE, REE (particularly HREE), and HFSE (except Nb). All of the melts share certain geochemical features. The concentrations of elements systematically increase from the mafic to acid melts (except only for the Sr and Eu concentrations, because of active plagioclase fractionation, and Ti, an element contained in ore minerals). The paper presents a review of literature data on volcanic rocks in the Kurile-Kamchatka area in which melt inclusions with high K2O concentrations (K2O/Na2O > 1) were found. K-rich melts are proved to be extremely widespread in the area and were found on such volcanoes as Avachinskii, Bezymyannyi, Bol'shoi Semyachek, Dikii Greben', Karymskii, Kekuknaiskii, Kudryavyi, and Shiveluch and in the Valaginskii and Tumrok Ranges.

Geochemistry of volcanic rocks of ODP Holes 126-792E and 126-793B

The Izu-Bonin forearc basement volcanic rocks recovered from Holes 792E and 793B show the same phenocrystic assemblage (i.e., plagioclase, two pyroxenes, and Fe-Ti oxides ±olivine), but they differ in the crystallization sequence and their phenocryst chemistry. All the igneous rocks have suffered low-grade hydrothermal alteration caused by interaction with seawater. As a result, only clinopyroxenes, plagioclases, and oxides have preserved their primary igneous compositions. The Neogene olivine-clinopyroxene diabasic intrusion (Unit II) recovered from Hole 793B differs from the basement basaltic andesites because it lacks Cr-spinels and contains abundant titanomagnetites (Usp38.5-46.4) and uncommon FeO-rich (FeO = 29%) spinels. It displays petrological and geochemical similarities to the Izu Arc volcanoes and, thus, can be considered as related to Izu-Bonin Arc magmatic activity. The titanomagnetites (Usp28.5-33) in the calc-alkaline andesitic fragments of the Oligocene volcaniclastic breccia in Hole 793B (Unit VI) represent an early crystallization phase. The Plagioclase phenocrysts enclosed in these rocks show oscillatory zoning and are less Ca-rich (An78.6-67.8) than the plagioclase phenocrysts of the diabase sill and the basement basaltic andesites. Their clinopyroxenes are Fe-rich augites (Fs ? 19.4; FeO = 12%) and thus, differ significantly from the clinopyroxenes of the Hole 793B arc-tholeiitic igneous rocks. The 30-32 Ma porphyritic, two-pyroxene andesites recovered from Hole 792E are very similar to the andesitic clasts of the Neogene breccia recovered in Hole 793B (Unit VI). Both rocks have the same crystallization sequence, and similar chemistry of the Fe-Ti oxides, clinopyroxenes, and plagioclases: that is, Ti-rich (Usp25.5-30.4) magnetites, Fe-rich augites, and intensely oscillatory zoned plagioclases with bytownitic cores (An86-63) and labradorite rims (An73-68). They display a calc-alkaline differentiation trend (Taylor et al., this volume). So, the basement highly porphyritic andesites recovered at Hole 792E, and the Hole 793B andesitic clasts of Unit VI show the same petrological and geochemical characteristics, which are that of calc-alkaline suites. These Oligocene volcanic rocks represent likely the remnants of the Izu-Bonin normal arc magmatic activity, before the forearc rifting and extension. The crystallization sequence in the basaltic andesites recovered from Hole 793B is olivine-orthopyroxene-clinopyroxene-plagioclase-Fe-Ti oxides, indicating a tholeiitic differentiation trend for these volcanic rocks. Type i is an olivine-and Cr-spinel bearing basaltic andesite whereas Type ii is a porphyritic pyroxene-rich basaltic andesite. The porphyritic plagioclase-rich basaltic andesite (Type iii) is similar, in most respects, to Type ii lavas but contains plagioclase phenocrysts. The last, and least common lava is an aphyric to sparsely phyric andesite (Type iv). Cr-spinels, included either in the olivine pseudomorphs of Type i lavas or in the groundmass of Type ii lavas, are Cr-rich and Mg-rich. In contrast, Cr-spinels included in clinopyroxenes and orthopyroxenes (Types i and ii lavas) show lower Cr* and Mg* ratios and higher aluminium contents. Orthopyroxenes from all rock types are Mg-rich enstatites. Clinopyroxenes display endiopsidic to augitic compositions and are TiO2 and Al2O3 depleted. All the crystals exhibit strong zoning patterns, usually normal, although, reverse zoning patterns are not uncommon. The plagioclases show compositions within the range of An90-64. The Fe-Ti oxides of the groundmass are TiO2-poor (Usp16-17). The Hole 793B basaltic andesites show, like the Site 458 bronzites from the Mariana forearc, intermediate features between arc tholeiites and boninites: (1) Cr-spinel in olivine, (2) presence of Mg-rich bronzite, Ca-Mg-rich clinopyroxenes, and Ca-plagioclase phenocrysts, and (3) transitional trace element depletion and epsioln-Nd ratios between arc tholeiites and boninites. Thus, the forearc magmatism of the Izu-Bonin and Mariana arcs, linked to rifting and extension, is represented by a depleted tholeiitic suite that displays boninitic affinities.

Sedimentology of ODP Leg 126 deep-water volcaniclastics

Three sites were drilled in the Izu-Bonin forearc basin during Ocean Drilling Program (ODP) Leg 126. High-quality formation microscanner (FMS) data from two of the sites provide images of part of a thick, volcaniclastic, middle to upper Oligocene, basin-plain turbidite succession. The FMS images were used to construct bed-by-bed sedimentary sections for the depth intervals 2232-2441 m below rig floor (mbrf) in Hole 792E, and 4023-4330 mbrf in Hole 793B. Beds vary in thickness from those that are near or below the resolution of the FMS tool (2.5 cm) to those that are 10-15 m thick. The bed thicknesses are distributed according to a power law with an exponent of about 1.0. There are no obvious upward thickening or thinning sequences in the bed-by-bed sections. Spaced packets of thick and very thick beds may be a response to (1) low stands of global sea level, particularly at 30 Ma, (2) periods of increased tectonic uplift, or (3) periods of more intense volcanism. Graded sandstones, most pebbly sandstones, and graded to graded-stratified conglomerates were deposited by turbidity currents. The very thick, mainly structureless beds of sandstone, pebbly sandstone, and pebble conglomerate are interpreted as sandy debris-flow deposits. Many of the sediment gravity flows may have been triggered by earthquakes. Long recurrence intervals of 0.3-1 m.y. for the very thickest beds are consistent with triggering by large-magnitude earthquakes (M = 9) with epicenters approximately 10-50 km away from large, unstable accumulations of volcaniclastic sand and ash on the flanks of arc volcanoes. Paleocurrents were obtained from the grain fabric of six thicker sandstone beds, and ripple migration directions in about 40 thinner beds; orientations were constrained by the FMS images. The data from ripples are very scattered and cannot be used to specify source positions. They do, however, indicate that the paleoenvironment was a basin plain where weaker currents were free to follow a broad range of flow paths. The data from sandstone fabric are more reliable and indicate that turbidity currents flowed toward 150? during the time period from 28.9 to 27.3 Ma. This direction is essentially along the axis of the forearc basin, from north to south, with a small component of flow away from the western margin of the basin.

Strontium and rubidium isotopes in serpentinites from ODP Site 195-1200

Subduction of the Pacific plate beneath the Mariana forearc releases fluids to the overlying mantle wedge that ascend, producing serpentinite "mud" that discharges on the ocean floor. As part of Leg 195 of the Ocean Drilling Program cores were obtained from drill-holes into the mud volcanoes. We report the isotopic composition of Sr in water squeezed from intervals of the cores, in the serpentinite mud, in leaches of the serpentinite mud, and in entrained small harzburgitic clasts. Except in the upper few meters below the seawater-mud interface, where pore water approaches seawater Sr concentration and isotopic ratio, Sr concentration and isotopic composition remain constant at 3-6 µmol/kg and ~0.7054. Because the elemental chemistry of the pore water is unlike seawater, this isotopic composition reflects fluids derived from the subducted slab, probably modified by reaction with mantle material during ascent. Higher Sr isotopic ratios, up to 0.7087, - but not with higher Sr concentrations in pore water - occur superimposed on an advection profile at 13-16 mbsf surrounding a thin layer of foraminiferal sand. Since the upward seepage velocity of slab fluids in the mud volcano vents is a few cm/yr, exchange of Sr between these carbonates and the rising fluids must have occurred within a maximum of a few hundred years, essentially instantaneously given the millions, or tens of millions, of years the mud volcanoes have been in existence. In contrast, the strontium isotopic compositions of leached serpentinite mud, and of small harzburgite clasts entrained in the mud, are always significantly greater than that of the pore water. In small harzburgite clasts the ratio reaches 0.7088, almost as high as the seawater value of 0.7092 and much higher than the value of typical mantle-derived strontium of ~0.704. The serpentinite muds and harzburgite clasts clearly equilibrated with seawater Sr when they were initially deposited at the surface of the seamount, but following burial they have not fully equilibrated with strontium in the pore water now discharging through the vents. These variations in the strontium isotopic composition of solids and pore waters are more consistent with episodic expulsion of fluids in the subduction zone than steady state flow. Whereas strontium in carbonates equilibrates isotopically within a few hundred years, strontium in buried harzburgite clasts does not equilibrate in the same time, assuming steady state rates of upward fluid flow. By inference, the harzburgite clasts and associated serpentinite mud must have been near the seafloor, unburied, for a yet undetermined but much longer period of time to have equilibrated from ~0.704 to 0.709 prior to subsequent burial. It may be possible to characterize at least the periodicity of fluid release in the mud volcano setting by investigating the zonation of strontium isotopic composition of hartzburgite clasts throughout the 60-meter deep composite cores.

Mean chemical characteristics and volcano signatures of ice cores from Belukha, the Everest and three Svalbard sites

Ice cores from outside the Greenland and Antarctic ice sheets are difficult to date because of seasonal melting and multiple sources (terrestrial, marine, biogenic and anthropogenic) of sulfates deposited onto the ice. Here we present a method of volcanic sulfate extraction that relies on fitting sulfate profiles to other ion species measured along the cores in moving windows in log space. We verify the method with a well dated section of the Belukha ice core from central Eurasia. There are excellent matches to volcanoes in the preindustrial, and clear extraction of volcanic peaks in the post-1940 period when a simple method based on calcium as a proxy for terrestrial sulfate fails due to anthropogenic sulfate deposition. We then attempt to use the same statistical scheme to locate volcanic sulfate horizons within three ice cores from Svalbard and a core from Mount Everest. Volcanic sulfate is <5% of the sulfate budget in every core, and differences in eruption signals extracted reflect the large differences in environment between western, northern and central regions of Svalbard. The Lomonosovfonna and Vestfonna cores span about the last 1000 years, with good extraction of volcanic signals, while Holtedahlfonna which extends to about AD1700 appears to lack a clear record. The Mount Everest core allows clean volcanic signal extraction and the core extends back to about AD700, slightly older than a previous flow model has suggested. The method may thus be used to extract historical volcanic records from a more diverse geographical range than hitherto.

Iodine and boron concentrations in pore fluids, ultramafic clasts and interstitial serpentinite mud of ODP Hole 125-779A and Site 195-1200

Iodine and boron were analyzed in pore fluids, serpentinized ultramafic clasts, and the serpentinized mud matrix of the South Chamorro Seamount mud volcano (Ocean Drilling Program Leg 195 Site 1200) to determine the distribution of these elements in deep forearc settings. Similar analyses of clasts and muds from the Conical Seamount mud volcano (Leg 125 Site 779) were also carried out. Interstitial pore fluids are enriched in boron and iodine without appreciable change in chloride concentration relative to seawater. Both the ultramafic clasts and the associated serpentinized mud present the highest documented iodine concentrations for all types of nonsedimentary rocks (6.3-101.7 µmol/kg). Such high iodine concentrations, if commonplace in marine forearc settings, may constitute a significant, previously unknown reservoir of iodine. This serpentinized forearc mantle reservoir may potentially contribute to the total crustal iodine budget and provide a mechanism for its recycling at convergent plate margins. Both clasts and mud show concurrent enrichments in boron and iodine, and the similarity in pore fluid profiles also suggests that these two incompatible, fluid-mobile elements behave similarly at convergent plate margins.

Geochemistry of Pliocene-Pleistocene volcanic sands of ODP Site 136-842

Several distinct, thin (2-7 cm), volcanic sand layers ("ashes") were recovered in the upper portions of Holes 842A and 842B. These holes were drilled 320 km west of the island of Hawaii on the outer side of the arch that surrounds the southern end of the Hawaiian chain. These layers are Pliocene to Pleistocene in age, graded, and contain fresh glass and mineral fragments (mainly olivine, plagioclase, and clinopyroxene) and tests of Pleistocene to Eocene radiolarians. The glass fragments are weakly vesicular and blocky to platy in shape. The glass and olivine fragments from individual layers have large ranges in composition (i.e, larger than expected for a single eruption). These features are inconsistent with an explosive eruption origin for the sands. The only other viable mechanism for transporting these sands hundreds of kilometers from their probable source, the Hawaiian Islands, is turbidity currents. These currents were probably related to several of the giant debris slides that were identified from Gloria sidescan images around the islands. These currents would have run over the ~500-m-high Hawaiian Arch on their way to Site 842. This indicates that the turbidity currents were at least 325 m thick. Paleomagnetic and biostratigraphic data allow the ages of the sands to be constrained and, thus, related to particular Hawaiian debris flows. These correlations were checked by comparing the compositions of the glasses from the sands with those of glasses and rocks from islands with debris flows directed toward Site 842. Good correlations were found for the 110-ka slide from Mauna Loa and the ~1.4-Ma slide from Lanai. The correlation with Kauai is poor, probably because the data base for that volcano is small. The low to moderate sulfur content of the sand glasses indicates that they were derived from moderately to strongly degassed lavas (shallow marine or subaerially erupted), which correlates well with the location of the landslide scars on the flanks of the Hawaiian volcanoes. The glass sands may have been formed by brecciation during the landslide events or spallation and granulation as lava erupted into shallow water.

Geochemistry of ash and pumice of ODP Hole 132-810C

Air-fall volcanic ash and pumice were recovered from 22 intervals in upper Miocene-Pleistocene nannofossil oozes cored in Hole 810C on Shatsky Rise, northwest Pacific. Shatsky Rise is near the eastern limit of ash falls produced by explosive volcanism in arc systems in northern Japan and the Kuriles, more than 1600 km away. Electron probe analyses establish that the ash beds and pumice pebbles are andesitic to rhyolitic in composition, and belong to both tholeiitic and high-alumina lineages similar to tephra from Japanese volcanoes. High-speed winds in the polar-front and subtropical jets are evidently what propelled the ash for such a distance. The pumice arrived by flotation, driven from the same directions by winds, waves, and currents. It is not ice-rafted debris from the north. One thick pumice bed probably was deposited when a large pumice mat passed over Shatsky Rise. Far more abundant ash occurs in sediments cored at DSDP Sites 578 through 580, about 500 km west of Shatsky Rise. Most of the ash and pumice at Shatsky Rise can be correlated with specific ash beds at 1, 2, or all 3 of these sites by interpolating to precisely determined magnetic reversal sequences in the cores. Most of the correlations are to thick ash layers (5.7 +/- 3.0 cm) at one or more sites. These must represent extremely large eruptions that spread ash over very wide areas. Whereas several of the thicker correlative ashes fell from elongate east-trending plumes directed from central Japan, the majority of them - dating from about 2 Ma - came from the North Honshu and Kurile arc systems to the northwest. This direction probably was in response to both long-term and seasonal fluctuations in the location and velocity of the polar-front jet, and to more vigorous winter storm fronts originating over glaciated Siberia.

Geochemistry of boninite and harzburgite of ODP Leg 125 basement samples

Holes drilled into the volcanic and ultrabasic basement of the Izu-Ogasawara and Mariana forearc terranes during Leg 125 provide data on some of the earliest lithosphere created after the start of Eocene subduction in the Western Pacific. The volcanic basement contains three boninite series and one tholeiite series. (1) Eocene low-Ca boninite and low-Ca bronzite andesite pillow lavas and dikes dominate the lowermost part of the deep crustal section through the outer-arc high at Site 786. (2) Eocene intermediate-Ca boninite and its fractionation products (bronzite andesite, andesite, dacite, and rhyolite) make up the main part of the boninitic edifice at Site 786. (3) Early Oligocene intermediate-Ca to high-Ca boninite sills or dikes intrude the edifice and perhaps feed an uppermost breccia unit at Site 786. (4) Eocene or Early Oligocene tholeiitic andesite, dacite, and rhyolite form the uppermost part of the outer-arc high at Site 782. All four groups can be explained by remelting above a subduction zone of oceanic mantle lithosphere that has been depleted by its previous episode of partial melting at an ocean ridge. We estimate that the average boninite source had lost 10-15 wt% of melt at the ridge before undergoing further melting (5-10%) shortly after subduction started. The composition of the harzburgite (<2% clinopyroxene, Fo content of about 92%) indicates that it underwent a total of about 25% melting with respect to a fertile MORB mantle. The low concentration of Nb in the boninite indicates that the oceanic lithosphere prior to subduction was not enriched by any asthenospheric (OIB) component. The subduction component is characterized by (1) high Zr and Hf contents relative to Sm, Ti, Y, and middle-heavy REE, (2) light REE-enrichment, (3) low contents of Nb and Ta relative to Th, Rb, or La, (4) high contents of Na and Al, and (5) Pb isotopes on the Northern Hemisphere Reference Line. This component is unlike any subduction component from active arc volcanoes in the Izu-Mariana region or elsewhere. Modeling suggests that these characteristics fit a trondhjemitic melt from slab fusion in amphibolite facies. The resulting metasomatized mantle may have contained about 0.15 wt% water. The overall melting regime is constrained by experimental data to shallow depths and high temperatures (1250? C and 1.5 kb for an average boninite) of boninite segregation. We thus envisage that boninites were generated by decompression melting of a diapir of metasomatized residual MORB mantle leaving the harzburgites as the uppermost, most depleted residue from this second stage of melting. Thermal constraints require that both subducted lithosphere and overlying oceanic lithosphere of the mantle wedge be very young at the time of boninite genesis. This conclusion is consistent with models in which an active transform fault offsetting two ridge axes is placed under compression or transpression following the Eocene plate reorganization in the Pacific. Comparison between Leg 125 boninites and boninites and related rocks elsewhere in the Western Pacific highlights large regional differences in petrogenesis in terms of mantle mineralogy, degree of partial melting, composition of subduction components, and the nature of pre-subduction lithosphere. It is likely that, on a regional scale, the initiation of subduction involved subducted crust and lithospheric mantle wedge of a range of ages and compositions, as might be expected in this type of tectonic setting.

Main components and isotopic composition of gases from mineral waters of the Taman Peninsula and Caucasus

Helium isotope composition as an indicator of the mantle-derived component was studied in gases from mineral springs, stratal waters, and mud volcanoes developed west of the Teberda River valley (10 objects) and two springs in the central segment of the Greater Caucasus orogen between the active El'brus and Kazbek volcanoes. In the western segment of the orogen ratios of 3He/4He = R_corr vary from 46x10**-8 to 114x10**-8 (from 0.33 to 0.81 R_atm, where R_atm = 1.4x10**-6 is the atmospheric ratio). They are substantially lower relative to ratios in the vicinity of El'brus and Kazbek and close to those in samples from the central segment (from 70x10**-8 to 134x10**-8 (from 0.50 to 0.96 R_atm), as well as to ratios previously recorded in the Caucasian Mineral Waters (CMW) area. Moreover, concentration of 3He in them is notably higher than its crustal radiogenic level characteristic of mud volcanoes in the Taman Peninsula, where 3He/4He varies from 1.4x10**-8 to 2.8x10**-8 (from 0.01 to 0.02 R_atm). Nitrogen-methane gas from northern piedmonts of the western Caucasus also contains nonatmogenic components including radiogenic 40Ar (40Ar/36Ar = 900), excessive nitrogen (~87% of total N2 concentration in sample) and mantle He. These data specify distribution of mantle derivates along the orogen strike and age of intrusive magmatic activity in its different segments.

Geochemistry of Pliocene-Pleistocene volcanic rocks from the Izu Arc

The igneous geochemistry of lavas and breccias from the basement of Sites 790 and 791, and pumice clasts from the Pliocene-Pleistocene sedimentary section of Sites 788, 790, 791, and 793 were studied. Arc volcanism became silicic about 1.5 m.y. before the inception of rifting in the Sumisu Rift at 2 Ma, but eruption of these silicic magmas reflects changes in stress regime, especially during the last 130,000 yr, rather than crustal anatexis. Arc magmas have had a larger proportion of slab-derived components since the inception of rifting than before, but are otherwise similar. Rift basalts and rhyolites are derived from a different source than are arc andesites to rhyolites. The rift source has less slab-derived material and is an E-MORB-like source, in contrast to an N-MORB-type source overprinted with more slab-derived material beneath the arc. Rift magma types, in the form of rare pumice and lithic clasts, preceded the rift, and the earliest magmas that erupted in the rift already differed from those of the arc. The earliest large rift eruption produced an exotic explosion breccia ("mousse") despite eruption at >1800 mbsl. Although this rock type is attributed primarily to high magmatic water content, the clasts are more MORB-like in trace element and isotopic composition than are modern Mariana Trough basalts. After rifting began, arc volcanism continued to be predominantly silicic, with individual pumice deposits containing clasts that vary in composition by about 5 wt% SiO2, or about as much as in historical eruptions of submarine Izu Arc volcanoes. The overall variations in magma composition with time during the inception of arc rifting are broadly similar in the Sumisu Rift and Lau Basin, though newly tapped OIB-type mantle seems to be present earlier during basin formation in the Sumisu than Lau case.