Geochemistry of Pleistocene volcanic rocks in the Mariana Forearc

Geophysical surveys of the Mariana forearc, in an area equidistant from the Mariana Trench and the active Mariana Island Arc, revealed a 40-m-deep graben about 13 km northwest of Conical Seamount, a serpentine mud volcano. The graben and its bounding horst blocks are part of a fault zone that strikes northwest-southeast beneath Conical Seamount. One horst block was drilled during Leg 125 of the Ocean Drilling Program (Site 781). Three lithologic units were recovered at Site 781: an upper sedimentary unit, a middle basalt unit, and a lower sedimentary unit. The upper unit, between 0 and 72 mbsf, consists of upper Pliocene to Holocene diatomaceous and radiolarian-bearing silty clay that grades down into vitric silty clay and vitric clayey silt. The middle unit is a Pleistocene vesicular, porphyritic basalt, the top of which corresponds to a high-amplitude reflection on the reflection profiles. The lower unit is a middle to upper (and possibly some lower) Pliocene vitric silty clay and vitric clayey silt similar to the lower part of the upper unit. The thickness of the basalt unit can only be estimated to be between 13 and 25 m because of poor core recovery (28% to 55%). The absence of internal flow structures and the presence of an upper glassy chilled zone and a lower, fine-grained margin suggest that the basalt unit is either a single lava flow or a near-surface sill. The basalt consists of plagioclase phenocrysts with subordinate augite and olivine phenocrysts and of plagioclase-augite-olivine glomerocrysts in a groundmass of plagioclase, augite, olivine, and glass. The basalt is an island arc tholeiite enriched in large-ion-lithophile elements relative to high-field-strength elements, similar to the submarine lavas of the southern arc seamounts. In contrast, volcanic rocks from the active volcanoes on Pagan and Agrigan islands, 100 km to the west of the drill site, are calc-alkaline. The basalt layer, the youngest in-situ igneous layer reported from the Izu-Bonin and Mariana forearcs, is enigmatic because of its location more than 100 km from the active volcanic arc. The sediment layers above and below the basalt unit are late Pliocene in age (about 2.5 Ma) and normally magnetized. The basalt has schlierenlike structures, reverse magnetization, and a K-Ar age of 1.68±0.37 Ma. Thus, the basalt layer is probably a sill fed by magma intruded along a fault zone bounding the horst and graben in the forearc. The geochemistry of the basalt is consistent with a magma source similar to that of the active island arc and from a mantle source above the subducting Pacific plate.

(Table 1) Lead isotopic compositions of basalts from DSDP Hole 81-553

The passive continental margin south-west of Rockall Plateau is characterized by a thick sequence of oceanward-dipping seismic reflectors. During Leg 81 of the Deep Sea Drilling Project, these reflectors were sampled at Site 553 and proved to consist almost exclusively of basalt. Here we present lead isotope data which indicate that these basalts may have been contaminated by ancient uranium-depleted continental crust, or alternatively, derived from a sub-continental lithospheric mantle source. In either case, the implications are that the basalts of the south-west Rockall Plateau formed by eruption through and onto continental basement, not by 'subaerial seafloor spreading'. This conclusion is in accord with gravity models of the area, which predict stretched continental crust beneath the dipping reflector sequence.

Element analyses and isotope composition of basalts from the West Valley segment, Juan de Fuca Ridge, northeast Pacific

The 50 km-long West Valley segment of the northern Juan de Fuca Ridge is a young, extension-dominated spreading centre, with volcanic activity concentrated in its southern half. A suite of basalts dredged from the West Valley floor, the adjacent Heck Seamount chain, and a small near-axis cone here named Southwest Seamount, includes a spectrum of geochemical compositions ranging from highly depleted normal (N-) MORB to enriched (E-) MORB. Heck Seamount lavas have chondrite-normalized La/Sm en -0.3, 87Sr/86Sr = 0.70235 - 0.70242, and 206Pb/204Pb = 18.22 - 18.44, requiring a source which is highly depleted in trace elements both at the time of melt generation and over geologic time. The E-MORB from Southwest Seamount have La/Sm en -1.8, 87Sr/86Sr = 0.70245 - 0.70260, and 206Pb/204Pb = 18.73 - 19.15, indicating a more enriched source. Basalts from the West Valley floor have chemical compositions intermediate between these two end-members. As a group, West Valley basalts from a two-component mixing array in element-element and element-isotope plots which is best explained by magma mixing. Evidence for crustal-level magma mixing in some basalts includes mineral-melt chemical and isotopic disequilibrium, but mixing of melts at depth (within the mantle) may also occur. The mantle beneath the northern Juan de Fuca Ridge is modelled as a plum-pudding, with "plums" of enriched, amphibole-bearing peridotite floating in a depleted matrix (DM). Low degrees of melting preferentially melt the "plums", initially removing only the amphibole component and producing alkaline to transitional E-MORB. Higher degrees of melting tap both the "plums" and the depleted matrix to yield N-MORB. The subtly different isotopic compositions of the E-MORBs compared to the N-MORBs require that any enriched component in the upper mantle was derived from a depleted source. If the enriched component crystallized from fluids with a DM source, the "plums" could evolve to their more evolved isotopic composition after a period of 1.5-2.0 Ga. Alternatively, the enriched component could have formed recently from fluids with a lessdepleted source than DM, such as subducted oceanic crust. A third possibility is that enriched material might be dispersed as "plums" throughout the upper mantle, transported from depth by mantle plumes.

Mineraly of oceanic crust from the Kane Transform, Mid-Atlantic Ridge

Gabbroic cumulates drilled south of the Kane Transform Fault on the slow-spread Mid-Atlantic Ridge preserve up to three discrete magnetization components. Here we use absolute age constraints derived from the paleomagnetic data to develop a model for the magmatic construction of this section of the lower oceanic crust. By comparing the paleomagnetic data with mineral compositions, and based on thermal models of local reheating, we infer that magmas that began crystallizing in the upper mantle intruded into the lower oceanic crust and formed meter-scale sills. Some of these magmas were crystal-laden and the subsequent expulsion of interstitial liquid from them produced '"cumulus" sills. These small-scale magmatic injections took place over at least 210 000 years and at distances of ~3 km from the ridge axis and may have formed much of the lower crust. This model explains many of the complexities described in this area and can be used to help understand the general formation of oceanic crust at slow-spread ridges.

Geochemistry of ODP Leg 125 peridotites

Ocean Drilling Program Leg 125 recovered serpentined harzburgites and dunites from a total of jive sites on the crests and flanks of two serpen finite seamounts, Conical Seamount in the Mariana forearc and Torishima Forearc Seamount in the Izu-Bonin forearc. These are some of the first extant forearc peridotites reported in the literature and they provide a window into oceanic, supra-subduction zone (SSZ) mantle processes. Harzbutrgites from both seamounts are very refractory with low modal clinopyroxene (<4%), chrome-rich spinels (cx-number = 0.40-0.80), very low incompatible element contents, and (with the exception of amphibole-bearing samples) U-shaped rare earth element (REE) profiles with positive Eu anomalies. Both sets of peridotites have olivine-spinel equilibration temperatures that are low compared with abyssal peridotites, possibly because of water-assisted diffusional equilibration in the SSZ environment However, other features indicate that the harzburgites from the two seamounts have very different origins. Harzburgites from Conical Seamount are characterized by calculated oxygen fugacities between FMQ (fayalite- magnetite- quartz) - 1.1 (log units) and FMQ + 0.4 which overlap those of mid-ocean ridge basalt (MORB) peridotites. Dunites from Conical Seamotmt contain small amounts of clinopyroxene, orthopyroxene and amphibole and are light REE (LREE) enriched. Moreover; they are considerably more oxidized than the harzburgites to which they are spatially related, with calculated oxygen fugacities of FMQ -0.2 toFMQ + 1.2. Using textural and geochemical evidence, we interpret these harzburgites as residual MORB mantle (from 15 to 20 % fractional melting) which has subsequently been modified by interaction with boninitic melt ivithin the mantle wedge, and these dunites as zones of focusing of this melt in which pyroxene has preferentially been dissolved from the harzbutgite protolith. In contrast, harzburgites from Torishima Forearc Seamount give calculated oxygen fugacities between FMQ + 0.8 and FMQ + l.6, similar to those calculated for other subduction-zone related peridotites and similar to those calculated for the dunites (FMQ + 1.2 to FMQ + 1.8) from the same seamount. In this case, we interpret both the harzburgites and dunites as linked to mantle melting (20-25 % fractional melting) in a supra-subduction zone environment The results thus indicate that the forearc is underlain by at least two types of mantle lithosphere, one being trapped or accreted oceanic lithosphere, the other being lithosphere formed by subduction-related melting. They also demonstrate that both types of mantle lithosphere may have undergone extensive interaction with subduction-derived magmas.

Chemical and mineral compositions of basalts and volcanic glasses from DSDP Sites 65-483 and 65-485

Major and rare earth element (REE) data for basalts from Holes 483, 483B, and 485A of DSDP Leg 65, East Pacific Rise, mouth of the Gulf of California, support a simple fractional crystallization model for the genesis of rocks from this suite. The petrography and mineral chemistry (presented in detail elsewhere) provide no evidence for magma mixing, but rather a simple multistage cooling process. Based on its lowest TiO2 content (0.88%), FeO*/MgO ratio (0.95 with total Fe as FeO), and Mg# (100 Mg/Mg + Fe" = 70), sample 483-17-2-(78-83) has been selected as the most primitive primary magma of the samples analyzed. This is supported by the REE data which show this sample has the lowest total REE content, a La/Sm_cn (chondrite-normalized) = 0.36, and Eu/Sm_cn = 1.05. Because other samples analyzed have higher SiO2, lower Mg#, and a negative Eu anomaly (Eu/Sm_cn as low as 0.89), they are most likely derivative magmas. Wright-Doherty and trace element modelling support fractional crystallization of 14.1% plagioclase (An88), 6.7% olivine (Fo86), and 4.7% clinopyroxene (Wo41En49Fs10) from 483-17-2-(78-83) to form the least differentiated sample with Mg# = 63. The La/Sm_cn of this derivative magma is almost identical to the parent magma (0.35 to 0.36), but the other samples have higher La/Sm_cn (0.45 to 0.51), more total REE, and lower Mg# (60 to 56). Both Wright-Doherty and trace element modelling indicate that the primary magma chosen cannot produce these more evolved samples. For the major elements, the TiO2 and P2O5 are too low in the calculated versus the observed (1.38 to 1.90; 0.11 to 0.17, respectively, for example). Rayleigh fractionation calculates a lower La/Sm_cn and requires about 60% crystal removal versus 40% for the Wright-Doherty. These more evolved samples must be derived from a parent magma different from the one selected here and, unfortunately, not sampled in this study. A magma formed by a smaller degree of partial melting with slightly more residual clinopyroxene left in the mantle than for sample 483-17-2-(78-83) is required.

Physical properties, velocity and attenuation characteristics, and mineralogy of gabbros from ODP Hole 176-735B

Laboratory compressional wave (Vp) and shear wave (Vs) velocities were measured as a function of confining pressure for the gabbros from Hole 735B and compared to results from Leg 118. The upper 500 m of the hole has a Vp mean value of 6895 m/s measured at 200 MPa, and at 500 meters below seafloor (mbsf), Vp measurements show a mean value of 7036 m/s. Vs mean values in the same intervals are 3840 m/s and 3857 m/s, respectively. The mean Vp and Vs values obtained from log data in the upper 600 m are 6520 and 3518 m/s, respectively. These results show a general increase in velocity with depth and the velocity gradients estimate an upper mantle depth of 3.32 km. This value agrees with previous work based on dredged samples and inversion of rare element concentrations in basalts dredged from the conjugate site to the north of the Atlantis Bank. Laboratory measurements show Vp anisotropy ranging between 0.4% and 8.8%, with the majority of the samples having values less than 3.8%. Measurements of velocity anisotropy seem to be associated with zones of high crystal-plastic deformation with predominant preferred mineral orientations of plagioclase, amphiboles, and pyroxenes. These findings are consistent with results on gabbros from the Hess Deep area and suggest that plastic deformation may play an important role in the seismic properties of the lower oceanic crust. In contrast to ophiolite studies, many of the olivine gabbros show a small degree of anisotropy. Log derived Vs anisotropy shows an average of 5.8% for the upper 600 m of Hole 735B and tends to decrease with depth where the overburden pressure and the age of the crustal section suggests closure of cracks and infilling of fractures by alteration minerals. Overall the results indicate that the average shear wave splitting in Hole 735B might be influenced by preferred structural orientations and the average value of shear wave splitting may not be a maximum because structural dips are <90°. The maximum fast-wave orientation values could be influenced by structural features striking slightly oblique to this orientation or by near-field stress concentrations. However, flexural wave dispersion analyses have not been performed to confirm this hypothesis or to indicate to what extent the near-field stresses may be influencing shear wave propagation. Acoustic impedance contrasts calculated from laboratory and logging data were used to generate synthetic seismograms that aid in the interpretation of reflection profiles. Several prominent reflections produced by these calculations suggest that Fe-Ti oxides and shear zones may contribute to the reflective nature of the lower oceanic crust. Laboratory velocity attenuation (Q) measurements from below 500 m have a mean value of 35.1, which is consistent with previous vertical seismic profile (VSP) and laboratory measurements on the upper 500 m.

Chemistry of peridotites from ODP Hole 109-670A

Two types of serpentinized peridotites from Hole 670A of Leg 109 were studied in detail. A small piece of relatively unaltered sample, 109-670A-9R-1, #3 (22-24 cm), is olivine websterite characterized by aluminous chromian spinel with Cr/(Cr + Al) ratio of about 0.2. The other minerals have compositions essentially identical with those in more commonly observed serpentinized harzburgite like 109-670A-9R-01, #12 (94-97 cm). The occurrence of pyroxene-rich peridotite with normal harzburgite suggests that small scale heterogeneity in modal compositions exists in the upper mantle beneath the Mid-Atlantic Ridge. Low Cr/Al ratios of spinel and pyroxenes of those peridotites indicate that they are relatively less refractory among peridotites ever recovered from the oceanic region. Textures and the estimated equilibration temperatures indicate that peridotites recovered from Hole 670A are recrystallized and reequilibrated at subsolidus temperature. The occurrence of serpentinized peridotites from the rift valley of the active mid-oceanic ridge may suggest that they represent direct exposure of upwelling mantle materials rather than serpentine diapirs.

Chemiscal composition of basalts from DSDP Legs 69 and 70 Holes

Chemical compositions and 1-atm. phase relations were determined for basalts drilled from Holes 501, 504A, 504B, 505, and 505B on Legs 68, 69, and 70 of the Deep Sea Drilling Project. Chemical, experimental, and petrographic data indicate that these basalts are moderately evolved (Mg' values from 0.60 to 0.70), with olivine plus Plagioclase and often clinopyroxene on the liquidus. Chemical stratigraphy was used to infer that sequential influxes of magma into a differentiating magma chamber or separate flows from different magma chambers or both had occurred. Two major types of basalt were found to be inter layered: Group M, a rarely occurring type with major element chemistry and magmaphile element abundances within the range of the majority of ocean-floor basalts (TiO2 = 1.3%, Na2O 2.5%, Zr = 103 ppm, Nb = 2.5 ppm, and Y = 31 ppm); and Group D, a highly unusual series of basalt compositions that exhibit much lower magmaphile element abundances (TiO2 = 0.75-1.2%, Na2O = 1.7-2.3%, Zr = 34-60 ppm, Nb = 0.5-1.2 ppm, and Y = 16-27 ppm). The liquidus temperatures of the Group D basalts are high (1230- 1260°C) compared with those of other ocean-floor basalts of similar Mg' values. They have high CaO/Na2O ratios (5-8) and are calculated to be in equilibrium with unusually calcic Plagioclase (An78-84). The two basalt groups cannot be related by fractionation processes. However, constant Zr/Nb ratios (>40) for the two groups suggest a single mantle source, with differences in magmaphile element abundances and other element ratios (e.g., Zr/Ti, Zr/Y, Ce/Yb) arising through sequential melting of the same source. Magmas similar to Group D, if mixed with more typical mid-ocean-ridge basalt (MORB) magmas in shallow magma chambers, could provide a source for the highly calcic Plagioclase phenocrysts that appear in more common (i.e., less depleted) phyric ocean-floor basalts.