B21A-0319
Discovery of Off-Axis Melt Lenses at the RIDGE-2000 East Pacific Rise Integrated Studies Site
The first multi-streamer 3D seismic reflection experiment conducted aboard the R/V Marcus Langseth was carried out in July-August 2008 at the RIDGE-2000 Integrated Studies Site at the East Pacific Rise between 9°37'N and 9°57'N. One of the major findings of this cruise is the presence of at least three regions with prominent near-axis crustal reflectors that we interpret as off-axis melt lenses (OAML) injected at mid-crustal levels to both the east and west of the ridge crest. In some instances these regions are associated with subtle topographic expressions of seafloor volcanism. The first set of OAML reflections was discovered near 9° 54'N at distances from the ridge axis that range from ~4 to ~7 km, and extending ~6.3 km in the ridge-parallel direction. Two-way travel times from the seafloor to the OAML are 750-830 ms, which correspond to depths of ~1,800-2,000 m (assuming an average upper crustal velocity of 4.7 km/s), although in some cases there is evidence for deeper (~1,420 ms, or ~3,330 m) reflections farther off-axis forming a staircase-like system of melt lenses. A second set of OAML reflections was discovered further south near 9° 39'N also within 3-6 km east of the spreading axis, and extending ~3.3 km in the ridge-parallel direction. Crustal travel times to these OAML reflections are 710-830 ms (1,700-2,000 m below seafloor). In this region the OAML is locally underlain by deeper reflections at depths of 1,300 ms (~3,100 m) that could indicate off-axis melt sill intrusions at multiple levels. A smaller OAML was found within 2 km west of the axis near 9° 52'N. In this paper we present preliminary cross-sections of this dataset to evaluate the extent and general characteristics of these off-axis events, which highlight the importance of off-axis magmatism in crustal accretion along this section of the East Pacific Rise.
B21A-0320
New observations of the magmatic segmentation of the East Pacific Rise from Siquieros to Clipperton from a multi-streamer seismic reflection imaging study
In summer 2008, we collected the first multi-streamer 3D seismic reflection dataset of the new national seismic imaging facility, the R/V Marcus G. Langseth, during cruise MGL0812. Our survey included a primary 3D grid extending from 9°57 to 9°42 centered on the EPR ISS "bulls eye" site at 9°50' and 3 parallel along axis lines extending from ~10°05' N to 9°40N. The central along-axis line was extended to encompass the entire length of the ridge from Clipperton to Siquieros fracture zones to facilitate regional studies of magmatic segmentation along the full length of this first order ridge segment. Multibeam bathymetry data were collected simultaneously with the seismic data using the 1°x1° beam EM120 available on the Langseth providing high quality bathymetry extending 30-40 km to either side of the axis north of 8°50'N. In this presentation we present preliminary results focused on axial segmentation from Siquieros to Clipperton. The data reveal fine-scale segmentation of the axial magma lens coincident with the volcanic segmentation of the ridge axis evident in the seafloor morphology. Each volcanic segment is associated with a discrete melt lens, ~5-10 km long, and, in several cases, defined by diffractions from the lens edges. Adjacent lenses differ in reflection strength, depth, and dip. At the discontinuities, lenses are offset from one another and overlap forming shingled lenses in along-axis view and multiple lenses in cross-axis view. These magma lens discontinuities correspond with offsets in the axial summit trough and changes in the volcanic morphology of the axial high, and point to a similar lifespan for these structures. The segmentation of the axial magma body observed in our new data is also apparent in early seismic reflection data collected with the R/V Conrad over 20 years ago, indicating persistent segmentation through the two volcanic eruptions that have occurred in this region since this time.
B21A-0321
Searching for Changes in AMC Characteristics on the EPR Using Comparisons of Reflection Images Obtained in 1985 and 2008
The recent 3D multi-streamer seismic reflection imaging experiment carried out on the East Pacific Rise at 9°50'N (R/V Langseth MGL0812) encompassed the region in which single and two-ship 2D multi- channel seismic data were acquired in 1985 (R/V Conrad). These two phases of seismic data acquisition therefore span the time of known volcanic eruptions (1991 and 2005-06) on this ridge segment. This invites a comparison of the seismic response of key features of the subsurface such as the axial magma chamber (AMC) to investigate if discernable changes have taken place that could be ascribed to the eruptions. Detailed comparison of the two datasets is hampered by differences in ship and streamer navigation accuracy and in acquisition parameters between the two experiments including airgun source array volumes and tuning, shot interval, streamer tow depth, streamer length and hydrophone group spacing. Though careful analysis will be required to minimize the effects of these differences, preliminary observations can be reported from the data in their initial stages of processing. Three cross-axis lines from the Conrad survey are compared with Langseth data: one at 9°50'N within the area of high temperature vents, a second south of 9°40'N near vent B and a third around 9°30' N near vent K, well south of the area of known eruptions. In the crossings near the sites of vents B and K we are unable to identify any significant changes in the AMC event. In contrast, at 9°50'N, significant differences can be recognized in subsurface features. The AMC appears to be shallower and less complexly structured in the 1985 data suggesting perhaps that the magma lens has been deflated somewhat by subsequent eruptions. Direct along-axis comparisons are less straightforward as the earlier lines do not directly match new along-axis profiles. Nevertheless, there are suggestions that reflection strength of the AMC has changed in a number of places between the two surveys in the most magmatically active stretch of the ridge, and these amplitude changes too are likely to be associated with the eruptions. One of the goals of the MGL0812 experiment was to establish a base for a repeat 3D (i.e., 4D) survey that would allow mapping temporal changes in the subsurface that can be compared to changes in features of the seafloor such as hydrothermal venting and volcanism. Our preliminary comparison of the 3D data with data collected in 1985 suggests that future changes in the subsurface will be detectable.
B21A-0322
3D Seismic Reflection Imaging of Crustal Formation Processes on the East Pacific Rise, 9°57-42'N
Between June 29th and August 19th 2008 the research vessel Marcus G Langseth carried out its first multi- streamer 3D seismic reflection imaging cruise, MGL08-12, by conducting a program research on the East Pacific Rise centered around 9°50'N. The primary goals were to create an accurate 3D seismic reflection image of the magmatic-hydrothermal system at this Integrated Study Site of the Ridge2000 program by imaging the structure of the axial magma chamber (AMC) lid and oceanic crust at a resolution, accuracy, and scale comparable to seafloor observations. The vessel acquired data with four, 6-kilometer solid streamers each comprising 468 active channels deployed with a total separation of 450 meters. Four gun strings with total volume of 3300 cubic inches in two groups fired alternately provide the source for a shot spacing of 37.5 meters. This configuration yields eight CMP lines for each of the sail lines that were spaced 300 m apart, and a static bin size of 6.25 m × 37.5 m in the along-track and across-track directions, respectively, providing a nominal fold of 40. The cruise accomplished the acquisition of ~3,782 km of sail line data. There are 111 across axis lines that required 10 repeated lines and 14 infills. Average feathering during the cruise was 0° ± 5° (one standard deviation), with maximum values of up to 11°. This means that 18% of the total cross axis acquisition was needed for reshoots and infilling. A 25% multiplier on planned lines for a 3D grid is probably a useful figure to use in cruise planning and is fairly standard in the seismic industry. Data quality meets or exceeds industry standards. 3D coverage was achieved in two areas. The larger comprises a set of 93 equally spaced lines forming the 3D grid between 9°57'N and 9°42'N. This grid is made up of lines from all of racetracks #1 and #2 and the northern lines of racetrack#3 and covers two principal hydrothermal vent areas in a continuous fashion. The second 3D area is comprised of 14 southern lines of the incomplete racetrack#3, and is separated from the larger grid by a gap of 11 lines or 3.3 km. South of this region we acquired a pair of lines that crossed the ridge at about 9°35'N and a single line just south of 9°30'N. Along axis lines also extend throughout the entire ridge segment from the Clipperton to Siqueiros Fracture Zones. Between the Clipperton Fracture zone and 9°41'N we acquired three parallel lines along the axis. Between this latitude and 9°20'N we obtained two parallel lines and one axial line south from that latitude, including both limbs of the Overlapping Spreading Center. Because each of the sail lines acquired represents eight CMP profiles CMP line acquisition is 30,255.60 km. This comprises 99,888 shots and 186,998,336 source-receive pairs and with a sampling rate of 2 milliseconds for a total of 957,890,520,320 data samples. In this overview presentation we will describe the work carried out and the lessons learnt from the cruise. Accompanying presentations will describe initial findings from the cruise and plans for data analysis.
B21A-0323
Deep reflection structure imaged by the 2008 3D seismic reflection Survey at the RIDGE- 2000 East Pacific Rise Integrated Studies Site
The first multi-source and multi-streamer 3D seismic reflection experiment carried out using academic resources was done aboard the R/V Marcus G. Langseth in Summer 2008 during cruise MGL0812. The targeted area was the RIDGE-2000 Integrated Studies Site at the East Pacific Rise. Our primary 3D survey grid extends from about 9° 57'N to 9° 42'N, with a smaller grid just to the south covering approximately from 9° 40'N to 9° 37.5'N. Additionally, about 1 and 0.5 km wide across-ridge-axis swaths of data were collected at 9° 36'N and 9°30'N respectively, as well as an along-ridge-axis swath about 1 km wide and extending from 10° 05'N to 9° 40'N. We here focus on a preliminary analysis of the reflection structure imaged within the lower crust and uppermost mantle. Moho reflection arrivals are imaged through much of the investigated area. The character of Moho reflection events varies from simple, single reflection wavelet to more complex arrivals indicating spatial changes in structure within the Moho transition zone. Particularly strong Moho reflections are observed in the southern half of the main 3D grid. In places, Moho reflection event appears to extend across the ridge axis potentially suggesting "zero-age" Moho development. Weak Moho arrivals are found at the north end of the main 3D box and within the smaller box to the south. Most notable place lacking Moho reflections is the Lamont seamount area where Moho is not observed on either side of the ridge axis, although the area lacking Moho reflections is wider on the western ridge flank. Further south, along the across-ridge-axis swaths, Moho reflections again become more pronounced. A suit of what mostly appear to be reflection events is recognized between the AMC and Moho. Many of them do not appear to be multiples, and their origin is not well understood. Possible origins for these events include: lower boundary of the AMC, S-converted waves, and lower crustal melt lenses. Along sections of the two 3D grids, possible subhorizontal, low signal-to-noise ratio mantle reflections are identified 1-2 s after the Moho reflection. Arrivals potentially originating from lower crustal melt sills, zero-age Moho, and mantle reflections are all of critical importance for understanding fast spreading centers. In the years to follow, we will put significant effort to analyze the identified arrivals in order to determine their origin and find where they fit within the crustal accretionary puzzle.
B21A-0324
Tidal Triggering of Microearthquakes Constrains Permeability at 9º50'N East Pacific Rise
The permeability structure within hydrothermal circulation systems at mid-ocean ridges is a primary control on the geometry of subseafloor fluid flow, which strongly influences the distribution of heat, chemistry, and biological productivity within new oceanic crust. Though the permeability structure of young oceanic crust plays an important role within hydrothermal systems, measurements of on-axis permeability have been elusive. We show that along-axis variations in the tidal triggering of microearthquake activity can be used to estimate bulk permeability of the subseafloor at 9°50'N East Pacific Rise. We focus on 6,050 earthquakes recorded between October 2003 and April 2004, and observe a systematic phase lag between peak microearthquake production and maximum tidal extension along-axis. We hypothesize that the pattern of earthquake activity results from poroelastic pore pressure gradients traveling through the hydrothermal system. Such pore pressure gradients may be generated in response to Earth tides and ocean tidal loading, when heterogeneities in the permeability structure or variations in the formation or fluid elastic properties exist within the hydrothermal circulation cell. We propose a one-dimensional poroelastic model to estimate the bulk permeability within the hydrothermal system, and find that the diffusive lengthscale of pore pressure perturbations is ~2.7 km. This estimate implies a permeability of ~10-13 to 10-12 m2, which is consistent with on-axis numerical modeling studies (~10-16 to 10-12 m2) and shallow off-axis in situ borehole measurements (~10-15 to 10-13 m2). The investigation of spatio- temporal patterns in tidal triggering of microearthquakes may provide a new technique to estimate permeability at mid-ocean ridges.
B21A-0325
Earthquake Monitoring at 9 deg 50'N on the East Pacific Rise: Latest Results and Implications for Integrated Models
Ocean bottom seismograph (OBS) data were recorded continuously between October 2003 and January 2007 at the Ridge 2000 Bull's Eye site at 9°50'N on the East Pacific Rise (EPR) using a 4 x 4 km array of up to 12 instruments with approximately annual turnaround. These data have provided exciting insights into fundamental processes at fast-spreading ridges including volcanism and hydrothermal circulation. They also are providing critical linkages for understanding the geological, chemical and biological data at this site. Results from the first OBS deployment have shown that we are able to monitor microseismicity on a fine enough scale to image the fundamental structure of a hydrothermal circulation cell, and we have identified an on-axis down-flow zone and a hydrothermal cracking front overlying the axial magma chamber (Tolstoy et al., 2008). Our results show that hydrothermal circulation at the EPR is dominantly along-axis with narrowly focused down-flow at small kinks in the axial summit trough (AST). There appear to be two distinct circulation cells within the 9°49'N-9°51'N area, and these correlate well with temperature, chemical and biological observations. The rate of seismic events recorded at the array were ~2 orders of magnitude higher than anticipated based on prior results from this area (>320,000 events recorded versus ~4,500 anticipated), and therefore the processing task is considerable. In addition to hand-picking phase arrival times from periods of particular interest, we are also working on improved automatic detection tools to speed up processing of data from the remaining years and the use of waveform cross-correlation to improve event locations. Preliminary results to date suggest that the basic structure imaged in the 2003-2004 earthquake data persists, with seismicity rates continuing to climb leading up to the January 2006 eruption. We will present the most recent earthquake locations and discuss how they fit into results from the 2003-2004 data, as well as the implications for integrated models at this site.
B21A-0326
Seismic activity associated with temperature perturbations at Bio 9 hydrothermal vent on the East Pacific Rise at 9deg50'N
A brief seismic swarm north of high temperature vent Bio 9 on November 19th 2003 appears to have triggered a temperature response on a time scale of ~3 days, similar to previously observed events (Sohn et al., 1998; Wilcock, 2004). Temperatures initially rose by ~5degC over the course of 10 days and then fell by ~14degC over 8 days, before gradually recovering to temperatures slightly higher than before the event over the following 2-3 months. Near-by chimneys showed small or no response to this event suggesting that fluid pathways to individual vents are narrowly defined deep within the crust. The seismic swarm of ~130 relocated events forms a thin near-vertical line approximately 1 km tall with evidence of upward propagation from directly above the axial magma chamber to the top of hydrothermal cracking zone and into the interpreted upflow zone. The earthquakes initiate from an area close to the events associated with a similar temperature perturbation at Bio 9 in 1994 (Sohn et al., 1998), suggesting that fluid pathways are maintained on decadal time scales. However, the character of the temperature perturbation is different from the 1994 event in that after initially rising, temperatures decline to below pre- swarm level before recovering gradually. This may reflect a different phase in the evolution of the hydrothermal cell, since temperatures at Bio 9 in 2003 were higher and more stable than they were in 1994. The interpreted hydrothermal cell beneath Bio 9 shows a decrease in seismic b-value following recovery of the Bio 9 temperatures. A decrease in b-value may reflect an increase in stress, lower pore pressures, or variation in the distribution of fractures within the rock. Analysis of the seismic swarm within the context of the interpreted hydrothermal cell and previous seismic results will be presented, and possible models for triggering the temperature perturbations will be discussed.
B21A-0327
Is There a Deep Hydrothermal Circulation at the EPR?
Recent papers suggest the relatively narrow (4km) magma chamber observed beneath fast spreading ridges imaged using seismic tomographic, reflection and compliance techniques requires that the hydrothermal circulation extends through most of the crust within a few kilometers of the rise axis. Geothermometry provides conflicting data as to the rate of cooling of the lower crust under fast spreading ridges. Geochemical observations mostly find little water reaches the lower crust near fast spreading ridges. Ridge crest magma chamber theoretical models find that either that a deep circulation is necessary or that some kind of magma convection (eg. the gabbro glacier) must occur to explain the narrow magma chamber observed. I reexamine the seismic observations and conclude that the seismic data have insufficient accuracy to constrain the temperatures of the lower crust outside of the magma chamber so that the tomographic results are fully consistent with an absence of a deep circulation near axis. The absence of deep microseismicity or of any surface expression of the hydrothermal flow off axis suggest that there is no deep hydrothermal circulation near the EPR crest.
B21A-0328
The 2005-2006 Eruption at the East Pacific Rise 9°50ŒN: Detailed Flow Morphology Mapping and Insight Into Emplacement Processes
The volumes of individual eruptions and modes of lava distribution across the seafloor are crucial to understanding the construction of the upper oceanic crust at mid-ocean ridges. An unprecedented opportunity to gain insight into these processes arose on the East Pacific Rise (EPR) when an eruptive event was discovered to have occurred during late 2005 to early 2006 near 9°50ŒN. Since this new flow was documented in April 2006, a total of 41 deep-towed imaging surveys have been conducted with the Woods Hole Oceanographic Institutionfs TowCam system. More than 60,000 digital color images and high- resolution (+ 10 cm) bathymetric profiles were collected during the surveys. There has been significant progress in determining the flow boundaries of the 2005-2006 eruption and we are currently completing a detailed analysis of the lava morphology, composition, and structures within the flow. Using TowCam images and bathymetric data, the surface morphology of the flow is being analyzed at a level of detail that has never before been possible, giving insights into the spatial distributions of lava flow surface morphologies of the 2005-2006 eruption and how these various morphologies relate to the physical characteristics of the ridge and dynamics of flow emplacement. Initial results suggest that the interior of this recent flow at the EPR is characterized by lobate and sheet morphologies whereas pillow fronts tend to form at the termini of the flow lobes. The centers of lava channels that served as distribution pathways during the flow tend to be characterized by sheet flows while hackly flows that transition into lobate flows define the edges of the channels. However, the largest of these channels is dominated by a hackly flow
B21A-0329
Volcanic Eruptions of the EPR and Ridge Axis Segmentation: An Interdisciplinary View
The eruption of the EPR in 2005-06 provides an ideal window into the relationship between fine-scale segmentation of the ridge axis and individual eruptive episodes. Lava flow mapping of the eruption by visual and acoustic images, precise dates on multiple eruptive units, stress information from seismicity, long-term records of hydrothermal activity, and well known segment boundaries illustrate the relationships between eruptions and segmentation of mid-ocean ridges. Lava flows emerged from several sections of the axial summit trough (AST) during the eruption, presumably from en echelon fissures between 9 45'N and 9 57'N. Each en echelon fissure is a 4th order segment, and the overall area matches the 3rd Order segment between ~9 45'N and ~9 58'N. Within the eruption, the primary eruptive fissure jumped east by 600 m at 9 53'N, and ran along an inward facing fault scarp, although limited lava effusion also extended northward along the axial fissure. A zone of high seismicity connects the normal fault bounding the eastern fissure eruption with the main locus of eruption on the ridge axis to the south, suggesting that the offset eruption may have occurred in response to stress buildup on this fault. Radiometric ages indicate that the entire along-axis extent of the eruptive fissures activated initially, but that volcanic activity focused to a single fourth-order segment within 1-3 months. Previously indentified breaks in the AST and its overall outline were largely unchanged by the eruption. These observations support the hypothesis that fourth-order segments are offsets controlled by the mechanics of dike emplacement, whereas third-order segments represent discrete volcanic systems. Dike segmentation may be controlled by variations in underlying ridge structure or the magma reservoir. Hydrothermal systems disrupted as far south as 9 37'N may be responding to cracking due to stress interaction or share a common deeper magmatic source. Comparisons between the 1991 EPR eruption at the same site, and several mapped southern EPR eruptions, the 10 45'N EPR eruption in ca. 2003 all show similar relationships to segmentation
B21A-0330
Geochemical Characterization of Axial and Off-axis Lavas From the 2005-06 Eruption Along the East Pacific Rise (9°46-56' N): Implications for AMC Composition Over Decadal Time Scales and Mantle Heterogeneity Along Fast Spreading Centers
In 2005-06, axial and off-axis fissure eruptions produced ~22 × 106 m3 of lava along an ~18 km (9°46'-56' N) ridge segment of the northern East Pacific Rise (EPR) at the R2K Integrated Study Site (ISS). This event marked the first observed repeat eruption at a MOR as 2005-06 units largely bury flows from a smaller eruptive event in 1991-1992 at ~9° 50' N. In general, 2005-06 lavas have typical N-MORB signatures with strongly depleted incompatible element patterns. In comparison to the more primitive 1991 lavas (MgO = 8.7-8.4 wt %, Zr = 82 ± 7 ppm), 2005-06 lavas are slightly more evolved (MgO = 8.2-7.8 wt %; Zr = 94 ± 5 ppm) and exhibit marked negative Sr and Eu/Eu* anomalies. Spatial geochemical differences show that lavas from the northern and southern limits of the 2005-06 eruption are the most evolved. Sr, Nd, and Pb isotopic ratios for 2005-06 axial and off-axis lavas (87Sr/86Sr = 0.70247-0.70250; εNd = +10 to +11; 206Pb/204Pb = 18.25- 18.27; 207Pb/204Pb = 15.470-15.474; 208Pb/204Pb = 37.65-37.68) fall within larger isotopic fields defined for relatively young (~200 yrs to 1991-92) lavas from the 9°-10° N EPR (Sims et al., 2002). However, temporal variation in Pb isotopic ratios is apparent over the 7-month course of the eruption as determined from high-precision 210Po lava eruption ages (see Rubin et al., abstract, this session). Geochemical data from the 2005-06 eruption are consistent with (1) upper crustal processes (e.g., fractional crystallization and magma mixing) operating on decadal time scales within the axial magma chamber (AMC) and (2) an isotopically-variable mantle source. Products of the off-axis fissure eruption are chemically similar to the northernmost axial flows and are best explained by the eruption of cooler fractionated melts from the AMC. Basalts erupted in 2005-06 may reflect a regional depleted mantle end-member, as isotopic mixing models between a 2005-06 depleted mantle and various local enriched sources (evident in E-MORB and off-axis seamounts) can explain the overall isotopic heterogeneity along the 9°-10°N EPR.
B21A-0331
Magmatism at the 9o 03'N Overlapping Spreading Center, East Pacific Rise: The Xenolith Story
There is a growing body of evidence that the spectrum of volcanics observed at the 9o03'N overlapping spreading center (OSC) results from a complex interplay of crystal fractionation and magma mixing. Many of the lavas sampled during the 2007 Atlantis II/Jason cruise to the OSC contain xenoliths that bear witness to the complexities of magmatic processes occurring within the nascent crust. The xenoliths are largely composed of coarse (greater than 3mm) clots of plagioclase, olivine, and clinopyroxene, in various stages of disaggregation, often with sub-ophitic to poikilitic textures. The individual minerals show only very small variations in major element chemistry. Many xenoliths that have not been disaggregated contain patches of interstitial glass. A common characteristic is that the mineral assemblages are not in equilibrium with their host melts. The compositions of olivines demonstrate that most mineral assemblages have crystallized from melts that are more primitive than the host lavas, a conclusion consistent with the olivine Ni contents and the embayed outlines of individual olivine grains. An interesting aspect of the chemistry of clinopyroxenes is that the major and trace element abundances are decoupled. Individual clinopyroxene grains that are homogeneous with respect to major elements show a variety of REE normalized patterns, including patterns with smoothly varying REE abundances and those with substantial negative europium anomalies. The lack of correlation between Eu/Eu* values and total REE abundances in individual clinopyroxenes indicates that the REE variations cannot reflect in situ crystallization of clinopyroxene from a single melt undergoing substantial plagioclase precipitation. Instead, the growth of clinopyroxene must have been from unrelated melts with similar major element chemistry but with distinctly different trace element signatures.
B21A-0332
Geochemistry of the 2005-2006 East Pacific Rise Eruption: Chemical Changes and Possible Petrogenetic Relationships to the 1991-1992 Eruption
The first opportunity to study a well-documented temporal sequence of eruptions at one mid-ocean ridge (MOR) locality presented itself when a second eruption was discovered in 2006 on the East Pacific Rise (EPR) around 9 °50 'N, 15 years after an eruption was witnessed in 1991-1992. Major and trace element analyses of glass samples from the 2005-2006 eruption indicate that the flows are largely aphyric, relatively homogeneous N-type MORB but exhibit some geochemical variability outside of analytical uncertainties. Relative to the more than 1200 basalts that have been analyzed from this section of the Integrated Study Site (ISS), samples from the 2005-06 eruption are moderately evolved (7.5 to 8.3 weight percent MgO), and are more evolved than the basalts erupted in 1991-92 (8.3 to 8.9 weight percent MgO). However, both the 2005- 06 and 1991-92 eruptions appear to have originated from the same incompatible element-depleted mantle source. Modeling of the observed chemical variability suggests the 2005-06 flows could be related to the most primitive 1991-92 lava by up to ~10 percent low-pressure fractional crystallization. The compositions of the new flows fall along a liquid line of descent predicted by the crystallization of olivine and plagioclase in ratios that range, respectively, from approximately 6:1 to 1:2, and at crystallization temperatures that range from approximately 1200 °C to 1140 °C. Some gabbroic crystal clots enclosed in the young lavas support this model but others suggest complex crystal-melt interactions in the subaxial magma chamber. Additionally, chemical and spatial data from the 2005-06 eruption may indicate chemical differentiation of the lava along axis during the eruptive process. Although the data are consistent with small amounts of crystal fractionation of the 1991-92 parent over the past 15 years, it is possible the elemental trends were created by mixing between the parental melt and a more evolved magma from another part of the axial magma chamber.
B21A-0333
In-Situ Measurement of pH of Hydrothermal Vent Fluids at EPR 9 N EPR: Implications for Acidity in Subseafloor Reaction Zones
In-situ pH values obtained at the vent site offer an opportunity to constrain vent fluid pH in the deep reaction zone in comparison with results based on lab-measured pH and distribution of aqueous species calculations. pH is a master variable in all geochemical systems and it is of great importance to minimize uncertainties in its determination if hydrothermal alteration processes at depth are to be determined accurately. During a recent cruise (AT 15-28) in early 2008, in-situ measurements of pH and dissolved H2 and H2S were carried out along with conventional sampling of vent fluids, with efforts focused at the P-vent site at 9°N EPR. The measured fluid temperature was approximately 380°C, while in-situ pH value of 5.14 was obtained at the orifice for P structure. Based on measured fluid samples and the in-situ data, efforts were made to predict the pH value at depth from which the vent fluids were derived. Taking constraints imposed on dissolved chloride by phase separation effects, we estimate the current temperature and pressure conditions for P vent source fluid to be approximately 420°C and 330 bars. The pH value at these conditions is then predicted by taking explicit account of the measured in-situ pH data for the end-member fluid at the seafloor. In the calculation, adiabatic cooling is assumed for fluid ascending from the reaction zone at the depth to the seafloor vent site. During ascent, the acidity of the fluid is also assumed to behave conservatively. Our prediction suggests a pH of approximately 6.4 in the subseafloor reaction zone at the higher conditions of temperature and pressure. The predicted pH value indicates that the fluid is only slightly acidic and actually very near the neutrality condition. This method for predicting pH in the root zone of a seafloor hydrothermal system from in-situ pH values measured at the seafloor will be further utilized for the other vent sites at different conditions (temperature, composition) to help understand the evolution of acidity and redox of the vent system.
B21A-0334
Circulation in the EPR ISS Observed During the LADDER Project
Horizontal dispersal of hydrothermal "products," including the larvae of organisms endemic to vent fields, is accomplished via advection by the oceanic flow field. In the context of the ongoing LADDER project (LArval Dispersal along the Deep East pacific Rise), the hydrography and deep circulation near the EPR crest between 9 and 10N was sampled during 3 quasi-synoptic oceanographic surveys, traced with a neutrally buoyant float and a SF6 release, as well as monitored for an entire year with an array of moored instruments. The data reveal a low-frequency (on time scales exceeding a few months) westward drift of ~1cm/s across the EPR crest, consistent with helium observations obtained during WOCE. Close to the ridge axis, the cross-ridge flow is modified by the presence of north- and southward boundary currents along the western and eastern EPR flanks, respectively. On monthly time scales, which are particularly relevant for larval dispersal, mesoscale oscillations dominate the velocity measurements. The mesoscale currents are characterized by large directional variability and typical velocities of several cm/s, similar to the tidal velocities in this region. As a result, dispersal on monthly time scales can be in any direction, including against the mean flow. While the mesoscale velocity field is coherent over more than 50km in the along-axial direction, lack of significant cross-axial coherence across the 30km mooring spacing indicates that the mesoscale oscillations on axis are more likely the signatures of waves, rather than of coherent eddies. Near the northern end of the segment, the circulation is strongly affected by a chain of seamounts on the western ridge flank. As detailed in the neighboring poster by Lavelle et al., the hydrographic and velocity observations collected during LADDER have been integrated with a numerical model of the regional circulation in order to study dispersal in the EPR ISS. Comparisons with data from the tracer-release experiment indicate that the model is of sufficient quality to investigate dispersal on monthly time scales, which adds significant value to the observations.
B21A-0335
Deep Ocean Circulation and Transport Where the East Pacific Rise (9-10 N) Meets the Lamont Seamount Chain
Numerical modeling studies along with current and hydrographic measurements and a SF6 tracer experiment during LADDER (LArval Dispersal along the Deep East Pacific Rise) I and II cruises in November and December 2006 have yielded new insight into ocean flow and transport at the northern end of 9-10 N EPR segment. Long period flows tending westward over the ridge result in uplifted isopycnals over the ridge crest. The consequence is narrow geostrophic (plus rectified flow) 'jets' on each side of the ridge which can transport material north on the western and south on the eastern flanks of the ridge. When the flow reaches the gap between the ridge and the first Lamont seamount (Sasha) to the west of the ridge, flow moves north through the gap and/or along the southern side of the seamounts. A model SF6 tracer dispersion simulation meant to parallel the SF6 field experiment suggest that in late November, 2006, at the time when the tracer would have likely reached the gap in the northward flank flow, mass flux through the gap was atypically southward, partially stalling northward SF6 transport and sending the bulk of the SF6 westward. Measurements of actual SF6 dispersion initially suggested the westward transport just south of the seamounts. The model also indicates long term mean flow in an anticyclonic sense around the entire seamount chain and transport primarily north through the gaps between individual seamounts. If larvae were to disperse passively like the tracer in these experiments, only a small fraction would be retained along the ridge, with most transported to the Lamont Seamount chain where conditions are not suitable for larval settlement. The adjacent poster by Thurnherr et al. presents some of the LADDER field measurements underlying this modeling effort.
B21A-0336
Dike Interactions along the East Pacific Rise: Can Dike-Induced Static Stress Changes Trigger Diking at Adjacent Ridge Segments?
Although the physical conditions that control dike intrusion at spreading centers are subject to debate, it is generally agreed these include a crustal section stressed close to failure and an overpressured magma chamber. Along the East Pacific Rise (EPR), fast spreading rates (100-150 mm/yr) and the presence of a narrow, shallow and nearly continuous melt lens (0.5-1.5 km wide, 1-1.5 km deep) imply that these two criteria are frequently met. Hence, a small incremental stress change at the EPR axis may be sufficient to trigger dike intrusion along a "4rth-order segment"- the elemental 10-20 km-long section of ridge that ruptures in a given volcanic episode. We use the COULOMB 3.1 software to investigate whether static stress changes induced by dike emplacement at a 4rth-order segment may be sufficient to trigger diking at neighboring segments. In the absence of any observational data about dike triggering, we adopt some conclusions from studies of stress transfer in earthquake sequences, and assume that static stress changes of 1-2 bar can promote dike intrusion in a favorably pre-stressed crust. Modeling results indicate that tensile stress changes of 1-2 bars extend as far as 10 km beyond the tip of a newly emplaced dike, suggesting that along the EPR, the intrusion of a dike may be sufficient to promote concomitant diking at adjacent 4rth-order segments. This 10 km distance varies only slightly when experimenting with reasonable dike geometries: dike heights of 1-1.5 km and widths of 0.5-2 m are used, as narrowly constrained from observations at the EPR. Hence, provided that a second-order segment (50 km-long or more) is critically stressed and that the subjacent magma chamber is overpressured, eruption of a 4rth-order segment somewhere along that segment may trigger a sequence of diking events on a relatively short time-scale. This model may soon be tested at the EPR "Integrated Study Site" at 9N, by a series of recent and on-going experiments funded by the NSF-RIDGE2000 initiative. A diking event was documented in 2006 between 9 46'-56'N (Bull's Eye area), as evidenced from the pattern of microseismicity, a large hydrothermal plume, and fresh, extremely glassy basalts. During cruise AT15-16 early 2007, basalt samples were collected near 10 01'N that appear equally fresh and glassy, and a tow-yo survey mapped a sizable hydrothermal plume at 9 59'-10 03'N. This suggests that an eruption also occurred recently in that area, within the range of calculated significant stress change from the 2006 dike intrusion, but was undetected by the array of sensors clustered within the Bull's Eye. This would be compatible with a model in which one of these eruptions triggered the other. A linear array of 19 bottom pressure recorders (BPRs) are currently deployed along the 9N segment at roughly equal spacing, to be left on the seafloor for 2.5 years (project NSF-OCE0426575; coPIs M.H. Cormier, W.R. Buck, and S.C. Webb). BPRs can detect vertical motion of the seafloor with a resolution of just a few cm, much smaller than what is expected to occur above an intruding dike. The BPR array should "catch in the act" any diking event during that period, and constrain whether or not they occur in a cohesive spatio-temporal pattern.
B21A-0337
Temporal Variability of Hydrothermal Plumes Along the 9N Segment of the East Pacific Rise
Tow-yo profiling of the water column has been carried out continuously between 9 31'-10 04'N along the East Pacific Rise over the course of two expeditions of the R/V Atlantis, AT15-16 in March 2007 and AT15-32 in June 2008. These data were collected as ancillary programs to underwater geodetic and seafloor compliance experiments, when time permitted. Overall, the new data highlight a pattern of intense and continuous hydrothermal plumes between 9 31'-10 02'N, within the broad bull's eye area; this is similar to the pattern that emerged from tow-yo profiles acquired soon after the 1991 eruption at 9 45'-51'N (Baker et al., EPSL 1994), suggesting it either represents a long-term feature of this inflated portion of the ridge segment, or that it develops soon after an eruption and persists for at least a few years. Differences between the two datasets suggest that this later interpretation is more correct. First, the hydrothermal plume between 9 58'N and 10 02'N, north of the site of the documented 1991 and 2006 eruptions, is significantly more vigorous in 2007 than it was 1991. Lava samples dredged at 10 01'N during the 2007 expedition are visually undistinguishable from samples collected from the 2006 lava flow at 9 46'56'N, suggesting that a relatively recent eruption occurred north of the Bull's Eye area. Second, within the Bull's Eye itself, the 2008 tow-yo survey reveals light attenuation values that are significantly lower than values measured soon after the 2006 eruption (Cowen et al., EOS 2007), and the plume height is 100-140 m deeper overall than it was in 2006, confirming the strong temporal connection between hydrothermal venting and magmatic activity. We also investigated for possible hydrothermal plumes 8 km and 39 km east of the axis, at 9 30'N and 9 20'N, respectively. Although results from seismic tomography suggest the presence of crustal and subcrustal melt anomalies in these two areas (Toomey et al., Nature 2007), no hydrothermal plumes were identified from the two tow-yo profiles.
B21A-0338
Iron Isotope Systematics in Hydrothermal Plume Fall-out at EPR 9°50' N
We have initiated a study as part of NSF's Ridge 2000 program to investigate the fate of dissolved FeII released from high-temperature hydrothermal venting on the East Pacific Rise and its potential impact on local deep-ocean Fe-isotopic and biogeochemical budgets. Fe is released in abundance from high-temperature vents and plays a major role in modifying the gross flux of other elements from hydrothermal vents to the oceans. In addition, hydrothermal sources have the potential to dominate the global deep ocean budget of dissolved Fe, an essential micronutrient (Bennett et al., 2008; Chu et al., 2006). Here, we investigated Fe isotopes systematics in hydrothermal plume products, especially through Fe-sulfide precipitation and Fe oxidation, in order to determine whether biotic and/or abiotic processes might impose a characteristic hydrothermal Fe-isotope 'fingerprint' to the plume fall-out material deposited to the surrounding seabed. To achieve our goals we collected plume samples close to vent-sites at 9-10°N on the East Pacific Rise over a period extending from late 2004 to late 2007 bracketing an episode of recent volcanism (Winter 2005-2006). Thus, we are in a position to investigate both the "steady state" input from a deep-ocean vent- system and also how that system has been perturbed following volcanic activity out to a period of 1-2 years post-eruption. Sampling was conducted using time-series sediment traps deployed on long-term moorings which sampled particles settling from buoyant and non-buoyant plumes at 9°50'N. At EPR 9-10°N, we demonstrate that Fe isotope fractionation in the buoyant hydrothermal plume occurs during the formation of Fe-sulfide precipitates. The δ 56Fe values range from 0.0 to - 0.9‰ with an average of -0.56‰ (n=72), which is significantly lower than the end-member hydrothermal fluids (δ 56Fe between -0.3 to -0.4‰, Rouxel et al., 2008). These results contrast previous work at the Rainbow vent field on MAR, where sediments beneath the plume are found to be indistinguishable from the vent fluid, and where Fe particles within the buoyant and non-buoyant plume exhibit values from 0.1 to 1.1‰ relative to the vent fluid, consistent with partial oxidation of FeII to FeIII, the (Severmann et al., 2004). Our analyses of major and trace element flux variations suggest 5 major components in the plume: (i) hydrothermal Fe-oxides, scavenging P, V and REE (ii) Fe-sulfide precipitates enriched in Cu, Co, Zn and Cd (iii) hydrothermally-affected marine sediment enriched in Mn, Th and Ba (iv) lithogenic components with Al and Ti enrichments (v) carbonate controlling Ca and Sr enrichment. These results have been used to assess the relative contributions of each the plume components to Fe-isotope composition of plume fall-out and allowed to establish a preliminary Fe-isotopic mass balance of hydrothermally derived Fe in the deep sea.
B21A-0339
Hydrothermal plume particles deconstructed: evidence of biotic and abiotic interactions in particle formation at 9N East Pacific Rise
We are using non-buoyant hydrothermal plume samples collected at 9° 50' N East Pacific Rise to
conduct an unprecedented examination of the fine scale (μm- and nm-range) mineralogical and
biogeochemical composition of hydrothermal particulates. Recent findings from studies of sinking particulate,
suggest that East Pacific Rise hydrothermal particulates are inorganic/organic aggregates. We confirmed this
with suspended particulate samples collected, during a 3-day moored multi-sampler deployment in the non-
buoyant hydrothermal plume above Tica vent at 9° 50' N East Pacific Rise in November 2007. A
combination of high energy synchrotron x-ray absorption spectroscopy and bulk and trace elemental analysis
reveal that the >1 μm suspended particulates consist of inorganic Fe oxide grains in a pervasive
organic C matrix with carbon K-edge spectra consistent with proteins, lipids, and polysaccharides. This
aggregate structure may preserve reduced Fe phases in the presence of oxygenated seawater and alter our
basic assumptions about hydrothermal particle dispersion. Scavenging, by coprecipitation and surface
adsorption, of seawater nutrients and trace elements such as P, U, Mo, Cr, V, and As may also be influenced
by this particulate composition. These samples were collected with a newly developed Suspended Particulate
Rosette multi-sampling system designed to collect geochemical and microbial samples from rising deep-sea
hydrothermal plumes to enable investigations of abiotic and biotic plume processes.
http://www.whoi.edu/people/jbreier