Tectonophysics [T]

T51F  MW:3020   Friday
The Generation of Oceanic Lithospheric in Areas of Low Effusive Magmatism: Where Has All the Crust Gone? I
Presiding: R Searle, Durham University; M Tivey, Woods Hole Oceanographic Institution

T51F-01 INVITED 

Development of Oceanic Core Complexes on the Mid-Atlantic Ridge at 13-14N: Deep-Towed Geophysical Measurements and Detailed Seafloor Sampling

* Searle, R (r.c.searle@durham.ac.uk) MacLeod, C (macleod@cardiff.ac.uk) Murton, B (bjm@noc.soton.ac.uk) Mallows, C (Christopher.Mallows@durham.ac.uk) Casey, J (jfcasey@uh.edu) Achenbach, K (KAY@uwyo.edu) Unsworth, S (samwise@noc.soton.ac.uk) Harris, M (mh9@noc.soton.ac.uk)

The first scientific cruise of research vessel James Cook in March-April 2007 targeted the Mid-Atlantic Ridge at 13-14°N, to investigate details of lithospheric generation and development in a low-magmatic setting. Overall objectives were to 1) investigate the 3D pattern of mantle upwelling and melt focusing; 2) study how plate accretion and separation mechanisms differ between magma-rich and magma-poor areas; and 3) test mechanisms of detachment faulting and extensional strain localisation in the lower crust and upper mantle. Smith et al. (Nature 2006) had shown this to be an area of widespread detachment faulting and formation of oceanic core complexes (OCC), and published bathymetry showed an extensive area of blocky rather than lineated topography, which elsewhere has correlated with areas of low effusive magmatism. We conducted a TOBI deep-towed geophysical survey over a 70 km length of ridge extending to magnetic chron C2n (1.9 Ma) on each flank. This included sidescan sonar and high resolution bathymetry and magnetic measurements on 13 E-W tracks spaced 3 - 6 km apart. The area includes 1 active, 1 dying, and 1 defunct OCC and borders well-lineated, apparently magmatically robust seafloor to the north. The geophysical survey was complimented by recovery of 7 oriented and 18 unoriented core and 29 dredge samples, including some from a probable OCC south of the TOBI survey. Deep-towed sidescan, bathymetry and video show the OCCs typically comprise a steeply outward tilted volcanic ridge marking the breakaway (as suggested by Smith et al., 2006); a high, rugged central massif that is complexly deformed as a result of uplift and bending, and may be separated from the breakaway ridge by what we interpret as a late outward dipping normal fault; and a smooth, corrugated surface that generally dips c. 20° towards the ridge axis at the termination but gradually rotates to horizontal or gently outward dipping near its junction with the central massif. Older OCCs have flatter central areas than the youngest OCC, perhaps as a result of unbending late in their development. They also have steep, ridge-dipping normal faults at or near their terminations that may play a role in ending the detachment faulting. The axial neovolcanic zone is absent or poorly developed opposite the active and dying OCCs, and appears to be propagating towards them, thus playing an important part in terminating OCC activity. We recovered peridotite from the central parts of the three northern OCCs, the median valley walls near 13°00'N, and the neovolcanic ridge near the OCC at 13°20'N. Basalt was recovered from the breakaway ridges and often the central massifs. Most of the corrugated surfaces were extensively covered in sediment and basaltic rubble which prevented us obtaining drill samples, though dredging recovered peridotite, fault rocks and basalt rubble. The toes of the 13°30'N and 13°20'N OCCs showed evidence of extensive and some recent hydrothermal activity. Basalt and fresh glass were recovered from axial volcanic ridges. We recovered very little gabbro. A reasonably coherent magnetic central anomaly is seen in the sea-surface field, but the higher-resolution deep- towed magnetic data reveal a pattern of crustal magnetisation that does not obviously match a simple reversal history, even in areas away from obvious OCCs. We are modelling these results in terms of asymmetric spreading, ridge jumping and tectonic rotations varying in time and space.

T51F-02 INVITED 

Geological Context of Ultramafic-Hosted Hydrothermal Vent Fields in the 13-15°N Region of the Mid Atlantic Ridge : Preliminary Results of the Serpentine Cruise.

* Cannat, M (cannat@ipgp.jussieu.fr), Institut de Physique du Globe, CNRS, 4 place Jussieu, Paris, 75252, France, Metropolitan Ondreas, H (hondreas@ifremer.fr), Ifremer, BP 70, Plouzane, 29280, France, Metropolitan Fouquet, Y (yves.fouquet@ifremer.fr), Ifremer, BP 70, Plouzane, 29280, France, Metropolitan Silantiev, S (silantyev@geokhi.ru), Vernadsky Institute, Academy of Sciences, Kosygin av, Moscow, 119991, Russian Federation Hoise, E (hoise@geologie.ens.fr), Ecole Normale Superieure de Paris, 24 rue Lhomond, Paris, 75005, France, Metropolitan Fontaine, F (fontaine@ipgp.jussieu.fr), Institut de Physique du Globe, CNRS, 4 place Jussieu, Paris, 75252, France, Metropolitan Scientific Party, S (yves.fouquet@ifremer.fr

During the SERPENTINE cruise (feb. 25 to apr. 5, 2007) on board the new RV Pourquoi Pas?, we used the VICTOR 6000 ROV to conduct a multidiciplinary (geology, fluid studies, biology) exploration of the Ashadze 1 and 2 (12deg58N) and Logatchev 1 and 2 (14deg43N and 14deg45N) hydrothermal fields. ROV observations and sampling were complemented by a night program of dredging, water sampling and underway geophysics. The Ashadze and Logatchev fields are located respectively on the west and east flank of the Mid-Atlantic Ridge, on outcrops of mantle-derived peridotites and interspersed gabbroic rocks. We present our principal findings in terms of geology, focusing on the geological context of hydrothermal vents. Ashadze comprises two active vent fields located at different levels on the western wall of the axial valley near 13°N : Azhadze-1 at 4000 m depth, and Ashadze-2 at 3300 m. We also sampled extinct sulfide chimneys near the base of the axial valley wall at 4530 m depth. The top of the wall, at 2300 m, corresponds to the termination of a large fossil corrugated surface. The axial valley at this latitude is strongly asymmetric, with higher relief to the west. This asymmetry is reversed immediately to the south, where the axial magnetic anomaly appears offset by a few kilometers to the west. The active and extinct Ashadze vents are roughly aligned to the north of this minor axial discontinuity. We find similarities between this general context, and the setting of the two Logatchev active vent fields: Logatchev-1 on the east axial valley wall near 14deg45N, and the smaller Logatchev-2 in a seemingly off-axis position near 15deg43N. Both fields lie to the north of a small offset axial discontinuity, and in an inward position relative to fossil corrugated surfaces. Based on seafloor morphology, dive observations, and rock sampling, we develop a model whereby ultramafic-hosted hydrothermal venting in the 13-15N region of the MAR involves both large active normal faults, and an inside corner-type position relative to a small offset of the volcanic axis. We discuss the relevance of this model for the TAG hydrothermal field at 26N, and for the Rainbow field at 36deg10N.

T51F-03 

Diversity of Ultramafic Hosted Hydrothermal Deposits on the Mid Atlantic Ridge: First Submersible Studies on Ashadze, Logatchev 2 and Krasnov Vent Fields During the Serpentine Cruise.

* Fouquet, Y (fouquet@ifremer.fr), IFREMER, BP 70, Plouzane, 29280, France Cherkashov, G), VNIIOkeangeologia, Angliysky Avenue, St.Petersburg, 190121, Russian Federation Charlou, J), IFREMER, BP 70, Plouzane, 29280, France Ondreas, H), IFREMER, BP 70, Plouzane, 29280, France Cannat, M), IPGP, CNRS UMR 7154, 4 place Jussieu, Cedex 05, Paris, 75252, France Bortnikov, N), IGEM, Staromonetny per., 35, Moscow, 119017, Russian Federation Silantiev, S), Vernadsky Institute, 10 Kosigin Street, Moscow, 119991, Russian Federation Etoubleau, J), IFREMER, BP 70, Plouzane, 29280, France Scientific Party of the SERPENTINE cruise

During the Serpentine cruise (March 2007) we have explored and sampled, using the ROV Victor, new ultramafic hydrothermal fields between 13°N and 17°N on the Mid Atlantic Ridge (MAR). The Serpentine cruise was part of a 4 years cooperation agreement between France and Russia. Targets were Ashadze1 and 2 (12°58"N), Logatchev 1 (14°45"N) and 2 (14°43"N) and Krasnov (16°38"N) fields localized after several surface cruises of the R/V professor Logatchev. A significant portion of the dives was dedicated to detailed microbathymetry, 50 m and 20 m above the seafloor, and simultaneous physical and chemical plume studies and magnetic surveys. High resolution (30cm) maps were further used for geological, biological, microbiological and fluid sampling operations. The cruise identified three new very active black smoker fields (Ashadze 1 and 2, Logatchev 2) on serpentinized peridotites. One extensive low temperature inactive deposit (dominantly birnessite) was discovered 1 km east of the Logatchev 1 field. The basaltic hosted Krasnov field was inactive. The Ashadze 1 site at 4080m of water depth is the deepest active black smoker field so far known in the ocean. Inactive and basalt hosted sulfide chimneys (Ashadze 4) were found at the base of the rift valley at 4530 m. Extensive gravity sliding related to the emplacement of the ultramafic rocks is evident at all ultramafic sites (see abstract by Ondreas et al.). Fluids, enriched in H2 and hydrocarbon, confirm the originality of ultramafic environments (see abstract by Charlou et al.). Logatchev 2 is venting low salinity black smoker fluids indicating phase separation. In addition, its position 12 km off axis, moves from 8 (Logatchev 1) to 12 km the possibility to have off axis black smokers long the MAR. Basaltic hosted deposits are dominated by pyrite and silica at Krasnov (Fe:39%, Si:11%, Cu:2.2%, Zn:0.14%) and by sphalerite and pyrite at Ashadze 4 (Fe:24%, Si:1.5%, Cu:0.15%, Zn:32%). Ultramafic deposits are characterized by high copper concentration dominated by chalcopyrite and isocubanite. Ashadze 1 (Fe:33%, Si:1.3%, Cu:14%, Zn:14%) and Logatchev 2 (Fe:20%, Si:3%, Cu:14%, Zn:23%) are enriched in sphalerite. New samples at Logatchev 1 confirm that copper is largely dominant at this site (Fe:29%, Si:3%, Cu:28%, Zn:4%). The Ashadze 2 field is unusual. A small active crater can be interpreted as a hydrothermal volcano built up with a mixture of carbonates and secondary copper sulfides and chlorides. Massive sulfide chimneys are associated with the active smokers at the center of the crater. Many inactive carbonates/sulfides mounds are also aligned along a N-S depression. Two types of hydrothermal deposits are observed: massive copper-rich sulfides associated with the black smokers and carbonate/sulfides chimneys. Average composition of hydrothermal deposits for the field is Fe:26%, Si:11%, Cu:11%, Zn:5%, Ca:8%. The dominant carbonate is aragonite, Mg-Calcite is rare, and talc is common. Comparisons with other ultramafic sites along the MAR will also bee presented. http://www.ifremer.fr/serpentine

T51F-04 

High Hydrogen and abiotic hydrocarbons from new ultramafic hydrothermal sites between 12°N and 15°N on the Mid Atlantic Ridge- Results of the Serpentine cruise (March 2007)

* Charlou, J (charlou@ifremer.fr), Département Géosciences Marines, IFREMER C/Brest, Plouzané, 29280, France Donval, J), Département Géosciences Marines, IFREMER C/Brest, Plouzané, 29280, France Konn, C), Département Géosciences Marines, IFREMER C/Brest, Plouzané, 29280, France Konn, C), Department of Geology and Geochemistry, Stockholm University, Stockholm, SE-106 91, Sweden Birot, D), Département Géosciences Marines, IFREMER C/Brest, Plouzané, 29280, France Sudarikov, S), VNIOkeanologia, Anglivsky Avenue, St Petersburg, 190121, Russian Federation Jean-Baptiste, P), LSCE, CEA/Saclay, Gif-sur-Yvette, 91191, France Fouquet, Y), Département Géosciences Marines, IFREMER C/Brest, Plouzané, 29280, France Scientific Party of the SERPENTINE cruise

New hydrothermal fields were recently explored and sampled between 12° and 17°N (MAR) on the Mid-Atlantic Ridge as a part of the French-Russian cooperative SERPENTINE diving cruise. In addition to the Logachev I site (14°45N) previously studied and revisited during this cruise, new smoking areas, called Ashaze I and 2 (12°58N) and Logachev 2 (14°43N) were discovered in ultramafic environments, as previoulsly found at Rainbow (36°14N), Lost City (30°N), and Logachev 1 (14°45N). Very strong anomalies in temperature, nephelometry, CH4 (from 1 to 120 µl/l), helium, were found in the seawater column above the Ashaze and Logachev high-temperature fields. The fluid endmembers at these sites exhibit different temperature (310 to 370°C), and different chemical characteristics: pH (3.5 to 4.0), chloride (150 to 620 mM), signifying that phase separation is occurring and controlling the fluid chemistry. All fluids are issued from ultramafics and controlled by seawater-peridotite interaction They show low silica (5 to 10 mM), low H2S (<0.5 mM) and are extraordinary enriched in hydrogen gas. Gas bubbles are observed coming out from Ashaze 1 vents and pulses of clear fluid were observed venting from Logachev 2. All fluids issued from ultramafics contain very high concentrations in H2 (70 per cent of total gas), CO2, CH4. Preliminary calculations show that one vent at Ashaze 1 field produces 1 million of cubic meters of natural H2 per year. CH4 is clearly abiogenic with d13C varying from –6 to –14 per mil (PDB). In addition, the progressive isotopic trends for the series of C1 to C4 alkanes indicate that hydrocarbon formation occurs by way of polymerization of CH4 precursors. The serpentinization process is observed here up to 4080m at Ashaze 1, the deepest venting area so far know in ocean, and generates high hydrogen and abiogenic hydrocarbons during the hydration of olivine and pyroxen minerals through catalytic reactions (Fischer-Tropsch type reactions) as previously observed at all ultramafic sites on the slow-spreading mid Atlantic Ridge between 12°N and the Azores Triple Junction. More information is available at http://www.ifremer.fr/serpentine.

T51F-05 INVITED 

The Crustal Section Exhumed by Oceanic Detachment Faults

* Tucholke, B E (btucholke@whoi.edu), Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, United States Behn, M D (mbehn@whoi.edu), Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, United States Canales, J (jcanales@whoi.edu), Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, United States Xu, M (minxu@mit.edu), MIT/WHOI Joint Program in Oceanography, Massachusetts Institute of Technology, Cambridge, MA 02139, United States Buck, W R (buck@ldeo.columbia.edu), Division of Marine Geology and Geophysics, Lamont-Doherty Earth Observatory, Palisades, NY 10964, United States Lin, J (jlin@whoi.edu), Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, United States

Although normal faults are ubiquitous on mid-ocean ridges, they develop larger offsets and thus expose oceanic core complexes at slower-spreading ridges because the average amount of melt accreted at the spreading axis is reduced and tectonic extension consequently increases. The longest-lived (detachment) faults slip for as long as 1-2 m.y. and exhume hundreds of square kilometers of lower ocean crust and upper mantle in the fault footwalls. To maintain isostatic equilibrium, these footwalls roll over and form megamullions that are characterized by domed shapes and by large, enigmatic, fault-surface corrugations (mullions) as well as striations that parallel fault-slip direction. Over the past decade, dozens of megamullions have been identified on mid-ocean ridges spreading at slow to intermediate rates, and these features offer tremendous potential for study of the internal structure, lithologic architecture, and alteration of the oceanic lithosphere. Because detachment faulting suggests extreme tectonic extension, we expect to see little manifestation of magmatism in these tectonic windows. However, recent studies show that at least some megamullions form in association with emplacement of large gabbro bodies and thus with apparently elevated magmatism. Here we present geological and geophysical data, together with numerical modeling results, to show that long-lived detachment faults are likely to form megamullions only when ca. 30-50 percent of total extension is accommodated by magmatic accretion. Under these conditions magmatism may focus unevenly along the spreading axis and create an irregular brittle-plastic transition where detachments are rooted, thus explaining the initiation of large fault corrugations. If magmatism persists, it may continue to focus toward the original injection points, thus tending to orient gabbro bodies parallel to footwall corrugations. In the absence of continuing magmatism, corrugations may still be formed by the cool hanging-wall mold. The morphological and compositional characteristics of the oceanic lithosphere suggested by this study provide important constraints to assess the distribution of magmatic versus tectonic extension along mid-ocean ridges.

T51F-06 

The Sub-Seafloor Structure of Mid-Atlantic Ridge Core Complexes

* Canales, J (jpcanales@whoi.edu), Woods Hole Oceanographic Institution, 360 Woods Hole Rd., Woods Hole, MA 02543, United States Xu, M (minxu@mit.edu), Massachusetts Institute of Technology-Woods Hole Oceanographic Institution Joint Program, 77 Massachusetts Ave., Cambridge, MA 02139, United States Tucholke, B E (btucholke@whoi.edu), Woods Hole Oceanographic Institution, 360 Woods Hole Rd., Woods Hole, MA 02543, United States Collins, J A (jcollins@whoi.edu), Woods Hole Oceanographic Institution, 360 Woods Hole Rd., Woods Hole, MA 02543, United States Dubois, D L (ddubois@whoi.edu), Woods Hole Oceanographic Institution, 360 Woods Hole Rd., Woods Hole, MA 02543, United States

Oceanic core complexes form by long-lived extension along detachment faults (in some instances for periods more than a million years), exposing sub-volcanic lithosphere at the seafloor. Seafloor sampling and deep drilling indicate that highly altered mantle peridotites and lower crustal gabbros are the dominant lithologies in these structures, but their relative abundance and distribution beneath the detachment surfaces are not well constrained, which hinders our understanding of the origin, composition, and evolution of core complexes. Here we present results from a geophysical study aimed at determining the sub-seafloor P-wave seismic velocity structure of three of the best developed and most studied oceanic core complexes (CC) found on the Mid-Atlantic Ridge: Kane CC located on ~3.3-2.1 Ma lithosphere of the North American plate immediately to the south of Kane FZ (23° 20'-37'N), Dante's Domes CC located on ~2.1-0.7 Ma African-plate lithosphere at 26° 35'-45'N, and Atlantis CC located on ~2.0-0.8 Ma lithosphere of the North American plate immediately to the north of Atlantis FZ (30°05'-20'N). We use seismic data acquired in 2001 along and across the three core complexes using the 6-km-long hydrophone streamer and air-gun array of the R/V Ewing (cruise EW0102). The dense sampling of sources and receivers and the relatively shallow seafloor of the study areas allow us to perform traveltime tomography inversions to image lateral variations in seismic velocity at lateral scales of 1 km or less within the upper ~0.5-1.5 km of the lithosphere. Our results show that the seismic velocity structure within each CC is highly heterogeneous, but that remarkably similar patterns in velocity structure exist among the three CCs. In a broad sense, the velocity structure of each of the three CCs is characterized by three distinct patterns: (1) areas with relatively high velocities (>4 km s-1) and high velocity gradients (>3 s-1) near the seafloor, (2) areas with moderate seafloor velocities (3.5-4 km s-1) and less-pronounced velocity gradients (~1.25- 2.5 s-1), and (3) areas with low seafloor velocities (<3.5 km s-1) and low velocity gradients (<1.25 s-1). We interpret the lithological correlation of the velocity structures by integrating our results with the extensive seafloor geological sampling available from the Kane CC and the geological sampling and deep drilling results at the Atlantis CC (IODP Hole U1309D). The lowest seismic velocities/gradients generally correspond to volcanic terrain and an abundance of in situ pillow basalts; areas with moderate velocities/gradients correspond to predominance of highly altered mantle peridotites; and the highest velocities/gradients indicate a predominance of gabbros. If our interpretations are correct, then the observed seismic velocity patterns indicate that the corrugated CCs have exhumed both gabbro and mantle-derived serpentinite sections. These sections can be as large as 5-20 km wide in plan view, and they exhibit complex lateral variations in both strike and dip directions.

T51F-07 

Styles of Detachment Faulting at the Kane Oceanic Core Complex, 23°N Mid-Atlantic Ridge

* Hansen, L N), University of Wyoming, Department of Geology and Geophysics, Laramie, WY 82071, United States Cheadle, M J), University of Wyoming, Department of Geology and Geophysics, Laramie, WY 82071, United States John, B E), University of Wyoming, Department of Geology and Geophysics, Laramie, WY 82071, United States Swapp, S M), University of Wyoming, Department of Geology and Geophysics, Laramie, WY 82071, United States Dick, H J), Woods Hole Oceanographic Institution, Department of Geology and Geophysics, Woods Hole, MA 02543, United States Tucholke, B E), Woods Hole Oceanographic Institution, Department of Geology and Geophysics, Woods Hole, MA 02543, United States Tivey, M A), Woods Hole Oceanographic Institution, Department of Geology and Geophysics, Woods Hole, MA 02543, United States

In 2004, R/V Knorr Cruise 180-2 used ROV Jason II, the autonomous vehicle ABE, and dredges to collect samples and geophysical data from the Kane Oceanic Core Complex (OCC) on the Mid-Atlantic Ridge at 23°N. Examination of the deformed samples by hand-sample analysis, petrography, electron backscatter diffraction, and geothermometry in conjunction with the interpreted bathymetry suggests that the Kane OCC is bound by a detachment fault system that initiated at a moderate to high angle (45-60°), and rooted below the brittle-plastic transition. Constraints on the initial dip of the detachment fault come from the slopes of the ridge forming the breakaway (>23° to the west and >22° to the east). Assuming this ridge formed by flexural uplift, these slopes suggest the detachment fault formed with a dip >45°. Fault rocks, including peridotite mylonites and gabbro ultramylonites, reveal a history of deformation from granulite and amphibolite through sub-greenschist facies including brittle cataclasis. We present two cross sections through the detachment fault and footwall based on samples collected from secondary, high-angle normal faults that cut the detachment. One section, through Cain Dome in the central OCC, is dominated by peridotite and shows a ~450-m thick zone of discrete ductile shear zones with the uppermost portion overprinted by a 200-m zone of semi-brittle and brittle deformation. These are maximum shear zone thicknesses due to the possibility of down-scarp slumping/displacement. The other section, through Adam Dome on the southwest part of the OCC, is dominated by gabbroic rocks and shows little deformation. This section lies <4 km from the breakaway, and is therefore inferred to have undergone only brittle deformation in the shallow crust. A rheologic analysis, using LPO-deduced deformation mechanisms and geothermometry to construct deformation mechanism maps, suggests strain rates for the amphibole-bearing gabbros, the gabbronorites, and the peridotites of 10- 10s-1, 10-12s-1, and 10-13s-1 respectively. These strain rates reflect differing amounts of strain localization during the evolution of the detachment fault. The presence of deformed and undeformed diabase dikes, peridotite mylonites that have been intruded by gabbroic melts, and Fe-Ti oxide microstructures indicating deformation with melt present all suggest that detachment faulting was coeval with magmatism. In summary, the bathymetry of the Kane OCC, the initial moderate-steep dip of the fault, and the depth constraints provided by the fault rocks are most consistent with a rolling-hinge detachment fault model.

T51F-08 

Reanalysis of Geophysical Data With IODP Constraints at Atlantis Massif

* Blackman, D K (dblackman@ucsd.edu), Scripps Institution of Oceanography, UCSD, 0225, La Jolla, CA 92093, Collins, J A (jcollins@whoi.edu), WHOI, MGG, Woods Hole, MA 02543, Searle, R C (R.C.Searle@durham.ac.uk), Durham University, Dept Earth Sciences, Durham, DH1 3LE, United Kingdom Karner, G D (garry@ldeo.columbia.edu), Exxon Mobil, Upstream Research Co, Houston, TX 77252-2189,

The recovery of a dominantly gabbroic sequence during IODP Expeditions 304/305 at the oceanic core complex on the Mid Atlantic Ridge 30°N (Atlantis Massif, AM) motivated re-analysis of pre-existing geophysical data together with new borehole logging data. Re-assessment of prior interpretations of the domal high of AM as an ultramafic, flexurally uplifted footwall was warranted. Residual gravity anomalies corrected for seafloor topography and lithospheric temperature structure can be accounted for by a 3D gabbroic core, consisting of the central dome and southern ridge of AM, and a basaltic hanging wall. Mantle Bouguer anomalies are best fit by a model where the core terminates in the north at ~30°15'N where the corrugated detachment surface ends and a series of steep normal faults characterizes the median valley wall to the north. Prior OBS refraction analyses implied the existence of a 7.5 km/s layer (appropriate for fresh peridotite) at 600 m sub-seafloor and some along- strike variability in the shallower, lower velocity layers. This was consistent with known outcrops of serpentinized harzburgite on the southern ridge. A new analysis of the OBS data suggests somewhat greater along-strike variability in layer thicknesses but with velocities in the 500 m deep section constrained by the ray tracing generally less than 6 km/s, a mafic rather than ultra-mafic rock. This model is similar to a prior best-fit model of data along another NOBEL refraction profile to the east on the dome. Unfortunately, the shot spacing and short (2 km) NOBEL line length do not allow discrimination between the various refraction models. Analysis of refracted arrivals recorded on the MCS streamer using a new method (see Harding et al., Session S12, Fall AGU07) will provide the basis of an inversion for the shallow velocity structure of the central dome near IODP Hole U1309D. This tomographic model will be reported and, together with drilling results, the implications in terms of variability within the core of Atlantis massif will be discussed.