Case for Support: Part 2. Description of Project
EXECUTIVE SUMMARY
Central to models of ocean lithosphere generation is the concept that upwelling of mantle is focused beneath the centres of spreading segments, at least at slow-spreading mid-ocean ridges (MOR)1. As a result, the extent of mantle melting and melt migration are predicted to vary systematically along ridge segments. However, these concepts have never been tested directly and there is little agreement as to the scale, mechanism, or extent to which focusing of mantle and/or melt occur (Fig. 1)2-4. Few opportunities exist to test the models directly because the mantle is normally covered by a magmatic ocean crustal layer. Exposure of mantle peridotite is only widespread at a few ‘magma-starved’ regions where the magmatic crust is thin, discontinuous, or even absent altogether3,5. Fortunately, such areas are those at which focused upwelling should be most marked.
Focused upwelling beneath the ridge inevitably gives rise to along-axis variability in the thermal structure of the lithosphere, which in turn has a profound effect on its mechanical behaviour. Lithosphere at slow-spread segment ends is known to be thicker, hence cooler and stronger, and tectonism is inferred to play a much more significant role in accommodating plate separation3,6. The recent discovery of low-angle detachment faults in these environments7-10 provides a new insight into how this ‘amagmatic’ or ‘tectonic’ spreading occurs. The mechanisms by which such detachments form and operate are contentious8,10,12, yet important to understand the accommodation of strain in the lithosphere.
We propose to test the focused mantle/melt upwelling hypothesis directly by drilling a large number of orientated peridotite rock cores from the median valley of the Mid-Atlantic Ridge (MAR) in the vicinity of the Fifteen-Twenty fracture zone (FTFZ) using the ‘BRIDGE’ seabed rock drill. The FTFZ region is the best example of a ‘magma-starved’ mid-ocean ridge (MOR) yet known, with extensive exposures of peridotite for >50km either side of the fracture zone13,14. The area has been chosen as a top priority for international investigation15, and Ocean Drilling Program (ODP) Leg 209 was sited here in May-July 200316,17 (see below). With the cores we collect we aim to reconstruct spatial variations in the geometry of mantle upwelling, the degree of melting, and the distribution of melt channels and melt relicts, with respect to position in the spreading segments. The extent to which these vary is key to distinguishing the competing models (summarised in Fig. 1).
To accompany the drill sampling we propose a deep-towed sidescan sonar survey of the FTFZ area. The sonar images are necessary to enable drillsite selection and, more importantly, allow us to map out the areas of peridotite outcrop, reveal spatial variations in the distribution of magmatic bodies and, from variations in ridge morphology and distribution and nature of faulting, test models of strain accommodation during ‘amagmatic’ spreading.
Figure
1.
Cartoon summary of end-member models for mantle upwelling: (a) Broad
upwelling: mantle wells up passively everywhere beneath the ridge axis,
with variable melt supply and degree of melting along axis depending on the
depth of the melting zone. Mantle fabrics are predicted to be everywhere
vertical. (b) Focused upwelling: active (buoyancy-driven) mantle
upwelling is confined to the segment centre but then overturns and is
transported horizontally along axis at shallow levels. This model predicts
variation of flow fabrics from vertical to horizontal along axis, and a similar
degree of melting everywhere. Note that the radius of such upwelling zones is
unknown; estimates vary from 10s to 100s of kme.g.1-4,16.
PREVIOUS WORK
The FTFZ area has been extensively sampled by dredging, manned submersible, seabed rock coring and now ODP drilling4,10-14,16-18 (Fig. 2). Extensive areas of peridotite crop out continuously for tens of kilometres in and around the median valley (MV) both north and south of FTFZ. As expected, basalts predominate within the MV well away from FTFZ in the south (south of 14°40’N), but are also common immediately north of FTFZ, counter to simple ideas that ‘magmatic’ crust near segment boundaries gives way to ‘amagmatic’ crust near segment ends3. Gabbros have been recovered in ca. 25% of dredges and cores10,12,17-19, but their regional pattern of outcrop remains unclear. In view of this, the term ‘amagmatic’ may be misleading, but we use it, and ‘magma-starved’, as shorthand for areas with high positive gravity anomalies, extensive peridotite and an extrusive layer that is thin or absent.
High
quality multibeam bathymetry exists (Fig. 2), with accompanying sea-surface
gravity and magnetic data4,10,12,18.
North of 15°40’N and south of 14°30’N the area has the attributes of normal slow-spreading crust. There are negative residual mantle Bouguer anomalies (RMBA) and a strongly lineated abyssal hill terrain, as found elsewhere on the MAR and well understood by current models3,20. The areas of peridotite outcrop show a marked contrast, and we suggest that the ridge accretion and spreading processes there are significantly changed from the “normal” model. There is a strong, positive RMBA, suggesting that it is associated with ‘amagmatic’ crust4,18. Some basalt crops out in these areas, and both submersible observations and ODP site 1272 show basalts directly capping gabbro or ultramafics14,17, though the exact structural context is unknown. The topography of the ‘amagmatic’ areas is quite different: it is more rugged, with few abyssal hills but frequent short inward-, outward- and oblique-dipping scarps14, indicating absence of the usual well-organised ridge-parallel faulting. Such structures are seen sporadically at segment ends elsewhere at slow-spreading MOR; however at FTFZ they are sufficiently extensive to make a comprehensive study of them possible. The boundaries between lineated and rugged terrains have steep gravity gradients and appear to have migrated north and south along axis over periods of ~2Ma (Fig. 2).
Finally, the bathymetry reveals the presence of four corrugated surfaces interpreted as detachment faults over ultramafic core complexes4,10,12,18 (Fig. 2). Unusually, three of the detachments occur on RTI outside corners (OC), not inside corners (IC) as predicted by current models8,9. One of these detachments, at 15°45’N 46°55’W, was the subject of a highly successful TOBI and seabed rock drill study by one of us (CJM) in 200110,12. The sidescan data reveal a complexly striated detachment, a major escarpment flanking the MV floor, and minor normal faults more typical of segment centres. Only ODP sites 1274 and 1275 fall within the existing sidescan coverage. From 73 successful sites drilled during our seabed rock drill survey we showed that the detachment surface exposes peridotite and dolerite, overlying a discrete gabbro body, with extensive fault rock deformed entirely in the brittle regime10,12.
JUSTIFICATION for the PRESENT project
The objectives of ODP Leg 209 were broadly similar to our own first objective (see below): viz. to test mantle upwelling models by drilling an along-strike transect of holes either side of the FTFZ. Whereas the ODP drilling results will enormously improve our understanding of the upper mantle beneath ridges in a number of ways, the Leg 209 scientists19 confirm that they were unable to address this principal objective satisfactorily because of the technical limitations of the JOIDES Resolution drillship: it was difficult to get holes started, and most soon collapsed. Although eight sites were eventually drilled during Leg 209, substantial (>100m) sections of peridotite were recovered at only three, with minor peridotite in three more17. Spacing between peridotite sites was closer to 25km than the 10km planned in all but one case. Although preliminary results suggest there is no clear variation in peridotite major element composition with position within a spreading segment17,19, the major conclusion was that the spacing of ODP sites was much larger than the length scale of variability in peridotite composition and distribution of melt pathways and melt relicts.
Leg 209 results show that the mantle upwelling hypotheses can only be tested by drilling a significantly larger number of holes spaced much closer together.
The second significant problem encountered during Leg 209 was in reconstructing the geometry of ductile flow in the peridotites. ODP cores are not azimuthally orientated and, from observation of anomalous inclinations of magnetic remanence in Leg 209 cores, it appears that they were subjected to significant tectonic rotations during exhumation19. Palaeomagnetic restorations in abyssal peridotites are particularly uncertain, as magnetic remanence directions are set during serpentinisation, late in their uplift and cooling history21. Despite the best efforts of the Leg 209 scientists19 it is simply not possible to deconvolve these tectonic rotations and thus reconstruct high-temperature mantle flow directions in unorientated drillcore without making unjustifiable assumptions about the axes, magnitudes and/or timings of rotations. The only way of reorientating ODP core independent of palaeomagnetism – by systematically matching core structural measurements with their corresponding azimuthally orientated borehole wall images obtained from geophysical well logs, thus allowing remanence directions to be measured directly (and hence tectonic rotations inferred)25 – will not be possible except in a very few cases because the poor borehole conditions prevented a full logging programme from being undertaken.
Fully orientated core material is essential if we are to reconstruct the geometry of high-temperature mantle upwelling. It is likewise the only way of documenting the kinematics of the later deformation and quantifying the tectonic rotations that led to exhumation of the peridotites.
We propose to obtain fully orientated cores by employing the British Geological Survey (BGS) BRIDGE rock drill, which is currently the only tool that can recover fully geographically orientated material from the seabed.
Our proposed study, using seabed rock drilling with the BRIDGE rock drill24 within a structural context determined from TOBI sidescan sonar, will complement and greatly enhance the value of ODP Leg 209. While TOBI is essential to obtain the structural setting of the drill sites, it additionally offers us the opportunity of addressing a number of other important problems concerning the geology of ‘amagmatic’ ridge areas.
Specific objectives
We have three major scientific objectives:
1. Test the hypothesis that mantle upwelling and melting is focused at the centres of slow spreading ridge segments and transposed by sub-horizontal flow away from there.
2. Test the hypothesis that plate accretion and separation mechanisms are fundamentally different in ‘magma-starved’ areas.
3. Test mechanisms of detachment faulting and extensional strain localisation in the lower crust and upper mantle.
These will be achieved by linked structural and geochemical studies of carefully sited cores of mantle rocks, coupled with high resolution mapping and imaging with TOBI sidescan sonar and magnetometer, in areas of severely diminished magmatic crust. These micro-studies will be supported by regional structural analyses based on high-resolution sidescan sonar and magnetic field studies that will yield information about the different volcanic and tectonic processes operating between magmatic and ‘amagmatic’ spreading episodes (see below for detailed methodology).
1. Mantle upwelling and melt focusing
This was a major objective of Leg 209 and its background is outlined above. However, the 10-25km site spacing proved to give inadequate horizontal resolution. Existing submersible and dredged samples are all unoriented, so cannot be used to investigate mantle flow or kinematics of later deformation. We will test the length scale of mantle flow patterns, melt migration features and heterogeneity of peridotite composition by analysing approximately 70 oriented seabed rock drill cores from along-axis transects in the MV at a variety of spacings.
Different models of mantle upwelling predict different patterns of mantle shear fabric and source residue depletion along axis. If mantle flow is highly focused at the segment centre and the residue is then transposed by sub-horizontal flow towards segment ends, the shear fabrics will become sub-horizontal towards segment ends but the degree of melting should be similar throughout the segment. However, if upwelling is more broadly focused but with a reduction in the extent of melting at segment ends, the melt fraction should progressively decrease towards the segment ends and the shear fabrics will be steeply inclined there. The origin and shear direction of mantle fabrics will be determined from oriented peridotite cores using electron backscatter diffraction (EBSD41) micro-fabric analyses (see Methodology section).
Variations in mantle melt fraction will be evident from changes in incompatible element abundance and ratios in basaltic glass and mantle cpx, and decreases in Mg# and Cr# in mantle pyroxene and spinel (see Methodology section).
2. Mechanisms of plate accretion and separation
There are three aspects to this:
2a) Low melt production vs. asymmetric accretion
An untested, and often unstated, assumption is that ‘amagmatic crust’ results from low magma budget due to decreased mantle melt fraction as a result of changes in mantle temperature or composition. An alternative is that melt fraction remains unchanged, but melt is distributed asymmetrically about the rift axis via asymmetric faulting, resulting in magmatic crust accreted to one plate but not the other26-28. We will test these hypotheses by sampling a spreading flow line on both sides of the MV, crossing the transition from volcanic to magma-starved crust as identified by existing and new sonar data. Changes in mantle melt fraction, variation in fractional crystallisation history, and links to mantle heterogeneity will be determined from geochemical and isotopic analyses of basaltic glass and relict mantle mineralogy. Inversion of TOBI magnetic field measurements will be used to infer thicknesses of the extrusive layer26.
2b) Tectonism
We will test models of how the tectonic fabric changes between the magmatic and amagmatic areas. There is a strong contrast between the tectonic styles of the ‘magmatic’ and ‘amagmatic’ areas as outlined above. One explanation for this difference is a variation in lithology and therefore rheology. We will correlate recovered lithologies with variations in tectonic style mapped by TOBI sidescan. Another explanation is the amount of crustal extension taken up by magmatic intrusion. If the ‘amagmatic’ area has fewer dykes, we predict greater tectonic extension than the 15-20% commonly seen on the MAR29,30. We will test this by measuring tectonic strain across the MV from high-resolution, near bottom sidescan sonar images, as we did at the MAR at 29°N30.
2c) Crustal architecture
The nature and mechanisms of peridotite emplacements remain unresolved: are they normal fault scarps, detachment fault surfaces, protruded serpentinite bodies, or bare, unfaulted mantle outcrops? Similarly, what are the spatial and temporal relationships between extrusives and mantle rocks in ‘amagmatic’ crust? Are extrusives emplaced primarily via fissure volcanism in axial volcanic ridges (AVRs) as elsewhere on the MAR31-33, or is the mechanism different during ‘amagmatic’ spreading?
If the variations in crustal architecture and its surface expression are due to variations in magma budget, there should be a gradient in melt fraction, volcanic thickness and tectonic properties between ‘magmatic’ and ‘amagmatic’ areas. If lineated terrain expands due to increased magmatic supply, we predict that both AVRs and normal faults should propagate into peridotite-dominated seafloor of the ‘amagmatic’ terrain. But if low-magma spreading is characterised by asymmetric rifting and accretion26, the boundary between the two terrains should be marked tectonically, e.g. by lateral transfer faults. We will distinguish between these models by using TOBI sidescan sonar to map volcanic and tectonic terrains34, complemented by magnetic field measurements to delineate and date extrusive areas26, and geochemical studies of melt source and evolution as in objective 1. These studies will bridge the serious mismatch between the resolution of remote surface observations (e.g. the ~100m footprint of multibeam bathymetry) and the scale of seafloor visual observations and ODP holes.
3. Mechanisms of detachment faulting and extensional strain localisation in the lower lithosphere
Most tectonic models predict that while mantle is exhumed along low-angle detachments, there is asymmetric plate accretion with little or no spreading on the opposite ridge flank (usually the OC), and that detachment faulting is initiated by a decrease in magma supply, expressed by declining extrusive production8. This will be tested by extending TOBI near-bottom magnetic observations22 across the detachment surfaces and their conjugate seafloor on the opposite side of the MV and inverting magnetic field to give the thickness of the extrusive layer26.
The FTFZ region is unusual in having detachments on the OC and locally on both flanks. We will cover both IC and OC detachments north of FTFZ to investigate how the detachment on the IC gave way, via ‘normal’ spreading, to a detachment on the OC. Assuming serpentisation and the associated magnetisation occurred during unroofing, we will date mantle exhumation by measuring offsets of magnetic chrons across detachment boundaries. We will also constrain the history of rotation of the footwall by palaeomagnetically determining the dip of remnant magnetisation in oriented cores across a detachment.
Our previous drill study of the 15°45’N low-angle detachment fault showed only extremely low-temperature (greenschist facies) brittle deformation10,12. Deformation was localised by weak hydrous minerals formed along the fault, and almost certainly aided by high fluid pressures. The detachment apparently slipped at a low angle in cold, shallow mantle lithosphere beneath the MV rather than by high-temperature precursory ductile deformation (as predicted in other detachment modelse.g.8). However, the question how such detachments can accommodate extension in the strong lower lithosphere remains unresolved. We will document mechanisms of strain localisation in the lower lithosphere by seeking intermediate- to high-temperature shear zones in pre-serpentinisation mantle and evidence of syntectonic, melt-assisted deformation and/or melt channelling.
METHODOLOGY AND APPROACH
Our primary tools are the BRIDGE wireline rock drill and TOBI. The drill is deployed from a conventional research ship on a conducting cable, has an on-bottom digital camera to aid site selection, and takes metre-length cores from bare-rock substrates. Cores are scribed with reference to compasses on the drill frame, allowing geographical orientation. We have demonstrated on two previous cruises that the drill can reliably take an average of six gabbro cores per day in water depths >4500 m on slopes up to 44° 10,36, proving it to be the paramount tool for oriented sampling in hard-rock terrains. Supplementary sampling on slopes too steep for the rock drill to be easily used will be undertaken using USBL (acoustic navigation) controlled dredging.
We propose to complement rock drilling with high resolution sidescan sonar imagery (horizontal resolution of 6m, swath width 6km) and magnetic measurements using the deep-towed vehicle TOBI. TOBI is a superb tool for determining lithological and tectonic boundaries20,30-34,37. The sidescan data also enable a high degree of sampling control34 and provide vital geological context at much higher resolution than swath bathymetry and over a much wider range than manned submersibles. The near-source magnetometer provides much higher spatial resolution than can be obtained from sea-surface surveys, so can be used to investigate detailed variations in seafloor spreading, magmatic accretion histories and rock magnetisations26.
We have assumed a cruise with 27 days on station (as 10). We plan 15 days for two TOBI deployments either side of FTFZ (Fig. 2), to give 100% sidescan coverage with 6km spaced tracks The northern deployment abuts and extends that of 10. These are designed to provide complete coverage of all ODP holes, extensive coverage over areas of predominantly ultramafic outcrop, cover the transition from lineated to rugged terrain south of FTFZ, and provide continuous coverage of three detachments and the intervening and conjugate seafloor of two. Tow lines will be slightly oblique to spreading direction to give good magnetic profiles and to image ridge-parallel structures20,30. We allow 12 days for coring, which should allow acquisition of about 70 cores. Sample locations will be selected using TOBI data to target areas as indicated under ‘Specific Objectives’.
Geochemistry
Variations in the extent of mantle melting will be used to test the hypotheses of mantle focusing and that ‘amagmatic’ spreading is linked to a decrease in melt supply. Mantle melt fraction will be determined from the composition of relict mineralogy (e.g., orthopyroxene (opx), clinopyroxene (cpx) lamellae and unaltered chrome spinel) preserved in the highly serpentinised peridotites, and their associated basalts. Cr# in mantle spinels and Mg# in relict mantle olivine and pyroxenes varies with extent of melting. Incompatible element abundances and their ratios in mantle pyroxene (esp. Al, Ca, Ti, Sr, Nb, Zr and Y) vary with melt fraction. In basalts, systematic changes in the ratio of immobile, very incompatible elements to moderately incompatible elements (e.g. Nb/Zr and Ti/Y), and fractionation-corrected parameters such as Na[8] and Ti[8], also indicate variations in source melting conditions, with a systematic increase in Na[8], Ti[8] and Nb/Zr ratio indicating lower temperatures, and lower Dy/Yb ratios indicating a decrease in initial melting depth. Fourier Transform Infra-Red (FTIR) analyses of fresh glass for primary magmatic water content will show changes in mantle hydration state and allow any link to changes in melt supply to be modelled using the pMELTS code. We will use measures of decreasing melt fraction and dryer mantle to determine the threshold at which diminishing melt supply leads to ‘amagmatic’ spreading.
Isotopic studies will be used to investigate the nature and origin of mantle heterogeneity to complement interpretation of mantle melting history and hence flow geometry. At the Atlantis II transform fault isotopically distinct components within the MORB source have different melting properties, e.g. an ancient, inherited pyroxenite signature40. The FTFZ region has similar enriched signatures38,39. We will compare the variation in mantle melting indices (see above) in both peridotites and basalts with their isotopic ratios (e.g. 143/Nd/144Nd, 87Sr/86Sr and double-spike Pb) to test the hypothesis that the enriched mantle component resides as small-scale heterogeneities in disequilibrium with the bulk of the mantle. For example, correlation between 143/144Nd and Mg#, Ti/Y and La/Yb in mantle pyroxene would indicate removal of an enriched, lower melting temperature component, allowing isotopic studies to address melting processes. A systematic variation in isotopic ratio throughout a ridge segment will indicate the variation in degree of melting: homogeneous for focused upwelling vs. systematic variation for the broad upwelling model. Alternatively, if the transition to amagmatic spreading is a function of mantle source composition (e.g. a depletion in fusible enriched components that promote low temperature melting), we predict that the transition to low melt fraction will be accompanied by a decrease in isotopic ratios such as 87Sr/86Sr.
We anticipate a moderate to high percentage of alteration of ultramafic
rocks, as found on leg 209, so we will be largely limited to trace element
analyses on relict mineral grains by excimer laser
ablation MC-ICP-MS and electron microprobe analyses (at Southampton), mineral
(pyroxene) separates (at Cardiff) and ion probe at Edinburgh. Standard major-,
trace- and rare-earth-element analyses will be performed on fresh basaltic
glasses and gabbroic rocks by ICP-OES at Cardiff
and solution and laser ablation ICP- MS at Southampton .
FTIR analyses of magmatic water content will be made by transmission laser
spectroscopy on 120µm wafers of fresh glass at Southampton .
Structural studies
From textural studies of the peridotite cores we will assess the conditions of high-temperature deformation and likely deformation mechanisms. In our experience (from ophiolite studies and from our BRIDGE drill cores from 15°45’N) this is normally broadly possible even in highly serpentinised samples, except for those samples that have undergone later low-temperature shearing. Preserved microstructures and grain sizes and shapes permit the distinction between very high temperature (asthenospheric) from moderately high temperature (lithospheric) deformation conditionse.g.12,14.
Mantle flow directions may be determined in orientated thin sections from chrome-spinel elongation directions and olivine a-axis alignments, which form a high strain lineation that is approximately parallel to the ductile flow direction. Although these fabrics are often partially obscured by serpentinisation, crystal lattice orientations and deformation modes can be recovered from the geographically orientated thin sections using the electron backscatter diffraction technique (EBSD41). Crystallographic orientations of the petrographically fresh minerals may need to be extracted manually rather than using automated routines, but this should still be possible in most cases.
From our oriented cores we will reconstruct original foliation and lineation directions and their mode of formation to infer the history of mantle upwelling geometry. For example, we predict a correlation between the degree of focusing of upwelling4-6 and the strength of along-axis sub-horizontal flow fabrics. Similar microfabric studies will be made on discrete shear zones to determine pressure and temperature histories of mantle exhumation, testing the inference from Leg 209 drilling and our own observations from 15°45’N that much of the mantle uplift is a result of brittle antithetic faulting localised beneath the MV10,12,17.
From the BRIDGE drill cores we can determine in geographical coordinates the true orientation of the primary remnant magnetisation. We will measure magnetic susceptibility anisotropies and compare them with mantle flow directions. Cores will be sub-sampled, secondary magnetic components (including any acquired during drilling) will be removed using alternating frequency demagnetization, and the remanent magnetisation will be determined using standard techniques. This will be carried out in the palaeomagnetic laboratory of Dr. E. Hailwood (Core Magnetics, Sedburgh).
Deviations from the predicted directions will be interpreted in terms of tectonic rotations and the axes and magnitudes of rotation constrained. Considering these data together with geographically orientated kinematic information from brittle fault material may allow us to determine which structures are accommodating the rotations. The fault kinematic data will be put together with structural information gleaned from the TOBI survey to integrate fully the different scales of investigation and provide very powerful constraint on the tectonics of ‘amagmatic’ spreading.
Crustal architecture
Many authors30,32-34,37 have confidently used TOBI sidescan sonar images to distinguish extrusive volcanics from tectonised and other terrains. Post-cruise, TOBI sidescan will be processed and digitally mosaicked and analysed for lithology and tectonics30,34,37. We will calibrate the acoustic imagery using existing samples and especially the well-navigated ODP and seabed rock drill cores, and use image analysis to generate acoustic classification maps37. Hence we will map the distribution of mantle and magmatic crustal units, determining their boundaries and structural context.
Rock magnetisation, inferred by inverting the TOBI magnetic field measurements20,26, will aid lithological identification. Although most crustal magnetisation has been thought to reside in the extrusive layer, exhumed serpentinised peridotite often has significant remanent magnetisation and well-defined magnetic anomalies42-44. Even so, the magnetic properties of serpentinised peridotite and gabbro are poorly documented. Inversion of TOBI magnetic data will be used to estimate their magnetic signatures, helping us to map lithological boundaries. We will use the effective magnetisation (the product of magnetic intensity and source thickness) inferred from TOBI data, to estimate the thickness of the extrusive layer, assuming constant remanent magnetisation20,26, to test models of extrusive thickness variations, asymmetric accretion, and magma supply variations
Resources and justification
Support costs for TOBI and the rock drill, based on quotations from G. West (UKORS) and C. Brett (BGS), include transport to and from ports, technician support, maintenance and consumables.
The contributions of the two tied students are given in the
attached RS1As. We include essential travel of the PI, 2 CIs,
2 students and 4 further watch-keepers to and from port. The vicissitudes of
ship schedules preclude the use of APEX fares. Three shifts of 3 watch keepers
are required for efficient conduct of the cruise. Watch keepers will include
partners Cannat and Casey and marine research students from our institutions.
Other travel costs include 1 journey per year between institutions for all
staff to facilitate collaboration over data analysis, and travel to the
required pre-cruise planning meeting at
Technician
costs and laboratory consumables are for the essential geochemical analyses, costed at standard ‘internal’ rates. We estimate 100
samples for major and minor element analyses by ICP-OES and 60 for trace elements
by ICP-MS at
A
high-specification dedicated PC and plotter, licences
for Matlab, Arc-Info and FORTRAN, and three months
support for our computing officer are required at
We request costs for the students and watchkeepers to undertake a 1-day Basic Sea Survival course, following EU guidelines. We also ask for a contribution to cruise-related e-mail costs, as per the new UKORS guidelines.
PROJECT MANAGEMENT AND OTHER ISSUES
The PI will take overall responsibility for the project, but will liase closely with the CIs. We
request a grant split between
As
is our normal practice, we will submit our data to appropriate national and
international databanks, and will make it available to bona fide scientists on
request. Core material will be stored in the repository at
We have been proactive in taking our results to the wider public, include promotion of marine science through school-based cruise web sites (e.g.45), presentation at the British Association, Royal Society soirées, radio and TV, articles in the national and local press, University and other public lectures, live TV and web-casting. We will pursue similar opportunities for this work.
This
work will lead to a significant improvement in our understanding of
lithospheric processes at
REFERENCES
1 Whitehead, JA, et al., Nature, 312, 146-8, 1984
2 Parmentier, EM & Phipps Morgan, J, Nature, 348, 325-8, 1990
3 Tucholke, BE & Lin, J, J. Geophys. Res., 99, 11,937-11,58, 1994
4 Escartin, J & Cannat, M, Earth Planet. Sci. Lett., 171, 411-424, 1999
5 Cannat, M, J. Geophys. Res., 101, 2847-2857, 1996
6 Escartin, J, et al., Earth Planet. Sci. Lett., 151, 181-189, 1997
7 Cann, J, et al. Nature, 385, 329-332, 1997
8 Tucholke, B, et al., J. Geophys. Res., B, 103, 9857- 9866, 1998
9 Blackman, DK, et al., J. Geophys. Res., B, 103, 21,315-21,333, 1998
10 MacLeod, CJ, et al., Geology, 30, 879-82, 2002
11 Casey, JF, et al.,
Trans. Amer. Geophys.
12 Escartin, J, et al., Geochem. Geophy. Geosy., 4, art. no.-1067, 2003
13 Cannat, M, et al., Geology, 23, 49-52, 1995
14 Cannat, M & Casey, JF, in Mantle and Lower Crust Exposed in Oceanic Ridges and in Ophiolites, Kluwer, 5-34, 1995
15 Dick, HJB & Mével, C, The Oceanic Lithosphere & Scientific Drilling into the 21st Century, JOI/US-SSP 89pp, 1996
16 Kelemen, P, et al., Proposal to ODP, 20 pp., 6 figs., 1998
17 Shipboard Scientific Party, http://www-odp.tamu.edu/ publications/prelim/ 209_prel/ro9toc.html, 2003
18 Fujiwara, T, et al., Geochem., Geophys., Geosys., 4, 1024, 2002
19 Kelemen, P & Dick, H, Personal communication to RCS, 2003
20 Searle, RC, et al., Earth Planet. Sci. Lett., 154, 167-83, 1998
21 Kelso, P et al., Proc. ODP, Sci Res, 147, 405-16, 1996
22 Searle, RC, et al., Geochem. Geophys. Geosyst., 4, art. 9105, 2003
23 Parson, L, et al., Earth Planet. Sci. Lett., 178, 231-51, 2000
24 Allerton, S, et al.,
Trans. Amer. Geophys.
25 MacLeod, CJ et al., Geol. Soc. Lond. Spec. Publication 65, 235-246, 1992
26 Allerton, S, et al., Geology, 28, 179-82, 2000
27 Searle, RC & Bralee, AV, EOS, Suppl. to Vol. 82, No. 47, F1097, 2001
28 Cannat, M, et al., Geochem. Geophys. Geosyst., 4, art. 9104, 2003
29 Ballard, RD & Van Andel, TH, Bull. Geol. Soc. Am., 88, 507-30, 1977
30 Escartin, J, et al., J. Geophys. Res., 104, 10,421-10,37, 1999
31 Head, JW, et al., J. Geophys. Res., 101, 28,265-28,80, 1996
32 Parson, LM, et al., Earth Planet. Sci. Lett., 117, 73-87, 1993
33 Smith, DK, et al., J. Volc. Geotherm. Res., 67, 233-62, 1995
34 Lawson, K, et al., Geol. Soc. Lond. Spec. Publication 118, 61-102, 1996
35 Allerton,
36 MacLeod, CJ, et al., InterRidge News, 7, 39-43, 1998
37 Blondel, P & Murton, BJ Handbook of Seafloor Sonar Imagery Wiley 314pp
38 Dosso, L, et al., Earth Planet. Sci. Lett., 106, 29-43, 1991
39 Bonatti, E, et al., J. Geophys. Res., 97, 4461-76, 1992
40 Salters, VJM & Dick, HJB, Nature, 418, 68-72, 2002
41 Prior, DJ et al. Am. Mineral., 84,1741-1759, 1999
42 Pariso, JE & Johnson, HP, J. Geophys. Res., 98, 16,013-16,32, 1993
43 Nazarova, K, Mar. Geophys. Res., 16, 455-62, 1993
44 Sempéré, J-C, Earth Planet. Sci. Lett., 107, 389-405, 1991
45 http://www.soc.soton.ac.uk/CHD/classroom@sea/ carlsberg/index.html