Georges Ceuleneer

Characteristics and evolution of the segmentation of the Mid-Atlantic Ridge between 20°N and 24°N during the last 10 million years

Abstract High-resolution bathymetry and geophysical data collected along the slow-spreading axis and flanks of the Mid-Atlantic Ridge between 20°N and 24°N reveal the characteristics and history of different wavelengths of segmentation during the last 10 m.y. The bathymetric data exhibit a morphotectonic pattern dominated by ridge-normal and oblique bathymetric lows that partition the ridge flanks into rhomb-shaped areas of relatively high elevation. At least four different types of oblique bathymetric lows have been identified which represent the off-axis traces of axial discontinuities and suggest a complex and ongoing evolution of ridge-axis segmentation. One group of oblique structures is represented by two deep ridge-normal depressions with typical fracture zone characteristics that are connected to the present active transform by oblique depressions near the ridge axis. These oblique traces correspond to the southward shift of axial discontinuities associated with the propagation of the ridge axis, while maintaining a constant offset of the latter. Two other types of oblique structures correspond to elongate bathymetric lows and oblique alignments of ridge-parallel bathymetric lows symmetric about the ridge axis. Both types of oblique structures frequently change their orientation (from normal to subparallel to the ridge axis) and appear to merge and diverge off-axis. These oblique depressions are characterized by positive filtered mantle Bouguer anomalies, high magnetizations, complex magnetic anomaly patterns, and possible exposure of mantle lithologies. The ridge segments defined by these oblique depressions lengthen or shorten along the ridge axis, with propagation rates varying from 0 to 25 km m.y. −1 . The last and smallest discontinuities observed in this area correspond to small ridge-axis offsets and off-axis traces identified by alignments of the terminations of abutting abyssal hills. The ridge-flank morphotectonic patterns produced by the evolution of these elementary segments of accretion may represent temporally variable upwelling volumes of melt. The centres of the rhomb-shaped areas correspond to maximum crust production and thin lithosphere, and the discontinuities correspond to a thick lithosphere with very thin crust and possible outcrops of peridotites. We propose a model which accounts for the punctuated injection of magma and the evolution of elementary segments of accretion over periods of several million years.

Trace element and isotopic characterization of mafic cumulates in a fossil mantle diapir (Oman ophiolite)

This paper is devoted to an exploratory geochemical study (trace elements, SrNd isotopes) of a suite of cumulate features cropping out in the mantle harzburgites of Oman. The cumulates are concentrically distributed around a fossil mantle diapir. From the centre to the periphery of the diapir, their mineralogy becomes more and more evolved, from troctolite to olivine gabbro and gabbronorite, and their textural and field characteristics point to injection in a cooler environment. Nd isotopic data are consistent with a mantle origin for all these cumulates. The mantle source is heterogeneous on a small scale (6.09 <ϵNd < 1015) but this heterogeneity and the average ϵrmNd (≈8) are comparable to those of Indian Ocean mid-ocean ridge basalts (MORBs). An origin of this cumulate suite, largely by fractional crystallization from liquids with MORB characteristics, is supported by major and compatible trace element variations. However compatible elements show that it is unlikely that all these cumulates are derived by different degrees of fractional crystallization from the same liquid. This conclusion is corroborated by incompatible trace element data. The calculated liquids in equilibrium with these cumulates have MORB-like REE patterns. However, variations in REE patterns from one lithological group to the next, and within each lithological group, imply a pronounced heterogeneity in the REE content of the equilibrium liquids that clearly does not result from a simple fractional crystallization process. Liquids in equilibrium with the troctolites have a higher range in [La]n/[Yb]n than liquids in equilibrium with the olivine gabbros. The most evolved cumulates (gabbronorites) have REE patterns that might imply a more complex origin for their parental melt involving mixing of MORB-like liquids with melts from a LREE-depleted source.

Chromite crystallization in a multicellular magma flow : Evidence, from a chromitite dike in the Oman ophiolite

The investigated chromitite dike is located at the top of an upwelling mantle structure of the Oman ophiolite (Maqsad diapir), in undeformed dunites displaying evidence for magma impregnation and circulation, just below the paleo-ridge axis. The chromitite dike is undeformed, its shape is that of an upward widening tube. It exhibits an internal layering which is roughly perpendicular to the cavity axis and comprises a vertical succession of four main layers showing a graded-bedding. Chromitite magmatic structures are beautifully preserved and result from a progressive crystallization from small euhedral crystals to wide octahedron-shaped nodules; dissolution textures provide evidence for late magmatic desequilibrium; sedimentation structures include flattening of the largest nodules. The silicate matrix comprises poikilitic forsterite and a locally abundant association of primary pargasite and plagioclase and alteration minerals (vesuvianite-chlorite-dolomite); pargasite inclusions are very abundant in the chromite. Chromite composition changes from one layer to the other and from core to rim in the chromite nodules (chromium decreases and titanium increases); Ti contents are generally high (0.4 to 0.8 wt.% TiO2) with respect to podiform chromites. Platinum-group elements are not abundant but they show a strong fractionation at the scale of the orebody and of the main graded-bedded layers (Pd/Ir ratio varies from 0.5 to 11.5). REE patterns of chromitite parallel to those of gabbros and furthermore display a sea water related hydrothermal alteration (Ce negative anomaly). The chromitite dike of Maqsad provides evidence for the crystallization of chromitite bodies in subvertical magma conduits below oceanic ridges; it corroborates the model of Cassard et al. (1981) and Lago et al. (1982) concerning the formation of chromitite pods in ophiolites which were later deformed and transposed into the horizontal plane due to the plastic flow prevailing away from the paleo-axial zone. Layering and chromite compositional variations are ascribed to a multicellular convective system segregating various stocks of chromite particles either in the upwelling flow of fresh magma or in the convective cells of fractionated residual magma in the confined part of the cavity. The estimated life-time for the magma influx is very short (<2 months). The parent-magma was probably of MORB-type and already fractionated (Ti-rich and PGE-poor), which is consistent with the strong evidence of magma-peridotite interactions in the core of the Maqsad diapir. Hydrous fluids were present during chromite crystallization (pargasite inclusions) suggesting that fluid-rich melts occur in the upper mantle.

Silicate and base-metal sulfide inclusions in chromites from the Maqsad area (Oman ophiolite, Gulf of Oman): A model for entrapment

Abstract Al-rich chromites from the Maqsad area (Oman ophiolite) have been studied to propose a genetic model for silicate and base-metal sulfide (BMS) inclusion entrapment. Samples were collected in both concordant and discordant nodular pods and in a peculiar stratiform ore; chromite ore contains olivine+clinopyroxene+plagioclase as interstitial silicates. Regardless of the ore type, the silicate inclusions are, in decreasing order of abundance, composed of Ti- and Na-rich phlogopite, Ti-pargasite, enstatite, Mg-diopside, olivine, albite and secondary hydrothermal products. Chromite-hosted BMS are dominated by pentlandite, heazlewoodite, millerite and are nearly devoid of Fe- and Cu-sulfides. The samples which come from concordant ores are very poor in mineral inclusions, a consequence of the strain-induced grain boundary migration and recrystallization which they suffered during plastic deformation. In contrast, the nodular and stratiform samples which escaped post-magmatic deformation and densification are inclusion rich. The inclusions display negative crystal shape following the chromite symmetry. If present in nodular samples, they may be arranged as coronas which mimic the shape of the host chromite grains. All these textural features are considered to be evidence for mineral inclusion entrapment during magmatic precipitation of chromite. By reference to a previous dynamic model of chromite ore genesis, we consider that the nodules of chromite crystallized from basaltic magma were flowing inside a narrow cavity in which a non-turbulent convective system was established. We deduce from the presence of several inclusion coronas or -patches in the largest nodules and their aabsence in the smallest,that the inclusions were trapped preferentially at the point in which the primitive magma flow meets the convective current. The peculiar mineralogy of mineral inclusions dominated by hydroxyl-sodic phases indicates a complex interaction at the magmatic stage between the chromite-precipitating magma and an unrelated aqueous fluid phase. This fluid, which was sodium- and sulfur-bearing, travelled through the top of the mantle sequence and the basis of cumulate rocks of the ophiolite. It could have been injected at certain stages of its evolution in the same conduits as the chromite-precipitating magma, locally giving rise to volatile-enriched differentiated melts. The latter have been sealed together with the liquidus phases of magma (olivine, clinopyroxene, orthopyroxene and plagioclase) as inclusions during chromite growth and/or dissolution. Olivine and pyroxenes have partly reacted with the trapped volatile-rich melt, producing phlogopite and pargasite several hundred degrees below trapping temperatures. BMS have formed through sulfidization reactions with enclosed silicates and/or host chromite.

The remelting of hydrothermally altered peridotite at mid-ocean ridgesby intruding mantle diapirs

Most gabbroic cumulates found at ocean spreading centres are thought to have been generated by the fractional crystallization of melts with the composition of mid-ocean ridge basalt (MORB). There are exceptions, however, including some cumulates which appear to have come from melts that contain more silica than MORB and are much more depleted in the incompatible elements (those elements that do not readily substitute into the main mineral phases). These unusual rocks bear witness to relatively deep petrological processes that are not accessible through the study of melts erupted on the sea floor, and their origin is still debated. Fortunately, the same lithologies can be studied in detail in ophiolites (sections of oceanic crust accreted to a continent). In a fossil mantle diapir of the Oman ophiolite, we have observed the same dichotomy between a suite of ‘normal’, MORB-type, cumulates (‘N-cumulates’) and a suite of cumulates issued from silica-enriched but incompatible-element-depleted melts (‘D-cumulates’). While the N-cumulates crystallized inside the diapir, the D-cumulates occur essentially as intrusions surrounding the diapir. The combination of silica enrichment, extreme depletion in incompatible elements, and seawater isotopic signature indicates that the D-cumulates were formed by the remelting at low pressure of hydrated residual peridotites left after MORB extraction at the ridge axis. The distribution of the D-cumulates relative to the N-cumulates suggests that such depleted melts are produced episodically at ridge axes when the lithospheric mantle is reheated by a new diapiric pulse.

Tectonic setting for the genesis of oceanic plagiogranites: evidence from a paleo-spreading structure in the Oman ophiolite

Abstract In the Oman ophiolite, detailed structural and lithological mapping of the 60 × 45 km Maqsad area has revealed several plagiogranite intrusions, ranging from metre scale dykes to hectometre scale bodies. These plagiogranites are spatially related to kilometre scale mafic plutons, displaying generally pegmatitic textures and locally deformed in amphibolite facies conditions along shear zones up to a few hundred metres thick. Isotope data demonstrate the mantle origin of these pegmatites and plagiogranites: ϵ Nd ranges from 6.3 to 9.9 and has an average value (26 samples) of 8.1, within the field of present-day MORBs, and similar to the average ϵ Nd (7.8) of the Oman ophiolite primary igneous sequence. These intrusions are not restricted to a given structural level: their paleovertical extent exceeds 5–6 km, from 2 km below the Moho up to the base of the sheeted dyke complex. Their modal composition evolves up section, from predominantly pyroxenites and gabbro norites in the mantle harzburgites to gabbro norites, diorites and quartz diorites at crustal level. Petrography, whole rock and mineral chemistry show that this vertical succession of lithologies can be simply explained by low pressure fractional crystallization of a hydrated basaltic melt. Plagiogranites, although more abundant at upper crustal level, are not restricted to this horizon. Plagioclase composition in plagiogranites presents a remarkable evolution with paleodepth, from highly calcic (70–95% an ) in the mantle section, intermediate and highly scattered in the lower crust (10–85% an ), to quite sodic at upper crustal level (5–20% an ). Highly calcic plagioclases in quartz-bearing rocks is a puzzling character, inconsistent with fractional crystallization. Even if high degrees of fractional crystallization of a hydrated basaltic melt remain, the simplest way to account for the chemistry of the upper crustal plagiogranites, processes such as fluid-induced remelting or assimilation of country rocks must be invoked to account for the chemistry and field characteristics of the deeper plagiogranites. Mantle and crustal structures in the Maqsad area indicate that the pegmatites and plagiogranites emplaced in an ocean ridge setting, when a mantle diapir, soaked with basaltic melt, intruded a cool, hydrothermally altered, lithosphere. They were eventually transposed in an off-axis setting as the ascension of the diapir progressed up to Moho level. It is concluded that the formation of large volumes of buoyant leucocratic rocks is possible at ocean ridges, provided periods of amagmatic spreading are long enough to induce the growth and alteration of an axial lithospheric lid. This is likely more frequent at slow spreading centres where mantle upwelling is discontinuous in space and time.

Primitive layered gabbros from fast-spreading lower oceanic crust

Three-quarters of the oceanic crust formed at fast-spreading ridges is composed of plutonic rocks whose mineral assemblages, textures and compositions record the history of melt transport and crystallization between the mantle and the sea floor. Despite the importance of these rocks, sampling them in situ is extremely challenging owing to the overlying dykes and lavas. This means that models for understanding the formation of the lower crust are based largely on geophysical studies and ancient analogues (ophiolites) that did not form at typical mid-ocean ridges. Here we describe cored intervals of primitive, modally layered gabbroic rocks from the lower plutonic crust formed at a fast-spreading ridge, sampled by the Integrated Ocean Drilling Program at the Hess Deep rift. Centimetre-scale, modally layered rocks, some of which have a strong layering-parallel foliation, confirm a long-held belief that such rocks are a key constituent of the lower oceanic crust formed at fast-spreading ridges. Geochemical analysis of these primitive lower plutonic rocks—in combination with previous geochemical data for shallow-level plutonic rocks, sheeted dykes and lavas—provides the most completely constrained estimate of the bulk composition of fast-spreading oceanic crust so far. Simple crystallization models using this bulk crustal composition as the parental melt accurately predict the bulk composition of both the lavas and the plutonic rocks. However, the recovered plutonic rocks show early crystallization of orthopyroxene, which is not predicted by current models of melt extraction from the mantle and mid-ocean-ridge basalt differentiation. The simplest explanation of this observation is that compositionally diverse melts are extracted from the mantle and partly crystallize before mixing to produce the more homogeneous magmas that erupt.

Compaction in a mantle mush with high melt concentrations and the generation of magma chambers

Abstract Most compaction models in the partially molten mantle have addressed the case of low intergranular melt concentrations (up to a few percent). Here we develop a mathematical and numerical formalism adapted to the two-dimensional modeling of compaction in mushes with higher melt concentrations (up to a few tens of percents). Experimental data on mantle-like mushes (olivine crystals+basaltic melts) suggest that the mush viscosity depends in a complex way on the melt concentration: three rheological thresholds occur at melt concentrations of about 5%, 20%, and 40%, respectively. The first threshold corresponds to the establishment of full interconnectivity of the intergranular melt, the second to the formation of a very dense suspension of crystals and the last to the development of crystal clusters in the suspension. The present models take into account the stiff and drastic viscosity drops associated with these rheological thresholds. Intergranular melt migration associated with an initial melt pulse generates a horizon of high melt/crystal ratio. If the melt concentration in the initial pulse presents a local excess, the horizon becomes slightly tilted. As a consequence, melt percolates upslope inside the tilted horizon, pools at its summit and generates a ‘pocket-like’ zone. Due to the higher melt concentration, the upward Darcy velocity in the pocket markedly exceeds that in the horizon. The result is that the pocket-like impregnation is rapidly disconnected from the horizon and a new pocket develops at the summit of the partially fragmented horizon. Eventually, the intergranular melt contained in the horizon is completely redistributed into pockets. Increasing the background melt concentration in the mush from 5%, 20%, and 40% leads to an increase of the maximum melt concentration of 10%, 40%, and 100%, in the transient horizons and of 25%, 60%, and 100% in the pockets. These models suggest that magma chambers with a kilometer extent naturally result from the compaction of a mantle mush with an initial melt concentration exceeding 5%.

Viscosity and thickness of the sub-lithospheric low-viscosity zone: constraints from geoid and depth over oceanic swells

Abstract The medium-wavelength geoid to depth anomalies ratio (GDR) at oceanic hotspot swells has been found to increase from ∼ −0.5 m/km to ∼ 5 m/km according to the age of the lithosphere they occur on. In order to interpret this trend, the geoid and topography anomalies associated with mantle convective plumes crossing a sublithospheric low viscosity zone (LVZ) have been derived from numerical models and a systematic investigation of the GDR dependence on the viscosity and depth extent of the LVZ, on the thickness and thermal structure of the lithosphere and on the Rayleigh number has been conducted. It is shown that, for viscosity drops across the base of the LVZ, greater than one order of magnitude, the GDR is strongly dependent on the depth of shallow interfaces such as the lithosphere/ athenosphere boundary and on the LVZs thickness. Consequently, the empirical trend can be accounted for by the thickening of the lithosphere with age provided it occurs at the expense of a LVZ whose base is at a fixed depth (around 200 km). In such a frame, no significant variation with age of the LVZs viscosity is required by the GDR data. Best fit with the empirical trend is found for a LVZ about 50 times less viscous than the underlying mantle. The mantle flow starts to fluctuate when the local Rayleigh number of the low-viscosity layer exceeds the Rayleigh number of the underlying mantle. The fluctuations are initiated in the upper boundary layer, in the diverging part of the plume, at a distance of a few hundreds of kilometers from the main ascending current. For viscosity contrasts in the range of 40–60, deduced from the present study, the conditions for the development of these small-scale instabilities are realized only where the lithosphere has not yet grown significantly downwards (ages

Mapping of an ophiolite complex by high‐resolution visible‐infrared spectrometry

The Sumail massif of the Oman peridotite has been surveyed by the high spectral resolution imager HyMap. The field measurements have been taken simultaneously to the HyMap campaign with a GER 3700 spectrometer. To compare the two data sets, further calibration and atmospheric correction are made through empirical line corrections. A continuum removal by a modified Gaussian model on reflectance is then performed for each pixel to minimize atmospheric scattering and shadowing effects. Classification from spectral distance is established using field observations and comparisons at full spectral resolution to identify peridotite subunits (harzburgites and dunites) and to distinguish plagioclase-wehrlites from gabbros. Detections of hydrothermal transformations of olivine in serpentines and their alteration in carbonates along fracture networks, characterized by narrow spectral signatures, validate the classification. This demonstrates the utility of the high spectral resolution and the development of appropriate processing methods for geological identifications.