Plate tectonics describes oceanic transform faults as conservative strike-slip boundaries, where lithosphere is neither created nor destroyed. Seafloor accreted close to ridge-transform intersections (RTI) has therefore been expected to follow a similar subsidence trend with age as lithosphere that forms away from RTIs. Our recent combined analysis of high-resolution bathymetric data, satellite gravity, and three-dimensional numerical models from transform faults segmenting mid-ocean ridges across the entire spectrum of spreading rates challenges this concept.  One striking observation is that transform faults are systematically deeper than their adjacent fracture zones. Gravity data suggests that the underlying reason may be changes in crustal thickness, with transform valleys having thin and fracture zones ‘normal’ crustal thicknesses. Another observation is that outside corner crust often shows symmetric abyssal hills with intact flat top volcanoes, while the inside corner regions show intense and oblique tectonic deformation. Furthermore, so-called J-shaped ridges, volcanic ridges that bend towards the active transform, show that magmatic accretion occurs predominantly along the spreading axis, ‘feeling’ the rotating stress field only in the direct vicinity of the RTI. While these observations do show some dependence on spreading rate, they can be identified across a wide range of opening rates, suggesting that they are expressions of processes inherent to transform faulting.

In this contribution, we will review these observations before presenting numerical 3-D thermo-tectono-magmatic models designed to elucidate the underlying processes. These models use a dilation term to mimic magmatic accretion and resolve visco-elasto-plastic deformation. The simulations show that the tectonic deformation axis, the axis of plate separation, becomes oblique at depth resulting in extension and crustal thinning within the transform deformation zones. Complementing simulations that account for magmatic accretion and hydrothermal cooling show that a skew can develop between this oblique deformation axis and the axis of magmatic accretion, implying a possible disconnect between the main diking direction and the direction of tectonic deformation. Taken all evidence together, oceanic transform faulting appears to be much more complex than pure strike-slip motion. It shows a surprisingly complex pattern of tectonic faulting and hints at spill-over magmatism at the RTI.  Crustal accretion at ridge transform intersections may therefore be fundamentally different to accretions elsewhere along mid-ocean ridges.

How to cite: Ruepke, L., Grevemeyer, I., Guo, Z., Liu, S., Chen, M., Morgan, J., and Devey, C.: Oceanic transform faults revisited with models and data, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9093, https://doi.org/10.5194/egusphere-egu23-9093, 2023.

Observations suggest that the oceanic lithosphere is shaped by dike intrusions and faulting in proportions that depend on the spreading rate (Carbotte et al., 2016). Yet it remains unclear how the interplay between magmatism and faulting during seafloor spreading affects mid-ocean ridge (MOR) axial morphology, fault spacing, and the pattern of abyssal hills (Buck et al., 2005, Huybers et al., 2022). Here we present two-phase flow numerical models of oceanic lithosphere extension that reconcile the nonlinear brittle behaviour of the lithosphere with mantle melting and magma transport through the lithosphere. 

Fast-spreading ridges show symmetric normal faulting and axial highs, while slow-spreading ridges show an asymmetric fault pattern and axial valleys. Previous work has focused on explaining the MOR fault pattern by tectonic or magmatic-induced deformation. In the first scenario, faults result from tectonic stretching of the thin axial lithosphere during amagmatic periods (Forsyth 1992), while in the second scenario, dike-injection may create stresses that activate extensional faults (Carbotte et al., 2016). Current state-of-the-art models (i.e., Buck et al., 2005) use a single-phase formulation for the deformation of oceanic lithosphere in which a prescribed axial dike may accommodate both magmatic and tectonic extension. In these models, the fault pattern depends on M – the fraction of plate separation rate that is accommodated by magmatic dike opening. While M-models are able to explain a number of observations, M represents a simple parameterization of complex fracture dynamics of sills, dikes, and faults. In particular, M-value models neglect fault–dike interaction and other modes of melt transport and emplacement in the lithosphere (Keller et al., 2013). 

Here we build a 2-D oceanic lithosphere extension model that incorporates a new poro- viscoelastic–viscoplastic theory with a free surface (Li et al., in review) to robustly simulate plastic representations of dikes and faults in a two-phase magma/rock system. We hypothesise that magma supply controls the pattern of dike–fault interaction in oceanic extension settings. We present simplified model problems to compare results with those from M-value models. These enable us to address the significance of M in terms of fundamental magma and lithospheric processes. We then focus on development of fault patterns, magma pathways and crustal production at fast-/slow-spreading ridges.

References

Buck et al., 2005, Nature, doi:10.1038/nature03358.

Carbotte et al., 2016, Geol. Soc. London, doi:10.1144/SP420.

Forsyth, 1992, Geology, doi:10.1130/0091-7613(1992)020<0027:FEALAN>2.3.CO;2.

Huybers et al., 2022, PNAS, doi:10.1073/pnas.2204761119.

Keller et al., 2013, GJI, doi:10.1093/gji/ggt306.

Li, Y., Pusok, A., Davis, T., May, D., and Katz, R., (in review). Continuum approximation of dyking with a theory for poro-viscoelastic–viscoplastic deformation, GJI.

How to cite: Pusok, A. E., Li, Y., Katz, R. F., Davis, T., and May, D. A.: The role of magma supply in fragmentation of oceanic lithosphere, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-16161, https://doi.org/10.5194/egusphere-egu23-16161, 2023.

Mantle melting along mid-ocean ridges occurs in a segmented manner.  Melting and melt extraction are greatest within ridge segment interiors but near segment ends mantle upwelling decreases, cooling increases and melt extraction becomes inefficient.  Owing to the strong influence of water on mantle rheology, these effects have important consequences for the strength of oceanic lithosphere.  Residual mantle formed in ridge segment interiors is melt-depleted and dehydrated forming strong rheological bands.  Near segment ends, however, the formation of low-degree hydrous melts predominates, and these are inefficiently extracted from the mantle.  On solidification, these hydrous melts can re-fertilize surrounding mantle with water due to the high diffusivity of hydrogen in mantle material. This results in weak hydrous bands of mantle material near segment ends.  Thus, segmented mantle melting creates a corresponding segmented oceanic mantle rheological structure that favors the localization of shear deformation in the weak bands near segment ends.  Further strain localization within these weak zones may then facilitate additional weakening processes along discrete narrow transform fault zones. We Illustrate our model with geophysical observations from the Reykjanes Ridge and northern Mid-Atlantic Ridge south of Iceland.  The Reykjanes Ridge is a ~1000 km long linear axis without transform faults.  Rapid propagation of melting anomalies along its linear axis precludes a stable magmatic segmentation as shown by its linear mantle Bouguer anomaly.  Immediately south of the Reykjanes Ridge, the northernmost segments of the Mid-Atlantic Ridge have prominent mantle Bouguer anomaly lows indicating stable cells of segmented mantle melting. Transform and non-transform discontinuities immediately form at the ends of the mantle Bouguer anomaly lows.  This model can be extended to explain the occurrence (or absence) of transform faults over the full range of spreading rates from ultra-slow to ultra-fast ridges.

Reference: Martinez, F., and R. Hey (2022), Mantle melting, lithospheric strength and transform fault stability: Insights from the North Atlantic, Earth and Planetary Science Letters, 579, doi:10.1016/j.epsl.2021.117351.

How to cite: Martinez, F. and Hey, R.: Segmented Mantle Melting, Lithospheric Rheology and Transform Fault Formation, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9688, https://doi.org/10.5194/egusphere-egu23-9688, 2023.

Gravity signals over the mid-ocean ridge-transform system reflect the distribution of underlying crustal and upper mantle mass anomalies. The gravity measurement, especially ‘residual’ gravity anomalies, relies on the gravitational corrections of both seafloor relief and lithospheric thermal structure. Lithospheric thermal correction typically uses a 1D plate cooling approximation or a 3D passive flow model that assumes isoviscous mantle rheology. As this rheological approximation is oversimplified and physically complex, how sensitive gravity anomalies are to an increasingly complex/accurate approximation for mantle rheology is still unresolved. Here we systematically examine the residual gravity anomaly discrepancies caused by assumptions of different mantle rheologies on 16 natural ridge-transform systems ranging from ultraslow- to fast-spreading. Our calculations show that estimated residual gravity anomalies are significantly lower (e.g., ~21 mGal lower at mid-ocean ridges) in the isoviscous flow models than in the static plate cooling models, primarily due to the effects of lateral heat advection and conduction. When the assumed mantle rheology is changed from uniform viscosity to a non-Newtonian viscosity with brittle weakening in cooler (faulting) regions, the mantle upwelling intensifies and local near-surface temperature generally increases, resulting in an increase in the residual anomaly. This increase is distributed uniformly along the ultraslow-and slow-spreading ridge axes, but is concentrated along transform faults at intermediate- and fast-spreading ridges. The amount of the rheology-induced gravity difference is most closely linked to transform age offset instead of spreading rate or transform offset length alone. Our analysis reveals that oceanic transform faults exhibit higher gravity anomalies than adjacent fracture zones, which may reflect thinner crust in the transform deformation zone.

How to cite: Liu, S., Guo, Z., Rüpke, L., Morgan, J. P., Grevemeyer, I., and Ren, Y.: Gravity signature in the mid-ocean ridge-transform system: Insights from deep mantle rheology and shallow crustal structure, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-869, https://doi.org/10.5194/egusphere-egu23-869, 2023.

Recent studies on oceanic transform faults, one of the three fundamental types of plate boundaries, has suggested that they may not be purely conservative features and that the crust formed adjacent to them (on the "inside corners" of the ridge-transform intersection) may differ in structure and composition significantly from outside-corner crust. Here we present a geological map of the Oceanographer Transform (Atlantic Ocean, southwest of the Azores) created by combining an interpretation of multibeam bathymetry, rock sampling and seafloor visual observations. We find that outside- and inside-corner crust at the ridge transform intersection have distinctive morphologies and petrography: the outside corner shows rough seafloor, from which only pillow basalts are recovered, extending all the way to the fracture zone. The inside corners, in contrast, are characterized by both rough, basaltic seafloor and regions that are much smoother, from which serpentinized peridotite are often recovered. The width of the inside-corner region showing this variable morphology, bathymetry and petrography seems to vary over time from 10 to 25 km. In two places, oceanic core complex crust is recognized close to the transform in this inside-corner region. We emphasize that plate production at the inside corner appears to occur via a variety of magmatic and amagmatic processes.

How to cite: Unger Moreno, K. A., Devey, C. W., Rüpke, L., Beniest, A., Hansteen, T. H., and Grevemeyer, I.: Geological overview of the Oceanographer Transform Fault, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9241, https://doi.org/10.5194/egusphere-egu23-9241, 2023.

The significant discrepancy between the observed conductive heat flow and predictions by thermal models for oceanic lithosphere younger than 50 Ma is generally interpreted to result from hydrothermal circulation between basement outcrops. Numerical simulations of fluid flow between such outcrops performed in previous studies revealed that establishing horizontal pressure gradients to sustain a hydrothermal siphon requires high aquifer permeabilities and a contrast in the outcrops’ transmittance, which is the product of the outcrop permeability and the area of outcrop exposure. However, most previous studies focused on the model parameters needed to sustain a hydrothermal siphon, while the physical processes that create the horizontal pressure gradients in the first place remain poorly constrained.

In order to shed more light on the physics behind outcrop-to-outcrop flow, a simple synthetic 2D model of two outcrops connected by a permeable aquifer was set up. Fluid flow modelling was done by using hydrothermalFoam, a hydrothermal transport model, that is based on the open-source C++ computational fluid dynamics toolbox OpenFOAM. Our initial simulations focus on variations of the permeability of the outcrops and the aquifer. The results reveal two key points that are essential to generate a flow: First, the outcrops permeability has a fundamental effect on its average pressure. High permeabilities lead to a rather "cold" hydrostatic pressure regime with lower temperatures and hence higher average pressures. Lower outcrop permeabilities are accompanied with a rather "warm" hydrostatic pressure regime characterized by higher temperatures and lower average pressures. Secondly, fluid convection in the aquifer is necessary to establish a siphon flow. Therefore, the aquifer permeability must be sufficiently high to overcome Darcy resistance and yet low enough to prevent the flow from being solely diffusive.

How to cite: Kremin, I., Guo, Z., and Rüpke, L.: The effect of permeability on the pressure regime in 2D outcrop-to-outcrop submarine hydrothermal flow models, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-2737, https://doi.org/10.5194/egusphere-egu23-2737, 2023.

Lucky Strike volcano is the central edifice of the Lucky Strike segment, Mid-Atlantic Ridge. Its summit overlies an axial magma chamber (AMC), 3-3.8 km beneath the seafloor, and hosts one of the largest known deep-sea hydrothermal fields. Local seismicity beneath the hydrothermal field has been monitored since 2007 as a part of the EMSO (European Multidisciplinary Seafloor and water column Observatory)-Azores observatory by 5 OBSs with yearly redeployments. In a 12-year (2007-2019) earthquake catalog (noncontinuous), we observe continuous low magnitude seismicity (M_L ~ -1 to 0), focused mainly 0.5-2 km above the AMC, suggesting that thermal contraction of rocks, possibly combined to deformation induced by volume changes during hydrothermal alteration, at the base of a single limb along-axis hydrothermal cell is the primary source of this seismicity. We thus interpret the seismicity clusters, with horizontal extent 1200 to 1800 m², as zones of enhanced heat extraction, in the lower part of the hydrothermal downflow zone.

We present the evolution of this hydrothermally-induced seismicity over the 12 years of the catalog. We observe three lateral 400-800 m shifts of the main seismicity clusters. The first and second shifts are small and could be explained by a fortuitous combination of network-based biases, picking error and/or change in the shallow seismic velocity structure of the volcano. The third shift, occurring during a catalog gap between June 2013 and April 2015, is ~800 m eastward and corresponds to a change in the seismicity distribution from a patch above the AMC to a vertical pipe-like pattern, indicating a real change in the hydrothermal circulation. We propose that this shift is driven by recent magmatic injections above the AMC, and/or to the opening of new tectonic cracks, enhancing local permeability and allowing for more efficient cooling above the shallower region of the AMC roof.

We also observe three Higher Seismic Activity (HSA seismic rate > 18 events/week) periods: April-June 2009, August-September 2015, and April-May 2016. The 2009 HSA period was the most intense: it lasted ~13 weeks, starting with a relatively higher magnitude event (M_L = 1.7), and culminating in June after another higher magnitude (M_L = 1.8) event. Most of the events clustered 0 to 1 km above the AMC reflector, with a few deeper events (down to only 800 m below the AMC reflector) during the culmination period. Although we do not have focal mechanisms to test this hypothesis, we propose that this HSA period resulted from tectonic events opening enhanced local permeability channels for downgoing hydrothermal fluids, and leading to higher heat extraction by the hydrothermal system.

How to cite: Bohidar, S., Crawford, W., and Cannat, M.: Seismic constraints on the evolution of hydrothermal circulation beneath Lucky Strike volcano, Mid-Atlantic Ridge, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14327, https://doi.org/10.5194/egusphere-egu23-14327, 2023.

In recent years hydrothermal vent systems were found in unexpectedly high abundance along ultraslow spreading ridges, despite their overall decreased magma supply. Thin oceanic crust and resulting shallow heat sources can drive hydrothermal fluid circulation and detachment faults can act as fluid pathways, resulting in e.g., serpentinization of the oceanic crust. So far, no long-term recording of seismicity around hydrothermal vent systems along ultraslow spreading ridges have been reported. Here, we present results from a ~1-year local Ocean Bottom Seismometer deployment between 2019 - 2020 at Loki’s Castle hydrothermal vent field (LCVF) along the Arctic Mid Ocean Ridge. LCVF is located at a water depth of ~2500m on top of the axial volcanic ridge (AVR) at the Mohn-Knipovich Ridge bend where spreading is highly asymmetric from west to east.

For the processing we use a combination of an automatic event detection algorithm (Lassie), a deep-learning phase picking model (PhaseNet) and partial manual re-evaluation of phase picks. Additionally, selected clusters of events are cross-correlated and relocated using hypoDD. The resulting earthquake catalogue consists of a total of 12368 events with 6719 manually re-evaluated and 5649 automatically picked events.

From the results we see that most of the plate divergence at the Mohn-Knipovich Ridge bend is accommodated by a young detachment fault west of the AVR. Most of the seismicity occurs between depths of ~2-8km in a bended band that steepens up to 70° with depth and follows the local topography. However, the described detachment fault differs from reported mature detachment faults at the Mid-Atlantic Ridge or Southwest Indian Ridge. Within the footwall we observe episodical, clustered seismicity with extensional faulting mechanisms, indicating that the detachment could be cross-cut by normal faults. Along strike, the seismicity of the fault plane appears highly heterogeneous, with the central part showing only sparse seismicity at depths below 3km while other segments show episodical shallow seismicity. Towards LCVF seismicity below the AVR increases and the maximum depth of earthquakes shallows by about ~2km. This could indicate the presence of a shallow heat source below LCVF as a driving factor for the hydrothermal circulation.

How to cite: Pilot, M., Lien, M. E., Schlindwein, V., Ottemoeller, L., and Barreyre, T.: Highly Asymmetric Seismicity in a System of Tectonic Extension and Hydrothermal Venting at the Mohn-Knipovich Ridge Bend, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-2230, https://doi.org/10.5194/egusphere-egu23-2230, 2023.

On-going work on the Samail Ophiolite Volcanogenic Massive Sulfide (VMS) deposits and Oman Drilling Project (OmanDP) drill cores provide insights on sulfur and metal cycling during hydrothermal alteration and critical differences between ophiolitic and modern oceanic crust.

The volcanic section is pervasively overprinted by low-T oceanic metasomatism leading to variably depleted sulfur in sulfide (TS_sulf) concentrations reflecting leaching and oxidation of magmatic sulfides. Secondary sulfides incorporated mostly basaltic sulfur with minor sulfur addition via open-system bacterial sulfate reduction (BSR).

The sheeted dyke-gabbro transition (OmanDP GT3 drillhole) records a change from BSR open-system processes (d³⁴S>-12.8‰) towards addition of heavy hydrothermal sulfur via thermochemical sulfate reduction ~4km above the Moho Transition Zone-MTZ. Downward progression from d³⁴S=+13.6‰ to ~MORB values suggest decreasing water/rock ratios during hydrothermal alteration. Here, near complete recrystallization under greenschist/amphibolitic facies conditions (no magmatic sulfides), coupled with strong sulfur (TS_sulf>2 ppm) and copper leaching (>1 ppm), document the high-T reaction zone of the hydrothermal system overlying the axial melt lens, where S and metals are sourced to form VMS deposits. Although multiple sulfur isotope systematics for Oman VMS ores indicates a deep S-source within the range of GT3 reaction zone (d³⁴S ~4‰), REE patterns and trace metal endowments in the ores suggest that the footwall lavas are also a source of metals, in addition to those leached from the deep reaction zone. Crucially, metal leaching and S-isotopic shifts are far more extensive than those reported on in-situ oceanic crust, implying a net addition of seawater-S ~30% to the upper crustal section.

Differences between in-situ and ophiolitic lower crustal sections are seemingly less pronounced: the foliated and layered gabbros (GT2-GT1 drillholes) preserve small S-isotopic shifts relative to MORB, implying that formation of secondary sulfides involved minor S-seawater input (~7%) and mostly redistribution of magmatic-S. Wide fault zones of convincing oceanic origin preserve sulfates with composition similar to Cretaceous seawater (d³⁴S~+18‰) supporting the role of focused fluid flow corridors during deep crustal cooling. TS_sulf and Cu+Ni concentrations increase throughout the lower crust while strong Cu+S leaching characterize tectonized and low-T hydrothermally overprinted domains. Above the MTZ, the primitive layered gabbros and intercalated ultramafics (CM1 drillhole- Sequence SI) record metal and TS_sulf enrichments related with magmatic sulfide saturation/segregation from mantle melts upon entering the crust. Incompatible element rich pegmatoidal dikelets crosscutting SI include late, high-fS₂ sulfides formed during low-T BSR (δ³⁴S>-25.8‰).

The MTZ comprises 90m of fully serpentinised dunite (SII) underlain by dunite with rodingitized gabbro (SIII). The SII-dunites show vanishing TS_sulf and Cu concentrations, consistent with desulfurization producing alloy-bearing mineral assemblages formed during extremely low fS₂-fO₂ conditions, typical of early serpentinization stages. The dunites mark the onset of increasing S-isotopic shifts towards the SIII-rodingites The occurrence of both sulfides (δ³⁴S₌+1.4, +56.9‰) and sulfates (δ³⁴S_SO4=+19.4, +36.5‰) with δ³⁴S>>Cretaceous seawater sulfate can be explained by input of fluids at the top of SII-dunites which composition progressed towards extreme heavy values during closed-system, multi-staged evolution.

AJ acknowledges WWU International Visiting Scholars and EU-H2020 Marie Sklodowska-Curie #894599 Fellowships, and FCT I.P./MCTES PIDDAC–UIDB/50019/2020- IDL.

How to cite: Jesus, A. P.: Sulfur and metal fluxes in the oceanic crust: the Samail  ophiolite as proxy for fast spreading ridges., EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17478, https://doi.org/10.5194/egusphere-egu23-17478, 2023.

Based on a newly established profile through fast-spreading oceanic crust of the Oman ophiolite and on cores drilled within the ICDP Oman Drilling Project (OmanDP), we present here the results of 12 years research, focusing on the nature of the magmatic accretion of the deep crust beneath fast-spreading mid-ocean ridges. We established a 5 km long profile through the whole plutonic crust of the Oman ophiolite by systematic outcrop sampling in the Wadi Gideah (Wadi Tayin Block near Ibra), providing the reference frame for the 300 to 400 m long OmanDP drill cores GT1 and GT2 (lower crust, mid-crust), as well as CM1 and CM2 (crust-mantle boundary) drilled into the same area.
The results allow implication on the mechanism of accretion of fast-spreading lower oceanic crust. Depth profiles on bulk rock and mineral compositions, crystallization temperature and microstructures combined with petrological modeling reveal insights into the mode of magmatic formation of fast-spreading lower oceanic crust, implying a hybrid accretion mechanism. The lower 2/3 of the crust (mainly layered gabbros) was formed via the injection of melt sills and in situ crystallization. Here, upward moving fractionated melts mixed with more primitive melts through melt replenishments, resulting in an upward differentiation trend. Since the fraction of crystallization is only small, upmoving melts could easily transport the latent heat produced by deep crystallization upward. The upper third of the gabbroic crust is significantly more differentiated, in accord with a model of downward differentiation of a parental melt originated from the axial melt lens sandwiched between the gabbroic crust and the sheeted dike complex. While the 5 km long profile shed light on the overall magmatic accretion process, the Oman DP drill cores showing ~ 100% recovery allowing high density sampling provide incredible details on the magmatic accretion process. Examples are the identification of individual melt sills from which the layered gabbro section has been formed (drill core GT1) or the detailed observation of olivine accumulation at the base of the crust (drill cores CM1/CM2).

How to cite: Koepke, J., Garbe-Schönberg, D., Mock, D., and Merseburger, S.: Accretion of fast-spreading oceanic crust: benefits from large-scale sampling in the Oman ophiolite in combination with cores drilled by the ICDP Oman Drilling Project, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-11852, https://doi.org/10.5194/egusphere-egu23-11852, 2023.

At mid-oceanic ridges, mantle temperature and magma supply influence the structure of the neo-volcanic zone. Due to the large Romanche offset, a strong “cold edge” effect is present at its eastern intersection with the Mid-Atlantic Ridge. This effect decreases with the distance from the transform fault, making this region an ideal area to study the impact of the thermal gradient on the architecture of the neo-volcanic zone. We analyzed seafloor videos and photos from submersible dives, as well as bathymetry and backscatter data collected during the SMARTIES cruise (2019), from the Ridge-Transform Intersection (RTI) to approximately 80 km to the south of it. We produced maps at local and regional scales and quantified the morphology of volcanoes (height, diameter, height/diameter ratio, volume and surface). Visual observations have showed that the seafloor is mainly made up of pillows or elongated pillows and rare massive lava flows. Within 30 km of the RTI, the neo-volcanic area is characterized by clusters of volcanoes, affected by faults trending N120-130° E, oblique to the extension and to the transform fault orientation, and by faults trending E-W. At 30 km to 50 km from the RTI, the Central segment displays a robust volcanic ridge oriented N150°E built by a pilling of volcanoes and narrow ridges. Its eastern and southern parts are old and characterized by oblique faults and narrow ridges (N130-140°E), while the northwest portion is more recent, faults and ridges are almost normal to the extension. The southernmost segment, located at 80 km from the RTI, is orthogonal to the spreading direction and asymmetric, bounded at the west by a detachment fault. Recent volcanic edifices were observed from the center of the segment to the base of the detachment. Our observations suggest that the neo-volcanic area is fed by more and more magma from north to south. This increase in magma supply is marked by a more structured volcanic axis, volcanoes that are more voluminous and a change in the orientation of the segments and faults. Changes in the orientation of faults and off-axis abyssal hills also reveal variations in the magmatic supply over time.

How to cite: Grenet, L., Maia, M., Hamelin, C., Briais, A., and Brunelli, D.: Construction of the neo-volcanic area of a slow-spreading ridge in a cold mantle context: Mid-Atlantic Ridge Eastern Intersection with Romanche Transform Fault, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-742, https://doi.org/10.5194/egusphere-egu23-742, 2023.

Lithospheric inheritance is known to strongly influence the spatial and temporal patterns of continental deformation in all geodynamic contexts, emphasizing the role of rheological feedbacks between time-spaced geodynamic events. In principle, the transition from continental rifting to sea-floor spreading at diverging plate boundaries marks a threshold beyond which these long-term feedbacks no longer apply. This is because sea-floor spreading is accompanied by the creation of new lithosphere from melting and cooling of the underlying and uprising mantle, which should make lithospheric inheritance negligible at oceanic plate boundaries. However, whether and how lithospheric inheritance continues to affect oceanic plate boundary processes after the continental rifting to sea-floor spreading transition is reached has so far not been explored in detail.

As a young oceanic rift that broke up the Arabia-Nubia Shield and its mosaic of Proterozoic accreted blocks, the Red Sea (RS) represents an ideal case to study these specific lithospheric inheritance effects. We performed a quantitative morpho-structural analysis designed to track along-axis variations of the magmato-structural architecture of the RS plate boundary and to explore its relationships with the inherited structures of the rifted continental plates. Specifically, faults and sea-floor morphology have been mapped over the post-5.3Ma extent of oceanic crust from Global Multiresolution Synthesis (including multibeam surveys) bathymetry. The structural and magmatic patterns have then been extracted by quantifying four metrics: the axial depth, the slope of the central-trough flanks, the proportion of exposed volcanic sea-floor, and the distribution of normal-fault offsets.

This analysis reveals that anomalously deep segments bounded by steeper-than-average flanks bound the central RS in the North and South. Furthermore, it shows that this specific axial topography occurs where the structural pattern locally switches from regularly-spaced and moderate-displacement (~400m) normal faults to one dominant large-displacement (~1200m) fault as well as coinciding with a lower proportion of volcanic sea-floor (15-20% versus 70% on average along the rest of the axis). This distinct magmato-structural signature is commonly interpreted to reflect a decreased fraction of plate separation accommodated magmatically along slow and ultra-slow spreading ridges, in agreement with tectono-magmatic interaction models: individual faults that form near the axis remain active longer and accumulate more displacement when this fraction decreases. On the other hand, a decreased magma input would result in a thinner crust, and thus isostatically account for the anomalous depth of these segments.

The location of these two magma-starved segments appears unrelated to variations in spreading rate or to the segmentation of the RS axis, but stands in the prolongation of two major Proterozoic suture zones within the Arabia-Nubia Shield. On the Arabian side, both of these two inherited structures coincide with a rise of the lithosphere-asthenosphere boundary (LAB) as mapped from S-to-P receiver functions. We therefore propose that on-axis magma starving results from local outward spreading of the upper-mantle upwelling, in turn driven by its off-axis channeling along the LAB topographic highs. Thereby, heat and eventually melts would be transferred from beneath the axis to beneath the onshore suture zones, possibly fueling the Plio-Pleistocene volcanic activity observed there.

How to cite: Moulin, A. and Jónsson, S.: Lithospheric inheritance controls on early sea-floor spreading: new insights from magmato-structural patterns along the Red Sea axis, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-11070, https://doi.org/10.5194/egusphere-egu23-11070, 2023.

At ultra-slow ridges, tectonics, hydrothermalism, serpentinization and magmatism interact to build the oceanic crust. How this heterogenous crust forms and relates to faulting remains poorly understood, but is key for elucidating hydrothermal flow patterns and their implications for ocean-lithosphere element exchange. Along the melt-poor Southwest Indian Ridge (SWIR) at 64°30' East, crustal thickness varies across the ridge strike, with crustal thickening attributed to serpentinization extending downward along detachment faults, DFs. This observation calls into question the commonly assumed relationship between local crustal thickening and magma-supply increase. Here we use 2D numerical models to analyze how coupled tectonics, mantle melting, magma emplacement and serpentinization interact. Our model includes hydrothermal cooling, ocean loading, and the oceanic crust density. We reproduce the observed bathymetry at SWIR, 64°30'E, which is shaped by alternating DFs formed in flip-flop mode. Our results show that the offset and duration of DFs are controlled by ocean loading and crustal density. Importantly, shallow faulting and deeper mantle flow are coupled: long-lived DFs result in relatively slower mantle upwelling, lower melt supply, but crustal thickening due to deeper serpentinization, ~5 km, consistent with the observed thick ultramafic crust in nature. In between alternating DFs, mantle upwelling is faster, melt supply higher, and serpentinization shallower, < 2km. Since magmatic crustal thickness is overall very small, 1.5-2 km, changes in faulting-induced serpentinisation depth, are the main cause for observed variations in crustal thickness, 2-7 km. We conclude that, at melt-poor ridges, tectonics controls both crustal thickness variations and melt supply oscillations.

How to cite: Mezri, L., García-Pintado, J., Pérez-Gussinyé, M., Liu, Z., Bach, W., and Cannat, M.: Tectonics controls on melt production and crustal architecture during nearly amagmatic seafloor spreading, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-12073, https://doi.org/10.5194/egusphere-egu23-12073, 2023.

The eastern Southwest Indian Ridge (SWIR) is a melt-poor end-member region of the global MOR system. The available magma focuses to axial volcanoes, leaving >50 km-wide, nearly amagmatic ridge sections, where seafloor spreading occurs via large offset detachment faults. We present map to outcrop scale observations of the deformation associated to one of these detachments, in the 64°35'E region of the SWIR. This active detachment fault presently has a horizontal offset of ~4 km (Cannat et al., 2019), and accommodates nearly all the plate divergence (14 km/million year; Patriat and Segoufin, 1988). We focus on the lower slopes of the footwall, where this active fault currently emerges at an angle of ~35°. The emergence is traceable across a length of ~20 km on side-scan sonar and shipboard bathymetry data. It locally shows undulations at a wavelength of ~1-4 km. High-resolution bathymetry at and near the emergence area shows two morphological domains. In one domain, the exhumed fault surface bears distinct corrugations that trend at an 18° to 33° angle to the spreading direction, extends up to 300 m, are spaced by ~15-300 m, and are ~1 m to ~40 m in amplitude. In the other domain, the exhumed fault is not corrugated. Remotely operated vehicle (ROV) dive observations at the outcrop scale show discrete planar fault planes and brecciated and fractured rock forming the top ~1-4 meters of the corrugated exposures. In contrast, the non-corrugated fault exposures show up to ~8 m of gouge-bearing micro-brecciated domains, including several up to 1 m thick horizons of semi-brittle sheared serpentinites. Dive observations further suggest that: (1) there are several sigmoidal intercalations of such gouge-bearing horizons forming the upper few tens of meters of the non-corrugated fault zone, and (2) the horizons of sheared serpentine originated as brittle cracks that served as hydrous fluid pathways into the fault damage zone. We propose: (1) that the absence of corrugations is related to the overall weaker semi-brittle rheology of the emerging fault in this domain, compared to the purely brittle corrugated domain; and (2) that the two domains represent damage developed in distinct conditions of temperature and hydrothermal fluid availability. At the broader map scale, the non-corrugated domain to the east emerges about 1.2 km farther south than the corrugated domain, and the trace of emergence thus draws an indentation between the two domains. Given the ~35° fault emergence angle in the two domains, we infer that their across-fault distance is ~650 m. The detachment damage zone may thus be at least that broad, and comprised of distinct, probably anastomosing domains of more localized deformation, which would preferentially be exposed at the seafloor. This damage zone anatomy would be consistent with seismic refraction observations (Momoh et al., 2017) in the area.

How to cite: Mahato, S. and Cannat, M.: Anatomy of a detachment fault damage zone at a nearly amagmatic mid-ocean ridge: observations from outcrop to map scale, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-6200, https://doi.org/10.5194/egusphere-egu23-6200, 2023.

Hydrothermal alteration of seafloor basalts alters its elemental and isotopic composition. Studies on dredged basalts and ophiolite sequences using stable O, K, and radiogenic Sr ^[1,2,3] isotopes have documented the effect of seafloor alteration on such lithologies. Experimental studies of basalt-seawater interaction have also demonstrated exchange of Sr isotopic signatures between these two phases ^[4] while limited data for altered oceanic crust suggests incorporation of heavier Sr isotopes ^[5]. To further our understanding of the behaviour of Sr during seafloor alteration in natural settings, we measured ⁸⁷Sr/⁸⁶Sr and δ^88/86Sr in a suite of variably altered lithologies from Hess Deep Rift (HDR), which include basalt, norite, gabbro and troctolite.

Radiogenic Sr (⁸⁷Sr/⁸⁶Sr) was measured using TIMS (Thermo Scientific, Triton Plus), using internal normalization while stable Sr isotopes (δ^88/86Sr, reported relative to NIST SRM 987) were measured using a double spike (⁸⁴Sr-⁸⁷Sr) TIMS technique, both at the Centre for Earth Sciences, IISc, Bangalore. The δ^88/86Sr values of the HDR samples (0.308-0.810 ‰) are higher than the bulk silicate Earth (BSE) value (0.27 + 0.05 ‰) ^[6]; some samples show δ^88/86Sr values higher than modern-day seawater value (0.386 ‰) ^{[e.g., 7]}. The ⁸⁷Sr/⁸⁶Sr varies from ~0.703 in unaltered samples to ~0.709 in altered samples, the latter close to the modern-day seawater value. Overall, our data suggests incorporation of heavier isotopes of Sr in altered oceanic crustal samples; the heavier than seawater δ^88/86Sr values observed in some samples reflect formation of new mineral phases, consistent with high δ^88/86Sr observed in anhydrite formed in laboratory experiments of basalt-seawater interaction^[4].

[1]. Lamphere et al. (1981) Journal of Geophysical Research: Solid Earth, 86(B4), pp.2709-2720; [2]. McCulloch et al. (1981) Journal of Geophysical Research: Solid Earth, 86(B4), pp.2721-2735; [3]. Parendo et al. (2017) Proceedings of the National Academy of Sciences, 114(8), pp.1827-1831; [4]. Voigt et al. (2018) Geochimica et Cosmochimica Acta, 240, pp.131-151; [5] Klaver et al. (2020) Geochimica et Cosmochimica Acta, 288, pp.101-119; [6] Moynier et al. (2010) Earth and Planetary Science Letters, 300(3-4), pp.359-366; [7]. Ganguly and Chakrabarti 92022) Journal of Analytical Atomic Spectrometry, 37(10), pp.1961-1971.

How to cite: Haldar, U., Sensarma, S., and Chakrabarti, R.: Significant variability in 87Sr/86Sr and d88/86Sr in Hess Deep Rift lithologies due to hydrothermal alteration., EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-15377, https://doi.org/10.5194/egusphere-egu23-15377, 2023.

The Eastern Gulf of Guinea hosts several buried Cretaceous-aged oceanic fracture zones. 3D broadband seismic data acquired offshore São Tomé and Príncipe revealed a complex crustal architecture. Mapped oceanic fracture zones show low-angle reflectors that detach onto or eventually cross through the Moho boundary, overlain by strong reflectors that are interpreted as transform process related extrusive lava flows. Here, we use a high resolution shipborne free-air gravity and total field intensity magnetic data set to reassess whether previously defined seismic models of the crust are in conformity with potential field data. The study area is located offshore São Tomé with a size of c. 150x150 km. Using the software IGMAS+, we model the gravity and magnetic properties of the crust (i.e. density and susceptibility) in 3D. Long record length seismic sections plus mapped seismic horizons, which include bathymetry, sediments, upper and lower crust, are used as constraints. While the general trend of the free-air anomaly can be explained within a range of expected crustal densities, the magnetic field anomaly reflects high residuals that are predominantly oriented parallel to the transform faults. This indicates that gravity and magnetic data cannot be explained by the same simple source geometry. Therefore, we first perform sensitivity tests to isolate the source of the residual magnetic anomaly, followed by a structural analysis along the transform faults with special emphasis to the extrusive lava flows in the crustal domain. Our final model reconciles seismic horizons and potential field data and will stimulate a discussion on the architecture and evolution of transform faults and their signatures in different data sets.    

How to cite: Haas, P., Thomas, M., Heine, C., Ebbing, J., Seregin, A., and van Itterbeeck, J.: Oceanic transform faults offshore São Tomé and Príncipe highlighted by integrated density and magnetic modeling of the crust, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3138, https://doi.org/10.5194/egusphere-egu23-3138, 2023.