Abstract
Oceanic detachments are deep-rooted, long-lived, plate-scale structures and serve as fluid conduits introducing water into the oceanic lithosphere, impacting plate rheology and potentially inducing oceanic crustal melting. However, the extent and mechanisms of fluid ingress and crustal melting during detachment faulting remains poorly constrained. Here we examine felsic veins from the Atlantis Bank oceanic core complex, Southwest Indian Ridge, to elucidate oceanic detachment controls on crustal melting. We suggest that the felsic veins are products of strong fractionation of either primitive basalts or magmas generated by hydrous melting of gabbros (i.e., anatectic melts). The anatectic felsic veins are proximal to the fault plane, suggesting that detachment fault facilitated high-temperature (750–900 °C) seawater infiltration into deep oceanic crusts. Our findings highlight the essential role of detachment faulting played in the fluid ingress and melting of oceanic crust, bearing implications for chemical and heat exchange between seawater and oceanic lithosphere.
Similar content being viewed by others
Introduction
Oceanic lithosphere, the most important chemical reservoir recycled at subduction zones1,2, is created at mid-ocean ridges with a total length of ~60,000 km3. Comprehensive investigations of the tectonic, magmatic, and hydrothermal activities during the formation of oceanic lithosphere provides essential information of chemical exchanges on the Earth1,2. Slow- to ultraslow-spreading ridges, with a full spreading rate <40 mm yr-1, comprise more than three-fifths of the global ridge system and are important sites for mass and energy exchange between the interior and the exterior geo-spheres3,4,5. Different from the fast-spreading ridges, the slow- and ultraslow-spreading ridges are characterized by tectonic-controlled lithospheric accretion through detachment faulting6, with the development of oceanic core complexes (OCCs)6,7,8. Detachment faulting is a fundamental mode of lithospheric accretion that comprises of ~50% of slow-spreading ridges and ~40% of ultraslow-spreading ridges9,10. Detachment faults root deep into oceanic lithosphere (e.g., up to 13 km below the seafloor11), creating pathways for vigorous hydrothermal circulation11,12, crucial for cooling of lithosphere13 and formation of hydrothermal vent11, and might potentially trigger the hydrous melting of oceanic crust14,15,16,17. Fundamental variabilities in magma supply are exhibited by detachment faulting in slow- and ultraslow-spreading ridges8,18,19,20. Except for the nearly amagmatic OCCs [e.g., Southwest Indian Ridge (SWIR) at 61 − 67°E] where detachment faults accompanied nearly 100% plate divergence18,19, the formation of many OCCs are associated with variable magma supply8,20. Magma supply provides heat and elements to sustain long-lived hydrothermal activities21, and thus exerts a multifaceted influence on OCC development and the associated hydrothermal systems. Hence, the tectonic-magmatic interplay during detachment faulting is crucial for understanding of the chemical and heat budgets in oceanic lithosphere. Yet, how and to what extent the detachment faulting might influence the deep hydrothermal circulation and the magmatic processes, such as oceanic crustal melting, remain poorly understood.
Felsic rocks, comprising diorites, tonalites and trondhjemites17, are minor lithological components in oceanic crust22. They offer essential records of magmatism, tectonism, and hydrothermal processes15,23 and are potential to unravel the tectonic-magmatic interaction during detachment faulting17. Oceanic felsic rocks have been proposed to form via (i) hydrous partial melting of mafic rocks22, (ii) mid-ocean ridge basalt (MORB) fractional crystallization17,24, and (iii) liquid immiscibility25. However, discriminating these competing models remains challenging, and the mechanisms governing the formation and distribution of felsic melts in the oceanic lithosphere remains unclear.
Here, we explore the origin of felsic rocks from the Atlantis Bank OCC on the SWIR using zircon and apatite data coupled with petrochemical modeling. Our results demonstrate that detachment faulting was instrumental in triggering felsic melt generation and modulating their geochemical characteristics at the Atlantis Bank OCC. Additionally, the origin of felsic rocks from two OCCs on the Mid-Atlantic Ridge was examined to illustrate the complex interactions among magmatism, detachment faulting, and hydrothermal activities. The results suggest that oceanic felsic rocks serve as indicators of tectonic-magmatic interactions and deep hydrothermal circulation during OCC development.
Results
The footwall sequence of the Atlantis Bank OCC
The Atlantis Bank OCC is a 660-km2 dome that situates at 57oE of SWIR with a full spreading rate of ~14 mm yr−1, mainly comprising gabbros generated during 13.4 to 10.5 Ma (Fig. 1a)26. This dome rises from 5700 meters below sea level (mbsl) at the transform floor to 689 mbsl at a 25-km2 wave-cut platform. The Atlantis Bank OCC has been drilled at three sites on its summit, namely, 1508-m ODP Hole 735B, 158-m ODP Hole 1105 A, and 809-m IODP Hole U1473A27; Hole U1473A is 1 to 2 km away from the Hole 735B and Hole 1105 A (Fig. 1b, c). Atlantis Bank OCC was once uplifted above sea level by high-angle normal faulting, resulting in erosion near its surface28. As the three drilling sites are situated in the wave-cut platform, the effect of erosion should be assessed before reconstructing the footwall sequence.
a Topography of the SW Indian Ridge (SWIR) and bathymetric map of Atlantis II Transform Fault54, showing the ___location of Atlantis Bank OCC. b Bathymetric map of Atlantis Bank OCC showing the locations of IODP and ODP drilling sites28. c Stratigraphy of lithology and shear sense for ODP Hole 735B, 1105 A and IODP Hole U1473A17,27 and the schematical illustration of the reconstructed footwall sequence of Atlantis Bank OCC. Normal and reverse denote crystal-plastic fabric with normal and reverse shear sense, respectively. Also shown in panel c is the locations of felsic veins and oxide gabbros with zircon trace element data (Supplementary Data 12). The boundary between extension-dominated and compression-dominated zones is highlighted in dark dashed line.
During detachment faulting, the upper and lower portions of the footwall display extensional and compressional stress fields, respectively, driven by the plate bending due to footwall rotation17,27,29. The gabbros captured these stress regimes during their intruding into the footwall, resulting in the spatial distribution of normal and reverse shear senses within the gabbro bodies17,27,29 (Fig. 1c). Therefore, the stress state of gabbros can be utilized to infer their primary locations relative to the fault plane29; namely, the intervals with normal and reverse shear senses are proximal and distal to the fault plane, respectively (Fig. 1c).
Hole 735B exhibits remarkable stratigraphic variations in ductile shear sense and lithologies, and is, thus, grouped into two portions, i.e., 0–640 mbsf (meters below seafloor) and 640–1508 mbsf (Fig. 1c). Its upper portion (0–640 mbsf) show enrichments in oxide gabbros and crystal-plastic fabrics with normal sense indicators (Fig. 1c)17,27, whereas the intervals in its lower portion (640–1508 mbsf) contain less oxide gabbros and are characterized by reverse-sense crystal-plastic fabrics (Fig. 1c).
Hole U1473A exhibits stress state and lithological stratigraphy resembling the lower portion of Hole 735B (Fig. 1c). Hole U1473A comprises predominant olivine gabbros intruded by oxide gabbros, without massive oxide gabbro layers as seen in the upper portion of Hole 735B (Fig. 1c). Moreover, shear zones in Hole U1473A mainly display reverse sense17, a feature comparable to the lower portion of Hole 735B (Fig. 1c). The nearly absence of normal shear sense-dominant zones in Hole U1473A reflects more erosion at site U1473 than other sites (Fig. 1c), consistent with site U1473 being closer to the center of the wave-cut platform27. In contrast, Hole 1105 A comprises thick oxide gabbro layers as observed in the upper portion of Hole 735B (Fig. 1c). Such enrichments of oxide gabbros reflect Fe-Ti melt entrapments near the fault plane27, implying that both the Hole 1105 A and 0–640 mbsf of Hole 735B locate at the upper portion of the footwall29 (Fig. 1c).
Overall, the stress state of gabbros from different Atlantis Bank drilling sites reflects their relative distance to the detachment fault plane. Hole 1105 A and the top 640 mbsf of Hole 735B represent the upper portion of the footwall, while the 640–1508 mbsf of Hole 735B and the entire Hole U1473A exhibit reverse shear sense and correspond to the lower portion of the footwall (Fig. 1c). In this study, the felsic veins from the Atlantis Bank OCC are grouped based on their locations within the footwall, i.e., from the upper portion and lower portion of the footwall (Fig. 1c).
Sample description
Twenty-three felsic rocks and one oxide gabbro were collected from the 200–809 mbsf interval of Hole U1473A (Fig. 1b, c) for comprehensive geochemical measurements, and complete datasets are provided in Supplementary Data 1–16. Felsic rocks mainly intrude the gabbros as regular planar dikelets with sharp or irregular boundaries, or as irregularly shaped patches and networks (Fig. 2a, b and Fig. S1), with a thickness from a few millimeters to several centimeters. The felsic veins in the upper 600 meters exhibit stronger crystal plastic fabric intensities (3.1 ± 1.5) than those from the lower 200 meters (crystal plastic fabric intensities of 1.6 ± 1.1, Supplementary Data 1). By contrast, the host gabbros are mostly weakly deformed to undeformed, except for six samples with crystal plastic fabric (Supplementary Data 1). The host gabbros are mostly olivine gabbros and gabbros, whilst only three out of twenty-three felsic veins are associated with oxide gabbros (Supplementary Data 1). The felsic rocks mainly consist of diorite and trondhjemite, with minor tonalite, most of which composed predominantly of plagioclase and amphibole, with or without quartz (Supplementary Data 1). Diorites contain 3−34 vol% amphibole, 49−90 vol% plagioclase, 0 − 12 vol% quartz, and 0 − 8 vol% Fe-Ti oxides. The two tonalites contain 14 − 17 vol% amphibole, 62 vol% plagioclase, 20−24 vol% quartz, and 0 − 2 vol% Fe-Ti oxides. Trondhjemites are made up of 3−9 vol% amphibole, 36 − 88 vol% plagioclase, 17−25 vol% quartz, and 0−2 vol% Fe-Ti oxides (Supplementary Data 1). All U1473A felsic rocks contain abundant zircons in a subhedral to euhedral crystal shape and euhedral apatite is present in most felsic rocks. Other accessory minerals, including titanite, monazite, xenotime, and Fe-Ti oxides, are heterogeneously distributed (Fig. 2c−f).
a,b Representative felsic veinlets crosscutting the gabbros. c–f Representative textures of Hole U1473A felsic veins. The contacts between felsic veins and the host gabbros have been highlighted in yellow dashed curves. Ap apatite, Zr zircon, Mon monazite, Amp amphibole, ilm ilmenite, Xen xenotime.
Geochemical data of U1473A felsic rocks
Plagioclase (Pl) in felsic veins has lower anorthite content [An1–46, An=100×Ca/(Ca+Na)] than that in the host gabbros (An18-60) (Fig. 3a, b). Pl cores in the host gabbros exhibit An contents (An42–60) comparable to the main trends defined by the Pl in olivine gabbros, whereas Pl rims in host gabbros show low An (An18–44) that are consistent with Pl in the associated felsic veins (Fig. 3b). Brown amphibole (Amp) in felsic veins are pargasite to Mg hornblende (Fig. 3c) and has Mg# [100×Mg/(Mg+Fe)] of 60–75 (Fig. S2a). Amp in U1473A felsic vein exhibit lower Na2O and TiO2 abundances (Na2O = 1.2 to 3.0 wt.% and TiO2 = 1 to 3.5 wt.%) than those in olivine gabbros (Na2O = 2.5–3.5 wt.% and TiO2 = 2.0–4.2 wt.%, Fig. S2b). A weak Amp Mg#-Pl An% correlation is shown by the U1473A felsic veins, with higher Amp Mg# at a given Pl An than the main trend defined by Atlantis Bank gabbros (Fig. S2a). Based on the results of thermobarometers30,31, U1473A felsic vein Amp is inferred to crystallize under lower temperatures (683–837 oC; Fig. S2c) and pressures (0.5–3.6 kbar) than those from the gabbros ( > 850 oC, 2–4 kbar; Fig. 3d). Based on the published model30, the U1473A felsic vein Amp yield equilibrium melt H2O abundances of 5.0 ± 0.7 wt.% (Fig. S2d).
a Feldspar (anorthite [An]-albite [Ab]-orthoclase [Or]) compositional diagram, with compositions of plagioclase experimentally produced by hydrous partial melting of gabbros14 and calculated plagioclase compositions resulting from advanced fractional crystallization24 shown for comparisons. b Stratigraphic variations of plagioclase An for Hole U1473A felsic veins and the host gabbros, with the published olivine gabbro plagioclase89 data shown for comparisons. c Amphibole Na+K (f.u) versus Si (f.u) plot. d Calculated crystallization pressure and temperatures based on amphibole thermobarometry30,31.
Zircon in felsic veins show flat to weak oscillatory zoning, characteristic of igneous zircon crystallized from mafic magmas, whereas those in sample 34R-4-41-45 have porous textures (Fig. S3a−c). Zircon exhibits subtle chemical zoning, with rims enriched in Hf, U, and Y relative to cores (Fig. S3d−f). Zircon in U1473A felsic veins yield 206Pb/238U weighted average ages of 11.3–12.5 Ma (Figs. S4−S5), without robust stratigraphic variations (Fig. 4a). U1473A felsic vein zircon exhibits large variations in rare earth element (REE) compositions, i.e., Yb=251–8976 ppm, Eu/Eu*=0.1–0.6 [Eu/Eu*=EuN/(SmN*GdN)0.5, N denotes the CI-chondrite normalized values32. Figure 4b, c]. Crystallization temperatures for zircon in U1473A felsic veins were computed using the Ti-in-zircon thermometer33,34, assuming aTiO2 value of 0.7 and aSiO2 value of 135. Ti-in-zircon temperatures are variable (650–900oC, Fig. 4d), and comparable to those (683–837oC) derived from Amp-Pl thermometer31 and overlapping the temperatures for amphibolite veins of high-temperature hydrothermal origin36. Remarkably, zircon Yb contents, Eu/Eu* and Ti-in-zircon temperatures vary remarkably within individual sample (Fig. 4b−d). For instance, sample U1473A-24R-1-120-127 exhibits zircon Yb, Eu/Eu* and Ti-in-zircon temperatures spanning the entire compositional range of U1473A felsic veins (Fig. 4b−d), consistent with the protracted crystallization of zircon in oceanic magma reservoirs, particularly at slow spreading rates37,38. U1473A felsic veins display zircon εHf(t) of +13 ~ + 15 (t = 12 Ma, Fig. 4e), consistent with previous whole-rock results of Hole 735B24.
a zircon U-Pb ages (2σ error), b zircon Yb contents, c Eu/Eu*, d Ti-in-zircon temperatures33, e zircon εHf(t). Temperature spectrum for amphibolite veins has been previously reported36. εHf(t) denote the εHf values at age of 12 Ma. Hf isotopic data on ODP Hole 735B felsic veins have been previously reported24.
The apatite in U1473A felsic veins are classified to fluoroapatite and hydroxyapatite, with the molar proportion of chlorine (XCl=0.01–0.22) considerably lower than apatite in East Pacific Rise felsic rocks39 (XCl=0.1–0.8; Fig. S6). U1473A apatite exhibit variable REE contents (ΣREE = 1460–47401 ppm) and Eu/Eu* (0.06–0.50) from sample to hole scale (Fig. 4a, b). U1473A felsic veins display apatite εNd(t) of +9 ~ + 10 (t = 12 Ma, Fig. 5c).
a apatite total REE abundances (ΣREE), b apatite Eu/Eu*, c apatite εNd(t).
Zircon in most U1473A felsic veins exhibit mantle-like δ18O (5.1–5.6‰), with no correlation with indicators of magma differentiation like Eu/Eu* (Fig. 6a), except for the porous zircon with δ18O values (4.8–5.1‰) below mantle-like values40 (5.3 ± 0.6‰). In contrast, apatite displays more variable δ18O values of 1.4‰ − 7.1‰ (Fig. 6b, c). Moreover, apatite δ18O values display positive correlation with Eu/Eu* and negative correlation with Yb contents (Fig. 6b, c). Zircon-apatite δ18O fractionation factor (expressed as ∆δ18OZircon-apatite=δ18OZircon-δ18OApatite) increases from -1.0 to +1.0 with the increasing apatite Yb contents (Fig. 6d).
a Plot of zircon δ18OV-SMOW against Eu/Eu*, b apatite δ18OV-SMOW versus Eu/Eu*, c apatite δ18OV-SMOW versus Yb contents, d Plot of ∆δ18OZircon-apatite against apatite Yb contents. HPM=hydrous partial melting, FC=fractional crystallization.
Phase equilibria modeling of hydrous partial melting of gabbros
We performed a phase equilibria modeling of hydrous partial melting (HPM) of gabbros at 2 kbar. Gabbros were chosen as the protoliths for HPM modeling, owing to the scarcity of diabase in the Atlantis Bank OCC28. Four representative gabbro compositions, encompassing low-Al and high-Al olivine gabbros and low-Al and high-Al oxide gabbros, were selected to span the compositional range of U1473A gabbros (Fig. S7). Phase equilibrium (temperature-XH2O pseudosection) diagrams were conducted using GeoPS software41. More detailed information of activity models and parameters for phase equilibria modeling are provided in Methods.
Modeling reveals that the temperature and water content (XH2O) ranges for HPM depend on protolith compositions, i.e., 820–900oC and >0.4 wt.% XH2O for olivine gabbros and 750–840oC and >0.4–2.0 wt.% XH2O for oxide gabbros (Fig. 7). Higher Fe and Al contents in the protolith lowers the solidus temperature, thus facilitating melting (Fig. 7). The residual phases consist of Amp, Pl, pyroxenes, and ilmenite.
a Low-Al olivine gabbro. b High-Al olivine gabbro. c Low-Al oxide gabbro. d High-Al oxide gabbro. Compositions of the protoliths for HPM are shown in Supplementary Data 13 and Fig. S7. The red areas labeled in each panel represent the temperature-H2O ranges for HPM, which are utilized in the trace element modelings of equilibrium melting. Solid yellow lines represent the solidus at variable water contents, and the white lines with numbers label the weight percentage of residual liquids. Detailed mineral and melt fractions at different temperatures, and XH2O are given in Supplementary Data 13.
Trace element modeling of felsic melt formation
To quantitatively constrain the genesis of felsic veins quantitatively, a series of petrochemical modeling were employed, including hydrous partial melting (HPM), fractional crystallization of anatectic melts, and fractional crystallization of MORBs (Methods). Since the felsic veins represent mixtures of melts and incompletely segregated crystals, the bulk compositions of felsic veins would bias the corresponding melt compositions24. Rock-forming minerals like Amp and Pl in felsic veins are susceptible to post-magmatic alteration. Zircon remains mostly unaltered in felsic veins, therefore, offering insights into magmatic processes. The accessory mineral fractionation and oxygen fugacity have limited effects on the zircon REE compositions of Atlantis Bank felsic veins (see Methods for evaluation). Instead, the zircon Yb and Eu/Eu* for oceanic felsic melts mainly reflect the melt evolution, crystal fractionation, and initial melt compositions. Here, we performed petrochemical modeling of various processes to examine the mechanisms that control the zircon REE variability for Atlantis Bank felsic veins.
Trace element modeling of HPM was conducted using the batch melting equation, using the residual melt fraction and phase fractions at varying temperatures obtained by phase equilibrium modelings. Notably, only HPM of the high-Al oxide gabbro generates substantial amounts of Zr-saturated anatectic melts, whereas the partial melts of other protoliths are Zr-undersaturated (Supplementary Data 13). This implies that, except for those derived from oxide gabbros, the felsic melts produced by HPM of most oceanic gabbros without extensive fractional crystallization are unlikely to contain zircon. As shown in Fig. 8a, zircon in equilibrium with the anatectic melts show low Yb abundances ( < 1000 ppm) with Eu/Eu* decreasing from 0.4 to 0.2.
a Zircon Yb-Eu/Eu* plot for the felsic veins from reverse-stress-sense-dominant zones (the lower portion of footwall), i.e., Hole U1473A and 640–1508 mbsf of Hole 735B). b Zircon Yb-Eu/Eu* plot for the felsic veins from normal-stress-sense-dominant zones (the upper portion of footwall), including 0–640 mbsf of Hole 735B, Hole 1105 A, and one seafloor sample JR31-12-29. c Zircon Yb-Eu/Eu* plot for the Atlantis Bank oxide gabbros grouped by their reconstructed locations in detachment footwall. Insert shows the interpretations of different zircon REE trends. Literature data sources are given in Supplementary Data 12. FV felsic veins, OxG oxide gabbros, HPM hydrous partial melting, FC fractional crystallization.
Although the modeled anatectic melts are mostly zircon-undersaturated, they can reach zircon saturation via protracted fractional crystallization. We conducted trace element modeling of fractional crystallization of the anatectic melts. The weight averaged partial melts produced by HPM of each protolith were selected as the initial melts. The modeling results show that the Zr-undersaturated anatectic melts can reach zircon saturation after 20–91% fractional crystallization (Fig. 8b and Supplementary Data 14). Meanwhile, the modeled zircon by fractional crystallization of anatectic melts exhibit an increase in Yb contents with decreasing Eu/Eu*, with slopes becoming steeper at Eu/Eu*<0.2 (Fig. 8b).
Finally, we conducted trace element modeling of MORB fractional crystallization. Two representative Atlantis Bank MORB (MgO of 5.3–13.3 wt.%) are selected as the initial melts. Experimental results of low-pressure MORB crystallization42,43 were utilized to reconstruct the fractionating phase fractions and the melt compositions. As shown in Fig. 8a, during the fractional crystallization of MORBs, liquidus zircon has Eu/Eu* of 0.3–0.4, and exhibits a rapid increase in Yb contents with decreasing Eu/Eu*. Compared with zircon derived from fractional crystallization of anatectic melts or HPM of gabbros, zircon of MORB fractional crystallization origin show higher Yb contents at a given Eu/Eu (Fig. 8a, b).
Discussion
Hole U1473A felsic rocks formed by high-degree differentiation of MORB
Several lines of evidence argue against the formation of Hole U1473A felsic rocks by hydrous partial melting (HPM) of gabbros. Firstly, HPM of gabbros tends to produce melts with depletion in TiO2 and REE, due to the depleted nature of gabbroic cumulates22,42. Phase equilibria modeling suggests that felsic melts generated by HPM of olivine gabbros are unsaturated in zircon (Figs. 7 and 8). However, the U1473A felsic veins contain abundant Fe-Ti oxides and zircon, together with other U-bearing accessory minerals, e.g., apatite, monazite, and xenotime (Fig. 2c–f and Supplementary Data 1). Moreover, high REE contents of both apatite and zircon support that they crystallized from REE-rich parental magmas (Figs. 4 and 5), which are unlikely products of HPM processes. Petrochemical modelings also suggest that zircons crystallized from felsic melts of pure HPM origin should contain <1000 ppm Yb, which are far less than the Yb contents (1000 to 9000 ppm) of most U1473A zircons (Fig. 8a). Secondly, previous experiments on HPM of gabbros demonstrated the formation of Ca-rich Pl, which have higher An contents than plagioclase in both U1473A felsic veins and their host gabbros (Fig. 3a). The host gabbros for U1473A felsic veins display Pl An contents consistent with the main trends defined by the olivine gabbros (Fig. 3b), without anomalously high An content as predicted by HPM22. Thirdly, except for the porous zircons in one sample (34R-4-41-45), all U1473A felsic rocks exhibit high Yb contents in apatite and zircon and mantle-like zircon δ18O values (Fig. 6), reflecting limited involvement of seawater component44. The porous zircon in sample 34R-4-41-45 exhibit relatively low δ18O values of 4.8–5.1‰ and relatively low Ti-in-zircon temperatures ( ~ 700oC), which might be attributed to post-crystallization seawater overprinting45.
Therefore, U1473A felsic veins were most likely generated by high-degree fractional crystallization of either MORB17,24 or anatectic melts derived from HPM of gabbros46. Fe-Ti oxides appear on the liquidus after extensive MORB fractional crystallization. Further Fe-Ti oxide fractionation results in rapid SiO2 enrichments in residual melts, thus, producing felsic veins24. Advanced fractional crystallization leads to residual melts rich in REE, consistent with the abundance of REE-rich accessory minerals (Fig. 2; Supplementary Data 1) and high zircon and apatite Yb contents (Figs. 4, 5) in U1473A felsic veins. The protracted fractional crystallization process is also supported by the large variations of zircon and apatite REE contents within each U1473A felsic vein (Figs. 4, 5). The published zircon-apatite pairs for magmatic rocks exhibit a positive correlation between ∆δ18Ozircon-apatite and whole-rock SiO2 contents (Fig. S8). Thus, the positive correlation between ∆δ18OZircon-apatite and apatite Yb contents for Hole U1473A felsic rocks (Fig. 6d) supports the progressive Yb and SiO2 enrichments during magma fractional crystallization42.
Finally, we argue that parental magmas of the U1473A felsic veins are MORB rather than hydrous anatectic melts of gabbros. This is because zircon in felsic melts formed by fractional crystallization of MORBs and anatectic melts exhibit different Yb versus Eu/Eu* correlations (Fig. 8a, b); the MORB differentiates exhibit higher zircon Yb contents at a given Eu/Eu* than the high-degree differentiates of anatectic melts. Such a contrast reflects the more depleted REE abundances in the anatectic melts than those in MORBs, while the Eu/Eu* in zircon are mainly controlled by the extent of Pl fractionation during protracted magma evolution (Fig. S9). The zircon Yb-Eu/Eu* correlations of U1473A felsic veins are consistent with the trends of MORB fractional crystallization (Fig. 8a). This study does not exclude the potential role of liquid immiscibility in the genesis of felsic melts47. Felsic melts formed through liquid immiscibility would rapidly evolve into highly fractionated compositions, akin to those produced by fractional crystallization47. In summary, we propose that the U1473A felsic veins were most likely generated through extensive MORB differentiation.
Detachment fault controlling the felsic melt generation in the Atlantis Bank OCC
Our study on the U1473A felsic rocks suggests that zircon trace elements could be considered as the benchmark for MORB differentiation. Therefore, we use zircon data to decipher the origin of felsic veins from other two drilling holes on the Atlantis Bank OCC. The gabbro sections in the three Atlantis Bank drill holes exhibit different stress regimes (Fig. 1c), reflecting the evolving stress field associated with footwall rotation during detachment faulting17,28,29. Atlantis Bank gabbros captured different stress regimes, probably due to the different distance of individual gabbro intrusions to the detachment fault plane29. Extensional stress dominates the upper portions of footwall, i.e., the upper 640 meters of Hole 735B and the whole Hole 1105 A. In contrast, the 640–1508 mbsf of Hole 735B and 50–809 mbsf of Hole U1473A are compressional stress-dominant, representing the lower portions of the footwall (Fig. 1c). To simplify the discussion, zircon REE data of the Atlantis Bank felsic veins and oxide gabbros are classified to two groups according to their locations, i.e., Group 1 (the upper portions of footwall) and Group 2 (the lower portions of footwall, Figs. 8 and 9).
a,b Zircon Yb-Eu/Eu* plot for the Atlantis Bank felsic veins grouped by their reconstructed locations in detachment footwall. c Schematic cartoons illustrating the interplay among detachment faulting, magma intrusion, and high-temperature hydrothermal circulation during the formation of the Atlantis Bank OCC. Red dashed lines in panel c denote the qualitative isotherms. OxG oxide gabbros, HPM hydrous partial melting, FC fractional crystallization.
Zircon in Group 1 felsic veins, namely, the upper 640 meters of Hole 735B and Hole 1105 A, initially display low and uniform Yb contents along with decreasing Eu/Eu*, followed by a rapid increase in Yb contents at Eu/Eu* <0.2 (Fig. 8b, c); their zircon Yb contents are too low to match the trend of MORB fractional crystallization, but can be successfully reproduced by fractional crystallization of hydrous anatectic melts (Figs. 8c, 9a). Similarly, zircon in Group 1 oxide gabbros exhibit low Yb contents ( < 1000 ppm) with Eu/Eu* decreasing from 0.5 to 0.1, consistent with the trends of HPM of gabbros without protracted fractional crystallization (Fig. 8c). Thus, zircon in oxide gabbros from the upper portions of footwall may have crystallized in situ from the anatectic melts trapped in their protoliths due to incomplete melt extraction (Fig. 8c). In contrast, the Group 2 felsic veins, namely, Hole U1473A and lower part (640–1508 mbsf) of Hole 735B, exhibit indistinguishable zircon Yb-Eu/Eu* correlations that are consistent with an origin of fractional crystallization of MORB (Figs. 8a, 9b). Similar correlations are also observed for zircons in Group 2 oxide gabbros, i.e., a progressive increase in Yb contents (from <1000 to 4000 ppm) with Eu/Eu* decreasing from 0.5 to 0.2, which are consistent with the modeling trend of MORB fractional crystallization (Fig. 8c).
These observations indicate that Group 1 felsic veins were highly fractionated anatectic melts (Fig. 9a, c, d), whereas Group 2 felsic veins and oxide gabbros record an extraordinary high degrees ( > 90%) of fractional crystallization of MORB24 (Fig. 9b, c, e). These findings suggest that the origin of felsic veins is coupled to their distance to the fault plane in Atlantis Bank OCC (Fig. 9c); felsic melts of an anatectic origin are mostly adjacent to the fault plane (Group 1), while the felsic melts of MORB fractional crystallization origin are mostly further away from the fault plane (Group 2, Fig. 1c).
The complex interplay between the detachment faulting and magma intrusion is essential to seawater ingress into deep crust and, thus, triggering melting of gabbros. Previous studies on the Atlantis Bank OCC have suggested that the detachment fault was rooted within the melt-present zones of lower oceanic crust28,48. Amphibolite veins in the three drilling holes suggest the occurrence of high-temperature (700–850 oC) alteration by seawater ingress (Fig. 4d)36. Phase equilibria modeling indicates that ~0.4–2.0 % H2O in system is sufficient to induce hydrous melting of gabbros at temperatures exceeding the water-present solidus of gabbros (750–900oC; Fig. 7). Gabbros with >0.4 wt.% H2O are ubiquitous throughout the Holes 735B and U1473A49 but textures indicative of HPM are uncommon14,17. This suggests that elevated H2O contents of gabbros alone is not enough to induce hydrous melting. Atlantis Bank OCC exhibits striking Fe-Ti oxide enrichments near the detachment fault plane, reflecting the Fe-Ti melt injection into shear zones driven by the deformation28,50,51. Injection of such Fe-Ti melts with temperatures of ~920–1000 oC52 near the fault plane provided excess heat for the HPM of gabbros. In addition, the reaction of gabbros with Fe-Ti melts increased their FeO contents and, thus, decreased their solidus by 50–80 oC (Fig. 7), further facilitating HPM.
These observations highlight the essential role of detachment faulting in controlling felsic melt generation at Atlantis Bank OCC. Because the HPM of gabbros requires the seawater ingress at high temperatures of 750–900 oC (Fig. 7), the deep rooting of detachment fault and hydrothermal system into the melt-present zones of lower oceanic crust likely occurred during the formation of Atlantis Bank OCC (Fig. 9c−e). Despite distinct formation mechanisms, no obvious differences in lithological and structural features exist between Group 1 and Group 2 felsic veins17,53,54,55 (Fig. S10). Hence, these characteristics cannot be used to assess the potential involvement of HPM during their formation. Instead, these features of felsic veins reflect prolonged processes of melt differentiation, emplacement, and deformation. Specifically, the lithological characteristics of felsic veins primarily indicate the extent of fractional crystallization, while their structural features record the most recent deformation experienced during their formation. Additionally, syn-magmatic deformation might amplify the extent of fractional crystallization of felsic melts, as indicated by the negative correlation between mean Ti-in-zircon temperatures and the crystal plastic fabric intensity in the Hole U1473A felsic veins (Fig. S11).
Generation of felsic melts during formation of OCCs
Except for the nearly amagmatic OCCs with almost exclusive mantle rock exhumation in the SWIR at 61 − 67°E18,19, the formation of many OCCs are accompanied with variable magma intrusions8,20. To further assess the interplay between detachment faulting and magma intrusion during the formation of OCC, zircon REE data were applied to examine the origin of felsic veins from two additional OCCs along the Mid-Atlantic Ridge: the Atlantis Massif and 15o20’N.
The IODP Hole U1309D in the Atlantis Massif OCC, formed through detachment faulting under moderate magma supply (Fig. 10a)56, is composed of gabbros veined by minor felsic rocks. Zircon in most Atlantis Massif felsic rocks display rapid increase in Yb contents with decreasing Eu/Eu*, consistent with the trends of MORB fractional crystallization (Fig. 10a). The Hole U1309D gabbros were weakly deformed and were subjected to greenschist-facies alteration, in contrast to the extensive deformation and high amphibolite-facies alterations seen in Atlantis Bank OCC57. The localized low amphibolite-facies alterations in the gabbro bodies of the Atlantis Massif indicate seawater influx at temperatures of 450–750 oC58, below the water-present gabbro solidus ( > 750oC, Fig. 7). These characteristics indicate that the detachment fault localized along the margin of gabbros that have been sufficiently cooled57. Once solidified, the gabbro bodies would behave as barriers for downward seawater infiltration9. Therefore, the lack of gabbro HPM at Atlantis Massif OCC plausibly reflects that the gabbros intruded relatively far below the fault plane and has been sufficiently cooled ( < 750 oC) prior to interaction with hydrothermal system58 (Fig. 10a).
a The large gabbro pluton below the central dome of Atlantis Massif OCC (30oN of Mid-Atlantic Ridge), b 15o20’ FZ OCC (Mid-Atlantic Ridge), where gabbro intrusions are scattered and only a few meters to one hundred meters thick64. Literature data sources are given in Supplementary Data 12. HPM hydrous partial melting, FC fractional crystallization, MAR Mid-Atlantic Ridge.
The 15o20’N OCCs consist of peridotites with gabbro intrusions of several to less than one hundred meters in thickness59, formed under low magma supply (Fig. 10b). Zircon in the 15o20’N felsic veins show low and uniform Yb contents ( < 1000 ppm) with varying Eu/Eu* (0.2–0.6), consistent with HPM of gabbros without protracted fractional crystallization (Fig. 10b). Modeling results show that zircon-saturated felsic melts cannot be produced through HPM of olivine gabbros (Fig. 8a), but can be generated through melting of an Al-rich oxide gabbro at temperature of 750–830oC (Figs. 7d, 8a). A distinctive feature of the 15o20’ OCC is its enrichment in oxide gabbros, suggesting that most melts were trapped within the cold lithosphere rather than being erupted59. Such iron-rich gabbros commonly have relatively low solidus temperatures <800oC (Fig. 7c, d) and are easier to melt than other lithologies (such as olivine gabbros). Both peridotites and gabbros at 15o20’N OCCs contain high-temperature (i.e., 850–950 oC) mylonitic zones60,61, suggesting the juxtaposition of the detachment fault and magma intrusions near the brittle-ductile transition. Thus, the interaction of the gabbros trapped within the peridotites with the seawater released via the deep-rooted faults might have triggered the HPM of gabbros during the accretion of 15o20’N OCCs60.
Collectively, our studies offer insights into the petrogenesis of felsic melts at slow- and ultraslow-spreading ridges, highlighting the crucial role of detachment faulting in inducing hydrous partial melting of the lower oceanic crust at elevated temperatures ( > 750 oC, Fig. 7). The sensitivity of felsic melt formation to water activities and temperatures within magmatic systems offers a window into the complex processes involved in the formation of OCCs. Furthermore, the results indicate that the interplay among faulting, magmatism, and hydrothermal circulation manifest distinctly across various OCCs (Figs. 9 and 10). Notably, the formation modes of nearly amagmatic OCCs in the SWIR at 61 − 67°E18,19 differ from those of the more magmatic OCCs examined in this study (Fig. 10); the former likely resulted from interactions between ultramafic footwalls and melts/fluids at considerable depths ( ~ 15 km)19,62. The complex interactions associated with detachment, combined with varying magma supplies, could lead to diverse chemical and thermal exchanges between the oceanic lithosphere and seawater at different ocean ridges3,11,13,49,63,64. If correct, the construction modes and chemical and heat budgets of slow- and ultraslow-spreading oceanic lithosphere could be more complicated than previously anticipated1,2,27,64.
Methods
Mineral major and trace element analysis
The major elements of amphibole, plagioclase, and apatite were determined using a JEOL Superprobe (JXA 8100) electron microprobe in the Institute of Geology and Geophysics, Chinese Academy of Sciences (IGGCAS). The measurements were conducted with an accelerating voltage of 15 keV, a beam current of 10 nA and beam size of 3 μm for amphibole and plagioclase and 15μm for apatite. The elements analyzed in amphibole and plagioclase included Si, Ti, Al, Cr, Fe, Mn, Mg, Ca, Ni, Na, and K, while additional elements P, S, Cl, and F were measured in apatite.
Trace element abundances in zircon and apatite were determined by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS), using an Element XR HR-ICP-MS instrument (Thermo Fisher Scientific, USA) coupled with a 193 nm ArF excimer laser system (Geolas HD, Lambda Physik, Göttingen, Germany) at IGGCAS. The methodology followed the well-developed procedures65, employing a laser diameter of 32 μm, 5 Hz repetition rate, and 4 J/cm2 laser energy density. Helium was employed as the ablation gas to improve ablated aerosols’ transport efficiency. International standard SRM NIST 610 glass was used for external calibration66, and SRM NIST 612 glass was used for quality control monitoring. Silicon (29Si) and calcium (43Ca) were used as internal standards for zircons and apatites, respectively. Mineral major and trace element data for the studied IODP Hole U1473A felsic and gabbroic rocks are given in Supplementary Data 2–6.
Zircon and apatite oxygen isotope analysis
Zircon and apatite δ18O values were analyzed using a Cameca Secondary Ion Mass Spectroscopy (SIMS)-1280 at IGGCAS. Zircon and apatite grains separated from each felsic rock sample were embedded in epoxy mounts together with standards for oxygen isotope and U-Pb dating analysis. Specifically, the Penglai and Qinghu zircon standards are selected for zircon oxygen isotope analysis, and the Durango and Qinghu apatite standards are utilized for apatite oxygen isotope analysis. The in-situ apatite67 and zircon68 oxygen isotope analytical procedures have been previously reported. The Cs+ primary ion beam was accelerated at 10 kV with an intensity of 2 nA and a beam size of 10 μm. The 16O and 18O ion intensities were measured simultaneously in multi-collection mode using two off-axis Faraday cups. The δ18O values were normalized to Vienna Standard Mean Ocean Water (V-SMOW) values and corrected for the instrumental mass fractionation factor (IMF). The relative 2σ errors for the δ18O analyzes include both external and internal precision. The IMF was determined using the Durango apatite as a reference with a δ18O value of 9.4‰69 for apatite and the Penglai zircon as a reference with a δ18O value of 5.3‰ for zircon67. During the measurements of unknown samples, Qinghu zircons and apatites were analyzed as external standards, and yield mean values of 5.4 ± 0.2 ‰ for zircon (n = 55, 1σ) and 5.1 ± 0.3 ‰ for apatite (n = 41, 1σ), consistent with the recommended δ18O values of 5.4 ± 0.2 ‰ for zircon67 and 5.6 ± 0.2 ‰ for apatite68. The external reproducibility of 16O/18O ratios for repeated standard measurement was generally better than 0.3 ‰. Zircon and apatite oxygen isotope data are provided in Supplementary Data 7–8.
Zircon U-Pb dating
Before conducting U-Pb dating, the internal structure of zircon grains was examined using a Gatan Mono CL4 cathodoluminescence (CL) spectrometer equipped with a Nova Nano SEM 450 scanning microscope, with a 45-second scanning time at 10 kV and 120 nA at IGGCAS. Zircon U-Pb isotopic measurements were performed using a Cameca IMS-1280 secondary ion mass spectrometry (SIMS) at IGGCAS, following the previously reported procedure70. Pb/U calibration was conducted relative to the Plešovice zircon standard71. The measured 206Pb/238U error is generally around 1% (1σ). Measured compositions were corrected for common Pb using non-radiogenic 204Pb. Uncertainty on individual analysis is reported at a 1σ level; mean ages for pooled U/Pb analyzes are quoted at a 95% confidence interval. Zircon standard Qinghu was simultaneously analyzed with the studied samples to monitor the external uncertainties of SIMS U-Pb measurements. Data reduction was performed using the Isoplot/Ex V.4.0 program72. Zircon U-Pb dating results are provided in Supplementary Data 9.
Zircon Hf isotope and apatite Nd isotope analysis
In-situ zircon Hf and apatite Nd isotopic analyzes were conducted on a Thermo Fisher Neptune Plus multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS), coupled to a Coherent COMPex Pro 193 nm ArF Excimer laser-ablation system. Hf isotopic analysis was performed on the same zircon grain that was previously analyzed for SIMS U-Pb dating and oxygen isotopes73. Ablation time was about 26 s for each analysis with a beam diameter of 44–60 μm, depending on the sizes of the analyzed grains, and 8 Hz repetition rate at 20 J/ cm2. Zircon standards Plešovice were used for external calibration, yielding 176Hf/177Hf ratios of 0.282467 ± 14 (2σ, n = 64), which are in good agreement with those reported in the literature71. Nd isotopic analysis was performed on the same apatite grains previously analyzed for major elements, trace elements, and oxygen isotopes, following the reported procedure74. A beam size of 60–120 μm, and 8 Hz repetition rate with an energy density of 20 J/cm2 were employed. During the analyzes, the isobaric interference of 144Sm on the 144Nd signal was evaluated and corrected75. Repeated analyses of apatite standard AP1 and AP2 yield 143Nd/144Nd of 0.511336 ± 11 (1σ) and 0.510999 ± 89 (1σ) and 147Sm/144Nd of 0.0810 ± 6 (1σ) and 0.0755 ± 14 (1σ), respectively, consistent with recommended values74. Zircon Hf and apatite Nd isotope data are given in Supplementary Data 10 and Supplementary Data 11, respectively.
Phase equilibria and trace element modeling of hydrous partial melting (HPM)
Phase equilibrium (temperature-XH2O pseudosection) diagrams were conducted using GeoPS software41 with the internally consistent thermodynamic dataset ds6276 and the Na2O-CaO-K2O-FeO-MgO-Al2O3-SiO2-H2O-TiO2 (NCKFMASHTO) chemical system. The corresponding a-x models are clinoamphibole, augite, and metabasite melt77, orthopyroxene78, plagioclase and K-feldspar79, ilmenite80, and magnetite81. Quartz and rutile are treated as pure phases. The bulk compositions of four representative gabbros with various Fe2O3 and Al2O3 contents are given in Supplementary Data 13. The previous study has estimated the pressure of ~2–3 kbar and temperature of 820–900oC for the development of granulite-facies deformation in 631 mbsf of Hole U1473A48 and the thickness of layers that were eroded after exhumation at site U1473 is less than several hundred meters27. Based on these observations, the pressure for the gabbros in the three drilling holes (735B, 1105 A, U1473A) cooling to <900oC would vary between 1.5 and 2.5 kbar, depending on the depths of individual gabbro intrusion. Thus, a pressure of 2 kbar was utilized for phase equilibria modelings of HPM.
Trace element modeling of HPM was conducted using the batch melting equation Cliquid=Csource/[D + F×(1-D)], where Cliquid and Csource represent trace element contents in resultant liquid and source, respectively; D is the bulk partition coefficient calculated based on mineral partition coefficients (Kd) and corresponding residual phase modes (Xi); F is the degree of melting. F and Xi at varying temperatures are obtained by phase equilibrium modelings. Modes of apatite and zircon in the residues are computed using the solubility expressions for zircon82 and apatite83. To correct the temperature effect on zircon-melt REE partitioning, we calculated the zircon-melt REE partition coefficients utilizing the Kd-temperature correlations for zircons84. Input parameters and modeling results are given in Supplementary Data 13.
Trace element modeling of fractional crystallization of anatectic melts
Rayleigh fractional crystallization equation [Cliquid = C0/F(D-1)] was used to conduct the trace element modeling of fractional crystallization of anatectic melts, where C0, F, and D represent initial melt compositions, residual liquid fraction, and bulk partition coefficients, respectively. The weight-averaged compositions of partial melts of HPM of each protolith were selected as the initial melt compositions. The fractionating phase fractions were determined using phase equilibrium modelings. The behavior of zircon and apatite was modeled using the solubility expressions for zircon82 and apatite83 and applying the averaged Atlantis Bank felsic vein bulk compositions24,49,54 as the melt compositions. Since zircons in felsic veins from normal- and reverse-sense dominant zones of Atlantis Bank OCC show indistinguishable Ti-in-zircon temperature values (mean values of ~800 oC), temperature in the modeling was set to be 800oC (Fig. S12). Zircon-melt REE partition coefficients were calculated utilizing the zircon Kd-temperature correlations84. Input parameters and modeling results are given in Supplementary Data 14.
Trace element modeling of fractional crystallization of MORB
Rayleigh fractional crystallization equation [Cliquid = C0/F(D-1)] was used to conduct the trace element modeling of fractional crystallization of MORB magmas, where C0, F and D represent initial melt compositions, residual liquid fraction and bulk partition coefficients, respectively. Two representative Atlantis Bank MORB magmas (MgO of 5.3 and 13.3 wt.%, respectively) are selected as the initial melts. Experimental results of low-pressure MORB crystallization42,43 were utilized to reconstruct the fractionating phase fractions and the melt major element compositions at different temperatures. Modes of apatite and zircon appearing in liquidus are computed using the solubility expressions for zircon82 and apatite83. For temperature <900oC, the bulk analysis on representative U1473A felsic veins54 were selected to represent the melt major element compositions, which was utilized to calculate the apatite and zircon solubility. Zircon-melt REE partition coefficients were calculated utilizing the zircon Kd-temperature correlations84. Input parameters and modeling results are given in Supplementary Data 15.
Effect of accessory mineral fractionation and oxygen fugacity on zircon Yb-Eu/Eu* correlations
Because zircon and apatite are the major accessory minerals in Atlantis Bank felsic veins, their effects on the REE compositions of modeled zircons have been considered in the petrochemical modelings. Overall, the fractionation of zircon and apatite has a negligible effect on the modeled zircon Yb-Eu/Eu* trends, because of their very low fractions in the crystallizing phases. Titanite fractionation would lead to increase in both Eu/Eu* and (Yb/Gd)N ratios in zircons85. However, no correlations between Eu/Eu* and (Yb/Gd)N in Atlantis Bank felsic vein zircons are present (Fig. S13), implying limited titanite fractionation during zircon crystallization. Monazite and xenotime are only observed in several samples in Hole U1473A and occur as replacive rims of apatite and zircon, respectively, indicating they crystallized after apatite and zircon. Therefore, the effects of monazite and xenotime on zircon compositions are expected to be limited. Besides, the zircons in the felsic veins with monazite and xenotime show no difference from others.
Oxygen fugacity of felsic magmas is mostly controlled by processes rather than source86. At low pressure, due to the fractionation of Fe-Ti oxides, the felsic melts after basalt fractional crystallization typically have ∆QFM of -186. During hydrous partial melting of gabbros, the residues typically contain Fe-Ti oxides87 (Fig. 7). Thus, Fe-Ti oxide fractionation is prominent in the felsic melt formation at Atlantis Bank core complex. In this scenario, the felsic melts are expected to be reduced and have limited variations in oxygen fugacity. Then, the effect of oxygen fugacity is not prominent in the zircon REE compositions. Overall, we suggest the effect of accessory mineral fractionation and oxygen fugacity on zircon Yb-Eu/Eu* correlations are both negligible. The zircon Yb-Eu/Eu* correlations for felsic veins reflect the melt evolution and initial melt compositions.
Data availability
All Supplementary data (Supplementary Data S1−16) can be found in the on-line open-access repository (figshare)88 at https://doi.org/10.6084/m9.figshare.25911742.v2.
Code availability
The software (GeoPS) used to generate the phase diagrams presented herein can be downloaded at http://www.geops.org/en-us/.
References
Merdith, A. S., Atkins, S. E. & Tetley, M. G. Tectonic Controls on Carbon and Serpentinite Storage in Subducted Upper Oceanic Lithosphere for the Past 320 Ma. Frontiers in Earth Science 7, 1–23 (2019).
Merdith, A. S. et al. Pulsated Global Hydrogen and Methane Flux at Mid-Ocean Ridges Driven by Pangea Breakup. Geochemistry, Geophysics, Geosystems 21, e2019GC008869 (2020).
Früh-Green, G. L. et al. Diversity of magmatism, hydrothermal processes and microbial interactions at mid-ocean ridges. Nature Reviews Earth & Environment, (2022).
Dick, H. J. B., Lin, J. & Schouten, H. An ultraslow-spreading class of ocean ridge. Nature 426, 405–412 (2003).
Tao, C. et al. Sulfide metallogenic model for the ultraslow-spreading Southwest Indian Ridge. Science China Earth Sciences 66, 1212–1230 (2023).
Buck, W. R., Lavier, L. L. & Poliakov, A. N. B. Modes of faulting at mid-ocean ridges. Nature 434, 719–723 (2005).
Cannat, Mathilde Emplacement of mantle rocks in the seafloor at mid-ocean ridges. Journal of Geophysical Research Atmospheres 98, 4163–4172 (1993).
Tucholke, B. E., Behn, M. D., Buck, W. R. & Lin, J. Role of melt supply in oceanic detachment faulting and formation of megamullions. Geology 36, 455–458 (2008).
Jian H., Canales J. P., Dunn R., Nedimović M. R. Hydrothermal flow and serpentinization in oceanic core complexes controlled by mafic intrusions. Nature Geoscience, (2024).
Escartín, J. et al. Central role of detachment faults in accretion of slow-spreading oceanic lithosphere. Nature 455, 790–794 (2008).
Tao, C. et al. Deep high-temperature hydrothermal circulation in a detachment faulting system on the ultra-slow spreading ridge. Nature Communications 11, (2020).
Cannat, M., Mével, C. & Stakes, D. Stretching of the deep crust at the slow-spreading Southwest Indian Ridge. Tectonophysics 190, 73–94 (1991).
Coogan, L. A., Jenkin, G. R. T. & Wilson, R. N. Contrasting Cooling Rates in the Lower Oceanic Crust at Fast- and Slow-spreading Ridges Revealed by Geospeedometry. Journal of Petrology 48, 2211–2231 (2007).
Koepke, J., Feig, S. T., Snow, J. & Freise, M. Petrogenesis of oceanic plagiogranites by partial melting of gabbros: an experimental study. Contributions to Mineralogy and Petrology 146, 414–432 (2004).
Koepke J., Feig S., Snow J. Late stage magmatic evolution of oceanic gabbros as a result of hydrous partial melting: Evidence from the Ocean Drilling Program (ODP) Leg 153 drilling at the Mid-Atlantic Ridge. Geochemistry Geophysics Geosystems 6, (2005).
Pietranik A., Storey C., Koepke J., Lasalle S. Zircon record of fractionation, hydrous partial melting and thermal gradients at different depths in oceanic crust (ODP Site 735B, South-West Indian Ocean). Contributions to Mineralogy and Petrology 172, 10 (2017).
Ma, Q., Dick, H. J. B., Urann, B. & Zhou, H. Silica-Rich Vein Formation in an Evolving Stress Field, Atlantis Bank Oceanic Core Complex. Geochemistry, Geophysics, Geosystems 21, e2019GC008795 (2020).
Cannat, M. et al. On spreading modes and magma supply at slow and ultraslow mid-ocean ridges. Earth and Planetary Science Letters 519, 223–233 (2019).
Paquet M., Cannat M., Brunelli D., Hamelin C., Humler E. Effect of melt/mantle interactions on MORB chemistry at the easternmost Southwest Indian Ridge (61 to 67°E). Geochemistry Geophysics Geosystems 17, (2016).
Olive, J.-A., Behn, M. D. & Tucholke, B. E. The structure of oceanic core complexes controlled by the depth distribution of magma emplacement. Nature Geoscience 3, 491–495 (2010).
Van Dover, C. L., German, C. R., Speer, K. G., Parson, L. M. & Vrijenhoek, R. C. Evolution and Biogeography of Deep-Sea Vent and Seep Invertebrates. Science 295, 1253–1257 (2002).
Koepke, J., Berndt, J., Feig, S. T. & Holtz, F. The formation of SiO2-rich melts within the deep oceanic crust by hydrous partial melting of gabbros. Contributions to Mineralogy and Petrology 153, 67–84 (2007).
Zhang, C., Koepke, J., France, L. & Godard, M. Felsic Plutonic Rocks from IODP Hole 1256D, Eastern Pacific: Implications for the Nature of the Axial Melt Lens at Fast-Spreading Mid-Ocean Ridges. Journal of Petrology 58, 1535–1565 (2017).
Chen, Y. et al. Petrogenesis of ODP Hole 735B (Leg 176) Oceanic Plagiogranite: Partial Melting of Gabbros or Advanced Extent of Fractional Crystallization? Geochemistry, Geophysics, Geosystems 20, 2717–2732 (2019).
Dixon, S. & Rutherford, M. J. Plagiogranites as late-stage immiscible liquids in ophiolite and mid-ocean ridge suites: An experimental study. Earth and Planetary Science Letters 45, 45–60 (1979).
Baines, A. G., Cheadle, M. J., John, B. E. & Schwartz, J. J. The rate of oceanic detachment faulting at Atlantis Bank, SW Indian Ridge. Earth Planetary Science Letters 273, 105–114 (2008).
Dick, H. J. B. et al. Dynamic Accretion Beneath a Slow-Spreading Ridge Segment: IODP Hole 1473A and the Atlantis Bank Oceanic Core Complex. Journal of Geophysical Research: Solid Earth 124, 12631–12659 (2019).
Dick, H. J. B., Kvassnes, A. J. S., Robinson, P. T., MacLeod, C. J. & Kinoshita, H. The Atlantis Bank Gabbro Massif, Southwest Indian Ridge. Progress in Earth and Planetary Science 6, 64 (2019).
Parnell-Turner, R. et al. Oceanic detachment faults generate compression in extension. Geology 45, 923–926 (2017).
Ridolfi, F., Renzulli, A. & Puerini, M. Stability and chemical equilibrium of amphibole in calc-alkaline magmas: an overview, new thermobarometric formulations and application to subduction-related volcanoes. Contributions to Mineralogy and Petrology 160, 45–66 (2010).
Holland, T. & Blundy, J. Non-ideal interactions in calcic amphiboles and their bearing on amphibole-plagioclase thermometry. Contributions to Mineralogy and Petrology 116, 433–447 (1994).
Sun, S. S. & McDonough, W. F. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Geological Society London Special Publications 42, 313–345 (1989).
Watson, E. B., Wark, D. A. & Thomas, J. B. Crystallization thermometers for zircon and rutile. Contributions to Mineralogy and Petrology 151, 413 (2006).
Ferry, J. M. & Watson, E. B. New thermodynamic models and revised calibrations for the Ti-in-zircon and Zr-in-rutile thermometers. Contributions to Mineralogy and Petrology 154, 429–437 (2007).
Grimes, C. B. et al. On the occurrence, trace element geochemistry, and crystallization history of zircon from in situ ocean lithosphere. Contributions to Mineralogy and Petrology 158, 757 (2009).
Sanfilippo A., Tribuzio R., Antonicelli M., Zanetti A. Amphibole and felsic veins from the gabbroic oceanic core complex of Atlantis Bank (Southwest Indian Ridge, IODP Hole U1473A): When the fluids meets the melts. In: AGU Fall Meeting Abstracts) (2017).
Lissenberg, C. J., Rioux, M., Shimizu, N., Bowring, S. A. & Mével, C. Zircon Dating of Oceanic Crustal Accretion. Science 323, 1048–1050 (2009).
Lissenberg, C. J., Condon, D. J., Smye, A. J. & Anma, R. Crystallization of Superfast-Spreading Oceanic Crust (ODP Hole 1256D, Pacific Ocean): Constraints From Zircon Geochronology. Geochemistry, Geophysics, Geosystems 24, e2023GC010964 (2023).
Zhang, C., Koepke, J., Albrecht, M., Horn, I. & Holtz, F. Apatite in the dike-gabbro transition zone of mid-ocean ridge: Evidence for brine assimilation by axial melt lens. American Mineralogist 102, 558–570 (2017).
Valley, J. W. Oxygen Isotopes in Zircon. Reviews in Mineralogy and Geochemistry 53, 343–385 (2003).
Xiang, H. & Connolly, J. A. D. GeoPS: An interactive visual computing tool for thermodynamic modelling of phase equilibria. Journal of Metamorphic Geology 40, 243–255 (2022).
Brophy, J. G. La–SiO2 and Yb–SiO2 systematics in mid-ocean ridge magmas: implications for the origin of oceanic plagiogranite. Contributions to Mineralogy and Petrology 158, 99 (2009).
Berndt, J., Koepke, J. & Holtz, F. An Experimental Investigation of the Influence of Water and Oxygen Fugacity on Differentiation of MORB at 200 MPa. Journal of Petrology 46, 135–167 (2005).
Grimes, C. B., Ushikubo, T., John, B. E. & Valley, J. W. Uniformly mantle-like δ18O in zircons from oceanic plagiogranites and gabbros. Contributions to Mineralogy and Petrology 161, 13–33 (2011).
Schwartz, J. J. et al. Dissolution–reprecipitation of igneous zircon in mid-ocean ridge gabbro, Atlantis Bank, Southwest Indian Ridge. Chemical Geology 274, 68–81 (2010).
Sun, Z.-M. et al. Diversity of Felsic Rocks in Oceanic Crust: Implications From the Neoproterozoic Plagiogranites Within the Northeast Jiangxi Ophiolite, Southern China. Journal of Geophysical Research: Solid Earth 125, e2019JB017414 (2020).
Brunelli, D. et al. Origin of oceanic ferrodiorites by injection of nelsonitic melts in gabbros at the Vema Lithospheric Section, Mid Atlantic Ridge. Lithos 368-369, 105589 (2020).
Taufner R., Viegas G., Faleiros F. M., Castellan P., Silva R. Deformation mechanisms of granulite-facies mafic shear zones from hole U1473A, Atlantis Bank, Southwest Indian Ridge (IODP Expedition 360). Journal of Structural Geology, 104380 (2021).
Kendrick, M. A. Halogens in Atlantis Bank gabbros, SW Indian Ridge: Implications for styles of seafloor alteration. Earth and Planetary Science Letters 514, 96–107 (2019).
Zhang, W.-Q., Liu, C.-Z., Dick, H. J. B., Mitchell, R. N. & Liu, B.-D. Oceanic mantle beneath ultraslow spreading ridges metasomatized by variably evolved melts. Contributions to Mineralogy and Petrology 179, 15 (2024).
Zhang W.-Q., Dick H. J. B., Liu C.-Z., Lin Y.-Z., Angeloni L. M. MORB Melt Transport through Atlantis Bank Oceanic Batholith (SW Indian Ridge). Journal of Petrology 62, (2021).
Ferrando, C. et al. Brown Amphibole as Tracer of Tectono-Magmatic Evolution of the Atlantis Bank Oceanic Core Complex (IODP Hole U1473A). Journal of Petrology 63, (2022).
Dick, H. J., Natland, J. H., Miller, D. J. & aSS, Party SITE 735. Proceedings of the Ocean Drilling Program Scientific Results 176, 1–314 (1999).
MacLeod, C. J. et al. Site U1473. In MacLeod, CJ, Dick, HJB, Blum, P, and the Expedition 360 Scientists, Southwest Indian Ridge Lower Crust and Moho Proceedings of the International Ocean Discovery Program, 360: College Station, TX (International Ocean Discovery Program), (2017).
Nguyen, D. et al. Occurrence of Felsic Rocks in Oceanic Gabbros from IODP Hole U1473A: Implications for Evolved Melt Migration in the Lower Oceanic Crust. Minerals 8, 583 (2018).
Harding, A. J., Arnulf, A. F. & Blackman, D. K. Velocity structure near IODP Hole U1309D, Atlantis Massif, from waveform inversion of streamer data and borehole measurements. Geochemistry, Geophysics, Geosystems 17, 1990–2014 (2016).
Ildefonse, B. et al. Oceanic core complexes and crustal accretion at slow-spreading ridges. Geology 35, 623–626 (2007).
Nozaka, T. & Fryer, P. Alteration of the Oceanic Lower Crust at a Slow-spreading Axis: Insight from Vein-related Zoned Halos in Olivine Gabbro from Atlantis Massif, Mid-Atlantic Ridge. Journal of Petrology 52, 643–664 (2011).
Kelemen P., Kikawa E., Miller D., others A. Drilling Mantle Peridotite Along the Mid-Atlantic Ridge from 14 Degrees to 16 Degrees N. In: Proceedings of the Ocean Drilling Program. Initial Reports). Ocean Drilling Program (2004).
Jöns, N., Bach, W. & Schroeder, T. Formation and alteration of plagiogranites in an ultramafic-hosted detachment fault at the Mid-Atlantic Ridge (ODP Leg 209). Contributions to Mineralogy and Petrology 157, 625 (2008).
Albers, E, Schroeder, T & Bach, W Melt Impregnation of Mantle Peridotite Facilitates High-Temperature Hydration and Mechanical Weakening: Implications for Oceanic Detachment Faults. Geochemistry, Geophysics, Geosystems 20, 84–108 (2019).
Bickert, M., Cannat, M. & Brunelli, D. Hydrous fluids down to the semi-brittle root zone of detachment faults in nearly amagmatic ultra-slow spreading ridges. Lithos 442-443, 107084 (2023).
Kendrick, M. A., Marks, M. A. W. & Godard, M. Halogens in serpentinised-troctolites from the Atlantis Massif: implications for alteration and global volatile cycling. Contributions to Mineralogy and Petrology 177, 110 (2022).
Grimes C. B., Cheadle M. J., John B. E., Reiners P. W., Wooden J. L. Cooling rates and the depth of detachment faulting at oceanic core complexes: Evidence from zircon Pb/U and (U-Th)/He ages. Geochemistry, Geophysics, Geosystems 12, (2011).
Wu, S.-T., Huang, C., Xie, L.-W., Yang, Y.-H. & Yang, J.-H. Iolite Based Bulk Normalization as 100% (m/m) Quantification Strategy for Reduction of Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry Transient Signal. Chinese Journal of Analytical Chemistry 46, 1628–1636 (2018).
Pearce, N. J. G. et al. A Compilation of New and Published Major and Trace Element Data for NIST SRM 610 and NIST SRM 612 Glass Reference Materials. Geostandards Newsletter 21, 115–144 (1997).
Li, X.-H. et al. Penglai Zircon Megacrysts: A Potential New Working Reference Material for Microbeam Determination of Hf–O Isotopes and U–Pb Age. Geostandards and Geoanalytical Research 34, 117–134 (2010).
Li, Y. et al. Revisiting apatite SIMS oxygen isotope analysis and Qinghu-AP reference material. Chemical Geology 582, 120445 (2021).
Trotter, J. A., Williams, I. S., Barnes, C. R., Lécuyer, C. & Nicoll, R. S. Did Cooling Oceans Trigger Ordovician Biodiversification? Evidence from Conodont Thermometry. Science 321, 550–554 (2008).
Li X.-H., Liu Y., Li Q.-L., Guo C.-H., Chamberlain K. R. Precise determination of Phanerozoic zircon Pb/Pb age by multicollector SIMS without external standardization. Geochemistry, Geophysics, Geosystems 10, (2009).
Sláma, J. et al. Plešovice zircon — A new natural reference material for U–Pb and Hf isotopic microanalysis. Chemical Geology 249, 1–35 (2008).
Ludwig, K. R. Isoplot 3.00: A geochronological toolkit for Microsoft Excel. Berkeley Geochronology Center Special Publication 4, 70 (2003).
Wu, F.-Y., Yang, Y.-H., Xie, L.-W., Yang, J.-H. & Xu, P. Hf isotopic compositions of the standard zircons and baddeleyites used in U–Pb geochronology. Chemical Geology 234, 105–126 (2006).
Yang, Y.-H. et al. Sr and Nd isotopic compositions of apatite reference materials used in U–Th–Pb geochronology. Chemical Geology 385, 35–55 (2014).
McFarlane, C. R. M. & McCulloch, M. T. Coupling of in-situ Sm–Nd systematics and U–Pb dating of monazite and allanite with applications to crustal evolution studies. Chemical Geology 245, 45–60 (2007).
Holland, T. J. B. & Powell, R. An improved and extended internally consistent thermodynamic dataset for phases of petrological interest, involving a new equation of state for solids. Journal of Metamorphic Geology 29, 333–383 (2011).
Green, E. C. R. et al. Activity–composition relations for the calculation of partial melting equilibria in metabasic rocks. Journal of Metamorphic Geology 34, 845–869 (2016).
White, R. W., Powell, R., Holland, T. J. B., Johnson, T. E. & Green, E. C. R. New mineral activity–composition relations for thermodynamic calculations in metapelitic systems. Journal of Metamorphic Geology 32, 261–286 (2014).
Holland, T. & Powell, R. Activity–composition relations for phases in petrological calculations: an asymmetric multicomponent formulation. Contributions to Mineralogy and Petrology 145, 492–501 (2003).
WhitePowell, Holland & Worley The effect of TiO2 and Fe2O3 on metapelitic assemblages at greenschist and amphibolite facies conditions: mineral equilibria calculations in the system K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–Fe2O3. Journal of Metamorphic Geology 18, 497–511 (2000).
White, R. W. & Powell, R. Melt loss and the preservation of granulite facies mineral assemblages. Journal of Metamorphic Geology 20, 621–632 (2002).
Boehnke, P., Watson, E. B., Trail, D., Harrison, T. M. & Schmitt, A. K. Zircon saturation re-revisited. Chemical Geology 351, 324–334 (2013).
Harrison, T. M. & Watson, E. B. The behavior of apatite during crustal anatexis: Equilibrium and kinetic considerations. Geochimica et Cosmochimica Acta 48, 1467–1477 (1984).
Claiborne, L. L. et al. Zircon as Magma Monitor. In: Microstructural Geochronology) (2018).
Loader, M. A., Wilkinson, J. J. & Armstrong, R. N. The effect of titanite crystallisation on Eu and Ce anomalies in zircon and its implications for the assessment of porphyry Cu deposit fertility. Earth and Planetary Science Letters 472, 107–119 (2017).
Sun, C. & Lee, C.-T. A. Redox evolution of crystallizing magmas with C-H-O-S volatiles and its implications for atmospheric oxygenation. Geochimica et Cosmochimica Acta 338, 302–321 (2022).
Xu, X.-F., Gou, L.-L., Long, X.-P., Zhao, Y.-H. & Zhou, F. Phase Equilibrium and Trace-Element Modeling of the Partial Melting of Basaltic Rocks under Low Pressure: Implications for Plagiogranite Generation. Journal of Petrology 62, 1–29 (2021).
Zhang W.-Q., Liu C.-Z., MacLeod C. J., Lissenberg C. J. Supplementary Data for “The role of detachment faulting in the genesis of oceanic felsic melts” by Zhang et al.). figshare. Dataset (2024).
Zhang, W.-Q., Liu, C.-Z. & Dick, H. J. B. Evidence for Multi-stage Melt Transport in the Lower Ocean Crust: the Atlantis Bank Gabbroic Massif (IODP Hole U1473A, SW Indian Ridge). Journal of Petrology 61, 1–33 (2020).
Acknowledgements
This work was made possible by samples and data from the Integrated Ocean Discovery Program (IODP). We are indebted to the captain and shipboard crew of IODP Expedition 360 for their invaluable support in data acquisition at sea, and to the Scientific Party of IODP Expedition 360 for stimulating discussions during the cruise and post-cruise meeting. We thank Daniele Brunelli and one anonymous reviewer for their constructive comments and suggestions, and Carolina Ortiz Guerrero for editorial handling and helpful comments. This work received funding from the National Natural Science Foundation of China (41673038), National Science Fund for Distinguished Young Scholars (42025201), National Key Research and Development Project of China (2020YFA0714801), and Strategic Priority Research Program of the Chinese Academy of Sciences (XDB42020301).
Author information
Authors and Affiliations
Contributions
Wei-Qi Zhang acquired the data and wrote the manuscript. Chuan-Zhou Liu designed the study, acquired the samples, acquired fundings, and edited the manuscript. Christopher J. MacLeod and C. Johan Lissenberg revised and edited the manuscript. All authors contributed to the interpretation of the data.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Earth & Environment thanks Daniele Brunelli and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Derya Gürer and Carolina Ortiz Guerrero. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Zhang, WQ., Liu, CZ., MacLeod, C.J. et al. The role of detachment faulting in the genesis of oceanic felsic melts. Commun Earth Environ 6, 109 (2025). https://doi.org/10.1038/s43247-025-02098-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s43247-025-02098-3
