Heterogeneous Earth’s mantle drilled at an embryonic ocean | Nature Communications

Nature Communications volume 16, Article number: 2016 (2025) Cite this article

790 Accesses

3 Altmetric

Metrics details

Mantle processes control plate tectonics and exert an influence on biogeochemical cycles. However, the proportion of mantle sampled in-situ is minimal, as it is buried beneath igneous crust and sediments. Here we report the lithological characteristics of two mantle sections from an embryonic ocean drilled by the International Ocean Discovery Program (IODP) in the Tyrrhenian Sea. Contrary to the mantle drilled at Mid Ocean Ridges (MORs) and hyperextended passive margins, our findings reveal exceptionally heterogeneous and fertile mantle lithologies, ranging from fertile lherzolites to depleted harzburgites and dunites, interlayered with pyroxenites. Plagioclase- and clinopyroxene-rich layers, hydrous potassic magmatic veins, and mafic intrusions indicate substantial mantle refertilization and delayed inception of magmatic crust. We propose that magma-poor rifts do not require a chemically depleted mantle, too refractory to melt. Deep lithospheric processes such as mantle refertilization and prolonged lithospheric thinning delayed melt focusing and the formation of a steady-state spreading center.

Since the discovery of large portions of the upper mantle exhumed at slow and ultraslow spreading MORs, scientific ocean exploration programs have aspired to collect in-situ sections of the underlying mantle (Fig. 1a and Table 1). Currently, drilled sections of the oceanic mantle are limited to three locations along the central Mid-Atlantic Ridge - near the Kane Fracture Zone at 23°N (Ocean Drilling Program, ODP Leg 153)1, at the 15°20' N Fracture Zone (ODP Leg 209)2, and recently in the Atlantic Massif oceanic core complex at 30°N (IODP Expedition 399)3. The mid-oceanic mantle section in the Pacific Ocean has only been drilled in the Hess Deep4, a tectonic window along the East Pacific Rise, during ODP Leg 147. The abyssal mantle peridotites recovered at these sites are extensively serpentinized and geochemically depleted due to partial melting of the ascending asthenosphere. Clinopyroxene-poor peridotites, such as harzburgites and dunites, are the most commonly drilled mantle lithologies at these sites and rarely consist of plagioclase-rich varieties, formed in response to grain-scale percolation and crystallization of MOR basaltic (MORB) magmas5. In addition, a significant proportion of these depleted mantle sections consist of MORB-type magmatic intrusions of variable thickness (Table 1).

a 1 – ODP Leg 147 (Hess Deep, East Pacific Rise), 2 – ODP Leg 153 (Kane Fracture Zone, Mid-Atlantic Ridge), 3 – ODP Leg 209 (15°20' Fracture Zone, Mid-Atlantic Ridge), 4 – ODP Leg 173 (Western Iberian margin COT, Atlantic Ocean), 5 – IODP Expedition 399 (Atlantis Massif, Mid-Atlantic Ridge). 6 – IODP Expedition 402 (Tyrrhenian Sea), this study. The location of IODP Expeditions 367-368 in the South China Sea COT is also shown (7). b Bathymetrical map and location of the Expedition 402 sites (U1614 and U1616) in the Vavilov basin (Tyrrhenian Sea). Previous drill sites and results of dredging are also shown (Sites 651, 652, 655, and 656). Colors refer to the nature of the recovered basement lithologies. White lines mark the location of interpreted normal faults.

In addition to MORs, mantle peridotites in the ocean have also been recovered in the distal portion of a passive continental margin, a geological setting that formed during the last phase of continent rifting, which is well distinct from that of a mature ocean ridge. In the late 20th century, the discovery of these exhumed mantle rocks by drilling during ODP Legs 103, 149, 173, and 210 at the conjugate magma-poor Iberia-Newfoundland rifted margins in the Atlantic Ocean6,7, led to the formulation of a new model for the formation of Continent – Ocean Transition (COT). Prior to this, the COT was interpreted as a simple juxtaposition of stretched continental crust and new magmatic oceanic crust resulting from rapid lithospheric breakup and leading to a magmatically robust mid-ocean ridge system. Drilling offshore at the hyper-extended Atlantic-COT challenged this axiom, showing that continental breakup can be followed by exhumation of a chemically depleted sub-continental mantle with minimal or absent magma production, until the uprising asthenosphere produced enough melts and focused melt-extraction to generate a magmatic Penrose-type oceanic crust6,8. Over the past decades, COT has been thereby classified into two end-member models of magma-rich and magma-poor types, characterized by contrasting structural architectures9,10. Despite a global consensus that the magma-poor nature of hyper-extended COTs is caused by a combination of low extension rates and a depleted nature of the lower lithospheric mantle11,12,13,14,15,16,17, the debate is still ongoing due to a lack of direct geologic sampling to test geophysical observations and geodynamic models of these contrasting end-members. The IODP Expeditions 367-368 drilled the intermediate South China Sea (SCS) continental margin and recovered unexpectedly thick MORB-type basaltic sections close to the thinned continental crust, revealing that abundant magmatism was produced during the breakup of a thinned continental lithosphere18. The mechanical behavior of the lithosphere, which is also dependent on its inherited structures, exerts a key influence on the rifting evolution19. Hence, understanding the COT formation requires access to deep lower crustal and mantle rocks to define the link between deep tectono-magmatic processes and the development of a rifted margin.

Continuing the legacy of in-situ mantle drilling in COT, the penultimate IODP Expedition (402) on the R/V JOIDES Resolution was conducted to drill into the oceanic basement of the Vavilov Basin, representing an embryonic oceanic basin located in the central part of the Tyrrhenian Sea (Fig. 1b). Different from the Iberia-Newfoundland rifted margins, the COT exposed in the area never developed a steady-state ocean ridge, and shares geological features with both magmatic and amagmatic COTs. The Tyrrhenian Sea was formed as a result of the northwestward subduction of the Ionian oceanic lithosphere beneath the Calabrian Arc and slab roll-back20,21,22,23. This process stretched a thinned continental lithosphere, which may have included remnants of the Jurassic Alpine Tethys, to create an active back-arc basin. Lithospheric thinning began in the southern part of the Tyrrhenian during the Middle Miocene. With the rate of extension increasing progressively, lithospheric thinning moved southward until it breaks at latitude 41°N, leading to the opening of the Vavilov basin23 (Fig. 1b). Seismic data22,23 and previous drilling expeditions24 concur that the central portion of the Vavilov basin is formed by exhumed mantle peridotites covered by a thin layer of basalts and sediments. Although the basin opened at a full spreading rate ranging between 40 and 60 mm/yr21, which is higher than ultraslow and slow-spreading ridges where mantle exhumation has been largely reported, the absence of a continuous magmatic crust suggests regional igneous starvation. The main objectives of IODP Expedition 402 were to ascertain whether the mantle in this nascent ocean differs from the depleted mantle observed at MORs and at Atlantic-type hyper-extended COTs and to constrain the effect of mantle composition in the transition from continental rifting to ocean formation.

The post-Messinian (< 5 Ma) basement of the Vavilov basin was the focus of scientific drilling during the 1980s and 1990s when DSDP Site 373 and ODP Site 655 recovered exclusively < 150 m thick MORB-type tholeiitic basalts along linear ridges trending N10°E24. Only ODP Site 651 (Leg 107) recovered a 40 m thick mantle section, below 136 m of basalts and basalt breccias, intercalated by a dolerite-albitic intrusion25. The drilled peridotites were extensively serpentinized, and the original mantle fabric showed porphyroclastic textures with relicts of orthopyroxene and spinel in a serpentinized matrix. In contrast, IODP Expedition 402 recovered two continuous sections of moderately to highly serpentinized mantle peridotites (Sites U1614 and U1616) directly beneath a ~ 250 m thick sedimentary cover, with no intervening magmatic crust (Fig. 2). The contact between brecciated serpentinized peridotites and dolomitic sediments is exceptionally preserved (Fig. 3a) and lack a thick zone of cataclasites or tectonic gouges as observed at the top of exhumed mantle in other COT6,7. Ophicarbonates and tectonic breccias were nonetheless recovered at different stratigraphic levels throughout both Expedition 402 sites (Supplementary Figs. 1, 2). The 160 m thick mantle section sampled at Site U1614 is composed of approximately equal proportions of spinel-bearing lherzolite (24%) and harzburgite (26%), with minor dunite (10%), pyroxenites (4%) and mafic intrusions (3%) (Fig. 2a and Supplementary Fig. 1). Plagioclase-bearing lherzolites and harzburgites constitute the rest of the mantle section (27%). The 94 m thick mantle section cored at Site U1616 also recovered heterogeneous lithologies with spinel-bearing harzburgites and lherzolites making up to 50% and 19% of the recovered fraction, respectively (Fig. 2b and Supplementary Fig. 2). Mafic intrusions in both U1614 and U1616 mantle sections are relatively abundant and dispersed throughout the section (Figs. 2c, 3b–d). The different lithologies appear randomly distributed throughout the two sites. However, depleted harzburgites and dunites are prevalent at shallower depths at Site U1614 (< 320 mbsf). The results of Expedition 402 are in good agreement with the refractory nature of the peridotites previously sampled at ODP Site 65126, which were restricted to a depth of 30 m below the sedimentary cover. The thicker mantle sections drilled at Sites U1614 and U1616 now suggests that refractory peridotites (26% at Site U1614 and 50% at Site U1616) are not prevailing lithologies in the Tyrrhenian COT mantle, which is exceptionally heterogeneous and on average mineralogically fertile. The relative proportions of fertile lithologies in the Tyrrhenian mantle are indeed exceptionally higher than mantle peridotites available from back-arc and fore-arc oceanic basins, with a lithological heterogeneity covering the entire spectrum of MOR mantle peridotites (Fig. 4). Notably, the clinopyroxene modal proportions of Hole U1416C peridotites exceed by far that of Hole U1601C that drilled a 1268 m long section of depleted mantle at the Mid-Atlantic Ridge5.

a and (b) show lithostratigraphy of Holes U1614C and U1616E, respectively. c Depth-wise variation in magmatic vein intensity calculated at 2 m intervals for both holes. Magmatic veins group all veins with primary minerals including pyroxene-rich, plagioclase-rich, and gabbroic intrusions. Dip of magmatic veins (d) and mantle ductile fabric (e) relative to the core reference frame. Also indicated are the locations of K-Ti-rich mafic intrusions (pink bars). Depth-wise whole rock (f) Al2O3, and (g) K2O variations in peridotites. The gray bar in Hole U1616E indicates a drilled interval with no coring.

a The sediment-basement interface is defined by a direct contact (white dashed line). Peridotite is highly serpentinized with a weak mantle fabric defined by the alignment of pyroxene minerals. b Clinopyroxene and plagioclase-rich veins in partially serpentinized peridotite at a low angle to the mantle fabric orientation defined by orthopyroxene porphyroclasts. c Weathered plagioclase-bearing harzburgite intruded by a centimeter-scale spinel-websterite vein. The latter is rimmed by serpentine and carbonate veins and fractured by another, orthogonal network of the same veins. d Phlogopite-rich veins hosted in mantle peridotite.

a Relationships between orthopyroxene and clinopyroxene modal abundances and (b) histogram of clinopyroxene modal abundance compared to peridotites from mid-ocean ridges5,37 (gray plain bars), fore-arc54 (red contours) and back-arc55,56 (purple contours).

The recovered dunites likely formed as a response to focused melt migration27,28. This inference is based on the occurrence of large quantities of chrome-spinel organized in subparallel trails, defining a foliation that is geometrically consistent with the harzburgite contacts and mantle fabric dips (Supplementary Figs. 1, 2). In addition, some of the dunites contain plagioclase and diopside-rich veins, reflecting the reactive percolation of melts at lower temperatures29,30. Consistently, the TiO2 contents of the dunite locally reached 0.05 wt%, lower than those of the harzburgites (typically 0.01-0.02 wt%) suggesting chemical equilibration with migrating melts (Fig. 5a). Conversely, the sampled harzburgites rarely contain plagioclase, instead preserving a refractory chemical character (Supplementary Figs. 1, 2). These harzburgites are compositionally more depleted than most of the abyssal peridotites sampled from MORs and the Iberia-Newfoundland COT, approaching the composition of fore-arc peridotites from the Mariana Trench that have undergone hydrous-fluid flux melting forming arc-related magmas (Fig. 5). Although some of the depleted Tyrrhenian harzburgites seem to be similar to fore-arc peridotites, a suprasubduction tectonic setting would not explain the geochemical signature of the basalts recovered in the Vavilov Basin and Vavilov Seamount that range from MORB-like to alkaline24,26. The very low Al2O3, TiO2, K2O, and CaO contents of the Tyrrhenian harzburgites (Fig. 5 and Supplementary Fig. 3), instead, can be explained by high degree (~ 20–25%) fractional melting of a typical depleted mantle peridotite (Fig. 5; see detail of the melting model in figure caption and in Supplementary Data 4). Such a residual character is also revealed by the low incompatible trace element compositions (i.e., mid to heavy-Rare Earth Elements) of pyroxene from the Site 651 peridotites, indicating high degrees of partial melting and melt extraction. The paucity of magmatic features, such as volcanic ridges and seamounts, indicates a subdued melt supply in the Tyrrhenian Sea22, the refractory character could have been partly inherited from ancient melting events. Ancient refractory mantle domains are documented in mantle sections from the Jurassic Alpine ophiolites31 and the subcontinental lithospheric mantle (SCLM) exposed in the Ivrea-Verbano Zone of the Southern Alps32, proving that refractory portions of mantle are entrained in the circum-Tyrrhenian SCLM33. Therefore, although we cannot exclude that some of the collected peridotites melted to some degrees during the rifting stage, the existence of highly depleted harzburgites in the Tyrrhenian mantle may be related to ancient melting events that preceded continental rifting and opening of the basin, as seen in the surrounding subcontinental peridotites.

Covariations in Al2O3 versus (a) TiO2 and (b) K2O contents and comparison between (c) Al2O3 and (d) K2O contents between the Tyrrhenian Sea peridotites and abyssal peridotites from mid-ocean ridges (MOR)57,58, back-arc basin59, Continent-Ocean Transitions (COT) and fore-arc peridotites54. Compositions of orogenic peridotites and mantle xenoliths from the circum-Mediterranean area are compiled from the PetDB database (https://search.earthchem.org/). In (a), the compositions of abyssal peridotites are compared to the Al2O3 versus TiO2 compositions of mantle residues of fractional melting. Residual mantle compositions were thermodynamically calculated using pMELTS60 included in alphaMELTS software package61. Depleted-MORB Mantle (DMM) composition62 (K, P, and Cr were excluded) was chosen as source material. Isentropic near fractional melting of DMM was conducted at pressures ranging from 3 to 1 × 10-4 GPa at 1450 °C. The melting trend is depicted in pink, and values of melting degrees are in pink italic numbers (see values in Supplementary Data 4).

These depleted harzburgites are widespread in the Tyrrhenian COT mantle, where they are associated with significant quantities of mineralogically and chemically fertile peridotites (lherzolites) and pyroxenites. The lherzolites have a porphyroclastic texture with weak foliation defined by the preferential orientation of pyroxene and spinel (Fig. 3b). This coarse-grained, high-temperature protogranular texture is characteristic of asthenospheric flow34. This fabric locally evolves to highly deformed horizons with strong oriented crystal-plastic fabrics and mylonitic textures, likely related to exhumation and formation of low-angle normal faults (Supplementary Figs. 1 and 2). Most of the lherzolite layers show weak plastic deformation and have undeformed plagioclase organized in millimeter-scale trails, often aligned along the main foliation of the pyroxene porphyroclasts and spinels (Fig. 3b), attesting to melt impregnation following the deformation process. The lherzolites extend from refractory compositions to a very fertile character, represented by plagioclase-bearing lherzolites, which preserve covariations in Al2O3 versus MgO, CaO, TiO2, and K2O contents (Fig. 5a, b and Supplementary Fig. 3). Plagioclase crystallizes in response to refertilization or impregnation by migrating melts at lithospheric levels35, and is a well-known process at slow-spreading MORs36, COTs13, and back-arc basins35. However, the Tyrrhenian mantle also shows textural evidence for crystallization of secondary clinopyroxene in veins ranging in thickness from mm- to cm (Figs. 3b, 6), which are instead rare in the sub-ridge mantle37. Clinopyroxene can crystallize from mafic melts at high pressure or due to the interaction of alkaline melts with the host peridotites36,38,39, conditions that can convert initially depleted harzburgites into fertile lherzolites. Given that alkaline melts are widespread from the Vavilov and Magnaghi seamounts (Fig. 1b), both located in close vicinity to the drilled sites40, similar melts might have migrated and interacted with the peridotites since the early phases of exhumation of the mantle sequence. Evidence for early interaction with alkaline melts is indeed documented in ODP Site 651 peridotites, where interstitial clinopyroxene presenting selective enrichments in strongly incompatible elements is locally associated with magmatic amphibole, preserving evidence of high-temperature deformation related to mantle exhumation41. Refertilization by interaction with alkaline melts is also consistent with the slight but distinct K2O enrichment observed in many plagioclase-bearing peridotites (Fig. 5b, d) and is rarely seen in global MOR peridotites. The K2O enrichments are nonetheless also found in mantle peridotites from the Iberian COT and the Philippine Sea back-arc basin (Fig. 5d), suggesting the occurrence of alkaline melts during the refertilization process of the transitional mantle. Hence, although defining the timing of depletion and refertilization requires further investigations, the geometric relationships between deformation and melt-migration features and the geochemical affinity of refertilizing melt leads us to propose that the plagioclase and clinopyroxene-rich veins within the lherzolites were generated by localized events of refertilization during the exhumation of the mantle sequence to the Tyrrhenian seafloor. We cannot exclude, however, that some of the chemical variability observed here could have been acquired during ancient events of refertilization, similar to what has been suggested for fertile SCLM sections exposed in the circum-Mediterranean area42 (Fig. 5). In any case, ancient and recent refertilization processes resulted in a high chemical variability of this embryonic oceanic mantle sequence that encompasses the full range of chemical variability of abyssal peridotites from oceanic tectonic settings in only two drill sites (Figs. 4, 5 and Supplementary Fig. 4).

a Photograph of core U1614C-22R-1A (interval 135/150 cm). b Mantle fabric is usually at a high angle to the core reference frame (> 40°) and here consists in a weak deformation subparallel to pyroxene-rich veins (1). A mafic intrusion (2, in blue) cuts this mantle fabric at a high angle. c Two generations of centimeter-scale serpentine veins (3, in pale green) follow pre-existing lithological contrasts or cut the magmatic vein-peridotite contacts. d Carbonate veins dissect the pre-existing features, overprinting both the initial mantle fabric and the serpentine vein network (4, in yellow).

Concrete evidence supporting the heterogeneous character of the Tyrrhenian mantle is the presence of a minor (~ 4%) yet significant proportion of pyroxene-rich lithologies, such as websterites, orthopyroxenites, and clinopyroxenites. These fertile and readily fusible lithologies, grouped as ‘pyroxenites’, are recovered throughout both mantle sections, where they have thicknesses ranging from 3 cm to as high as 10 cm (Fig. 3c). The best-preserved pyroxenite occurrences are the subparallel layers of spinel websterites in plagioclase-bearing harzburgites recovered at ~ 230 mbsf at Site U1614 (Fig. 3c and Supplementary Fig. 1). Unlike the spinel trails in plagioclase-peridotites, the coarse pyroxene porphyroclasts composing the pyroxenites veins are weakly deformed, showing undulose extinction, which is consistent with the overall weak deformation of the host peridotite. The websterites contain abundant ameboid-shaped mm-sized Cr-spinels. Plagioclase rims around spinel within pyroxene-rich domains attest to melt impregnation, or sub-solidus re-equilibration from spinel to plagioclase-facies mantle conditions43. The subtle enrichments in CaO, Al2O3, and TiO2 at constant Mg# distinguish these chemical fertile mafic lithologies from most of the peridotites (Fig. 5a, b and Supplementary Fig. 3). In orogenic massifs, pyroxenites are common lithologies, and appear stretched, boudinaged, and intermingled with the host SCLM peridotites. These pyroxenites are either considered to be formed as a result of recycling crustal material44, or from high-pressure segregation of melts with remarkable crustal components45. At MORs, pyroxenite layers are rare and commonly interpreted as deep segregations of melt contemporary to the MORB magmatism46. None of these hypotheses can be discounted here, but, in all cases, the coexistence of these fertile lithologies with highly refractory peridotites and lherzolites attest that a compositional heterogeneity was attained during events of refertilization processes following the last melting event.

Most of the Tyrrhenian samples have retained a weak, porphyroclastic texture, typical of the high temperature, and low stress conditions expected in the asthenospheric mantle or the deep lithosphere47. Peridotites with higher degrees of deformation, such as protomylonites to mylonites, are mostly associated with mafic veins that intrude the exhumed peridotites and rarely reported in the depleted lithologies (Supplementary Figs. 1, 2). In the refertilized peridotites, this high-temperature fabric is sub-parallel to the plagioclase and clinopyroxene-rich trails formed as response to the refertilization process (Figs. 2, 3, 6a, b). The prevalence of this high-temperature low-stress deformation in refertilized lherzolites and of mafic veins is a fundamental observation, as it suggests that detachment faults exhuming these mantle rocks likely root into a domain in which the mantle was weakened by regular melt injections. In addition, similar geological observations attesting for melt-rock interaction and melt-assisted deformation are observed in lower crust and mantle rocks exposed at oceanic core complexes from slow-spreading MORs48, where the sporadic addition of melts or fluids has been invoked as a necessary trigger mechanism of deep detachment faulting49. Last, mafic veins and deformation fabrics seem to control the distribution of serpentine and carbonate-rich veins, formed by interaction with seawater during the later stages of exhumation of the sequence to the seafloor (Fig. 6c, d).

A key finding of Expedition 402 is the absence of recovered igneous crust, while the previous ODP Leg 107 recovered two ∼ 100 m long basaltic sections in the vicinity of the Gortani ridge (Site 655) and the central Vavilov basin (Site 651)24 (Fig. 1b). The Tyrrhenian oceanic plutonic material drilled so far is restricted to limited mafic intrusions and veins (olivine gabbro to gabbro, norite, diorite, and diabase), making < 10% of the entire recovered basement (Supplementary Fig. 4a, b). The MORB-type plutonic magmatism in the Tyrrhenian COT was thereby limited to a few sporadic mafic intrusions, often converted to rodingites, mainly documented at Site U1616. In line with the scarcity of igneous crust, our drilled section agrees with seismic data20,21,22,23 and show that the MORB-like basalt flows sampled by ODP Leg 10726 (Fig. 1b) do not continuously overlie the exhumed peridotites. Thus, despite magma was produced by mantle upwelling in the underlying asthenosphere, the nascent Tyrrhenian oceanic igneous crust was formed by sporadic intrusions in an exhuming, refertilized mantle. Indeed, some proportion of magma produced in the asthenosphere were retained within the mantle without the formation of a continuous plutonic crust16. The occurrence of melt-bearing zones could have promoted strain localization and the formation of lithospheric shear zones, leading to an efficient process of mantle exhumation18. While the infiltration of small aliquots of magma facilitates mantle exhumation, the development of deep-rooted detachment faults produces a cool thermal structure beneath the footwall, possibly decreasing melt production50. Mantle refertilization and prolonged lithospheric thinning through detachment faulting, hence, might have delayed the development of a steady-state spreading center.

Limited melt focusing is further indicated by another remarkable discovery of the IODP Expedition 402: the occurrence of phlogopite, plagioclase, amphibole-, and apatite-rich magmatic intrusions (2 wt% K2O, 0.6 wt% TiO2, 31.2 wt% MgO, 8.1 wt% Al2O3, and 14.7 wt% CaO; Supplementary Data 2) crosscutting the exhumed mantle section. At Site U1614, these intrusions occur either as irregular patches or as late-stage cm-scale veins cross-cutting the serpentinite fabric of the host peridotite (Figs. 3d; 6a, b and Supplementary Fig. 4c, d). The occurrence of hydrous potassic intrusions elevates the Al2O3 and K2O content in the associated lherzolite and dunite samples, such as those located at depths 350 and 360 mbsf at Site U1614 (Fig. 2 and Supplementary Fig. 3). The widespread refertilization recorded by the lherzolites and the injections of these potassic melts are linked, but the crosscutting relationships suggest two different stages of melt percolation. Indeed, while hydrous potassic intrusions are rare and randomly widespread in the entire section, the high Al-Ti-K contents of the refertilized peridotites are a common feature of the plagioclase-bearing lherzolite, with no spatial relationships with the late-stage potassic magmatic veins (Fig. 2). The location of the drilled sections in a back-arc environment suggests a contribution from the subducted slab in the formation of these hydrous potassic melts. It is worth noting, however, that magmatic intrusions with the same mineralogical association are reported in the Iberia-Newfoundland COT15, where initial decompression melting taps these fertile reservoirs before pervasive partial melting and mantle depletion beneath a mature spreading center15. Magmatism with an enriched geochemical signature seems thereby a characteristic of embryonic stages of oceanic evolution. In the Tyrrhenian Sea, these melts impart a potassic flavor to the mantle and may be involved in the origin of widespread MgO-rich potassic-ultrapotassic magmatism in the Italian peninsula51.

In conclusion, we show here that the penultimate IODP Expedition on the R/V JOIDES Resolution has recovered some exceptional mantle lithologies from the Tyrrhenian embryonic COT. Contrary to the conventional view that the amagmatic or magma-poor nature of rifts is due to prolonged mantle upwelling of the melt-depleted refractory mantle, drilling deep into the Tyrrhenian embryonic oceanic mantle disclosed its real nature: a heterogeneous mantle produced by injections of heterogeneous magmas. These findings may prove that the inception of oceanic crust in this magma-poor margin was deferred by the exhumation of a hybridized mantle lithosphere. Hence, the two drillholes into the embryonic Tyrrhenian Sea will provide the opportunity to re-evaluate models of the birth of a new ocean before widespread asthenospheric upwelling allows volumetrically significant melt production leading to steady-state seafloor spreading. This study shows that our enhanced understanding of several fundamental geological processes has been achieved following oceanographic expeditions, particularly, those aimed at drilling the Earth’s mantle.

All the data shown in this article have been produced onboard JOIDES Resolution during IODP Expedition 402. Bulk-rock geochemical analyses were obtained following the standard analytical protocols of ODP52, also followed in previous IODP Expeditions (e.g., ref. 53), and will be hereafter briefly described. Geochemical data are reported in Supplementary Data 1 and 2.

Major, minor, and trace elements were obtained using an Agilent 5110 inductively coupled plasma-atomic emission spectrometry (ICP-AES). Rock samples consisted of ~ 2 to 8 cm3 of material cut from the core with a diamond saw blade. Saw marks, alteration rinds, and a by-product of the drilling process, were carefully removed from the outer surfaces of the rock samples by grinding on a diamond-impregnated grinding wheel. Each sample was then sonicated for 15 min in acetone. After decanting of the acetone, the samples were then sonicated twice more in nanopure deionized water for 10 min. The cleaned samples were dried for 10–12 h at 110 °C and afterward were crushed to < 1 cm chips using two disks of Delrin plastic installed in a Jaw Crusher hydraulic press. A SPEX 8515 Shatterbox powdering system with a tungsten carbide mill was then used to grind the chips to a fine powder for 3 min. Then, a ~ 5 g aliquot of the sample powder was weighed on a Mettler Toledo balance and ignited at 950 °C using a quartz crucible for 4 h to determine loss on ignition (LOI) with an estimated precision of 0.02 g (0.4%).

Ignited powders of 100.0 ± 0.2 mg were weighed and mixed with 400.0 ± 0.5 mg of lithium metaborate (LiBO2) flux. Samples and standards were weighed on a Cahn C-31 microbalance designed to measure on a moving platform, with weighing errors of ± 0.05 mg under relatively smooth sea conditions. Standard rock powders and full procedural blanks were included with unknowns in each ICP-AES run. To prevent cooled glass beads from sticking to a crucible, a 10 µL aliquot of 0.172 mM aqueous LiBr solution was added to the mixture of flux and rock powder as a nonwetting agent. Samples were fused individually in Pt-Au (95:5) crucibles for 12 min at a maximum temperature of 1050 °C in an internally rotating induction furnace (Bead Sampler NT-4100). After cooling, beads were transferred to 125 mL high-density polypropylene (HDPE) bottles and dissolved in 50 mL of 10% (by volume) dilution of concentrated trace-metal grade HNO3 (10% HNO3, hereafter), aided by shaking with a Burrell wrist-action bottle shaker for 2 h. After digestion of the glass beads, the solutions were passed through a 0.45 μm filter into a clean 60 mL wide-mouth HDPE bottle. The solutions for major, minor, and trace element analyses were prepared from the filtered solutions, with a final solution-to-sample dilution factor of 5000 ×. The final solutions were prepared by pipetting 0.5 mL into a polyethylene centrifuge tube, which was diluted with 4.4 mL of 10 wt% HNO3 and 0.1 mL of an internal standard solution containing Be, In, and Sb (final concentration of 100 μg/L Be and In and 200 μg/L Sb).

Major (Al, Ca, Fe, K, Mg, Mn, Na, Si, and Ti) and trace (Ba, Co, Cr, Cu, Ni, P, Sc, Sr, V, Y, Zn, and Zr) element concentrations of standards and samples were determined using an Agilent 5110 ICP-AES. Plasma was ignited for at least 30 min before each run of samples to allow the instrument stabilization and warm-up time. Each sample was analyzed three times from the same dilute solution within a given sample run. Typically, 12 unknown samples were analyzed during a single run. A 10% HNO3 rinse solution was run for 90 s between each sample analysis. Procedural blank solutions were run at the beginning and end of each run. Certified international rock reference materials run as calibration, drift, and quality check standard solutions were included with the unknown samples for each sample run. Certified international rock reference materials for calibration standard solutions were peridotites JP-1 and DTS-1, basalts BIR-1, BCR-2, JB-1b, and JB-2a, gabbros MRG-1 and JGb-1, andesite AGV-1, granite JG-2 and carbonate NBS-1C. Two certified international rock reference materials of basalt BHVO-2 and peridotite JP-1 were chosen for their compositional similarity to that of the analyzed material and were run as unknowns to check the precision and accuracy of major and trace analyses.

After drift correction using internal standards (Be, In, Sb) and subtraction of the procedural blank, a calibration line for each element was calculated using the results for the certified rock standards by Agilent’s ICP Expert software. Concentrations used for the calibrations were compiled from literature52. Total Fe oxide concentrations were reported as Fe2O3T (Supplementary Data 1). Estimates of accuracy and precision of major and trace element analyses were based on the standard deviation between the three analyses of each sample and the replicate analyses of check standards, compared to published values. During the expedition, run-to-run relative standard deviation (RSD) by ICP-AES was generally ± 1–5% for major elements and ± 1–15% for trace elements. The RSD of P2O5 in peridotite JP-1 was exceptionally high, up to 21%, because P2O5 content is extremely low down to 0.005 wt% in JP-1 (Supplementary Data 3).

Total inorganic carbon (TIC) content was measured by quantifying the carbon dioxide (CO2) released from an acidification process on carbonate minerals using a UIC Coulometrics 5017 CO2 coulometer. Approximately 10– 12 mg of dried rock powders prepared for ICP-AES analyses were accurately weighed into a clean glass vial using two electronic balances and a computer averaging system that corrects for ship motion (Cahn C-31 automated electrobalance). Then, the rock powders were reacted with 5.0 mL of 1 M HCl on a hot plate set at 50°C, resulting in a release of CO2. This reaction converts any present carbonate to CO2. The liberated CO2 was then transferred with an N2 carrier gas to a coulometer cell filled with a solution containing excess monoethanolamine and a colorimetric pH indicator. As the gas passed through the coulometer cell, CO2 was quantitatively absorbed and reacted with the monoethanolamine to form the carbamate salt of monoethanolamine. This solution was titrated with electrochemically generated OH- to a colorimetric endpoint, and the corresponding change in light transmittance in the coulometric cell was monitored with a photodetection cell. This change in light transmittance was proportional to the TIC content of the sample. If a sample only contains calcium carbonates, all the evolved CO2 was derived from the dissolution of calcium carbonate, and the TIC content was measured 5 min after the beginning of the reaction. The weight percent of calcium carbonate was calculated from the inorganic carbon content using the following equation:

A weighed amount of standard calcium carbonate (> 99.9%, CaCO3, Fisher Scientific) and standard dolomite (MgCa(CO3)2, British Chemical Standards no. 368) was used to monitor analytical precision (1.46%) and accuracy (less than 1%). The measured data is presented in Supplementary Data 2.

The measured geochemical data on the rocks and standards reported in this study are presented in Supplementary Data 1, 2 and 3. Results of alphaMELTS modeling is presented in Suppementary Data 4. The data are available in Figshare: https://doi.org/10.6084/m9.figshare.27153891. Additional structural and lithological data are available in the Preliminary Report of IODP Expedition 402 at: http://publications.iodp.org/preliminary_report/402/. More extensive data sets are available in the Expedition 402 Proceedings volume at: http://publications.iodp.org/proceedings/402/402/title.html an https://doi.org/10.14379/iodp.proc.402.2025

Cannat, M. et al. Proceedings of the ocean drilling program. Init. Rep. 153, 5–13 (1995).

Google Scholar

Kelemen, P. B. et al. Leg 209 Summary. 209 (2004).

McCaig, A. et al. Expedition 399 preliminary report: Building blocks of life, atlantis massif. International Ocean Discovery Program. https://doi.org/10.14379/iodp.pr.399.2024 (2024).

Gillis, K. et al. Proceedings of the Ocean drilling program. Init. Rep. 147, 109–159 (1993).

MATH Google Scholar

Lissenberg et al. A long section of serpentinized depleted mantle peridotite. Science 385, 623–629 (2024).

Article CAS PubMed Google Scholar

Boillot, G. et al. Proceedings of the Ocean drilling program. Sci. Res. 103, 505–512 (1988).

Whitmarsh, R. B. et al. Site 1068. Proc. Ocean Drill. Prog. Init. Rep. 173, 163–218 (1998).

Google Scholar

Whitmarsh, R. B., Manatschal, G. & Minshull, T. A. Evolution of magma-poor continental margins from rifting to seafloor spreading. Nature 413, 150–154 (2001).

Article ADS CAS PubMed Google Scholar

Mohn, G. et al. Mode of continental breakup of marginal seas. Geology 50, 1208–1213 (2022).

Article ADS MATH Google Scholar

Pérez-Gussinyé, M. et al. Towards a process-based understanding of rifted continental margins. Nat. Rev. Earth Env. 4, 166–184 (2023).

Article MATH Google Scholar

Pérez-Gussinyé, M. & Reston, T. J. Rheological evolution during extension at nonvolcanic rifted margins: Onset of serpentinization and development of detachments leading to continental breakup. J. Geophys. Res. 106, 3961–3975 (2001).

Article ADS MATH Google Scholar

Lavier, L. L. & Manatschal, G. A mechanism to thin the continental lithosphere at magma-poor margins. Nature 440, 324–328 (2006).

Article ADS CAS PubMed Google Scholar

Müntener, O. & Manatschal, G. High degrees of melt extraction recorded by spinel harzburgite of the Newfoundland margin: The role of inheritance and consequences for the evolution of the southern North Atlantic. Earth Planet. Sci. Lett. 252, 437–452 (2006).

Article ADS Google Scholar

Péron-Pinvidic, G. & Manatschal, G. The final rifting evolution at deep magma-poor passive margins from Iberia-Newfoundland: a new point of view. Int. J. Earth Sci. 98, 1581–1597 (2009).

Article Google Scholar

Jagoutz, O. et al. The rift-to-drift transition in the North Atlantic: A stuttering start of the MORB machine? Geology 35, 1087–1090 (2017).

Article ADS MATH Google Scholar

Gillard, M. et al. Birth of an oceanic spreading center at a magma-poor rift system. Sci. Rep. 7, 15072 (2017).

Article ADS PubMed PubMed Central MATH Google Scholar

Pérez-Gussinyé, M., Morgan, J. P., Reston, T. J. & Ranero, C. R. The rift to drift transition at non-volcanic margins: Insights from numerical modelling. Earth Planet. Sci. Lett. 244, 458–437 (2006).

Article ADS Google Scholar

Larsen, H. C. et al. Rapid transition from continental breakup to igneous oceanic crust in the South China Sea. Nat. Geosci. 11, 782–789 (2018).

Article ADS CAS MATH Google Scholar

Sapin, F., Ringenbach, J.-C. & Clerc, C. Rifted margins classification and forcing parameters. Sci. Rep. 11, 8199 (2021).

Article ADS CAS PubMed PubMed Central MATH Google Scholar

Malinverno, A. & Ryan, W. B. F. Extension in the Tyrrhenian Sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere. Tectonics 5, 227–245 (1986).

Article ADS MATH Google Scholar

Faccenna, C., Funiciello, F., Giardini, D. & Lucente, P. Episodic back-arc extension during restricted mantle convection in the Central Mediterranean. Earth Planet. Sci. Lett. 187, 105–116 (2001).

Article ADS CAS Google Scholar

Prada, M., Ranero, C. R., Sallarès, V., Zitellini, N. & Grevemeyer, I. Mantle exhumation and sequence of magmatic events in the Magnaghi–Vavilov Basin (Central Tyrrhenian, Italy): New constraints from geological and geophysical observations. Tectonophys 689, 133–142 (2016).

Article ADS Google Scholar

Loreto, M. F., Zitellini, N., Ranero, C. R., Palmiotto, C. & Prada, M. Extensional tectonics during the Tyrrhenian back‐arc basin formation and a new morpho‐tectonic map. Basin Res. 33, 138–158 (2021).

Article ADS Google Scholar

Beccaluva, L et al. in Proceedings Ocean Drilling Program, Scientific Results. 107, (1990).

Kastens, K. & Mascle, J. The geological evolution of the Tyrrhenian Sea: An introduction to the scientific results of ODP Leg 107. Proc. Ocean Drill. Prog. Sci. Res. 107, 3–26 (1990).

MATH Google Scholar

Bonatti, E., Seyler, M., Channell, J., Giraudeau, J. & Mascle, G. Peridotites drilled from the Tyrrhenian Sea, ODP Leg 107. Proc. Ocean Drill. Prog. Sci. Res. 107, 37–47 (1990).

Google Scholar

Kelemen, P. B. Reaction between ultramafic rock and fractionating basaltic magma I. Phase relations, the origin of calc-alkaline magma series, and the formation of discordant dunite. J. Petrol. 31, 51–98 (1990).

Article ADS MATH Google Scholar

Suhr, G. Melt migration under oceanic ridges: Inferences from reactive transport modelling of upper mantle hosted dunites. J. Petrol. 40, 575–599 (1999).

Article ADS CAS MATH Google Scholar

Dygert, N. & Liang, Y. Temperatures and cooling rates recorded in REE in coexisting pyroxenes in ophiolitic and abyssal peridotites. Earth Planet. Sci. Lett. 420, 151–161 (2015).

Article ADS CAS Google Scholar

Sanfilippo, A., Salters, V., Tribuzio, R. & Zanetti, A. Role of ancient, ultra-depleted mantle in Mid-Ocean-Ridge magmatism. Earth Planet. Sci. Lett. 511, 89–98 (2019).

Article ADS Google Scholar

Ongunyele, A. C. et al. Transition from orogenic-like to anorogenic magmatism in the Southern Alps during the Early Mesozoic: Evidence from elemental and Nd-Sr-Hf-Pb isotope geochemistry of alkali-rich dykes from the Finero Phlogopite Peridotite, Ivrea–Verbano Zone. Gondwana Res. 129, 201–219 (2024).

Article ADS Google Scholar

González-Jiménez, J. M. et al. The architecture of the European-Mediterranean lithosphere: A synthesis of the Re-Os evidence. Geology 41, 547–550 (2013).

Article ADS MATH Google Scholar

Nicolas, A., Bouchez, J. L. & Mercier, J. C. Textures, structures and fabrics due to solid state flow in some European lherzolites. Tectonophysics 12, 55–86 (1971).

Article ADS Google Scholar

Dick, H. J. B. & Bullen, T. Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contrib. Mineral. Petrol. 86, 54–76 (1984).

Article ADS CAS MATH Google Scholar

Loocke, M., Snow, J. E. & Ohara, Y. Melt stagnation in peridotites from the Godzilla megamullion oceanic core complex, Parece Vela basin Philippine Sea. Lithos 182–183, 1–10 (2013).

Article ADS Google Scholar

Sani, C., Sanfilippo, A., Peyve, A. A., Genske, F. & Stracke, A. Earth mantle’s isotopic record of progressive chemical depletion. AGU Adv. 4, https://doi.org/10.1029/2022AV000792 (2023).

Warren, J. M. Global variations in abyssal peridotite compositions. Lithos 248–251, 193–219 (2016).

Article ADS MATH Google Scholar

Borghini, G. et al. Fast REE re-distribution in mantle clinopyroxene via reactive melt infiltration. Geochem. Perspect. Lett. 26, https://doi.org/10.7185/geochemlet.2323 (2023).

Warren, J. M. & Shimizu, N. Cryptic variations in abyssal peridotite compositions: Evidence for shallow-level melt infiltration in the oceanic lithosphere. J. Petrol. 51, 395–423 (2010).

Article ADS CAS Google Scholar

Trua, T., Clocchiatti, C., Schiano, P., Ottolini, L. & Marani, M. The heterogeneous nature of the Southern Tyrrhenian mantle: Evidence from olivine-hosted melt inclusions from back-arc magmas of the Marsili seamount. Lithos 118, 1–16 (2010).

Article ADS CAS Google Scholar

Zitellini, N., Malinverno, A. & Estes, E. R. International Ocean discovery program. Expedition 402 scientific prospectus: Tyrrhenian continent–Ocean transition. https://doi.org/10.14379/iodp.sp.402.2023 (2023).

Bodinier, J.-L., Garrido, C. J., Chanefo, I., Bruguier, O. & Gervilla, F. Origin of pyroxenite–peridotite veined mantle by refertilization reactions: Evidence from the Ronda peridotite (southern Spain). J. Petrol. 49, 999–1025 (2008).

Article ADS CAS Google Scholar

Rampone, E., Piccardo, G. B., Vannucci, R., Bottazzi, P. & Ottolini, L. Subsolidus reactions monitored by trace element partitioning: the spinel- to plagioclase-facies transition in mantle peridotites. Contrib. Mineral. Petrol. 115, 1–17 (1993).

Article ADS CAS Google Scholar

Garrido, C. J. & Bodinier, J. L. Diversity of mafic rocks in the Ronda peridotite: Evidence for pervasive melt–rock reaction during heating of subcontinental lithosphere by upwelling asthenosphere. J. Petrol. 40, 729–754 (1999).

Article ADS CAS Google Scholar

Morishita, T., Arai, S., Gervilla, F. & Green, D. H. Closed-system geochemical recycling of crustal materials in alpine-type peridotite. Geochim. Cosmochim. Acta 67, 303–310 (2003).

Article ADS CAS Google Scholar

Ferrando, C. et al. Deep segregation and crystallization of ultra-depleted melts in the sub-ridge mantle. Chem. Geol. 644, 121840 (2024).

Article CAS MATH Google Scholar

Ceuleneer, G. & Cannat, M. High-temperature ductile deformation of site 920 peridotites. Proc. Ocean Drill. Prog. Sci. Res. 153, 23–34 (1997).

CAS Google Scholar

Dick, H. J. B. et al. Dynamic accretion beneath a slow-spreading ridge segment: IODP Hole 1473A and the Atlantis Bank oceanic core complex. J. Geophys. Res. 124, 12631–12659 (2019).

Article ADS MATH Google Scholar

Bickert, M., Lavier, L. & Cannat, M. How do detachment faults form at ultraslow mid-ocean ridges in a thick axial lithosphere? Earth Planet. Sci. Lett. 533, 116048 (2020).

Article CAS Google Scholar

Mezri, L. et al. Tectonic controls on melt production and crustal architecture during magma-poor seafloor spreading. Earth Planet. Sci. Lett. 628, 118569 (2024).

Article CAS Google Scholar

Conticelli, S. et al. Trace elements and Sr–Nd–Pb isotopes of K-rich, shoshonitic, and calc-alkaline magmatism of the Western Mediterranean Region: Genesis of ultrapotassic to calc-alkaline magmatic associations in a post-collisional geodynamic setting. Lithos 107, 68–92 (2009).

Article ADS CAS Google Scholar

Murray, R., Miller, D. J., Kryc, K. Analysis of major and trace elements in rocks, sediments, and interstitial waters by inductively coupled plasma–atomic emission spectrometry (ICP-AES). ODP Technical Note 29 https://doi.org/10.2973/odp.tn.29.2000 (2000).

Planke, S., Berndt, C., Alvarez Zarikian, C. A., and the Expedition 396 Scientists. Mid-Norwegian margin magmatism and Paleoclimate implications. In Proceedings of the International Ocean Discovery Program, 396: College Station, TX (International Ocean Discovery Program, 2023).

Parkinson, I. J. & Pearce, J. A. Peridotites from the Izu–Bonin–Mariana forearc (ODP Leg 125): Evidence for mantle melting and melt–mantle interaction in a supra-subduction zone setting. J. Petrol. 39, 1577–1618 (1988).

Article ADS Google Scholar

Ohara, Y. et al. Peridotites from the Mariana Trough: first look at the mantle beneath an active back-arc basin. Contrib. Mineral. Petrol. 143, 1–18 (2002).

Article ADS CAS MATH Google Scholar

Ohara, Y., Fujioka, K., Ishii, T. & Yurimoto, H. Peridotites and gabbros from the Parece Vela backarc basin: Unique tectonic window in an extinct backarc spreading ridge. Geochem. Geophys. Geosys. 4, 8611 (2003).

Article ADS Google Scholar

Debret, B., Andreani, M. & Godard, M. A review of abyssal serpentinite geochemistry and geodynamics. Earth Sci. Rev. 258, 104910 (2024).

Article CAS Google Scholar

Niu, Y. Bulk-rock major and trace element compositions of abyssal peridotites: Implications for mantle melting, melt extraction and post-melting processes beneath mid-ocean ridges. J. Petrol. 45, 2423–2458 (2004).

Article ADS CAS MATH Google Scholar

Akizawa, N. et al. Geochemical characteristics of back-arc basin lower crust and upper mantle at final spreading stage of Shikoku Basin: an example of Mado Megamullion. Prog. Earth Planet. Sci. 8, 65 (2021).

Article ADS Google Scholar

Ghiorso, M. S., Hirschmann, M. M., Reiners, P. W. & Kress, V. C. The pMELTS: A revision of MELTS for improved calculation of phase relations and major element partitioning related to partial melting of the mantle to 3 Gpa. Geochem. Geophys. Geosys. 3, 1–35 (2002).

Article Google Scholar

Smith, P. M. & Asimow, P. D. Adiabat_1ph: A new public front-end to the MELTS, pMELTS, and pHMELTS models. Geochem. Geophys. Geosys. https://doi.org/10.1029/2004GC000816 (2005).

Workman, R. K. & Hart, S. R. Major and trace element composition of the depleted MORB mantle (DMM). Earth Planet. Sci. Lett. 231, 53–72 (2005).

Article ADS CAS MATH Google Scholar

Download references

This research used samples and data generated during the International Ocean Discovery Program (IODP) Expedition 402 on the JOIDES Resolution. We are grateful to the JOIDES Resolution Science Operator staff as well as the Siem Offshore crew members who made the expedition possible. E.R.E. was supported through NSF award OCE–1326927. L.M.’s PhD scholarship is supported by Geoscience Australia. W.M. is supported by the H2020 MSCA-IF – TURBOMUD project – GA No. 101018321. Support for E. Estes was provided by NSF grants OCE1326927 (IODP JRSO) and OCE2412279 (IODP Closeout). Gianreto Manatschal, Darin Schwartz, and three anonymous reviewers are thanked for their comments and suggestions that improved the earlier versions of the manuscript.

Department of Earth and Environmental Science, University of Pavia, Pavia, Italy

Alessio Sanfilippo

School of Earth, Environmental and Sustainability Sciences, Indian Institute of Science Education and Research Thiruvananthapuram, Kerala, India

Ashutosh Pandey

Atmosphere and Ocean Research Institute, University of Tokyo, Tokyo, Japan

Norikatsu Akizawa

Department of Geology and Geophysics, Louisiana State University, Baton Rouge, LA, USA

Eirini Poulaki & Brandon D. Shuck

Department of Geology and Geophysics, University of Utah, Salt Lake City, USA

Emily Cunningham & Victoria L. Pavlovics

Geo-Ocean, Univ Brest, CNRS, IFREMER, Plouzané, France

Manon Bickert

College of Marine Science and Engineering, China University of Geosciences, Wuhan, China

Chao Lei

Earth Sciences Department, University of Florence, Florence, Italy

Paola Vannucchi

International Ocean Discovery Program, Texas A&M University, College Station, Texas, USA

Emily R. Estes

Lamont-Doherty Earth Observatory, Columbia University, Palisades, USA

Alberto Malinverno

Tono Geoscience Center, Japan Atomic Energy Agency (JAEA), Tokyo, Japan

Noriaki Abe

Department of Biological, Geological and Environmental Science, Earth Science Division, University of Catania, Catania, Italy

Agata Di Stefano

Earth and Atmospheric Sciences, University of Nebraska, Lincoln, USA

Irina Y. Filina

Department of Earth and Atmospheric Sciences, University of Houston, Houston, USA

Qi Fu

Laboratoire d’Océanographie et du Climat: Expérimentations et Approches Numériques, Sorbonne Université, Campus Pierre et Marie Curie, Paris, France

Swanne B. L. Gontharet

University of Texas Institute for Geophysics, University of Texas at Austin, Austin, TX, USA

Lorna E. Kearns

Department of Earth Sciences, Binghamton University, New York, USA

Ravi Kiran Koorapati

Institute of Marine Sciences - ISMAR, Italian National Research Council, Bologna, Italy

Maria Filomena Loreto & Nevio Zitellini

Institute for Marine and Antarctic Studies (IMAS), University of Tasmania, Hobart, Australia

Luca Magri

Department of Marine Geosciences, ICM-CSIC, Barcelona, Spain

Walter Menapace

Geosciences Montpellier, Montpellier, France

Philippe A. Pezard

Marine Biology, Texas A&M University at Galveston, Galveston, USA

Milena A. Rodriguez-Pilco

School of Oceanography, Shanghai Jiao Tong University, Shanghai, China

Xiangyu Zhao

Instituto Andaluz de Ciencias de la Tierra (IACT), Armilla, Granada, Spain

Carlos Garrido

Dipartimento di Scienze Chimiche e Geologiche, Università di Modena e Reggio Emilia, Modena, Italy

Daniele Brunelli

School of Geosciences and Civil Engineering, College of Science and Engineering, Kanazawa University, Kanazawa, Japan

Tomoaki Morishita

You can also search for this author in PubMed Google Scholar

N.Z., M.F.L., D.B., and C.G. conceived the project and contributed to the original IODP proposal. E.R.E., A.M., and N.Z. led Expedition 402. M.B., C.L., T.M., A.P., E.P., A.S., and P.V. performed the petrographic and structural description of mantle peridotites and associated lithologies. Nak, E.C., S.G., and Q.F. performed the whole-rock geochemical analysis. Nab, I.F., M.F.L., B.S., P.P., V.P., and X.Z. measured the physical and magnetic properties of the peridotites. L.K., L.M., and W.M. described the peridotite-sediments interface and sedimentary lithologies. A.D.S. and R.K.K. constructed the preliminary age model of the sediments. M.R.P. quantified biogeochemical cycling of elements. All authors contributed equally to the interpretation of the data. A.S. and A.P. wrote the text with contributions from all authors.

Correspondence to Alessio Sanfilippo.

The authors declare no competing interests.

Nature Communications thanks Gianreto Manatschal and Darin Schwartz for their contribution to the peer review of this work. A peer review file is available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

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/.

Reprints and permissions

Sanfilippo, A., Pandey, A., Akizawa, N. et al. Heterogeneous Earth’s mantle drilled at an embryonic ocean. Nat Commun 16, 2016 (2025). https://doi.org/10.1038/s41467-025-57121-0

Download citation

Received: 05 October 2024

Accepted: 11 February 2025

Published: 27 February 2025

DOI: https://doi.org/10.1038/s41467-025-57121-0

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative