Alexandra Yang Yang

Subduction transports carbon into Earth's interior, yet how subducted carbon influences arc magma ascent dynamics remains unclear. Here we combine volatile compositions of olivine-hosted melt inclusions with olivine diffusion chronometry at Pagan volcano (Mariana arc) to reveal a direct link between carbonate-capped seamount subduction and rapid magma ascent. Melt inclusions record high CO₂ contents (2000-6000 ppm), and magma storage near Moho depths (~20 km). Olivine zoning profiles constrain magma ascent from near-Moho to the surface on timescales of weeks to months. Elevated CaO/Al₂O₃ ratios (>1) in erupted volcanics, heavy δ¹³C values (~0.5‰) in monitored volcanic gases, and the occurrence of seamount carbonate in the forearc mud volcano collectively point to subducted seamount carbonates as the dominant CO₂ source feeding Pagan magmatism. We propose that early CO₂ exsolution near the Moho generates overpressure that enables hydrous arc magmas to bypass crustal stalling. These results identify CO₂ as a key driver of rapid magma ascent in arcs and highlight carbonate-capped seamount subduction as an efficient pathway for deep carbon cycling.

The role of subducted lower oceanic crust gabbro in the water cycle of subduction zones has been underappreciated. This study conducted boron and strontium isotope analyses for basaltic glasses from the Eastern Lau Spreading Center (ELSC) at the Tonga subduction zone. Our results indicate that ELSC1 magmas, with heavy B isotopes and elevated H2O/Ce ratios, exhibit the lowest 87Sr/86Sr ratios among Lau Basin basalts, suggesting contribution of water released from lower crust gabbro. Globally, the lower crust gabbro, hosting at least 20% of total water budget of the subducted ocean plate, would release water continuously up to depths of ~250 km, implying a globally important role in subduction zone water cycle. Similar signals identified in the Izu, Lesser Antilles, and Cascades arcs support the widespread water contribution from gabbro. Our findings call for a reassessment of the lower crust’s role in the global water budget of subduction zones.

Subduction of oceanic plates drives carbon cycling between Earth’s interior and surface, yet the influence of subducted carbon on arc magma ascent dynamics remains poorly understood. Here we integrate volatile analyses of olivine-hosted melt inclusions and diffusion chronometry at Pagan volcano (Mariana arc) to reveal a direct link between carbonate-capped seamount subduction and rapid magma ascent. Melt inclusions record high CO₂ contents (2000-6000 ppm), and storage at Moho depths (20 km). Diffusion profiles of Fo, Ni, and Mn in olivine constrain magma ascent from the Moho to the surface to have occurred within ~100 days. Elevated CaO/Al₂O₃ ratios (>1) in melts, heavy δ¹³C values (~0.5‰) in volcanic gases, and the presence of seamount carbonate in the forearc mud volcano collectively indicate subducted seamount carbonates as the main source of CO₂-rich magma at Pagan. We propose that early CO₂ exsolution at Moho depths generates overpressure, enabling hydrous arc magmas to bypass stalling within crust. This mechanism provides a new perspective on CO₂-driven arc eruptions and highlights the role of seamount subduction in deep carbon cycling.

North Atlantic Deep Water (NADW), the return flow component of the Atlantic Meridional Overturning Circulation (AMOC), is a major inter-hemispheric ocean water mass with strong climate effects but the evolution of its source components on million-year timescales is poorly known. Today, two major NADW components that flow southward over volcanic ridges to the east and west of Iceland are associated with distinct contourite drift systems that are forming off the coast of Greenland and on the eastern flank of the Reykjanes (mid-Atlantic) Ridge. Here we provide direct records of the early history of this drift sedimentation based on cores collected during International Ocean Discovery Programme (IODP) Expeditions 395C and 395. We find rapid acceleration of drift deposition linked to the eastern component of NADW, known as Iceland–Scotland Overflow Water at 3.6 million years ago (Ma). In contrast, the Denmark Strait Overflow Water feeding the western Eirik Drift has been persistent since the Late Miocene. These observations constrain the long-term evolution of the two NADW components, revealing their contrasting independent histories and allowing their links with climatic events such as Northern Hemisphere cooling at 3.6 Ma, to be assessed.

Global ocean crust has an average thickness of 6–7 kilometers, suggesting a globally pervasive, rather uniform mantle composition. At some ultra-slow spreading ridges, crustal thickness is highly variable and mantle peridotite can be found at the surface. The peridotites, however, are mostly recovered in fracture zones that expose the deeper crust, or at the edges of ridge segments where there is a central volcano. The Gakkel Ridge is unusual in this regard because it contains a 400-kilometer-long sparsely magmatic zone (SMZ) with extensive mantle peridotite exposure, negligible crustal thickness and limited basaltic lava outcrops. This segment is also bracketed by two other sections of ridge that have active volcanism, including the adjacent Western Volcanic Zone (WVZ) where no peridotites were recovered. What is the origin of this enigmatic expanse of ridge, and is it simply a curiosity or does it have global implications for ocean ridges and mantle recycling? We have undertaken systematic geochemical analysis of 267 basaltic glass samples from the WVZ and the few recovered basalts from the SMZ. The WVZ has normal-thickness oceanic crust and predominantly produces depleted normal mid-ocean ridge basalt (N-MORB). Gradients in chemical composition can be accounted for by a combination of more depleted mantle and lower extents of melting as the SMZ is approached. Across an abrupt boundary, the SMZ has negligible crustal thickness and is dominated by exposed mantle peridotite and a few samples of enriched mid-ocean ridge basalt (E-MORB). Quantitative models suggest the SMZ is the result of cold, ancient ocean mantle lithosphere that has been metasomatized by enriched, low degree melts. While the SMZ is a rare occurrence, simple mass balance considerations suggest such occurrences should instead be very common. While recycled ocean crust is commonly called upon, sometimes as an isolated lower mantle reservoir, the mass of depleted ocean mantle lithosphere would be more than ten times greater. Indeed, using current ridge production rates, over the last 2.5 billion years the total volume of recycled mantle lithosphere would be equivalent to the volume of the entire lower mantle. While vestiges of such lithosphere are frequently invoked from Os isotopes or melt inclusions, almost all of these occurrences are coincident with predominant basalts, and occur in regions with normal crustal thickness. Why are there not vast regions dominated by depleted lithosphere, negligible crust, or common occurrences of basalts that come from highly depleted reservoirs? An obvious solution is that mantle convection is highly efficient at mixing crustal and mantle components on a scale finer than is sampled by melting, permitting relatively uniform crustal thickness and composition on a global basis.

The water cycle between Earth’s surface and interior plays a critical role in maintaining long-term sea-level stability. The discovery of ~24-km-thick serpentinized mantle in the Pacific plate near the Mariana Trench suggests that water influx could be up to three times greater than previous estimates. However, the contributions of thick slab serpentinite to the magma genesis in the subduction zones remain uncertain. We integrate new boron isotopic data from Mariana arc volcanics and Lau Basin basalts with Sr isotopic, halogen, and trace element data. One rear arc volcano shows evidence of contributions from slab serpentinite, which otherwise has a limited role in the overall chemical variations of arc-basin magmatism. Instead, water-rich fluxes from lower crustal gabbro contribute to back-arc magmatism in both the Mariana and Tonga subduction zones. The geochemical signature of lower crustal gabbro-derived fluxes is also evident in volcanoes from the Izu rear-arc, the Lesser Antilles Arc, and the Cascades arc. Water flux calculations demonstrate that the slab crust alone provides sufficient water influx to balance outflux through Mariana magmatism, raising the question of the fate of slab serpentinite. It appears that most of the water from serpentinite does not contribute to the arc, and hence would lead to a large deep water return flux to the mantle. If applied globally, however, such subduction would imply a sea-level drop of ~500 m per 100 million years, contradicting geological evidence of long-term sea-level stability. To reconcile this discrepancy, we propose that the water budget of slab serpentinite is either overestimated or not representative of cold slabs. A globally averaged thickness of ~2 km of slab serpentinite with 15 vol% serpentinization (total water budget within 3.4 × 10⁷ Tg/Myr) is permitted to maintain the sea-level stability during the Phanerozoic.

Extinct spreading ridges are globally widespread and are crucial to understanding the lifespan of oceanic plates. Yet the nature of the LAB beneath extinct ridges remains enigmatic. In this study, we investigate the LAB structure beneath the SCS basin, where a ~700-km-long extinct ridge system stopped spreading at ~15 Ma. A 120 km long marine magnetotelluric (MT) transect perpendicular to the extinct mid-ocean ridge in the southwest sub-basin of SCS was carried out at September 2021. The electrical resistivity model reveals a relatively low-resistivity layer at depths of 50-80 km, potentially corresponding to 0.1%-0.9% partial melts. This low-resistivity layer is heterogeneous and absent directly beneath the extinct ridge axis. This observation supports a model in which melts are efficiently extracted beneath the ridge axis, leaving the central region depleted, while partial melts are retained in the surrounding areas on either flank. Additionally, residual melts at shallower depths have likely solidified due to plate cooling, while deeper melts indicate the depth of the LAB. These findings propose a new mechanism for the emplacement of long-lived partial melts at the LAB and suggest that a discontinuous melt-rich layer may commonly occur near extinct spreading ridges globally.