Remobilization of sedimentary carbonate in subduction zones modulates arc volcanism emissions and thus Earth’s climate over geological timescales. Although limestones (or chalk) are thought to be the major carbon reservoir subducted to subarc depths, their fate is still unclear. Here we present high-pressure reaction experiments between impure limestone (7.4 wt.% clay) and dunite at 1.3–2.7 GPa to constrain the melting behaviour of subducted natural limestone in contact with peridotite. The results show that although clay impurities significantly depress the solidus of limestone, melting will not occur whilst limestones are still part of the subducting slab. Buoyancy calculations suggest that most of these limestones would form solid-state diapirs intruding into the mantle wedge, resulting in limited carbon flux to the deep mantle (< ~10 Mt C y⁻¹). Less than 20% melting within the mantle wedge indicates that most limestones remain stable and are stored in subarc lithosphere, resulting in massive carbon storage in convergent margins considering their high carbon flux (~21.4 Mt C y⁻¹). Assimilation and outgassing of these carbonates during arc magma ascent may dominate the carbon flux in volcanic arcs.

The generation of strongly potassic melts in the mantle requires the presence of phlogopite in the melting assemblage, while isotopic and trace element analyses of ultrapotassic rocks frequently indicate the involvement of subducted crustal lithologies in the source. However, phlogopite-free experiments that focus on melting of sedimentary rocks and subsequent hybridization with mantle rocks at pressures of 1–3 GPa have not successfully produced melts with K₂O >5 wt%–6 wt%, while ultrapotassic igneous rocks reach up to 12 wt% K₂O. Accordingly, a two-stage process that enriches K₂O and increases K/Na in intermediary assemblages in the source prior to ultrapotassic magmatism seems likely. Here, we simulate this two-stage formation of ultrapotassic magmas using an experimental approach that involves re-melting of parts of an experimental product in a second experiment. In the first stage, reaction experiments containing layered sediment and dunite produced a modally metasomatized reaction zone at the border of a depleted peridotite. For the second-stage experiment, the metasomatized dunite was separated from the residue of the sedimentary rock and transferred to a smaller capsule, and melts were produced with 8 wt%–8.5 wt% K₂O and K/Na of 6–7. This is the first time that extremely K-enriched ultrapotassic melts have been generated experimentally from sediments at low pressure applicable to a post-collisional setting.

Sediments play a key role in subduction. They help control the chemistry of arc volcanoes and the location of seismic hazards. Here, we present a new model describing the fate of subducted sediments that explains magnetotelluric models of subduction zones, which commonly show an enigmatic conductive anomaly at the trenchward side of volcanic arcs. In many subduction zones, sediments will melt trenchward of the source region for arc melts. High-pressure experiments show that these sediment melts will react with the overlying mantle wedge to produce electrically conductive phlogopite pyroxenites. Modelling of the Cascadia and Kyushu subduction zones shows that the products of sediment melting closely reproduce the magnetotelluric observations. Melting of subducted sediments can also explain K-rich volcanic rocks that are produced when the phlogopite pyroxenites melt during slab roll-back events. This process may also help constrain models for subduction zone seismicity. Since melts and phlogopite both have low frictional strength, damaging thrust earthquakes are unlikely to occur in the vicinity of the melting sediments, while increased fluid pressures may promote the occurrence of small magnitude earthquakes and episodic tremor and slip.