Marine carbon dioxide removal is increasingly seen as a necessary complement to emissions cuts, and ocean alkalinity enhancement (OAE) is among the most widely proposed approaches. OAE works by raising seawater alkalinity (A_­T), and with it the calcium carbonate (CaCO₃) saturation state, to draw additional CO₂ into the ocean as dissolved inorganic carbon. The shelf seafloor, however, is a large A_T source, recycling ~22 Tmol A­_T per year through CaCO₃ dissolution, which is dependent on the CaCO₃ saturation state of the overlying water. Because OAE deliberately shifts the saturation state, it could affect this natural source.

Here, we combine 25 years of shelf and Subantarctic carbonate-system observations from the Munida time series off Aotearoa New Zealand with experimental sediment incubations from three different shelf locations across Aotearoa (east Otago, Te Matau-a-Māui, and Taranaki). The incubation experiments show that CaCO₃ dissolution in the seafloor increases with decreasing saturation state, which constitutes a "fast climate feedback" that can explain the increase in shelf alkalinity over recent decades observed in the Munida time series. Inversely, increasing alkalinity suppresses seafloor CaCO₃ dissolution, indicating that OAE reduced natural A_T generation. Initial modelling suggests this could lead to an OAE efficiency loss of 1-10%, indicating that accounting for the shelf seafloor feedback will be essential to gain realistic estimates of the carbon sequestration potential of OAE.