Regional ocean dynamics regulate the efficiency of ocean alkalinity enhancement

Ocean alkalinity enhancement (OAE) is a promising marine carbon dioxide removal (mCDR) method that aims to reduce atmospheric CO 2 by increasing the ocean's storage capacity. While global ocean and Earth system models are necessary to track air-sea CO 2 equilibration and far-field alkalinity transport on extended timescales, OAE efficiency is strongly influenced by local oceanography and climate. Regional ocean modeling is a powerful tool for capturing the effects of these influences on near-field plume dispersion and CO 2 uptake variability. Here, we used a high-resolution, three-dimensional hydrodynamic and biogeochemical model (2 km) to evaluate the effects of deployment location and interannual climate variability on OAE efficiency in Bass Strait, southeast Australia. We simulated the addition of 113.21 Gmol of alkalinity (a theoretical uptake capacity of ~4.2 Mt CO 2 ) over one month, via four infrastructure-constrained pathways: a desalination outfall, a shipping lane, a ferry track, and a series of coastal outfalls. These additions were repeated across three Southern Annular Mode (SAM) phase end-member winters: 2017 (positive SAM–low winds), 2021 (neutral SAM–moderate winds), and 2023 (negative SAM–strong winds). CO 2 uptake efficiency variability (mol CO 2 /mol TA) is primarily influenced by delivery method and location (95.2%) rather than by interannual climate variability (3.9%). At the shelf-break, 70.67% ± 10.53% of added alkalinity is subducted below the mixed layer before equilibration with the atmosphere; this alkalinity is then exported from the Bass Strait region at depth. The subduction and loss of alkalinity from the region before equilibration reduces the realised CO 2 uptake and contributes to efficiencies (0.11–0.27) that are lower than those simulated by a global model in the same region (0.31). This mismatch, driven by regionally specific oceanographic processes, has implications for OAE deployment and equilibration timescales in other dynamic shelf environments. To resolve these scale-dependent limitations, we recommend integrated monitoring, reporting, and verification frameworks that combine observational networks, regional models, and global models (with regional model exports as input to global models). This approach is necessary to accurately quantify net carbon removal and to constrain the long-term fate of added alkalinity.