Revisiting ocean carbon sequestration by direct injection: A global carbon budget perspective Fabian Reith, David P. Keller & Andreas Oschlies

Abstract:

Marchetti [1977] proposed that CO₂ could be directly injected into the deep ocean to mitigate its rapid build-up in the atmosphere. Although previous studies have investigated biogeochemical and climatic effects of injecting CO₂ into the ocean, they have not looked at global carbon cycle feedbacks and backfluxes that are important for accounting. Using an Earth System Model of intermediate complexity we simulated the injection of CO₂ into the deep ocean during a high CO₂ emissions scenario. At seven sites 0.1 GtC yr^-1 was injected at three different depths (3 separate experiments) between the years 2020 and 2120. After the 100-year injection period, our simulations continued until the year 3020 to assess the long-term dynamics. In addition, we investigated the effects of marine sediment feedbacks during the experiments by running the model with and without a sediment sub-model. Our results, in regards to efficiency (the residence time of injected CO₂) and seawater chemistry changes, are similar to previous studies. However, from a carbon budget perspective the targeted cumulative atmospheric CO₂ reduction of 70 GtC was never reached. This was caused by the atmosphere-to-terrestrial and/or atmosphere-to-ocean carbon fluxes (relative to the control run), which were effected by the reduction in atmospheric carbon. With respect to global oceanic carbon, the respective carbon cycle-climate feedbacks led to an even smaller efficiency than indicated by tracing the injected CO₂. The ocean also unexpectedly took up carbon after the injection at 1500 m was stopped because of a deep convection event in the Southern Ocean. These findings highlighted that the accounting of CO₂ injection would be challenging.