Combining and scaling up the application of terrestrial and marine CDR methods does not compromise CDR efficiency

Reaching the Paris Agreement’s 1.5°C climate goal will require the large-scale deployment of Carbon Dioxide Removal (CDR). Relying on single CDR methods, however, risks exceeding sustainability thresholds as compared to CDR portfolios that integrate both land- and marine-based methods. Therefore, Integrated Assessment Models have already started to include diverse CDR portfolios in modelled future pathways. While Earth System Models (ESMs) have been used to explore the climate and carbon cycle feedbacks under the deployment of individual methods, no study has yet examined the co-application of land- and marine-based CDR methods using an ESM.

Here, we use two fully coupled Earth System Models (MPI-ESM and FOCI) to investigate scaling up and/or combining land- and marine-based CDR methods under a high-emissions scenario (SSP3-7.0). Specifically, we examine the whole spectrum of Afforestation/Reforestation (AR) (0-927 Mha) and Ocean Alkalinity Enhancement (OAE) (0-16 Pmol) using a multifactorial setup encompassing seven scenarios and an ensemble of 42 simulations. The AR scenario includes ambitious forestation within the range of country pledges and has been developed based on 1,259 scenarios generated by Integrated Assessment Models, while considering biodiversity constraints and restoration potential maps. The OAE scenario includes the continuous application of alkalinity across ice-free coastline gridcells globally, with up to 16 Pmol applied – an amount sufficient to sequester as much carbon in the ocean as is the sequestration on land in the AR scenario.

Our results suggest that the efficiency of CDR, expressed as the fraction of removed carbon that remains out of the atmosphere, is ~0.85-0.87 for both AR and OAE and is independent of the magnitude of the CDR application. Overall, scaling up and/or combining the two CDR methods results in a linear scaling of carbon flux responses, despite the emerging feedbacks in the Earth system. Specifically, compared to a counterfactual no-CDR scenario, the simulated AR and OAE reduce atmospheric carbon by up to 429 and 503 GtCO2, respectively, and co-applying the two results in a reduction of 856 GtCO2. Halving the application of AR and OAE results in a reduction of atmospheric carbon by 220 and 225 GtCO2 respectively, while their combination yields 443 GtCO2.

Our findings suggest flexibility in designing CDR portfolios, as incorporating both land- and marine-based CDR methods does not compromise one or the other method’s efficiency in the two models applied. This may address sustainability concerns around large-scale deployment of single methods and can alleviate the pressure on the water-food-land nexus.