Operational assessment of atmospheric carbon dioxide removal and carbon sequestration from a giant kelp farm in Luderitz, Namibia

Giant kelp has been identified as a promising nature-based carbon dioxide removal (CDR) pathway (Pessarrodona et al., 2024) and several scientific studies have analyzed the processes involved both in the different carbon sequestration pathways and the atmospheric carbon dioxide removal (Krause-Jensen and Duarte, 2016). Giant kelp is well-known for its rapid growth of up to tens of centimeters per day, which is fueled by the consumption of nutrients and dissolved inorganic carbon (DIC). As a result, the local concentration of CO2 in the seawater decreases, enabling the uptake of atmospheric CO2. The consumed CO2 is transformed by the kelp to grow and increase its biomass, but a fraction of the carbon is exported to the ocean as dissolved organic carbon (DOC) and particulate organic carbon (POC). Both POC and DOC are transported by ocean currents and can eventually be effectively sequestered for a long time. In the case of POC, fragments of kelp are exported into the deep ocean or are buried in the continental shelf sediment (Pessarrodona et al., 2024). A fraction of the DOC, known as recalcitrant DOC, is highly stable for very long periods of time and is also considered as an effective carbon sequestration pathway (Zhang et al., 2023).In 2024, Kelp Blue operated a pilot giant kelp farm in Luderitz, Namibia and was selected as a CDR XPRIZE finalist. One of the goals for the project was to assess the ability of kelp farms to remove atmospheric CO2 and sequester carbon for over 100 years.In this study, we present the results of atdepth’s independent assessment of Kelp Blue’s project. One of the main challenges for marine nature-based solutions is to assess the additionality of CDR and the sequestration of carbon through POC and DOC export. To tackle the problem, we first developed a detailed MRV methodology leveraging prior scientific studies. The methodology combines monitoring data with advanced physical and biogeochemical simulations of the ocean and the kelp. A giant kelp model developed by our collaborators at University of Cambridge was implemented to assess the net primary production of kelp, and the resulting change in surface ocean CO2 and nutrient concentrations.The model was informed by kelp biomass and ocean biogeochemistry data and integrated with a high-resolution physical and biogeochemical model of the ocean using OCM1, a novel simulation code developed by atdepth, that leverages advanced numerical frameworks from MIT and University of Cambridge.OCM1 integrates data from the project site, and from global ocean and atmospheric data products such as Copernicus’ GLORYSv12, or the European Centre for Medium Range Weather Forecast’s (ECMWF) ERA5 data. The simulation system achieved unprecedented performance compared to legacy ocean modeling systems; the simulation was run on a single NVIDIA A100 GPU, compared to the 1300 CPUs required to run the equivalent simulation using existing ocean modeling codes, which results in a computational cost reduction of over 350x.About atdepth: atdepth is an MIT spinoff company supported by the U.S. Department of Energy through the SEA-CO2 program and by NASA through the CMS program.