Global biogeochemical tracers simulated by the CESM2-MARBL under present and future climate conditions

To reproduce our results, users should first download and correctly install CESM2.1.3 (see the CESM official repository for installation guidance). Next, you should download All_simulations_Initial_conditions.rar as the initial forcing and execute the commands provided in All_simulation_setups.docx. If you want to perform SensitivityExperiments , you should download SensitivityExperiments_CodeReviseSetup.rar . Detailed guidance on how to activate and use this revised code can be found in the SensitivityExperiments_CodeReviseSetup_Introduction.md. Please note that file paths must be updated to match the user's own environment. The simulation duration can be adjusted according to specific research objectives. If you have any questions, you could ask for help in introductory files and DiscussCESM Forums. In summary, a comprehensive set of materials has been provided to fully reproduce the experiments and outcomes. Annually averaged outputs of ocean biogeochemical tracers and tracer fluxes are stored in tar.gz archives named BHIST_**, BSSP370_**, BSSP585_**, and GIAF_**. A detailed description of all dataset structure and content is available in Annual_DataSets_introduction.md. The default model code and configuration are archived in Default_CESM213.tar.gz, which contains the Community Earth System Model version 2, coupled with the Marine Biogeochemistry Library (CESM2-MARBL). The complete description of each experimental design and the code modifications is provided in SensitivityExperiments_CodeReviseSetup.rar, with the related instructions available in SensitivityExperiments_CodeReviseSetup_Introduction.md. Four GIAF_** experiments were conducted using the active ocean-sea ice configuration of CESM2 under present-day climate forcing. The default model was forced with the Japanese 55-year reanalysis dataset (JRA-55) for the period 1958–2018 and was first spun up for four full cycles to reach a quasi-steady state in the upper ocean. In the fifth cycle, the ocean state from the year 1990 was used as the initial condition for the sensitivity experiments. From this point, the fifth cycle proceeded along two branches: (1) the control simulation (GIAF_control): a continuation of the default run without any temperature sensitivity modifications; and (2) three sensitivity experiments, each incorporating different temperature parameter modifications. Each experiment ran through the end of the fifth cycle. The outputs presented here are the annual averages for each of the final ten years (2009–2018) of tracer concentrations and fluxes. Eight B** experiments were performed using the fully coupled CESM2 under both historical and future climate forcings. Future climate simulations were conducted under IPCC’s SSP3-7.0 and SSP5-8.5 emission pathways. Three B**_control experiments were conducted using the default model forced by historical (BHIST_control) or future (BSSP370_control and BSSP585_control) climate conditions. Five sensitivity experiments are branched from the years 1975 (BHIST_** experiments) and 2075 (BSSP370_** and BSSP585_** experiments), and extended for 25 years. This study analyzed data from the BSSP370_** and BSSP585_** experiments for the period 2090–2099, and from the BHIST_** experiment for the period 1990–1999. Among them, B**_PftZooRemin_variedq10 experiments incorporate both plankton-specific Q10 values and temperature-dependent POM remineralization, while B**_PftZooOMRemin_q10 experiments additionally include temperature-dependent DOM remineralization.