This repository contains the processed datasets, analysis scripts, and figure-generation code sufficient to reproduce the results published in the article:
Convection injects labile particulate organic carbon to the deep ocean. Science Advances (in press). Martí Galí*, María Andrea Orihuela-García, Yohan Ruprich-Robert, Vladimir Lapin, María Sánchez-Urrea, Marcos Fontela, Joan Llort, Valentina Sicardi, Raffaele Bernardello *Corresponding author. Email: mgali@icm.csic.es
Source code in this repository is distributed under the MIT License. Processed datasets are distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.
Input data available at DOI:10.5281/zenodo.20611379
./ (root directory within this repository)
- Data analysis and figure-generation scripts for Figs. 1-5, S1-S16 and S20.
input_data/
- Files required to reproduce Figs. 1-5, S1–S16 and S20
- NOTE: Datasets that can be downloaded from official public repositories (e.g., Argo, OC-CCI, CMEMS) are not included.
output/
- Default directory for the figures and tabulated data generated by the scripts
GLORYS12v1_code_data_MSU/
- Scripts for GLORYS12v1 reanalysis data processing, generating input data for Fig. 2, S14 and S15.
OPERA_sims_code_data_MAOG/
- Scripts for processing NEMO4.0.4-PISCESv2_RC simulation data from the OPERA project, generating input for Fig. 3B, 4, 5, and S17-S19.
./analyze_plot_mergedProfilesTrajSat.R
- Reproduces Figs. S3-S9
- Classifies annual BGC-Argo time series into pre-convection, convection and post-convection periods
- Produces BGC-Argo and satellite merged datasets for their use in posterior plots and analyses
- Developed with R 4.3.3
./fig_conv_events.R
- Reproduces Fig. 1, S1, S10, S11, and S12
- Generates intermediate summary dataset for Table S3
- Developed with R 4.3.3
./fig_chl_sat_traj_mld.R
- Reproduces Fig. 2 and Movie S1
- Developed with R 4.3.3
./fig_conv_events_pisces.R
- Reproduces Fig. 3 and S13
- Developed with R 4.3.3
./fig_conv_fluxes_budgets.R
- Reproduces Fig. 4, 5A, and S16A
- Generates data and stats for Fig. 5B (schematic) and Tables S5 and S6
- Developed with R 4.3.3
./fig_anomaly_conv.R
- Reproduces Fig. S16 B and C
- Developed with R 4.3.3
./fig_S2.ipynb
- Reproduces Fig. S2
- Developed with Python X (Jupyter notebook)
./fig_mlotst_opera_glorys.R
- Reproduces Fig. S14 and S15
- Developed with R 4.3.3
./OPERA_sims_code_data_MAOG/mega_plot_phydiat_phymisc/megaplot_phydiat_phymisc_estadisticos-Zenodo.py
- Reproduces Fig. S17-S19
- Developed with Python 3
./fig_S20.R
- Reproduces Fig. S20
- Developed with R 4.3.3
./f_*.R
- Generic functions called in many scripts
- Some hard-coded paths may need to be updated to execute the scripts in a different computing environment.
- The code and analyses were developed over multiple stages and across various projects, eventually converging into this manuscript, explaining the complex structure of this unified repository.
- NEMO-PISCES simulations performed under the OPERA project (PID2019-107952GA-I00) were executed on the Barcelona Supercomputing Center (BSC) high-performance computing infrastructure. Simulation workflows used the autosubmit software, enabling reproducibility (see SM references), and are stored at the institutional GitLab repository of the BSC. Full raw simulation outputs are archived within BSC storage systems and are too large (~3TB) for public redistribution. The processed datasets required to reproduce the analyses and figures presented in the manuscript are provided in this repository.
- NEMO-PISCES simulations are identified with an experiment ID in the simulation workflow (abbreviated "expid"). These expids are used in the simulation post-processing scripts, and correspond to the simulations mentioned in the main manuscript as follows: a67o = REF (reference simulation) a683 = SAT (surface phytoplankton biomass toward satellite observations; intermediate restoring strength) a5xl = SAT_extreme (surface phytoplankton biomass toward satellite observations; high restoring strength, results reported in SM only)