Datasets, Codes, and Demos: paper of 'Observed large-scale and deep-reaching compound ocean state changes over the past 60 years'

Below are the codes to reproduce the results presented in the paper entitled 'Observed large-scale and deep-reaching compound ocean state changes over the past 60 years' by Tan et al. 2025 published in Nature Climate Change. https://www.nature.com/articles/s41558-025-02484-x

Author: Zhetao Tan (tanzhetao19@mails.ucas.ac.cn; zhetao.tan@lmd.ipsl.fr)

Contributors: Lijing Cheng, Karina von Schuckmann, Sabrina Speich, Laurent Bopp, Jiang Zhu

🎥 Dashboard: A dashboard is available here to enable the climate scientists and climate policy-makers to check the dynamic evolution (since 1985) and current state of the simultaneous change in the global ocean.

▶️ Video: an evolution of the changing compound CIDs (ToE and its exposure) from 1980 to 2023 can be reached here

♻️ More information and useful materials could be found on my personal webpage

1. Dataset: IAP Compound CIDs Dataset

1.1 Introduction

The IAP Ocean Compound Climatic impact-drivers (CIDs) Monitoring Dataset integrates global, three-dimensional observations of physical and biogeochemical variables to capture long-term, compound ocean state changes with global coverage (1-degree) from the surface to mesopelagic zone. This dataset, combining time-of-emergence (ToE) analyses and exposure metrics from 1960 to 2023 for the concurrent change of multiple Essential Climate Variables (ECVs) includes ocean temperature, salinity, dissolved oxygen, and surface pH, provides a robust foundation for assessing the interplay between individual and compound CIDs. This dataset, together with the analysis framework shown in the corresponding paper, can server as a science-policy interface tool and indicators to facilitate the integration of our understanding of oceanic environmental change with broader knowledge of the compound impacts on the ocean and human societies.

1.2 Data Information

• Format: netCDF

• Time: 1980 – onward

• Temporal resolution: Annual

• Spatial resolution: 1-degree

• Spatial coverage: Global ocean

• Vertical resolution: surface, epipelagic zone (0-200m), mesopelagic zone (200-1000m).

• Version: v0.1 (an interim version)

This dataset is stored at ./IAP_CompoundCID_dataset/. You can also download the dataset from IAP data protocol: http://www.ocean.iap.ac.cn/ftp/cheng/Compound_CIDs/

1.3 Video of the evolution of compound ocean state change

We also provide a dynamic view of the compound climatic impact-drivers (Temperature, salinity, dissolved oxygen, and surface pH) from 1980 to 2023, which are represented by the compound time-of-emergence and its exposure. You can find the evolution video in this folder ./dynamic_video/ or via http://www.ocean.iap.ac.cn/ftp/cheng/Compound_CIDs/dynamic_video/

This video is produced by using the codes in Section 2.3 and Section 2.4

We also provide a dashboard to show the dynamic evolution (since 1985) of compound CIDs here: http://www.ocean.iap.ac.cn/pages/dataV/dataV.html?navAnchor=dataV

2. Codes to reproduce the figures

2.1 System and software requirements

• MATLAB (>R2020a)
• Python >3.9
  • numpy (version: >= 1.24.3)
  • scipy (version: >= 1.10.1)
  • cartopy (version: >= 0.21.1)
  • matplotlib (version: >= 3.7.2)
  • seaborn (version: >= 0.12.2)

2.2 Codes for Fig. 1a-c (Time of emergence figures)

(1) Fig. 1A

Working folder: Codes_Fig1

Input data are stored in ./Codes_Fig1/ToE_data

Please run the following script plot_Fig1a.m with MATLAB.

(2) Fig. 1B

Working folder: Codes_Fig1

Input data are stored in ./Codes_Fig1/ToE_data

Please run the followings script plot_Fig1b.m with MATLAB.

(3) Fig. 1C

Working folder: Codes_Fig1

Input data are stored in ./Codes_Fig1/ToE_data

Simply run plot_Fig1c.m with MATLAB.

2.3 Codes for Fig. 2 (Time of emergence figures)

Working folder: Codes_Fig2

Input data are stored in ./Codes_Fig2/Demo_Fig2_compound_emergence.py/

You can also run the script Demo_Fig2_compound_emergence.ipynb with Jupyter Notebook under the installation of Python 3.8. This will have the same output as *.py file.

2.4 Codes for Fig. 3 (Exposure figures)

Working folder: Codes_Fig3

Input data are stored in ./Codes_Fig2/Demo_Fig3_exposure/

You can also run the script Demo_Fig3_exposure.ipynb with Jupyter Notebook under the installation of Python 3.8. This will have the same output as *.py file.

2.5 Codes for Fig. 5 (compound effects figures)

(1) Fig. 5A

Working folder: Codes_Fig5

Input data are stored in ./Codes_Fig5/Input_data/

Simply run plot_Fig5a.ipynb (with Jupyter Notebook) under the installation of Python 3.8.

(2) Fig. 5B

Working folder: Codes_Fig5

Input data are stored in ./Codes_Fig5/Input_data/

Please run the plot_Fig5b.ipynb (with Jupyter Notebook) under the installation of Python 3.8.

(3) Fig. 5C

Working folder: Codes_Fig5

Please run the following script plot_Fig5c_HighSea.py .

Note: The Box 1 Figure and Fig. 4 are schematic (conceptual illustrations). The codes for these two figures can be obtained upon request.

3. Demos

3.1. Demo for estimating the time of emergency (ToE) for individual climatic impact-drivers

Here, taking temperature emergence as an example, we provide the codes to calculate its individual ToE, as presented in the manuscript.

We used the IAPv3 temperature 1-degree gridded product (the analysis mean field) as the input data, which can be assessed via http://www.ocean.iap.ac.cn/ftp/cheng/CZ16_v3_IAP_Temperature_gridded_1month_netcdf/

Working folder: demo_individual_ToE

Input data are stored in ./demo_individual_ToE/Input_data

Some internal functions for used are available in ./demo_individual_ToE/functions

Running order

1. N01_cal_baseline.m: This is to calculate the 1960-1989 climatology (baseline). Here, the input data is the IAPv3 1-degree grid dataset from 1960 to 2023 (monthly netCDF format). **Due to GitHub's data storage limitation, you should download the netCDF data before running the code via http://www.ocean.iap.ac.cn/ftp/cheng/CZ16_v3_IAP_Temperature_gridded_1month_netcdf/
2. N02_anomaly_monthly.m: This is to calculate the anomaly field relative to 1960-1989 baseline
3. N03_gloal_avg_temp.m: This is to calculate the global temperature 3-D average and its time series
4. N04_global_singal_noise.m: This is to calculate the global average signal (G(t)) and noise
5. N05_local_singal_noise.m: This is to calculate the local signal and noise in each 1-degree box at each standard depth level following Hawkins et al., 2020
6. N06_ToE_calculate.m: This is to calculate the temperature ToE in each 1-degree box (each standard depth level, epipelagic zone, and mesopelagic zone). We also provide some codes for plotting the spatial ToE maps.
7. N07_ToE_depth_percentage.m: This is to calculate the global percentage of temperature emergence as a function of depth
8. N08_ToE_year_percentage_upper2001000.m: This is to calculate the global percentage of temperature emergence as a function of year

Note: If you don't download the full IAP dataset, you can run this demo from the Step 6.

Citation

Tan Z., K. von Schuckmann, S. Speich, L. Bopp, J. Zhu, L. Cheng*., 2025: Observed large-scale and deep-reaching compound ocean state changes over the past 60 years. Nature Climate Change.

License

These codes are licensed under the MIT License

Questions and feedbacks

If you have any questions/suggestions about this manuscript, or if you find some bugs in this demo, please feel free and do not hesitate to tell us via:

• Create an issue in the Github community
• Pull requests your debugged/improved codes in the Github community
• Send an email to us: tanzhetao19@mails.ucas.ac.cn or zhetao.tan@lmd.ipsl.fr
• More information could be found in Zhetao Tan's webpage: https://zqtzt.github.io/

Update logs

11/03/2024: The initialization to upload the reproduction and demo codes (version 0.1.0)

25/11/2025: Updated version (version 1.0.0)