ACC Standing Meander SST Trends

Overview

Python analysis scripts and derived data products for the manuscript:

Liu, X. (2026). Topographic Hotspots Shape the Spatial Pattern of Southern Ocean SST Trends: Satellite Evidence from ACC Standing Meanders, 2026.

This study co-locates 33 years (1993–2025) of satellite sea surface temperature (SST) trends with structural trends of Antarctic Circumpolar Current (ACC) standing meanders at four major topographic hotspots. We demonstrate that SST trends at meander sites differ significantly from quiescent ACC sections, with meander sites warming 0.047 °C per decade less than control regions (D = 0.181, p < 0.001).

Study Sites

Site	Abbreviation	Longitude Range	Dynamical Regime
Campbell Plateau	CP	150°E–150°W	Plateau-controlled
Pacific-Antarctic Ridge	PAR	150°W–80°W	Ridge-controlled
Southeast Indian Ridge	SEIR	130°E–152°E	Ridge-controlled
Southwest Indian Ridge	SWIR	15°E–45°E	Ridge-controlled

Repository Structure

acc-meander-sst-trends/
├── README.md
├── LICENSE
├── .gitignore
├── environment.yml
├── project_config.json
│
├── scripts/                    # Analysis pipeline (numbered by execution order)
│   ├── setup_meander_sst_project.py   # Project directory setup and mask generation
│   ├── download_cmems_sst.py          # CMEMS OSTIA SST data download
│   ├── NB03b_spatial_eke.py           # Spatial EKE computation per site
│   ├── NB12_regrid_sst.py             # SST regridding (0.05° → 0.125°)
│   ├── NB13_sst_trends.py             # Per-grid-point SST trend computation
│   ├── NB14_fix_1995_and_par_transect.py  # 1995 fix + along-ACC transect
│   ├── NB15_manuscript_figures.py     # Main-text figure generation
│   ├── NB16_fix_1995_complete.py      # 1995 SST re-download and reprocessing
│   ├── NB17_supporting_information.py # Supporting Information outputs
│   └── verify_step01.sh               # Infrastructure verification
│
├── pbs/                        # NCI Gadi PBS job submission scripts
│   ├── download_sst.pbs
│   ├── run_spatial_eke.pbs
│   ├── run_regrid.pbs
│   ├── run_sst_trends.pbs
│   └── run_fix.pbs
│
├── data/                       # Derived data products (masks, trends, statistics)
│   ├── masks/
│   │   ├── meander_envelope_mask_CP.nc
│   │   ├── meander_envelope_mask_PAR.nc
│   │   ├── meander_envelope_mask_SEIR.nc
│   │   ├── meander_envelope_mask_SWIR.nc
│   │   ├── control_mask_CTRL_SE_PAC.nc
│   │   ├── control_mask_CTRL_S_IND.nc
│   │   ├── control_mask_CTRL_S_ATL.nc
│   │   ├── control_mask_CTRL_CP_SEIR.nc
│   │   └── circumpolar_meander_envelope.nc
│   │
│   ├── eke_trends/
│   │   ├── spatial_eke_CP_trend.nc
│   │   ├── spatial_eke_PAR_trend.nc
│   │   ├── spatial_eke_SEIR_trend.nc
│   │   └── spatial_eke_SWIR_trend.nc
│   │
│   ├── sst_trends/
│   │   ├── sst_trend_map.nc
│   │   └── sst_trend_along_acc_transect.nc
│   │
│   └── statistics/
│       ├── sst_trend_stats.csv
│       ├── ks_test_meander_vs_control.csv
│       ├── sst_eke_correlation.csv
│       ├── along_acc_site_anomalies.csv
│       ├── decadal_decomposition.csv
│       ├── table_s1_control_regions.csv
│       └── table_s2_detailed_stats.csv
│
├── manuscript/
│   └── table_s1_control_regions.tex
│
└── logs/                       # PBS job output logs (for reproducibility)
    ├── spatial_eke.out
    ├── spatial_eke.err
    ├── regrid_sst.out
    ├── regrid_sst.err
    ├── sst_trends.out
    ├── sst_trends.err
    ├── fix_1995_par.out
    └── fix_1995_par.err

Analysis Pipeline

The analysis proceeds in the following order:

Step	Script	Description	PBS Script
0	verify_step01.sh	Verify NCI Gadi infrastructure	—
1	setup_meander_sst_project.py	Create directory structure, meander envelope masks, control region masks	—
2	download_cmems_sst.py	Download CMEMS OSTIA SST (0.05°, daily, 1993–2025)	download_sst.pbs
3	NB03b_spatial_eke.py	Compute per-grid-point monthly EKE and Sen's slope trends	run_spatial_eke.pbs
4	NB12_regrid_sst.py	Regrid SST from 0.05° to 0.125° monthly	run_regrid.pbs
5	NB13_sst_trends.py	Compute per-grid-point SST trends (Sen's slope + Mann-Kendall)	run_sst_trends.pbs
6	NB14_fix_1995_and_par_transect.py	Recompute with all 33 years + corrected along-ACC transect	run_fix.pbs
7	NB15_manuscript_figures.py	Generate Figures 1–3 for the manuscript	—
8	NB17_supporting_information.py	Generate Supporting Information tables and Figure S1	—

Data Sources

Dataset	Resolution	Period	Source
CMEMS DUACS L4 ADT	0.125°, daily	1993–2025	DOI: 10.48670/moi-00148
CMEMS OSTIA SST	0.05°, daily	1993–2025	DOI: 10.48670/moi-00168
GMRT Bathymetry	Variable	—	GMRT

Requirements

• Python 3.11+
• Key packages: xarray, numpy, scipy, matplotlib, cartopy, pymannkendall, statsmodels, pandas
• NCI Gadi account (for PBS job submission)
• copernicusmarine CLI tool (for SST data download)

Installation

git clone https://github.com/xinlongliu0307/acc-meander-sst-trends.git
cd acc-meander-sst-trends
conda env create -f environment.yml
conda activate meander-sst

Key Results

• Meander–control SST trend difference: −0.047 °C/decade (D = 0.181, p < 0.001)
• PAR meander envelope: −0.008 °C/decade (net cooling; 13.4% significant)
• CP meander envelope: +0.217 °C/decade (strongest warming; 84.4% significant)
• EKE–SST sign reversal: CP (R = +0.104) vs. PAR (R = −0.082)

Citation

If you use these scripts or data products, please cite:

@article{liu2026meander_sst,
  title={Topographic Hotspots Shape the Spatial Pattern of Southern Ocean SST Trends: Satellite Evidence from ACC Standing Meanders},
  author={Liu, Xinlong},
  year={2026},
  publisher={Wiley Online Library}
}

License

This project is licensed under the MIT License — see LICENSE for details.

Acknowledgments

Computational resources were provided by the Australian National Computational Infrastructure (NCI). This study used E.U. Copernicus Marine Service Information. XL acknowledges support from the University of Tasmania and the Australian Antarctic Program Partnership.