This repository contains codes and processed datasets for a manuscript entitled "Oceanic heat delivery to the Antarctic continental shelf: Large-scale, low-frequency variability" (2018), by A. Palóczy, S. T. Gille and J. L. McClean (Journal of Geophysical Research – Oceans). This Jupyter notebook provides an overview of the contents.

The directory plot_figs/ contains the Python codes used to produce all figures in the main manuscript (Figures 1-10) and Figures S3-S10 and S1 in the Supplementary Materials. The codes depend on the data files in the data_reproduce_figs/ directory. Some of these are too large to be included in this repository, but are available for download from the links listed on the accompanying README files. Please contact André Palóczy if you have issues downloading the files.

• The contributions from eddy advection, eddy stirring and mean flow advection to the total onshore heat transport vary regionally.
• The time-mean component governs the seasonal variability of the total heat transport and largely cancels the eddy component.
• Circumpolar heat transports and total heat content of the Antarctic continental margin follow SAM, but ENSO prevails in West Antarctica.

Onshore penetration of oceanic water across the Antarctic continental slope (ACS) plays a major role in global sea level rise by delivering heat to the Antarctic marginal seas, thus contributing to the basal melting of ice shelves. Here, the time-mean ($\Phi^\text{mean}$) and eddy ($\Phi^\text{eddy}$) components of the heat transport ($\Phi$) across the 1000 m isobath along the entire ACS are investigated using a 0.1\textdegree global coupled ocean/sea ice simulation based on the Los Alamos Parallel Ocean Program (POP) and sea ice (CICE) models. Comparison with \textit{in situ} hydrography shows that the model successfully represents the basic water mass structure, with a warm bias in the Circumpolar Deep Water layer. Segments of on-shelf $\Phi$, with lengths of O(100-1000 km), are found along the ACS. The circumpolar integral of the annually-averaged $\Phi$ is O(20 TW), with $\Phi^\text{eddy}$ always on-shelf, while $\Phi^\text{mean}$ fluctuates between on-shelf and off-shelf. Stirring along isoneutral surfaces is often the dominant process by which eddies transport heat across the ACS, but advection of heat by both mean flow-topography interactions and eddies can also be significant depending on the along- and across-slope location. The seasonal and interannual variability of the circumpolarly-integrated $\Phi^\text{mean}$ is controlled by convergence of Ekman transport within the ACS. Prominent warming features at the bottom of the continental shelf (consistent with observed temperature trends) are found both during high-SAM and high-Niño 3.4 periods, suggesting that climate modes can modulate the heat transfer from the Southern Ocean to the ACS across the entire Antarctic margin.

In the Southern Ocean, warm Circumpolar Deep Water (CDW) is carried by the Antarctic Circumpolar Current (ACC), which flows in the open ocean usually far from cooler Antarctic coastal waters. When the ACC approaches the Antarctic coast, CDW can supply heat to the ice shelves. We use a realistic computer model of the global ocean and sea ice to study these processes. The model indicates that winds, ocean eddies and current interactions with the seafloor all contribute to moving heat toward Antarctica. Seasonal changes in the winds around Antarctica affect the heat transport near the surface and due to current-seafloor interactions, but not so much the heat transport due to eddies. Over multi-year periods, some of the changes in the heat transport are related to climate variability taking place in the tropics and around Antarctica. Both the processes responsible for bringing heat to the Antarctic coast and the variability of this heat delivery represent knowledge needed to improve computer model simulations of the melting of the Antarctic ice cap. In turn, improving these simulations is likely to reduce uncertainties in projections of sea level rise, allowing for the development of adaptation and mitigation policies that better address global societal impacts.

• André Palóczy (apaloczy@ucsd.edu)
• Sarah T. Gille
• Julie L. McClean

This work was funded by the U.S. Department of Energy (DOE, grant # DE–SC0014440) in the scope of the "Ocean and Sea Ice and their Interactions around Greenland and the West Antarctic Peninsula in Forced Fine–Resolution Global Simulations" project and high-performance computing support from Yellowstone (ark:/85065/d7wd3xhc) provided by National Center for Atmospheric Research (NCAR)'s Climate Simulation Laboratory (CSL), sponsored by the National Science Foundation (NSF) and other agencies. J.L. McClean was supported by the U.S. DOE Office of Science grant entitled "Ultra-High Resolution Global Climate Simulation" via a Los Alamos National Laboratory subcontract to carry out the POP/CICE simulation. The analyses of the model output performed in this study were enabled by computing resources provided by Oak Ridge Leadership Computing Facility (OLCF).