Heat transport across the Antarctic Slope Front controlled by cross-slope salinity gradients

Please see the latest version:

https://doi.org/10.5281/zenodo.7651177

This release contains updates on analysis code and products.

• MITgcm_ASF-heat-ver2/newexp/: the Matlab scripts used to generate and run the MITgcm simulations
• MITgcm_ASF-heat-ver2/analysis/cross_slope/ and MITgcm_ASF-heat-ver2/analysis/plots/: the Matlab scripts used to analyze model output and make plots.
• exps_configuration.zip: the configurations of the MITgcm simulations.
• products_new-ver2.zip: the products calculated from MITgcm diagnostics, including 7-year means of all the model outputs, overturning streamfunctions, neutral density, shoreward heat transport, kinetic energy, temporal decomposition, etc. 
• ThicknessFlux_FreshShelf.zip: products of isopycnal thickness flux, used to calculate the decomposition of eddy/tidal heat advection/diffusion, for the "fresh-shelf" simulation. 
• ThicknessFlux_ref.zip: as above, but for the reference simulation.
• ThicknessFlux_DenseShelf.zip: as above, but for the "dense-shelf" simulation. 

The source code of the Massachusetts Institute of Technology General Circulation Model (MITgcm) is available at: http://mitgcm.org.

All the raw data of the model output are available at: https://doi.org/10.15144/S47P49.

To reproduce MITgcm_ASF simulations: 

1. Start each simulation with a 20-year spin-up integration. Before running each simulation, you need to substitute &OBCS_PARM04 with &OBCS_PARM05 in the file input/data.obcs, and substitute &EXF_NML_05 with &EXF_NML_OBCS  in the file input/data.exf. For simulations with very fresh shelf waters (e.g., shelf salinity = 33 psu), you need to spin up the simulation with a very small time step (e.g., 60s) for ~ two months, and then use a larger time step. 

2. Initialize the production run from the corresponding spin-up run, using the Matlab script initialize.m in the folder MITgcm_ASF-heat-ver2/newexp/. When using the LAYERS package, you need to substitute numperlist = 1 with numperlist = 2 in the file code/DIAGNOSTICS_SIZE.h before running the simulations.

Notes on calculationg the overturning streamfunction and its mean/eddy/tidal decomposition using the MITgcm LAYERS package: 

• avg_t: Calculate time averages. It has been modified since the vertical number of layers can be different from Nr. 
• calc_Overturning_pt, usscar_plot_overturning_pt: calculate and plot eddy/mean/isopycnal overturning streamfunction using potential temperature layer fluxes.

• calc_Overturning_rho, usscar_plot_overturning_rho: calculate and plot eddy/mean/isopycnal overturning streamfunction using potential density layer fluxes.

• calc_Overturning_pt_Aocean, usscar_pt_overturning_rho_Aocean (recommended if your bathymetry is not flat): calculate and plot eddy/mean/isopycnal overturning streamfunction using potential temperature layer fluxes. For each latitude, use the total ocean area below a certain level to interpolate the streamfunction from pt space to z space. 

• calc_Overturning_rho_Aocean, usscar_plot_overturning_rho_Aocean (recommended if your bathymetry is not flat): calculate and plot eddy/mean/isopycnal overturning streamfunction using potential density layer fluxes. For each latitude, use the total ocean area below a certain level to interpolate the streamfunction from potential density space to z space.

• calc_decomposition_OT, plot_OT_rho_Aocean_TidalEddyMean: decompose the isopycnal overturning streamfunction into tidal/eddy/mean components, using potential density layer fluxes.

Feel free to contact Yidongfang Si via ysi@g.ucla.edu if you have any questions.