A mathematical model (Bennetts et. al, 2022, doi.org/10.1029/2022GL100868) is used to analyse the ocean wave transfer to the Ross Ice Shelf flexure. The results showed significant impact of shelf geometry and wave period on flexural strain. The presented study further investigates the impact of geometrical variations on shelf flexure by doing a case study on the Larsen C Ice Shelf. The analysis also extended to multiple ice shelves to providing statistical relationships between shelf flexure and ice shelf thickness and seabed variation across the range of wave periods (swell, infra-gravity, and extremely long period waves). The geometry data of ice shelves are collected from the Bedmap2 dataset (Fretwell et al, 2013, doi.org/10.5194/tc-7-375-2013) and feed into our model to calculate the corresponding flexural strain and stored in our data files. Our dataset consists of MAT-files which require MATLAB to read the data. The folder includes a data folder and some MATLAB functions to produce the figures as shown in the paper (Jie et. al, (submitted 2023)). The data folder consists 7 MAT files named "Fig_number" where each file is for generating the corresponding figure. Data file named "3km_divided_thick_updated" includes all the input/output data collected of 15 ice shelves for statistical analysis. We also made a comparison between BedMachine3 and BEDMAP2 (BedMachine3 (Morlighem, Mathieu, et al. 2020, https://www.nature.com/articles/s41561-019-0510-8) is another dataset containing Antarctic geometry data). The data "Larsen_bedmachine3" and "Larsen_bedmap2" include the outputs (flexure strain) of the Larsen C Ice Shelf using the geometry from BedMachine3 and BEDMAP2 and "Larsen_map_Bedmachine" and "Map_Larsen_c" and are for producing the maps of Larsen C from BedMachine3 and BEDMAP2 respectively. Eight MATLAB scripts are included to recreate results from Jie et al (submitted). "Fig_1_model" is for producing a geometry of Larsen C using BEDMAP2, "Fig_2_3_varying_geometry" is for geometry variation analysis for Larsen C, "Fig_4_strain_profile" is to show the strain for all the transects covering Larsen C, "Fig_5_Box_plot_with_period_Outliners" is for producing box-plot of strain with outliers for Larsen C, "Fig_6_Box_plot_with_period_all_shelf" is for producing median strain over wave periods for 15 ice shelves, "Fig_7_8_Box_plot_with_shelves_correlation_log" and "Fig_9_correlation_and_periods" are showing the correlation of strain and shelf front thickness / water cavity depth. "Fig_10_Box_plot_with_period_Outliners" is showing a comparison of BedMachine3 and BEDMAP2 using Larsen C as an example.

These data are contained within a netcdf file of the multi-model mean (MMM) calculated as part of the Realistic ice-shelf/ocean state estimates (RISE) project, with the following variables calculated on a 2 kilometer grid, from the ten contributing models: - longitude degrees east EPSG:4326 - latitude: degrees north EPSG:4326 - easting: meters east EPSG:3031 - northing: meters_north EPSG:3031 - mask: grounded=1,iceshelf=2,conshelf=3,ocean=4 - iceshelf_id: NSIDC iceshelf-id - h: Depth (m) - zice: Ice draft depth (m) - ismr: Average basal iceshelf melt (m/year) - salt_bar: Depth averaged practical salinity (psu) - salt_zice: Average practical salinity (icedraft) (psu) - salt_zice_sa: Average absolute salinity (icedraft) (g/kg) - temp_bar: Average potential temperature - \"theta\" (water column) (degrees C) - temp_tw_zice: Average in-situ temperature (icedraft) (degrees C) - temp_tf_zice: Average in-situ temperature which seawater freezes (icedraft) (degrees C) - tstar_zice: Average thermal driving (degrees C) - u_bar: Average East-west velocity (u) ocean current speed (m/s) - v_bar: Average North-south velocity (v) ocean current speed (m/s) - u_zice: Average East-west velocity (u) ocean current speed (icedraft) (m/s) - v_zice: Average North-south velocity (v) ocean current speed (icedraft) (m/s) - rho_zice: Average in-situ seawater density (icedraft) (kg/m3) - ustar_zice: Average ice-water friction velocity Contextual information taken from the abstract of the referenced paper: Societal adaptation to rising sea levels requires robust projections of the Antarctic Ice Sheet’s retreat, particularly due to ocean-driven basal melting of its fringing ice shelves. Recent advances in ocean models that simulate ice-shelf melting offer an opportunity to reduce uncertainties in ice–ocean interactions. Here, we compare several community-contributed, circum-Antarctic ocean simulations to highlight inter-model differences, evaluate agreement with satellite-derived melt rates, and examine underlying physical processes. All but one simulation use a melting formulation depending on both thermal driving (T ⋆) and friction velocity (u⋆), which together represent the thermal and ocean current forcings at the ice–ocean interface. Simulated melt rates range from 650 to 1277 Gt year−1 (m = 0.45 − 0.91 m year−1), driven by variations in model resolution, parameterisations, and sub-ice shelf circulation. Freeze-to-melt ratios span 0.30 to 30.12 %, indicating large differences in how refreezing is represented. The multi-model mean (MMM) produces an averaged melt rate of 0.60 m year−1 from a net mass loss of 842.99 Gt year−1 (876.03 Gt year−1 melting and 33.05 Gt year−1 refreezing), yielding a freeze-to-melt ratio of 3.92 %. We define a thermo-kinematic melt sensitivity, ζ = m/(T ⋆ u⋆) = 4.82 × 10−5 °C−1 for the MMM, with individual models spanning 2.85 × 10−5 to 19.4 × 10−5 °C−1. Higher melt rates typically occur near grounding zones where both T ⋆ and u⋆ exert roughly equal influence. Because friction velocity is critical for turbulent heat exchange, ice-shelf melting must be characterised by both ocean energetics and thermal forcing. Further work to standardise model setups and evaluation of results against in situ observations and satellite data will be essential for increasing model accuracy, reducing uncertainties, to improve our understanding of ice-shelf–ocean interactions and refine sea-level rise predictions.

Metadata record for data from ASAC Project 2504 See the link below for public details on this project. In this project a sea-ice model for application in Southern Ocean climate and forecasting studies will be developed to amend identified deficiencies in numerical models (i.e. unaccounted short-term dynamics; or non-suitable ice rheology). In-situ deformation and ice-stress data will be used to derive parameterisations suitable for the Southern Ocean pack. Antarctic sea ice is an important component of the Southern Hemisphere climate. It provides a habitat for algae, plankton and for larger species such as mammals or penguins. It is a transport medium for freshwater and biological matter. On the other hand it acts like a barrier between ocean and atmosphere in regard to the exchange of thermal energy, water vapour and gases. Sea ice affects the polar climate in many ways: E.g., by effectively insulating the ocean from the colder atmosphere the sea ice enables an advection of relatively warm water onto the shallow Antarctic continental shelf. This warmer water is then available to interact with other components of the climate system, such as by basal melting of the continental ice shelves [Jenkins and Holland, 2002]. Also, due to its high albedo, the sea ice has a large-scale effect on the net incoming solar radiation [Ebert et al., 1995] and reduces the absorption of solar energy into the upper ocean. The thermodynamic growth of seaice and the consequent desalination of the ice gives rise to a transport of salt from the ice into the ocean, which increases the water density over the shelf, thereby driving the deep vertical overturning cell in the global ocean circulation. High ice-growth rates (e.g., in regions of polynyas) are generally concentrated in small areas in shallow waters. These regions are often insufficiently resolved or even unresolved in coupled climate models, which are generally configured to run at a spatial resolution of 2 degree longitude by 1 degree latitude or coarser [Zhang and Hunke, 2001]. The specific objectives of this project are to: identify the variabilities in the sea-ice characteristics and the underlying physical processes; identify the time scales, at which the sea ice interacts with the ocean and atmosphere; assess the contribution of sub-daily ice motion and deformation due to tidal forcing and inertial response to changes within the Antarctic ocean-ice-atmosphere system; derive the impact of sub-daily ice dynamics on the sea-ice area, extent and mass on interannual and decadal time scales; determine the scale effect of dynamic processes on the accuracy of modelled sea-ice parameters using a global high-resolution model; identify model uncertainties through comprehensive validation studies. However, logistical problems prevented the project from collecting any data in the field. To overcome the paucity of planned buoy data we used the following data sets to address some of the aspects of the original proposal: 1) Sea-ice buoy data: ISPOL 2004: See AAS #2500 for metadata. 2) Numerical investigations: We have investigated the failure of sea ice using an isotropic model [Hibler, 1979], where ice strength is modelled as a random variable in the model space. In situ weakening was prescribed by a fracture-based Coulombic rheology [Hibler and Schulson, 2000]. We realised this by parameterising weakening with an ice-strength parameter of 1000 and initialising the ice strength across the model grid by random. The simulations were run over a 2000 km by 2000 km region and forced, from rest, with an idealised wind field. We analysed the sensitivity of failure to ice strength and wind stress as well as the intersection angle of the wind stress, and conducted idealised 2D failure experiments.

This dataset contains numerical simulation results of the wave fields in the Davis Sea from end of December 2019 to start of February 2020. Hindcasts were obtained through the third-generation spectral wave model WAVEWATCH-III (hereafter WW3). A high resolution Davis Sea regional grid (resolution 0.1 degree, 60-80E longitude, 70-60S latitude) was nested into global grid domain (resolution 0.5 degree, 80S-80N latitude). The global model is forced with 0.5 degree sea ice concentration and 10m-wind fields from ECMWF's ERA5 reanalysis. The Davis sea model is forced with 0.1 degree 10m-wind fields from ECMWF's archived forecasts, and high-resolution (3.125km) AMSR2 satellite data for sea ice concentration (Beitsch et al., 2013 updated). Ice-induced wave attenuation is parameterized following Sutherland et al. (2019, doi:10.1016/j.apor.2019.03.023) whilst the break-up of sea ice is parameterized as 'broken' or 'unbroken' based on the break-up parameter of Voermans et al.(2020, doi:10.5194/tc-14-4265-2020). The numerical simulations have been calibrated using the buoy-observations of Voermans (2022, dataset, doi:10.26179/cdmx-n995). Sensitivity of the simulations to sea ice properties was tested and all results are provided in the dataset. The data tree: * global: model outputs for the global domain - ncfield: gridded wave and ice data for this domain in netCDF-4 format - nests: binary data used by WW3 for boundary conditions for the Davis Sea grid - restarts: binary data used by WW3 for restarting this domain * davis_sea: model outputs for the Davis Sea domain - ncfield: gridded wave and ice data for this domain in netCDF-4 format - ncpoint: spectral wave data for a few points in the Davis Sea in netCDF-4 format - nctrack: spectral wave data following the wave buoys of Voermans et al (2022) in the Davis Sea in netCDF-4 format - restarts: binary data used by WW3 for restarting this domain - IHOT: binary text field of broken and unbroken ice for restarting this domain File naming convention (by example): ww3.20200101_20200103_M3D_IHOT_H0P0325_A0P01_YY9P0_SS0P1_HH0P55.nc * 20200101_20200103 identifies the datespan of the simulation in YYYYMMDD format * A0P01 refers to the attenuation coefficient of the model (where P stands for 'point'), in this case, A=0.01 * YY is the Young's Modulus timed 10^9, here, Y-9.0e9 Pa * SS is the ice strength 'sigma' times 10^6, here sigma=0.1e6 * HH is the ice thickness, here h=0.55 m * H0P0325 is proportional to the epsilon calibration coefficient (H=0.5ice_thicknessepsilon). * M3D refers to the 3rd instantiation of the model * IHOT refers to hot start using the ice breakup field from the previous week. ww3.*_M3D_IHOT_H0P065_A0P05_YY6P0_SS0P55.nc is considered the baseline file (note, this simulation only covers the first two weeks of the study period). Reference: Beitsch, A., Kaleschke, L. and Kern, S. (2013). "AMSR2 ASI 3.125 km Sea Ice Concentration Data, V0.1", Institute of Oceanography, University of Hamburg, Germany, digital media

A Langrangian free drift model is developed, including a term for geostrophic currents that reproduces the 13 h period signature in the ice motion observed in the data (CLSC_WIIOS_2017; parent data). The calibrated model is shown to provide accurate predictions of the ice drift for up to 2 days, and the calibrated parameters provide estimates of wind and ocean drag for pancake floes under storm conditions. Model setup is described in "Drift of pancake ice floes in the winter Antarctic marginal ice zone during polar cyclones", Alberello et. al [https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2019JC015418; pre-print https://arxiv.org/pdf/1906.10839.pdf]. The dataset includes model data. Six model outputs are included. (i) "full_t00" includes the full 10 days simulation, with all the forcing switched on (ii) "noge_t00" includes the full 10 days simulation, but the geostrophic current is suppressed (iii) "full_t25_noup" includes the simulation with start at 2.5 days, all the forcing switched on, no update of the drag coefficients (iv) "full_t25_newn" includes the simulation with start at 2.5 days, all the forcing switched on, the drag coefficients are recalibrated (v) "full_t50_noup" includes the simulation with start at 5 days, all the forcing switched on, no update of the drag coefficients (vi) "full_t50_newn" includes the simulation with start at 5 days, all the forcing switched on, the drag coefficients are recalibrated In each file: - rho_a the air density (1.3 kg/m3) - rho_w the water density (1028 kg/m3) - rho_i the ice density (910kg/m3) - C_w the water drag coefficient (calibrated) - C_a the air drag coefficient (calibrated) - turn the turning angle (25 degrees) - Nansen the Nansen number evaluated using C_a and C_w - aalpha a model parameter (proportional to air and ice parameters) - abeta a model parameter (proportional to water and ice parameters) - ag amplitude of the geostrophic current (U_g=0.125m/s) - tg initial phase of the geostrophic current (in radians) - to start time (in matlab format, use "datestr(to)" ), after which model resolution is 60 seconds - wo components of wind in the East and North direction (m/s) - wi components of wind in the East and North direction (m/s) - uo components of modelled ice drift speed in the East and North direction (m/s) - lo longitude and latitude of modelled ice position (degrees) - xo position of modelled ice in the East and North direction (m), given with respect to the initial position (0,0) - wco components in the East and North direction of geostrophic current (m/s)

Although the floating sea ice surrounding the Antarctic damps ocean waves, they may still be detected hundreds of kilometres from the ice edge. Over this distance the waves leave an imprint of broken ice, which is susceptible to winds, currents, and lateral melting. The important omission of wave-ice interactions in ice/ocean models is now being addressed, which has prompted campaigns for experimental data. These exciting developments must be matched by innovative modelling techniques to create a true representation of the phenomenon that will enhance forecasting capabilities. This metadata record details laboratory wave basin experiments that were conducted to determine: (i) the wave induced motion of an isolated wooden floe; (ii) the proportion of wave energy transmitted by an array of 40 floes; and (iii) the proportion of wave energy transmitted by an array of 80 floes. Monochromatic incident waves were used, with different wave periods and wave amplitudes. The dataset provides: (i) response amplitude operators for the rigid-body motions of the isolated floe; and (ii) transmission coefficients for the multiple-floe arrays, extracted from raw experimental data using spectral methods. The dataset also contains codes required to produce theoretical predictions for comparison with the experimental data. The models are based on linear potential flow theory. These data models were developed to be applicable to Southern Ocean conditions.

A numerical model of ocean wave interactions with Antarctic sea ice cover, including: (i) attenuation of wave energy due to the ice cover (based on the empirical model of Meylan, Bennetts, Kohout, 2014, Geophys Res Lett, doi:10.1002/2014GL060809); and (ii) breakup of the ice cover into smaller floes due to strains imposed by wave motion (based on the theory of Williams et al, 2013, Ocean Model., doi:10.1016/j.ocemod.2013.05.010). The model is coded in FORTRAN90 for use as a module in a standalone version of the CICEv4.1 sea ice model (http://oceans11.lanl.gov/trac/CICE). It requires incident wave forcing to be specified at some constant latitude outside the ice cover, which can be user chosen or imported from data files (e.g. data given by Wavewatch III hindcasts, see http://doi.org/10.4225/08/523168703DCC5). Modifications to the existing CICE routines are given to allow integration of the broken floe sizes into its lateral melting scheme, and for incorporation of a floe bonding scheme. Bennetts, O'Farrell and Uotila (submitted) use the model to study the impact of wave-induced ice breakup on model predictions of the concentration and volume of Antarctic sea ice.

Direct Numerical Simulation (DNS) was used to study the effect of sloping the ice-shelves on the dissolution/melt rate at the ice-ocean interface. The simulations were done on the HPC Raijin at NCI, Canberra over March 2015 to June 2017. Numerical experiments were carried out over a range of slope angle (5 degrees – 90 degrees) of the ice-shelves measured from the horizon. Turbulent flow field is simulated over the domain length of 1.8 m, (for slope angle greater than or equal to 50 degrees) and 20 m (for slope angle less than or equal to 20 degrees) respectively; the flow-field is laminar otherwise. A constant ambient temperature 2.3 degrees C and salinity 35 psu is maintained throughout the simulations. The DNS successfully resolved all possible turbulence length scales and relative contributions of diffusive and turbulent heat transfer into the ice wall is measured. Data available: Excel file Profile_salinity_temperature_velocity.xlsx contains along-slope velocity, temperature and salinity as a function of wall normal distance for slope angle 50 degrees, 65 degrees and 90 degrees respectively for the domain length 1.8 m.

The AA4528 corridor dataset contains the Matlab scripts for the corridor algorithm, ice shelf locations and file extensions. The corridor algorithm is designed to calculate the parts of the ocean which can directly propagate swell into an exposed ice shelf. The algorithm achieves this as an expansion of the coastal exposure algorithm (Reid and Massom, 2021), with the details of the inner working of the algorithm work presented in the paper attached with this dataset. Corridors can be used to calculate the frequency of swell reaching an ice shelf per year and can be combined with hindcasts to extract relevant wave data to an ice shelf for modelling or data analysis purposes. The corridor algorithm requires sea ice concentration data, which was provided by the NSIDC Sea ice concentrations from the Nimbus-7 SMMR and DMSP SSM/I-SSMIS Passive Microwave Data, Version 1 (https://nsidc.org/data/nsidc-0051). Ice shelf coordinates were extracted from the gfsc_25s.msk that come with the sea ice data, with the aid of Antarctic Mapping Toolbox (Greene et al., 2017), and were attached separately to make editing more consistent. As this is designed to use daily sea ice data from the 1st of January 1979 onwards, I’ve also attached the sea ice files for the off-days when the sea-ice data was taken every 2nd day. Th file extensions script was also included to be able to switch through off-day files and changes that occur with the NSIDC file format. The ocean hindcast that the corridor algorithm was built around is the CAWCR Wave Hindcast – Aggregated Collection (https://data.csiro.au/collections/collection/CI39819v005). The corridor algorithm uses daily data to make it consistent with the sea ice data and calculated the maximum significant wave height for each cell present in the hindcast. Data that was extracted from it was the maximum daily significant wave height recorded in the corridor and the direction of that cell. Data was taken from 01/09/1979 to 31/08/2019 giving 40 years of data which accounts for seasonality of corridors. The excel spreadsheet attached contains relevant corridor data for each ice shelf with an area greater than 500 km^2. Area was determined by either the supplementary files from Rignot et. al., 2013, or ice shelf areas from the Antarctic mapping toolbox (Greene et al., 2017). Angle1 and Angle2 were the ones used in the direction filter, and there should be a comment in the filter with how it handles if Angle 1 is greater than Angle 2 or vice versa. Ac is the corridor area, PA is potential corridor area (i.e. the absolute max it could be with the settings we used, Ac_max is the maximum corridor area, D_cor is the days that corridors were present, Hs is significant wave height and LW (large waves) is counting days per year when significant wave heights greater than or equal to 6 m (Morim et al., 2021). Refs: Greene, C. A., Gwyther, D. E. and Blankenship, D. D. (2017) ‘Antarctic Mapping Tools for MATLAB’, Computers and Geosciences, 104, pp. 151–157. doi: 10.1016/j.cageo.2016.08.003. Morim, J. et al. (2021) ‘Global-scale changes to extreme ocean wave events due to anthropogenic warming’, Environmental Research Letters, 16(7), p. 074056. doi: 10.1088/1748-9326/ac1013. Reid, P. and Massom, R. (2021) ‘Change and Variability in Antarctic Coastal Exposure , 1979-2020’. In pre-print (https://assets.researchsquare.com/files/rs-636839/v1/02002d0b-2c6c-402b-8e14-7f77075d8f90.pdf?c=1631885736) Rignot, E. et al. (2013) ‘Ice-shelf melting around antarctica’, Science, 341(6143), pp. 266–270. doi: 10.1126/science.1235798.

Direct Numerical Simulation (DNS) was used to study the effect of sloping the ice-shelves on the dissolution/melt rate at the ice-ocean interface. The simulations were done on the HPC Raijin at NCI, Canberra over March 2015 to June 2017. Numerical experiments were carried out over a range of slope angle (5 degrees – 90 degrees) of the ice-shelves measured from the horizon. Turbulent flow field is simulated over the domain length of 1.8 m, (for slope angle greater than or equal to 50 degrees) and 20 m (for slope angle less than or equal to 20 degrees) respectively; the flow-field is laminar otherwise. A constant ambient temperature 2.3 degrees C and salinity 35 psu is maintained throughout the simulations. The DNS successfully resolved all possible turbulence length scales and relative contributions of diffusive and turbulent heat transfer into the ice wall is measured. Data available: Excel file Meltrate_vs_slopeangle_lam_turb.xlsx contains both simulated laminar and turbulent dissolution/melt rate as a function of slope angle along with their analytical values based on laminar and turbulent scaling theory respectively.