• Energy Research

# Dataset: Subglacial freshwater driven speedup of East Antarctic outlet glacier retreat [https://doi.org/10.5061/dryad.1vhhmgr0b](https://doi.org/10.5061/dryad.1vhhmgr0b) Journal: Journal of Geophysical Research: Earth Surface Principle Investigator: * Tyler Pelle, Scripps Institution of Oceanography, University of California San Diego, [tpelle@ucsd.edu](mailto:tpelle@ucsd.edu) Co-Authors: * Dr. Jamin Greenbaum, Scripps Institution of Oceanography, University of California San Diego * Dr. Shivani Ehrenfeucht, Department of Geography and Environmental Management, University of Waterloo * Prof. Christine Dow, Department of Geography and Environmental Management, University of Waterloo * Dr. Felicity S. McCormack, Securing Antarctica's Environmental Future, School of Earth, Atmosphere, & Environment, Monash University Created on October 4, 2023 ## Description of the data and file structure ### File description: 1. runme.m - MATLAB script used to run coupled ISSM-GlaDS SSP5-8.5_{F,M} simulation - includes melt rate parameterization. 2. ssp585.mat – Yearly ice sheet model output from 2017-2100 for SSP5-8.5 simulation. 3. ssp585_F.mat – Yearly ice sheet model output from 2017-2100 for SSP5-8.5_{F} simulation. 4. ssp585_M.mat – Yearly ice sheet model output from 2017-2100 for SSP5-8.5_{M} simulation. 5. ssp585_FM.mat – Yearly ice sheet model output from 2017-2100 for SSP5-8.5_{F,M} simulation. 6. ssp126.mat – Yearly ice sheet model output from 2017-2100 for SSP1-2.6 simulation. 7. ssp126_F.mat – Yearly ice sheet model output from 2017-2100 for SSP1-2.6_{F} simulation. 8. ssp126_M.mat – Yearly ice sheet model output from 2017-2100 for SSP1-2.6_{M} simulation. 9. ssp126_FM.mat – Yearly ice sheet model output from 2017-2100 for SSP1-2.6_{F,M} simulation. 10. ssp585_Totten_T.mat - Bi-weekly ocean temperature (Ta) for Totten Glacier from January 1, 2017 to December 31, 2099 (high emission). 11. ssp585_Moscow_T.mat - Bi-weekly ocean temperature (Ta) for Moscow University Glacier from January 1, 2017 to December 31, 2099 (high emission). 12. ssp585_Vander_T.mat - Bi-weekly ocean temperature (Ta) for Vander Glacier from January 1, 2017 to December 31, 2099 (high emission). 13. ssp585_Totten_S.mat - Bi-weekly ocean salinity (Sa) for Totten Glacier from January 1, 2017 to December 31, 2099 (high emission). 14. ssp585_Moscow_S.mat - Bi-weekly ocean salinity (Sa) for Moscow University Glacier from January 1, 2017 to December 31, 2099 (high emission). 15. ssp585_Vander_S.mat - Bi-weekly ocean salinity (Sa) for Vander Glacier from January 1, 2017 to December 31, 2099 (high emission). 16. ssp126_Totten_T.mat - Bi-weekly ocean temperature (Ta) for Totten Glacier from January 1, 2017 to December 31, 2099 (low emission). 17. ssp126_Moscow_T.mat - Bi-weekly ocean temperature (Ta) for Moscow University Glacier from January 1, 2017 to December 31, 2099 (low emission). 18. ssp126_Vander_T.mat - Bi-weekly ocean temperature (Ta) for Vander Glacier from January 1, 2017 to December 31, 2099 (low emission). 19. ssp126_Totten_S.mat - Bi-weekly ocean salinity (Sa) for Totten Glacier from January 1, 2017 to December 31, 2099 (low emission). 20. ssp126_Moscow_S.mat - Bi-weekly ocean salinity (Sa) for Moscow University Glacier from January 1, 2017 to December 31, 2099 (low emission). 21. ssp126_Vander_S.mat - Bi-weekly ocean salinity (Sa) for Vander Glacier from January 1, 2017 to December 31, 2099 (low emission). 22. TotBasin.exp - Polygon that contains Totten Glacier over which Totten's ocean temperature is applied. 23. MuisBasin.exp - Polygon that contains Moscow University Glacier over which Totten's ocean temperature is applied. 24. VandBasin.exp - Polygon that contains Vanderford Glacier over which Totten's ocean temperature is applied. ### File specific information: **ASB_IceHydroModel.mat**: All data associated with the ice sheet and subglacial hydrology model initial state is held in ASB_IceHydroModel.mat, which contains a MATLAB ‘model’ object (for more information, see [https://issm.jpl.nasa.gov/documentation/modelclass/](https://issm.jpl.nasa.gov/documentation/modelclass/). In MATLAB, the model can be loaded and displayed by running load(‘ASB_IceHydroModel.mat’), which will load in the model variable ‘md’. Of particular interest will be the following data contained in md: md.mesh (mesh information), md.geometry (initial ice sheet geometry, ice shelf geometry, and bed topography), md.hydrology (initial hydrology model fields), md.initialization (model initialization fields) and md.mask (ice mask and grounded ice mask). Note that all fields are defined on the mesh nodes, and one can plot a given field in MATLAB using the ISSM tool ‘plotmodel’ (e.g., plotmodel(md,'data',md.geometry.bed) will plot the model bed topography). For more information on plotting, please see [https://issm.jpl.nasa.gov/documentation/plotmatlab/](https://issm.jpl.nasa.gov/documentation/plotmatlab/). **Model output files (e.g. ssp585_FM.mat)**: Yearly ice sheet model results between 2017-2100 for all model simulations described in the paper. Fields appended with '*' are included in results with changing subglacial hydrology (ssp126_F, ssp126_M, ssp126_FM, ssp585_F, ssp585_M, ssp585_FM). Fields appended with '**' are included in results where ice shelf melt is enhanced by subglacial discharge (ssp126_M, ssp126_FM, ssp585_M, ssp585_FM). These files contain a MATLAB variable that is the same as the file name, which is a model object of size 1x83 that contains the following yearly variables: * \* Vel (velocity norm, m/yr) * \* Thickness (ice sheet thickness, m) * \* Surface (ice sheet surface elevation, m) * \* Base (ice sheet base elevation, m) * \* BasalforcingsFloatingiceMeltingRate (ice shelf basal melting rate field, m/yr) * \* MaskOceanLevelset (ground ice mask, grounded ice if > 0, grounding line position if = 0, floating ice if < 0) * \* IceVolume (total ice volume in the model domain, t) * \* IceVolumeAboveFloatation (total ice volume in the model domain that is above hydrostatic equilibrium, t) * \* TotalFloatingBmb (Total floating basal mass balance, Gt) * \* \\*ChannelDischarge\\_Node (GlaDS-computed channel discharge interpolated onto model node, m3/s) * \* \\*ChannelDiameter\\_Node (GlaDS-computed channel diameter interpolated onto model node, m) * \* \\*ChannelArea (GlaDS-computed channel area defined on model edges, m2) * \* \\*ChannelDischarge (GlaDS\\_computed channel discharge defined on model edges, m3/s) * \* \\*EffectivePressure (GlaDS-computed ice sheet effective pressure, Pa) * \* \\*HydraulicPotential (GlaDS computed hydraulic potential, - * \* \\*HydrologySheetThickness (GlaDS-computed after sheet thickness, m) * \* \\*GroundedIceMeltingRate (Grounded ice melting rate defined on all grounded nodes, m/yr) * \* \\*\\*melt\\_nodis (ice shelf basal melting rate computed when discharge is set to zero, m/yr) * \* \\*\\*zgl (grounding line height field, m) * \* \\*\\*glfw (grounding line fresh water flux field, m2/s) * \* \\*\\*chan\\_wid (Domain average subglacial discharge channel width, m) * \* \\*\\*maxdist (5L' length scale used in melt computation, m) * \* \\*\\*maxis (maximum discharge at each subglacial outflow location, m2/s) * \**\\*\\_T.mat**: Bi-weekly ocean temperature extracted from an East Antarctic configuration of the MITgcm (Pelle et al., 2021), where '\\*' ssp126 (low emission) or ssp585 (high emission). Ocean temperature was averaged adjacent to each target ice front in both depth and in the contours shown in figure 1b. * \**\\*\\_S.mat**: Same as above, but for salinity in units on the Practical Salinity Scale (PSU). * \***.exp**: Exp files that contain coordinates that outline a polygon for the drainage basins of each major glacier in this study (Vanderford Glacier contains the drainage basins for Adams, Bond, and Underwood Glaciers as well). Recent studies have revealed the presence of a complex freshwater system underlying the Aurora Subglacial Basin (ASB), a region of East Antarctica that contains ~7 m of global sea level potential in ice mainly grounded below sea level. Yet, the impact that subglacial freshwater has on driving the evolution of the dynamic outlet glaciers that drain this basin has yet to be tested in a coupled ice sheet-subglacial hydrology numerical modeling framework. Here, we project the evolution of the primary outlet glaciers draining the ASB (Moscow University Ice Shelf, Totten, Vanderford, and Adams Glaciers) in response to an evolving subglacial hydrology system and to ocean forcing through 2100, following low and high CMIP6 emission scenarios. By 2100, ice-hydrology feedbacks enhance the ASB’s 2100 sea level contribution by ~30% (7.50 mm to 9.80 mm) in high emission scenarios and accelerate retreat of Totten Glacier’s main ice stream by 25 years. Ice-hydrology feedbacks are particularly influential in the retreat of the Vanderford and Adams Glaciers, driving an additional 10 km of retreat in fully-coupled simulations relative to uncoupled simulations. Hydrology-driven ice shelf melt enhancements are the primary cause of domain-wide mass loss in low emission scenarios, but are secondary to ice sheet frictional feedbacks under high emission scenarios. The results presented here demonstrate that ice-subglacial hydrology interactions can significantly accelerate retreat of dynamic Antarctic glaciers and that future Antarctic sea level assessments that do not take these interactions into account might be severely underestimating Antarctic Ice Sheet mass loss. In this data publication, we present the model output and results associated with the following manuscript recently submitted to the Journal of Geophysical Research: Earth Surface: “Subglacial discharge accelerates ocean driven retreat of Aurora Subglacial Basin outlet glaciers over the 21st century”. We include yearly ice sheet model output between 2017-2100 for eight numerical ice-subglacial hydrology model runs. We also include the ice sheet and subglacial hydrology model initial states. In addition, we include all ocean forcing time-series (temperature and salinity for the low emission and high emission climate forcing scenarios for three glacial regions), which are used as input into the melt parameterization. Lastly, we include a MATLAB script that contains the code used to couple the ice-subglacial hydrology models as well as a "readme" file with further information on all data in this publication. Ice sheet model results: Direct results taken from the Ice-sheet and Sea-level System Model (ISSM, Larour et al. 2012) with no processing applied, provided yearly as *.mat files. Ice sheet and subglacial hydrology model initial states: Initial state of the ice sheet model (ice geometry, mesh information, inversion results, etc.) and subglacial hydrology model (steady-state water column thickness, effective pressure, channelized discharge, etc.) containing Aurora Subglacial Basin outlet glaciers with no processing applied, provided as a *.mat file. The contents of the *.mat file is a MATLAB variable of class "model", which is compatible with ISSM. Model coupling script: Documented MATLAB script ready to run with the provided data sets. Ocean temperature and salinity timeseries: Bottom ocean temperature (°C) and salinity (PSU) timeseries (January 1st, 2017 through December 31, 2099) extracted from an East Antarctic configuration of the ocean component of the MITgcm (Pelle et al., 2021). Temperature and salinity are provided bi-weekly and averged both in depth and along the ice fronts of Moscow University, Totten, and Vanderford Glaciers (see white dashed contour in figure 1b of the main manuscript text). Data are provided as *.mat files. Polygons that provide locaion to apply ocean temperature and salinity: Polygons provided as a list of x/y coordinates (meters) are provided in three *.exp files that cover the drainage basins of Moscow University, Totten, and Vanderford Glaciers (the polygon for Vanderford also includes the drainage basins of Adams, Bond, and Underwood Glaciers). 

&lt;p&gt;Containing ~52 m sea level rise equivalent ice mass (SLRe), the East Antarctic Ice Sheet (EAIS) is a major component of the global sea level budget; yet, uncertainty remains in how this ice sheet will respond to enhanced atmospheric and oceanic thermal forcing through the turn of the century. To address this uncertainty, we model the most dynamic catchments of EAIS out to 2100 using the Ice Sheet System Model. We employ three basal melt rate parameterizations to resolve ice-ocean interactions and force our model with anomalies in both surface mass balance and ocean thermal forcing from both CMIP5 and CMIP6 model output. We find that this sector of EAIS gains approximately 10 mm SLRe by 2100 under high emission scenarios (RCP8.5 and SSP585), and loses mass under low emission scenarios (RCP2.6). All basins within the domain either gain mass or are in near mass balance through the 86-year experimental period, except the Aurora Subglacial Basin. The primary region of mass loss in this basin is located within 50 km upstream of Totten Glacier&amp;#8217;s grounding line, which loses up to 6 mm SLRe by 2100. Glacial discharge from Totten is modulated by buttress supplied by a 10 km ice plain, located along the southern-most end of Totten&amp;#8217;s grounding line. This ice plain is sensitive to brief changes in ocean temperature and once ungrounded, glacial discharge from Totten accelerates by up to 70% of it present day configuration. In all, we present plausible bounds on the contribution of a large sector of EAIS to global sea level rise out to the end of the century and target Totten as the most vulnerable glacier in this region. In doing so, we reduce uncertainty in century-scale global sea level projections and help steer scientific focus to the most dynamic regions of EAIS.&lt;/p&gt;

# Data from: Subglacial discharge accelerates future retreat of Denman and Scott Glaciers, East Antarctica [https://doi.org/10.7280/D1X12S](https://doi.org/10.7280/D1X12S) Journal: Science Advances Principle Investigator: * Tyler Pelle, Scripps Institution of Oceanography, University of California San Diego, [tpelle@ucsd.edu](mailto:tpelle@ucsd.edu) Co-Authors: * Dr. Jamin Greenbaum, Scripps Institution of Oceanography, University of California San Diego, [jsgreenbaum@ucsd.edu](mailto:jsgreenbaum@ucsd.edu) * Dr. Christine Dow, Department of Geography and Environmental Management, University of Waterloo, [christine.dow@uwaterloo.ca](mailto:christine.dow@uwaterloo.ca) * Dr. Adrian Jenkins, Department of Geography and Environmental Sciences, Northumbria University, [adrian2.jenkins@northumbria.ac.uk](mailto:adrian2.jenkins@northumbria.ac.uk) * Dr. Mathieu Morlighem, Department of Earth Sciences, Dartmouth College, [Mathieu.Morlighem@dartmouth.edu](mailto:Mathieu.Morlighem@dartmouth.edu) Created on September 5, 2023 ## Description of the data and file structure ### File description: 1. runme.m - MATLAB script used to compute melt rates with and without considering subglacial discharge. 2. ice_ctrl_sd.mat – Yearly ice sheet model output from 2017-2100 for the control subglacial discharge simulation. 3. ice_ctrl_nosd.mat – Yearly ice sheet model output from 2017-2100 for the control non-subglacial discharge simulation. 4. ice_ssp126_sd.mat – Yearly ice sheet model output from 2017-2100 for the low emission (SSP1-2.6) subglacial discharge simulation. 5. ice_ssp126_nosd.mat – Yearly ice sheet model output from 2017-2100 for the low emission (SSP1-2.6) non-subglacial discharge simulation. 6. ice_ssp585_sd.mat – Yearly ice sheet model output from 2017-2100 for the high emission (SSP5-8.5) subglacial discharge simulation. 7. ice_ssp585_nosd.mat – Yearly ice sheet model output from 2017-2100 for the high emission (SSP5-8.5) non-subglacial discharge simulation. 8. T_MITgcm_ctrl.mat – Bi-weekly ocean temperature (Ta) for the control simulation from January 1, 2017 to December 31, 2299 averaged at the ice shelf terminus of Denman and Scott Glaciers, used as input into the melt parameterization. 9. S_MITgcm_ctrl.mat – Bi-weekly ocean salinity (Sa) for the control simulation from January 1, 2017 to December 31, 2299 averaged at the ice shelf terminus of Denman and Scott Glaciers, used as input into the melt parameterization. 10. T_MITgcm126.mat - Same as T_MITgcm_ctrl.mat, but for the low emission scenario. 11. S_MITgcm126.mat - Same as S_MITgcm_ctrl.mat, but for the low emission scenario. 12. T_MITgcm585.mat - Same as T_MITgcm_ctrl.mat, but for the high emission scenario. 13. S_MITgcm585.mat - Same as S_MITgcm_ctrl.mat, but for the high emission scenario. 14. discharge_den.xyz - 2017 modeled channelized subglacial discharge flux output from GlaDS, used as input into the melt parameterization. 15. DenModel.mat – Ice sheet model initial state (January 1, 2021), including all mesh information, ice sheet and ice shelf geometry, inversion fields for basal friction and ice stiffness, and initial state variables. ### File specific information: **DenModel.mat**: All data associated with the ice sheet model initial state is held in DenModel.mat, which contains a MATLAB ‘model’ object (for more information, see [https://issm.jpl.nasa.gov/documentation/modelclass/](https://issm.jpl.nasa.gov/documentation/modelclass/)). In MATLAB, the model can be loaded and displayed by running load(‘DenModel.mat’), which will load in the model variable ‘md’. Of particular interest will be the following data contained in md: md.mesh (mesh information), md.geometry (initial ice sheet geometry, ice shelf geometry, and bed topography), and md.mask (ice mask and grounded ice mask). Note that all fields are defined on the mesh nodes, and one can plot a given field in MATLAB using the ISSM tool ‘plotmodel’. Once IceSheetModel.mat is loaded, we can plot the ice shelf basal melting rate by running the following command: plotmodel(md, ’data’, md.results.TransientSolution(1).BasalforcingsFloatingiceMeltingRate). For more information on plotting, please see [https://issm.jpl.nasa.gov/documentation/plotmatlab/](https://issm.jpl.nasa.gov/documentation/plotmatlab/). **ice_*_sd.mat**: Yearly ice sheet model results between 2017–2300 for the subglacial discharge experiments, where ‘*’ is either ctrl (control), ssp126 (low emission scenario), or ssp585 (high emission scenario). These files contain a MATLAB variable that is the same as the file name, which is a model object of size 1x283 that contains the following yearly variables: * Vel (velocity norm, m/yr) * Pressure (N/A since we use a 2D ice flow model) * Thickness (ice sheet thickness, m) * Surface (ice sheet surface elevation, m) * Base (ice sheet base elevation, m) * BasalforcingsFloatingiceMeltingRate (ice shelf basal melting rate field, m/yr) * MaskOceanLevelset (ground ice mask, grounded ice if > 0, grounding line position if = 0, floating ice if < 0) * IceVolume (total ice volume in the model domain, t) * IceVolumeAboveFloatation (total ice volume in the model domain that is above hydrostatic equilibrium, t) * TotalFloatingBmb (Total floating basal mass balance, Gt) * melt_nodis (ice shelf basal melting rate computed when discharge is set to zero, m/yr) * zgl (grounding line height field, m) * glfw (grounding line fresh water flux field, m2/s) * chan_wid (Domain average subglacial discharge channel width, m) * maxdist (5L' length scale used in melt computation, m) * maxdis (maximum discharge at each subglacial outflow location, m2/s) **ice_*_nosd.mat:** Same file information as ice_*_sd.mat, but for the non-subglacial discharge ice sheet model simulations without the following variables: melt_nodis, zgl, glfw, chan_wid, max_dist, and max_dis. **T_MITgcm*.mat**: Bi-weekly ocean temperature extracted from an East Antarctic configuration of the MITgcm from Pelle et al. (2021), where '*' is ctrl (control), ssp126 (low emission scenario), or ssp585 (high emission scenario). Ocean temperature was averaged at the ice front of Denman and Scott Glaciers (see contour in Figure 1 in the main text) at the lowest ocean level. Ocean temperature data is in units of degrees Celsius. **S_MITgcm*.mat**: Same as above, but for salinity in units on the Practical Salinity Scale (PSU). **discharge_den.xyz:** GlaDS output present-day channelized subglacial discharge flux (m3/s). To load the data and interpolate onto the model mesh, use: ``` [x,y,dis] = xyz2grid('discharge_den.xyz'); dis_den = InterpFromGridToMesh(x(1,:)',flipud(y(:,1)),flipud(dis),md.mesh.x,md.mesh.y,0); ``` where 'InterpFromGridToMesh' is a module built into ISSM. Ice sheet model results: Direct results taken from the Ice-sheet and Sea-level System Model (ISSM, Larour et al. 2012) with no processing applied, provided yearly as *.mat files. Ice sheet model initial state: Initial state (ice geometry, mesh information, inversion results, etc.) of the ice sheet model containing Denman and Scott Galciers with no processing applied, provided as a *.mat file. The contents of the *.mat file is a MATLAB variable of class "model", which is compatible with ISSM. Melt parameterization script: Documented MATLAB script ready to run with the provided data sets. Ocean temperature and salinity timeseries: Bottom ocean temperature (°C) and salinity (PSU) timeseries (January 1st, 2017 through December 31, 2299) extracted from an East Antarctic configuration of the ocean component of the MITgcm (Pelle et al., 2021). Temperature and salinity were averaged bi-weekly along the ice fronts of Denman and Scott Glaciers (see white dashed contour in figure 1a of the main manuscript text) along the sea floor. Data are provided as *.mat files. Note that the ocean model in Pelle et al. (2021) was run through 2100. In the control and low emission scenarios, we repreat the last 20-years of simulated ocean (2079-2099) conditions through 2300. In the high emission scenario, where a clear warming trend was evident in the ocean temperature after 2050, we extrapolate this warming trend with a square-root function and add this onto the retreated 2079-2099 repearted forcing. Channelized discharge flux data: Present-day output from the Glacier Drainage Systems (GlaDS) model in units of m3/s over grounded elements of the domain. Data is provided in a *.xyz file (see README.txt for instructions on how to load the data and interpolate onto model mesh using MATLAB). No processing has been applied other than subglacial flux values less than 0.001 m3/s have been removed from the dataset. Ice shelf basal melting is the primary mechanism driving mass loss from the Antarctic Ice Sheet, yet it is unknown how the localized melt enhancement from subglacial discharge will impact future Antarctic glacial retreat. Here, we develop a parameterization of ice shelf basal melt that accounts for both ocean and subglacial discharge forcing and apply it in future projections of Denman and Scott Glaciers, East Antarctica, through 2300. In forward simulations, subglacial discharge accelerates retreat of these systems into the deepest continental trench on Earth by 25 years. During this retreat, Denman Glacier alone contributes 0.33 mm/yr to global sea level rise, comparable to half of the contemporary sea level contribution of the entire Antarctic Ice Sheet. Our results stress the importance of resolving complex interactions between the ice, ocean, and subglacial environments in future Antarctic Ice Sheet projections. In this data publication, we present the model output and results associated with the following manuscript submitted to Science Advances: “Subglacial discharge will accelerate retreat of Denman and Scott Glaciers, East Antarctica”. We include yearly ice sheet model output between 2017-2300 for models that do and do not resolve subgalcial discharge in the melt calculation. We also include the ice sheet model's initial state. In addition, we include all ocean forcing time-series (temperature and salinity for the control, low emission, and high emission climate forcing scenarios) and the present-day chanellized subglacial discharge flux field over the Denman and Scott Glacier model domain, which are used as input into the melt parameterization. Lastly, we include a MATLAB script that contains the code used for ice shelf melt rate computation as well as a "README" file with further information on all data in this publication.  Ice sheet modeling results and initial states are compatible with the open source, NASA funded Ice-sheet and Sea-level System Model (ISSM, Larour et al. 2012), which is freely available for download here. In addition, the data files provided in the publication are available as *.mat files, which are compatible with MATLAB but can be accessed using most scripting languages.