Articles | Volume 17, issue 2
https://doi.org/10.5194/tc-17-499-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/tc-17-499-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Predicting ocean-induced ice-shelf melt rates using deep learning
Sebastian H. R. Rosier
CORRESPONDING AUTHOR
Department of Geography and Environmental Sciences, Northumbria University, Newcastle Upon Tyne, UK
WSL Institute for Snow and Avalanche Research SLF, Davos, Switzerland
Christopher Y. S. Bull
Department of Geography and Environmental Sciences, Northumbria University, Newcastle Upon Tyne, UK
Wai L. Woo
Department of Computer and Information Sciences, Northumbria University, Newcastle Upon Tyne, UK
G. Hilmar Gudmundsson
Department of Geography and Environmental Sciences, Northumbria University, Newcastle Upon Tyne, UK
Related authors
Sebastian H. R. Rosier, G. Hilmar Gudmundsson, Adrian Jenkins, and Kaitlin A. Naughten
The Cryosphere, 19, 2527–2557, https://doi.org/10.5194/tc-19-2527-2025, https://doi.org/10.5194/tc-19-2527-2025, 2025
Short summary
Short summary
Glaciers in the Amundsen Sea region of Antarctica have been retreating and losing mass, but their future contribution to global sea level rise remains highly uncertain. We use an ice sheet model and uncertainty quantification methods to evaluate the probable range of mass loss from this region for two future climate scenarios. We find that the rate of ice loss until 2100 will likely remain similar to present-day observations, with little sensitivity to climate scenario over this short time frame.
Christian T. Wild, Reinhard Drews, Niklas Neckel, Joohan Lee, Sihyung Kim, Hyangsun Han, Won Sang Lee, Veit Helm, Sebastian Harry Reid Rosier, Oliver J. Marsh, and Wolfgang Rack
EGUsphere, https://doi.org/10.5194/egusphere-2024-3593, https://doi.org/10.5194/egusphere-2024-3593, 2024
Short summary
Short summary
The stability of the Antarctic Ice Sheet depends on how resistance along the sides of large glaciers slows down the flow of ice into the ocean. We present a method to map ice strength using the effect of ocean tides on floating ice shelves. Incorporating weaker ice in shear zones improves the accuracy of model predictions compared to satellite observations. This demonstrates the untapped potential of radar satellites to map ice stiffness in the most critical areas for ice sheet stability.
Cristina Gerli, Sebastian Rosier, G. Hilmar Gudmundsson, and Sainan Sun
The Cryosphere, 18, 2677–2689, https://doi.org/10.5194/tc-18-2677-2024, https://doi.org/10.5194/tc-18-2677-2024, 2024
Short summary
Short summary
Recent efforts have focused on using AI and satellite imagery to track crevasses for assessing ice shelf damage and informing ice flow models. Our study reveals a weak connection between these observed products and damage maps inferred from ice flow models. While there is some improvement in crevasse-dense regions, this association remains limited. Directly mapping ice damage from satellite observations may not significantly improve the representation of these processes within ice flow models.
Emily A. Hill, Sebastian H. R. Rosier, G. Hilmar Gudmundsson, and Matthew Collins
The Cryosphere, 15, 4675–4702, https://doi.org/10.5194/tc-15-4675-2021, https://doi.org/10.5194/tc-15-4675-2021, 2021
Short summary
Short summary
Using an ice flow model and uncertainty quantification methods, we provide probabilistic projections of future sea level rise from the Filchner–Ronne region of Antarctica. We find that it is most likely that this region will contribute negatively to sea level rise over the next 300 years, largely as a result of increased surface mass balance. We identify parameters controlling ice shelf melt and snowfall contribute most to uncertainties in projections.
Sebastian H. R. Rosier, Ronja Reese, Jonathan F. Donges, Jan De Rydt, G. Hilmar Gudmundsson, and Ricarda Winkelmann
The Cryosphere, 15, 1501–1516, https://doi.org/10.5194/tc-15-1501-2021, https://doi.org/10.5194/tc-15-1501-2021, 2021
Short summary
Short summary
Pine Island Glacier has contributed more to sea-level rise over the past decades than any other glacier in Antarctica. Ice-flow modelling studies have shown that it can undergo periods of rapid mass loss, but no study has shown that these future changes could cross a tipping point and therefore be effectively irreversible. Here, we assess the stability of Pine Island Glacier, quantifying the changes in ocean temperatures required to cross future tipping points using statistical methods.
Sebastian H. R. Rosier, G. Hilmar Gudmundsson, Adrian Jenkins, and Kaitlin A. Naughten
The Cryosphere, 19, 2527–2557, https://doi.org/10.5194/tc-19-2527-2025, https://doi.org/10.5194/tc-19-2527-2025, 2025
Short summary
Short summary
Glaciers in the Amundsen Sea region of Antarctica have been retreating and losing mass, but their future contribution to global sea level rise remains highly uncertain. We use an ice sheet model and uncertainty quantification methods to evaluate the probable range of mass loss from this region for two future climate scenarios. We find that the rate of ice loss until 2100 will likely remain similar to present-day observations, with little sensitivity to climate scenario over this short time frame.
Ole Richter, Ralph Timmermann, G. Hilmar Gudmundsson, and Jan De Rydt
Geosci. Model Dev., 18, 2945–2960, https://doi.org/10.5194/gmd-18-2945-2025, https://doi.org/10.5194/gmd-18-2945-2025, 2025
Short summary
Short summary
The new coupled ice sheet–ocean model addresses challenges related to horizontal resolution through advanced mesh flexibility, enabled by the use of unstructured grids. We describe the new model, verify its functioning in an idealised setting and demonstrate its advantages in a global-ocean–Antarctic ice sheet domain. The results of this study comprise an important step towards improving predictions of the Antarctic contribution to sea level rise over centennial timescales.
Jing Jin, Antony J. Payne, and Christopher Y. S. Bull
The Cryosphere, 19, 1873–1896, https://doi.org/10.5194/tc-19-1873-2025, https://doi.org/10.5194/tc-19-1873-2025, 2025
Short summary
Short summary
The Amery Ice Shelf cavity is one of the largest cold cavities filled by relatively cold Dense Shelf Water. However, in this study, we show that warm intrusion of modified Circumpolar Deep Water flushes the Amery cavity, which changes it from a cold cavity to a warm cavity and leads to an abrupt increase in the basal melt rate in the 2060s. The shift to a warm cavity is attributed to a freshening-driven current reversal in front of the ice shelf.
Jowan M. Barnes, G. Hilmar Gudmundsson, Daniel N. Goldberg, and Sainan Sun
EGUsphere, https://doi.org/10.5194/egusphere-2025-328, https://doi.org/10.5194/egusphere-2025-328, 2025
Short summary
Short summary
Calving is where ice breaks off the front of glaciers. It has not been included widely in modelling as it is difficult to represent. We use our ice flow model to investigate the effects of calving floating ice shelves in West Antarctica. More calving leads to more ice loss and greater sea level rise, with local differences due to the shape of the bedrock. We find that ocean forcing and calving should be considered equally when trying to improve how models represent the real world.
Christian T. Wild, Reinhard Drews, Niklas Neckel, Joohan Lee, Sihyung Kim, Hyangsun Han, Won Sang Lee, Veit Helm, Sebastian Harry Reid Rosier, Oliver J. Marsh, and Wolfgang Rack
EGUsphere, https://doi.org/10.5194/egusphere-2024-3593, https://doi.org/10.5194/egusphere-2024-3593, 2024
Short summary
Short summary
The stability of the Antarctic Ice Sheet depends on how resistance along the sides of large glaciers slows down the flow of ice into the ocean. We present a method to map ice strength using the effect of ocean tides on floating ice shelves. Incorporating weaker ice in shear zones improves the accuracy of model predictions compared to satellite observations. This demonstrates the untapped potential of radar satellites to map ice stiffness in the most critical areas for ice sheet stability.
Richard Parsons, Sainan Sun, G. Hilmar Gudmundsson, Jan Wuite, and Thomas Nagler
The Cryosphere, 18, 5789–5801, https://doi.org/10.5194/tc-18-5789-2024, https://doi.org/10.5194/tc-18-5789-2024, 2024
Short summary
Short summary
In 2022, multi-year landfast sea ice in Antarctica's Larsen B embayment disintegrated, after which time an increase in the rate at which Crane Glacier discharged ice into the ocean was observed. As the fast ice was joined to the glacier terminus, it could provide resistance against the glacier's flow, slowing down the rate of ice discharge. We used numerical modelling to quantify this resistive stress and found that the fast ice provided significant support to Crane prior to its disintegration.
Xianwei Wang, Hilmar Gudmundsson, and David Holland
EGUsphere, https://doi.org/10.5194/egusphere-2024-2790, https://doi.org/10.5194/egusphere-2024-2790, 2024
Short summary
Short summary
Understanding why iceberg calved during drifting in the ocean is important to understand the life cycle and the influence on the surrounding ocean of an iceberg. This study explains why iceberg A68a calved when approaching the South Georgia Island in late 2020 during its drifting in the Southern Ocean using satellite observation and modeling, which was caused by collision with seamount.
Brad Reed, J. A. Mattias Green, Adrian Jenkins, and G. Hilmar Gudmundsson
The Cryosphere, 18, 4567–4587, https://doi.org/10.5194/tc-18-4567-2024, https://doi.org/10.5194/tc-18-4567-2024, 2024
Short summary
Short summary
We use a numerical ice-flow model to simulate the response of a 1940s Pine Island Glacier to changes in melting beneath its ice shelf. A decadal period of warm forcing is sufficient to push the glacier into an unstable, irreversible retreat from its long-term position on a subglacial ridge to an upstream ice plain. This retreat can only be stopped when unrealistic cold forcing is applied. These results show that short warm anomalies can lead to quick and substantial increases in ice flux.
Gong Cheng, Mathieu Morlighem, and G. Hilmar Gudmundsson
Geosci. Model Dev., 17, 6227–6247, https://doi.org/10.5194/gmd-17-6227-2024, https://doi.org/10.5194/gmd-17-6227-2024, 2024
Short summary
Short summary
We conducted a comprehensive analysis of the stabilization and reinitialization techniques currently employed in ISSM and Úa for solving level-set equations, specifically those related to the dynamic representation of moving ice fronts within numerical ice sheet models. Our results demonstrate that the streamline upwind Petrov–Galerkin (SUPG) method outperforms the other approaches. We found that excessively frequent reinitialization can lead to exceptionally high errors in simulations.
J. Rachel Carr, Emily A. Hill, and G. Hilmar Gudmundsson
The Cryosphere, 18, 2719–2737, https://doi.org/10.5194/tc-18-2719-2024, https://doi.org/10.5194/tc-18-2719-2024, 2024
Short summary
Short summary
The Greenland Ice Sheet is one of the world's largest glaciers and is melting quickly in response to climate change. It contains fast-flowing channels of ice that move ice from Greenland's centre to its coasts and allow Greenland to react quickly to climate warming. As a result, we want to predict how these glaciers will behave in the future, but there are lots of uncertainties. Here we assess the impacts of two main sources of uncertainties in glacier models.
Cristina Gerli, Sebastian Rosier, G. Hilmar Gudmundsson, and Sainan Sun
The Cryosphere, 18, 2677–2689, https://doi.org/10.5194/tc-18-2677-2024, https://doi.org/10.5194/tc-18-2677-2024, 2024
Short summary
Short summary
Recent efforts have focused on using AI and satellite imagery to track crevasses for assessing ice shelf damage and informing ice flow models. Our study reveals a weak connection between these observed products and damage maps inferred from ice flow models. While there is some improvement in crevasse-dense regions, this association remains limited. Directly mapping ice damage from satellite observations may not significantly improve the representation of these processes within ice flow models.
Emily A. Hill, Benoît Urruty, Ronja Reese, Julius Garbe, Olivier Gagliardini, Gaël Durand, Fabien Gillet-Chaulet, G. Hilmar Gudmundsson, Ricarda Winkelmann, Mondher Chekki, David Chandler, and Petra M. Langebroek
The Cryosphere, 17, 3739–3759, https://doi.org/10.5194/tc-17-3739-2023, https://doi.org/10.5194/tc-17-3739-2023, 2023
Short summary
Short summary
The grounding lines of the Antarctic Ice Sheet could enter phases of irreversible retreat or advance. We use three ice sheet models to show that the present-day locations of Antarctic grounding lines are reversible with respect to a small perturbation away from their current position. This indicates that present-day retreat of the grounding lines is not yet irreversible or self-enhancing.
Ronja Reese, Julius Garbe, Emily A. Hill, Benoît Urruty, Kaitlin A. Naughten, Olivier Gagliardini, Gaël Durand, Fabien Gillet-Chaulet, G. Hilmar Gudmundsson, David Chandler, Petra M. Langebroek, and Ricarda Winkelmann
The Cryosphere, 17, 3761–3783, https://doi.org/10.5194/tc-17-3761-2023, https://doi.org/10.5194/tc-17-3761-2023, 2023
Short summary
Short summary
We use an ice sheet model to test where current climate conditions in Antarctica might lead. We find that present-day ocean and atmosphere conditions might commit an irreversible collapse of parts of West Antarctica which evolves over centuries to millennia. Importantly, this collapse is not irreversible yet.
Bertie W. J. Miles, Chris R. Stokes, Adrian Jenkins, Jim R. Jordan, Stewart S. R. Jamieson, and G. Hilmar Gudmundsson
The Cryosphere, 17, 445–456, https://doi.org/10.5194/tc-17-445-2023, https://doi.org/10.5194/tc-17-445-2023, 2023
Short summary
Short summary
Satellite observations have shown that the Shirase Glacier catchment in East Antarctica has been gaining mass over the past 2 decades, a trend largely attributed to increased snowfall. Our multi-decadal observations of Shirase Glacier show that ocean forcing has also contributed to some of this recent mass gain. This has been caused by strengthening easterly winds reducing the inflow of warm water underneath the Shirase ice tongue, causing the glacier to slow down and thicken.
Jowan M. Barnes and G. Hilmar Gudmundsson
The Cryosphere, 16, 4291–4304, https://doi.org/10.5194/tc-16-4291-2022, https://doi.org/10.5194/tc-16-4291-2022, 2022
Short summary
Short summary
Models must represent how glaciers slide along the bed, but there are many ways to do so. In this paper, several sliding laws are tested and found to affect different regions of the Antarctic Ice Sheet in different ways and at different speeds. However, the variability in ice volume loss due to sliding-law choices is low compared to other factors, so limited empirical knowledge of sliding does not prevent us from making predictions of how an ice sheet will evolve.
Tom Mitcham, G. Hilmar Gudmundsson, and Jonathan L. Bamber
The Cryosphere, 16, 883–901, https://doi.org/10.5194/tc-16-883-2022, https://doi.org/10.5194/tc-16-883-2022, 2022
Short summary
Short summary
We modelled the response of the Larsen C Ice Shelf (LCIS) and its tributary glaciers to the calving of the A68 iceberg and validated our results with observations. We found that the impact was limited, confirming that mostly passive ice was calved. Through further calving experiments we quantified the total buttressing provided by the LCIS and found that over 80 % of the buttressing capacity is generated in the first 5 km of the ice shelf downstream of the grounding line.
Emily A. Hill, Sebastian H. R. Rosier, G. Hilmar Gudmundsson, and Matthew Collins
The Cryosphere, 15, 4675–4702, https://doi.org/10.5194/tc-15-4675-2021, https://doi.org/10.5194/tc-15-4675-2021, 2021
Short summary
Short summary
Using an ice flow model and uncertainty quantification methods, we provide probabilistic projections of future sea level rise from the Filchner–Ronne region of Antarctica. We find that it is most likely that this region will contribute negatively to sea level rise over the next 300 years, largely as a result of increased surface mass balance. We identify parameters controlling ice shelf melt and snowfall contribute most to uncertainties in projections.
Jowan M. Barnes, Thiago Dias dos Santos, Daniel Goldberg, G. Hilmar Gudmundsson, Mathieu Morlighem, and Jan De Rydt
The Cryosphere, 15, 1975–2000, https://doi.org/10.5194/tc-15-1975-2021, https://doi.org/10.5194/tc-15-1975-2021, 2021
Short summary
Short summary
Some properties of ice flow models must be initialised using observed data before they can be used to produce reliable predictions of the future. Different models have different ways of doing this, and the process is generally seen as being specific to an individual model. We compare the methods used by three different models and show that they produce similar outputs. We also demonstrate that the outputs from one model can be used in other models without introducing large uncertainties.
Sebastian H. R. Rosier, Ronja Reese, Jonathan F. Donges, Jan De Rydt, G. Hilmar Gudmundsson, and Ricarda Winkelmann
The Cryosphere, 15, 1501–1516, https://doi.org/10.5194/tc-15-1501-2021, https://doi.org/10.5194/tc-15-1501-2021, 2021
Short summary
Short summary
Pine Island Glacier has contributed more to sea-level rise over the past decades than any other glacier in Antarctica. Ice-flow modelling studies have shown that it can undergo periods of rapid mass loss, but no study has shown that these future changes could cross a tipping point and therefore be effectively irreversible. Here, we assess the stability of Pine Island Glacier, quantifying the changes in ocean temperatures required to cross future tipping points using statistical methods.
Bertie W. J. Miles, Jim R. Jordan, Chris R. Stokes, Stewart S. R. Jamieson, G. Hilmar Gudmundsson, and Adrian Jenkins
The Cryosphere, 15, 663–676, https://doi.org/10.5194/tc-15-663-2021, https://doi.org/10.5194/tc-15-663-2021, 2021
Short summary
Short summary
We provide a historical overview of changes in Denman Glacier's flow speed, structure and calving events since the 1960s. Based on these observations, we perform a series of numerical modelling experiments to determine the likely cause of Denman's acceleration since the 1970s. We show that grounding line retreat, ice shelf thinning and the detachment of Denman's ice tongue from a pinning point are the most likely causes of the observed acceleration.
Jan De Rydt, Ronja Reese, Fernando S. Paolo, and G. Hilmar Gudmundsson
The Cryosphere, 15, 113–132, https://doi.org/10.5194/tc-15-113-2021, https://doi.org/10.5194/tc-15-113-2021, 2021
Short summary
Short summary
We used satellite observations and numerical simulations of Pine Island Glacier, West Antarctica, between 1996 and 2016 to show that the recent increase in its flow speed can only be reproduced by computer models if stringent assumptions are made about the material properties of the ice and its underlying bed. These assumptions are not commonly adopted in ice flow modelling, and our results therefore have implications for future simulations of Antarctic ice flow and sea level projections.
Kate Winter, Emily A. Hill, G. Hilmar Gudmundsson, and John Woodward
Earth Syst. Sci. Data, 12, 3453–3467, https://doi.org/10.5194/essd-12-3453-2020, https://doi.org/10.5194/essd-12-3453-2020, 2020
Short summary
Short summary
Satellite measurements of the English Coast in the Antarctic Peninsula reveal that glaciers are thinning and losing mass, but ice thickness data are required to assess these changes, in terms of ice flux and sea level contribution. Our ice-penetrating radar measurements reveal that low-elevation subglacial channels control fast-flowing ice streams, which release over 39 Gt of ice per year to floating ice shelves. This topography could make ice flows susceptible to future instability.
Cited articles
Asay-Davis, X. S., Cornford, S. L., Durand, G., Galton-Fenzi, B. K., Gladstone, R. M., Gudmundsson, G. H., Hattermann, T., Holland, D. M., Holland, D., Holland, P. R., Martin, D. F., Mathiot, P., Pattyn, F., and Seroussi, H.:
Experimental design for three interrelated marine ice sheet and ocean model intercomparison projects: MISMIP v. 3 (MISMIP +), ISOMIP v. 2 (ISOMIP +) and MISOMIP v. 1 (MISOMIP1), Geosci. Model Dev., 9, 2471–2497, https://doi.org/10.5194/gmd-9-2471-2016, 2016. a, b
Barnier, B., Madec, G., Penduff, T., Molines, J., Treguier, A.-M., Le Sommer, J., Beckmann, A., Biastoch, A., Boning, C., Dengg, J., Derval, C., Durand, E., Gulev, S., Remy, E., Talandier, C., Theetten, S., Maltrud, M., McClean, J., and de Cuevas, B.: Impact of partial steps and momentum advection schemes in a global ocean circulation model at eddy-permitting resolution, Ocean Dynam., 56, 543–567, https://doi.org/10.1007/s10236-006-0082-1, 2006. a
Beucler, T., Pritchard, M., Rasp, S., Ott, J., Baldi, P., and Gentine, P.:
Enforcing Analytic Constraints in Neural Networks Emulating Physical Systems, Phys. Rev. Lett., 126, 098302, https://doi.org/10.1103/PhysRevLett.126.098302, 2021. a
Boyer, T. P., García, H. E., Locarnini, R. A., Zweng, M. M., Mishonov, A. V., Reagan, J. R., Weathers, K. A., Baranova, O. K., Paver, C. R., Seidov, D., Smolyar, I. V.: World Ocean Atlas 2018, decav, NOAA National Centers for Environmental Information [data set], https://www.ncei.noaa.gov/archive/accession/NCEI-WOA18 (last access: 10 June 2021), 2018. a, b
Brenowitz, N. D. and Bretherton, C. S.:
Prognostic Validation of a Neural Network Unified Physics Parameterization, Geophys. Res. Lett., 45, 6289–6298, https://doi.org/10.1029/2018GL078510, 2018. a
Brenowitz, N. D. and Bretherton, C. S.:
Spatially Extended Tests of a Neural Network Parametrization Trained by Coarse-Graining, J. Adv. Model. Earth Sy., 11, 2728–2744, https://doi.org/10.1029/2019MS001711, 2019. a
Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., and Yuille, A. L.:
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs, IEEE T. Pattern Anal., 40, 834–848, https://doi.org/10.1109/tpami.2017.2699184, 2018. a
De Rydt, J., Holland, P. R., Dutrieux, P., and Jenkins, A.:
Geometric and oceanographic controls on melting beneath Pine Island Glacier, J. Geophys. Res.-Ocean., 119, 2420–2438, https://doi.org/10.1002/2013JC009513, 2014. a
De Rydt, J., Reese, R., Paolo, F. S., and Gudmundsson, G. H.:
Drivers of Pine Island Glacier speed-up between 1996 and 2016, The Cryosphere, 15, 113–132, https://doi.org/10.5194/tc-15-113-2021, 2021. a
Donat-Magnin, M., Jourdain, N. C., Spence, P., Le Sommer, J., Gallée, H., and Durand, G.:
Ice-Shelf Melt Response to Changing Winds and Glacier Dynamics in the Amundsen Sea Sector, Antarctica, J. Geophys. Res.-Ocean., 122, 10206–10224, https://doi.org/10.1002/2017JC013059, 2017. a
Dupont, T. K. and Alley, R. B.:
Assessment of the importance of ice-shelf buttressing to ice-sheet flow, Geophys. Res. Lett., 32, L04503, https://doi.org/10.1029/2004GL022024, 2005. a
Edwards, T. L., Nowicki, S., Marzeion, B., Hock, R., Goelzer, H., Seroussi, H., Jourdain, N. C., Slater, D. A., Turner, F. E., Smith, C. J., McKenna, C. M., Simon, E., Abe-Ouchi, A., Gregory, J. M., Larour, E., Lipscomb, W. H., Payne, A. J., Shepherd, A., Agosta, C., Alexander, P., Albrecht, T., Anderson, B., Asay-Davis, X., Aschwanden, A., Barthel, A., Bliss, A., Calov, R., Chambers, C., Champollion, N., Choi, Y., Cullather, R., Cuzzone, J., Dumas, C., Felikson, D., Fettweis, X., Fujita, K., Galton-Fenzi, B. K., Gladstone, R., Golledge, N. R., Greve, R., Hattermann, T., Hoffman, M. J., Humbert, A., Huss, M., Huybrechts, P., Immerzeel, W., Kleiner, T., Kraaijenbrink, P., Le clec’h, S., Lee, V., Leguy, G. R., Little, C. M., Lowry, D. P., Malles, J.-H., Martin, D. F., Maussion, F., Morlighem, M., O’Neill, J. F., Nias, I., Pattyn, F., Pelle, T., Price, S. F., Quiquet, A., Radić, V., Reese, R., Rounce, D. R., Rückamp, M., Sakai, A., Shafer, C., Schlegel, N.-J., Shannon, S., Smith, R. S., Straneo, F., Sun, S., Tarasov, L., Trusel, L. D., Van Breedam, J., van de Wal, R., van den Broeke, M., Winkelmann, R., Zekollari, H., Zhao, C., Zhang, T., and Zwinger, T.: Projected land ice contributions to twenty-first-century sea level rise, Nature, 593, 74–82, https://doi.org/10.1038/s41586-021-03302-y, 2021. a
Favier, L., Jourdain, N. C., Jenkins, A., Merino, N., Durand, G., Gagliardini, O., Gillet-Chaulet, F., and Mathiot, P.:
Assessment of sub-shelf melting parameterisations using the ocean–ice-sheet coupled model NEMO(v3.6)–Elmer/Ice(v8.3) , Geosci. Model Dev., 12, 2255–2283, https://doi.org/10.5194/gmd-12-2255-2019, 2019. a, b
Feldmann, J., Reese, R., Winkelmann, R., and Levermann, A.:
Shear-margin melting causes stronger transient ice discharge than ice-stream melting in idealized simulations, The Cryosphere, 16, 1927–1940, https://doi.org/10.5194/tc-16-1927-2022, 2022. a
Garbe, J., Albrecht, T., Levermann, A., Donges, J., and Winkelmann, R.:
The hysteresis of the Antarctic Ice Sheet, Nature, 585, 538–544, https://doi.org/10.1038/s41586-020-2727-5, 2020. a
Goldberg, D. N., Gourmelen, N., Kimura, S., Millan, R., and Snow, K.:
How Accurately Should We Model Ice Shelf Melt Rates?, Geophys. Res. Lett., 46, 189–199, https://doi.org/10.1029/2018GL080383, 2019. a
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y.:
Generative Adversarial Nets, in: Advances in Neural Information Processing Systems, edited by: Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N., and Weinberger, K. Q., vol. 27, Curran Associates, Inc., https://proceedings.neurips.cc/paper/2014/file/5ca3e9b122f61f8f06494c97b1afccf3-Paper.pdf (last access: 1 April 2021), 2014. a
Gudmundsson, H.: GHilmarG/UaSource: Ua2019b (Version v2019b), Zenodo [code], https://doi.org/10.5281/zenodo.3706623, 2020. a
Gudmundsson, G. H., Krug, J., Durand, G., Favier, L., and Gagliardini, O.:
The stability of grounding lines on retrograde slopes, The Cryosphere, 6, 1497–1505, https://doi.org/10.5194/tc-6-1497-2012, 2012. a
Gudmundsson, G. H., Paolo, F. S., Adusumilli, S., and Fricker, H. A.:
Instantaneous Antarctic ice sheet mass loss driven by thinning ice shelves, Geophys. Res. Lett., 46, 13903–13909, https://doi.org/10.1029/2019GL085027, 2019. a
Gurvan, M., Bourdallé-Badie, R., Chanut, J., Clementi, E., Coward, A., Ethé, C., Iovino, D., Lea, D., Lévy, C., Lovato, T., Martin, N., Masson, S., Mocavero, S., Rousset, C., Storkey, D., Vancoppenolle, M., Müeller, S., Nurser, G., Bell, M., and Samson, G.: NEMO ocean engine, Institut Pierre-Simon Laplace (IPSL), Zenodo, https://doi.org/10.5281/zenodo.1464816, 2019. a
He, K., Zhang, X., Ren, S., and Sun, J.:
Deep Residual Learning for Image Recognition, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770–778, https://doi.org/10.1109/CVPR.2016.90, 2016. a
Holland, P. R., Jenkins, A., and Holland, D. M.:
The Response of Ice Shelf Basal Melting to Variations in Ocean Temperature, J. Climate, 21, 2558 – 2572, https://doi.org/10.1175/2007JCLI1909.1, 2008. a, b
Hu, J., Shen, L., and Sun, G.:
Squeeze-and-Excitation Networks, in: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, pp. 7132–7141, https://doi.org/10.1109/CVPR.2018.00745, 2018. a
Intergovernmental Oceanographic Commission, Scientific Committee on Oceanic Research, and International Association for the Physical Sciences of the Oceans: The International thermodynamic equation of seawater – 2010: calculation and use of thermodynamic properties, includes corrections up to 31 October 2015, Paris, France, UNESCO, Intergovernmental Oceanographic Commission Manuals and Guides, 56, 196 pp., https://doi.org/10.25607/OBP-1338, 2015. a, b
IPCC:
IPCC, 2021: Climate Change 2021: The Physical Science Basis, Cambridge University Press, Cambridge, United Kingdom, 2021. a
Jenkins, A.:
A one-dimensional model of ice shelf-ocean interaction, J. Geophys. Res.-Ocean., 96, 20671–20677, https://doi.org/10.1029/91JC01842, 1991. a
Jenkins, A., Shoosmith, D., Dutrieux, P., Jacobs, S., Kim, T. W., Lee, S. H., Ha, H. K., and Stammerjohn, S.: West Antarctic Ice Sheet retreat in the Amundsen Sea driven by decadal oceanic variability, Nat. Geosci., 11, 733–738, https://doi.org/10.1038/s41561-018-0207-4, 2018. a
Jha, D., Smedsrud, P. H., Riegler, M. A., Johansen, D., Lange, T. D., Halvorsen, P., and Johansen, D. H.:
ResUNet++: An Advanced Architecture for Medical Image Segmentation, in: 2019 IEEE International Symposium on Multimedia (ISM), San Diego, CA, USA, 2019, pp. 225–2255, https://doi.org/10.1109/ISM46123.2019.00049, 2019. a, b
Jordan, J. R., Holland, P. R., Goldberg, D., Snow, K., Arthern, R., Campin, J.-M., Heimbach, P., and Jenkins, A.:
Ocean-Forced Ice-Shelf Thinning in a Synchronously Coupled Ice–Ocean Model, J. Geophys. Res.-Ocean., 123, 864–882, https://doi.org/10.1002/2017JC013251, 2018. a, b
Jourdain, N. C., Asay-Davis, X., Hattermann, T., Straneo, F., Seroussi, H., Little, C. M., and Nowicki, S.:
A protocol for calculating basal melt rates in the ISMIP6 Antarctic ice sheet projections, The Cryosphere, 14, 3111–3134, https://doi.org/10.5194/tc-14-3111-2020, 2020. a
Jouvet, G., Cordonnier, G., Kim, B., Lüthi, M., Vieli, A., and Aschwanden, A.:
Deep learning speeds up ice flow modelling by several orders of magnitude, J. Glaciol., 68, 651–664, https://doi.org/10.1017/jog.2021.120, 2022. a, b
Khairoutdinov, M. F. and Randall, D. A.:
Cloud Resolving Modeling of the ARM Summer 1997 IOP: Model Formulation, Results, Uncertainties, and Sensitivities, J. Atmos. Sci., 60, 607 – 625, https://doi.org/10.1175/1520-0469(2003)060<0607:CRMOTA>2.0.CO;2, 2003. a
Kreuzer, M., Reese, R., Huiskamp, W. N., Petri, S., Albrecht, T., Feulner, G., and Winkelmann, R.:
Coupling framework (1.0) for the PISM (1.1.4) ice sheet model and the MOM5 (5.1.0) ocean model via the PICO ice shelf cavity model in an Antarctic domain, Geosci. Model Dev., 14, 3697–3714, https://doi.org/10.5194/gmd-14-3697-2021, 2021. a
Lazeroms, W. M. J., Jenkins, A., Gudmundsson, G. H., and van de Wal, R. S. W.:
Modelling present-day basal melt rates for Antarctic ice shelves using a parametrization of buoyant meltwater plumes, The Cryosphere, 12, 49–70, https://doi.org/10.5194/tc-12-49-2018, 2018. a
Madec, G., Delecluse, P., Imbard, M., and Lévy, C.:
OPA 8.1 Ocean General Circulation Model reference manual, Note du Pole de Modelisation, Institut Pierre-Simon Laplace, Paris, France, 91 pp., 1998. a
Mathiot, P., Jenkins, A., Harris, C., and Madec, G.:
Explicit representation and parametrised impacts of under ice shelf seas in the z∗ coordinate ocean model NEMO 3.6, Geosci. Model Dev., 10, 2849–2874, https://doi.org/10.5194/gmd-10-2849-2017, 2017. a
Naughten, K. A., De Rydt, J., Rosier, S. H. R., Jenkins, A., Holland, P. R., and Ridley, J. K.:
Two-timescale response of a large Antarctic ice shelf to climate change, Nat. Commun., 12, 2041–1723, https://doi.org/10.1038/s41467-021-22259-0, 2021. a
Nilsson, J., Jakobsson, M., Borstad, C., Kirchner, N., Björk, G., Pierrehumbert, R. T., and Stranne, C.:
Ice-shelf damming in the glacial Arctic Ocean: dynamical regimes of a basin-covering kilometre-thick ice shelf, The Cryosphere, 11, 1745–1765, https://doi.org/10.5194/tc-11-1745-2017, 2017. a
O'Gorman, P. A. and Dwyer, J. G.:
Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events, J. Adv. Model. Earth Sy., 10, 2548–2563, https://doi.org/10.1029/2018MS001351, 2018. a
Pritchard, H. D., Ligtenberg, S. R. M., Fricker, H. A., Vaughan, D. G., van den Broeke, M. R., and Padman, L.:
Antarctic ice-sheet loss driven by basal melting of ice shelves, Nature, 484, 502–505, https://doi.org/10.1038/nature10968, 2012. a
Radford, A., Metz, L., and Chintala, S.:
Unsupervised Representation Learning with Deep Convolutional Generative
Adversarial Networks, in: 4th International Conference on Learning Representations, ICLR 2016, 2–4 May 2016,
San Juan, Puerto Rico, Conference Track Proceedings, edited by: Bengio, Y. and LeCun,
Y., http://arxiv.org/abs/1511.06434, 2016. a
Ramachandran, P., Zoph, B., and Le, Q. V.:
Searching for Activation Functions, CoRR, arXiv [cs.Ne], arXiv:1710.05941, 2017. a
Rasp, S., Pritchard, M. S., and Gentine, P.:
Deep learning to represent subgrid processes in climate models, P. Natl. Acad. Sci. USA, 115, 9684–9689, https://doi.org/10.1073/pnas.1810286115, 2018. a
Reese, R., Albrecht, T., Mengel, M., Asay-Davis, X., and Winkelmann, R.:
Antarctic sub-shelf melt rates via PICO, The Cryosphere, 12, 1969–1985, https://doi.org/10.5194/tc-12-1969-2018, 2018a. a, b
Reese, R., Gudmundsson, G. H., Levermann, A., and Winkelmann, R.:
The far reach of ice-shelf thinning in Antarctica, Nat. Clim. Change, 8, 53–57, https://doi.org/10.1038/s41558-017-0020-x, 2018b. a, b
Ronneberger, O., Fischer, P., and Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation, in: Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, edited by: Navab, N., Hornegger, J., Wells, W., and Frangi, A., MICCAI 2015, Lecture Notes in Computer Science, Springer, Cham, vol. 9351, https://doi.org/10.1007/978-3-319-24574-4_28, 2015. a
Rosier, S. H. R.: shrrosier/MELTNET: v1.0.0 (1.0.0), Zenodo [code], https://doi.org/10.5281/zenodo.7018247, 2022. a
Seroussi, H., Nakayama, Y., Larour, E., Menemenlis, D., Morlighem, M., Rignot, E., and Khazendar, A.:
Continued retreat of Thwaites Glacier, West Antarctica, controlled by bed topography and ocean circulation, Geophys. Res. Lett., 44, 6191–6199, https://doi.org/10.1002/2017GL072910, 2017. a
Smith, R. S., Mathiot, P., Siahaan, A., Lee, V., Cornford, S. L., Gregory, J. M., Payne, A. J., Jenkins, A., Holland, P. R., Ridley, J. K., and Jones, C. G.:
Coupling the U. K. Earth System Model to Dynamic Models of the Greenland and Antarctic Ice Sheets, J. Adv. Model. Earth Sy., 13, e2021MS002520, https://doi.org/10.1029/2021MS002520, 2021. a
Stevens, B. and Bony, S.:
What Are Climate Models Missing?, Science, 340, 1053–1054, https://doi.org/10.1126/science.1237554, 2013. a
Thoma, M., Determann, J., Grosfeld, K., Goeller, S., and Hellmer, H. H.:
Future sea-level rise due to projected ocean warming beneath the Filchner Ronne Ice Shelf: A coupled model study, Earth Pl. Sc. Lett., 431, 217–224, https://doi.org/10.1016/j.epsl.2015.09.013, 2015.
a
Thomas, R. H.:
Ice Shelves: A Review, J. Glaciol., 24, 273–286, https://doi.org/10.3189/S0022143000014799, 1979. a
Zhang, K., Zuo, W., Chen, Y., Meng, D., and Zhang, L.:
Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising, IEEE T. Image Process., 26, 3142–3155, https://doi.org/10.1109/tip.2017.2662206, 2017. a
Short summary
Future ice loss from Antarctica could raise sea levels by several metres, and key to this is the rate at which the ocean melts the ice sheet from below. Existing methods for modelling this process are either computationally expensive or very simplified. We present a new approach using machine learning to mimic the melt rates calculated by an ocean model but in a fraction of the time. This approach may provide a powerful alternative to existing methods, without compromising on accuracy or speed.
Future ice loss from Antarctica could raise sea levels by several metres, and key to this is the...