EGU24-1729, updated on 08 Mar 2024
https://doi.org/10.5194/egusphere-egu24-1729
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Revisiting the iceberg thickness distribution in Southern Ocean simulations.

Anna Olivé Abelló1, Pierre Mathiot1, Nicolas Jourdain1, Yavor Kostov2, and Paul Holland2
Anna Olivé Abelló et al.
  • 1Université Grenoble Alpes/CNRS/IRD/INRAE/G-INP, Institut des Géosciences de l’Environnement, Grenoble, France., IGE, Grenoble, France (anna.olive-abello@univ-grenoble-alpes.fr)
  • 2British Antarctic Survey, Cambridge, UK

The acceleration of glaciers in the Antarctic ice sheet amplifies the flow of icebergs into the Southern Ocean. The presence of these icebergs has a significant impact on the penetration of warm water toward ice shelves and can also induce the formation of large polynyas when grounded, thereby promoting dense water formation. The existing implementation of the Lagrangian iceberg module in NEMO does not consider the Antarctic ice-shelf thicknesses from which the icebergs are originated, so the model cannot represent whether icebergs are thin enough to cross the shallow bathymetric ridges. In the present model, the iceberg masses, thicknesses, and size distribution are prescribed a priori as input parameters irrespective of the source ice-shelf characteristics. In addition, the categorization of iceberg classes and the scaling of smaller icebergs are not optimized, which is a strong limitation for climate modelling. Hence, the main aim of this study is to improve the thickness distribution of the calved icebergs based on ice shelf characteristics, decrease the computational cost of the model, and assess how these improvements alter the lifespan of the icebergs and their freshwater flux distribution across Antarctica.

The new approach has been implemented in a 0.25° Southern Ocean configuration of the NEMO ocean–sea-ice model. We used a power-law probability distribution function of iceberg occurrence as a function of iceberg area and a tabular iceberg definition to establish the thickness distribution for the small iceberg categories, imposing that each class exhibits the same total mass. In order to reduce computational costs, we constrained the frequency of icebergs released per class so that the smaller classes of multiple icebergs are gathered into one particle. Our preliminary results show that the iceberg thickness distribution, implemented as a function of areas, is supported by in-situ observations measured from high-resolution SAR-1 satellite images. The released icebergs display a typical thickness per class depending on the ice shelf's source, with a broader distribution when more calving classes are established. Ultimately, the findings reveal that accounting for realistic Antarctic ice-shelf thicknesses leads to thicker icebergs, particularly in larger classes, consequently increasing the mass that each transports westward around Antarctica. Future iceberg simulations, carried out for 25 years, will assess the iceberg's sensitivity to the maximum iceberg area, the number of different sizes and the area bounds used to define each size, among others. It is also expected that these simulations will also unveil high-melting regions and high iceberg densities in regions where icebergs ground, and will determine if fragmentation processes are needed to achieve realistic iceberg lifespans.

How to cite: Olivé Abelló, A., Mathiot, P., Jourdain, N., Kostov, Y., and Holland, P.: Revisiting the iceberg thickness distribution in Southern Ocean simulations., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1729, https://doi.org/10.5194/egusphere-egu24-1729, 2024.

Supplementary materials

Supplementary material link

Comments on the supplementary material

AC: Author Comment | CC: Community Comment | Report abuse

supplementary materials version 1 – uploaded on 14 Apr 2024, no comments