A hybrid sea ice model to estimate the impact of floe scale sea ice-ocean-atmosphere coupling on the Antarctic sea ice evolution
A number of mechanisms have been proposed to explain the decrease and increase of the Antarctic sea ice extent in the recent years. But the processes that drive this evolution are not well understood. The simulation of Antarctic sea ice in current climate models remains a fundamental problem and the reason for this shortcoming is a current research question. But there is some evidence that this stems, in addition to the formulation of atmospheric and oceanic processes, also from the description of the sea ice physics in the Southern Ocean. Even though much of the current sea ice cover in the Southern Ocean resembles a marginal ice zone, continuum sea ice models usually do not resolve sea ice floes nor parameterize this regime and neglect important feedbacks on climate and weather. Furthermore, the application of continuum sea ice models at or below the resolution of individual floes is questionable as the underlying continuum assumption of those sea ice models likely breaks down. In this proposal we will address these shortcomings of current continuum sea ice models used in climate models by developing a hybrid sea ice model, that explicitly describes atmosphere, sea ice and ocean interactions up to the floe scale. The hybrid approach provides a seamless model framework to predict the sea ice state, ranging from interacting sea ice floes in the marginal ice zone up to pack ice. The development of new numerical models and their validation to improve the understanding of Polar Processes and Mechanisms is a central aspect of the current call. Our hybrid model, which combines particle with continuum methods, will contribute to a better understanding and prediction of the Antarctic climate system by explicitly including coupling and feedbacks between atmosphere, sea ice and ocean at the floe scale. Small scale processes related to individual floes are important to the polar climate, but their parameterization in continuum sea ice models remains a research question.To understand the impact of floe scale interactions on the evolution of the sea ice cover in the Southern Ocean, we will develop a Discrete Element sea ice model, based on the description of DESIgn and the Princeton DEM, and nest it into the continuum sea ice formulation in the climate model ICON. Our goal is to explicitly simulate discrete sea ice floes in a subdomain of interest such as the marginal ice zone. In regions where a high spatial resolution is not required, the simulation is based on the continuum model, which is a suitable, computational efficient, approach to describe the sea ice evolution at large scales and low resolutions. Using the hybrid approach, we will explicitly take the floe size distribution into account, which significantly impacts the simulated sea ice volume. Finally, owing to the particle approach used in the hybrid sea ice model, this project opens a pathway towards exascale sea ice modeling, including the possible use of GPUs.
Dr. Carolin Mehlmann
Institut für Analysis und Numerik
Tel.:+49 391 6752027