[2022-present] The Southern Ocean Feshwater Input from Antarctica (SOFIA) initiative is a collabaritve effort of a group of initially 9—but quickly growing—climate modeling groups across the globe to access the role of ocean model formulations in climate system response to enhanced melting of the Antarctic ice sheet. The initiative was started by curent and former members of the CLIVAR/CliC/SCAR Southern Ocean Region Panel (SORP) to reconcile recently published work on this topic using varius models and scenarios. An initial paper providing background information and describing the experiment protocol is accepted and soon to appear in GMD. With a set of Tier 1 pre-industrial control experiments we currently quantify model uncertainty. Tier 2 experiements will provide projections of ocean and climate repsones until 2100. And Tier 3 collects experiments focused at improving the realism of meltwater release into climate model oceans. The initiative is recognized as an official CLIVAR task team. And since Nov. 2023 SOFIAMIP has been registered with CMIP.   

[2022-2024] The Helmholtz Changing Earth project Storyline Scenarios of Extreme Weather, Climate, and Environmental Events along with their Impacts in a Warmer World (SCENIC) is an innovative multi-center community project with the goal "to develop and apply a novel storyline approach to examine how extreme events would unfold in different climates, thereby complementing classical climate scenario methods. Furthermore, novel ways of communicating uncertainty will be enabled by differentiating between dynamic (high uncertainty) and thermodynamic drivers of change." At GEOMAR we will specifically study the cascade of events leading to extreme melting rates on Greenland, their impact on the ocean, and their implications under global warming.

[2020-2023] With PalMod going into round 2, we now apply our nested ocean global climate model to the Southern Ocean (FOCI-ORION10X) to investigate redistribution pathways and consequences of enhanced meltwater input from Antarctica. For the interaction of the meltwater with the open ocean, the Antarctic slope current plays in important role, which can only be simulated properly by high resolution models. We study the importance of mesoscale dynamics in heat uptake as well as redistribution and freshwater export routes from the Antarctic Circumpolar Current to the continental shelf. 

[2020-2024] Greenland Ice Sheet–melting exiting ocean extremes is an ongoing, DFG-funded project in which we seek to identify the time of emergence of enhanced ice-sheet melt in the open ocean and its potential to cause unusual, extreme conditions in the ocean at Europe's doorstep, the North Atlantic. We use the global climate model FOCI-VIKING10 with grid refinement in the subpolar North Atlantic enabling explicit simulation of mesoscale ocean processes. While validating the model with gridded data products of ocean salinity and temperature, we also study ways and usefullness of defineing extreme events. Various scenarios of realistically enhanced meltwater runoff into the ocean are explored to study the impact of accelreated Greenland Ice Sheet–melting over the next couple of decades. 

[2015-2019] The PalMod project is an ambitious national German initiative funded by the BMBF to simulate an entire glacial cycle, i.e. from 135,000 years ago to the present, with complex Earth system models. As a sidecar in phase 1, we study the role of ocean model formulation, atmosphere coupling and mesoscale ocean dynamics in redistribution and impact of enhanced meltwater input to the subpolar North Atlantic. With FOCI (see below) we can run centennial sensitivity simulations with an eddy-rich ocean in the region of interest, here 30-85N (FOCI-VIKING10).

A model incomparison of Greenland freshwater release at 0.05 Sv focusing on the ocean components identifies robust responses but also the non-negligible impact of internal variability and model uncertainty. We also showed the relatively rapid reversibility of upper ocean reponses but centennial-scale lasting effects on the deep ocean. 

Rivers and meltwater from ice sheets are important sources of freshwater to the ocean and are often associated with distinct locations such as river mouths and outlet glaciers. As ocean models are applied to increasingly finer spatial grids, it is crucial  that the runoff provided as boundary conditon matches the  actual model coastline.  Because of its descrete spatial distribution runoff forcing fields cannot be simply interpoalted onto the model grid but must be remapped and assigned to matching coastal ocean grid nodes. 

Starting from the JRA55-do  atmospheric forcing product for ocean models I coded a versatile, conservative remapping routine suitable for transferring runoff to ocean grids applicabale for coarse to high-resolution (e.g. 1/20˚) destination grids. The routine  reassigns runoff based on distance  between source and destination location keeping the focus on major runoff locations. Spreading of runoff at river mouths proportional to magnitude is also enabled. 

[2012-2014] The drastic thinning and retreat of the Arctic sea-ice cover over the past decades has dynamical consequences for the ocean underneath. Working with Mike Steele at the Polar Science Center I found and quantified that the weakened sea-ice cover causes an enhanced momentum influx into the Arctic Ocean on annual average and is responsible for a 23% increase of the ocean surface stress from 1979 to 2012. However, we also discovered that the extensive retreat of the sea ice, which has a rougher surface due to its deformations and floe egdes compared to open water, yields a reduction of ocean surface stress in the summer months. As sumer ice conditions expand into fall with global warming, a new regime of momentum transfer can be expected to emerge.

A follow up study based on a model with variable sea-ice roughness highlights the competing effects of decreasing roughness and ice strength--both due to sea ice thinning--on the momentum transfer.

[2010-2012] Large polynyas, kind of "windows" in the pack ice, due to open ocean deep convection in the Southern Ocean are a rare phenomenon. As the high southern latitudes still pose a challenge to global climate models, these feature various expressions of this process. In the Kiel Climate Model (KCM) we identified a mechanism driving the recurrence of such events, involving slow heat accumulation at intermediate ocean depths and its rapid release to the atmosphere in polynya years. This variability is present in many CMIP simulations and it turns out that sea ice volume and stratification of the Southern Ocean are its key controls common to different models. The phenomenon— if strong enough—drives low frequent variability of the Atlantic meridional overturning circulation and warming over Antarctica. The variability patterns associated with the shutdown of the dee convection have the potential to mask global warming trends.

[2008-2009] Icebergs have been overlooked as an important part of the freshwater cycle in polar ocean and cliamte modelling. They carry land ice and thus freshwater away from the coast of their source region and alter watermass properties farther offshore when they melt. 

As a postdoc at GFDL together with Alistair Adcroft I took a Lagrangian approach to simulate up to 100,000 individual icebergs in a complex coupled climate model.  We showed that it was possible to integrate this new component into the closed water cycle of the coupled model and create a more realistic distribution of meltwater. This distinct pattern of meltwater from drifting icebergs impacts both sea-ice thickness and bottom water formation around Antarctica, where larger icebergs, which resist melt and erosion longer, can carry the freshwater all the way into the Antarctic Circumpolar Current. The impact of icebergs from Greenland is less pronounced because icebergs are smaller here, thus have a reduced lifetime and mostly melt in the boundary currents joining the liquid runoff from Greenland. 

The iceberg model code has been picked up by the NEMO ocean model community.