Resolving Leads in Sea-Ice Models : New Analysis Methods for Frontier Resolution Arctic Simulations
|Other Titles:||Modellieren von Meereisrinnen : Neue Analysemethoden für Arktische Simulationen mit extremen Auflösungen||Authors:||Hutter, Nils||Supervisor:||Jung, Thomas||1. Expert:||Jung, Thomas||2. Expert:||Haas, Christian||Abstract:||
Sea ice deforms constantly under the forcing of winds and ocean currents. Eventually the ice cover of the Arctic Ocean breaks into a multitude of ice floes. Strips of open ocean, so-called leads, and pressure ridges, where the collision of floes piled up the ice, are found along the floe boundaries. These features have a strong impact on the interaction of sea ice with the atmosphere and the ocean, as they affect heat loss and surface drag. Currently, climate models do not resolve leads and pressure ridges in simulated sea ice fields due to their coarse resolution. They parameterize the effects of leads on the Arctic climate, if at all. The goal of this thesis is to develop Arctic simulations that reproduce leads sufficiently to be used in climate simulations. By decreasing the horizontal grid-spacing, a numerical ocean sea-ice model is shown to resolve leads explicitly. To test how realistic these lead-resolving sea-ice simulations are, the following research questions are addressed: (1) what are good metrics to evaluate the simulated leads with observational data? (2) Which observed characteristics of sea ice deformation and deformation features are reproduced by the model? In a first step, the sea ice deformation in a 1-km lead-resolving sea-ice simulation is analyzed with a spatio-temporal scaling analysis. The simulated sea ice deformation is strongly localized in failure zones and dominated by spontaneous fracture. This heterogeneity and intermittency of sea ice deformation shows that the simulation captures the fracture processes that form leads. In a second step, two new algorithms are described that detect and track leads and pressure ridges, combined into Linear Kinematic Features (LKFs). Both algorithms are applied to deformation data observed from satellite to establish a data set of deformation features that can be used as a reference in model evaluation. LKFs in two lead-resolving sea-ice simulations are extracted with the same algorithms, and found to agree with the LKF data set with respect to their spatial characteristics and temporal evolutions. In conclusion, high resolution sea-ice simulations can explicitly resolve leads. These simulations reproduce the characteristics of sea ice deformation and the representation of LKFs that are both observed from satellite. In future work, these simulations could be used as prototypes for the configuration of the sea-ice component in a climate model to directly simulate air-ice-ocean interaction processes in the Arctic.
|Keywords:||Arctic, sea ice, sea-ice modelling, high-resolution, remote sensing, sea-ice observations, MITgcm, LKFs, leads, pressure-ridges, Linear Kinematic Features, lead-resolving sea-ice simulation, scaling, power-law, scaling analysis, fractal, multi-fractal, image detection, lead detection, automated detection, evaluation, validation||Issue Date:||15-Aug-2019||URN:||urn:nbn:de:gbv:46-00107794-12||Institution:||Universität Bremen||Faculty:||FB1 Physik/Elektrotechnik|
|Appears in Collections:||Dissertationen|
checked on Sep 23, 2020
checked on Sep 23, 2020
Items in Media are protected by copyright, with all rights reserved, unless otherwise indicated.