Davor Dundovic

• Dundovic, Davor; Wallwork, Joseph G.; Kramer, Stephan C.; Gillet-Chaulet, Fabien; Hock, Regine & Piggott, Matthew D. (2025). Anisotropic metric-based mesh adaptation for ice flow modelling in Firedrake. Geoscientific Model Development. ISSN 1991-959X. 18(13), s. 4023–4044. doi: 10.5194/gmd-18-4023-2025. Fulltekst i vitenarkiv Vis sammendrag

  Glaciological modelling is a computationally challenging task due to its high cost and complexity associated with large spatial-scale and long timescale simulations. In this paper we propose feature-based anisotropic mesh adaptation methods and demonstrate their effectiveness for time-dependent glaciological modelling on a Marine Ice Sheet Model Intercomparison Project (MISMIP+) experiment. Our methods use the Python-based Firedrake finite element library and the mmg2d remeshing software. We show that we are able to achieve solution accuracy comparable to uniform 0.5 km resolution mesh simulations using a sequence of adapted meshes with, on average, 30 times fewer vertices when adapting meshes based on the basal stress and 8–12 times fewer when adapting based on ice thickness and velocity. We further introduce a novel hybrid time-dependent fixed-point mesh adaptation algorithm that reaches mesh convergence approximately twice as fast compared with the existing global fixed-point algorithm. Since the fixed-point algorithms require that the problem is solved multiple times, the reported reduction in the number of vertices ultimately translates into a 3–6 times lower overall computational cost compared to uniform mesh simulations.

• Jordan, Connor; Dundovic, Davor; Fragkou, Anastasia K.; Deskos, Georgios; Coles, Daniel S. & Piggott, Matthew D. [Vis alle 7 forfattere av denne artikkelen] (2022). Combining shallow-water and analytical wake models for tidal-array micro-siting. Journal of Ocean Engineering and Marine Energy. ISSN 2198-6444. doi: 10.1007/s40722-022-00225-2. Fulltekst i vitenarkiv

Se alle arbeider i NVA