Towards seasonal prediction: stratosphere-troposphere coupling in the atmospheric model ICON-NWP

• Main Author: Köhler, Raphael (M.Sc.)
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: 2020
• Subjects: ddc:530; Institut für Physik und Astronomie; Arctic; Climate change
• Online Access: https://publishup.uni-potsdam.de/frontdoor/index/index/docId/48723; https://nbn-resolving.org/urn:nbn:de:kobv:517-opus4-487231; https://doi.org/10.25932/publishup-48723; https://publishup.uni-potsdam.de/files/48723/koehler_diss.pdf

Stratospheric variability is one of the main potential sources for sub-seasonal to seasonal predictability in mid-latitudes in winter. Stratospheric pathways play an important role for long-range teleconnections between tropical phenomena, such as the quasi-biennial oscillation (QBO) and El Niño-Southern Oscillation (ENSO), and the mid-latitudes on the one hand, and linkages between Arctic climate change and the mid-latitudes on the other hand. In order to move forward in the field of extratropical seasonal predictions, it is essential that an atmospheric model is able to realistically simulate the stratospheric circulation and variability. The numerical weather prediction (NWP) configuration of the ICOsahedral Non-hydrostatic atmosphere model ICON is currently being used by the German Meteorological Service for the regular weather forecast, and is intended to produce seasonal predictions in future. This thesis represents the first extensive evaluation of Northern Hemisphere stratospheric winter circulation in ICON-NWP by analysing a large set of seasonal ensemble experiments. An ICON control climatology simulated with a default setup is able to reproduce the basic behaviour of the stratospheric polar vortex. However, stratospheric westerlies are significantly too weak and major stratospheric warmings too frequent, especially in January. The weak stratospheric polar vortex in ICON is furthermore connected to a mean sea level pressure (MSLP) bias pattern resembling the negative phase of the Arctic Oscillation (AO). Since a good representation of the drag exerted by gravity waves is crucial for a realistic simulation of the stratosphere, three sensitivity experiments with reduced gravity wave drag are performed. Both a reduction of the non-orographic and orographic gravity wave drag respectively, lead to a strengthening of the stratospheric vortex and thus a bias reduction in winter, in particular in January. However, the effect of the non-orographic gravity wave drag on the stratosphere is stronger. A third ...