Norway's marine and terrestrial climate mapped with dynamical downscaling

Long-term numerical reconstructions on high spatial resolution of past weather are essential tools for studies of the local climate and climate extremes. The focus of this thesis has been to resolve the wind, precipitation, temperature and wave climate of Norway and Norwegian waters by high-resoluti...

Full description

Bibliographic Details
Main Author: Haakenstad, Hilde
Format: Doctoral or Postdoctoral Thesis
Language:English
Published: The University of Bergen 2022
Subjects:
Online Access:https://hdl.handle.net/11250/2998279
Description
Summary:Long-term numerical reconstructions on high spatial resolution of past weather are essential tools for studies of the local climate and climate extremes. The focus of this thesis has been to resolve the wind, precipitation, temperature and wave climate of Norway and Norwegian waters by high-resolution dynamical downscaling. Known as a hindcast archive, this is a well-known method to obtain local information based on more coarse atmospheric fields, typically reanalyses. Such reanalyses provide the state of the atmosphere as accurately as possible on meso-beta scale (20-200 km) whereas flow over complex terrain and along irregular coastlines requires resolutions on meso-gamma scale (2-20 km) or microscale (1 km or less) to be well represented. By using numerical weather prediction models tailored to high resolution modelling to downscale the reanalyses, we obtain far more detailed information than what a global reanalysis alone can give. In this thesis I focus on a convection-permitting non-hydrostatic downscaling and compare it to a hydrostatic hindcast as well as the host reanalysis. We see improvement in performance of the wind speed in both downscaling procedures, compared to the large scale reanalysis. However, extreme winds and precipitation are much better resolved by the convection-permitting non-hydrostatic model with better representation of convective features and the wind field in steep terrain and along irregular coastlines. We also find that the representation of polar lows is improved. Both atmospheric hindcasts are accompanied by wave hindcasts. We find that the wave field is sensitive to strong winds, and indeed the strongest winds (realistically) rendered by the non-hydrostatic NORA3 hindcast yields too strong wave growth. A new parameterization of the Charnock coefficient is explored and successfully used to generate a high-resolution wave hindcast based on the NORA3 atmospheric hindcast. Lange historiske rekonstruksjoner med høy romlig oppløsning av været som har vært, er et viktig verktøy ...