Influences of surface layer modelling approaches on model performance

• Main Author: Ries, Hinnerk
• Other Authors: Schlünzen, K. Heinke (Prof. Dr.)
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: Staats- und Universitätsbibliothek Hamburg Carl von Ossietzky 2010
• Subjects: SEA-ICE ZONE; MESOSCALE MODEL; CYCLONE; LAND-SEA BREEZE; VERTICAL RESOLUTION; LAND-SURFACE-MODEL; 550 Geowissenschaften; 38.84 Meteorologie: Sonstiges; Meereis; mesoskaliges Model; Land-See-Wind; vertikale Auflösung; Bodenmodell; ddc:550; Sea ice
• Online Access: http://nbn-resolving.de/urn:nbn:de:gbv:18-48720; https://ediss.sub.uni-hamburg.de/handle/ediss/3835

In this thesis the impact of surface layer modelling on the development of atmospheric phenomena is investigated. The main objective is to determine well suited model setups for the surface layer. The specifications are based on sensitivity studies of the number of atmospheric vertical levels, on the choice of the land surface model and on various ground parameters for a variety of atmospheric situations. To some extent mesoscale systems are always exposed to the synoptic scale forcing. As in this thesis only real-case scenarios are investigated, the model performance is assessed in relation to the dominating scale of the meteorological situation, its representation in the driving model and its consideration by the mesoscale model in form of initial values and boundary value nudging. Thus, another objective of this thesis is to give a recommendation on an appropriate nudging strength in dependence of the dominating scale of the meteorological situation. The impact of the number of atmospheric vertical levels and of the land surface model used is investigated with two different setups of the fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) for the bay of Valencia. In this region the predominant mesoscale circulation is the land-see-breeze. Setup M1 is characterised by 34 atmospheric vertical sigma levels combined with a five-ayer land surface model (5L LSM). M2 has 52 levels and the Noah LSM is used. Results of 71 forecasts of 2006 are evaluated with a focus on the accuracy of meteorological surface data. The accuracy improves for all meteorological parameters except pressure with the M2 setup. Detailed analysis of results for 4 days simulated with all combinations of vertical resolutions and LSMs shows that the increase of vertical resolution is responsible for the improved forecast performance. The Noah LSM generally improves temperature performance, but reduces the performance of all other parameters, especially for days with sea-breeze conditions. After this general evaluation of two ...