 |
| Titel |
Determination of the Effective Resolution of Regional Climate Models |
| VerfasserIn |
Kathrin Lisa Kapper, Andreas Gobiet, Heimo Truhetz |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250040870
|
|
|
|
|
|
| Zusammenfassung |
| The spatial resolution of numerical weather prediction and climate models is generally
determined by their grid spacing (Îx) or spectral truncation and the numerical
implementation of dynamical core and model parametrisations. For example features of the
scale 2Îx and 3Îx are smoothed to avoid numerical instabilities (e.g., aliasing effects) and
parameterisations in connection to advection, pressure gradient force, and subgrid-scale
diffusion can only be well represented at dimensions of at least four times the grid spacing.
Some parametrisations, however, generate energy at the grid-spacing scale. These
multiple effects on the effective resolution of models are investigated in this study for
three high resolution regional climate models (RCMs) in dependence of their grid
spacing.
In order to determine the effective resolution variance spectra of the model results at
various grid spacings are derived. The variance spectra are constructed by applying the
Discrete Cosine Transform (DCT) on the model fields. This Fourier-type transform can,
unlike many other methods, handle aperiodic fields. The variance is displayed in terms of
scales (wavelengths) of the analysed atmospheric field.
Variance spectra of three non-hydrostatic high-resolution models (PSU/NCAR model
MM5, its succesor WRF, and the German model of the national weather services CCLM)
with three different grid spacings (1km, 3km, 10km) on two vertical levels (near the surface
and at 700hPa) are compared. Furthermore, the model variance spectra are compared to
variance spectra of highly resolved gridded observational-based data (1km grid spacing), the
nowcasting system of the Austrian meteorological service (INCA) and a dense
observation network in Eastern Styria, Austria (Wegener Net, www.wegenernet.org),
and the Integrated Forecast System (IFS, -25km grid spacing) of the European
Centre for Medium-Range Weather Forecasts (ECMWF). Preliminary results show
that the effective resolution varies from 3Îx to 7Îx, depends on the investigated
parameter as well as on the model formulation, and is particularly sensitive to the
distance from the surface. More results of this analysis are going to be discussed in
detail. |
|
|
|
|
|
|