Sensitivity experiments to mountain representations in spectral models

This paper describes a set of sensitivity experiments to several formulations of orography. Three sets are considered: a "Standard" orography consisting of an envelope orography produced originally for the ECMWF model, a"Navy" orography directly from the US Navy data and a "...

Full description

Bibliographic Details
Main Authors: U. Schlese, E. Roeckner, S. Gualdi, M. Biasutti, A. Navarra, U. Shulzweida
Format: Article
Language:English
Published: Istituto Nazionale di Geofisica e Vulcanologia (INGV) 2000-06-01
Series:Annals of Geophysics
Subjects:
Online Access:http://www.annalsofgeophysics.eu/index.php/annals/article/view/3658
_version_ 1811261341700194304
author U. Schlese
E. Roeckner
S. Gualdi
M. Biasutti
A. Navarra
U. Shulzweida
author_facet U. Schlese
E. Roeckner
S. Gualdi
M. Biasutti
A. Navarra
U. Shulzweida
author_sort U. Schlese
collection DOAJ
description This paper describes a set of sensitivity experiments to several formulations of orography. Three sets are considered: a "Standard" orography consisting of an envelope orography produced originally for the ECMWF model, a"Navy" orography directly from the US Navy data and a "Scripps" orography based on the data set originally compiled several years ago at Scripps. The last two are mean orographies which do not use the envelope enhancement. A new filtering technique for handling the problem of Gibbs oscillations in spectral models has been used to produce the "Navy" and "Scripps" orographies, resulting in smoother fields than the "Standard" orography. The sensitivity experiments show that orography is still an important factor in controlling the model performance even in this class of models that use a semi-lagrangian formulation for water vapour, that in principle should be less sensitive to Gibbs oscillations than the Eulerian formulation. The largest impact can be seen in the stationary waves (asymmetric part of the geopotential at 500 mb) where the differences in total height and spatial pattern generate up to 60 m differences, and in the surface fields where the Gibbs removal procedure is successful in alleviating the appearance of unrealistic oscillations over the ocean. These results indicate that Gibbs oscillations also need to be treated in this class of models. The best overall result is obtained using the "Navy" data set, that achieves a good compromise between amplitude of the stationary waves and smoothness of the surface fields.
first_indexed 2024-04-12T19:02:47Z
format Article
id doaj.art-0dbc46a8363441e481f99fa45789cdf2
institution Directory Open Access Journal
issn 1593-5213
2037-416X
language English
last_indexed 2024-04-12T19:02:47Z
publishDate 2000-06-01
publisher Istituto Nazionale di Geofisica e Vulcanologia (INGV)
record_format Article
series Annals of Geophysics
spelling doaj.art-0dbc46a8363441e481f99fa45789cdf22022-12-22T03:20:06ZengIstituto Nazionale di Geofisica e Vulcanologia (INGV)Annals of Geophysics1593-52132037-416X2000-06-0143310.4401/ag-3658Sensitivity experiments to mountain representations in spectral modelsU. SchleseE. RoecknerS. GualdiM. BiasuttiA. NavarraU. ShulzweidaThis paper describes a set of sensitivity experiments to several formulations of orography. Three sets are considered: a "Standard" orography consisting of an envelope orography produced originally for the ECMWF model, a"Navy" orography directly from the US Navy data and a "Scripps" orography based on the data set originally compiled several years ago at Scripps. The last two are mean orographies which do not use the envelope enhancement. A new filtering technique for handling the problem of Gibbs oscillations in spectral models has been used to produce the "Navy" and "Scripps" orographies, resulting in smoother fields than the "Standard" orography. The sensitivity experiments show that orography is still an important factor in controlling the model performance even in this class of models that use a semi-lagrangian formulation for water vapour, that in principle should be less sensitive to Gibbs oscillations than the Eulerian formulation. The largest impact can be seen in the stationary waves (asymmetric part of the geopotential at 500 mb) where the differences in total height and spatial pattern generate up to 60 m differences, and in the surface fields where the Gibbs removal procedure is successful in alleviating the appearance of unrealistic oscillations over the ocean. These results indicate that Gibbs oscillations also need to be treated in this class of models. The best overall result is obtained using the "Navy" data set, that achieves a good compromise between amplitude of the stationary waves and smoothness of the surface fields.http://www.annalsofgeophysics.eu/index.php/annals/article/view/3658General circulationGCM
spellingShingle U. Schlese
E. Roeckner
S. Gualdi
M. Biasutti
A. Navarra
U. Shulzweida
Sensitivity experiments to mountain representations in spectral models
Annals of Geophysics
General circulation
GCM
title Sensitivity experiments to mountain representations in spectral models
title_full Sensitivity experiments to mountain representations in spectral models
title_fullStr Sensitivity experiments to mountain representations in spectral models
title_full_unstemmed Sensitivity experiments to mountain representations in spectral models
title_short Sensitivity experiments to mountain representations in spectral models
title_sort sensitivity experiments to mountain representations in spectral models
topic General circulation
GCM
url http://www.annalsofgeophysics.eu/index.php/annals/article/view/3658
work_keys_str_mv AT uschlese sensitivityexperimentstomountainrepresentationsinspectralmodels
AT eroeckner sensitivityexperimentstomountainrepresentationsinspectralmodels
AT sgualdi sensitivityexperimentstomountainrepresentationsinspectralmodels
AT mbiasutti sensitivityexperimentstomountainrepresentationsinspectralmodels
AT anavarra sensitivityexperimentstomountainrepresentationsinspectralmodels
AT ushulzweida sensitivityexperimentstomountainrepresentationsinspectralmodels