A Method for Extrapolation of Atmospheric Soundings

Meteorological models use terrain databases with resolutions on the order of model grid intervals (e.g., a terrain grid spacing of 4 km for a model with 4-km spacing). Generally, model software smoothes the terrain to a coarser resolution for computational stability (e.g., a 4-km grid model may have...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cogan, James, Reen, Brian
Format: Report
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Meteorological models use terrain databases with resolutions on the order of model grid intervals (e.g., a terrain grid spacing of 4 km for a model with 4-km spacing). Generally, model software smoothes the terrain to a coarser resolution for computational stability (e.g., a 4-km grid model may have terrain smoothed to about 10 km). In more complex terrain, the difference between the smoothed terrain and actual elevation could reach hundreds of meters, perhaps 1000 m for mountainous regions. To obtain meteorological information between the actual height and the model height or a nearby rawinsonde, current methods often use the temperature at the lowest level (nominal surface) and assume a standard lapse rate down to the level of interest, which is normally the elevation of the actual surface or of a finer terrain database. Pressure is computed from temperature and height. Many methods keep the surface wind speed and direction constant throughout the extrapolated section of the profile. These methods are often used to extrapolate an observed sounding (e.g., radiosonde) to a nearby lower elevation. Our method extracts information from the model-generated atmospheric profile or measured sounding to help generate the extrapolated profile. It takes into account surface-based inversions, although, at present, does not explicitly account for elevated inversions. The original document contains color images.