Premium
Predictive skill of the UKMO unified model in the lower stratosphere
Author(s) -
Lahoz W. A.
Publication year - 1999
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1002/qj.49712555813
Subject(s) - stratosphere , troposphere , geopotential height , polar vortex , forecast skill , atmospheric sciences , environmental science , climatology , meteorology , geology , precipitation , geography
This paper studies the short‐term (up to 10 days) predictive skill of the UK Meteorological Office Unified Model (UM) in the lower stratosphere by applying conventional verification statistics used in numerical weather prediction studies. Simulations from four versions of the UM, each of them having a different representation of the stratosphere, are used to assess the dependence of the model's skill on stratospheric vertical resolution, location of the highest model full‐level and location of the highest model half‐level. Results are shown for two winter periods: February 1994 (northern winter with a weak wave‐number two warming) and October 1994 (southern winter with a typical top‐down break‐up of the stratospheric polar vortex). the results of this paper indicate that the model's skill in the lower stratosphere depends on a number of factors, including flow structure, stratospheric vertical resolution and the location of the highest model full‐level. By contrast, the model forecast skill in the mid troposphere is insensitive to the stratospheric representation on the 10‐day time‐scale. the highest skill (10 days and above for temperature and geopotential height at 50 hPa) is associated with steady and deep stratospheric flows which span the upper troposphere. the poorest skill is associated with rapidly varying flows; flows with shallow vertical structure also tend to have relatively poor skill. In all cases the model's skill is a significant improvement over persistence. the model has higher forecast skill in the lower stratosphere than in the mid troposphere; it also has higher skill in northern winter than in southern winter: these differences in skill can be ascribed to the flow regime in the lower stratosphere being dominated by lower wave numbers than in the mid troposphere, and to larger initialization errors in the southern hemisphere. It is shown that higher stratospheric vertical resolution does not necessarily give higher skill in the lower stratosphere; this is particularly true in northern winter. the model results show evidence of links between tropospheric and stratospheric errors.