II sure wish we were allowed to do this sort of thing in the SOFT sciences.
A new tropical cyclone bogus scheme introduced by The Met. Office
In the last issue of nwp Gazette a new scheme for creating bogus observations around tropical cyclones was mentioned. Following successful trials this has now been introduced operationally at The Met. Office, Bracknell.
For several years forecasters have been able to automatically generate a set of bogus points representing a simple symmetric tropical cyclone vortex. Although this system has yielded acceptable results in terms of verification of forecast cyclone tracks in the short and medium term, it was felt that further improvements could be made.
Firstly, the radial structure of the wind is calculated analytically by using information given in Tropical Cyclone Advisory Messages received over the Global Telecommunication System (GTS), and by incorporating real observations when available. The second improvement is to introduce an asymmetric wind structure to the vortex by including a steering-flow vector.
A parallel trial lasting 19 days started at the end of August 1994. This involved running a five-day forecast once a day from data assimilation using the new scheme, in parallel with the operational forecast in real time. There were 12 active tropical cyclones during the trial period seven in the western north Pacific, four in the eastern and central north Pacific and one in the North Atlantic. The improvement in verification statistics, averaged over all cases, was striking. Analysis errors were cut by 65% (to 39 km); 72-hour forecast errors fell by 25% (to 394 km) and 120-hour errors fell by 46% (to 440 km). The model's skill over climatology/persistence methods improved dramatically from 4% to 40% at 24 hours, and from 7% to 37% at 72 hours. On the basis of these results the new bogus scheme was introduced operationally on 25 October 1994 at 12 UTC. The quality of tropical cyclone track forecasts issued to Regional Specialised Meteorological Centres (RSMCs) after this date is expected to show a noticeable improvement.
National meteorological services are reminded that for guidance on the development of tropical cyclones in their area they should always contact the appropriate Regional Specialised Meteorological Centre.
Nicholas S. Thompson
Professor of Psychology and Ethology
Clark University
[hidden email]
http://home.earthlink.net/~nickthompson/[hidden email]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/friam_redfish.com/attachments/20050817/2d973a8c/attachment.htm
Locked
