GFS Evaluation - Part 2

GFS evaluation experiment - Part 2
Yesterday I showed the GFS t+144 hour charts valid for 12GMT on Thursday 20th September. Twenty four hours later we're looking at the t+120 hour charts from todays GFS 12z run which are valid for the same time and date.

12z charts generated on 15/09/2007

Temperature(C)


Rainfall rate (mm/hr)


Wind speed (km/hr) and direction


Finally, here's a link to an image showing the synoptic set up expected.
Chart example

Summary
The GFS on this run is continuing to show the UK in a rather unsettled westerly flow. To this extent it looks as though there is a good deal of consistency with yesterdays 12z run at least synoptically. Looking at the temperatures profile also shows a reasonable amount of consistency, although there are some local differences which are quite significant. The fun and games really begin when looking at the precipitation (rainfall) charts, and again there is some consistency, but there are some quite big changes with most of Scotland now set for a soaking if this run is correct. So small changes in the synoptic set up can result in very large changes in the actual weather we experience. This illustrates how difficult it is to accurately forecast rainfall more than a couple of days ahead, and it is quite likely that the GFS run we'll be looking at tomorrow will show something slightly different once again. Hopefully now you can start to see how difficult it can be to make accurate forecasts more than a few days ahead!

Issued 19:50, 15/09/2007

 

Brian Gaze's blog

MORE



COMPUTER MODELS

INFO

Short range
Short to medium range
Medium to long range
Deterministic
Ensemble

See the Model inventory for the full list of model charts and data