Jump to content
Snow?
Local
Radar
Cold?
IGNORED

Lies, Damned lies and Statistics!


Norway Nut

Recommended Posts

Posted
  • Location: Austevoll Kommune North of 60 deg N
  • Weather Preferences: Cold with a metre of lying snow
  • Location: Austevoll Kommune North of 60 deg N

Has anyone the stats on verification of models at different "T+" intervals? I've just seen a post from SM with the stats for the various models at T+144 and they range from just over 80% to about 77%, what I would like to know is the verification stats out to the limit of the model run.

 

The reason for this post is that I'm tired of reading about a "great looking T+284 chart from GFS" only to see it disappear at T+240 and never reappear.

 

I'm expecting something like 100% at T0, falling linearly to roughly 80% @ T144 and then falling exponentially to something like 5% at T284.

 

I hope someone can oblige!

 

Many thanks in anticipation

 

NN

Link to comment
Share on other sites

Posted
  • Location: Eastbourne, East Sussex (work in Mid Sussex)
  • Location: Eastbourne, East Sussex (work in Mid Sussex)
Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

Has anyone the stats on verification of models at different "T+" intervals? I've just seen a post from SM with the stats for the various models at T+144 and they range from just over 80% to about 77%, what I would like to know is the verification stats out to the limit of the model run.

 

The reason for this post is that I'm tired of reading about a "great looking T+284 chart from GFS" only to see it disappear at T+240 and never reappear.

 

I'm expecting something like 100% at T0, falling linearly to roughly 80% @ T144 and then falling exponentially to something like 5% at T284.

 

I hope someone can oblige!

 

Many thanks in anticipation

 

NN

 

coast has given the links but I would pretty much agree with you suggested accuracy levels, perhaps a touch lower at 144h?

Link to comment
Share on other sites

Posted
  • Location: Sheffield South Yorkshire 160M Powering the Sheffield Shield
  • Weather Preferences: Any Extreme
  • Location: Sheffield South Yorkshire 160M Powering the Sheffield Shield

It's a pity they don't have any scores for something T240. May stop some of the nailed on nonsense we see when the model forum goes on hyperdrive.

Link to comment
Share on other sites

Posted
  • Location: The Netherlands
  • Location: The Netherlands

I hope this isn't already included, but the Canadian warning centre has verification stats for all models (I don't know how those statistics are compiled, though)

The verification stats are for 1 day and 5 days in advance over the last few years.

 

Posted Image

The yearly verification stats at T24h.

 

Posted Image

The yearly verification stats at T120h.

 

Furthermore, error growth charts can be found on the same site for each month. For example, the verification stats for the month October are as follows:

Posted Image

 

 

And from the link below, error propagation charts of different months in 2013 can be found.

http://weather.gc.ca/verification/error_growth_e.html

 

Finally, the World Meteorological Organization also does research on model verification, and they might have some interesting articles as well:

http://www.wmo.int/pages/prog/arep/wwrp/new/Forecast_Verification.html

 

I hope this helped a little.

 

Sources: 

http://weather.gc.ca/verification/monthly_ts_e.html

http://www.wmo.int/pages/prog/arep/wwrp/new/Forecast_Verification.html

Edited by Vorticity0123
  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...