Which ecmwf files/fields needed to initialize and bc ems?

All issues/questions about EMS v3.4 package, please ask here.
Post Reply
theocarter2911
Posts: 65
Joined: Fri Mar 08, 2013 5:22 am

Which ecmwf files/fields needed to initialize and bc ems?

Post by theocarter2911 » Mon May 12, 2014 6:38 am

Hi,
Am going to get a quote from ecmwf to intialize and bc a 48 hour run of EMS but I need to work out what the minimum data necessary is in order to minimize the cost (ecmwf data is quite dear!). If anyone else is or has done this, can you help?

I realise one can limit the area of data, and also limit the steps (either 6 hrly or 3 hrly), but it looks like the cost is also heavily dependant on amount of actual fields needed.

My aim here is to have two seperate global scale models feeding into mesoscale level for the same area. We already use gfs and feed that both into nmm and into arw. But it would be better to have a totally separate global MOAD data stream.

From first impressions it looks like model level parameters needed are z/q/u/v/t/d/vo and then surface level parameters are sp/msl/skt/2t/10u/10v/2d/z/lsm/sst/ci/sd/stl1/stl2/stl3/stl4/swvl1/swvl2/swvl3/swvl4..

Can anyone confirm that this holds true?

Thanks,
Theo

meteoadriatic
Posts: 1566
Joined: Wed Aug 19, 2009 10:05 am

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by meteoadriatic » Mon May 12, 2014 8:45 am

Well, you need those fields that are listed in Vtable.ECMWF

I guess that some can be skipped (if you don't use noah lsm I think that those underground layers things are not mandatory).

Code: Select all

cat wrfems/data/tables/vtables/Vtable.ECMWF
I personally see no commercial point using ECMWF data, it is just too expensive for advantage it gives. If you can get it for free then it is probably better choice than GFS ;)

theocarter2911
Posts: 65
Joined: Fri Mar 08, 2013 5:22 am

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by theocarter2911 » Mon May 12, 2014 10:16 am

Hi meteoadriatic,

I agree it is very expensive, but there is a small chance it may be viable for me! Fingers crossed..

Quick question, I see there is a Vtable for both model levels and for normal levels - can you think of any reason why one would be preferable over the other, the only thing I can think of is that the model levels would be a huge amount more than standard levels, and so may be more costly perhaps? For feeding a mesoscale model I cannot offhand think of another reason to prefer one over the other?

Theo

meteoadriatic
Posts: 1566
Joined: Wed Aug 19, 2009 10:05 am

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by meteoadriatic » Mon May 12, 2014 6:22 pm

Never worked with ECMWF so I just guessing. I don't know how much (sigma or postprocessed) levels it has and would it be better to use sigma vtable or "normal"... probably better is that what has more levels available ;) (= more expensive?)

theocarter2911
Posts: 65
Joined: Fri Mar 08, 2013 5:22 am

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by theocarter2911 » Thu May 15, 2014 6:24 pm

Hi Meteoadriatic,
The model (sigma) levels are something like 134! Whereas standard levels will be significantly less, similar to gfs. And in any case, ungrib will just interpolate those 134 levels to standard ones anyway for wrf run. So I will go for the standard levels. Thanks for your thoughts!
Theo

julien1188
Posts: 36
Joined: Fri Mar 09, 2012 1:02 pm

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by julien1188 » Fri May 16, 2014 12:36 pm

use pressure level. It is more easier to integrate to WPS

theocarter2911
Posts: 65
Joined: Fri Mar 08, 2013 5:22 am

Re: Which ecmwf files/fields needed to initialize and bc ems?

Post by theocarter2911 » Sun May 18, 2014 1:56 pm

Thanks Julien, will do that. Theo

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests