Page 1 of 1

gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 1:26 pm
by Carsten
Hello,

I currently run a forecast over Central Europe initialized by the gfsp25ptile data set. This worked fine for the last 2 years, but since yesterday I get much too low air temperatures alonge the costline of the north sea and the baltic. It seems that the water temperatures are much too cold. Does anybody know about this and how it can be fixed?
Thanks for help!

Carsten

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 3:47 pm
by meteoadriatic
Hello,

Do you run old EMS version? I ask because you put question in this section of forum, and your symptoms could also lead to such conclusion.

If yes you will need to upgrade because fro July 19 12z there is new gfs output that requires some changes in EMS.

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 4:25 pm
by Carsten
Hello,

I run 3.4.1.15.16. Probably this is the reason as the problems occure since July 19th...
Does this mean I have to upgrade to UEMS? Will the existing runs (directories from the 3.4.1.15.16 still work or do I have to setup new runs?

Thanks for your help!

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 4:32 pm
by meteoadriatic
I didn't tried to upgrade 3.4 with this, but try to download update file, extract manually somewhere outside your EMS, and replace your current ungrib binary, Vtable.GFS and METGRID.TBL.ARW files with those from update file. That might be enough..

Update file is here:
http://soostrc.comet.ucar.edu/strc/uems ... ly2017.tgz

I don't promise it will work!

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 4:55 pm
by Carsten
Thank you... I can't find the METGRID.TBL.ARW unless the other two files are in the package you mentioned. Is it somewhere else to be find?

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 6:06 pm
by Carsten
...and it seems to work without that missing file. Thanks once more!

Re: gfsp25ptile too cold water temperatures

Posted: Fri Jul 21, 2017 8:33 pm
by meteoadriatic
Yes really there is no METGRID.TBL.ARW update. Then it is OK that what you have most probably. Do detailed check of snow on the ground field, ground temperature (skin and lower levels if you can and also soil moisture - the best will be to use ncview to look into met_em* files - or wrfinput file). But if you don't want to do all that just verify that your snow on the ground looks without VERY strange things and then you are most probably fine.

Re: gfsp25ptile too cold water temperatures

Posted: Mon Jul 24, 2017 10:39 am
by Carsten
I will check the snow coverage - actually I do not expect a lot of snow in my domain...
By the way, do you know how to get an overview on the ptile datasets that are currently available? The list in the manual is quite old.

Thanky you very much for your help once more!

Re: gfsp25ptile too cold water temperatures

Posted: Mon Jul 24, 2017 11:10 am
by meteoadriatic
There is snow that gets somehow in domain from nowhere if land mask is not correct; you will see kilometers of snow on small areas near coast land in ncview (SNOW; SNOWH variables)- It will be obvious immediatelly and surface scheme will most probably either crash the model or temperatures there will be much colder than around...

I believe ems_prep --dslist will get list of datasets available from EMS servers.