Reanalysis based on ECMWF

Forum dedicated to older versions of EMS package (WRFEMS v3.2, v3.1 or older). Support is user-to-user based, so please help others if you can.
Posts: 1
Joined: Thu Feb 21, 2013 10:44 pm

Re: Reanalysis based on ECMWF

Post by quernic » Thu Feb 21, 2013 10:54 pm

Dear All

I am new at wrfems and trying to run era interim for the last a few years. I succeeded to run it with the data in However, i want to continue with 2012 data from ecmwf data portal. From that portal i could get required 3d and surface data with example python scripts but i could not use them in ems_prep. In surface and 3d data files are appendded into one file and it could be directly ungribed by ems_prep. Could you please explain the way for appending grib files or adding two files to ems_prep.

Thanks in advance

Posts: 94
Joined: Sat May 21, 2011 7:34 am

Re: Reanalysis based on ECMWF

Post by smartie » Fri Feb 22, 2013 9:46 pm

Download your data from the ECMWF portal.

For 3D fields you need only u wind, v wind, temperature, Rel Humidity and Geopotential Height.
You should have two files- one for 3D pressure level fields , the other for 2D surface files.

Put these in a temp dir with individual file names eg grib1.grib , grib2.grib. Cat the two files together-

Code: Select all

$ cat grib1.grib grib2.grib > ecmwf.grb
In .../wrfems/util/scripts you will find a PERL exe-
Copy or link this to your temp dir and run the script-

Code: Select all

$ ./   ecmwf.grb
The script should split the single file into separate date/time files which ems_prep can process.


Posts: 7
Joined: Sat Mar 22, 2014 12:25 am

Re: Reanalysis based on ECMWF

Post by Ponce10 » Wed Mar 26, 2014 1:30 am

Hi all,

I just entered into the world of wrf and especially wrf-ems. I am able to run simulations and create forcast outputs. So far so good. The next challange is the creation of historical analysis. On the long run I want to create time series of wind speed and wind direction for one cell.

As initial dataset(s) I want to use either ECMWF or MERRA (not quite sure yet).

This is where my problem starts: How can I integrate "near-to-realtime" data such as ECMWF or Merra as initial dataset for a long term simulation? When I download the data from the ECMWF homepage ist round about 1 GB per day (!), which is much more than I expected...

So the question is: How to do? Which files to modify? I tried the procedure described by smartie, but I couldn't locate the folder nor the file you mentioned.



By the way: I am using version 3.4.

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 3 guests