Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

Initializing WRF with Global ECMWF

This post was from a previous version of the WRF&MPAS-A Support Forum. New replies have been disabled and if you have follow up questions related to this post, then please start a new thread from the forum home page.

Gabriel Cassol

New member
Hello everyone, yesterday the European center released the raw data in grib2 format of some variables of the operational model. I would like to know if with these variables available, it is possible to initialize the WRF model? If yes, what is the vtable configuration I can use?

Data is available at this link: https://data.ecmwf.int/forecasts
 
Hi Gabriel,
Unfortunately we haven't had an opportunity to dig into that new dataset yet to do any testing. If you're interested in using it, I would recommend downloading it and making sure you are able to get all the mandatory variables for running WRF. You can then use g2print utility available in WPS/util to look at the codes for each variable and see if they match any of the existing Vtables. If not, I would advise using one of those Vtables to modify the codes to the new ones. If you do need to do that and are successful, please feel free to pass along the new Vtable and any other useful information you're gathered in the process!
 
Hi there! I was just reading these messages as I was looking for a Vtable for the publicly available grib2 files from ECMWF (https://data.ecmwf.int/forecasts/). I created the Vtable for these grib2 files which is attached to this message. Feel free to use it.

Cheers,
Casper
 

Attachments

  • Vtable_ECMWF_grib2.txt
    2 KB · Views: 258
CasperK said:
Hi there! I was just reading these messages as I was looking for a Vtable for the publicly available grib2 files from ECMWF (https://data.ecmwf.int/forecasts/). I created the Vtable for these grib2 files which is attached to this message. Feel free to use it.

Cheers,
Casper

Casper, first thank you very much for your contribution!

Were you able to successfully run the WRF with only the currently available ECMWF data, or do you need to complete the missing ECMWF data with the GFS?
 
Hi Gabriel,

To be honest, I am not using the ECMWF data to initialize WRF (I am still using GFS for that) but I use it for plotting the difference between my WRF output and ECMWF data. Since the WRF model runs at a different resolution and the WRF is a regional model, I cannot simply subtract both netcdf/grib files but I use ungrib.exe and metgrid.exe to do that for me. And for ungrib.exe and metgrid.exe I needed this Vtable. Did you try to run it with this data only?

Cheers,
Casper
 
CasperK said:
Hi Gabriel,

To be honest, I am not using the ECMWF data to initialize WRF (I am still using GFS for that) but I use it for plotting the difference between my WRF output and ECMWF data. Since the WRF model runs at a different resolution and the WRF is a regional model, I cannot simply subtract both netcdf/grib files but I use ungrib.exe and metgrid.exe to do that for me. And for ungrib.exe and metgrid.exe I needed this Vtable. Did you try to run it with this data only?

Cheers,
Casper


I did some tests in the last few days and I was able to conclude some preliminary points:

- It is not possible to initialize the WRF with only these data available from the ECMWF because some important variables are missing, such as more levels of temperature and soil moisture;

- However, it is possible to do a mixed boot, using the GFS data to fill in the missing gaps in the ECMWF data;

- In the simulations I did, I observed that if I use the 0-7cm soil temperature data available in the ECMWF gribs, the 2 meter WRF temperature field is very unrealistic with absurd temperature values, much colder in some points and much hotter in other spots (I still don't understand the real reason for this). In the simulations I made removing only the 0-7cm soil temperature data, this problem of extreme values ​​of T2M does not happen.

My next goal is to test the performance of WRF with this mixed boot (ECMWF+GFS) on days of severe weather to see the difference in storm simulation compared to WRF booted with GFS alone, I'm curious about that.
 
Hi Gabriel,

May I ask how you are running a mixed boot? What are the rough steps to make that work?

Kind regards,
Casper
 
Top