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

Global hourly forecast

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.

xkcd_xkcd

New member
Hi,
Can I use WRF to do hourly global forecast with one domain?
I was wondering if it is enough to run ungrib.exe only with the initial condition because of no boundary to global run. For example, I run WRF from 01.01.2020, is it enough just to download the data of 01.01.2020.

I am a new WRF runner, can anyone point out whether my idea is feasible.
Thank you.
Peng
 
Peng,
You are right. When running global WRF, you don't need any boundary forcing data. The only data you need to provide is the initial condition, and thus you only need to ungrib a single large-scale input datafile.
 
Ming,

Thanks for the reply.
I will try to do a global hourly forecast with the ERA5 dataset for a short time, then use the hourly forecast temperature to study issues on forecast predictability.

Thank you again for your kind help.
Peng
 
Ming Chen said:
Peng,
You are right. When running global WRF, you don't need any boundary forcing data. The only data you need to provide is the initial condition, and thus you only need to ungrib a single large-scale input datafile.

Ming,

I have just provided the initial condition for a global run, but the error seems to indicate that I should provide the boundary forcing data.
I have attached the namelist.wps file and the error is as follows.
Can you help me with this matter, thank you very much for your support and time.

Subroutine DATINT: Interpolating 3-d files to fill in any missing data...
Looking for data at time 2019-01-02_00
Found file: FILE:2019-01-02_00
Looking for data at time 2019-01-02_06
ERROR: Data not found: 2019-01-02_06:00:00.0000


best,
Peng
 

Attachments

  • namelist.wps
    1.1 KB · Views: 44
This is the error message from running which program? It is not from running wrf.exe, I guess. Thanks.
 
Ming, I'm sorry that my question was not clear.
The error message came from running ungrib.exe. Did this mean that I should provide the boundary forcing data? Or there was a problem with the setting of my global domain.
Thank you very much for your patient reply.
 
Please change the settings (in namelist.wps) from:

start_date = '2019-01-02_00:00:00','2006-08-16_12:00:00',
end_date = '2019-01-03_00:00:00','2006-08-16_12:00:00',

to

start_date = '2019-01-02_00:00:00','2006-08-16_12:00:00',
end_date = '2019-01-02_00:00:00','2006-08-16_12:00:00',

The rerun the case. In the previous settings, WPS will look for data from '2019-01-02_00 to '2019-01-03_00, and errors occur once it cannot find data.
 
Dear Ming,

Thank you very much for your reply.
Now I can run the ungrib.exe successfully. But when I executed the real.exe program then the error occurs again (start_date = '2019-01-02_00:00:00'), which seems that I should provide the boundary forcing data in this global run. I don't know what's wrong.

What I want is to do a forecast in which I can initial the WRF model with any initial hour. This idea may be constrained by the data if I do a regional forecast. I think I can do a global run so there is no need to provide the boundary forcing data.

The error is as follows when I executed the real.exe and I have attached my namelist.wps and namelist.input.
I am sorry to bother you with such obvious questions.

best wishes,
Peng

FATAL CALLED FROM FILE: <stdin> LINE: 406
error opening met_em.d01.2019-01-02_06:00:00.nc for input; bad date in namelist or file not in directory
 

Attachments

  • namelist.wps
    1 KB · Views: 42
  • namelist.input
    4.1 KB · Views: 37
When running REAl for global WR run, you shouldn't specify time period that covers time beyond the initial time. This is the same error as that in your namelist.wps. Please change the options below from:
run_days = 1,
run_hours = 0,
run_minutes = 0,
run_seconds = 0,
start_year = 2019,2019,
start_month = 01,01,
start_day = 02,02,
start_hour = 00,00,
start_minute = 00,00,
start_second = 00,00,
end_year = 2019,2019,
end_month = 01,01,
end_day = 03,03,
end_hour = 00,00,
end_minute = 00,00,

to:

run_days = 0,
run_hours = 0,
run_minutes = 0,
run_seconds = 0,
start_year = 2019,2019,
start_month = 01,01,
start_day = 02,02,
start_hour = 00,00,
start_minute = 00,00,
start_second = 00,00,
end_year = 2019,2019,
end_month = 01,01,
end_day = 02,02,
end_hour = 00,00,
end_minute = 00,00,
 
By the way, we provide a sample namelist.input.global saved in /WRF/test/em_real/namelist.input.global

You may copy this file, make modification based on your case.
 
If the end time is the same as the start time in the namelist.input.global, then what confuses me now is how can I specify the end time of the global forecast.

Kind regards,
Peng
 
When you run REAL, the start and end time should be the same. When running WRF, you need to specify the start nd end time to cover your integration period.
 
Top