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

How to reduce the amount of wrfout data

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.

hiroto

New member
Hello. I am looking for a way to reduce the amount of wrfout data. For example, is there a way to prevent unused variables from being output?
Is there any other good way?
 
Hi,
There are 2 ways to do this:

1) Go into the WRF/Registry/Registry.EM_COMMON file, and then look for the variables you don’t want. You can remove the ‘h’ from out I/O column. After you do this for all of the variables you want removed, you will need to save the file, go back to the WRF/ directory, issue a ‘clean -a’ and then reconfigure and recompile

2) You can also use the Runtime I/O option, which will allow you to add a couple lines to your namelist, and create a text file that the model will read when it is run. With this method, you don't have to recompile the code, but you have to be careful to type everything correctly, and have no spaces in the text file. You can read about it here:
http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_v4/v4.1/users_guide_chap5.html#runtimeio

Just be careful removing variables. If you ever want to go back and plot specific things, or if you will use the files for anything else later, you may be missing something you didn't realize you would need.
 
Thank you very much.
I would like to try it with method 1.

Is the variable displayed by ncdump -h wrfout_d <domain> _ <date> currently calculated?

According to the User's Guide, if the value of the field is 0, it is not calculated,
Which is the value of the field?


In addiction,
The WRF version uses 3.3.1. I can't find Registry.EM_COMMON. Is it Registry.EM in version 3.3.1?
 
Hi,
Yes, because that version of the code is so old, things were done differently then. The file you should be looking in is Registry/Registry.EM.

Regarding your question about 'ncdump,' I'm not sure what you are asking. You will simply need to open up the Registry.EM file and make manual modifications to the file, then save it before recompiling.
 
Thank you for reply.

I don't understand how to use Resistry.EM.
I use only the following variables in the ctl file.
Is it okay to delete the 'h' for all other variables?

U 34 0 x-wind component (m s-1)
V 34 0 y-wind component (m s-1)
W 34 0 z-wind component (m s-1)
PH 34 0 perturbation geopotential (m2 s-2)
PHB 34 0 base-state geopotential (m2 s-2)
T 34 0 perturbation potential temperature (theta-t0) (K)

P 34 0 perturbation pressure (Pa)
PB 34 0 BASE STATE PRESSURE (Pa)
P_HYD 34 0 hydrostatic pressure (Pa)
Q2 1 0 QV at 2 M (kg kg-1)
T2 1 0 TEMP at 2 M (K)

PSFC 1 0 SFC PRESSURE (Pa)
U10 1 0 U at 10 M (m s-1)
V10 1 0 V at 10 M (m s-1)
QVAPOR 34 0 Water vapor mixing ratio (kg kg-1)
QCLOUD 34 0 Cloud water mixing ratio (kg kg-1)
QRAIN 34 0 Rain water mixing ratio (kg kg-1)
QICE 34 0 Ice mixing ratio (kg kg-1)
QSNOW 34 0 Snow mixing ratio (kg kg-1)
QGRAUP 34 0 Graupel mixing ratio (kg kg-1)
QNICE 34 0 Ice Number concentration ( kg-1)
QNSNOW 34 0 Snow Number concentration ( kg(-1))
QNRAIN 34 0 Rain Number concentration ( kg(-1))
QNGRAUPEL 34 0 Graupel Number concentration ( kg(-1))
LANDMASK 1 0 LAND MASK (1 FOR LAND, 0 FOR WATER) (-)

SST 1 0 SEA SURFACE TEMPERATURE (K)
SSTSK 1 0 SKIN SEA SURFACE TEMPERATURE (K)

TSK 1 0 SURFACE SKIN TEMPERATURE (K)
RAINC 1 0 ACCUMULATED TOTAL CUMULUS PRECIPITATION (mm)
RAINSH 1 0 ACCUMULATED SHALLOW CUMULUS PRECIPITATION (mm)
RAINNC 1 0 ACCUMULATED TOTAL GRID SCALE PRECIPITATION (mm)

QFX 1 0 UPWARD MOISTURE FLUX AT THE SURFACE (kg m-2 s-1)
 
I can't say with 100% certainty that it will be okay to delete all others, but you can try it and run a quick/small test case to see if the model is able to run, and whether you get the results you expect. It's probably best to save a copy of the original Registry.EM file somewhere else in case you need it again later.
 
I forgot to ask one. In Registry.EM, are the variables that can be used to delete h only variables listed in the output field in Chapter 5 of the User's Guide?

Does it make sense to delete variables that are not listed in Chapter 5 of the User's Guide?
 
I'm not certain that is an all-inclusive list. I would recommend doing a very short test, prior to deleted the 'h' in the registry file. Take a look at your output to see which variables are output. Then you can delete the 'h' in the registry for the ones that make sense.
 
Top