Obs Nudging with DFI

[wallis]

Hi,

I'm trying to setup obs nudging for my WRF runs, with a nest, but I'm hitting this error.
Does anyone know how to proceed?


OBSERVATION NUDGING IS ACTIVATED FOR MESH 2
WRF scheme selected for horizontal spreading of surface obs
Regime-based VIF scheme selected for vertical spreading of surface obs

*** TIME WINDOW SETTINGS FOR NEST 2
TWINDO (hrs) = 0.500 SFCFACT = 1.000 SFCFACR = 1.000

*** MAX PRESSURE GAP (cb) for interpolation between soundings = 20.0

File open on obs nudging unit ( 61 ) with wrong name
File open on obs nudging unit is named /home/wrf/OBS_DOMAIN101 but it should be named OBS_DOMAIN101
This likely means this unit number was opened elsewhere in WRF
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: wrf_fddaobs_in.b LINE: 461
wrf_fddaobs_in: in4dob STOP Obs nudging file name mismatch
-------------------------------------------

 

[kwerner]

Hi,
I've moved this post to the "Special Running Options" section since obs-nudging is not a standard running option. Can you let me know the version of the model you are running, and please attach the following:
1) namelist.input
2) the full error log (e.g., rsl.error.0000)
3) Issue the following commands (from the directory where you are running wrf):

 ls -ls >& ls.log

 ncdump -v Times wrffdda_d01 >& d01.log

 ncdump -v Times wrffdda_d02 >& d01.log

and please also attach those 3 *.log files. If you need instructions on attaching log files, please see the home page for this forum. Thanks!

 

[wallis]

Hi,
So I've debugged this further and determined it's related to dfi.

With dfi_opt = 3
It crashes with
File open on obs nudging unit ( 61 ) with wrong name
File open on obs nudging unit is named /home/wrf/OBS_DOMAIN101 but it should be named OBS_DOMAIN101
This likely means this unit number was opened elsewhere in WRF

If I change to dfi_opt = 0, everything works.
So I think there is a bug with how DFI handles obs nudging. Let me know if I need to provide any more information for this.

 

[kwerner]

Hi,
I'm glad you were able to pinpoint the problem! There's not a bug with DFI and nudging. It's just that the 2 options should not be used together. When you run DFI, you change the initial condition file (with noise removed), but when you nudge the model, you are forcing the model with un-filtered analysis. Additionally, DFI is really only useful for short-range runs, which is not typically what nudging is used for.

 

[wallis]

Hi,
Is this bad interaction documented anywhere?
I'm doing short range (0-24 hour) forecasts from GFS output +0, but by the time that is available there is already 4 hours of observations available, so I nudge the model with those observations.

 

[kwerner]

To my knowledge, it's not documented anywhere. I have it in my own notes, based on a previous support conversation I had, and based on conversations with other colleagues. It probably should be documented, though. I'll make a note to add it!