The model does not read the Forcing file for the Child grid

Report or discuss software problems and other woes

Moderators: arango, robertson

Post Reply
Message
Author
User avatar
Eugen
Posts: 36
Joined: Wed Apr 06, 2016 4:36 pm
Location: University of Liege

The model does not read the Forcing file for the Child grid

#1 Unread post by Eugen »

Hello, everyone!

For my nesting simulations, I made a forcing files for both, the parent and the child grids.
However, I noticed, that ROMS, instead of reading and processing the child forcing file, interpolates data from the parent forcing file into the child grid.

What could be the reason?

Here is a line from my .in file. And also I attach two headers from both netcdf forcing files (unfortunately, I haven't found an option here to post them directly on the forum in spoilers).
NFFILES == 1 1 ! number of unique forcing files

FRCNAME == /media/sf_Swap-between-windows-linux/New_Grid/Interannual_simulation/Forcing.nc \
/home/eivanov/coawst_data_prrocessing/GRID_RECTANGULAR/Meteo/Forcing_nested_2.nc
Thanks in advance!
Attachments
Forsing_nested.txt
Forcing header for the nested grid
(9.57 KiB) Downloaded 239 times
Forcing_parent.txt
Forcing header for the parent grid
(8.74 KiB) Downloaded 250 times
Evgeny Ivanov

PostDoc,
MAST, University of Liege, Belgium

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: The model does not read the Forcing file for the Child g

#2 Unread post by jcwarner »

how do you know that "[the model] interpolates data from the parent forcing file into the child grid."?
I would need to see the std out text to see what is happening.

-john

User avatar
Eugen
Posts: 36
Joined: Wed Apr 06, 2016 4:36 pm
Location: University of Liege

Re: The model does not read the Forcing file for the Child g

#3 Unread post by Eugen »

Thanks, John,

I can assume this, on the basis of netcdf pictures. For example, short-wave radiation, the variable I have to define.
On the left hand-side - how it does look in the Forcing file. On the right hand-side - how it does look at the history outputs.
Evgeny Ivanov

PostDoc,
MAST, University of Liege, Belgium

User avatar
wilkin
Posts: 875
Joined: Mon Apr 28, 2003 5:44 pm
Location: Rutgers University
Contact:

Re: The model does not read the Forcing file for the Child g

#4 Unread post by wilkin »

To the best of my knowledge we did not code any option to interpolate forcing data of the parent grid onto the child grid other than by explicitly repeating the name(s) of the parent forcing file(s) in the FRCNAME input.

Look closely at your stdout (the "log file") because it reports all the forcing file reads for both grids. There will be lines like:
GET_2DFLD - surface u-wind component, t = 2922 06:00:00
(Rec=0000003, Index=1, File: Uwind_nam_3hourly_MAB_and_GoM_2014.nc)
(Tmin= 2922.0000 Tmax= 3287.0000)
(Min = -1.63235843E+00 Max = 1.77264092E+01)
If you don't see your file Forcing_nested_2.nc reported in the File: tag then you are correct that it was not read.

This experience suggests that for added clarity we should have stdout explicitly report which Grid it is processing when making those GET_2DFLD calls, and whether or not it is invoking the regrid option.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu

User avatar
Eugen
Posts: 36
Joined: Wed Apr 06, 2016 4:36 pm
Location: University of Liege

Re: The model does not read the Forcing file for the Child g

#5 Unread post by Eugen »

Thank you, wilkin

Yes, I confirm, that the Forcing file for the parent grid is being read twice: for the parent and for the child grids.

But how can I make ROMS reading the Forcing file for the child grid?
This experience suggests that for added clarity we should have stdout explicitly report which Grid it is processing when making those GET_2DFLD calls, and whether or not it is invoking the regrid option.
So I have to make corrections into the Nesting subroutine to make it reading both of my Forcing files? Or there is another option that will allow ROMS to do it automatically?
Evgeny Ivanov

PostDoc,
MAST, University of Liege, Belgium

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: The model does not read the Forcing file for the Child g

#6 Unread post by jcwarner »

not sure where you are getting the forcing data from, but when you made the forcing files, did you interpolate the raw data to the grid resolutions? If you keep the raw data at the same resolution it came from, then the model will interpolate to the parent and child resolutions. Can you just make a forcing file with the resolution of the raw data and just use that for both parent and child?

User avatar
wilkin
Posts: 875
Joined: Mon Apr 28, 2003 5:44 pm
Location: Rutgers University
Contact:

Re: The model does not read the Forcing file for the Child g

#7 Unread post by wilkin »

The stdout ("log file") reports a list of Output/Input Files: for each grid, e.g.

Physical Parameters, Grid: 02
=============================

all the other parameters, then ...

Output/Input Files:

Output Restart File: r11/pioneer_rst.nc
Prefix for History Files: r11/pioneer_his
Prefix for Averages Files: r11/pioneer_avg
Output Stations File: r11/pioneer_sta.nc
Input Grid File: /home/om/roms/doppio/pioneer/grid_pioneer_v3.nc
Nesting grid connectivity File: /home/om/roms/doppio/pioneer/contactfile_doppio_pioneer_array_v3.nc
Input Nonlinear Initial File: /home/om/roms/doppio/pioneer/ini_pioneer_20140101_atmpress_NAVD88_v3bathy.nc
Input Sources/Sinks File: /home/om/roms/doppio/pioneer/pioneer_rivers_2007-2015.nc
Input Forcing File 01: /home/om/roms/doppio/frc/nam/lwrad_down_nam_3hourly_MAB_and_GoM_2014.nc
/home/om/roms/doppio/frc/nam/lwrad_down_nam_3hourly_MAB_and_GoM_2015.nc
Input Forcing File 02: /home/om/roms/doppio/frc/nam/Pair_nam_3hourly_MAB_and_GoM_2014.nc
/home/om/roms/doppio/frc/nam/Pair_nam_3hourly_MAB_and_GoM_2015.nc
Does the list for your Grid 02 (i.e. the child) show the parent forcing files (or any other parent files) when the ocean.in file lists the correct child grid forcing files? This inconsistency is inexplicable to me. If you have the wrong syntax in your ocean.in file for FRCNAME it's possible (but unlikely) that ROMS assumes the parent files for all child grids.

We might need to see your logfile to deduce the problem. I always say that more than 90% of the time the answer is in the logfile if you just read it carefully enough.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu

Post Reply