Opened 8 months ago

Last modified 7 months ago

#3489 new help

ARCHER2, easyaerosol data, spectral radiation data, wmcginty

Reported by: pmcguire Owned by: um_support
Component: UM Model Keywords: ARCHER2, easyaerosol data, spectral radiation data, wmcginty
Cc: mtodt Platform: ARCHER2
UM Version: 11.5

Description

Hello CMS support:
I am working on a UM11.5 suite for ARCHER2. These paths for easyaerosol data and for spectral radiation data from ARCHER do not exist anymore on ARCHER2 (from wmcginty's account). Is there another copy somewhere? If not, do you know what I should do?

easyaerosol_dir=/work/n02/n02/wmcginty/hadom/easyaerosol
SPECTRAL_FILE_DIR=/work/n02/n02/wmcginty/hadmn/cmip6spectralmonthly

Thanks,
Patrick McGuire

Change History (12)

comment:1 Changed 8 months ago by ros

Hi Patrick,

These either came from inside the Met Office and/or from Monsoon2.

What sub-directories are you expecting to see under these directories?

I can see a directory /projects/hirescm/hadom/cmip6spectralmonthly in Malcolm's area rather than Martin's.

rhatcher@xcs-c$ ls
mcica_data  sp_lw_ga7  sp_lw_ga7_k  sp_sw_ga7  sp_sw_ga7_k  sp_sw_ga7_var

Does this look right? If so do you have a Monsoon account to copy them over?

I don't know where else to look for the easyaerosol directory. Being at home I don't have access to the Met Office network, so you might need to ask Malcolm.

Cheers,
Ros.

comment:2 Changed 8 months ago by ros

Hi Patrick,

You could also talk to Markus Todt as he was using the files under these directories on ARCHER and I had told him if he needed them on ARCHER2 he would need to copy them over.

Cheers,
Ros.

comment:3 Changed 8 months ago by pmcguire

Hi Ros:
Thanks.
Those files look right, in Malcolm's directory.
I will ask Markus first.
Patrick

comment:4 Changed 8 months ago by annette

Patrick,

Just to add there is also data in $UMDIR on Archer2 which might be what you are looking for:

$UMDIR/highresmip/cmip6spectralmonthly
$UMDIR/highresmip/ancil/easyaerosol

The README.txt says it is from /data/d05/hadom which is Malcolm's directory, but I don't know aything more about the provenance of the data.

Annette

comment:5 Changed 8 months ago by ros

Well spied Annette. :-) The data that Markus was using was definitely from that directory of Malcolm's at Met Office so looking good!

comment:6 Changed 8 months ago by mtodt

Hi Patrick, Ros

I think Ros already found the solution, but just in case I looked up the content of the corresponding directories on MONSOON:

For easyaerosol_dir

volc_aer_absorption_lw.nc  volc_aer_asymmetry_lw.nc  volc_aer_extinction_lw.nc
volc_aer_absorption_sw.nc  volc_aer_asymmetry_sw.nc  volc_aer_extinction_sw.nc

For SPECTRAL_FILE_DIR

mcica_data  sp_lw_ga7  sp_lw_ga7_k  sp_sw_ga7  sp_sw_ga7_k  sp_sw_ga7_var

Hope that helps!
Markus

comment:7 Changed 7 months ago by pmcguire

  • Cc mtodt added

comment:8 Changed 7 months ago by pmcguire

Hi Ros & Markus & Annette:
I tried running 1month on ARCHER2 with $UMDIR/highresmip/ancil/easyaerosol as the path to the easyaerosol's. But the run didn't succeed. I hadn't checked beforehand to see if there are n216e ancillaries at that path like there were in the path from wmcginty on ARCHER. There are only n512e versions there. Do any of you know if the n216e easyaerosol ancillaries exist on ARCHER2 somewhere? Or should I try to get them from Markus's cache of them on Monsoon?
Patrick

comment:9 Changed 7 months ago by mtodt

Hi Patrick

Here's the full link on MONSOON:

/projects/ancils/cmip6/ancils/n216e/timeseries_1850-2014/VolcanicAod/v3

There are lots more subdirectories within /projects/ancils/cmip6/ancils/n216e/ but I suppose they aren't necessary for us at the moment.

Cheers
Markus

comment:10 Changed 7 months ago by pmcguire

Thanks, Markus!

Is the timeseries_1850-2014/VolcanicAod/v3 directory the same thing as hadom/easyaerosol/1949-2015?

I will start looking at those files.

Patrick

comment:11 Changed 7 months ago by mtodt

Oh, I see. Right. There are 2 namelists, one called easy_aerosol in app/um/opt/rose-app-cmip6_aerosols.conf (which we use) and one called easyaerosol in app/um/rose-app.conf. They look for different files:

CMIP6_ANCILS=/projects/ancils/cmip6/ancils
...
[namelist:easy_aerosol]
easyaerosol_dir='$CMIP6_ANCILS/n216e/timeseries_1850-2014/VolcanicAod/v3'
easyaerosol_files(1)='volc_aer_extinction_sw.nc'
easyaerosol_files(2)='volc_aer_absorption_sw.nc'
easyaerosol_files(3)='volc_aer_asymmetry_sw.nc'
easyaerosol_files(4)='volc_aer_extinction_lw.nc'
easyaerosol_files(5)='volc_aer_absorption_lw.nc'
easyaerosol_files(6)='volc_aer_asymmetry_lw.nc'

and

[namelist:easyaerosol]
easyaerosol_dir='/projects/hirescm/hadom/ancil/easyaerosol/1949-2015/n216e'
easyaerosol_files(1)='easy_extinction_sw_PI-MACv2-SP-CMIP6-Volc_1949-2015_v2_N216e_AW_1949-2015.nc'
easyaerosol_files(2)='easy_absorption_sw_PI-MACv2-SP-CMIP6-Volc_1949-2015_v2_N216e_AW_1949-2015.nc'
easyaerosol_files(3)='easy_asymmetry_sw_PI-MACv2-SP-CMIP6-Volc_1949-2015_v2_N216e_AW_1949-2015.nc'
easyaerosol_files(4)='easy_extinction_lw_PI-MACv2-SP-CMIP6-Volc_1949-2015_N216e_AW_1949-2015.nc'
easyaerosol_files(5)='easy_absorption_lw_PI-MACv2-SP-CMIP6-Volc_1949-2015_N216e_AW_1949-2015.nc'
easyaerosol_files(6)='easy_asymmetry_lw_PI-MACv2-SP-CMIP6-Volc_1949-2015_N216e_AW_1949-2015.nc'
easyaerosol_files(7)='easy_cdnc_PI-MACv2-SP_1949-2015_v2_N216e_AW_1949-2015.nc'

To be honest, I don't know whether only one is used or both are. I just changed the paths when setting up the suite on MONSOON.

Cheers
Markus

Last edited 7 months ago by mtodt (previous) (diff)

comment:12 Changed 7 months ago by pmcguire

Hi Markus:
Since we use the cmip6_aerosol opt flag, the files in app/um/opt/rose-app-cmip6_aerosols.conf are the ones that are used, not the other ones. So I guess we don't need (right now at least) the ones I was looking for that used to be in wmcginty's account on ARCHER, and which I guess are now in /projects/hirescm/hadom/ancil/easyaerosol/1949-2015/n216e on MONSOON.

And since we're using the cmip6_aerosol opt flag, those easyaerosol files already exist on ARCHER2, and I think my path for them was already right. So the error message I am getting with the easyaerosol's is due to some other problem, which I will investigate now.
Patrick

Note: See TracTickets for help on using tickets.