init_atmosphere_model step - generating init.nc and sf_update.nc files

[jpen]

I am migrating from MPASv7.3 to the more recent version (v8.2.2). I've questions regarding the sequence of processes followed in the init_atmosphere_model stage given in the MPAS tutorial:
1. While generating the init.nc file, I am wondering how much is its dependence on the static fields (static.nc). For instance, for a given grid, say 60-3km variable resolution, can the static file generated by v7.3 be used for generating init.nc in v8.2.2?
2. Similarly, the generation of sfc_update file can be done independent of the creation of init.nc file? Does it require the static.nc file only?
3. I am using ERA5 fields, and I observe that SST and SEAICE variables are present in the WPS ungrib.exe processed file (FILE:YYYY-MM-DD_HH). Can I use this file for creating sfc_update file (config_sfc_prefix = 'FILE')?

 

[Ming Chen]

Please see my answers below:

I am migrating from MPASv7.3 to the more recent version (v8.2.2). I've questions regarding the sequence of processes followed in the init_atmosphere_model stage given in the MPAS tutorial:
1. While generating the init.nc file, I am wondering how much is its dependence on the static fields (static.nc). For instance, for a given grid, say 60-3km variable resolution, can the static file generated by v7.3 be used for generating init.nc in v8.2.2?

if you stay with Noah surface scheme, then I expect that the static data from v7.3 could be used in v8.2.2. However, note that NoahMP is added to V8.2.2 and if you want to use NoahMP, you will need to generate a new static datafile.

2. Similarly, the generation of sfc_update file can be done independent of the creation of init.nc file? Does it require the static.nc file only?

The generation of sfc_update is independent of generation of init.nc. sfc_update includes information from static file as well as from intermediate input files.

3. I am using ERA5 fields, and I observe that SST and SEAICE variables are present in the WPS ungrib.exe processed file (FILE:YYYY-MM-DD_HH). Can I use this file for creating sfc_update file (config_sfc_prefix = 'FILE')?

Yes you need these files as input to create sfc_update.

 

[jpen]

Thank you Ming for your response. I have a follow-up question:

1. For the case of Thompson aerosol aware scheme, what needs to be done in the pre-processing stage? I have followed WRF Model Users Site for preparing the met files but how would this be included in the pre-processing?
2. Are any tables and/ or namelist edits need to be included for the model integration step?

 

[Ming Chen]

Please see my answers below:

Thank you Ming for your response. I have a follow-up question:

1. For the case of Thompson aerosol aware scheme, what needs to be done in the pre-processing stage? I have followed WRF Model Users Site for preparing the met files but how would this be included in the pre-processing?

The aerosol climatology file (QNWFA_QNIFA_SIGMA_MONTHLY.dat) is used when running the init_atmosphere_model program to produce initial and lateral
boundary conditions for nifa and nwfa. This is totally independent of WPS/WRF met_em files.

1. Are any tables and/ or namelist edits need to be included for the model integration step?

I don't think so. You only need to set
config_microp_scheme = 'mp_thompson_aerosols' in the &physics namelist group.

 

[jpen]

Thanks again. I need to just place the climatology file in the working directory and run init_atmopshere_model for generating static and init files? No need to edit namelist.init_atmosphere?

 

[Ming Chen]

If you are talking of the climatological data "QNWFA_QNIFA_SIGMA_MONTHLY.dat", then yes you are correct. If you look at the file "mpas_init_atm_core_interface.F", you can find the following piece of code:

     if (config_init_case == 9) then
         lbcs = .true.
         mp_thompson_aers_in = .false.
         inquire(file="QNWFA_QNIFA_SIGMA_MONTHLY.dat",exist=lexist)
         if(lexist) mp_thompson_aers_in = .true.
      else
         lbcs = .false. 
         mp_thompson_aers_in = .false.
      end if

This is how the model works to read aerosol data.

 

[jpen]

Thanks again. I placed the climatological file and generated the init.nc file which showed 2 additional aerosol variables in its header file

variables:
float qv(Time, nCells, nVertLevels) ;
qv:units = "kg kg^{-1}" ;
qv:long_name = "Water vapor mixing ratio" ;
float qc(Time, nCells, nVertLevels) ;
qc:units = "kg kg^{-1}" ;
qc:long_name = "Cloud water mixing ratio" ;
float qr(Time, nCells, nVertLevels) ;
qr:units = "kg kg^{-1}" ;
qr:long_name = "Rain water mixing ratio" ;
float nifa(Time, nCells, nVertLevels) ;
nifa:units = "nb kg^{-1}" ;
nifa:long_name = "Gocart ice-friendly aerosol number concentration" ;
float nwfa(Time, nCells, nVertLevels) ;
nwfa:units = "nb kg^{-1}" ;
nwfa:long_name = "Gocart water-friendly aerosol number concentration" ;

I hope this is correct. In your response above, you have pointed to a code segment which indicates "config_init_case == 9". Was this required? I simply ran with config_init_case = 7.

 

[Ming Chen]

Your data is correct. This data is also applied when config_init_case = 7.