operadar

Computes dual-pol variables (Z_H, Z_DR, K_DP, Rho_HV) in the 3D model grid for Arome or MesoNH model using existing T-matrix tables

• INPUT : T-matrix tables and model file (Arome .fa or MesoNH netcdf)

Tmatrix tables directory at CNRM on belenos: /home/augros/TmatCoefInt_SCXW/

• OUTPUT : netcdf file with $lat/lon$ (or $X/Y$) + $Z_{H}$ , $Z_{DR}$ , $K_{DP}$ , $\rho_{HV}$ , $T$, and altitude for each model level

How to get the code

For belenos: create a .gitconfig file (on your home directory) with:

[http]
       sslVerify = false

• If you only need to run the code:

git clone https://github.com/UMR-CNRM/operadar.git

• If you want to suggest slight modifications
  1. fork the code (select: Fork, create a new fork)
  2. get the name of your new fork repository by clicking on the Code button (copy the https url) e. g. :

     git clone https://github.com/augros/operadar.git 
     

• If you want to be part of the main developpers (ask [email protected])

How to run operadar

1. Create a study case file in ./study_cases/ (eg: CORSE_Arome.csv or CORSE_MesoNH.csv) ==> you copy one of the examples available that best fits to your simulation

start_time;end_time;radar_id_list;radar_band;run_model;latmin;latmax;lonmin;lonmax;

• start_time;end_time => period of your study (each time corresponds to 1 Arome or MesoNH file, the time step is defined in the configuration file below)
• radar_id_list => not used in this version
• radar_band => to specify which tables to read (S, C or X band)
• run_model => to specify at what time the simulation starts (for MesoNH) or the Arome run (00, 03 ...)
• latmin;latmax;lonmin;lonmax => to restrict a specific region where to compute the radar variables (only for Arome)

2. Create a configuration file in ./configFiles/ (eg: conf_AROME_ICE3_CORSEbe.py or conf_MesoNH_ICE3_CORSEbe.py)

You need to change:

• htypes_model and list_types_tot

  • ICE3 or LIMA (without hail) :
    • htypes_model = ['vv','cc','rr','ii','ss','gg'] # model's hydrometeors related variables
    • list_types_tot = ['rr','ii','ss','gg','wg'] # idem + wet hydrometeors (computed)
  • ICE4 or LIMA with hail :
    • htypes_model = ['vv','cc','rr','ii','ss','gg','hh']
    • list_types_tot = ['rr','ii','ss','gg','wg','hh','wh']

• directories and file name options (T-matrix table and model file directories)

• forward operator options: save_netcdf = True (keep netcdf format) step = dt.timedelta(hours=1) or dt.timedelta(minutes=15) or dt.timedelta(minutes=5) ... depending of the time step between 2 consecutive model files in your simulation

• radar options:

  • distmax_rad (m) => maximum range (m) until which variables are computed (with the default option corresponding to a radar located in the center of the domain)
  • radarloc="center" (center of the domain by default, no other option yet)

3. Run with exec_operad.sh

takes 4 arguments in this order :

• 1 Arome or MesoNH

• 2 date into the yyyymmdd format or "all"

• 3 microphysics scheme name in capital letter (ICE3, LIMA, ICE4, LIMAAG)

• 4 Config file specifying directories and forward operator options

  Before running, select the option with or without nohup in exec_operad.sh (directly with python3 -i if you need to debub, or with nohup if the script is running well)

Examples :

>>> ./exec_operad.sh MesoNH 20220818 ICE3 conf_MesoNH_ICE3_CORSEbe.py

>>> ./exec_operad.sh MesoNH 20220818 LIMA conf_MesoNH_LIMA_CORSEbe.py

>>> ./exec_operad.sh Arome 20220818 ICE3 conf_Arome_ICE3_CORSEbe.py

>>> ./exec_operad.sh MesoNH 20220818 LIMAAG conf_MesoNH_LIMH_CORSEbe.py

To Do

• find a way to read the ice concentration in AROME file ?
• add the radar geometry option (with elevations and beam filtering with gaussian)
• link this project with tmatrix DPOLSIMUL git (which produces the required tables !)
• put Tmatrix tables online so that every one can get them
• put one MesoNh and one AROME example test files