Version 32 (modified by nillod, 8 years ago) (diff) |
---|
LMDZOR6 RRTM with CMIP6 beta release volcanic forcing defined with 3 params (AOD, cg, piz on both SW and LW)
Model installation
- Start to import modIPSL: svn co http://forge.ipsl.jussieu.fr/igcmg/svn/modipsl/trunk modipsl
- Change LMDZ revision to trunk 2541 in mod.def
- Install the model: cd modipsl/util ; ./model LMDZOR_v6
3b. To enable debug version: edit modipsl/config/LMDZOR/Makefile and add debug option:
lmdz: ../../modeles/LMDZ (cd ../../modeles/LMDZ ; ./makelmdz_fcm -debug -d $(RESOL_LMDZ) [...] (cd ../../modeles/LMDZ ; ./makelmdz_fcm -debug -d $(RESOL_LMDZ) [...]
- Activate RRTM mode in modipsl/modeles/LMDZ/makelmdz_fcm: rrtm=true
- Compile with L=79: cd modipsl/config/LMDZOR ; gmake LMD144142-L79
Activate and configure (RRTM, read aerosol with CMIP beta 0 release forcing containing 3 parameters)
- RRTM: see process below (don't forget to compile with rrtm flag). Active flag_rrtm=1 in corresponding PARAM/physiq.def file.
- Read aerosol: Edit corresponding PARAM/physiq.def file and add flag_aerosol_strat=2. Then add aerosols files to read in COMP/lmdz.card as below:
[BoundaryFiles] List= (/path/to/input/dir/tauswstrat.2D.${year}.nc, tauswstrat.2D.nc), \ (/path/to/input/dir/taulwstrat.2D.${year}.nc, taulwstrat.2D.nc), \ (/path/to/input/dir/${CREATE}_${year}_limit.nc, limit.nc),\ ...
TESTS
Nomenclature:
2D: corresponds to CMIP6 beta0 release forcing dataset zonally averaged for the 3 parameters.
AOD: corresponds to SATO zonal AOD. The other 2 parameters are fixed.
Name of simulations:
- LMDZOR6.VOLC.2D.2 and LMDZOR6.VOLC.2D.3: same boundary conditions (ERSST) and other forcings (ie CMIP6 beta release). Only difference is on period length (1M or 1Y). See table and figures below. To note unexplained differences in radiation balance in temperature response. We note about 0.5W/m2 between test 2 and 3 that go back to 0 at the end of the year (see figure) => problem with temporal interpolation ??? To be checked in coupled model framework.
- LMDZOR6.VOLC.AOD and LMDZOR6.VOLC.AOD.2: same boundary conditions (ERSST) and other forcings (ie SATO AOD). Only difference is on period length (1M or 1Y). See table and figures below. Under analysis.
Name | #MPI | OMP | flag_aer | PerLen? | PackFq? | Physic | RRTM | Out | Forcings | Results | Comment | |
LMDZOR6.VOLC.2D.2 | 47 | 8CPU | 2 | 1Y | 3Y | NPv5.4 | Yes | DEVT | CMIP6 3par | ok 3Y | phylmd rev 2550 | |
LMDZOR6.VOLC.2D.3 | 47 | 8CPU | 2 | 1M | 3Y | NPv5.4 | Yes | DEVT | CMIP6 3par | ok 3Y | phylmd rev 2550/queue test | |
LMDZOR6.VOLC.AOD | 47 | 8CPU | 1 | 1Y | 3Y | NPv5.4 | Yes | DEVT | SATO | ok 3Y | phylmd rev 2550 | |
LMDZOR6.VOLC.AOD.2 | 47 | 8CPU | 1 | 1M | 3Y | NPv5.4 | Yes | DEVT | SATO | run... | phylmd rev 2550 | |
LMDZOR6.NOVOLC | 47 | 8CPU | 0 | 1Y | 3Y | NPv5.4 | Yes | DEVT | TSI only | ok 3Y | phylmd rev 2550/ctrl sim |
Outputs
All outputs were made with LMDZ6 rev 2550, physic=NPv5.4. Anomalies are performed by removing control run values (ie LMDZOR6.NOVOLC) in 1991-1993 period.
Forcings:
- 2D (or 3 params): read g, omega and piz in SW + absorption in LW.
- AOD: only the zonal AOD is reading
LMDZOR6.VOLC.2D.2 (CMIP6 beta release0, 3 param)
TOPSWAD (LMDZOR6.VOLC.2D.2-LMDZOR6.VOLC.2D.3)