Anta-Clarisse Sarr

Anta-Clarisse Sarr

PostDoc interested in geodynamics & paleoclimate interactions. Climate Modeling. Cenozoic.

PALEO-PISCES offline Model (old)

4 minute read

Reference paper for the PISCES model is Aumont et al. 2015. You will find additional information about parameterization and appropiate references within this paper as well.

Download & Compile the code

  • In the $CCCWORKDIR directory, create a new PALEO-PISCES directory
  • In PALEO-PISCES directory, do a svn to load the required files from the server
#Download Modipsl
cd $CCCWORKDIR
mkdir PALEOPISCES
cd PALEOPISCES

svn co http://forge.ipsl.jussieu.fr/igcmg/svn/modipsl/trunk modipsl

# Download PISCES
cd modipsl/util
./model NEMO_v6.1

# if NEMO version do not exists use ./model -h to check the existing version

# Compile XIOS (on Irene)
cd ../modeles/XIOS
./make_xios --arch X64_IRENE --full --prod --job 8

# ARCH file / key_xios
cp YOUR_WORKDIR/IPSLCM5A2/modipsl/modeles/NEMOGCM/ARCH/arch-X64_IRENE.fcm YOUR_WORKDIR/PALEOPISCES/modipsl/modeles/NEMOGCM/ARCH/.

vi PALEOPISCES/modipsl/modeles/NEMOGCM/CONFIG/ORCA2_OFF_PISCES/cpp_ORCA2_OFF_PISCES.fcm
- add key_xios2 to the list
  • Modify PISCES source code to have the updates about nutrient input
  • Compile Nemo
# Add modified source_file for PISCES 
mkdir YOUR_WORKDIR/PALEOPISCES/modipsl/modeles/NEMOGCM/CONFIG/ORCA2_OFF_PISCES/MY_SRC
cp MODIFIED_FILE_PATH YOUR_WORKDIR/PALEOPISCES/modipsl/modeles/NEMOGCM/CONFIG/ORCA2_OFF_PISCES/MY_SRC/.

# Compile  NEMO
cd PALEOPISCES/modipsl/modeles/NEMOGCM/CONFIG/
./makenemo -n ORCA2_OFF_PISCES -m X64_IRENE -j 8

#Executable
cp PALEOPISCES/modipsl/modeles/NEMOGCM/CONFIG/ORCA2_OFF_PISCES/BLD/bin/nemo.exe  PALEOPISCES/modipsl/bin/orca2offpisces.exe

Launch a new simulation

Get the boundary conditions files from the coupled simulation

  • In case it is the firs time you run an offline PISCES simulation, create a directory (ex: BC_PISCES_OFFLINE) where you will store the boundary conditions for the simulations.

  • Create a directory for the simulation and transfer grid_T.nc, grid_V.nc, grid_W.nc and grid_U.nc from OCE/ , ptrc.nc from MBG/ and icemod.nc from ICE/

In case siconc and wocetr_eff variables do not exist in you grid_T.nc anc grid_W.nc model output :

You will probably need to change the writing rights on the files by doing chmod +w file.nc

  • (“clim” simulation) Use the script UpdateVariable.ksh to add them (you need to run the script twice : for woce and sic)
  • (“IA” simulation) Use the script CopyVariable.ksh to add them

Best is to directly add those variables in the simulation setup : In file PARAM/file_def_nemo-opa.xml :

- Replace 

 <field field_ref="iceconc"      name="siconc"  />  (l.205)
 
  by 
  
 <field field_ref="iiceconc"      name="siconc"  />
 
- And add 

 <field field_ref="wocetr_eff"   name="wocetr_eff"  /> (l.265)

Generate initial state for MBG/ component

  • You need to have the ptrc.nc file of the coupled simulation
  • To generate initial state for the ocean-biogeochmistry use the script ScaleNutrient.ksh
cp /ccc/work/cont003/gen2212/MARIE_LOGIN/BC_PISCES_OFFLINE/ScaleNutrients.ksh  .

vi ScaleNutrients.ksh 
# you need to change file paths at least

Total nutrient input similar to present-day

In this case you just need to extract value from the ptrc.nc file and to scale the O2 variable to have the right unit.

Nutrient input adapted to climate

In this case you have, in addition, to be sure that each file has been scaled by the factor you calculated from the runoff of the coupled simulation.

Create a coastal mask

  • Copy bathymetry file from coupled simulation (PALEORCA file)
  • in BC_PISCES-OFFLINE copy the create_coastline.f90 script
  • Into the script you need to modify input and output file
  • Compile and execute
cd BC_PISCES_OFFLINE

cp bathy_paleorca_from_my_coupled_simulation.nc .

cp /ccc/work/cont003/gen2212/MARIE_LOGIN/BC_PISCES_OFFLINE/COASTLINE/create_coastline.f90

vi create_coastline.f90
# Need to modify lines 30,34,67

ifort -o "createcoast" $NETCDFFORTRAN_LDFLAGS $NETCDFC_LDFLAGS $NETCDFFORTRAN_FFLAGS $NETCDFC_CFLAGS create_coastline.f90

./createcoast

Compute scaling factor from runoff

To do so you need the grid_T.nc file from the coupled simulation and the script runoffs.jnl to obtain annually cumulated runoff. To get the scaling factor, you have to divide this value by value from a modern simulation (2352948)

Total nutrient input similar to present-day

This scaling factor in use to calculate initial concentration of dissolved elements in the river. To do so you have to modify file PARAM/NAMELIST/namelist_pisces_cfg (l.70-76)


<    dicconc = xxxe-6    ! dissolved inorganic carbon concentration in rivers (in Mmol/m3)
<    docconc = xxxe-6   ! dissolved organic carbon concentration in rivers (in Mmol/m3)
<    dinconc = xxxe-8   ! dissolved inorganic nitrogen concentration in rivers (in Mmol/m3)
<    donconc = xxxe-8    ! dissolved organic nitrogen concentration in rivers (in Mmol/m3)
<    dipconc = xxxe-9    ! dissolved inorganic phosphorus concentration in rivers (in Mmol/m3)
<    dopconc = xxxe-9   ! dissolved organic phosphorus concentration in rivers (in Mmol/m3)
<    dsiconc = xxxe-6  ! dissolved silicate concentration in rivers (in Mmol/m3)

Total nutrient input similar to present-day

This scaling factor is used to scale damping value (l.81-84)

  alkmean = 2426.    ! Mean alkalinity concentration
  po4mean = 2.165    ! Mean phosphate concentration
  no3mean = 30.90    ! Mean nitrate concentration
  silmean = 91.51    ! Mean silicate concentration

as well as to modifiy initial state for the MBG/ component as explained below

Define value of nutrient concentration

Total nutrient input similar to present-day

At the end, the total supply should roughly correspond to the one from the control simulation (ORCA2_OFF_PISCES/PROD/ORCA2clim/PISC-Control-RunOff/DEBUG/MBG_PISC-Control-RunOff_05000101_05001231_ocean.output)

N Supply : 36.096027 TgN/yr
Si Supply : 154.645665 TgSi/yr
P Supply : 2.572411 TgP/yr
Alk Supply : 35.141530 Teq/yr
DIC Supply : 586.800500 TgC/yr

So it is better to run the simulation for one year only and to adjust initial concentration if it does not fit. In this case, after modifying concentrations in PARAM/NAMELIST/namelist_pisces_cfg, change the DateEnd in the config.card and the JobStatus to ‘OnQueue’ in the run.card and re-submit the Job.

Nutrient input adapted to climate

Updated: