How can I extend FHIST time period?

[zhang@nuist_edu_cn]

Dear team,I noticed that FHIST was validated for 1979-2015. I'm wondering what should I do if I want make a recent-year run(like 2017-2018) or real-time run?I'm aware that I should provide OBS SSTs and Sea Ice for DOCN but what else? I set up a start-up run setting RUN_STARTDATE at 2017-12-01 and the atm.log says (GETFIL): attempting to find local file  LBC_17500116-20150116_CMIP6_0p5degLat_c180227.nc (GETFIL): using  /home/export/****/****/SBCs/CESM_inputdata/inputdata//atm/waccm/lb/LBC_ 17500116-20150116_CMIP6_0p5degLat_c180227.nc flbc_inti: time out of bounds for dataset =  /home/export/****/****/SBCs/CESM_inputdata/inputdata//atm/waccm/lb/LBC_ 17500116-20150116_CMIP6_0p5degLat_c180227.nc
ERROR: Unknown error submitted to shr_abort_abort.I entered that directory and it seems LBC file contains Aerosol forcing? How should I prepare that for years that after 2015-12-01?

Best regards,Oscar

 

[kristineom@met_no]

I am currently facing the same issue, did you find a solution for this? Cheers,Kine

 

[zhang@nuist_edu_cn]

yep. Sort of.i manually extend those aerosol related files using most recent 3-year-mean annual cycle. And write them back to original files. It works. This method is kind of bold because it assumes no interannual changes in aerosol for years you extend. best,Oscar

 

[Yuan Sun]

Dear Zhang, I made the same error as you had mentioned. Could you share more instructions on how to extend FHIST time period or modify input data? Thanks

 

[islas]

You can extend FHIST past 2014 by manually altering the namelist files to give it the forcings associated with a projection scenario e.g., SSP2-4.5 or SSP3-7.0. The best way to do this is to set up a coupled case with the SSP scenario e.g., using the BSSP370 compset. Preview the namelist by running ./preview_namelists from the case directory. Then go into ./CaseDocs from your case directory and look at atm_in and lnd_in and look at the forcing files that are needed. There will be many of them, especially from the atmosphere.

Then you can set up your FHIST case and paste these namelist settings into user_nl_cam and user_nl_clm in your case directory for the FHIST run. You may also need to give it an observation-based SST file that extends up to when your run ends.

 

[Yuan Sun]

Hi islas, Thank you for your detailed reply. I would have a try to edit the namelist!!

 

[simon_tett@ed_ac_uk]

I've just come across this as I am trying to run FHIST closer to date. My plan is:
Step 1: extend the driving datasets using the BSSP585cmip6 compset.
Step 2: Run 2016 as a startup simulation. That should use the existing SST's and Sea Ice values.
Step 3: Extend the SST & Sea-ice 's to "today" using HadISST/Reynolds or OSTIA.

For Step 1 I've followed Islas suggestion making, "obvious" guesses (does it look like a file) for what are drivers (as opposed to configurations) to set up user_cam_nl. However, user_clm_nl looks more tricky. I think that FHIST is using a different version of CLM from BSSP585cmip6. In particular, the full model is, I think, using 78 PFTs (crops and other things I imagine) while FHIST is using 16 PFTs. For CLM I think that the only different driving dataset is flanduse_timeseries.

So is there a SSP585 equivalent to lnd/clm2/surfdata_map/landuse.timeseries_0.9x1.25_hist_16pfts_Irrig_CMIP6_simyr1850-2015_c170824.nc?

Should it make a difference, I am trying to run on the UK's national super-computer: ARCHER2.

Simon

 

[simon_tett@ed_ac_uk]

And googling a bit. I think my best shot is:
https://svn-ccsm-inputdata.cgd.ucar...x1.25_rcp6.0_16pfts_simyr1850-2100_c160127.nc (I can live with poor land use changes...)
Simon

 

[Yuan Sun]

And googling a bit. I think my best shot is:
https://svn-ccsm-inputdata.cgd.ucar...x1.25_rcp6.0_16pfts_simyr1850-2100_c160127.nc (I can live with poor land use changes...)
Simon

Using this land file might meet an error: 'mismatch of input dimension 17 with expected value 15'. I used 'cesm/inputdata/lnd/clm2/surfdata_map/release-clm5.0.18/landuse.timeseries_0.9x1.25_SSP3-7.0_16pfts_Irrig_CMIP6_simyr1850-2100_c190214.nc'.

Thanks again Islas and Simon,
Yuan