14 | | asterix*) |
15 | | echo "sys source asterix/obelix lib." |
16 | | . ${libIGCM}/libIGCM_sys/libIGCM_sys_obelix.ksh;; |
| 15 | cd ~/ORCHIDEE-CN-CAN/config/ORCHIDEE_OL/ENSEMBLE |
| 16 | vi config.card |
| 17 | }}} |
| 18 | Change the following lines |
| 19 | {{{ |
| 20 | JobName=FLUXNET |
| 21 | ARCHIVE=/home/scratch01/mmcgrath |
| 22 | }}} |
| 23 | then create the job scripts |
| 24 | {{{ |
| 25 | ../../../libIGCM/ins_job |
| 26 | }}} |
| 27 | this creates Job\_FLUXNET. Notice that this job will pull from the SPINUP directory as well. ins\_job used to create Job files in every directory, but that functionally changed a while ago. Therefore, the following is now necessary (OOL\_SEC\_STO because we will run a job with sechiba and stomate). |
| 28 | {{{ |
| 29 | cd ../SPINUP |
| 30 | ../../../libIGCM/ins_job |
| 31 | cd SUBJOB/OOL_SEC_STO/ |
| 32 | ../../../../../libIGCM/ins_job |
| 33 | cd ../../../ENSEMBLE |
| 34 | }}} |
| 35 | Now modify the run directory. |
| 36 | {{{ |
| 37 | vi Job_FLUXNET |
| 38 | (change RUN_DIR_PATH=/home/scratch01/mmcgrath/RUN_DIR) |
| 39 | mkdir /home/scratch01/mmcgrath/RUN_DIR |
| 40 | }}} |
| 41 | Now change the options for the sites to run against. |
| 42 | {{{ |
| 43 | vi fluxnet.card |
| 44 | }}} |
| 45 | Need to change the number of PFTs, which sites we are running against (do a small testcase with a single site first), and the PFT characteristics of that site. |
| 46 | {{{ |
| 47 | NbPFTs= 28 |
| 48 | Groups= ( TEST ) |
| 49 | TEST = ( BR-Sa3 , BR-Sa3_2000-2003.nc , 2000 , 4 , 0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 ) |
21 | | In addition, make sure you run in the config/ORCHIDEE_OL/ directory. The script looks for some files under the assumption that it is in that directory (for example, when it copies ../SPINUP). |
| 61 | Now we need to copy the CAN-CN run.def that we want to use. XXXXXX is the run.def file from whatever configuration directory you have that is currently working. |
| 62 | {{{ |
| 63 | cp XXXXXX ../SPINUP/SUBJOB/OOL_SEC_STO/PARAM/ |
| 64 | cp XXXXXX ../ENSEMBLE/PARAM/ |
| 65 | }}} |
| 66 | It seems that AUTO values are not being removed in the run.def during processing. I set XIOS\_ORCHIDEE\_OK=N, DRIVER\_reset\_time=y, taking values from a previous successful run I did (looking in the RUN\_DIR). I only replaced a value when the code crashed with a complaint. |
| 67 | |
| 68 | {{{ |
| 69 | RESTART_FILEIN= NONE |
| 70 | SECHIBA_restart_in= NONE |
| 71 | STOMATE_RESTART_FILEIN= NONE |
| 72 | }}} |
| 73 | |
| 74 | Need to check the differences between the sechiba.card in our working configuration directory and that in SPINUP/SUBJOB/OOL\_SEC\_STO/COMP. I had to change the following in SPINUP/SUBJOB/OOL\_SEC\_STO/COMP/sechiba.card |
| 75 | {{{ |
| 76 | [InitialStateFiles] |
| 77 | List= (${R_IN}/SRF/routing.nc, .), \ |
| 78 | (${R_IN}/SRF/cartepente2d_15min.nc, .), \ |
| 79 | (${R_IN}/SRF/floodplains.nc, .), \ |
| 80 | (${R_IN}/SRF/albedo/alb_bg_modisopt_2D.nc, alb_bg.nc), \ |
| 81 | (${R_IN}/SRF/PFTmap_1850to2005_AR5_LUHa.rc2/PFTmap_IPCC_2000.nc, PFTmap.nc) |
| 82 | }}} |
| 83 | |
| 84 | in SPINUP/SUBJOB/OOL\_SEC\_STO/COMP/sechiba.card, I made the following change: |
| 85 | {{{ |
| 86 | [BoundaryFiles] |
| 87 | List= () |
| 88 | ListNonDel= (${R_IN}/SRF/reftemp.nc, .), \ |
| 89 | (/home/orchidee04/aslanso/Files/soils_param.nc, soils_param.nc), \ |
| 90 | (${R_IN}/SRF/soils_param_usdatop.nc, soils_param_usda.nc) |
| 91 | }}} |
| 92 | |
| 93 | I had to modify the run.def a bit due to crashes (AUTO values were not getting replaced). |
| 94 | |
| 95 | I removed the following lines from ENSEMBLE/PARAM/run.def, since those variables didn't appear anywhere in the working config directories for the TRUNK. |
| 96 | |
| 97 | {{{ |
| 98 | TIME_LENGTH = _AUTO_ |
| 99 | TIME_SKIP = _AUTO_ |
| 100 | SECHIBA_HISTFILE2=_AUTO_ |
| 101 | WRITE_STEP=_AUTO_ |
| 102 | STOMATE_HIST_DT=_AUTO_ |
| 103 | STOMATE_IPCC_HIST_DT=_AUTO_ |
| 104 | ATM_CO2 =_AUTO_: DEFAULT = 350. |
| 105 | STOMATE_CFORCING_NAME= _AUTO_ : DEFAULT = NONE |
| 106 | }}} |
| 107 | |
| 108 | Even though the next ones weren't in the TRUNK run.def, I assigned a value since I don't like the default. |
| 109 | {{{ |
| 110 | STOMATE_OK_STOMATE=y |
| 111 | }}} |
| 112 | |
| 113 | With my debug flags on, the code threw a runtime error on the following line for an unknown reason, so I just commented it out (it's a debugging write statement, and I had l\_dbg activated...no harm in removing for the moment). |
| 114 | {{{ |
| 115 | modeles/IOIPSL/src/calendar.f90 |
| 116 | |
| 117 | IF (l_dbg) THEN |
| 118 | ! WRITE(ipslout,*) "isittime 2.0 ", & |
| 119 | ! & date_next_check,date_next_act,ABS(dt_action-freq), & |
| 120 | ! & ABS(dt_action+dt_check-freq),dt_action,dt_check, & |
| 121 | ! & next_check_itau,do_action |
| 122 | ENDIF |
| 123 | }}} |
| 124 | |
| 125 | out\_orchidee has become huge, seemingly because of histfile debugging output. So I am turning off ioipsl\_debug in modeles/IOIPSL/src/errioipsl.f90. |
| 126 | |
| 127 | We need to make sure all of the following lines are commented out (or do not exist in the run.def, since the script will change the vegetation for each site by adding lines at the end of the run.def and if these lines are present they will override them: |
| 128 | |
| 129 | {{{ |
| 130 | IMPOSE_VEG=y |
| 131 | SECHIBA_VEG__01=0.0769230769231 |
| 132 | ... |
| 133 | SECHIBA_VEGMAX__01=0.0769230769231 |
| 134 | ... |
| 135 | }}} |
| 136 | |
| 137 | I often set the JobType=DEV in my Job file so that it will overwrite old output files in the archive. Else, you have to delete the IGCM\_OUT/OL2/PROD/ensemble/ files for every site before every run. Which is no problem if everything works, but it is annoying if you are not yet sure they will work. |
| 138 | |
| 139 | Launch the job (from the README file). |
| 140 | |
| 141 | {{{ |
| 142 | ./Job_ENSEMBLE fluxnet > out.Job_ENSEMBLE |
| 143 | }}} |
| 144 | |
| 145 | This submits jobs to the queue which create new directories in the ENSEMBLE directory, each one having the name of the test site. |
| 146 | |
| 147 | First simulation finishes with no problems! However, it complains of an error in copying the restart file. I tried deleting the whole archive and rerunning. |
| 148 | |
| 149 | It takes 20 minutes or so to run the STOI (it runs over four years or so, the length of the forcing file for that site). |
| 150 | |
| 151 | |
| 152 | |