Not all data available – in #10: INT2LM

in #10: INT2LM

<p> Dear all, <br/> I have a problem with simulation from <span class="caps"> MPI </span> - <span class="caps"> ESM </span> -LR historical experiment data, 2005 year, region of Chuckchi Peninsula. I have become the following errors (in attached log-file slurm-923246.out), including the final ‘Not all data available’. Unfortunately, I didn’t understand, what kind of data aren’t available in my caf-file… I’m attaching the <span class="caps"> INPUT </span> file, maybe some specifications for these datasets (i.e., model experiments) are needed. <br/> Thank you very much for any suggestions! </p>

  @vladimirplatonov in #870da4f

<p> Dear all, <br/> I have a problem with simulation from <span class="caps"> MPI </span> - <span class="caps"> ESM </span> -LR historical experiment data, 2005 year, region of Chuckchi Peninsula. I have become the following errors (in attached log-file slurm-923246.out), including the final ‘Not all data available’. Unfortunately, I didn’t understand, what kind of data aren’t available in my caf-file… I’m attaching the <span class="caps"> INPUT </span> file, maybe some specifications for these datasets (i.e., model experiments) are needed. <br/> Thank you very much for any suggestions! </p>

Not all data available

VP wrote

Dear all,
I have a problem with simulation from MPI - ESM -LR historical experiment data, 2005 year, region of Chuckchi Peninsula. I have become the following errors (in attached log-file slurm-923246.out), including the final ‘Not all data available’. Unfortunately, I didn’t understand, what kind of data aren’t available in my caf-file… I’m attaching the INPUT file, maybe some specifications for these datasets (i.e., model experiments) are needed.
Thank you very much for any suggestions!

View in channel
<p> Can you provide the output of <br/> ncdump -h /mnt/msu/users/vplatonov/COSMO_CLM/CLM/IN_DIR/historical/caf2005010100.nc <br/> please? </p>

  @burkhardtrockel in #da9098c

<p> Can you provide the output of <br/> ncdump -h /mnt/msu/users/vplatonov/COSMO_CLM/CLM/IN_DIR/historical/caf2005010100.nc <br/> please? </p>
BR replied

Can you provide the output of
ncdump -h /mnt/msu/users/vplatonov/COSMO_CLM/CLM/IN_DIR/historical/caf2005010100.nc
please?

View in channel
<p> Yes, I have attached output of: <br/> ncks -m caf…nc <br/> It is more simple for some reasons… Does it matter? <br/> Thank you. </p>

  @vladimirplatonov in #4e7d84f

<p> Yes, I have attached output of: <br/> ncks -m caf…nc <br/> It is more simple for some reasons… Does it matter? <br/> Thank you. </p>
VP replied

Yes, I have attached output of:
ncks -m caf…nc
It is more simple for some reasons… Does it matter?
Thank you.

View in channel
<p> There is no T_SKIN in your boundary caf data file. Therefore please set luse_t_skin = .FALSE., <br/> in the int2lm namelist. </p>

  @burkhardtrockel in #435a131

<p> There is no T_SKIN in your boundary caf data file. Therefore please set luse_t_skin = .FALSE., <br/> in the int2lm namelist. </p>
BR replied

There is no T_SKIN in your boundary caf data file. Therefore please set luse_t_skin = .FALSE.,
in the int2lm namelist.

View in channel
<p> OK, thank you! Could you suggest, which namelist parameters in <span class="caps"> COSMO </span> - <span class="caps"> CLM </span> <span class="caps"> INPUT </span> have to be switched on/off (or another values) due to this ‘luse_t_skin = .FALSE.’ setting? <br/> Thanks. </p>

  @vladimirplatonov in #9f6bf35

<p> OK, thank you! Could you suggest, which namelist parameters in <span class="caps"> COSMO </span> - <span class="caps"> CLM </span> <span class="caps"> INPUT </span> have to be switched on/off (or another values) due to this ‘luse_t_skin = .FALSE.’ setting? <br/> Thanks. </p>
VP replied

OK, thank you! Could you suggest, which namelist parameters in COSMO - CLM INPUT have to be switched on/off (or another values) due to this ‘luse_t_skin = .FALSE.’ setting?
Thanks.

View in channel
<p> Hi all, I have a new queston assoiated with the cited above. I’m using the <span class="caps"> NCEP </span> - <span class="caps"> CFSR </span> driving data from <span class="caps"> DKRZ </span> . INT2LM run doesn’t work, I have attached the log-file (slurm-1054836.out). As I have understand, the problem is in “ T_SKIN is used for T_S” and the following “forrtl: severe (174): <span class="caps"> SIGSEGV </span> , segmentation fault occurred” etc. I have changed luse_t_skin from <span class="caps"> TRUE </span> to <span class="caps"> FALSE </span> , but the result is the same: *** T(ke) is used for T_S *** (see file slurm-1055109.out). So, the problem is that <span class="caps"> NCEP </span> - <span class="caps"> CFSR </span> data doesn’t contain T_S parameter? How can I run INT2LM without using T_S or, maybe, it can be used in another way? <br/> Thank you very much for any suggestions! </p>

  @vladimirplatonov in #ab015a8

<p> Hi all, I have a new queston assoiated with the cited above. I’m using the <span class="caps"> NCEP </span> - <span class="caps"> CFSR </span> driving data from <span class="caps"> DKRZ </span> . INT2LM run doesn’t work, I have attached the log-file (slurm-1054836.out). As I have understand, the problem is in “ T_SKIN is used for T_S” and the following “forrtl: severe (174): <span class="caps"> SIGSEGV </span> , segmentation fault occurred” etc. I have changed luse_t_skin from <span class="caps"> TRUE </span> to <span class="caps"> FALSE </span> , but the result is the same: *** T(ke) is used for T_S *** (see file slurm-1055109.out). So, the problem is that <span class="caps"> NCEP </span> - <span class="caps"> CFSR </span> data doesn’t contain T_S parameter? How can I run INT2LM without using T_S or, maybe, it can be used in another way? <br/> Thank you very much for any suggestions! </p>
VP replied

Hi all, I have a new queston assoiated with the cited above. I’m using the NCEP - CFSR driving data from DKRZ . INT2LM run doesn’t work, I have attached the log-file (slurm-1054836.out). As I have understand, the problem is in “ T_SKIN is used for T_S” and the following “forrtl: severe (174): SIGSEGV , segmentation fault occurred” etc. I have changed luse_t_skin from TRUE to FALSE , but the result is the same: *** T(ke) is used for T_S *** (see file slurm-1055109.out). So, the problem is that NCEP - CFSR data doesn’t contain T_S parameter? How can I run INT2LM without using T_S or, maybe, it can be used in another way?
Thank you very much for any suggestions!

View in channel
<p> Can you provide the namelist? </p>

  @burkhardtrockel in #49dcda0

<p> Can you provide the namelist? </p>
BR replied

Can you provide the namelist?

View in channel
<p> Yes, I have attached the <span class="caps"> INPUT </span> file containing all namelists. </p>

  @vladimirplatonov in #c8955d0

<p> Yes, I have attached the <span class="caps"> INPUT </span> file containing all namelists. </p>
VP replied

Yes, I have attached the INPUT file containing all namelists.

View in channel
<p> The <span class="caps"> CFSR </span> data are on pressure levels and not on hybrid pressure-sigma levels which is used as default by INT2LM for interpolation. Therefore the data are interpolated to hybrid-pressure first (see subroutine src_pressure_to_hybrid.f90). The interpolation is triggered by the namelist parameter ke_hybrid. ke_hybrid can only take two values 31 and 38 presently. In the case of <span class="caps"> CFSR </span> you need to set ke_hybrid=38 in the <span class="caps"> GRID </span> _IN part of the namelist (see example below). <br/> Please also set the following in the <span class="caps"> CONTRL </span> of the namelist lprog_qi = .FALSE., luse_t_skin = .TRUE., </p> <pre> &amp;GRID_IN pcontrol_fi =30000., ie_in_tot = 720, je_in_tot = 361, ke_in_tot = 36, ke_hybrid=38, startlat_in_tot = -90., startlon_in_tot = -180., endlat_in_tot = 90., endlon_in_tot = 179.5, pollat_in = 90.0, pollon_in = 180.0, lushift_in=.FALSE.,.FALSE., lvshift_in=.FALSE.,.FALSE., ke_soil_in=3, czml_soil_in=0.05, 0.25, 0.7, 1.5, /END </pre>

  @burkhardtrockel in #8614fa7

<p> The <span class="caps"> CFSR </span> data are on pressure levels and not on hybrid pressure-sigma levels which is used as default by INT2LM for interpolation. Therefore the data are interpolated to hybrid-pressure first (see subroutine src_pressure_to_hybrid.f90). The interpolation is triggered by the namelist parameter ke_hybrid. ke_hybrid can only take two values 31 and 38 presently. In the case of <span class="caps"> CFSR </span> you need to set ke_hybrid=38 in the <span class="caps"> GRID </span> _IN part of the namelist (see example below). <br/> Please also set the following in the <span class="caps"> CONTRL </span> of the namelist lprog_qi = .FALSE., luse_t_skin = .TRUE., </p> <pre> &amp;GRID_IN pcontrol_fi =30000., ie_in_tot = 720, je_in_tot = 361, ke_in_tot = 36, ke_hybrid=38, startlat_in_tot = -90., startlon_in_tot = -180., endlat_in_tot = 90., endlon_in_tot = 179.5, pollat_in = 90.0, pollon_in = 180.0, lushift_in=.FALSE.,.FALSE., lvshift_in=.FALSE.,.FALSE., ke_soil_in=3, czml_soil_in=0.05, 0.25, 0.7, 1.5, /END </pre>
BR replied

The CFSR data are on pressure levels and not on hybrid pressure-sigma levels which is used as default by INT2LM for interpolation. Therefore the data are interpolated to hybrid-pressure first (see subroutine src_pressure_to_hybrid.f90). The interpolation is triggered by the namelist parameter ke_hybrid. ke_hybrid can only take two values 31 and 38 presently. In the case of CFSR you need to set ke_hybrid=38 in the GRID _IN part of the namelist (see example below).
Please also set the following in the CONTRL of the namelist lprog_qi = .FALSE., luse_t_skin = .TRUE., 

 &GRID_IN
  pcontrol_fi =30000., 
  ie_in_tot = 720, je_in_tot = 361, ke_in_tot = 36, ke_hybrid=38,
  startlat_in_tot = -90., startlon_in_tot = -180.,
  endlat_in_tot = 90., endlon_in_tot = 179.5,
  pollat_in = 90.0, pollon_in = 180.0,
  lushift_in=.FALSE.,.FALSE., lvshift_in=.FALSE.,.FALSE.,
  ke_soil_in=3,   
  czml_soil_in=0.05, 0.25, 0.7, 1.5,
 /END
View in channel
<p> Thank you very much! It works. Could you suggest, should I change any namelist parameters in the following cosmo5 run? <br/> Thanks. </p>

  @vladimirplatonov in #7c32ff5

<p> Thank you very much! It works. Could you suggest, should I change any namelist parameters in the following cosmo5 run? <br/> Thanks. </p>
VP replied

Thank you very much! It works. Could you suggest, should I change any namelist parameters in the following cosmo5 run?
Thanks.

View in channel
<p> Since there is no cloud ice water <code> lprog_qi = .FALSE. </code> in your INT2LM output, make sure to set in <span class="caps"> GRIBIN </span> of the <span class="caps"> CCLM </span> namelist lana_qi = .FALSE., llb_qi = .FALSE., </p>

  @burkhardtrockel in #bc8cba0

<p> Since there is no cloud ice water <code> lprog_qi = .FALSE. </code> in your INT2LM output, make sure to set in <span class="caps"> GRIBIN </span> of the <span class="caps"> CCLM </span> namelist lana_qi = .FALSE., llb_qi = .FALSE., </p>
BR replied

Since there is no cloud ice water lprog_qi = .FALSE. in your INT2LM output, make sure to set in GRIBIN of the CCLM namelist lana_qi = .FALSE., llb_qi = .FALSE.,

View in channel