Configuring CCLM for Middle East – in #10: INT2LM

  @burkhardtrockel in #c175b39

<p> We talked about this in the recent trainings course. <br/> Please check items 3 and 4 on the troubleshooting page (login requested) <br/> http://www.clm-community.eu/index.php?menuid=164 <br/> and change the int2lm namelist accordingly. <br/> However, a better alternative may be to recreate the external data file with the <span class="caps"> EXTPAR </span> option instead of the <span class="caps"> PEP </span> one. In the near future the <span class="caps"> PEP </span> option will become obsolete when the new <span class="caps"> EXTPAR </span> version will be released. </p>

  @redc_migration in #b9d94c8

<p> Dear colleagues, <br/> I am attempting now to start runiung cclm-sp_1.4 package over <span class="caps"> CORDEX </span> _MENA domain on our linux machine. <br/> Some progress may be reported – the original cclm-sp_1.4 has been successfully implemented (by Alon Shtivelman with some my participation). I have included the Fopts files used in the tar file attached. <br/> Our efforts to start running the new version over the <span class="caps"> CORDEX </span> _MENA domain (using <span class="caps"> ERA </span> -Int) driving data have been much less successful however. <br/> Since I already have a cclm-sp-1.3.3 version for this experiment ready I attemted to make exactly (I think) the same changes to the template/cclm.job template/int2lm.job and subchain file as those I made for the cclm-sp-1.3.3 case. Please see the results in the ccint2lm.o600038. <br/> Please let me know your recommendations. <br/> Regards, <br/> Simon Krichak </p>

  @burkhardtrockel in #50e6e52

<p> The grid points of the external data and the int2lm grid do not match. The external data set can be larger than the domain defined for the <span class="caps"> CCLM </span> , but in the overlapping region the grid points must match. <br/> In your output you can read <br/> <pre> OPEN: ncdf-file: /Tmp_Research/CLIMAT/Giora/COSMO-CLM/data/ext/MENA50km/domain2013112115045.nc external data file namelist input ie_ext 332 332 je_ext 218 218 lonll -48.4000015258789 -38.0000000000000 latll -28.6000003814697 -7.00000000000000 lonur 97.2399978637695 61.4400000000000 latur 66.8799972534180 45.8000000000000 dlon 0.439999998156038 0.440000000000000 dlat 0.439999989100865 0.440000000000000 pollat 90.0000000000000 90.0000000000000 pollon -180.000000000000 -180.000000000000 polgam 0.000000000000000E+000 0.000000000000000E+000 wrong grid for LM external parameters </pre> <br/> Here (-48.4 – (-38.0))/0.44 = -23.63637 and (-28.6 – (-7))/0.44= 40.09091, but the results have to be whole numbers. </p> <p> You did not send the template int2lm.job.tmpl for version 14. Please check if the following <br/> <pre> ie_ext=332, je_ext=218, ylmext_lfn='domain2013112115045.nc', ylmext_cat='/Tmp_Research/CLIMAT/Giora/COSMO-CLM/data/ext/MENA50km/', </pre> </p>

  @redc_migration in #7c10b1a

<p> Hi, <br/> Thank you. Still do not know what to do. My problem, as I see it, is that I am using exactly (I hope) same setup data in my 1.3.3 and 1.4 versions. With the only difference – the 1.4 crashes. Please see attached all the data I have. <br/> Best, <br/> Simon </p>

  @burkhardtrockel in #495029e

<p> I am not sure what happens in 1.3.3, but the int2m job can definitely not run with the startlon=-38. and startlat=-7. and domain2013112115045.nc unless you made some manipulations in the int2lm code. If you use startlon=-38.28 and startlon=-7.04 it should work. See the ncdump of the rlon/rlat in the following from domain2013112115045.nc. I does not contain a -38. for rlon and not a -7. for rlat: <br/> <pre> rlat = -28.6, -28.16, -27.72, -27.28, -26.84, -26.4, -25.96, -25.52, -25.08, -24.64, -24.2, -23.76, -23.32, -22.88, -22.44, -22, -21.56, -21.12, -20.68, -20.24, -19.8, -19.36, -18.92, -18.48, -18.04, -17.6, -17.16, -16.72, -16.28, -15.84, -15.4, -14.96, -14.52, -14.08, -13.64, -13.2, -12.76, -12.32, -11.88, -11.44, -11, -10.56, -10.12, -9.68, -9.24, -8.8, -8.36, -7.92, -7.48, -7.04, -6.6, -6.16, -5.72, -5.28, -4.84, -4.4, -3.96, -3.52, -3.08, -2.64, -2.2, -1.76, -1.32, -0.88, -0.44, 0, 0.44, 0.88, 1.32, 1.76, 2.2, 2.64, 3.08, 3.52, 3.96, 4.4, 4.84, 5.28, 5.72, 6.16, 6.6, 7.04, 7.48, 7.92, 8.36, 8.8, 9.24, 9.68, 10.12, 10.56, 11, 11.44, 11.88, 12.32, 12.76, 13.2, 13.64, 14.08, 14.52, 14.96, 15.4, 15.84, 16.28, 16.72, 17.16, 17.6, 18.04, 18.48, 18.92, 19.36, 19.8, 20.24, 20.68, 21.12, 21.56, 22, 22.44, 22.88, 23.32, 23.76, 24.2, 24.64, 25.08, 25.52, 25.96, 26.4, 26.84, 27.28, 27.72, 28.16, 28.6, 29.04, 29.48, 29.92, 30.36, 30.8, 31.24, 31.68, 32.12, 32.56, 33, 33.44, 33.88, 34.32, 34.76, 35.2, 35.64, 36.08, 36.52, 36.96, 37.4, 37.84, 38.28, 38.72, 39.16, 39.6, 40.04, 40.48, 40.92, 41.36, 41.8, 42.24, 42.68, 43.12, 43.56, 44, 44.44, 44.88, 45.32, 45.76, 46.2, 46.64, 47.08, 47.52, 47.96, 48.4, 48.84, 49.28, 49.72, 50.16, 50.6, 51.04, 51.48, 51.92, 52.36, 52.8, 53.24, 53.68, 54.12, 54.56, 55, 55.44, 55.88, 56.32, 56.76, 57.2, 57.64, 58.08, 58.52, 58.96, 59.4, 59.84, 60.28, 60.72, 61.16, 61.6, 62.04, 62.48, 62.92, 63.36, 63.8, 64.24, 64.68, 65.12, 65.56, 66, 66.44, 66.88 ;</pre> </p> rlon = -48.4, -47.96, -47.52, -47.08, -46.64, -46.2, -45.76, -45.32, -44.88, -44.44, -44, -43.56, -43.12, -42.68, -42.24, -41.8, -41.36, -40.92, -40.48, -40.04, -39.6, -39.16, -38.72, -38.28, -37.84, -37.4, -36.96, -36.52, -36.08, -35.64, -35.2, -34.76, -34.32, -33.88, -33.44, -33, -32.56, -32.12, -31.68, -31.24, -30.8, -30.36, -29.92, -29.48, -29.04, -28.6, -28.16, -27.72, -27.28, -26.84, -26.4, -25.96, -25.52, -25.08, -24.64, -24.2, -23.76, -23.32, -22.88, -22.44, -22, -21.56, -21.12, -20.68, -20.24, -19.8, -19.36, -18.92, -18.48, -18.04, -17.6, -17.16, -16.72, -16.28, -15.84, -15.4, -14.96, -14.52, -14.08, -13.64, -13.2, -12.76, -12.32, -11.88, -11.44, -11, -10.56, -10.12, -9.68, -9.24, -8.8, -8.36, -7.92, -7.48, -7.04, -6.6, -6.16, -5.72, -5.28, -4.84, -4.4, -3.96, -3.52, -3.08, -2.64, -2.2, -1.76, -1.32, -0.88, -0.44, 0, 0.44, 0.88, 1.32, 1.76, 2.2, 2.64, 3.08, 3.52, 3.96, 4.4, 4.84, 5.28, 5.72, 6.16, 6.6, 7.04, 7.48, 7.92, 8.36, 8.8, 9.24, 9.68, 10.12, 10.56, 11, 11.44, 11.88, 12.32, 12.76, 13.2, 13.64, 14.08, 14.52, 14.96, 15.4, 15.84, 16.28, 16.72, 17.16, 17.6, 18.04, 18.48, 18.92, 19.36, 19.8, 20.24, 20.68, 21.12, 21.56, 22, 22.44, 22.88, 23.32, 23.76, 24.2, 24.64, 25.08, 25.52, 25.96, 26.4, 26.84, 27.28, 27.72, 28.16, 28.6, 29.04, 29.48, 29.92, 30.36, 30.8, 31.24, 31.68, 32.12, 32.56, 33, 33.44, 33.88, 34.32, 34.76, 35.2, 35.64, 36.08, 36.52, 36.96, 37.4, 37.84, 38.28, 38.72, 39.16, 39.6, 40.04, 40.48, 40.92, 41.36, 41.8, 42.24, 42.68, 43.12, 43.56, 44, 44.44, 44.88, 45.32, 45.76, 46.2, 46.64, 47.08, 47.52, 47.96, 48.4, 48.84, 49.28, 49.72, 50.16, 50.6, 51.04, 51.48, 51.92, 52.36, 52.8, 53.24, 53.68, 54.12, 54.56, 55, 55.44, 55.88, 56.32, 56.76, 57.2, 57.64, 58.08, 58.52, 58.96, 59.4, 59.84, 60.28, 60.72, 61.16, 61.6, 62.04, 62.48, 62.92, 63.36, 63.8, 64.24, 64.68, 65.12, 65.56, 66, 66.44, 66.88, 67.32, 67.76, 68.2, 68.64, 69.08, 69.52, 69.96, 70.4, 70.84, 71.28, 71.72, 72.16, 72.6, 73.04, 73.48, 73.92, 74.36, 74.8, 75.24, 75.68, 76.12, 76.56, 77, 77.44, 77.88, 78.32, 78.76, 79.2, 79.64, 80.08, 80.52, 80.96, 81.4, 81.84, 82.28, 82.72, 83.16, 83.6, 84.04, 84.48, 84.92, 85.36, 85.8, 86.24, 86.68, 87.12, 87.56, 88, 88.44, 88.88, 89.32, 89.76, 90.2, 90.64, 91.08, 91.52, 91.96, 92.4, 92.84, 93.28, 93.72, 94.16, 94.6, 95.04, 95.48, 95.92, 96.36, 96.8, 97.24 ; }

  @redc_migration in #1f968af

<p> Hi, <br/> I’ve made the correction suggested (startlon=-38.28 and startlon=-7.04)and have now another problem with segmentation fault during the int2lm stage. I re-ran it with higher debug level. Please have a look at the file attached with (apparently) all the info possible. <br/> Best, <br/> Simon </p>

  @burkhardtrockel in #376df15

<p> Since your boundary data are <span class="caps"> ERAI </span> nterim, you have QI available in the boundary data. Therefore instead of lprog_qi = .FALSE., <br/> please try with lprog_qi = .TRUE., </p>

  @redc_migration in #e08e5d0

<p> Hi, <br/> I have successfully ran the cclm-sp-1_4 for 2 months but then there is a problem at the end of third month of my experiment. <br/> As far as I know I use exactly the same <span class="caps"> ERA </span> -Interim files as those I use with the cclm-sp-1.3.4. <br/> In the arttachment you please find the logouts from my cclm run with high level of debugging and also restart (no debbuging). <br/> Please let me know your recommendations. </p> <p> Simon </p>

  @hans-jürgenpanitz in #1caf817

<p> Hi Simon, </p> <p> That’s very nice that you fell into that trap because since yesterday I am discussing the problem again with <span class="caps"> DWD </span> (I did it already several times). <br/> The problem is that you want to (over)write restart files which already exist in your restart directory. <br/> This fails since the status of such a file in the corresponding open-statement (in io_utilities.f90) is <span class="caps"> NEW </span> . </p> <p> To my opinion this is unlucky, but <span class="caps"> COSMO </span> don’t want to change it from <span class="caps"> NEW </span> to, for example, <span class="caps"> UNKNOWN </span> . </p> <p> So, what can you do? <br/> 1. delete the already existing restart files in your restart directory </p> <p> or, if you want to keep these restart files </p> <p> 2. create a new restart directory and write into this new dir. </p> <p> Hans-Juergen </p>