Custom Query (210 matches)

Does FLEXWRF v3.2 use the namelist.input or namelist.wps files? Should copies of these files be placed in any particular folder? Is there a FLEXWRF sub-routine that compares the meteorological fields defined in the namelist files with the field defined in the flexwrf.input file?

Hi there, My job output file contains the following error text when I attempt to run FLEXPART. I don't know what the error means. I have checked the values for each variable in my flexwrf.input file and they appear to be set using the correct type. What can I do to further troubleshoot the issue?

Thank you

the input file used is flexwrf.input in the local folder of the executable

Pre-generating random numbers Calling readinput Opening ' flexwrf.input

' for reading

Reading pathnames

forrtl: severe (59): list-directed I/O syntax error, unit 1, file /project/6006014/kcmrowse/FLEX/flexpart_wrf-3.3/flexwrf.input Image PC Routine Line Source libifcoremt.so.5 00002ACB5B91CDF2 forio_return Unknown Unknown libifcoremt.so.5 00002ACB5B96582F for_read_seq_lis_ Unknown Unknown flexwrf32_intel_o 00000000004C35CA Unknown Unknown Unknown flexwrf32_intel_o 00000000004607E7 Unknown Unknown Unknown flexwrf32_intel_o 000000000040327E Unknown Unknown Unknown libc-2.24.so 00002ACB5DB8B2E0 libc_start_main Unknown Unknown flexwrf32_intel_o 000000000040317A Unknown Unknown Unknown

Dear all,

in FLEXPART 9.2 I stumbled upon what is most probably a bug, and it is still present in the FLEXPART git dev branch.

Comparing verttransform.f90 (now seems to be verttransform_ecmwf.f90) to verttransform_gfs.f90, around line 500, one notices pieces of code which probably should be identical, but they aren't:

for ECMWF it looks like:

     if (vv(nx/2-1,0,iz,n).lt.0.) then
        ddpol=atan(uu(nx/2-1,0,iz,n)/ &
             vv(nx/2-1,0,iz,n))+xlonr
      else if (vv(nx/2-1,0,iz,n).gt.0.) then
        ddpol=pi+atan(uu(nx/2-1,0,iz,n)/ &
             vv(nx/2-1,0,iz,n))+xlonr

(+ + xlonr)

whereas for GFS:

    if(vv(nx/2-1,0,iz,n).lt.0.) then
        ddpol=atan(uu(nx/2-1,0,iz,n)/vv(nx/2-1,0,iz,n))+xlonr
     elseif (vv(nx/2-1,0,iz,n).gt.0.) then
        ddpol=pi+atan(uu(nx/2-1,0,iz,n)/vv(nx/2-1,0,iz,n))-xlonr

(+ - xlonr)

I would thus propose changes as you can see in the attached picture (i.e. changing to "+ +" in the GFS version, and modifying the infinity case to be consistent with both approaches to the limit).

Unfortunately, I am neither an atmospheric transport expert nor do I have an idea anymore why I fixed it this way (I looked into that like several months ago, and I think I did a comparison with corresponding code pieces in FLEXTRA).

However, I think I write this here now as "as is" and some more expert people maybe check and fix this. Thanks very much!

Stephan

contact: ​http://shaching.userweb.mwn.de