GRID_DOT_2D column mismatch error when running two-way WRF-CMAQ, problem with griddesc?

asmontgomery · July 21, 2021, 6:44pm

Hello everyone,

I’m running a two-way wrf v3.8 - cmaq v5.2. I’ve been experimenting with various physics options and start days, but I keep getting cfl=2 errors and other grid mismatch errors but I can’t figure out where I went wrong and why my run gets up to the 6 minute mark. I’ve attached my my namelist.input , if needed, but the rsl.out.000 file was too large …

In brief, I’ve identified several weird things:

It says it fails on BC but theres nothing wrong with BC
There exist cfl=2 errors from time 00:02:36 until it stops running around 00:05:12
The met_cro_3d and pmet_cro_3d (and griddot, gricro) files have different dimensions for some reason, and I cannot figure out why – the GRIDDESC file is not created during the run however, I did pull in a GRIDDESC file into the output directory
It can’t find the variable uwind/vwind inside metcro files, but inside the pmetcro files exist uwindc, vwindc – why is this different?

Basically, seems like a lot of things are going wrong but I don’t know if these problems are all related or totally disjointed. Something may be wrong with my griddesc too.

Any insight is very much appreciated, thank you.

Thanks,
Anastasia

PS: Extra rsl excerpts to support points 1-4

tail rsl.out.0000

     >>--->> WARNING in subroutine GET_EMIS:INTERPX
     Variable "TOLU" not in file EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine GET_EMIS on PE 000
     Could not read TOLU from EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine GET_EMIS:INTERPX
     Variable "HGNRVA" not in file EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine GET_EMIS on PE 000
     Could not read HGNRVA from EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine GET_EMIS:INTERPX
     Variable "HGIIGAS" not in file EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine GET_EMIS on PE 000
     Could not read HGIIGAS from EMIS_1
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
          Temporal BEIS at time 0:05:12
     InLine LNOx production
     LNOx production step-starting date and time: 2015357 000448

     VDEMIS_LT sum:  2015357:000512   1.41742E-03
  
     Calculating emissions point source layer fractions for 000512
     
     >>--->> WARNING in subroutine READMD3:INTERPX
     Variable "UWIND" not in file MET_DOT_3D
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
     
     >>--->> WARNING in subroutine READMD3:INTERPX
     Variable "VWIND" not in file MET_DOT_3D
     M3WARN:  DTBUF 0:05:12   Dec. 23, 2015
  

     Maximum eddy diffusivity of:   421.89     (m**2/sec)
     at col, row, layer:   1,  37,  13
     corresponding to a free tropospheric wind shear of:  5.26842E-03 (/sec),
                            a bulk Richardson Number of:  -15.343    ,
     and pot. temps. in layer and layer+1:   292.29       291.62    
     
 after     VDIFF G  1.5830660E-01 A  3.7181008E+07 N  2.6839930E-06
 after COUPLE_WR G  1.3684100E+03 A  6.2489808E+11 N  2.3392845E-02
     
     >>--->> WARNING in subroutine Read BCs on PE 000
     Applying fix to aerosol Boundary Conditions for aerosol modes. Turn on verbose_rdbcon preprocessor flag to learnmore.
     M3WARN:  DTBUF 0:04:48   Dec. 23, 2015
 after      HADV G  1.3657651E+03 A  6.2040460E+11 N  2.3434713E-02
 after       ADV G  1.3627283E+03 A  6.1979073E+11 N  2.3418942E-02

     H-eddy DT & integration steps:   4.8000000E+01       1
 after     HDIFF G  1.3627278E+03 A  6.1979112E+11 N  2.3418935E-02
 after DECOUPLE_ G  1.5810940E-01 A  3.6935504E+07 N  2.7005131E-06

It also has cfl=2 errors in this file (rsl.out.0000
line 61137):

d01 2015-12-23_00:05:36            1          37          24 vert_cfl,w,d(eta)=   1.088588      -8.037120      5.0000008E-02
d01 2015-12-23_00:05:36            1          37          25 vert_cfl,w,d(eta)=   1.085756      -6.878208      5.0000008E-02
d01 2015-12-23_00:05:36            1          37          26 vert_cfl,w,d(eta)=   1.061352      -4.679369      5.0000008E-02
d01 2015-12-23_00:05:36            1          37          27 vert_cfl,w,d(eta)=   1.020186      -1.217384      4.9999982E-02
d01 2015-12-23_00:05:36          197 points exceeded cfl=2 in domain d01 at time 2015-12-23_00:05:36 hours
d01 2015-12-23_00:05:36  MAX AT i,j,k:           35           1          13 vert_cfl,w,d(eta)=   2.339636      -1.148410      1.0000050E-02

This mismatch between the met_cro_3d and pmet_cro_3d files where the ncols and nrows do not match – note this error repeats for GRID_CRO_2D, GRID_DOT_2D, (rsl.out.0000 line 9264 -):

   "GRID_CRO_2D" opened as OLD:READ-WRITE
     File name "BUFFERED"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 37 rows, 73 cols, 1 lays, 47 vbles
     BUFFERED "file" opened as READWRITE
     Time-independent data.
     PGRID_CRO_2D    :/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/GRID_CRO_2D20151223

     >>--->> WARNING in subroutine OPEN3
     File not available.

  Error: Could not open file PGRID_CRO_2D    for update

     Value for IOAPI_CHECK_HEADERS not defined;returning default:   FALSE
     Value for IOAPI_OFFSET_64:  Y returning TRUE

     "PGRID_CRO_2D" opened as NEW(READ-WRITE )
     File name "/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/GRID_CRO_2D20151223"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 726 rows, 1155 cols, 1 lays, 47 vbles
     NetCDF ID:    393216  opened as VOLATILE READWRITE
     Time-independent data.
     Value for IOAPI_LOG_WRITE:  F returning FALSE
     Value for IOAPI_LOG_WRITE:  F returning FALSE

     "GRID_DOT_2D" opened as OLD:READ-WRITE
     File name "BUFFERED"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 38 rows, 74 cols, 1 lays, 1 vbles
     BUFFERED "file" opened as READWRITE
     Time-independent data.
     PGRID_DOT_2D    :/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/GRID_DOT_2D20151223

     >>--->> WARNING in subroutine OPEN3
     File not available.

  Error: Could not open file PGRID_DOT_2D    for update

     "PGRID_DOT_2D" opened as NEW(READ-WRITE )
     File name "/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/GRID_DOT_2D20151223"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 727 rows, 1156 cols, 1 lays, 1 vbles
     NetCDF ID:    458752  opened as VOLATILE READWRITE
     Time-independent data.

     "MET_CRO_3D" opened as OLD:READ-WRITE
     File name "BUFFERED"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 39 rows, 75 cols, 35 lays, 15 vbles
     BUFFERED "file" opened as READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          000048 (0:00:48 hh:mm:ss)
     Maximum current record number         0
     PMET_CRO_3D     :/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_CRO_3D20151223

     >>--->> WARNING in subroutine OPEN3
     File not available.

  Error: Could not open file PMET_CRO_3D     for update

     "PMET_CRO_3D" opened as NEW(READ-WRITE )
     File name "/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_CRO_3D20151223"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 728 rows, 1157 cols, 35 lays, 15 vbles
     NetCDF ID:    524288  opened as VOLATILE READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          010000 (1:00:00 hh:mm:ss)
     Maximum current record number         0

     "MET_DOT_3D" opened as OLD:READ-WRITE
     File name "BUFFERED"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 40 rows, 76 cols, 35 lays, 4 vbles
     BUFFERED "file" opened as READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          000048 (0:00:48 hh:mm:ss)
     Maximum current record number         0
     PMET_DOT_3D     :/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_DOT_2D20151223

     >>--->> WARNING in subroutine OPEN3
     File not available.

  Error: Could not open file PMET_DOT_3D     for update

     "PMET_DOT_3D" opened as NEW(READ-WRITE )
     File name "/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_DOT_2D20151223"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 727 rows, 1156 cols, 35 lays, 4 vbles
     NetCDF ID:    589824  opened as VOLATILE READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          010000 (1:00:00 hh:mm:ss)
     Maximum current record number         0

     "MET_CRO_2D" opened as OLD:READ-WRITE
     File name "BUFFERED"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 39 rows, 75 cols, 1 lays, 32 vbles
     BUFFERED "file" opened as READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          000048 (0:00:48 hh:mm:ss)
     Maximum current record number         0
     PMET_CRO_2D     :/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_CRO_2D20151223

     >>--->> WARNING in subroutine OPEN3
     File not available.
  Error: Could not open file PMET_CRO_2D     for update

     "PMET_CRO_2D" opened as NEW(READ-WRITE )
     File name "/projects/b1045/wrf-cmaq/output/CONUS4K/output_CONUS4K_d02_4km_sf_rrtmg_12_8_1_v3852/MET_CRO_2D20151223"
     File type GRDDED3
     Execution ID "????????????????"
     Grid name "CONUS4K_d02"
     Dimensions: 726 rows, 1155 cols, 1 lays, 32 vbles
     NetCDF ID:    655360  opened as VOLATILE READWRITE
     Starting date and time  2015357:000000 (0:00:00   Dec. 23, 2015)
     Timestep                          010000 (1:00:00 hh:mm:ss)
     Maximum current record number         0
d01 2015-12-23_00:00:00  DEBUG wrf_timetoa():  returning with str = [2015-12-23_00:00:00]
d01 2015-12-23_00:00:00  DEBUG wrf_timetoa():  returning with str = [2015-12-23_00:00:00]
d01 2015-12-23_00:00:00  DEBUG wrf_timetoa():  returning with str = [2015-12-24_00:00:00]
d01 2015-12-23_00:00:00  DEBUG wrf_timeinttoa():  returning with str = [0000000000_000:000:006]

asmontgomery · July 28, 2021, 6:40pm

And as a further update, I was trying to recreate the error from above, and now I am successfully creating the GRIDDESC file – not sure what was happening before (potentially missing libraries as our supercomputer was being updated), but I reran with this namelist.input namelist.input.txt (5.7 KB)
.

However, this GL_NCOLS problem still persists. I think this is the key problem, beause if the metcro files are not being created properly, then wouldn’t that also cause cfl errors? If someone could help me troubleshoot why I’m getting this error.

     WARNING: Domain extents different from model for MET_CRO_3D
 >>  GL_NCOLS:         1155   NCOLS3D:           75

     WARNING: Domain extents different from model for MET_CRO_3D
 >>  GL_NROWS:          726   NROWS3D:           39

asmontgomery · July 30, 2021, 8:39pm

So I got an answer from someone at the EPA via email (thank you) and another from @jschnell , and the answer was quite simple.

BC was fine, not the actual error, as expected.
cfl =2 errors happen at boundaries and this is the key that cracked it open: I needed to adjust the spec_bdy_width = 5, spec_zone = 1, relax_zone = 4 because the errors were on the boundaries.
the GL_NROWS/NCOLS error apparently can be ignored.
The uwindc/vwindc error disappeared.

Further, I was also advised to change my cmaq_wrf_freq to 10 instead of 8 because the rule of thumb is TSTEP must be divisible by WRF_CMAQ_FREQ x time_step.

So yea, I imagine more errors in the future but for now I’ve gotten past 5:12. Solution.

Topic		Replies	Views
GRIDDESC error when running twoway WRF-CMAQ Run Time Errors and Issues	7	1130	July 1, 2021
Fortran runtime error CMAQ	6	115	May 14, 2024
WRF-CMAQ error in two day test-data run WRF-CMAQ	5	442	January 19, 2023
Cmaq - cctm error Run Time Errors and Issues	3	848	March 11, 2019
WRF 4.3 module_sf_noahdrv.f90 error WRF-CMAQ	2	967	June 14, 2021

GRID_DOT_2D column mismatch error when running two-way WRF-CMAQ, problem with griddesc?

Related topics