Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

NLCD40 WRF run crashes because of module_sf_noahdrv.f90 (unit = 19, file = 'VEGPARM.TBL')

ashreeva

New member
Hello,

I am trying to simulate a heatwave over California using WRF (NLCD Landuse). In order to make it work with NLCD, I added the line: geog_data_res = 'nlcd2011_30m+5m', 'nlcd2011_30m+2m', 'nlcd2011_30m+urbfrac_nlcd2011+30s', to my namelist.wps. I then did the following remapping procedure to make sure it works with the Urban Canopy Model (UCM, sf_urban_physics=1):
31 (Dwarf scrub) --> 7 (Open shrublands)
32 (Shrubs) --> 7 (Open shrublands)
33 (Grasslands) --> 10 (Grasslands)
23 (Developed open space) --> 31
24 (Low intensity) --> 31
25 (Mid intensity) --> 32
26 (High intensity) --> 33
With these changes, I was able to run real.exe successfully, but now my wrf.exe is crashing with the following message:

"INPUT LandUse = "NLCD40"
LANDUSE TYPE = "NLCD40" FOUND 40 CATEGORIES 2 SEASONS WATER CATEGORY = 17 SNOW CATEGORY = 15
Using NLCD40 for Noah, redefine urban categories
INITIALIZE THREE Noah LSM RELATED TABLES
Skipping over LUTYPE = USGS
Skipping over LUTYPE = MODIFIED_IGBP_MODIS_NOAH
LANDUSE TYPE = NLCD40 FOUND 40 CATEGORIES
At line 1981 of file module_sf_noahdrv.f90 (unit = 19, file = 'VEGPARM.TBL')
Fortran runtime error: Bad integer for item 1 in list input "

I have attached my rsl.error and namelist.input files to this post. Please let me know how can I proceed and what else can I share to make the debugging easier?

Thanks, Anamika
 

Attachments

  • Archive.zip
    14.9 KB · Views: 33
Anamika,
Would you please run this case using WRFv4.2 and let me know whether it works?
I am trying to find a solution for this issue and hope to get back to you soon. Thanks for your patience.
 
Anamika,
Would you please use the VEGPRM.TBL I attached here to replace the one in WRFV4.3, then rerun your case? Please let me know whether your case can work. Thanks.
 

Attachments

  • VEGPARM.TBL
    22.5 KB · Views: 134
Thank you, Ming Chen, for your response.
I executed my WRF run with the attached VEGPARM.TBL. With this, I was able to get past my initial error and I got a new error this time.

This is the last line of my rsl.error.0000 file:
-------------- FATAL CALLED ---------------
USING 10 WUDAPT LCZ WITHOUT URBPARM_LCZ.TBL. SET USE_WUDAPT_LCZ=1
------------------------------------------------

View attachment rsl files.zip

I have the URBPARM_LCZ.TBL in the same WRF/run/ folder though. So I am not sure what the error is.I am using NLCD urban classification system so could you please tell me if it is compatible with the LCZ urban morphology parameters (URBPARM_LCZ.TBL)? Also, I had remapped the urban classes as discussed in my original post. Could you please tell me if that is correct?
 
Please look at the code phys/module_physics_init.F, find the following piece of code:
Code:
               IF (config_flags%use_wudapt_lcz.eq.0 .and. max_utype_urb2d.gt.3.0) THEN  !new LCZ
                 CALL wrf_error_fatal &
                 ('USING 10 WUDAPT LCZ WITHOUT URBPARM_LCZ.TBL. SET USE_WUDAPT_LCZ=1')
               ENDIF

You can comment out the above lines, then recompile WRF and run your case.
Let me know whether it works.
 
Hello, Ming Chen,

I believe that I am past the previous error that we discussed here. But I am now getting a Segmentation error.


Here's the rsl.error.0000 file:

USING DEFAULT URBAN MORPHOLOGY
USING DEFAULT URBAN MORPHOLOGY
USING DEFAULT URBAN MORPHOLOGY
USING DEFAULT URBAN MORPHOLOGY
USING DEFAULT URBAN MORPHOLOGY
USING DEFAULT URBAN MORPHOLOGY
ThompMP: read qr_acr_qgV3.dat instead of computing
ThompMP: read qr_acr_qsV2.dat instead of computing
ThompMP: read freezeH2O.dat instead of computing
mediation_integrate.G 1944 DATASET=HISTORY
mediation_integrate.G 1945 grid%id 1 grid%oid 1
Timing for Writing /scratch_lg/uhi-modeling/WRF_Home/OUTPUTS/LA2020/wrfout_d01_2020-08-01_00:00:00 for domain 1: 0.65176 elapsed seconds
d01 2020-08-01_00:00:00 Input data is acceptable to use: wrfbdy_d01
Timing for processing lateral boundary for domain 1: 0.03110 elapsed seconds
Tile Strategy is not specified. Assuming 1D-Y
WRF TILE 1 IS 1 IE 234 JS 1 JE 180
WRF NUMBER OF TILES = 1

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0 0x2aaaab11021d in ???
#1 0x2aaaab10f463 in ???
#2 0x2aaaabcf03ff in ???
#3 0x2aaaabd3af3b in ???

Could it be a time_step issue? I am attaching my namelist.input file here. Please advice.

Anamika
 

Attachments

  • namelist.input
    3.9 KB · Views: 33
Hi, Anamika,
I looked at your namelist.input and everything looks just fine.
In WRFV4.3, a new option (LCZ) is introduced and some codes are modified. Despite the tests with Modis and USGS as input, I guess there is no test is conducted using NLCD as input. This is why we have troubles in your case.
With the changes in VEGPARM.TBl and phys/module_physics_init.F, I suppose the first two issues you reported have been resolved. Please let me know if I am wrong.
I also would like to know whether the same case works when using WRRFV4.2? We need to confirm that this problem is specific for WRFV4.3, which will give us some hints what is wrong.
 
UPDATE: I found a solution to the SEGMENTATION FAULT issue.

I noticed that my model ran just fine if I only un the largest domain, but only when I add the smallest sub-domain does it break down. I then found the following resource: https://forum.mmm.ucar.edu/phpBB3/viewtopic.php?t=8508 which suggests that it might be a wrong choice of processors. I then followed the guidelines provided here on the post "How many processors should I use to run WRF?": https://forum.mmm.ucar.edu/phpBB3/viewtopic.php?f=73&t=5082 which helped me figure out that a square number of processors is optimal for me.

Running the job on 36 processors on 4 nodes seems to be working just fine.

All this to say that the records kept here at the WRF forum are pretty cool and I hope that this thread helps someone too. :)

Cheers,
Anamika
 
Anamika,
Would you please use the VEGPRM.TBL I attached here to replace the one in WRFV4.3, then rerun your case? Please let me know whether your case can work. Thanks.


I am running WRF V4.4 and I was getting the same errors as Anamika's first post. After replacing the V4.4 vegparm.tbl with the one you gave here, it worked then. I am not sure if anyone else has faced the same problem with V4.4, but I thought I should let you guys know.

Thanks!
 
Hi Anamika, I'm curious what you exactly meant by "I then did the following remapping procedure". Did you change some netcdf files manually? If so, which files? Thanks in advance. Dan
 
Top