WRF stopped when running in SLUCM-LCZ mode.

[fc7102]

Hello, my WRF and WPS version is 4.3.3 and 4.3.1. I was running SLUCM in LCZ classification which is newly built in WRF. So I modified namelist like this:

sf_urban_physics         = 1,	1,	1,
use_wudapt_lcz	     = 1,
num_land_cat             = 41,
sf_sfclay_physics        = 1,        1,        1,
sf_surface_physics       = 2,        2,        2,
bl_pbl_physics           = 1,        1,        1,

And I also modified URBPARM_LCZ.TBL to adapt my simulation region. When I modified the parameter ROAD_WIDTH in URBPARM_LCZ.TBL, wrf always stopped with no error in this place (the whole crash log is attached below in wrf.log):

calling inc/HALO_PWP_inline.inc
in SFCLAY
in NOAH DRV

The default ROAD_WIDTH is

ROAD_WIDTH: 98.9, 39.2, 108.0, 108.0, 108.0, 108.0, 108.0, 108.0, 108.0, 108.0, 108.0

.
And I modified it to

ROAD_WIDTH: 32.0, 17.5, 9.0, 32.0, 20.0, 10.5, 7.0, 28.8, 10.0, 17.5, 10.0

Then I modified one parameter in a time, I found when I change the road width of the classification LCZ1 and LCZ4 to larger than 90, wrf can run successfully. But 90m road width is too large to the real condition in my simulation region.
I will attach some files which I think is useful, the road_width in URBPARM_LCZ.TBL was changed to default value to run wrf.

Thanks in advance,
Yao.

 

[Ming Chen]

Hi,
I am sorry I don't have an immediate answer to your question. LCZ is relatively new and we don't have many experiences with it.
Would you please take a look at these publications and hopefully you can find some helpful information:
Stewart ID, Oke TR. Local Climate Zones for Urban Temperature Studies. Bull Am Meteorol Soc. 2012;93(12):1879-1900. doi:10.1175/BAMS-D-11-00019.1
Stewart ID, Oke TR, Krayenhoff ES. Evaluation of the ‘local climate zone’ scheme using temperature observations and model simulations. Int J Climatol. 2014;34(4):1062-1080. doi:10.1002/joc.3746
Varentsov M, Samsonov T, Demuzere M. Impact of Urban Canopy Parameters on a Megacity’s Modelled Thermal Environment. Atmosphere (Basel). 2020;11(12):1349. doi:10.3390/atmos11121349

 

[fc7102]

Hi,
I am sorry I don't have an immediate answer to your question. LCZ is relatively new and we don't have many experiences with it.
Would you please take a look at these publications and hopefully you can find some helpful information:
Stewart ID, Oke TR. Local Climate Zones for Urban Temperature Studies. Bull Am Meteorol Soc. 2012;93(12):1879-1900. doi:10.1175/BAMS-D-11-00019.1
Stewart ID, Oke TR, Krayenhoff ES. Evaluation of the ‘local climate zone’ scheme using temperature observations and model simulations. Int J Climatol. 2014;34(4):1062-1080. doi:10.1002/joc.3746
Varentsov M, Samsonov T, Demuzere M. Impact of Urban Canopy Parameters on a Megacity’s Modelled Thermal Environment. Atmosphere (Basel). 2020;11(12):1349. doi:10.3390/atmos11121349

Thanks for your reply. I read the paper you recommended, they are useful for me, but I can't find the solution of wrf simulation in these papers. Actually I also tried SLUCM with 31~33 classification, the ROAD_WIDTH is no problem. And I also read some wrf-lcz papers, their ROAD_WIDTH are also close to real situation (about 10~50m). Due to my lackness of fortran, I also can't find the reason in module_sf_urban.F. What I want to know is why the ROAD_WIDTH is so big in WRF? Is this a bug? Or ROAD_WIDTH should be dm in unit?

 

[Ming Chen]

Hi,
We have talked to developers who work in the field of land/urban schemes. They said that most of those parameter values are specified based on limited in-situ observations and/or specific case studies. Therefore, the default values provided in WRF can be treated as kind of 'reference'. Users should be able to modify these values based on their specific cases. I hope this makes sense to you.

 

[fc7102]

Hi,
We have talked to developers who work in the field of land/urban schemes. They said that most of those parameter values are specified based on limited in-situ observations and/or specific case studies. Therefore, the default values provided in WRF can be treated as kind of 'reference'. Users should be able to modify these values based on their specific cases. I hope this makes sense to you.

Thank you. But my problem is the ROAD_WIDTH parameter in URBPARM_LCZ.TBL. The road width has to be large than 90 meters to make wrf run successfully, which restricted me to modified it. I think developers may know why, because the default road width in URBPARM.TBL is normal (about 10m to 30m). Maybe the unit of the road width in URBPARM_LCZ.TBL is not meter? Or something else?

 

[fc7102]

After a few days working, I think I can answer the question by myself.
1.About the sudden stop of WRF:
When the road width is too small, some problems will happened in boundary layer processing by the large HGT_TBL, R_TBL or some other variables(I can not confirm which variable now) related to road width.
2.About the large road width in default URBPARM_LCZ.TBL
At first, I just scale the roof width and road width in Google Earth and think this is the value. But after I read more papers, I realized that these two parameters show the ratio of building and impervious surface exclude building respectively. Then I scale the area of building and other impervious surface, I found that the default parameter is responsible. But some changes is still necessary to make the simulation better.
Thank the answer by Chen in previous again.
Yao.

 

[Ming Chen]

Yao,
Many thanks for the update and the kind information. Please keep me updated if you make any changes to get better results.