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

Interruption in running WRF4.3.3 with LCZ map

fpw

New member
Hello everyone! I am trying to use a single-layer urban canopy model and LCZ map in WRF4.3.3. I used the w2w program to input the LCZ information into the geo_cem.d01 file. After WPS finished running, I used mpirun-np 20 to run real and wrf. After successfully running the real program, there was a message as shown in Figure 1. When running the wrf program, it quickly ended, but there were no obvious errors. I know that the lowest model elevation in SLUCM is higher than the highest building height, so I set dzbot=53 in namglist. The attachment contains the namelist. input file and the generated rsl. error. 0000 file that I ran wrf on.
I hope someone can help me, this is very important to me.
Please any help would be much much much appreciated!
Wishing everything all the best!
Thank you!
 

Attachments

  • 图片1.png
    图片1.png
    94.2 KB · Views: 5
  • namelist.input
    5.4 KB · Views: 4
  • rsl.error.0000
    4.2 KB · Views: 1
Hi,
Apologies for the delay in response while our team tended to time-sensitive obligations. Thank you for your patience.

I have a couple of thoughts:

1) Can you issue the following command in the directory where your rsl* files reside?
Code:
grep cfl rsl*
. If it finds any CFL errors, see Segmentation Faults and CFL Errors
2) According to the rsl.error.0000 file you shared, you are using a total of 16 processors, which may not be enough for the size of your domain, especially with the high resolution and complex physics. See Choosing an Appropriate Number of Processors to determine a reasonable number to use if you tried to increase the number of processors.
3) I see the grid-spacing you're using for your domain is 1x1 km. What is the resolution of the input data? The recommendation is that d01 should not have a resolution (dx/dy) with a larger ratio than about 1:5. If the ratio is much larger, you likely need to consider having a more coarse domain parent around the domain you have set up now.
4) If none of the above help, can you try running this without using the LCZ map and turning off urban physics? This will give us an idea of whether those options are causing the issue.

Thanks!
 
Hi,
Apologies for the delay in response while our team tended to time-sensitive obligations. Thank you for your patience.

I have a couple of thoughts:

1) Can you issue the following command in the directory where your rsl* files reside?
Code:
grep cfl rsl*
. If it finds any CFL errors, see Segmentation Faults and CFL Errors
2) According to the rsl.error.0000 file you shared, you are using a total of 16 processors, which may not be enough for the size of your domain, especially with the high resolution and complex physics. See Choosing an Appropriate Number of Processors to determine a reasonable number to use if you tried to increase the number of processors.
3) I see the grid-spacing you're using for your domain is 1x1 km. What is the resolution of the input data? The recommendation is that d01 should not have a resolution (dx/dy) with a larger ratio than about 1:5. If the ratio is much larger, you likely need to consider having a more coarse domain parent around the domain you have set up now.
4) If none of the above help, can you try running this without using the LCZ map and turning off urban physics? This will give us an idea of whether those options are causing the issue.

Thanks!
Thank you for your reply!

I have already solved the above problem, but I have a new one recently. If possible, I hope you can help me.
I plan to use WRF's SLUCM model to study the impact of anthropogenic heat flux changes, and to simulate scenarios with and without AH by modifying the parameters of AH and the ahoption switch in URBPARM.TBL. At the beginning, I used single-layer nesting as a practice to simulate the above two scenarios, and there were significant differences in the HFX parameters in the WRF output results. Later, I used double-layer nesting for the same scenario simulation, but regardless of whether I set "AH=0 and ahoption=0" or "AH is not equal to 0 and ahoption=1", the output results of HFX in both scenarios remained the same without any change. I am very confused about this. The attachment contains the namelist. input used for double-layer nesting.
I have tested both WRF4.5 and 4.6 versions, and they both contain the above-mentioned issues.
Looking forward to your reply!

A lot of thanks.
 

Attachments

  • namelist.input.txt
    6.7 KB · Views: 3
Last edited:
I have already solved the above problem, but I have a new one recently. If possible, I hope you can help me.
Hi,
I'm so glad to hear your original issue is resolved. Since this new issue is different than the one orignally posted here, please start a new thread for this problem. Keeping each post focused on one issue makes the forum easier to read and search in the future. Thanks!
 
Top