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

Large precipitation bias after ndown

This post was from a previous version of the WRF&MPAS-A Support Forum. New replies have been disabled and if you have follow up questions related to this post, then please start a new thread from the forum home page.

I am using ndown to generate boundary conditions for a finer simulation. The coarser domain generally well reproduces the reference data but the precipitation of the nested domain is heavily overestimated (please look at the attached figures). The ndown namelist and the run namelists are attached.

I suspect this issue is generated when I run ndown, as I am attempting to change spec_bdy_width (and relax_zone) from 10 points of the parent domain to 5 points in the nested domain. I wonder if it is allowed to change the number of boundary points in ndown? If not, I have use a spec_bdy_width of 10 points even in the nested domain or I can change the width in some way? Besides, are 3h output from the coarse grid run enough to provide good boundary specifications?

Thanks in advance for your support.


  • WRF_12km.png
    435.4 KB · Views: 474
  • WRF_36km.png
    346.9 KB · Views: 474
  • namelist.input_NDOWN.txt
    6 KB · Views: 29
  • namelist.input_WRF_12km.txt
    5.7 KB · Views: 36
  • namelist.input_WRF_36km.txt
    5.9 KB · Views: 30
To answer your question about the boundary settings, yes you can change spec_bdy_width and relax_zone when you run down. A larger number of spec_bdy_width allows more smoother transition between the forcing and the model results. I won't expect that will dramatically change the results.
Regarding the update of the lateral boundary, in general, are frequent update will yield better results. In your case, 3-hour is fine, but you probably can try 1-hour and see whether the result can be improved.
All the physics options in your 12km and 36km run are the same except radt. Can you run the two cases with the same radt and compare the results? I am curious how it will affect the result. However, I am thinking the large difference could possibly be attributed to the difference in topography, which often has more significant impacts.
Hi Ming,
thanks for cross checking namelists and confirming that I can change the boundary size in ndown.
By the way, the wrong precipitation was due to an issue in the script I used to produce the maps; now all works fine.

Thanks again for your support,