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

T2 not change with time

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.


New member
I did a seven-day WRF run. Except for the temperature at 2 m, all of the variables in the output file vary with time (T2). T2 remains constant over the whole period. I couldn't find what was causing the issue. I attached the namelist.

Thanks to anybody who has offered suggestions.


  • namelist.input
    5.1 KB · Views: 33
  • plotted_wrfout_d03_2020-01-01_00%3A00%3A000.png
    88.2 KB · Views: 476
Are you running a standard real simulation, or an idealized run? Can you send a plot showing constant temp at a few locations?

I don't think this is causing your issues, but I do see some troublesome settings in your namelist.input file.

mp_physics               = 4,        6,        16,
You should be using the same physics setting across all domains. You'll need to choose one and set all 3 domains to that value.

radt                     = 30,       15,       10,
This should also be the same value for all domains. Set it to radt = 30, 30, 30.

3) Your domains are pretty small. We suggest using domains that are at LEAST 100x100 in order to be able to reasonably simulate any large-scale phenomena.

j_parent_start           = 1,        5,       39,
You have the boundary of domain 02 very close to the edge of domain 01. This is likely going to cause problems. Your domains are so small that if you need to have d02 in the location where you have it, you could simply make d01 larger. You're not saving very much computationally by having a small outer domain. The majority of computation cost is on the higher-resolution domains.

Take a look at this page that discusses some of the best practices for setting up your domain. You can also see this page regarding best practices for namelist.input.
As I mentioned in my above post, I didn't think the microphysics and radt settings were the culprit to the problem. Can you please look back at my previous post and address the other questions and requests I mention? Please send me an updated namelist.input file when you've made the changes. Thanks.
I used larger domains in my last run. The bottom and right boundaries of domains 1 and 2 are, however, still close together and cannot be changed.
I've attached the most recently used WRF namelist.



  • namelist .input
    5.1 KB · Views: 27

1) What is the reason you cannot change the domains? With your current set-up, NCL will not even plot the domains because d01 and d02 are too close together. It's not reasonable to have them so close and you will not obtain reasonable results with that set-up. You can make d01 (much) larger so that d02 will be positioned in a safe place within d01. As I previously mentioned, you're not saving very much computationally by having a small outer domain. The majority of computation cost is on the higher-resolution domains.

2) Please see my post from Oct 4, and please let me know if this is an idealized simulation and which one, if so?

3) Please see my post from Oct 4, and once you modify the domains, if you're still having the T2 issue, please send a plot showing constant temperature (T2) at a few locations.