Tracer non-linear mass

[Gorrono]

Hi,

I have recently generated a set of simulations with WRF in LES mode including a tracer for methane.
I have followed the detailed documentation from other colleagues e.g. Brian Blaylock, UofU

I successfully run the simulations and got results such as mass growth and concentration maps:

But I calculated the methane flux rates and were not exactly matching what I was expecting.
After a while, I discovered that when I zoom in the initial 5 minutes of the simulation are non linear (red shows ideal vs blue actual trend).


There is no apparent explanation for this non-linear mass growth (in those initial minutes the plume is inside the domain) and I thought about inconsistencies in the temporal interpolation or similar numerical issues.
I attach an example of the namelist file in case you have some ideas and can help me with this issue.

Thanks in advance
Javier Gorroño

 

[Gorrono]

Hi, just a more detailed explanation. I have made several tests including different temporal/spatial resolutions and configurations. The non-linearity remains.
Interestingly, I had access to simulations from other colleagues using WRFv3 and this effect does not exist (i.e. the mass grows perfectly linear).
The major change in namelist.input v4 wrt to v3 is that you must specify the case:
&ideal
ideal_case = 9
/
I am not aware of the details related to code changes but at this point I wanted to raise as a hypothesis.
Hope you can help me with this. Thanks