Using adaptive_time_step in 24hr fitch turbine runs.

[jk2011]

Hi! I'm running WRF ARM v4.4.1 using the fitch parameterization with a 3-way nested domain at 1km. I have been able to get decent results but haven't had the time to play with all the time functions that I've heard can alter the output quite a bit. Would someone mind sharing recommended time interval settings for 1km turbine runs? I see that adaptive_time_step can create unrealistic results and numerical noise when interesting weather features roll through, but can speed up run time and potentially receive better solutions in some cases. Is there a middle-ground that can be achieved by toning down the effects adaptive time step? or is sticking with a fixed time step recommended.

 

[kwerner]

Hi,
Apologies for the delay in response. We have been out for the holidays and are trying to get caught up. This scheme was provided to our group from an outside collaborator, so we don't have a lot of experience with it. If you haven't already, I would recommend taking a look at the literature for the scheme to see if there is any useful information.