[Wrf-users] cu_physics for dx = 7km

[Thomas Kloetzke]

Hi Jonas,

indeed, between 3 km (or 4 km) and 10 km it is a bit hard to find out what suits best.

I was running WRF for a simulation of an observed tropical cyclone (for reproduction) and my domains were 10 km and 3.3 km (following the 1:3 grid ratio). For the outer domain, I always used a cumulus scheme, for the inner domain I tried both cumulus scheme on and off and there were only tiny differences regarding track and intensity. But this may differ from case to case.

From my personal point of view, there is no thumb rule saying: You have to use a CU scheme or not for resolutions below 10 km. As I said before, this can be different depending what are you trying to simulate. I would recommend to try both, first leave the CU scheme on for the 4 km domain, second, turn it off and compare the results. From this sort of simulation you will learn much more about the behaviour of the CU scheme and it’ll help you to evaluate which parameterisation might fit better.

Which pressure levels do you use for your wind forecasts? For low-level winds, the choice of the planetary boundary layer scheme will be important as well.

Hope this helps. I am happy to discuss more about that, as I am in the learning process as well and I think many people might be interested in sharing some experiences as well.

Regards,
Thomas

From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On Behalf Of Jonas Kaufmann
Sent: Wednesday, 29 June 2016 6:09 AM
To: wrf-users at ucar.edu
Subject: [Wrf-users] cu_physics for dx = 7km

Hello,

I have set up a domain over Europe with 7km resolution. My main focus is to generate wind forecasts with this domain.

I am comparing outputs between different physics settings, and there is one thing with Cumulus schemes that I do not understand at the moment. I have found presentations online that said that below 4km resolution you probably won't need a cumulus scheme, between 4km and 10km it is not sure and above 10km you need a cumulus scheme. However I have also found that the Grell 3D scheme (cu_physics = 5) is also supported for resolutions below 10km.

Now I am not sure how to proceed with this. Can you give a recommendation when to use the Grell 3D scheme vs. not using a Cumulus scheme for my use case?


Thank you for your help!

Best regards,

Jonas


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20160628/9055147f/attachment-0001.html