Cumulus parameter

Forum dedicated to older versions of EMS package (WRFEMS v3.2, v3.1 or older). Support is user-to-user based, so please help others if you can.
Post Reply
wettersat
Posts: 32
Joined: Fri Aug 21, 2009 9:30 pm

Cumulus parameter

Post by wettersat » Mon Oct 05, 2009 12:02 pm

What cumulus parameter ist best for <4km grid resolution?
I have seen some 4km runs with cumulus parameter = 2 instead of "no cumulus"
In my case I have not any convective rain with "no cumulus".

Gerald

meteoadriatic
Posts: 1515
Joined: Wed Aug 19, 2009 10:05 am

Re: cumulus parameter

Post by meteoadriatic » Mon Oct 05, 2009 1:56 pm

In theory, none of cumulus schemes are designed for such fine girds. There is, however, some difference in real world, where you will very likely get unstable model if you don't use convective scheme, especially around slope terrain and near nested domain borders, which will often manifest as mesoscale convective cyclone forming where in nature will not be such features, not even chances for. You will see lot of precipitation and cyconic air flow around these areas, often with hurricane wind speeds.

Why is that? Because primary purporse of convective scheme is to relieve instability for air column. That way your model keep stability and go on without building these nonexistent mesoscale features with time.

But the question is - how do you know when to use scheme and when not? Answer is not simple and is subject of try and error principe. If you run anything finer than about 6 kilometers, you should try to run this grid without convective scheme. If your model stays stable in synoptic situations that does not likely produce mesoscale cyclonic systems with hurricane winds :) then, you are most probably fine. Check then, also, if your model produces enough quantity of rain...

If your model builds unexpected MCS systems, especially around high slope terrain or near domain borders, or your model does not produce enough rain, you shall turn on convective scheme. It will keep your model stable and it will produce some rain that model dynamics does not see. I run 4km grid WITH Cu=BMJ, because if I don't, my model become crazy over Italian mountains (Apenines). I will see if it will perform better in winter time (probably will), so I could switch off Cu scheme for this grid.

Yes, it is normal that you don't have convective rain in WRF output gribs if you don't use cumulus scheme, because, this variable show only rain that is produces by cumulus scheme :) But, that does not mean that model does not see convective rain, it does, through it's dynamic equations, just it is not distinctive from non-convective precipitation in model output.

Hope this helps.

wettersat
Posts: 32
Joined: Fri Aug 21, 2009 9:30 pm

Re: cumulus parameter

Post by wettersat » Mon Oct 05, 2009 7:12 pm

ok, thanks
I plot this on my meteogramm

*Total Precipitation
'd APCPsfc'

* Snow
'd WEASDsfc'

* Freezing rain
'd APCPsfc * cfrzrsfc'

* Convective Precipitation
'd ACPCPsfc'

* Ice pellets
'd APCPsfc * cicepsfc'

Is freezing rain/ice pellets/convective/snow part of total precipitation or additional?

Gerald

meteoadriatic
Posts: 1515
Joined: Wed Aug 19, 2009 10:05 am

Re: cumulus parameter

Post by meteoadriatic » Tue Oct 06, 2009 1:46 am

It should be under total precipitation. Also, convective precipitation should also be under total precipitation. That is, if say all precipitation is produced by convective scheme, then total precipitation = convective precipitation...

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 3 guests