[Wrf-users] default sigma level settings [SEC=UNCLASSIFIED]
Eric Altshuler
ela at cola.iges.org
Mon Jan 12 16:11:14 MST 2009
Perhaps the existing terminology of 'staggered' and 'unstaggered' doesn't work very well for the vertical dimension. Also, I have seen the terms 'half levels' and 'full levels' have different meanings in different situations, causing confusion. To me, it seems the most sensible usage of the terminology is as follows:
'Staggered' == 'eta LAYER INTERFACE values' == 'eta values on full (w) levels' (ZNW)
'Unstaggered' == 'eta LAYER MIDPOINT values' == 'eta values on half (mass) levels' (ZNU)
Although some sources define 'full levels' to be the layer midpoint (mass) levels, with 'half levels' being the layer interface (w) levels. Clearly the terms 'half' and 'full' levels are not used consistently by everyone. Also, modelers often speak of levels and layers interchangeably: "The model has 40 vertical levels" or "The model has 40 vertical layers". When we speak of layers, the meaning is clear: there would be 40 layers and 41 levels (layer interfaces). However, saying the model has 40 levels is not so clear: it could mean 40 layer midpoint levels (i.e. 40 layers) or 40 layer interface levels (i.e. 39 layers). It is probably best to describe the vertical discretization of a model in terms of layers rather than levels.
Eric
----- Original Message -----
From: "Chris Thomas" <Chris.Thomas at ga.gov.au>
To: wrf-users at ucar.edu
Sent: Monday, January 12, 2009 5:09:56 PM GMT -05:00 US/Canada Eastern
Subject: RE: [Wrf-users] default sigma level settings [SEC=UNCLASSIFIED]
I guess it's just a matter of terminology. When I do ncdump -h wrfinput_d01 I get (in part)
float ZNU(Time, bottom_top) ;
ZNU:FieldType = 104 ;
ZNU:MemoryOrder = "Z " ;
ZNU:description = "eta values on half (mass) levels" ;
ZNU:units = "" ;
ZNU:stagger = "" ;
float ZNW(Time, bottom_top_stag) ;
ZNW:FieldType = 104 ;
ZNW:MemoryOrder = "Z " ;
ZNW:description = "eta values on full (w) levels" ;
ZNW:units = "" ;
ZNW:stagger = "Z" ;
which I took to mean that the term "staggered" (when applied to the vertical dimension) refers to the eta levels given by ZNW (the full levels), and "unstaggered" to ZNU. Am I missing something here?
Regards
Chris
-----Original Message-----
From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On Behalf Of Mikhail Titov
Sent: Tuesday, 13 January 2009 8:17 AM
To: Thomas Chris; wrf-users at ucar.edu
Subject: RE: [Wrf-users] default sigma level settings [SEC=UNCLASSIFIED]
Hello,
Would like to join this discussion. I agree with Chris about levelling and checking levels but can not agree about staggered and unstaggered levels.
>From ARW WRF basic description (page 17, Figure 3.2) and my own
>experience the command:
ncdump -v ZNW wrfinput_d01
Will give unstaggered levels (basic levels for W, P, PB ...), while the command:
ncdump -v ZNU wrfinput_d01
Will give you staggered levels (intermediate levels for U,V, ...).
It is easy to check using in WRF namelist.input eta_levels and apply "ncdump -v ZNW wrfinput_d01": eta_levels (basic, unstaggered) will be identical in namlist.input and for ZNW=.
Regards,
Michael
--------------------------------------
Dr. Mikhail Titov
Connell Wagner Ltd.
VP of the NZ Met.Society
University of Canterbury
Christchurch, New Zealand
E-mail: Mikhail.Titov at canterbury.ac.nz
Phone: (03) 364 2987 Ext: 3088
Mobile: (021) 106 5563
--------------------------------------
-----Original Message-----
From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On Behalf Of Chris.Thomas at ga.gov.au
Sent: Monday, 12 January 2009 2:53 p.m.
To: wrf-users at ucar.edu
Subject: FW: [Wrf-users] default sigma level settings [SEC=UNCLASSIFIED]
As an addendum to my previous post, if you are interested in how the eta levels are defined for idealized cases (as was Leo, the author of the original post) then look at the corresponding module_initialize_xxxx.F file. For example look at module_initialize_quarter_ss.F and look for where grid%znw and grid%znu are assigned.
You can also read off the eta levels straight from the wrfinput_d01 file that is created by ideal.exe (or real.exe as the case may be). For example
ncdump -v ZNW wrfinput_d01
will give you the staggered eta levels and
ncdump -v ZNU wrfinput_d01
Will give you the unstaggered levels.
Regards
Chris
-----Original Message-----
From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On Behalf Of Thomas Chris
Sent: Monday, 12 January 2009 9:04 AM
To: wrf-users at ucar.edu
Subject: FW: [Wrf-users] default sigma level settings [SEC=UNCLASSIFIED]
I think that the relevant code can be found in module_initialize_real.F (assuming you are doing a real case). Look for the subroutine compute_eta.
Regards
Chris
-----------------------------------------------------------
Christopher Thomas
Computational Modeller
Risk and Impact Analysis Group
Geoscience Australia
Ph: +61 2 6249 9757
Email: chris.thomas at ga.gov.au
-----Original Message-----
From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On Behalf Of Eric Altshuler
Sent: Saturday, 10 January 2009 5:08 PM
To: wrf-users
Subject: Re: [Wrf-users] default sigma level settings
Hello Jorge,
I think the point of the original question was this: What are the sigma levels chosen by default when eta_levels is NOT specified? Suppose I set e_vert = 41 but don't specify eta_levels. What is the resulting distribution of sigma level values? There must be some formula in the code that calculates the 41 sigma values in this case, but it isn't clear how the default levels are calculated for a given value of e_vert.
Best regards,
Eric L. Altshuler
Assistant Research Scientist
Center for Ocean-Land-Atmosphere Studies
4041 Powder Mill Road, Suite 302
Calverton, MD 20705-3106
USA
E-mail: ela at cola.iges.org
Phone: (301) 902-1257
Fax: (301) 595-9793
----- Original Message -----
From: "Jorge Alejandro Arevalo Borquez" <jaareval at gmail.com>
To: "anis 980" <anis_980 at sina.com>
Cc: wrf-users at ucar.edu
Sent: Thursday, January 8, 2009 12:37:39 PM GMT -05:00 US/Canada Eastern
Subject: Re: [Wrf-users] default sigma level settings
Hi,
In WRF you can give explicitly the values for vertical levels, to do this you need add eta_levels values like this:
&domains
eta_levels = 1.0000, 0.9960, 0.9940, 0.9920, 0.9900,
0.9880, 0.9850, 0.9820, 0.9780, 0.9740, 0.97, 0.965,
0.96, 0.949, 0.937, 0.924, 0.909,
0.892, 0.873, 0.851, 0.826, 0.798,
0.768, 0.736, 0.702, 0.666, 0.629,
0.5915, 0.5536, 0.5153, 0.4773, 0.44,
0.404, 0.3695, 0.3375, 0.3085, 0.2845,
0.2645, 0.2465, 0.2305, 0.2165, 0.2035,
0.1915, 0.1792, 0.1667, 0.1539, 0.1407,
0.1272, 0.1134, 0.0995, 0.0855, 0.0713,
0.0571, 0.0520, 0.0480, 0.0429, 0.038,
0.032, 0.0287, 0.022, 0.018, 0.0145, 0.007, 0.000,
time_step = 200,
...
...
Best regards
2008/12/25 <anis_980 at sina.com>
> Hello,
> I am trying to run WRF ideal case em_quarter_ss, and I set
> "s_vert=1, e_vert=41 " in namelist.input file. But I want to know the
> sigma value of each level, does anyone know how does WRF model
> generate sigma values according the total number of vertical levels?
> What is the default sigma value according vertical level settings?
> Best, Leo
>
> _______________________________________________
> Wrf-users mailing list
> Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
>
>
--
Atentamente
Jorge Arévalo Bórquez
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu http://mailman.ucar.edu/mailman/listinfo/wrf-users
_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu
http://mailman.ucar.edu/mailman/listinfo/wrf-users
More information about the Wrf-users
mailing list