Vertical grid Spacing

Miscellaneous questions you have about anything related to graphically displaying WRF output

Vertical grid Spacing

Hello,

I am using the wrfout_d03 issue of the model.
I am trying to establish an approximated wind gradient for the ground to around 300m.

In order to do so variables "U" (wind velocity along x-axis) and "V" (wind velocity along y-axis) seem appropriate. The measures are given at latitude and longitude points but also at different altitudes of 7 different layers.
Unfortunately I can't find the height of these layers, therefore I am missing the z-coordinate of the measure.

I went trought the ARWUserGuide but i only found that Vertical grid-spacing can vary with height. Although no actual function is given to compute this height.

Thank you for help,
Manuel Duprey
ManuelD

Posts: 2
Joined: Wed Apr 13, 2011 9:20 am

Re: Vertical grid Spacing

There are two wrf output variables that you need to compute the height: PB and PHB. PB is the base state geopotential and PHB is the perturbation geopotential. I underline these because they are not height. To get height, the calculation is (PB + PHB)/g where g is the acceleration of gravity. If you're using NCL to compute this, there is a built in function to calculate height, located here: http://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_getvar.shtml.

HTH

Dave
dbh409

Posts: 89
Joined: Wed Mar 24, 2010 11:46 am

Re: Vertical grid Spacing

Thank you very much.
Can use PHT instead of (PB +PBH) ?
Does this mean that the first layer of the PHT variable is equal to PB (hence that it explains the extra layer) ?
Last edited by ManuelD on Fri Apr 22, 2011 2:54 am, edited 1 time in total.
ManuelD

Posts: 2
Joined: Wed Apr 13, 2011 9:20 am

Re: Vertical grid Spacing

I'm not sure what PHT is in your wrfout file. That variable name is not listed in the wrf output variables described in the ARW manual.

The reason there is an extra layer is because the vertical levels are staggered. I'm not sure if that answers your question or not. The first height level is in between the surface and the lowest mass point. If you'd like a more detailed understanding of grid staggering, check out http://www.atmos.ucla.edu/~fovell/AOS18 ... es2005.pdf. This is from Robert Fovell's course on numerical modeling. Section 16.2 would be of the most interest to you.

Dave
dbh409

Posts: 89
Joined: Wed Mar 24, 2010 11:46 am

Re: Vertical grid Spacing

Some corrections to this thread IMHO

As far as I can tell there isn't any built-in NCL function to calculate WRF heights (the link above has no such tool).

Furthermore, the variable names stated above are wrong.
Base State Geopotential = PHB (bot PB)
Perturbation Geopotential = PH (not PHB)
farrenthorpe

Posts: 2
Joined: Wed May 02, 2012 4:56 pm

Re: Vertical grid Spacing

Correction to the above correction:

ncl does in fact have a built in function, wrf_user_getvar, that can get the model heights in meters when you pass in "z" as the second argument. In the code WRFUserARW.ncl, z is computed as

z = (PH + PHB)/9.81

then unstaggering in the vertical.
lgstarn

Posts: 3
Joined: Wed Aug 15, 2012 7:43 pm

Who is online

Users browsing this forum: No registered users and 1 guest