Re: Sigma To z coordinate

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi Masoud,
I would check that you have the definition of variable depth
correct.  The demo file sigma_coordinate_demo.jnl is somewhat
complex because it is first creates a 4-dimensional field to use in
the regridding.  All you need is the input Z layers, the bathymetry,
and the desired output depth axis.

Does your variable dep contain the depth of the sigma levels?
If so then you can use the function like this:

 use test.cdf      ! has s and dep on 5 levels 
 define axis/z=0:100:5/unit=meters/depth zdepth
  let salt=ZAXREPLACE(s,dep,z[gz=zdepth])

If dep is the thickenss of each layer, then you would need to do something
like this

  let depths = LAYER_THICKNESS[k=@rsum]-layer_thickness/2
  let salt=ZAXREPLACE(s,depths,z[gz=zdepth])

If dep is the bottom depth only,  and you are creating
depths for each sigma level from it, then I am not sure what
you are doing with the command

  let h0=dep[k=s]

You could set up a set of depths ranging down to the value
of variable dep with commands like this:

  let nlev = `s, return=ksize`
  let depths = dep * (k[gz=s] / ndep)

  !! Try listing depths at a single value of X, Y, T
  !  list/x=0/y=180/L=1 depths 
  let salt=ZAXREPLACE(s,depths,z[gz=zdepth])

I hope these ideas will get you started.

Ansley

Masoud Sadrinasab wrote:

> Dear Ferreters,
>
> I am trying to convert results from my model (sigma coord) to  z 
> coordinates.
>
> When I tried use ZAXREPLACE, some problems  occur. I  followed the 
> sigma_coordinate_demo.jnl as example. The following is  my own journal:
>
> use test.cdf                                         !(5 sigma levels)
> let flow=s ! flow takes place of salinity - s
> let 
> h0=dep[k=s]                                                            
> !dep -> my bathmetry
> let h_nor=h0/h0[k=1:5@sum]
> let h=h_nor*(-1*dep)
> let depth=h[k=@sum]-h/2
> define axis/z=0:100:5/unit=meters/depth zdepth
> let salt=ZAXREPLACE(s,depth,z[gz=zdepth])
> shade salt[i=131,l=1]
>
> Any comment would be appreciated.
>
> Masoud Sadrinasab
> School of Earth Sciences
> Faculty of Science and Engineering
> Flinders University of South Australia
> South Australia 5042
> Phone: +61 8 82013055
> Fax: +61 8 82012676
> Mobile: +61 404 125 524
> E-mail: Masoud.Sadrinasab@flinders.edu.au