Re: [ferret_users] regridding data between curvilinear grids

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

Hi Hein,
The CURV_TO_RECT regridding function will let you regrid both fields to 
a common recilinear grid so you can compare them.

I also wonder, since you know the projection, if you couldn't perhaps 
"reverse" the projection analytically?  Perhaps take a look at the 
script mp_lambert.jnl which can be used to set up the Lambert projection 
projection from a rectangular grid.  Just talking off the top of my head 
here, I have not really thought this through.

Ansley

Hein Zelle wrote:

> Dear Ferret users,
>
> I am currently working with output from the WRF atmosphere model. The
> model uses a Lambert projection for its computatational grid. This
> means that the output fields do not have a simple longitude and
> latitude axis, but instead two Ni*Nj fields are provided with longitude
> and latitude values. Plotting the fields is no problem with the
> curvilinear versions of shade, fill and vector:
>
> 	    fill u[k=1,l=1], xlong, xlat
>
> works fine, except perhaps that I am now plotting a lambert projection
> on a rectangular lon,lat grid, which looks funny but should be
> correct.
>
> The problem appears when I want to compare two model runs with a
> different grid: one is larger than the other. Ferret doesn't know how
> to regrid from the large grid to the small grid, because the axes are
> defined in i and j coordinates instead of lon and lat. Is there a way
> to have ferret regrid/interpolate data from one curvilinear grid to
> another using the longitude and latitude fields provided with each
> grid?
>
> I can do this in matlab relatively easy, but it would be nice if my
> analysis and plot scripts could remain ferret-only.
>
> Thanks!
> 	Hein Zelle
>
> --
>
> Unix is user friendly. It's just very particular about who 
> it's friends are.
>
> Hein Zelle                     hein@icce.rug.nl
> 	                        http://www.icce.rug.nl/~hein
>
>