Re: "datatondc" coordinate conversion for lambert conformal

Hi Patrick,

You have your arrays switched. datatondc wants the longitude array
first:

       datatondc(plot,boxlon, boxlat,x_out,y_out)

Then, the "x_out" will be associated with lon, and the "y_out" with lat.

--Mary

>
On Mar 25, 2009, at 3:26 PM, Patrick Reuter wrote:

> Dear NCL users,
>
> Does coordinate conversion not work with lambert conformal projection
> ? I am using the
> datatondc function to transform a boxlat/boxlon array.
>
> ....
>
> plot = gsn_csm_map(xwks,res)
>
> draw(plot)
> frame(xwks)
> x_out = new(dimsizes(boxlat),float)
> y_out = new(dimsizes(boxlat),float)
> datatondc(plot,boxlat, boxlon,x_out,y_out)
> print("data: (" + boxlon + "," + boxlat + ") NDC: (" + x_out + ","
> + y_out +")")
>
> .....
>
> When I use robinson projection of my plot, I get the following
> desired output:
>
>
> (0) data: (-1,45) NDC: (0.801934,0.392597)
> (1) data: (-4,45) NDC: (0.801738,0.374937)
> (2) data: (-4,43) NDC: (0.791349,0.374937)
> (3) data: (-1,43) NDC: (0.791537,0.392597)
> (4) data: (-1,45) NDC: (0.801934,0.392597)
>
> However, with Lambert Conformal projection, it turns into :
>
> (0) data: (-1,45) NDC: (1e+12,1e+12)
> (1) data: (-4,45) NDC: (1e+12,1e+12)
> (2) data: (-4,43) NDC: (1e+12,1e+12)
> (3) data: (-1,43) NDC: (1e+12,1e+12)
> (4) data: (-1,45) NDC: (1e+12,1e+12)
>
>
> As you can see, the values are not converted.
>
> Could you please help me ? Is it not possible ? Or is there another
> way to do this ?
>
> Thanks in advance
>
> Patrick
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Mar 25 2009 - 15:35:04 MDT