Re: [ferret_users] contour/fill and contour/line - different algorithm?

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

 Hi Paul,
The FILL and CONTOUR commands use the same algorithm. What's the difference between your symbols ($my_levels) and ($lv)?  Maybe it's just a typo in what you sent.

   fill/lev=($my_levels)
   contour/over/nolab/lev=($lv)

If you want the same contour lines as colors, try doing it in one command with FILL/LINE.  If that shows any differences, then it may be in the rendering of the colors, or in any case we should hunt further.

Ansley

On 9/29/2010 1:56 PM, Paul Young wrote:
> Hi all,
>
> I've been doing some plots where I do a "fill" with a "contour/line" over the top. This has worked out fine for the most part, but today I've been trying it on some data where I transform the pressure axis to log pressure. In essence, the problem is that the contour lines do not line up with the shading, mostly at higher altitudes. (see attached) The vertical model levels are certainly more spread out there, but I would have guessed that the contour/line would still follow contour/fill.....
>
> Commands follow:
>
> use my_water_vap_climatology.nc                             !Model water vapor output
>
> define axis/depth/from_data/name=lnpz/z ln(z[gz=my_data])    !Put on log pressure axis
> let x_lnp = my_data[gz=lnpz@asn]
>
> let zmin = 250                                            !Set up plot (range and axes etc)
> let zmax = 25
> ppl axlabp -1, 0
> ppl tics 0.125, 0.25, 0, 0
>
> set region/z=`ln(zmin)`:`ln(zmax)`
> fill/lev=($my_levels)/nokey \
>   (x_lnp[l=@max]-x_lnp[l=@min])                              !Contour/fill yearly range
> contour/over/nolab/lev=($lv) \
>   (x_lnp[l=@max]-x_lnp[l=@min])                            !Over plot contour lines - PROBLEM HERE?!
>
> Thanks for any pointers,
>
> Paul