Re: Masking fields by topography

[D Shankar <shankar@xxxxxxxxxxxxx>]

The following script does the job for the Levitus climatology.  It works
for both horizontal and vertical sections.
______________________________________________________________________
! NOTE: for a contour plot in the XY plane BELOW the surface, you must supply
! the first argument to this GO script.
! usage: GO fland_levitus k=N  (where N is the required level)
! or     GO fland_levitus z=Z  (where Z is the required depth)
!        GO fland_levitus      (in other cases)
\cancel mode verify
! fland_levitus.jnl - a land mask for the LEVITUS dataset (overlay or "basemap")
! 1/97

! Description: Plot/overlay an an area-filled continental map using LEVITUS mask

! usage:               arg1      arg2        arg3         arg4      arg5
!  GO fland_levitus [z=lo:hi] [palette] [overlay_style] [x=lo:hi] [y=lo:hi]

! arg1  - depth limits - specify as for any FERRET depth range
!                        use either k=*:* or z=*:*
! arg2  - palette - may be any palette on your system.  Normally solid colors
!	           like black, gray,red, etc. (see Fpalette '*') default=gray
! arg3  - overlay_style - "basemap", "transparent", or "overlay" (default)
! arg4  - longitude limits - specify as for any FERRET longitude range
! arg5  - latitude  limits - specify as for any FERRET latitude  range
 
! examples:
!	  GO fland_levitus 	          ! default - color gray, overlay
!       GO fland_levitus k=1            ! MASK for surface
!	  GO fland_levitus k=1 red        ! palette red.spk, overlay
!	  GO fland_levitus z=0:1000 red basemap ! color red, basemap
!	  GO fland_levitus z=1000 gray transparent ! color gray, transparent mask
!       GO fland k=1 gray overlay x=30e:100w y=20s:20n

! Note : The transparent option creates a see-through overlay land mask

! check the overlay/basemap/transparent option
query/ignore $3%overlay|overlay|basemap|transparent<3rd argument of GO fland_levitus must be "overlay", "basemap", or "transparent"%

! save data set
set data/save

! save region
define region/default levitus

! set data for MASK
use levitus_mask.nc

set region/l=1

shade/nolab/nokey/levels="(-0.5,0.5,1.0)"$1" |*>/*"/pal=$2"gray"$3"/overlay|basemap> |*>/overlay"$4" |*>/*"$5" |*>/*" $3"MASK|transparent>if mod(i+j,2) eq 0 then MASK|*>MASK"

! cancel MASK data set
cancel data levitus_mask.nc

! restore region
set region levitus

! restore original data set
set data/restore

set mode/last verify
________________________________________________________________________

D. Shankar                           Ph. (O): [91](832) 22 6253/4/5/6
Physical Oceanography Division,                         * 4400/4224
National Institute of Oceanography       (R): [91](832) 23 8208    
Dona Paula,                          Fax    : [91](832) 22 3340/1360/9102
Goa 403 004,                         email  : shankar@csnio.ren.nic.in
India.                                        shankar@darya.nio.org

On Tue, 24 Nov 1998, David Oxilia wrote:

> As we have had mail problems due to severe rain I am forwarding
> this message of yesterday. AGU is approaching; any help would
> be appreciated!
> 
> thanks, David.
> 
> David Oxilia wrote:
> > 
> > Dear Ferreters,
> > 
> > One of the things I've always had trouble doing with ferret is
> > masking a field to be plotted as a section (ie, lat or lon vs
> > depth) by bottom topography. Say I want to plot a section of
> > sigma-0. I read in my data and generate the plot. However, I
> > then want to overlay a plot of the bathymetry.
> > 
> > How does one do this?
> > 
> > It is straight forward to simply draw a line plot of depth vs
> > latitude say over the sigma-0 field. There is a problem though;
> > that is, the line plot simply places a line over the field
> > without covering up (in a uniform color) the underlying colored
> > field. In essence if there was a means to do "filled curves"
> > this would solve the problem. One would say something like:
> > 
> > yes? plot/vs/over/yfilled/nolab/line=1 lat,zb
> > 
> > where the yfilled would instruct to fill the curve down to the
> > "y-axis" (defined by the grid) in the color specified by "line".
> > Alternatively, if one could mask the field by the topography
> > with something like:
> > 
> > yes? let masked_field = if (field_depth lt zb) then field else missing
> > 
> > This would allow one to contour the field with the lower bound
> > masked off by the topographic height. One could then plot the
> > topography line to make it stand out. This approach limits the
> > topography "color" to be white though which is not entirely
> > satisfactory.
> > 
> > Any suggestions?
> > 
> > David.
> > 
> > *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> > David Oxilia                    Tel: (541) 737-2368
> > COAS:Oregon State University    Fax: (541) 737-2064
> > Ocean Admin. Bldg. 104          WWW: http://www.oce.orst.edu/po
> > Corvallis, OR 97331-5503        Email: oxilia@kundun.oce.orst.edu
> > *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
>