| Satellite projection contour plots [message #75592] |
Wed, 30 March 2011 09:10  |
![Matt[3] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Matt[3] is currently offline")
Matt[3]
Messages: 23
Registered: April 2011
|
Junior Member |
|
|
Hi All,
I'm trying to create a few contour plots on a satellite map
projection. Mostly they look great, but from some viewing angles (the
lon/lat that the 'satellite' is over), there seems to be some strange
artifacts in my eps and png output. For example, I've put a plot up
here (http://dl.dropbox.com/u/13779929/h00200.png) which looks good in
the tropics, but goes a little crazy at the high-latitudes.
I've checked the output with antialiasing turned off (or on, whichever
it is), but same result. There are hints of similar behavior for
other projections, but not as severe as with the satellite. Other
color tables and number of contour levels etc. seem to lead to the
same result.
The map has been created using Dave Fanning's routines and the old
fashioned map_set (see below). If there's something I'm missing, or
if there's a better way to do this with the new graphics, I'd love to
hear it!
Cheers,
Matt
ps_start, filename=outeps, /nomatch, ysize=ysize, xsize=xsize,
xoffset=0., yoffset=0.
!p.multi=[0, 2, 1]
pos=[0.05, 0.1, 0.8, 0.9]
cgloadct, 0
MAP_SET, angle_lat, angle_lon, 0., position=pos, /noborder, /
satellite, SAT_P=[3., 0., 0.]
cgLoadCT, 25, /brewer, nColors=CtN, /reverse
cgcontour, IM_append, lon_append, lat, levels=indgen(ctN)+ctB, /
fill, /overplot
;IM_append is the image, scaled to be a floating point number
between 0. and 255. (CtN=256, CtB=0).
cgloadct, 0
map_continents, thick=2
;Color bar plotted here
ps_end, /png
|
|
|
|
| Re: Satellite projection contour plots [message #75693 is a reply to message #75592] |
Mon, 04 April 2011 17:04  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Matt writes:
> I only just got around to trying this, but it seems like you're right,
> using /cell_fill is the answer here! Everything looks OK at the date
> line for me too (I've added an extra longitude to my data so that it
> 'wraps around', but I'm assuming you did this in your plot?).
No, the "wrapping around," which I tried first turned
out to be unnecessary when the data is gridded correctly.
As I say, I'm working on an article to explain all this.
(The darn book needs updating, too. Sigh...)
The gridding also apparently has a lot to do with the
"scratch marks" in the PostScript file. I have some
extremely interesting figures I am trying to understand
right now. Big basketball game tonight, but maybe I'll
finish this in the morning. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
