| Re: map_proj_* help [message #66803 is a reply to message #66802] |
Sun, 07 June 2009 19:38   |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Ken Mankoff writes:
> I began converting from using map_set to map_proj_* about a year ago.
> I've used it successfully in a few situations, but still frequently
> find myself wrestling with it and not getting small details to line
> up. Not that small details always line up with MAP_SET...
Well, I would say about 25% of the time, details not lining
up probably has to do with limitations in IDL's map projection
code. But probably 75% of the time, it has to do with not
completely understanding map projections. I will say this,
they are a LOT harder to understand than they appear to be.
I've been doing map projection stuff of well over a year
now, and just when I begin to think I really understand what
I am doing, I find evidence to the contrary. (Maybe you have
read some of my map articles. :-)
One reason your data may not line up is that the lat/lon
values were produced with one datum, and you are doing
the projection with another datum. (If you don't know what
a datum is, the probability of getting things to line up
goes down by another factor of five.) IDL does not really
do datum shifts, and without them points can definitely be
located in different places. (A "latitude" and a "longitude"
are not the same thing as a "meter" or an "acre" and the point
on the Earth they are meant to pinpoint depends on what datum
you are using, and how that datum is oriented with respect to
the center of the Earth.
But, all that said, that is usually not the problem. More often
one of two things has happened. First, and considerably more
likely in my experience, the published "corner" points are
simply wrong. I don't know if it is because of typos or because
undergraduates who know even less about map projections than you
do are doing the calculations, but whatever it is, those numbers
are frequently figments of someone's imagination.
The second thing that happens frequently is that you (perhaps
naively, if you are not that familiar with map projections)
simply assume that the center of the map projection must be
located half way between the maximum and minimum latitude and
longitude. Nothing is ever that simple on a map projection,
believe me. :-)
So, typically, you have to spend several hours (or not infrequently,
several days) tracking down information and confirming every single
bit of it so you know *exactly* what you are doing. Then--and only
if you have been living your life in an exemplary fashion--things
line up like they are suppose to.
Am I discouraging you? I don't mean to. I'm just trying to
point out that you haven't yet tried hard enough to figure
it all out. There are lots of variables. Code, some of it
written by the good folks at ITTVIS, is written incorrectly
(and not fixed in updates of the software, but that is a topic
better left for another time). There are no good maps (a joke)
of the problem terrain.
Perhaps one good place to start is to read all of the map
articles on my web page. There is a LOT of hard earned
experience in those articles, believe me. If you understand
everything you find there, I would say your are probably at
least half way to your destination. Just grit your teeth and
keep going. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
