Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!uwm.edu!cs.utexas.edu!ut-emx!walt.cc.utexas.edu!tim
From: tim@walt.cc.utexas.edu (Tim Eakin)
Newsgroups: comp.sys.mac.misc
Subject: Re: Software to calculate areas needed...
Message-ID: <34721@ut-emx.UUCP>
Date: 31 Jul 90 05:22:40 GMT
References: <23374@dartvax.Dartmouth.EDU>
Sender: news@ut-emx.UUCP
Reply-To: tim@walt.cc.utexas.edu (Tim Eakin)
Organization: The University of Texas at Austin, Austin, Texas
Lines: 17

In article <23374@dartvax.Dartmouth.EDU> thomas@eleazar.dartmouth.edu (Thomas Summerall) writes:

>A geography professor here has the following problem:  If he has a map of
>a country, with each state as a separate object, and he overlays a map of
>rainfall, with each rainfall area as a separate object, is there software
>whic will tell him the area of the intersection of a state object with a
>rainfall object?
 
One problem may be that there are so many different types of projections.
Each would require its own algorithm.  In the usual Mercator conformal
projection you can get an overlap in the northern part of a state and
one in the southern part which are equal in map area but unequal in
actual surface area, and vice versa.  Any results obtained from generic
software that assumes isotropic preservation of area in projections would
not be too meaningful.  All this assumes of course that the professor's
maps are planar.  If he is working with 3-D objects that lie on the surface
of a sphere, then his overlap areas will all be linearly related.