[R-sig-Geo] projecting Lat and Long to 2D

Harry Kim infohunter at gmail.com
Wed May 14 08:43:16 CEST 2008


Dear R-sig-Geo listers,

     These are very newbie like questions but if i could get any kind of
help, i would very much appreciate it.

     I am trying to project points encoded in latitude and longitude to
hopefully Cartesian coordinates with a reasonable unit of distance
(meters or kilometers.) When i encountered this problem previously,
i've used PBSmapping package and convUL() to convert the points to
UTM coordinates.
     However, the points that i wish to project now are all over eurasia
and Africa. I feel that if i just naively convert all the points to
UTM, then i will run into a trouble as the points have different UTM
zones. Is this correct? If i have some points in zone 10 and zone 14,
and wish to find the distance between them using UTM coordiates and
euclidean norm, would it provide a reasonable approximation of their
distance? If not, is there a projection you would recommend that will
give me a reasonable results?

     I would also like to overlay the points with a world map excluding
America and Antartica. I know that i can plot the map of the world
using maps package and map("world",regions="") but i don't know how i
would exclude the two continents.

     I tried to export the world map into a shapefile then read it in to R
using readShapPoly() in maptools package, then change the coordinate
system into UTM using spTransform, but this ended up giving me a very
weird looking figure.

     In sum, i would like to know if there is a good way to project these
points and overlay a map such that i can use euclidean norm to find
the distance, and the map won't look completely bizzare. Also i would
like to know if i could exclude the two continents mentioned: the
America and the Antarctica.

I thank you very much in advance and i hope you are having a pleasant day.

Harry




More information about the R-sig-Geo mailing list