[R-sig-Geo] Good projection for N/S America?

[Tim Keitt]

On 4/9/07, Roger Bivand <Roger.Bivand at nhh.no> wrote:
> On Mon, 9 Apr 2007 White.Denis at epamail.epa.gov wrote:
>
> > For preview graphics and for large areas such as continents, large
> > countries, hemispheres, or the whole earth, spherical projections are
> > often adequate.  I can provide some of the ones I have used.  For
> > detailed work at sites and small areas, ellipsoidal projections such as
> > UTM are usually used, and then the coding gets more complicated with
> > choices of datums and so forth.
> >
>
> The attached script shows how to do the interrupted sinusoidal projection
> using spTransform in rgdal, for the whemi.lin data posted with the
> free-standing functions by Denis White a couple of days ago. Once the
> lines are converted into SpatialLines objects, the rest is robust and
> simple, as is the use of gridlines() in sp. The one catch is calculating
> the offset, here in an x_0= offset along the Equator in metres between the
> two central longitude values. The output is attached as a PNG image. The
> point about the sp objects is that they contain enough metadata (here a
> PROJ.4 projection description) to let them be moved to other R packages or
> external software.
>
> The half-dozen basic projections are easy to specify in PROJ.4, for
> example from the geotiff list:
>
> http://www.remotesensing.org/geotiff/proj_list/
>
> which is what I used here. The other projections mentioned are:
>
> Lambert Cylindrical Equal Area "+proj=cea +lon_0=-80"
>
> Lambert Azimuthal Equal Area "+proj=laea +lat_0=0 +lon_0=-80"
>
> while the Northern hemisphere sinusoidal is:
>
> Sinusoidal "+proj=sinu +lon_0=-100"
>
> So I'd argue that PROJ.4 projection descriptions are not difficult to use,
> and with sp objects, do stay stuck to the data (has anyone else ever
> forgotten what projection was used when revisiting data, not just me?).

Yes, that's brilliant (keeping the proj4string). My point was not that
proj is hard to use -- its simple -- rather that depending on proj
means more build time dependencies. On the other hand, if its not
broke....

THK

-- 
Timothy H. Keitt, University of Texas at Austin
Contact info and schedule at http://www.keittlab.org/tkeitt/
Reprints at http://www.keittlab.org/tkeitt/papers/
ODF attachment? See http://www.openoffice.org/