[Proj] Need help converting Goode to latlon

[Gerald I. Evenden]

On Wednesday 22 July 2009 2:39:52 am David Turover wrote:
> I have shapefiles made in ArcGIS that use the Interrupted Goode Homolosine
> projection. I am trying to use libproj to reproject them to lat/lon, but
> the resulting shapes do not align correctly to the corresponding features
> in Google Earth. Everything is pulled to the west by a small amount near
> the equator and a larger amount further away from the equator.
>
> Example: http://segfault.cs.sonoma.edu/~dturover/gisbug/projectionError.jpg

The projection in the above jpeg looks like Near-sided perspective.  At any 
rate, it certainly is not Goode's pseudocylindrical Homolosine.

> The projection in ESRI format is Goode%20Homolosine%206378137%20sphere.prj
> in that directory. The proj arguments that I am using to define Goode are:
>
> +proj=goode +lon_0=0 +x_0=0 +y_0=0 +a=6378137 +b=6378137 +units=m +no_defs

+b= is not necessary as Goode is only available as a spherical earth and +b is 
the minor axis of an ellipsoidal earth.

> What sort of things might be going wrong here? How might I be able to
> adjust whatever algorithm is being run to attempt to fix it? I am too new
> to projections to know what is possible in these areas.
>
> I see that the .prj file has a "<custom>" datum, but I do not know what
> this implies or how to make sure proj is using it. The custom datum is:
> DATUM[
>         "<custom>",
>         SPHEROID[
>                 "WGS_1984_Major_Auxiliary_Sphere",
>                 6378137.0,0.0
>         ]
> ]

At this point, and especially at global scales, I do not think you need to 
worry about datums.

> Also, proj -lP lists PCyl and Sph as different projection classes under
> +proj=goode but I cannot find anything in the proj4 documentation on
> how to specify a projection class.

PCyl : pseudocylindrical

Sph: spherical earth

Just FYI info and not something you specify in selecting a projection.

> The .prj file says that I should be 
> using a spherical projection. If proj defaults to a pseudocylindrical
> algorithm for Goode, would it produce different results? If so, how
> could I force it to run a spherical algorithm?
>
> Thanks in advance for any help.

Help is very difficult because of the confusion caused by the jpeg figure.  
Certainly a Goode projection is not going to come anyway come near matching 
your referenced jpeg figure except in a very limited form near the center of 
the projection.  See:

http://www.csiss.org/map-projections/Pseudocylindrical/Goode_Homolosine_a.pdf

-- 
The whole religious complexion of the modern world is due
to the absence from Jerusalem of a lunatic asylum.
-- Havelock Ellis (1859-1939) British psychologist