[OSRS-PROJ] Goode projection discrepancy

[Morten Hulden]

On Wed, 4 Oct 2000, Gerald I. Evenden wrote:

> The seam between Sinusoidal and Mollweide is where the scales are equal
> I suppose the 40:44:11.8 is a little more accurate than the 40:44 that the
> original sources from Snyder but who cares.  The typical scale of Goode
> is so small that 11.8" ain't goin to make much difference!

Unless someone uses a Goode map to program a cruise missile near 40N and
hits the wrong emb^H^H^Htarget ;-) 

> The Y correction is because the two projections do not produce the same
> y value at 40:44.  Equitorial-Polar aspect varies between 
> pseudocylindricals.

So if PH.LIM is changed to 40 44 11.8 then the correction value should
change too? 

> Beware, there are at least two versions of Goode's interuptions.  
> Also, there are gores that duplicate regions.  Almost impossible to
> make an automatic version. Take a look at one of Goode's atlases.
> 
> Interrupted projections are a nightmare to automate!

I get it. I was just thinking on using some of the USGS global/regional
Goode maps as a source for a GRASS5 location in the same projection (which
would require a PROJ goode_i projection, since GRASS relies on PROJ). A
mask (in GRASS) to blacken the duplicated regions would solve part of the
problems, but then it would be a GRASS-specific solution.

Morten Hulden 

----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.