[gdal-dev] Spherical Mercator projects and overviews?

[Klokan Petr Přidal]

Stephen,

if you are interested to visually see the extent of tiles of Spherical
Mercator (EPSG:900913) as it is used by Google Maps and similar
services you can play a bit with this mashup:

http://www.maptiler.org/google-maps-coordinates-tile-bounds-projection/

When you click it shows you an info-balloon with the extents of the
clicked tile.

The standard maximal extent is:

WGS84 datum (longitude/latitude):
-180 -85.05112877980659
180 85.0511287798066

Spherical Mercator (meters):
-20037508.342789244 -20037508.342789244
20037508.342789244 20037508.342789244

Best,

Klokan

On Thu, Apr 9, 2009 at 5:26 PM, Christopher Schmidt
<crschmidt at metacarta.com> wrote:
> On Thu, Apr 09, 2009 at 11:25:04AM -0500, Stephen Woodbridge wrote:
>> Frank Warmerdam wrote:
>> >Stephen Woodbridge wrote:
>> >>Hi all,
>> >>
>> >>I have a question regarding imagery in spherical mercator projection
>> >>in geotiff images. Do overviews work correctly?
>> >>
>> >>For example, I use EPSG:4326 and build lots of overviews and I can
>> >>zoom in and out and pan without a problem. Can I do the same thing
>> >>with EPSG:900913 (spherical mercator projection).
>> >>
>> >>I'm working with LandSat imagery over large area and scale ranges.
>> >
>> >Steve,
>> >
>> >I can't see any reason that overviews would not work on spherical
>> >mercator when they work on images in other coordinate systems.  The
>> >overview handling is independent of the georeferencing.
>> >
>>
>> Frank,
>>
>> Thanks this is what I was hoping the answer would be.
>>
>> My fear was that I would run into something like UTM zones and that you
>> can not have one global data set in UTM. So spherical Mercator is
>> projects the world into a single x-y plane similar to the way
>> geographical is treating the global as stretched onto a single x-y
>> plane.  This seems reasonable, I just never thought about it before.
>
> Where 'the whole world' in mercator can't actually be 'the whole world';
> you have to pick a reasonable north/south cutoff since the tangent of 90
> is infinity/undefined.
>
> -- Chris
>
>> Now I get to reproject everything yet again :)
>>
>> Thanks,
>>   -Steve
>> _______________________________________________
>> gdal-dev mailing list
>> gdal-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>
> --
> Christopher Schmidt
> MetaCarta
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
>



-- 
http://blog.klokan.cz/
http://www.maptiler.org/
http://www.oldmapsonline.org/