[Proj] Large scale mapping & Google

[Irwin Scollar]

Gerald I. Evenden wrote:

>Another minor point with google, I bet they are 
>*not* using a elliptical projection software.

Indeed they are not, and this is documented on 
support sites for KML and for the GE API's both 
for COM and the ActiveX Plugin. They use Plate 
Carée with sphere A = 6378137.0.

Clifford J Mugnier wrote:

 > It has been observed by Military Geodesists 
that "A major battle will occur at every Grid Zone Boundary,"

In a number of publications and in the help for 
my AirPhoto programme which was written for those 
practicing low-level oblique aerial photography 
of archaeological sites visible as crop or soil 
discolorations to record them on large scale 
(1:1000 - 1:5000) maps for monument protection 
purposes,  I noted this phenomenon with respect 
to the original Gauss-Krueger 3 degree grid used 
on German maps (Deutsche Grundkarte 1:5000), and 
the corollary that all important aerial 
archaeological sites are located at the 
intersection of four map sheets whose boundaries are on that grid.

An even worse example is the airport of the city 
of Basle (Bâle) in Switzerland which is partly in 
France and overlooking Germany with three 
incompatible map projections (Swiss Oblique TM, 
French Lambert Polyconic and German GK 3 degree), 
so converting these on the fly to UTM WGS84 with 
the very decent datum transformation parameters 
available from each country is the only sensible thing to do.

In the former German Democratic Republic, which 
was forced to change from the GK 3 degree Bessel 
Ellipsoid grid to the Soviet 6 degree Krassovsky 
after WW2, a change is now being made to WGS84 
UTM 6 degrees for re-mapping.  In Bavaria, a 
Soldner-Cassini projection and grid centered on 
the main church in Münich is still used in many 
cadastral records, so that there are at least 
three incompatible systems within Germany today that I have encountered.

I think that the most common application of large 
scale and largely unpublished mapping in all 
developed countries is cadastral survey for 
property boundary records many of which are some 
variant of TM.  The cost of converting the 
enormous bulk of data accumulated over centuries 
which this represents is perhaps the main reason 
why few countries are converting to something 
more modern, although pious wishes in this respect are frequently expressed.

Irwin Scollar