[Proj] Convert coordinates from ED50 to Greek Grid

[Mikael Rittri]

Hello Leonidas,
you wrote:

> I need to convert some coordinates from ED50 (EPSG:4230) 
> to Greek Grid (EPSG:2100).I'm using the following parameters
> for cs2cs:
> 
>     cs2cs +init=epsg:4230 +to +init=epsg:2100 
>
> but compared to already valid results, the output greek grid 
> coordinates are wrong. Do I have to use any datum shift. And
> if so, what are the appropriate parameters for that? 


Short answer: 

I suggest using

    cs2cs +init=epsg:4230 +towgs84=-86,-92.2,-127.5 +to +init=epsg:2100

but the datum shift may not be accurate enough (should be better than
nothing, though).


Long answer:

The Greek Grid is based on GGRS87 (Greek Geodetic Reference System 1987), 
and this datum has an appropriate datum shift in the file nad/epsg, at
least since Proj version 4.6.1.  

However, ED50 has a datum shift in the trunk, but I think not in Proj 
version 4.7. So you need to add one.  

At http://www.epsg-registry.org , I have searched with "query by filter", using

Type: Coordinate Transformation
Area: Greece

This gives three published datum shifts from ED50 to WGS 84:

Proj syntax			 Name			   Code	   Accuracy	 Area of Use				Information Source

towgs84=-87,-98,-121	 ED50 to WGS 84 (1)  EPSG:1133	 10 m	 West Europe (including Greece)	U.S. DMA TR8350.2

towgs84=-84,-95,-130 	 ED50 to WGS 84 (8)  EPSG:1140	 44 m	 Greece - onshore and offshore	U.S. DMA TR8350.2

towgs84=-86,-92.2,-127.5 ED50 to WGS 84 (19) EPSG:1440	999 m	 Greece - onshore and offshore	Geodesy Department; Hellenic Petroleum s.a. 

If you would use Proj from the trunk, you will automatically get the 
first datum shift for ED50. 

Otherwise, you must choose manually. In this case, the Accuracy field 
isn't much help when choosing. The confidence of the accuracy is not 
specified by EPSG, but varies with the information source.  When it 
happens to be U.S. DMA TR8350.2, I think it is one sigma, so the error
is worse about one third of the time. If so, then the accuraries of 
the first two datum shifts cannot be usefully compared, even though 
they come from the same information source, since they cover different
areas.  
   On the other hand, since DMA has bothered to give a specific Greek
datum shift, in addition to the general for West Europe, they must 
think it is more accurate in Greece. 
   Personally, I would use the third datum shift, simply because it
comes from a Greek information source. The Accuracy 999 m is used by
EPSG to mean "don't know".  (This datum shift doesn't differ from the 
Greek DMA datum shift by more than about 4 meters horizontally.) 

I don't know why the accuracy for ED50 in Greece should be as bad as
44 m. Maybe it is a pessimistic estimate, or maybe all those islands 
made the ED50 realisation more difficult. 

Best regards,
Mikael Rittri
Carmenta
Sweden
http://www.carmenta.com

-----Original Message-----
From: proj-bounces at lists.maptools.org [mailto:proj-bounces at lists.maptools.org] On Behalf Of leonidas
Sent: den 23 juni 2011 11:43
To: proj at lists.maptools.org
Subject: [Proj] Convert coordinates from ED50 to Greek Grid

I need to convert some coordinates from ED50 (EPSG:4230) to Greek Grid (EPSG:2100).I'm using the following parameters for cs2cs:
cs2cs +init=epsg:4230 +to +init=epsg:2100 but compared to already valid results, the output greek grid coordinates are wrong. Do I have to use any datum shift. And If so, what are the appropriate parameters for that? 


--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Convert-coordinates-from-ED50-to-Greek-Grid-tp6507750p6507750.html
Sent from the PROJ.4 mailing list archive at Nabble.com.
_______________________________________________
Proj mailing list
Proj at lists.maptools.org
http://lists.maptools.org/mailman/listinfo/proj