Esri Mapping Center - Ask a Cartographer

Ferro meridian

November 10 2011 | 7 comments
Categories: ArcGIS Methods, Map Data

For a Ferro equidistant conic projection ArcGIS 10 still expect that in "new" projection files the central meridian is specified in Greenwich degrees rather than Ferro degrees as it was the case for the "old" projection files.

Why?

This is contrary to current rules.

For reference check http://spatialreference.org/ref/epsg/31251/
PARAMETER["central_meridian",28] is universally expressed in Ferro degrees and not Greenwhich degrees

Mapping Center Answer:

It has to do with what prime meridian is being used. Ferro (El Hierro) is at 17deg40min West. Austria often uses it rather than Greenwich as the prime meridian. When defining a projected coordinate system, angular projection parameters like the central meridian are given relative to the connected GCS's prime meridian (thus 28E of Ferro rather than 10deg20min East of Greenwich) AND in the unit of the GCS. In this case, MGI (Ferro) uses degrees.

Because you refer to the equidistant conic projection, I think you might be defining a custom coordinate system. If so, and the definition uses MGI (Ferro), the central meridian must be relative to Ferro, not to Greenwich. I wonder if you tried to use MGI as the GCS. It uses Greenwich.

Note that we have not changed how the units of the angular/linear projection parameters are defined.

"central meridian must be relative to Ferro, not t posted by Maurizio de Felice on Nov 11 2011 11:37AM

EXACTLY
Why then ArcMap 10 expects the central meridian relative to Greenwich and nor Ferro?

example posted by Maurizio de Felice on Nov 11 2011 12:04PM

for equidistant conic with Ferro central meridian 44.5 ArcGIS 10 expects PARAMETER["Central_Meridian",26.83333333333333]

PROJCS["Potsdam-Ferro",GEOGCS["Potsdam-Ferro",DATUM["D_Potsdam_1983",SPHEROID["Bessel_1841",6377397.155,299.1528128]],PRIMEM["Ferro",-17.66666666666667],UNIT["Degree",0.0174532925199433]],PROJECTION["Equidistant_Conic"],PARAMETER["False_Easting",0.0],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",26.83333333333333],PARAMETER["Standard_Parallel_1",52.0],PARAMETER["Standard_Parallel_2",49.5],PARAMETER["Latitude_Of_Origin",0.0],UNIT["Meter",1.0]]

Ferro versus Greenwich posted by Melita Kennedy on Nov 11 2011 12:39PM

You must set a datum transformation if you are trying to project data on-the-fly in ArcMap from a Greenwich-based GCS to a Ferro-based GCS. Because you're using a custom GCS, you'll have to either define a custom transformation in ArcMap using the geographic transformation dialog, or use the Create Custom Geographic Transformation tool.

Where are you getting 44.5 for the central meridian? With Ferro as the prime meridian, 44.5 converts to 26.833333333333334 relative to Greenwich and located in Ukraine, not in Austria. A central meridian of 26.83333333333 in a Ferro-based GCS converts to 9deg 10min East of Greenwich which is at the western edge of Austria. Is that what you want to use?

Where 44.5 came from posted by Melita Kennedy on Nov 11 2011 12:44PM

Is it a mistake in a conversion? Because Ferro is at 17deg 40min West, it must be added to any Greenwich-based longitude. If we use 13 East of Greenwich as the center (roughly!) of Austria, the Ferro-based longitude is 30deg 40min East.

Ferro versus Greenwich posted by Maurizio de Felice on Nov 11 2011 1:45PM

My question from beginning: why (within a Ferro system) does ArcGIS expect the central meridian to be specified in Greenwich degrees rather than Ferro degrees? This is unfortunately what ArcGIS expects and everyone seems to agree that this is incorrect.

Conclusion: ArcGIS is not handling correctly projection files.
Possible reason: ArcGIS still expects central meridian to be expressed in Greenwich degrees as it was the case in "old" projection files.

Where 44.5 came from posted by Maurizio de Felice on Nov 11 2011 1:55PM

I had used a standard projection system ( http://spatialreference.org/ref/epsg/31251/ )to make the point.

In my particular case I am not dealing with Austrian maps but with Russian maps in equidistant conic projection compiled in the 18th and 19th century. The central meridian value of 44.5 Ferro degrees applies to one of such maps.

Ferro vs Greenwich posted by Melita Kennedy on Nov 11 2011 3:01PM

Maurizio,

Thank you for correcting my assumption that the data was in Austria.

ArcGIS/ArcMap does correctly handle non-Greenwich-based GCS. I just tried it in ArcGIS 10. The PCS central meridian value has to be relative to the prime meridian of its GCS.

The two prj files that I created to check ArcMap are below. HOWEVER, if you have data in MGI (using Greenwich) or WGS84, you must also set a geographic/datum transformation to handle the conversion (including the prime meridian change). I had WGS84-based data, so I had to set the transformation, MGI_to_WGS_1984 or MGI_Ferro_To_WGS_1984 to get the degree values on the status bar to display correctly.

The only problem that I think you might be running into is if you're using rasters and there's some problem when converting into the style that a particular raster format uses to store coordinate system information.

As we have longer and longer comments, would you mind switching to email. We can post a summary here. Please email me at mkennedy at esri dot com. Thanks!

PROJCS["Austria_equi_conic_ferro",GEOGCS["GCS_MGI_Ferro",DATUM["D_MGI",SPHEROID["Bessel_1841",6377397.155,299.1528128]],PRIMEM["Ferro",-17.66666666666667],UNIT["Degree",0.0174532925199433]],PROJECTION["Equidistant_Conic"],PARAMETER["False_Easting",0.0],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",28.66666666666667],PARAMETER["Standard_Parallel_1",43.0],PARAMETER["Standard_Parallel_2",62.0],PARAMETER["Latitude_Of_Origin",30.0],UNIT["Meter",1.0]]

PROJCS["Austria_equi_conic_greenwich",GEOGCS["GCS_MGI",DATUM["D_MGI",SPHEROID["Bessel_1841",6377397.155,299.1528128]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Equidistant_Conic"],PARAMETER["False_Easting",0.0],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",14.0],PARAMETER["Standard_Parallel_1",43.0],PARAMETER["Standard_Parallel_2",62.0],PARAMETER["Latitude_Of_Origin",30.0],UNIT["Meter",1.0]]

If you would like to post a comment, please login.