[PROJ] Matching EPSG:3005

[Even Rouault]

Hi Nyall,

> I suspect I'm overlooking something obvious, but I've read and re-read
> this multiple times and can't spot it!
> 
> When attempting the identify the WKT string
> 
> PROJCS["NAD83 / BC Albers", GEOGCS [ "NAD83", DATUM ["North American
> Datum 1983 (EPSG ID 6269)", SPHEROID ["GRS 1980 (EPSG ID 7019)",
> 6378137.0, 298.257222101]], PRIMEM [ "Greenwich", 0.000000 ], UNIT
> ["Decimal Degree", 0.0174532925199433]], PROJECTION ["Albers Conical
> Equal Area"], PARAMETER ["Latitude_Of_Origin", 45.0], PARAMETER
> ["Central_Meridian", -126.0], PARAMETER ["Standard_Parallel_1", 50.0],
> PARAMETER ["Standard_Parallel_2", 58.5], PARAMETER ["False_Easting",
> 1000000.0], PARAMETER ["False_Northing", 0.0], UNIT ["Meter", 1.0]]
> 
> I get a 25% match to EPSG:3005. Yet (obvious differences in the
> various names aside) this DOES look equivalent to the definition
> returned by projinfo EPSG:3005 -o WKT1_ESRI:
> 
> PROJCS["NAD_1983_BC_Environment_Albers",GEOGCS["GCS_North_American_1983",DAT
> UM["D_North_American_1983",SPHEROID["GRS_1980",6378137.0,298.257222101]],PRI
> MEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Albers"]
> ,PARAMETER["False_Easting",1000000.0],PARAMETER["False_Northing",0.0],PARAME
> TER["Central_Meridian",-126.0],PARAMETER["Standard_Parallel_1",50.0],PARAMET
> ER["Standard_Parallel_2",58.5],PARAMETER["Latitude_Of_Origin",45.0],UNIT["Me
> ter",1.0]]
> 
> Why do I only get a 25% confidence match here?

Do not hesitate to give the full context for your question. I suspect this is related to https://
github.com/qgis/QGIS/issues/36216 and Oracle WKT.

First, there's no particular effort in PROJ to deal with and "normalize" (that is use EPSG 
official names) Oracle WKT, so I'm not surprised that it doesn't work out of the box.

So let's play the "spot the difference" game.

One thing that I can see, is that, as PROJ doesn't (yet ?) include an IA, "North American 
Datum 1983 (EPSG ID 6269)" and "North American Datum 1983" are considered to be 
different.

Another more annoying reason is that when actually feeding the Oracle WKT in projinfo it 
reports "Unsupported conversion method: Albers Conical Equal Area", because it doesn't 
know this variant of the projection method name.

So I was rather surprised you got any identification at all. This is only due to just "NAD83 / BC 
Albers" matching exactly the EPSG name for this CRS.
Generally, the identification is done with the "normalized" names, so even if the Oracle WKT 
and ESRI WKT look similar, only the proximity of the ingested & potentially normalized WKT 
you try to identify with the WKT2_2019 output of the candidates CRS matters.

Even


-- 
Spatialys - Geospatial professional services
http://www.spatialys.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20200506/a2f9bdf6/attachment.html>