[FWTools] projecting with ogr2ogr v. ArcInfo

[Bryan Keith]

Frank,

Thanks so much for your help!  That worked.  Now I wonder which is 
correct, but consistency between other software packages seems important 
because ultimately I want my data translated to the same county 
coordinate system of the data I've received.

What I'm actually trying to do is go from UTM (zone15 NAD83) to the 
county coordinate system.  It's my understanding that ArcInfo requires a 
two-step process:

from UTM to geographic with this information:

INPUT
Projection UTM
Zone 15
Datum NAD83
Units METERS
Parameters
OUTPUT
Projection GEOGRAPHIC
Datum NAD83
Units DD
Parameters
END

and from geographic to the county coordinate system with this 
information (same as in my previous e-mail):

INPUT
Projection GEOGRAPHIC
Spheroid GRS1980
Units DD
Parameters
OUTPUT
projection lambert conformal conic
units feet
parameters 6378418.941 6357033.310
44.8833333333
45.1333333333
-93.3833333334
44.7911111111
152400.3048
30480.0610
END

I can now duplicate this two-step process with ogr2ogr.  However, since 
your solution required specifying the source of the lat-long information 
with the spheroid parameters, it seems that this coordinate 
transformation is also a two-step process with ogr2ogr.  Does that seem 
correct to you?

I'll send actual numbers if necessary.

Bryan

Frank Warmerdam wrote:

>On 12/15/05, Bryan Keith <bryan at geomega.com> wrote:
>  
>
>>Hello,
>>
>>I'm attempting to project data from geographic coordinates to a
>>Minnesota county coordinate system (Hennepin County) using ogr2ogr and
>>ArcInfo.  The results are slightly different, and I'm not sure why.  X
>>coordinates are about the same (within 0.1 feet), but Y coordinates
>>differ by ~3.1 feet.  This difference seems consistent though I haven't
>>verified that thoroughly.  I include a small subset of the points in
>>this e-mail.
>>
>>Input coordinates are long-lat:
>>
>>-93.17356183         45.04758353
>>-93.36960993         45.08857070
>>-93.27394885         44.97432735
>>
>>Output from ogr2ogr:
>>
>>554221.66861940 193587.15836056
>>503544.69316568 208462.19508218
>>528309.75142366 166825.34464421
>>
>>Output from ArcInfo:
>>
>>554221.6767         193584.0575
>>503544.6937         208459.0942
>>528309.7556         166822.2438
>>    
>>
>
>Bryan,
>
>I believe that Arc/Info is just doing the reprojection
>step on the custom ellipsoid, but not taking any action
>to try and translate that to the NAD83 datum.
>
>If I just do this with PROJ.4 directly, I get the same
>results as Arc/Info.
>
>% proj +proj=lcc +lat_1=44.88333333333333 +lat_2=45.13333333333333
>+lat_0=44.79111111111111 +lon_0=-93.38333333333334 +x_0=152400.3048
>+y_0=30480.06096 +a=6378418.941 +b=6357033.31   +units=us-ft +no_defs
>-93.17356183         45.04758353
>554221.68       193584.06
>
>But if I do a translation from NAD83 to that coordinate system I get
>the values you saw.
>
>% cs2cs +proj=latlong +datum=NAD83 +to +proj=lcc
>+lat_1=44.88333333333333 +lat_2=45.13333333333333
>+lat_0=44.79111111111111 +lon_0=-93.38333333333334 +x_0=152400.3048
>+y_0=30480.06096 +a=6378418.941 +b=6357033.31 +units=us-ft +no_defs
>-93.17356183         45.04758353
>554221.67       193587.16 -281.47
>
>I believe this is because PROJ.4 is doing an ellipsoid to ellipsoid
>conversion.
>
>So, if you just want to do the reprojection with no datum shifting
>you should specify "+proj=latlong +a=6378418.941 +b=6357033.3"
>as your source coordinate system.
>
>Note, this automatic ellipsoid to ellipsoid conversion (when there is
>no explicit datum shift info) is something I am likely to remove from
>PROJ.4 in the next revision since it is causing so much problem.  It
>definately seems to *not* be what other software packages do.
>
>Best regards,
>
>--
>---------------------------------------+--------------------------------------
>I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
>light and sound - activate the windows | http://pobox.com/~warmerdam
>and watch the world go round - Rush    | Geospatial Programmer for Rent
>
>
>  
>