Re: Re: Campo Inchauspe to WGS84 parameters

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Re: Campo Inchauspe to WGS84 parameters

Clifford J Mugnier




ALL datum shifts are approximations, particularly those that only use the 3
translations.  Accuracy increases as the number of parameters increase
and/or the size of the region being transformed decreases.

One generalization that is ALWAYS correct is that EVERY datum shift will
decrease the accuracy of the original data once it is transformed to the
other datum.

EPSG-published parameters are directed to regions that are predominately
hydrocarbon-producing ... e.g. coastal Argentina.  NIMA-published
parameters are directed to regions that are potential battle zones on land
... e.g. interior AND coastal Argentina.  None of them are "exact."
Classical datums have systematic errors that change greatly over vast
regions.  Argentina is not a small country.  Consider also the many sets of
3-parameter transformations available for the Continental U.S.

For those that "cherry pick" parameters from my columns, I suggest reading
the entire column with the associated background information for the
country of interest.  The primary living authority on Argentine Geodesy is
Mr. Ruben Rodriguez, as referenced in my column.  He is a practicing
Geodesist that lives in Argentina.

Datum shifts are "starters" for general reconaissance.  When one wants the
EXACT transformation parameters for a specific area, one puts a Geodesist
on the ground to find out.  Practical surveying beats out office theory
every time.

Cliff Mugnier

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

RE: Re: Campo Inchauspe to WGS84 parameters

Noel Zinn-2
Good points, Cliff, but you should caution your readers about the perils of
deriving 7-parameter shifts in "small" areas (and the use of 7-parameter
shifts so derived).  The smaller the area of the derivation the higher are
the correlations among the translations (as a group) and the rotations and
the scale (as another group).  In a small area the user should be satisfied
with deriving and using a 3-parameter shift, since the 3 orthogonal
translations are uncorrelated.  Introducing the rotations and scale to the
adjustment leads to (natural and expected) coordinate "noise" (inaccuracies)
driving big numbers for the solved parameters, which tend to countervail one
another and which are, therefore, physically meaningless.  In a small area
the 7-parameter derivation is an ill-conditioned adjustment that should be
avoided.  Of course, if you've already got a 7-parameter shift (given to
you) you need to use all 7 parameters (or you're in big trouble).  -Noel
Zinn  

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Clifford J Mugnier
Sent: Friday, November 11, 2005 10:45 AM
To: PROJ.4 and general Projections Discussions
Subject: Re: [Proj] Re: Campo Inchauspe to WGS84 parameters

ALL datum shifts are approximations, particularly those that only use the 3
translations.  Accuracy increases as the number of parameters increase
and/or the size of the region being transformed decreases.

One generalization that is ALWAYS correct is that EVERY datum shift will
decrease the accuracy of the original data once it is transformed to the
other datum.

EPSG-published parameters are directed to regions that are predominately
hydrocarbon-producing ... e.g. coastal Argentina.  NIMA-published
parameters are directed to regions that are potential battle zones on land
... e.g. interior AND coastal Argentina.  None of them are "exact."
Classical datums have systematic errors that change greatly over vast
regions.  Argentina is not a small country.  Consider also the many sets of
3-parameter transformations available for the Continental U.S.

For those that "cherry pick" parameters from my columns, I suggest reading
the entire column with the associated background information for the
country of interest.  The primary living authority on Argentine Geodesy is
Mr. Ruben Rodriguez, as referenced in my column.  He is a practicing
Geodesist that lives in Argentina.

Datum shifts are "starters" for general reconaissance.  When one wants the
EXACT transformation parameters for a specific area, one puts a Geodesist
on the ground to find out.  Practical surveying beats out office theory
every time.

Cliff Mugnier

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj


_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Re: Campo Inchauspe to WGS84 parameters

OvV_HN
In reply to this post by Clifford J Mugnier
>From mr. Warmerdam:
> Looking at Cliff's column, it seems EPSG shift is quite
> different than the NIMA one Cliff indicates.  This would
> make me quite nervous about picking just one of them as
> authoritative.

NGA-NIMA gives the -148,136,90 solution as a solution with respect to the
Campo Inchauspe 1969 datum.
The other solution mr. Mugnier's article mentions (-62,-1,-37) is the SAN-A
solution, I believe a mean solution for the whole of Argentina.
See the NGA-NIMA page:
<http://earth-info.nga.mil/GandG/coordsys/onlinedatum/CountrySouthAmericanTa
ble.html>


_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Re: Campo Inchauspe to WGS84 parameters

OvV_HN
In reply to this post by Noel Zinn-2
>From mr. Zinn:
> Good points, Cliff, but you should caution your readers about the perils of
> deriving 7-parameter shifts in "small" areas (and the use of 7-parameter
> shifts so derived).  The smaller the area of the derivation the higher are
> the correlations among the translations (as a group) and the rotations and
> the scale (as another group).  In a small area the user should be satisfied
> with deriving and using a 3-parameter shift, since the 3 orthogonal
> translations are uncorrelated.

Or use a 10-parameter Molodenksy-Badekas transformation. See EPSG coordinate
operation method 9636.
Example: the Dutch Cadastre has calculated coefficients for transformation
between Bessel 1841 and ETRS89. Rotation is around datum point Amersfoort.
See section 7.2.1 in "Het stelsel van de Rijksdriehoeksmeting en het
European Terrestrial Reference System 1989 - Gebruik en definitie"
(The RD system and the ETRS89 system - use and definitions; in Dutch
language).
http://www.rdnap.nl/download/rdnaptrans.pdf



  Introducing the rotations and scale to the

> adjustment leads to (natural and expected) coordinate "noise" (inaccuracies)
> driving big numbers for the solved parameters, which tend to countervail one
> another and which are, therefore, physically meaningless.  In a small area
> the 7-parameter derivation is an ill-conditioned adjustment that should be
> avoided.  Of course, if you've already got a 7-parameter shift (given to
> you) you need to use all 7 parameters (or you're in big trouble).  -Noel
> Zinn  
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Clifford J Mugnier
> Sent: Friday, November 11, 2005 10:45 AM
> To: PROJ.4 and general Projections Discussions
> Subject: Re: [Proj] Re: Campo Inchauspe to WGS84 parameters
>
> ALL datum shifts are approximations, particularly those that only use the 3
> translations.  Accuracy increases as the number of parameters increase
> and/or the size of the region being transformed decreases.
>
> One generalization that is ALWAYS correct is that EVERY datum shift will
> decrease the accuracy of the original data once it is transformed to the
> other datum.
>
> EPSG-published parameters are directed to regions that are predominately
> hydrocarbon-producing ... e.g. coastal Argentina.  NIMA-published
> parameters are directed to regions that are potential battle zones on land
> ... e.g. interior AND coastal Argentina.  None of them are "exact."
> Classical datums have systematic errors that change greatly over vast
> regions.  Argentina is not a small country.  Consider also the many sets of
> 3-parameter transformations available for the Continental U.S.
>
> For those that "cherry pick" parameters from my columns, I suggest reading
> the entire column with the associated background information for the
> country of interest.  The primary living authority on Argentine Geodesy is
> Mr. Ruben Rodriguez, as referenced in my column.  He is a practicing
> Geodesist that lives in Argentina.
>
> Datum shifts are "starters" for general reconaissance.  When one wants the
> EXACT transformation parameters for a specific area, one puts a Geodesist
> on the ground to find out.  Practical surveying beats out office theory
> every time.
>
> Cliff Mugnier
>
> _______________________________________________
> Proj mailing list
> [hidden email]
> http://lists.maptools.org/mailman/listinfo/proj
>
>
> _______________________________________________
> Proj mailing list
> [hidden email]
> http://lists.maptools.org/mailman/listinfo/proj

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

RE: Re: Campo Inchauspe to WGS84 parameters

Noel Zinn-2
Mr. van Vlijmen,

Excellent!  The 10-parameter Molodenksy-Badekas transformation is the right
solution for small areas.

Is it supported by Proj.4?

Noel Zinn

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Oscar van Vlijmen
Sent: Friday, November 11, 2005 5:55 PM
To: PROJ.4 and general Projections Discussions
Subject: Re: [Proj] Re: Campo Inchauspe to WGS84 parameters

>From mr. Zinn:
> Good points, Cliff, but you should caution your readers about the perils
of
> deriving 7-parameter shifts in "small" areas (and the use of 7-parameter
> shifts so derived).  The smaller the area of the derivation the higher are
> the correlations among the translations (as a group) and the rotations and
> the scale (as another group).  In a small area the user should be
satisfied
> with deriving and using a 3-parameter shift, since the 3 orthogonal
> translations are uncorrelated.

Or use a 10-parameter Molodenksy-Badekas transformation. See EPSG coordinate
operation method 9636.
Example: the Dutch Cadastre has calculated coefficients for transformation
between Bessel 1841 and ETRS89. Rotation is around datum point Amersfoort.
See section 7.2.1 in "Het stelsel van de Rijksdriehoeksmeting en het
European Terrestrial Reference System 1989 - Gebruik en definitie"
(The RD system and the ETRS89 system - use and definitions; in Dutch
language).
http://www.rdnap.nl/download/rdnaptrans.pdf



  Introducing the rotations and scale to the
> adjustment leads to (natural and expected) coordinate "noise"
(inaccuracies)
> driving big numbers for the solved parameters, which tend to countervail
one

> another and which are, therefore, physically meaningless.  In a small area
> the 7-parameter derivation is an ill-conditioned adjustment that should be
> avoided.  Of course, if you've already got a 7-parameter shift (given to
> you) you need to use all 7 parameters (or you're in big trouble).  -Noel
> Zinn  
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Clifford J Mugnier
> Sent: Friday, November 11, 2005 10:45 AM
> To: PROJ.4 and general Projections Discussions
> Subject: Re: [Proj] Re: Campo Inchauspe to WGS84 parameters
>
> ALL datum shifts are approximations, particularly those that only use the
3

> translations.  Accuracy increases as the number of parameters increase
> and/or the size of the region being transformed decreases.
>
> One generalization that is ALWAYS correct is that EVERY datum shift will
> decrease the accuracy of the original data once it is transformed to the
> other datum.
>
> EPSG-published parameters are directed to regions that are predominately
> hydrocarbon-producing ... e.g. coastal Argentina.  NIMA-published
> parameters are directed to regions that are potential battle zones on land
> ... e.g. interior AND coastal Argentina.  None of them are "exact."
> Classical datums have systematic errors that change greatly over vast
> regions.  Argentina is not a small country.  Consider also the many sets
of

> 3-parameter transformations available for the Continental U.S.
>
> For those that "cherry pick" parameters from my columns, I suggest reading
> the entire column with the associated background information for the
> country of interest.  The primary living authority on Argentine Geodesy is
> Mr. Ruben Rodriguez, as referenced in my column.  He is a practicing
> Geodesist that lives in Argentina.
>
> Datum shifts are "starters" for general reconaissance.  When one wants the
> EXACT transformation parameters for a specific area, one puts a Geodesist
> on the ground to find out.  Practical surveying beats out office theory
> every time.
>
> Cliff Mugnier
>
> _______________________________________________
> Proj mailing list
> [hidden email]
> http://lists.maptools.org/mailman/listinfo/proj
>
>
> _______________________________________________
> Proj mailing list
> [hidden email]
> http://lists.maptools.org/mailman/listinfo/proj

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj


_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Re: Campo Inchauspe to WGS84 parameters

Frank Warmerdam
On 11/11/05, Noel Zinn <[hidden email]> wrote:
> Mr. van Vlijmen,
>
> Excellent!  The 10-parameter Molodenksy-Badekas transformation is the right
> solution for small areas.
>
> Is it supported by Proj.4?

Noel,

No, I'm afraid not.  Not that I am well read, but this is the
first time I had heard of it.

Cliff's points on the dangers of accepting approximate
datum transformations is also well taken.  I would however
note that most mappers cannot afford to place geodesists on
the ground.   So we scratch around hoping to find something
reasonable.

Best regards,
--
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, [hidden email]
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj