D'Arcy,

There are just a few public-access sites that have more or 
less "original" data on various datums.  The most 
comprehensive data set is that of the U.S. military.

With regard to that, when you go to the http://www.nima.mil 
site, there are a variety of documents that you can 
download as well as software.  For instance, Technical 
Report TR8350.2, Third Edition, 4 July 1997, "Department of 
Defense World Geodetic System 1984" lists all of the datum 
relations that are unclassified and currently available 
from the U.S. military.  It is 170 pages long and is 
available free for downloading in Adobe Acrobat ".pdf" 
format.  (There is also a link to the Adobe site for those 
users that do not have a copy of the free Acrobat program 
that runs on Windows95/98/NT.)  Printouts on a laser 
printer are gorgeous.  An item usually ignored by most 
people (from my observations), is that this reference work 
gives the accuracy estimators of the transformation 
parameters.  There are many places in the world where 
3-parameter datum shift values are given and the geocentric 
components are +/- 25 meters in X, and in Y, and in Z! 
 When you do a three-dimensional solution for the 
Pythagorean Solution, that equates to 57+ meters on the 
ground!  I commonly see comments on the "net" to the effect 
that, "Golly, my data doesn't seem to fit when I use 
__________ software package to do a datum shift ..."  Well, 
guess where _________ software package got the parameters 
for the datum shift ...   Yep, TR8350.2, and nowhere else. 


For the most part, there are no "secrets" regarding datum 
shift parameters when you consider 3-parameter solutions. 
 The U.S. military has a long-established custom of calling 
that sort of thing a "Molodensky" 3-parameter datum shift" 
or on occasion, an "Abridged Molodensky 3-parameter datum 
shift" if the user is not interested in ellipsoid heights. 
 This is so much gobbledygook.  Without going into the 
subject deeply, this sort of transformation thing is 
intended for military users to sling bombs, missiles, 
napalm, and various related types of ordinance at the 
enemy.  The military does NOT need hair-splitting accuracy 
to conduct war.  Therefore, the National Imagery and 
Mapping Agency (NIMA) publishes only what is necessary to 
conduct its operations.  In those cases that better 
accuracy is published, it is absolutely ALWAYS with the 
consent of that particular country.  (Notice that you can 
get pretty good accuracy in the U.S., U.K., and Europe, but 
the parameters for the C.I.S. and many former members of 
the U.S.S.R. are good only to 25 meters in each geocentric 
component.)  There are many, many places in the world that 
are too BIG to offer better accuracy that +/-25 meters for 
a single set of 3 parameters for a datum shift!  Note that 
even in the U.S. there is an entire collection of 
parameters that you have to use depending on where you are 
in the country!  For that reason, most people use the NA  
DCON technique in the U.S. rather than the military 
technique.  The same sort of thing goes for Canada, and the 
Europeans are doing the same sort of things from country to 
country.

TR8350.2 also has the Multiple Regression Equation (MRE) 
coefficients for various continental land masses such as 
Europe that will accomplish a transformation from the 
European Datum of 1950 to WGS84 with better accuracy than a 
simple 3-parameter datum shift.  Note however, that NATO 
commanders have been warned not to use the MRE coefficients 
for naval artillery support applications because the 
reliablity of this method is seriously defficient!  The 
moral of the story is that (the use of a) MRE is useful 
only in areas of "interpolation" and not in areas of 
"extrapolation."  In the U.S., I think the fools and idiots 
in the U.S. Bureau of Land Management never were aware of 
this little "factoid" as applied to the lease blocks in the 
Gulf of Mexico.  Such MRE coefficients are also available 
for Brazil, Argentina, Australia, etc. - all only suitable 
for use "onshore" and not for offshore or navigational 
applications.

A secondary and VERY derivative source of miscellaneous 
datums is the European Petroleum Studies Group (EPSG) site 
hosted by PetroConsultants.com in Switzerland.  This is a 
group of (European) oil companies that have gathered 
together and agreed to share some of their data.  The 
majority of the data they have so graciously agreed to 
share is from TR8350.2 along with some parameters (mostly 
correct) regarding national Grid Systems.  However, their 
data collection is woefully small because they only use 
data that is "open to the public domain and not 
proprietary."  The data sets are copyrighted and are free 
to anyone that cares to download the stuff, but the proviso 
is that if one uses the stuff, one must credit the EPSG as 
the source. Their stated accuracy criterion is +/- one 
meter, and that is oftentimes met, but definitely not al  
ways met.  As a result, I consider it interesting poison. 
 I periodically download it and look at it, but I do not 
use it!

The NIMA site also offers software for free.  There is 
MADTRAN which implements all the published TR8350.2 datum 
shift parameters, along with another package called MUSE. 
 MUSE is something else.  It contains DCCP, a similar 
package to MADTRAN, but is for windows only, and various 
versions will work in WIndows95/98/NT, maybe Win3.1, and 
also for various flavors of UNIX such as Sun, Silicon 
Graphics, and whatever else I don't remember.   MUSE, 
however, will also do lots of GIS types of displays that 
include vector and raster formats, but all peculiarly U.S. 
Department of Defense (and NATO) stuff.  They (at present) 
could not care less about commercial GIS formats.

Whatever other site you go to, unless it is a national 
government (like Australia) will only list the TR8350.2 
data sets.  There is nothing else offered anywhere for 
free.  In regard to Grid transformation software, the free 
stuff is limited to the General Cartographic Transformation 
Package (GCTP) available in Fortran77 from the U.S. 
National Geodetic Survey sites as well as various U.S. 
Geological Survey sites.  The U.S.G.S. also offers PROJ4, a 
version of GCTP that has been extended for a few foreign 
countries beyond GCTP as well as a number of additional 
"pretty picture" projections and all translated into ANSI 
"C" language.

The National Geodetic Survey offers "NADCON" for free 
download which is a datum shift package among NAD27, NAD83, 
and NAD83 HARN.

The U.S. Army Corps of Engineers runs the Topographic 
Engineering Center (TEC) at Fort Belvoir, Virginia.  As a 
service to the Civil Works Division of the Corps of 
Engineers, they took the NADCON source code (trivial) along 
with the data files (really big deal), and wrote a Windows 
package that will also do State Plane Coordinate 
Transformations for every state and possession in the Un  
ited States!  It is available for free and is called 
CorpsCon.  Highly recommended!  That is what I use to teach 
with.  I consider CorpsCon the "cat's meow" in geodetic 
utility software for the United States.  I am constantly 
amazed at the number of people that spend good money to buy 
"commercial" software that is no better than CorpsCon!

The Canadian Government sells their version of CorpsCon. 
 Different country, different way of doing things.  Similar 
arrangements are to be found in England, Ireland, Northern 
Ireland, Belgium, etc.  Outside of the U.S. and Australia, 
free is rare.  Damned rare.

The biggest free site in the world is http://www.nima.mil . 
 That has about 110+ datums, and a few transformations 
beyond and including UTM.  EPSG has some free data but no 
free software.  I wrote a package a few years ago called 
"GRIDS 2.0".  It was reviewed by John P. Snyder and he used 
it for his private consulting practice after he retired 
from USGS for a few years before he passed away.  GRIDS 2.0 
has information on 1,100+ datums and 3,200+ Grid systems 
used in absolutely every country in the world for the past 
150 years.  I sold a few copies to oil companies and some 
"special" agencies in the U.S. government (nobody else 
could afford it), and have not sold any since.  I have been 
looking for software houses to license it from me and have 
been unsuccessful so far.  I have decided that selling for 
volume is the only way to go.  No more 5-digit prices.

Anybody out there want to license a DLL that is a bazillion 
times better than anything else, including the U.S. 
military?

$100 per copy?  (minimum 10 licenses)

Anybody?

Hello?

Miscellaneous datums?   You betcha!  Read my columns on 
Grids and Datums at:

http://www.asprs.org/resources.html

Cliff

Clifford J. Mugnier ([EMAIL PROTECTED])
The Topographic Engineering Laboratory
Department of Civil and Environmental Engineering
UNIVERSITY OF NEW ORLEANS
New Orleans, Louisiana  70148

Voice and Facsimilie: (504) 280-7095

>>>>>>>Business phones (504) 443-2355 or (504) 
286-1200<<<<<<<

--------------------------------------------------------  
-----------------------------------

On Tuesday, 29 June, 1999 6:02 PM, D'Arcy Emery 
[SMTP:[EMAIL PROTECTED]] wrote:
> Hi:
>
>       Would anyone happen to know where I could get numeric
>       data on the various geodetic datums?
>
>               Thanks Very Much,
>
>                       --D'Arcy
>
> -------
> Si Deus pro noblis quis contra nos?
> HomePage: http://www.APlusAdmiralty.com
> [EMAIL PROTECTED]
>
> 
**********************************************************
>
----------------------------------------------------------------------
To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put
"unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]

Reply via email to