[Proj] Using a Molodensky transformation instead of NADCON

[Clifford J Mugnier]

Tim,

Yes there is because PROJ4 is used for a lot of international work.
However, using it with U.S. data is not a good idea.  NADCON will yield
typical transformation accuracies of plus or minus a tenth of a meter.  A
three-parameter transformation designed SPECIFICALLY FOR A SINGLE COUNTY
will yield typical accuracies of  plus or minus ONE to THREE METERS.  It is
obviously worse for an entire state and is comparatively awful for a
region.

Only a 7-parameter Molodensky Model that also uses the datum origin, Meades
Ranch, KS in the equations can yield accuracies equal to NADCON, and then
only after the transformation parameters have been computed for a specific
county.  It can be done, but it's a pretty "hairy" and difficult task to
accomplish.  I had over a dozen Graduate Students do it (back in 1984-85)
for the State of Louisiana, each student had about 5 Parishes (counties) to
do, and they were quite busy for an entire semester.

The NADCON technique is preferable for reliablity in GIS applications,
especially since data sets can subsequently used by others that do not read
the metadata files and/or do not understand that the Molodensky-transformed
data is substantially inferior in positional accuracy.

All datum shift techniques will DEGRADE the accuracy of original survey
data.  Only the re-computation of original data on a different datum will
preserve the original accuracy.  For some GIS applications of a regional or
thematic objective it does not matter.  Usually the most reliable technique
is preferable and that's NADCON for the U.S., and NTv2 for Canada, the UK,
and Australia.

Regards,

Clifford J. Mugnier
Chief of Geodesy
CENTER FOR GEOINFORMATICS
Department of Civil Engineering
LOUISIANA STATE UNIVERSITY
Baton Rouge, LA 70803
Academic Office Voice and Facsimile:   (225) 578-8536
Research Office Voice and Facsimile:   (225) 578-4474
======================================================
http://www.ASPRS.org/resources.html
http://www.cee.lsu.edu/facultyStaff/mugnier/index.html
======================================================



Hey folks,

Is there a way to use a Molodensky transformation with Proj.4 when changing
from NAD83 to NAD27?

Thanks in Advance.

Tim

Timothy N.
Loesch
Minnesota Department of Natural Resources
GIS Operations Supervisor
MIS Bureau
(651)
296-0654                                                                         
 500 Lafayette Road
tim.loesch at dnr.state.mn.us
Saint Paul, MN  55155
   Lat   44.95609
Long -93.08388