NAD Conversions

[Gerald I. Evenden]

>Date: Thu, 20 Oct 1994 10:39:53 -0600
>From: a16dhoover at attmail.com (David  Hoover)
>Received: from a16dhoover by attmail; Thu Oct 20 16:39:53 GMT 1994
>Subject: NAD Conversions
>Sender: grass-lists-owner at moon.cecer.army.mil
>To: grassu-list at max.cecer.army.mil
>
>Greetings from Idaho !
>
>I have some concerns about explaining datum conversions.  I have had 
>someone who has gotten six different answers from six different people 
>(of course, I think my answer was a correct one!), but I think that is 
>more a result of asking the right questions in the right way and giving 
>the right answer in an equally right way.  I`ll pose the question to you 
>folks who I have come to rely upon for sound cartographic answers as well 
>as GRASS answers.
>
>By the way, the question arose when the person was attempting to transform
>NAD27 soil data on a 7.5 minute quad base to NAD83.
>
>Given:  The Alpha quad, in NAD27, of soils data, has a NE corner of 
>       -114 00 00 longitude and 44 00 00 latitude.  The corresponding 
>       UTM easting and northing values are 547100 and 4341400.  (These 
>       values are made up just for discussion purposes.)
>
>Given:  A transformation of the entire file is made to the NAD83 datum.
>
>Given:  All coordinates in that file have changed due to the
>                transformation.
>
>Question:  What are the longitude and latitude of the NE corner now?
>           Is it still -114 00 00 and 44 00 00?
>           Have the easting and northing of the NE corner changed?
>           Would they now be 547032 and 4341479 (or whatever the
>                actual shift for that quad is)?

The geographic *and* UTM coordinates will change.  A datum change
in the case of NAD *is a change in the geographic location*.
UTM coordinates are a red herring and only reflect geographic
projection to cartesian for a particular ellipsoid.

>I tried to tell this person that latitude longitude are surface 
>expressions of angular measurements and will not change the same 
>way we think of coordinates systems such as UTM changing.  For 
>example, in the geographic coordinate system, 90 degrees will still 
>be straight from the center of the earth to the north pole and 0 
>degrees will be perpendicular to that, and datum shifts will not 
>change that.  Thus, the NE corner would still have the same spherical 
>(lat/long) coordinate values under both datums.  What does change is 
>the location on the earth's surface of the NE corner of that quad in 
>relation to a plane coordinate system such as a UTM.

The above is loaded with mistakes.  For one thing, geographic latitude
is not geocentric.  Geographic coordinates refer to the point's location
on the earth.  Also datum shifts *will* and normally do change
geographic location.  Datum, per se, has nothing to do with a projection
system, i.e. UTM.

>If all the quads in an area are converted individually, then there will
>be no gaps in the data.

A meanless statement.

>How does this fit with your interpretation of datum shifts?  Maybe we
>are mixing apples and oranges here in talking about point coordinate
>changes and how we graphically represent them on a map.  Of course the
>plane coordinates for a quad's corner change, but we will still publish
>that map on a 7.5 minute base and have evenly divided corner locations
>in 7.5 minute increments.

Quadrangles and their boundaries have *nothing* to do with datum shifts.
But by virtue of a datum shifts, data on one sheet may be move to another
sheet *if* the boundaries of the sheet reflect the new datum.

>Thanks in advance for shedding some light on how I might explain this
>business better.  Even if you agree with the way I am presenting this,
>let me know so I can get some feedback!
>
>David Hoover, Manager
>Cartography and GIS Section
>USDA - Soil Conservation Service
>Idaho State Office (Boise)
>internet!attmail!a16dhoover
>208 334 1525

To repeat myself: changes in datum cause shift in the geographic
coordinates of the data points.  Standard datum shift procedures (like
NADCON) perform their conversions in geographic space thus for data in
a projection system like UTM must be inverted to geographic, datum
shifted and then reprojected back to cartesian.

Projection systems, like UTM, are merely artifacts for human
convenience in portrayal of geographic data.  They should not be
viewed as primary definitions of geographic location and/or confused
with geographic operations like datum shifts.

Gerald (Jerry) I. Evenden   Internet: gie at charon.er.usgs.gov
voice: (508)563-6766          Postal: P.O. Box 1027
  fax: (508)457-2310                  N.Falmouth, MA 02556-1027