Click here to get back home

Error from unknown datum
 Home | NewsGroups | Search | About
Login Password
Register Now!
 comp.infosystems.gis    Post an article   get this group's latest topics as an RSS feed add this group's latest topics to your My MSN content add this group's latest topics to your My Yahoo content
Subject Author Date
Error from unknown datum Eric Peterson 08-29-2005
Posted by Eric Peterson on August 29, 2005, 6:42 pm
Please log in for more thread options
 I sometimes work with data from other people where the datum is unknown.
Thus the datum shift becomes positional error. I have no formal training in
GIS and I'm wondering if someone can explain a quandry to me, or direct me
to an explanatory web page?

When (for example) the data were collected in GCS, NAD83 and I incorrectly
assume GCS, NAD27, the error would be about 87 meters West, by about 7
meters South.

Alternatively, if the data were collected in UTM, NAD83 and I incorrectly
assume GCS, the error is about 200 meters North, by about 80 meters West.

Somehow the logic circuits in my brain think that errors arrising from using
the wrong datum should be independent of the coordinate system, but
apparently I am wrong. What am I missing?




Posted by colinr23 on September 7, 2005, 11:01 pm
Please log in for more thread options
 Hi Eric,

I'm not totally sure i understand your confusion. If by Geographic
Coordinate System (lats and lons) is a spherical system and is not
useful for measuring distances. Projected coordinate systems (UTM etc)
project the coordinates from the sphere (the earth) onto the plain (2D
surface) so distances can be measured consistantly.

I think this would account for irregularities between your results
between GCS and UTM datum shifts.

Hope it helps,

Colin
Eric Peterson wrote:
> I sometimes work with data from other people where the datum is unknown.
> Thus the datum shift becomes positional error. I have no formal training in
> GIS and I'm wondering if someone can explain a quandry to me, or direct me
> to an explanatory web page?
>
> When (for example) the data were collected in GCS, NAD83 and I incorrectly
> assume GCS, NAD27, the error would be about 87 meters West, by about 7
> meters South.
>
> Alternatively, if the data were collected in UTM, NAD83 and I incorrectly
> assume GCS, the error is about 200 meters North, by about 80 meters West.
>
> Somehow the logic circuits in my brain think that errors arrising from using
> the wrong datum should be independent of the coordinate system, but
> apparently I am wrong. What am I missing?



Posted by HotRod on September 8, 2005, 9:57 am
Please log in for more thread options
I'm not GIS either so if this sounds wierd SORRY. In some cases choosing the
wrong data type will result in an ouright error however what your seeing is
data based on some type of projection system. This means that all of your
points are relative to the Projection area and system being used, in UTM you
will notice that each projected point is in reference to a Northing and
Easting correction. Without the Northing and Easting correction your point
will be out. In most cases data files should come with a .prj file that when
opened in note pad will give you an idea of the system being used.

Did that make sense??




Posted by Marc Pelletier on September 8, 2005, 3:55 pm
Please log in for more thread options
$nB6.443@newssvr30.news.prodigy.com:

> Alternatively, if the data were collected in UTM, NAD83 and I incorrectly
> assume GCS, the error is about 200 meters North, by about 80 meters West.
>
>

Eric,

Its really not clear what you mean by GCS.

In any event, the different datums represent different models of the shape
of the earth. Converting the same lat long numbers into utm assuming
different datums will result in different sets of utm coords. Locally each
set should be okay for measuring distance or area within a reasonable small
area, but the problem comes back when you try to merge different data sets,
since points or lines that are coincident in the real world will now have
the differences you've noticed.

The only way of certain way of determining the correct datum is to get that
info from whoever created the data, but you can often deduce it by
comparing an unknown dataset to a known one, if they contain similar data.

Also don't forget to consider the accuracy of your source data. The digital
chart of the world or vmap0 data are only accurate to about a km to start
with, so the errors introduced by assuming the wrong datum are minimal
compared to that.

good luck

Marc Pelletier
Goldak Airborne Surveys


Similar ThreadsPosted
error for arcgis 9.0 January 21, 2006, 9:44 am
error for arcgis January 21, 2006, 9:44 am
How to properly specify NAD27 datum August 26, 2004, 9:02 pm
Gis Query about Datum and Projection. February 22, 2005, 9:22 pm
Re: Converting co-ordinates from one datum to an other? July 20, 2006, 1:46 pm
USGS DLG encoded with wrong datum? May 17, 2004, 3:58 pm
georeferencing a raster in ArcMAp and transforming datum September 26, 2005, 7:19 am
Conversion WGS84 to East Timor Local Datum (1967) January 19, 2005, 8:18 pm

Our other projects:

Art Dolls, Fairies and Mermaids - Sunnyfaces.net

Roy's Linux, Programming and Search Engines messages

1-Script XML SitemapXML Sitemap