GPSFileDepot Forums

General Category => Using The Maps/Garmin Software => Topic started by: hwstock on June 01, 2012, 02:34:23 PM

Title: Error in maps for cebtral NV
Post by: hwstock on June 01, 2012, 02:34:23 PM
The recommended NV 24k maps have a registration error for one region of central NV near the Desatoya Twins (east of route 50, WSW of Austin).

I'd like to show you exactly where, but this brain-dead BBS software will not allow me to post external links

This looks like a datum error when the contour map was created from the DEM.  I have used the NV 24k maps in many parts of Nevada, and this is the 1st error that I have seen.

E-mail me if you want to undeniable examples of the error.
Title: Re: Error in maps for cebtral NV
Post by: -Oz- on June 01, 2012, 03:20:48 PM
if you post one or two more post it'll let you post links.  just reply to this thread with bump and see if it'll let you link.  It is part of the anti-spam measures (I know it totally screws the humans).

are these the links you're referring to?
The NV 24k maps put this track west of the peak:
http://hwstock.org/nverr/00Nevada24k.jpg

The track is actually right on the peak ridge, as seen by ExpertGPS:
http://hwstock.org/nverr/02Nevada24kExpert.jpg
Title: Re: Error in maps for cebtral NV
Post by: hwstock on June 01, 2012, 05:28:56 PM
Those are two of them... but enough to show the error.  The other two files (that I tried to link) just verify that 2 other sources give the correct placement of the track (i.e., a different placement than the NV 24k maps on this site).

I can do a search out from Desatoya Twins, by putting waypoints on peaks in mapping software, then plotting those waypoints on the NV 24k maps from GPSFileDepot, if that might help.  Then I could binary search in to try to locate the border between right and wrong maps... may be more tedious than I imagine, especially in flat areas.
Title: Re: Error in maps for cebtral NV
Post by: -Oz- on June 04, 2012, 07:28:34 AM
Since that appears to be a road this error is likely from the U.S. Census Bureau data used.
Title: Re: Error in maps for cebtral NV
Post by: hwstock on June 04, 2012, 02:02:00 PM
?? The 24k map contours are shifted east by several hundred feet.  The "road" is NOT a road at all, but is a track recorded on my gps, as I was walking/climbing.  The track plots correctly on USGS (via expert GPS), memmap, and Maptech software.  It plots incorrectly on the 24k NV maps available on this site.  I was there; I traveled directly up the ridge, and did not sidehill  for miles. I get this on 2 gps units that have the 24k maps from this site. Your 24k maps have proved to be "on" in many, many other NV locations.  But they are out-of-whack here.
Title: Re: Error in maps for cebtral NV
Post by: maps4gps on June 04, 2012, 04:00:50 PM
OZ,

   Looking at the two jpgs in your post, I agree with hwstock.
Look at Rock Creek on the west side, Smith Creek on the east side and North Fork Milkhous
Creek on the south side..  The contours defining the 'valley' bottoms are abour as much to the east of the streams as the contours for the ridge line that hwstock was walking.  However,
the contours for the unnamed stream to the north-northwest appear OK.  It also appears that
the countours around the lower portion of the stream feeding into Smith Creek from the north
are off while those for the upper portion are correct.  I take the color shift at the thin black line slightly to the north of the T17N/T18N red line and the very slight offset of the contour lines along that line to indicate a 7-1/2' quad boundary here.  The contours to the north appear to be correctly positioned, those to the south not.  Likely such a change is occuring along the boundary of a 7-1/2' quad; and more than one quad could be misregistered.

I have seen misregistered NED data from EROS Data Center before are occasionally they
do correct it.  Other than reporting to EROS and hoping they will correct it some time in the
future, I do not know if there is anything a map author could easily do to correct it.
Title: Re: Error in maps for cebtral NV
Post by: hwstock on June 05, 2012, 09:32:12 AM
Thanks much for your analysis!  The author of ExpertGPS found that many of the NV USGS topos off their web server were misregistered by about that amount.  Initially he had to apply the corrections before tying them in to his display. I guess the most you can do is keep a list of the quads that are off, leave it for folks to see (in the notes associated with the download), and communicate the error to the source of the DEMs.