-------- Original Message --------
Subject:        Re: Google Earth miss-matches
Date:   Sun, 25 May 2008 11:56:08 +0100
From:   Nicholas Verge <[log in to unmask]>
To:     [log in to unmask]



Tom,

Such mismatches could have many causes. In global terms, I think it is in
most instances they may be result of differences in the degree of
positional accuracy that different datsets have been captured with. For
example Google Earth uses pseudo-natural coloured Landsat imagery to
provide its medium resolution rendering. This is derived from the
orthorectified Geocover 2000 Landsat 7 data, with a supposed RMSE
positional error of 14m. If this is overlaid by vector drawing layers
captured from high resolution orthorectified imagery with a positional
error that is very much less, then discrepancies will be apparent. For
regions of the world where accurate data is available and is used by GE,
your suspeicions may also be right. Who knows how GE transforms data
between projections and datums.

This issue is one reason why I think it is very dangerous to use GE for
anything other than as a plaything. If you are undertaking work which
require that you confidently know the accuracy of the data you are working
with, how it was captued and processed, you have to work with data from
original data providers, not via a third party that may have messed around
with it in ways that are not declared and so unknown to you.




On Fri, 23 May 2008 21:21:21 +0100, Maps-L Moderator <[log in to unmask]> wrote:

> -------- Original Message --------
> Subject:        Google Earth miss-matches
> Date:   Fri, 23 May 2008 13:08:57 -0700 (PDT)
> From:   Tom Hanley <[log in to unmask]>
> To:     MAPS-L <[log in to unmask]>
>
>
> Are the miss-matches between the Google Earth maps and various polygons
> and lines like some water features, roads and some municipal boundaries
> due to data sets that have used different datums? Is there an effort to
> reconcile the miss-matches?
>
>
> Tom Hanley, Professor Emeritus of Geology
>
>
> Please use this Yahoo mail account for communications.
> Links to the ACRES projects and to Panama photos may be found at:
> http://chemgeo.ColState.edu/th_hp.htm
> "Rock Trails in Central Park" co-authored with M.M. Graff, is available
> for download at http://www.greenswardparks.org
>



--
----------------------------------------------------------------------
Nicholas J. Verge BSc. FGS
Geologist and geological remote sensing/GIS consultant

CEO,
Earthscience Technologies,
Henley-on-Thames, Oxfordshire, UK.

Voice: ++ 44 (0)1491 572022 (office hours 0900-2200UTC, Monday - Saturday)
Email: [log in to unmask]