Georeferencing and spatial literacy

[Robert Mesibov]

The Phyloinformatics article by Murphey et al. (to which Robert Guralnick
kindly pointed) is an excellent overview of the problems involved in
'retrospective georeferencing'. When we redesigned the zoology collection
database at our Museum a couple of years ago, we spent a lot of time chewing
over these problems. In the end we opted to do the job manually. 'Manually'
in our case meant 'by someone who knows the local geography and something,
at least, about the collectors and their recording habits'. Needless to say,
the job isn't finished.

There was another reason for doing it manually. Despite the widespread use
of GPS units these days and the availability of high-quality, UTM-gridded
maps, we sometimes get numerical location data on specimen labels which make
no sense at all. Some present-day collectors simply haven't a clue about
spatial data. Some of our Museum data-enterers were likewise 'challenged'.
Our partial solution was to put a spatial data primer (written in HTML, with
pretty pictures) on Museum workstations as a reference for data-enterers.

Another approach to this problem is to ask present-day collectors to record
a good verbal location as well as a numerical location for each collecting
site. Collectors here tend to record a nearest named place (NNP) and a
string of UTM numbers. NNPs are generally towns or landscape features, and
they have a mysterious attractive power in the minds of some collectors.
Collections made 5.5 km NW of a hill are recorded verbally as being at 'Hill
X', instead of '5.5 km NW of Hill X'. The collector seems to be thinking:
'I've recorded the site accurately with the UTM numbers, so I can generalise
the verbal location.'

We are also having problems with the datum shift from AGD66 (used as the
grid base on all Tasmanian maps until 2003) to GDA94 (more or less the same
as WGS84). I sometimes get blank looks when I ask collectors which datum
they used when recording localities with a GPS unit. It isn't widely
appreciated that the datum shift of ca. 200 m is more than 10 times the
precision of typical GPS readings.
--
Dr Robert Mesibov
Honorary Research Associate
Queen Victoria Museum and Art Gallery
and School of Zoology, University of Tasmania
Home address: PO Box 101, Penguin, Tasmania, Australia 7316
Home phone: (03) 6437 1195