[OSM-talk] Mapping and surveying long road tunnels

Andy Robinson Andy_J_Robinson at blueyonder.co.uk
Thu Mar 22 14:45:09 GMT 2007

Steve Chilton
>Sent: 22 March 2007 2:00 PM
>To: OSM
>Subject: [OSM-talk] Mapping and surveying long road tunnels
>Last weekend I went to the Open Knowledge Foundation OpenGeodata session
>in East London (UK), and whilst there took the opportunity to map a
>little around there, including Limehouse Basin. It struck me that we
>haven't yet captured the Limehouse Link which is a mile long tunnel
>opened in 1993 to make the Docklands regeneration area more accessible,
>and curves gracefully right UNDER the Limehouse Basin. Whilst writing up
>the session for the Society of Cartographers Newsletter, I started
>musing on how the various mapping sites/agencies cope with showing the
>Limehouse Link. The Ordnance Survey are fine, as are Multimap and
>Streetmap (which both use proprietary UK mapping, from I think the same
>source). However, Google Maps, Map24, Yahoo and Maporama - which are
>mapping "remotely" - have more of a problem. Yahoo uses Navteq/TeleAtlas
>data and compromises by showing the Basin as smaller and implying that
>the link skirts the North side of it above ground. Map24 does the same
>thing, with a different style and symbolisation, as does Maporama. Live
>Local takes the Navteq data (showing the basin wrongly) and then
>symbolises the link as underground, with dashed lines. And if you want
>to see some really shoddy cartography then go in to max zoom in Live
>Local! However, the best of all is Google, which has it splendidly
>shooting across the basin, with no indication that it is underground.
>So could OSM do better? Well, leaving aside the fact that the Mapnik
>layer doesn't yet have tunnels sorted, I am not sure it could. On the
>Osmarender layer the nearby Rotherhithe tunnel is shown as a dashed line
>but is a straight line connecting the roads N and S of the river. From
>memory, the tunnel is in two bores, and curves in some kind of S shape
>and I think the bores separate from each other significantly. The
>Blackwall tunnel just east of here is not yet depicted, and the road
>stops just by the Millenium Dome to south, and also to the north of the
>river. But my actually point is, how can we get an accurate alignment
>for tunnels such as this? How would you survey the curving underground
>alignment of the Limehouse Link? I suspect even the best GPS handheld
>devices won't get a signal. You can't get any useful data from the Yahoo
>satelite imagery. You can't use any published map sources that you
>consider reliable as it is too new for o-o-cop maps.
>Any thoughts?

If SteveC's plans to get the UK road tunnels surveyed by others does not
reach fruition I shall personally make it my goal to get them surveyed with
traditional methods. As a tunnel engineer that's the least I can do for the



>Steve Chilton, Learning Support Fellow
>Learning and Technical Support Unit Manager
>School of Health and Social Sciences
>Middlesex University
>phone/fax: 020 8411 5355
>email: steve8 at mdx.ac.uk
>Chair of the Society of Cartographers:
>Mind the (Map) Gap:
>talk mailing list
>talk at openstreetmap.org

More information about the talk mailing list