These come from the USGS's seamless imagery server. They are scanned and probably hand-georeferenced, so I'm not surprised they're not perfect.
I don't have an immediate solution for this, but I do have some contacts at the USGS that might be interested in making sure it's georeferenced correctly. Feel free to find other areas that have a similar problem. On Sun, Jul 15, 2012 at 1:13 PM, Apollinaris Schoell <[email protected]>wrote: > Ian, > > something is wrong with the USGS scanned topo map layer. At low zoom the > alignment is perfect and when zooming in it suddenly jumps approx 50-100m > east. I can see this in San Francisco and from there south to Mountain > View. It looks like an L-shaped area with SF at the north end. North and > East of SF is good. then extends further south-east to at least to Mountain > View but no more than Sunnyvale. > I have started to trace the areas but not yet finished. One place in SF > reminded me of the same problems on the terra server. Is this data a mirror > of the terra server? I know a couple other places to check but it will take > some time to verify. > For California and I think also for the rest of US there are far better > sources of scanned topo maps. Any chance to fix these? I can offer some > help to get the scanned maps and eventually with tiling too if there is > some documentation available. >
_______________________________________________ Talk-us mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk-us
