[Talk-us] Low-quality NHD imports

Christoph Hormann osm at imagico.de
Fri Oct 13 08:53:02 UTC 2017

On Friday 13 October 2017, Frederik Ramm wrote:
> I haven't researched who added them and when, but they would
> certainly not clear the quality standards we have for imports today.
> Most of this information can be properly modelled in usual OSM tags,
> and where it cannot, it probably shouldn't be in OSM in the first
> place.

Those are several regional imports made by different people, see 


to get an idea of the distribution.

In general the NHD water line features are next to impossible to 
properly model in OSM since the natural waterway lines (feature codes 
46000-46007) and the artificial waterway lines (feature codes 
33600-33603) include all sizes of waterways.  They were usually all 
imported as waterway=stream and waterway=ditch which is incorrect in 
many cases.

The geometries in the NHD source data are usually fine (at least in 
recent NHD versions) - though conflation is often poor, lacking 
connectivity to pre-existing data - like here:


And there are of course also obvious errors like this:


or this:


> Is there any systematic (or even sporadic) effort of cleaning up
> these old imports? Is there reason to believe that the neglect
> extends to more than just the tags - do geometry and topology usually
> work well on these, or are the funny tags a huge "this whole area
> hasn't had any love in a long time" sign?

I think this is probably a good example for imports discouraging manual 
mapping.  If this data was not there mappers would probably meanwhile 
have added at least the larger rivers but with the dense network of NHD 
geometries with a lot of cryptic tags and all flatly tagged as 
waterway=stream it is quite hard for mappers to identify the larger 
rivers and improve mapping there.  Like here (NHD import on the left, 
newer manual mapping on the right):


Christoph Hormann

More information about the Talk-us mailing list