[Talk-us] NHD data conversion

[Ben Supnik]

Hi Ben,

On 8/21/11 10:40 AM, Ben Miller wrote:
> Great! Thanks Ben.
>
> I downloaded the data for the two areas that I'm interested in
> (04060104 and 04060105) and stuck them in JOSM. I'm not sure I feel
> comfortable just dumping the whole thing in (especially if it might
> cause problems) so I was planning on doing it more or less item by
> item. Is there a flaw in that plan?

I don't think so...my hope was to get the data into a form editors could 
use, not to promote bulk importing; in some cases some but not all of 
the water data for a region may already be present from hand-mapping or 
other data sources, so merging is necessary.

> Also, is there an explanation somewhere of what the various files
> represent? XXX_nhdarhi0.xml, XXX_nhdflh0.xml, etc.

The names come right off of the NHD shapefile export.

http://nhd.usgs.gov/documentation.html

(That stuff gets pretty terrifying pretty quickly.  Basically the data 
is partitioned by data quality and topology type, so nhdarhi is NHD 
area, highest res data, fl is flow lines, etc.  The files are broken 
into sequence within a HUC to avoid any one file being too huge.

> And one question about methodology: There are a few larger lakes that
> were added as part of the PGS process. They appear to have been left
> pretty much untouched (except by me) and the NHD data is
> significantly more accurate. Would it be acceptable to replace the
> PGS ways with NHD ways, assuming I make sure to connect up any
> rivers, add them to appropriate relations, etc?

I can't comment on that - I'm not sure there is really a single "right 
thing" to do for OSM; others may at least have better informed opinions 
than I do. :-)

cheers
ben