[Imports] US Forest Service Data
tgis at bendbroadband.com
Mon Feb 7 04:41:05 GMT 2011
Thanks for the various responses to the question about importing Forest Service Data.
My impression is that the Forest Service data can come from different sources and often is listed in the attribute information. In the case of trails some maybe GPSed some may have been digitized off of DRGs (digital topo quads) etc. I asked someone from the Umpqua NF about trails there and that is what they said. In the case of the roads in the middle of nowhere it may be in fact TIGER roads that the FS used. In that regard probably not all national forests are the same. In the case of western Oregon there are millions of miles of logging roads that the Forest Service probably had to map out themselves and constantly up date out of necessity in order to get all those trees "harvested". Now the task for them is deleting roads as they decommission them. There is little road building going on now due to how little logging activity there is in previously untouched areas. National Forests with less logging activity like say Inyo NF in the Eastern Sierras probably had different needs and methods for mapping roads. If you just have a road that goes up some canyon to a mine you can draw that out on a paper napkin and it will work. That is just an educated guess.
TGIS at BendBroadband.com
From: Samat K Jain [mailto:lists at samat.org]
Sent: Friday, February 04, 2011 1:29 PM
To: imports at openstreetmap.org
Subject: Re: [Imports] US Forest Service Data
On Thursday, February 03, 2011 07:17:46 PM Tom Ponte wrote:
> The FS data is much better than the tiger data spatially and it has
> the road numbers which correspond to the signs the Forest Service uses
> for signage unless they are shot to pieces by second amendment
> intellectuals. I moved a few roads into place with the imagery west
> of town but quickly decided this would be a large undertaking
> especially adding road names. It appears that no one has touched the
> tiger data in the area of the FS land. I am guessing that is the case
> for much of the west but I have not checked that out. So if it has not
> been touched then importing new data would no doubt be a lot easier
> that if it had. Many of the existing roads showing in OSM with the tiger data have been decommissioned.
I'm the OSM mapper that's been (slowly) importing FS trails into New Mexico.
One thing pointing out is how wildly inconsistent TIGER data is—but you probably know this already. In some areas, TIGER appears to have used forest service data for roads (and is fairly accurate); in others, the TIGER dataset is divergent (and inaccurate).
The only resemblance of consistency I've found is that forest roads in the middle of nowhere appear to be identical in TIGER and FS data; roads in/near urban areas are divergent. I suppose this makes sense: urban area roads with residences/etc are likely to have things that need be monitored by the Census, so the US Census keeps their own, separate data.
> But before I get too far along with that is there any talk of updating
> these roads with newer data like the newer tiger data -(for all forest
> service districts)? I haven't looked into how much of that there is in
> the areas managed by the FS to see it is available or if it is any
> better - as good as etc. I sort of doubt it is better or as good
> unless it is now derived from FS data somehow. I am not an expert in
> all things tiger in case you couldn't tell.
There's been a *lot* of talk, but no solutions. Updating/replacing imported data is a major unsolved problem with OpenStreetMap.
My strategy with New Mexico is to just fix things, manually, as I see/survey them. Fortunately, (most) TIGER data contains forest service road numbers so most of the work required is just deleting nonexistent roads and fixing topology.
Samat K Jain <http://samat.org/> | GPG: 0x4A456FBA
What is wanted is not the will to believe, but the will to find out, which is the exact opposite.
-- Bertrand Russell (555)
More information about the Imports