[GRASSLIST:3089] Re: TIGER/Line Data

[Eric G. Miller]

On Sun, 3 Feb 2002 21:27:11 -0500, Ben Logan <ben at wblogan.net> wrote:

> On Sun, Feb 03, 2002 at 01:30:33PM -0700, Roger Miller wrote:
> > On Sunday 03 February 2002 12:50, you wrote:
> > 
> > > Does GRASS support the TIGER/Line data format used by the Census
> > > Bureau?  I looked on the GRASS homepage, but couldn't find anything
> > > about it.
> > 
> > I did this recently with Tiger 2K county data.  I used v.in.tig.basic.  I got 
> > the lines themselves, but I didn't get the topology or the attributes.
> 
> I created a new GRASS location using lat/long as a coord system, grs80
> as the spheroid, and nad83 as the datum.  I used the default region
> and figured I could always change that later.  I also created a mapset
> called "carroll" (the name of my county) to hold the maps.
> 
> I then downloaded the TIGER/Line 1997 file of my county (tgr51035.zip)
> and unzipped it into a directory.  Here's the command I used to import
> them:
> 
> v.in.tig.basic t1=TGR51035.RT1 t2=TGR51035.RT2 out=carroll spheroid=grs80
> 
> (I assumed that I didn't need to use the 'zone' option since my
> location uses lat/long.)
> 
> It ran for awhile and didn't produce any errors.  I then set the
> region:
> 
> g.region vect=carroll
> 
> And used d.vect to display the map.  Nothing showed up, though, until
> I ran v.support on the new map.
> 
> Even now, the map looks slightly distorted (could be my imagination
> though) and when I try to zoom in with d.zoom, it won't let me get
> close at all.  Did I do something wrong with the coordinate system?

I'm thinking TIGER data is in state plane.  But note, the state plane
stuff is hardcoded to the old NAD27, whereas you'd probably want the
NAD83.  It's a simple matter to rename some files in grass5/etc/.
Move state27 to state27.bak and rename state83 to state27 before
creating a your locations (or rerunning g.setproj).  It'll act
like it's doing state plane 27, but will use the 83 parameters.
Note: contrary to the original intent, some states still use feet
for state83.  I'm not sure if the TIGER data respects that, or 
uses meters for all.

Each county potentially has a different projection (usually lambert
conformal conic or transverse mercator).

-- 
Eric G. Miller <egm2 at jps.net>