[GRASSLIST:4996] Re: US Census boundary file problem

GRASS GRASS user
1

I'm asking those who created the export files to look into this vector. I
had loaded the e00 into ArcMap, and it did not complain or show this
vector, but then, it's not needed, topologically, to build the polygon so
it seems Arc import just discarded it.

Charles Dingman, Geographer/301-457-4307/Rm 425, WP-1
Geography Division, US Census/Washington, DC 20233-7400

|---------+---------------------------->
| | Radim Blazek |
| | <blazek@itc.it> |
| | Sent by: |
| | owner-GRASSLIST@b|
| | aylor.edu |
| | |
| | |
| | 11/19/2002 04:37 |
| | AM |
| | |
|---------+---------------------------->
  >--------------------------------------------------------------------------------------------------------------------------|
  | |
  | To: Michael Ash <mash@econs.umass.edu> |
  | cc: <GRASSLIST@baylor.edu> |
  | Subject: [GRASSLIST:4993] Re: US Census boundary file problem |
  >--------------------------------------------------------------------------------------------------------------------------|

On Monday 18 November 2002 10:44 pm, Michael Ash wrote:

Using d.what.vect and moving due east from northern New
Mexico, U.S.A., I find New Mexico, Nothing where Texas
Should Be, and Oklahoma:

...

I think that the problem is an unclosed area.

I was curious where the problem is and I found dangle on the boundary:
http://mpa.itc.it/radim/texas.png

For next occasions and others:
m.in.e00 input=st99_d00.e00 action=vector
v.support -r map=st99_d00 err=err
v.out.ascii input=err output=err

dig_ascii/err:
A 2
26.25857424 -98.6253
26.25219177 -98.6253

and it is also in st99_d00.e00:
      2094 12234 2039 2040 196 196 2
-9.86252999999997E+01 2.62585742374075E+01
-9.86252999999975E+01 2.62521917668542E+01

so the problem is in st99_d00.e00. I think that such dangles are
allowed in Arc/Info, but as it is not present in st99_d00.shp.
I consider it to be an error in data. You can write to census.gov.

Solution:
v.rm.dangles input=st99_d00 output=us
cp dig_att/st99_d00 dig_att/us
cp dig_cats/st99_d00.name dig_cats/us
v.support -r map=us

d.what.vect ->
-98.94589247(E) 32.37349011(N)
us in us Area - Category 196 Texas

Radim

2

I would like to thank Radim Blazek of the GRASS project for
his extraordinary effort in helping me track down a problem
in this map AND for explaining how to assess and fix such
problems in general.

(1) Assessing the problem
(a) v.support with the err= option creates a vector map of
inputs that create problems; and
(b) v.out.ascii on the vector map output by the err= option
permits identification of the lines of data that created
problem; and
(c) v.support on the vector map output by the err= option
above permits visual inspection of the problem.

(2) Fixing the problem: v.rm.dangle removes dangles.

I would also like to thank Charles Dingman of the U.S.
Bureau of the Census for his assistance with this problem.
The support of the U.S. Census, for example, by making .e00
files available, is enormously important to my research and
my ability to pursue it in GRASS, and I am deeply grateful.

Sincerely,

Michael Ash, Assistant Professor
  of Economics and Public Policy
Department of Economics and CPPA
University of Massachusetts
Amherst, MA 01003
Tel 413-545-6329 Fax 413-545-2921
Email mash@econs.umass.edu
http://www-unix.oit.umass.edu/~maash