[GRASSLIST:4575] RE: A million and a half questions.
John Gillette
JGillette at rfmd.com
Tue Sep 24 11:49:54 EDT 2002
> > For starters, I'm wondering if there are any other
> surveyors on this list
> using GRASS or WinGRASS who might give me testimonials -- or
> surveyors in
> states with a public land survey system (Townships and
> Sections) that might
> give me tips.
I am not a surveyor. However, I am looking at cadastral mapping
in GRASS in NC where the "metes and bounds" system is used.
> >
> > I'm fairly familiar with Linux/Unix -- though my boss and coworkers
> aren't, which makes WinGRASS a reasonable compromise.
> (Though I haven't
> ruled out the possibility of building a Linux box for the
> office network,
> which is currently running Win2k.)
I use a Sun workstation with a samba connection on the network
to be able to transfer information back and forth to Windows 2000.
This is a separate issue, though. I don't have experience with
the windows version of GRASS.
> > I'm considering creating a GIS in which each Section is a
> plane; land
> surveys in that Section will accordingly have x,y coordinates
> within that
> plane. This would, of course, create a gigantic soccer-ball of
> Section-Planes. Say then, that I acquired a huge area of data
> that spanned
> multiple sections in a different projection, like UTM or
> State Plane -- am I
> going to be able to use the two together without a lot of
> hairy conversions?
You will have to do conversions. Whether they are "hairy"
or not, I am not sure. GRASS has no built in conversions for
vectors from X,Y to other projections. You will have to be able
to convert the x,y data to some other geo-referenced coordinate
system. I am looking at state plane coordinates. You can export
the data to an ASCII file and then write a script (using GAWK
for example) to add an offset and/or rotation to the X,Y data
and then import the data back into the geo-referenced location.
So, for example, I have a plan to use the metes and bounds description
to create a drawing with some (free) cad program with the origin at
a know reference point. The survey starts at the intersection of 2
roads that I know (or can find) the state plane coordinates for.
I export the data in dxf to an X,Y location in GRASS. I export
the X,Y data in ASCII to a file.
I then rotate the data for declination with respect to the state
plane and offset the data to correspond the coordinates of the
road intersection. I then import the new ASCII data into GRASS,
this time in a location which is in state plane projection.
> > Also, how would something like this be set up in GRASS?
> Would I have to
> create a separate location for each Section, giving it an x,y
> projection --
> or would I be able to get by making a location for each
> Township, with each
> Section as a mapset?
X,Y is not a projection. Every X,Y location would be the same
in the sense that it is just a simple X,Y coordinate system.
You could put all the X,Y data in 1 location if you want.
This would simplify looking at the data because you have to
restart GRASS to change locations. (Or at least you did,
this has changed, I think.) You could create different mapsets
for different Townships or sections. The problem is that you are
going to manually have to keep track of the rotation/offset
for a given set of data.
Once in a geo-referenced location, you can re-project the data to
any supported projection and combine it with an other vector or raster
data.
(Note that I say above that I have a "plan to ..." I haven't
actually done this yet. I am still gathering metes and bounds
data and looking into an appropriate ground reference. I have talked
to the city GIS department and the method should work in principal.
The city and county do this all the time. HOWEVER YOU NEED TO CHECK
ON THE ACCURACY AND YOUR REQUIREMENTS WITH RESPECT TO LEGAL STUFF.
I.E. YOU PROBABLLY DON'T WANT TO REPRESENT THE DATA AS BEING OF
LEGAL ACCURACY BUT ONLY FOR DISPLAY PURPOSES, ETC.)
My application is the display of cadastral data on a DEM (both 2D and
3D using NVIZ and display of cadastral data on color infrared aerial
photography.
Hope this helps.
John Gillette
North Carolina
More information about the grass-user
mailing list