[GRASSLIST:4615] A few more questions.

[Dave Graham]

Thanks, Victor -- you've answered some of my questions, and raised a few more.  If anyone else has any input on this, I'll gladly take it.

> I'm not sure that this would be all that beneficial.  Most states operate 
> in State Plane projection, so if you ever have to back-correlate your 
> field data to records (for something like survey boundaries or property 
> lines) you'll still have to end up converting it to State Plane anyway.  

Legal descriptions in Monroe County, Indiana are in metes and bounds -- generally with a Section corner as a point of beginning.  At this time, these Section corners do not have good state plane coordinates -- at least, the county surveyor hasn't made any available.  Of course, they could be scaled off USGS quad sheets with questionable accuracy.

Heres part of my problem (and I'm not a geodesist, so please bear with me:)  To get coordinates from State Plane to something comparable to a legal description (ground), you have to apply a sea-level correction and scale factor for the area in which you are working; To my understanding, this is basically because the State Plane is such a large area.  

Sometimes the combined sea-level/scale-factor correction is negligible, sometimes it isn't.  What I was wondering was if it would be possible to create a plane for a Section, apply an average sea-level/scale-factor correction to that plane, and come up with something close enough to ground coordinates within that plane that could then be used for staking property lines, laying out subdivisions, what-have-you.  

I suppose what I'm saying is that I'm not quite convinced that State Plane is a close enough projection for what I want to do (though it might be better than most.)

Theres a couple stoppers in my little scheme -- which you've pointed out.  The first, of course, is the oddball job that covers area in several Sections (let alone several Counties.)

> My experience with sections also is that they are very often not ortho-
> rectified, which means that if you set a particular corner at 0,0, you're 
> likely going to end up with negative coordinates along some edge or 
> other.  Most engineers frown on negative coordinates.

The second stopper has to do with what you're talking about here.  Most Sections are not, in fact, perfect squares, and I'm assuming that in an x,y "projection" you're always working with a rectangular surface -- which would potentially create negative coordinates (which I tend to agree, is probably a bad idea.)

Another thing, is that my Section-planes probably wouldn't be seamless.  If two adjacent planes had different elevations (which they probably would,) I'd have a mess -- especially when jobs crossed Section lines.  This is unless a plane was twisted like a potato chip; but thats more mathematically complex than I really want to get.

If I work in State Plane, do I have to tell GRASS that I'm working in x,y when I set up my location?  Indiana's state plane is broken in half -- so would that create two locations?  In UTM, however, the entire state falls in one zone -- but I'll still have to convert UTM coordinates to ground; do you have any idea how thats done?

> 
> Those are just my own thoughts based on my own limitations.  For 
> those who are math geniuses (unlike me), keeping track of hundreds 
> of coordinate systems probably isn't any sweat.  For me, it would keep 
> me up nights!

I certainly appreciate your input.  Right now I'm still in the process of learning and evaluating GIS, not simply GRASS -- that alone is keeping me up nights!

  - dave graham
-- 
Get your free email from www.linuxmail.org 


Powered by Outblaze