UTM and Grid/Ground

I use UTM for work in the USACE Pittsburgh District, it is all in one zone whereas there are at least six SPC zones (maybe seven or eight).

The drawback is that the grid scale factor is MUCH larger in UTM than in state plane, 1:2500 on the central meridian and decreasing from there. Of course the elevation factor is the same. As long as you are aware of the magnitude of the combined factor, it can be accounted for.

Have you though of using a LDP (Low Distortion Projection) which would minimize the scale factor for that area?

Allen Wrench, post: 416669, member: 6172 wrote: I'm setting up a new coordinate base map, ... 60 miles on each side. .... so SPC and county coordinate systems are out of the question..

I really see no reason that SP should be out of the question. So it's in the overlap between 2 zones. So what. Choose one. And it seems to me that a custom "county" zone would be 'da bomb for such a project except for the problem of communicating the parameters with users, who may be less sophisticated than you. You should do some testing of points for CSF and convergence before selecting a zone. UTMs cover large areas and therefore have scale factors that depart from 1 by much more than SP zones do, even at the edges.

Allen Wrench, post: 416669, member: 6172 wrote: I can do all my work in UTM coordinates, but when it comes time to publish a map, or stakeout ground distances, how do you figure out the scale factor?

You don't have to. You set your dc to the Zone you choose. It handles the scaling issues. But long before you get to that you would have some sort of adjustment report from whatever software you use that gives parameters for each and every point.

Allen Wrench, post: 416669, member: 6172 wrote: I'd like a single scale factor that I could use anywhere in that entire 60 mile square - it's relatively flat across.

You are going to have some big variation in CSF over a 60 mile area. That may not work very well.

John Hamilton, post: 416671, member: 640 wrote: I use UTM for work in the USACE Pittsburgh District, it is all in one zone whereas there are at least six SPC zones (maybe seven or eight).

The drawback is that the grid scale factor is MUCH larger in UTM than in state plane, 1:2500 on the central meridian and decreasing from there. Of course the elevation factor is the same. As long as you are aware of the magnitude of the combined factor, it can be accounted for.

Have you though of using a LDP (Low Distortion Projection) which would minimize the scale factor for that area?

The LDP is essentially what I'm trying to come up with, and that's pretty much what the county coordinate systems are - a system where grid = ground (roughly). I think the difference is maybe 0.30'-0.40' per mile with the county coordinates, whereas the with UTM is like 3.00' or more per mile.

[USER=9981]@Norman Oklahoma[/USER] gave you great advice. SPC zones are somewhat arbitrary, so no reason to exclude them just because a project crosses the published boundary of the zone. I'm with [USER=640]@John Hamilton[/USER] and Norman regarding LDP, I'm a big fan of LDP. Like Norman says, though, the main overhead with LDP is getting those projection parameters to end users.

What kind of vertical relief is there within the design area?

I also agree that you could just choose one of the state plane zones and stick with it, the scale factor would still be much less than UTM. I prefer to stay in SPCS for the same reason Shawn stated; not just getting the parameters of an LDP to end users, but also having to explain the basis for the coordinates to someone with no knowledge of geodesy.

All good advice, there likely isn't a perfect solution that will give you low distortion over the entire project.

SPC, UTM, LDP, all likely just a "local" case of either a TM or Lambert projection (I say likely, because there have been known to be bastardized LDP designs), almost all software is able to handle those and do the scale factor math behind the scenes. Beware that some software/data collections are NOT able to handle a single parallel Lambert, BUT if your project is square, then there probably isn't a preference for Lambert over a TM and I would go with a TM LDP if you decide to design your own. Besides large scale factors with UTM, you can end up with large convergence angles too depending on location in zone, those are likely to be mitigated quite a bit with a CM in middle of project unless the CM of the UTM zone splits your project.

SHG

A properly designed and implemented LDP should result in "grid-ground" combined scale factors around +/- about 14 ppm (0.07 ft. per mile), SUBJECT of course to height (elevation) variances within your projection. A 60 mile x 60 mile "zone" out here, could easily have 6-7 THOUSAND feet of relief, so the "projection induced scale factor" would be the least of your worries. Remember, ~21 feet of elevation change = ~ 1ppm of "elevation factor" change.

Loyal

It's deep subjects like this that helped me decide I needed to get back in college and further my studies. Watching this thread with great interest.:stakeout:

Shawn Billings, post: 416678, member: 6521 wrote: What kind of vertical relief is there within the design area?

Most of the elevations in the site would fall in a band of about 100' vertically - pretty flat.

I'm trying to avoid re-inventing projections, since I'll be dealing with 3 different county surveyors and who knows how many other private surveyors and government agencies. I don't really want to explain to them all the parameters of my home brew coordinate system.

Wouldn't the distortion issues of extreme CSF's affect mainly the lengths of lines on a drawing? For example, the XYZ inverse between two coordinates wouldn't match the ground length of a line drawn between them, but you would only have to multiply that inverse distance by a (consistent) scale factor to get it to match the ground distance label. That's the idea right? Or am I thinking about this wrong?

You are going to find it hard to get a good project combined scale factor to work on a 60 mile square. The purpose of the scaling true grid coordinates by a single CFS is to make them look closer to the actual ground distance. This only works in areas where the project CFS is close to the actual CFS at any given point in the subject area. To much vertical relief and your sea level correction is going to bite you. To big of an area and the scale factor will do you in (unless it is a linear project running parallel to the defining secant lines). If you do not want to develop a LDP then choose one of the zones, preferably the one with the central meridian of parallel closer to your project. Remember that state plane coordinate system boundaries are based on political boundaries. Just make sure to document what you used.

After seeing your comment of trying to work with others Lee D is absolutely correct. Take it from a GIS guy, an LDP will absolutely take effort in serious documentation and passing those coordinate files into your flavor of software. Know how to Modify your CSD (Trimble), PRJ's for ESRI transmission of data. You may know the LDP parameters but does the software off the shelf and those 3 surveyors. And in 10 more years. Are there any other Lambert conformal projections that Others have used? UTM and your stuck w meters. !

Sent from my iPhone using Tapatalk

UTM can be used with feet, although that to me is blasphemy. It is not restricted to meters only.

Thanks everyone for the comments. I love getting feedback from so many different points of view. I'm thinking maybe I should just keep using the county coordinate systems. It's three separate base maps, but it will be on a known LDP system built in to CAD, GIS, survey, etc. Everyone can relate to it without explanation - including the DOT's and county surveyors.

I would think that it's not that big of a deal. The adjustment shouldn't be effected by utm or SPC. And output projection is generally a check box.

Publish all in lat long, all SPC zones, and utm. Feet and meters. And your phone number.

Larry Scott, post: 417658, member: 8766 wrote: Publish all in lat long, all SPC zones, and utm. Feet and meters. And your phone number.

Actually, all joking aside, publishing in LLh isn't a bad idea...