Ground versus Grid (Land Surveying)

My personal feeling is that if you're doing your design survey in grid, you should do your engineering on grid, and your subsequent construction staking on grid.

If you want to use a ground-based coordinate system, ideally that decision is made up-front, then you can do everything in the ground-based system.

You can flip back and forth if you have to, but you eliminate a lot of potential for error if you can make a coordinate decision up-front and then stick with it for the duration of the project.

What you really should do is collect raw data in the native format of the tool used then process it in a Least Squares program of your choice (GPS simultaneous with conventional) then you can put out final coordinates in whatever flavor you want. Don't let data collectors and controllers mess with your raw data.

Forget this grid bullcrap and get up on the ground where you're working at. You guys are getting so confused because of all your new toys you're forgetting what you're doing and are further confusing the client and the public that you're providing your services to.

i am a fan of 'put it on grid and leave it on grid'. but i work small sites and scale factors don't mean much - normally.

whatever you do establish a standard and do it that way every time. leave yourself notes in the autocad file or somewhere in the project file so you can remember what you did and why.

A friend told me I should collect everything in grid then as i bring the data into CAD I should let the software "scale-up" the drawing so everything is ground.

That's the way I do it, however, there is no cut-and-dried method that will work for "everything." Every job you do needs to be evaluated on 1) what is the purpose, 2) what are the needs of the client, and 3) what system will meet those needs.

I'm a boundary surveyor. Boundaries are on the ground, so an SPC grid 4,000 to 10,000 feet below the boundaries isn't going to help me out much at all. In fact, it'd be a complete pain in the rear. That's why, everything I've done since starting business in 1988 is on a local ground system.

I've been using GPS since 1990. Incorporating that equipment required that I understand the operation of equipment, the operation of the data collector, and the operation of my CAD software. It also required me to play around with the different possible configurations in order to determine A) what each software system was doing, and B) whether or not it was doing it correctly.

After figuring it all out, I found that the best system for me (generally) was to first, predefine a local coordinate system for my project, and second, to know how to translate (not scale) my data from one system to another. Understanding the translation process enabled me to not have to worry about "grid," "ground," "localizing," "calibrating," or "scaling." You set up the transformation parameters and you forget about it (well, not really).

If you're confused about which way to go, it's likely because you haven't taken the time (yet) to figure out the equipment and to learn the transformation projection process. Study up on it, test it out, and decide which works best for the type of work YOU do.

JBS

--
*** May your boundaries fall in pleasant places *** Ps. 16:6
http://www.cplsinc.com/id1.html

Well, I cheat. When I do what you're wanting to do, I do it like this.

1. Collect everything on the grid. Never molest your coordinates.

2. Draw everything in grid, and I mean EVERYTHING except for annotation. (wait wait.)

3. Scale your LINEWORK ONLY up by the CSF.

4. Annotate your map.

5. Write your field notes.

Now, all of your coordinates are still state plane. Your map is surface. You have no problems.

Good luck dude.

--
"You don't have to be a good surveyor if you find all the corners."

Kris

Ground v Grid
Modified State Plane Coordinates - recently by Loyal

Surveyor 1: I always use grid
Surveyor 2: I always use ground
Surveyor 3: What's the difference between ground and grid?
Surveyor 4: What's a scale factor
Surveyor 5: What do you mean "grid factor?"
Engineer 1: My surveyor says ...
Engineer 2: My surveyor Says ...
Engineer 3: My Suveryor SAYS ...
Engineer 4: Where is the project?

ML Schumann: Personally, I wish State Plane Coordinates had never come into being. I have seen more money wasted because of the ensuing confusion than I would ever want to have to pay.

Here in Soz we have coordinated reference marks PMs all adjusted in MGA 94 coordinates and we still show on cadastral plans the traditional ground distances as the typical survey area is not affected by scale factor. For cadastral work I usually use a TS so have always adopted one MGA PMs coordinates then a second for line bearing orientation. Then being able to compute and apply the scale factor to the computed join between the two PMs to be able to compare with the distance measured in the field with the TS . (We must tie into a minimum of 3 PMs to ensure no problems with their location or if any moved.)

That said I work then solely in ground distances with one PM holding true coordinates, the bearings orientated to MGA. Meaning all coordinates are plane.

That said on larger jobs where more efficient to use combination of GPS and TS , like at Coober Pedy in outback Soz, where strangely there is an incredible overkill of coordinated PMs,I work in MGA 94 coordinates as the GPS work in grid. Liscad software enables me to initially set transformation, zone etc when starting a job so that all coordinates are MGA grid. However a Liscad software feature is that you can toggle to display and enter ground distances, with the software auto computing grid coordinates. So when you want a join between two points when display is awt ground distances then you can immediately compare to existing plan data.

Of course when it comes to plan printout and you want ground distances then one ensures distances are flagged ground and the cad dump will reflect just that.

When returning to the field to set out or locate further detail then if using GPS upload MGA coordinates, so I have immediate grid comparison check in field. If then using TS for set out I then apply the scale factor for the area to rescale coordinates to ground coordinates as they enter the TS from Liscad so that I can immediately determine field comparison of ground existing and measured distances. Up loading data from TS to a grid job data base in Liscad it auto converts all ground data to grid coordinates.

One needs to be aware at all times of distance type set as it is possible to flag grid and manually enter ground distances by computations. Then it requires rescaling those added coordinates using the correct scale factor.

So I guess it rally means you must have an understanding of what you are doing, how you collect the raw data and how the end data is being used that determines the method of working in plane, grid or a mixture.

RADU

--
<i>Value added surveying</i>

At least with respect to GPS, what you really collect is ECEF coordinates. This locates your position in 3D in a reference frame to the earth (requires proper location of base etc.) So beyond that what the so called grid coordinates are is some sort of projection of these coordinates onto a surface (the grid) that can be rolled out flat to produce a simple 2D system that is easy to get around in (simple math). Nothing really new here we been using various ways to look at a round type world on a flat surface for centuries.

So there is some complicated math going on to do this and in a past attempt to get around that various grid systems were invented and standardized. SPC's come to mind and were implemented over quite large areas. So to keep elevations positive they set the grid surface quite low, usually about sea level or below (you can't keep a grid surface on a constant elevation).

Now pretty much everyone has computers doing the math so the computations are really not an issue anymore other than when using them you should have an understanding of what is going on in the background. There is no need to have the grid down around sea level, you may set up a projection with a grid near the average elevation where you are working and doing so on a small area is easy do. You just need to learn and understand what is going on and how to do it.

Once you do that your grid can essentially be equal your ground. The trick to keeping it all straight is to publish your projection with your work so the next guy can do the same if they need to follow. All this should be a fundamental thing to surveyors and I believe it is being taught in the schools. Those that didn’t get it in school can learn it on their own with the required effort.

As far as converting back and forth between various projections, as long as your basic data is properly recorded in true ECEF coordinates, it's just a few key strokes on a computer. I don't scale it in CAD anymore, that's way to much trouble fraught with the potential for screwup. Make the conversion in a program designed to do that and using the basic ECEF data to work with. I use TGO but there are many such flavors to work with.

I live and work at about 6000 feet. My projects are small (no where near a 3rd of a state). Using SPC's just makes it a nightmare (grid elevation to far away, grid/ground too far apart, I like my bearings to be closer to geodetic, like the GLO). If someone insists on SPC's, it's only a few computer clicks away.

Okay...at the risk of making a fool out of myself (again), I am going to say a couple of things about GPS and Total Station similarities from a practical standpoint (or what do the two technologies actually “measure”).

First, for our purposes, GPS means Post Processed Static, Rapid Static, Kinematic, and/or RTK.

Second, Total Station means Robot, “Total Station,” theodolite + EDMI, or even transit and chain I suppose (when measuring slope distance from axis of transit to remote point, or even breaking horizontal chain for that matter).

Now I DON'T want to get into the argument about whether GPS “returns” a position or a vector (chicken or the egg), MOST folks realize that the vector scenario is the best representation of just what emerges from behind the wizard's curtain (with few exceptions).

OKAY...The first (and most dangerous) ASSUMPTION to be made, is that the field operator actually KNOWS how to properly use the equipment, but I willing to make that assumption for now.

In the case of the Total Station, this means that he/she is using a properly adjusted (and checked) instrument (and/or instruments), AND is making the proper allowances for temperature, barometric pressure, prism offset (or in the case of a chain/tape, temperature, calibration, tension & sag etc.) AND the proper UNITS are selected (US Survey Foot, International Foot, Meter) because ALL EDMI(s) measure internally in METERS.

In the case of GPS, he/she is NOT warping the GPS data around via a “calibration/localization.” This would have no effect on the raw data, but it could pollute the final results IF you don't go back to the RAW DATA.

In both cases we will assume that he/she can measure a proper/accurate HI/rod (something that I seem to have a LOT more trouble doing than I should).

NOW, the only real difference between the two techniques as far as Land Surveyors are (or should be) concerned, is that the Total Station requires TWO points from which to start from (origin w/coordinate estimates [Pt-1], and a backsight (azimuth mark) to align to [Pt-1a]), whereas GPS only needs ONE (a point with a coordinate estimate [Pt-1]). There is of course the canopy, EM field, and multipath issue with GPS, but Total Stations can have problems with local physical (and electromagnetic) conditions as well.

When all the dust settles, BOTH technologies basically “return” a direction and distance between TWO POINTS (Pt-1 and Pt-2). Now in the “raw form” this “vector” is expressed very differently by the two technologies.

In the case of the Total Station, we would be talking about an azimuth (or angle relative to the backsight) and a [slope] distance from the axis of the instrument (Pt-1), to the prism (or end of the tape) at Pt-2, along with a vertical angle (normal to the geoid). This information is then used to COMPUTE a new coordinate at Pt-2 (along with the HI and Rod height of course, because we are talking about a 3d world).

In the case of GPS, this vector is returned as dX/dY/dZ in meters (converted to feet in some cases) between the L1 phase center of Pt-1 and the L1 phase center at Pt-2, expressed in the geocentric reference frame of the ephemeris used (WGS84[G1150], IGS05, IGS08, or maybe even NAD83). This information is then used to COMPUTE a new coordinate at Pt-2 (along with the HI and Rod height of course, because we are talking about a 3d world).

What seems to plumb evade some (maybe many) folks, is that this whole “grid-v-ground” BS is the result of what happens to the raw data (both technologies) between its collection, and when the “results” get burped out on the screen. GPS does NOT “work on the [SPC] grid” UNLESS you tell it to (and even then, only AFTER the raw data is collected and manipulated). In fact, in it's rawest form, it doesn't even “work” on any particular ellipsoid!

Remember, the GPS and the Total Station BOTH “measure” a vector (direction and distance) more or less along the SURFACE of the Earth. The only real difference, is in how that vector is derived, AND the format in which it is expressed (in raw form).

I will repeat...GPS DOES NOT inherently “work on the grid” any more than the Total Station does. BOTH can be configured to do so, BUT that involves SOFTWARE in BOTH cases (various transformations and computations).

Not only that, but there is no such thing as “THE grid.” A “grid” (THE grid) can be ANY defined surface or 3 dimensional construct. The SPC Grid is NOT coincident with the UTM Grid, any more than it is with the “spherical” Latitude/Longitude “grid” (or graticule). THE Grid, is whatever YOU tell the software to project/transform the raw data ON TO or IN TO.

One more thing... “GPS Coordinate”

Wazzat?

I would argue (and it's a point that can be argued for sure), that a “GPS Coordinate” would look something like:

REF FRAME: ITRF00 (EPOCH:2011.2491)

X: -1849361.379(m) 0.010(m)
Y: -4498433.644(m) 0.020(m)
Z: 4114802.071(m) 0.029(m)

Although you CAN generate that with a Total Station too!

Loyal