Control problems?

I suspect that you're not the one with the errors, but whoever gave you the coordinates are.

Whenever I see the term "Modified State Plane", I get the heebee-jeebees. You never know what you're really getting when you see that, until you go out and do actual field work, and try to verify EVERYTHING.

It's especially bad if you see someone giving you "Modified State Plane", yet also giving you geoid heights, etc... That's a major sign that they didn't know what they were doing.

Sounds like they did the original points with a handheld GPS ?

Vern, you have a problem. And the problem SOUNDS like the person who did the prev work. I recommend you shoot all the control, 3x, at 3 different times of the day. And keep up with your pdop. Then, average the shots, so long as they are within tolerance. (about 0.05' and a max of 0.10' for x,y, and z should be about 2-3x that)

And, that you have a long and serious sit down with the guy you are working for.

Something is wrong, and it will be cheaper to fix it now.

Nate

Whatever happens; DON'T calibrate anything!

You shouldn't need to if the GPS control was done correctly, and if it hasn't been done correctly, then calibrating to it will just create a bigger mess.

I have found that most of the time control is stated to be on a state plane system or a local datum plane with the information to get you back to state plane, it is WRONG! Typically one can calibrate to the given control and it will fit together within its self (which is not true in your case). The worst part of it is that the company that provided control was probably time and material and you have a fixed contract. Make sure you get paid for lost time if the problem is not yours. Good luck! Jp

> Something is wrong, and it will be cheaper to fix it now.

Very wise advise.

I agree with the above replies with one exception, I wouldn't spend another dime of your budget until the author's of the control work are notified and you have recieve an appropriate response.

I would agree with the above comment, except I would definitely find out if I did anything wrong first. Nothing worse than telling everyone they are all screwed up, that you need more money to fix it, and then finding out you pushed the wrong button.

Don't do a lot of GPS myself. However, I've had to unravel one or two and have worked with some of the best GPS surveyors in my area. I agree with most of the advice above. I would add that one local county that published extensive control always tried to have two points in a larger group that could be "ground truthed" with conventional equipment. They then applied the combined factor (grid and elevation average of the two points) to the distance and check the GPS distance. If you can do that with one or two pairs of the points then compare it with your GPS and the plan coordinates computed distances you will have more confidence in your own work.

EDIT: "When I first started looking for points I plugged a couple of the WGS coordinates into my hand held for searching. It was off miles, not even on the site." In Ohio that type of error might indicate using the wrong zone. That would not account for the distance errors.

> When I first started looking for points I plugged a couple of the WGS coordinates into my hand held for searching. It was off miles, not even on the site.
>
As a ruff and dirty check to make yourself feel better about your work: plug a couple of YOUR new WGS coords into the hand held and I bet it takes you right to the points.

I ONLY feel comfortable with such a system if an LDP is created for the area, and then verified. These systems work so well that I feel very good about them, even when I know they exceed the vertical limitations of a CSF, since I know that in most areas, staying within 400' of vertical limitation will only amount to hundredths of a foot of horizontal error. Then tying it with a good Geoid model, you also get very good ellipsoid elevations.

Although that depends on the site... In some cases, you're perfectly fine using a CSF. You might also be perfectly fine using one of those nasty "Modified State Plane" coordinate systems. I suppose the most important thing is to just know what you're doing, what the ramifications are, and whether or not your screwing it up. Another implication can come from anyone else who has preceded you, and if they already screwed it up. Then you have to use a judgement call as to whether you "fix" everything, or follow the errors of the other person...

If Vern is working with panel points that implies aerial control. One of the jobs I got involved in had a GPS error in aerial control. The control was for aerial covering a several mile long corridor. The result was distorted aerial mapping on the order of one meter in some locations. The required correction was new coordinates on all panel points and redo the aerial mapping. This initially did not make the client happy until they realized the cost that a design error of this magnitude would represent. Murphy's Law applied and the area that was most distorted was also most critical to design.

Another nit pic, but it's my soap box - there's no such thing as "WGS Coordinates (NAD 83 (92)). You either have WGS 84 coordinates (in one of the 5 different iterations) or you have NAD 83 (1992). NAD 83 and WGS 84 positions will vary by about 1 m each in the horizontal and ellipsoid height. If we as surveyors can't get this right how can we expect the other great unwashed in the geosaptial community to figure it out.

:good:

I was following a large national mapping firm's work once on a DoD installation that had requested elevations based on EGM96 on the previous firms work.

The previous firm had simply taken NAD83(xx) ellipsoid heights and applied the EGM96 model to those, OOPS!!!

Unbelievable how many times I see Latitude/Longitude referred to as the WGS84 control when it is really NAD83. Part of this is due to most of the major software providers labeling their LLH coordinates as WGS84, BUT a surveyor should know better.

SHG

Like I said I am very new to GPS, I need a little help with the definitions of the acronyms LDP and CSF.

I will go out again tomorrow and do it all over in a brand new file just like I had never been there before. I will also set the total station up as suggested and shoot between all the inter-visible points. Being salaried and unaccountable has its advantages.B-)

Thanks for all the support and suggestions.

Since I have no budget constraints, I will do it all again from a different point and compare the results.

This is a "route" project, approximately 3 miles long with the outlying panel points being about 1500 feet off-site both directions. My thought is that I want to set my own control points along the route where they will more accessible, Would using an assumed local coordinate system and tie all theirs in be a good idea or not?

I will also do as much as possible with the total station to confirm the GPS distances.

This would have been a great project to get my Trimble training on. 😉

LDP = Low Distortion Projection.

It's basically the same thing as a grid system like State Plane or UTM, but you typically center your projection on your site (although not always), and you scale up the ellipsoid to your average project elevation so that when you inverse between your grid coordinates, you get pretty much the exact same thing as you ground distances. And since it's just another coordinate system, you can always reproject anything in your LDP into something like State Plane or UTM, or vice-versa. So you can bring in aerial imagery or GIS shapefiles, and they fall right into place. Yet you can still inverse between coordinates and get ground distances. At least, as long as you have maybe 400' maximum of vertical difference over your project. Once your vertical distances start getting larger than that, you can start to see error in your horizontal distances. But since you're actually using a "real" coordinate system with a floating scale factor, the same as with something like State Plane or UTM, you get much tighter results, even if you go outside your elevation range. If you have a 2000' hill in the middle of your project, you may not be able to inverse between coordinates and get an exact ground distance, but your coordinates will all be correct, and can always be converted to any other coordinate system (such as State Plane or UTM).

On the other hand, if you use a CSF (Combined Scale Factor), then your error increases as you get further from your basis point, and also as your elevation difference changes. It can work fine for small sites with little elevation change - actually, it can work for relatively good-size sites, I tend to say anything smaller than 6 miles across and with less than 400' elevation change - but has other problems. First is that such a system often involves multiplying/dividing the coordinates by the CSF to convert between "Project Coordinates" and "State Plane Coordinates". If you're smart, you also offset your project coordinates (often called "truncated" coordinates, although this term isn't particularly accurate), so it's easy to tell the difference between the two sets of coordinates. (This is the system lots of people like to call "Modified State Plane", a term I personally abhor.) Not only is this system weaker than an LDP, working over a more-limited range, but it often also provides problems when trying to integrate things like aerial imagery and GIS data. It also can lead to SEVERE problems when trying to integrate many surveys over a large range, such as a highway corridor, canal, pipeline, or utility line that covers a large distance. Especially if anyone in the process is confused by the term "Modified State Plane", and thinks all surveys in the area are on the same coordinate system. Copy-and-paste or XREF do not work in this sort of situation. We ran into this situation while staking a power plant, and the previous surveyors did the overlot grading in the wrong spot by 600'x400' because they messed all this up. Rather than regrade the whole site (which would have been VERY expensive), we built the plant on the pre-graded pad and jogged the major transition lines by the 600'x400' as they came into the site. The owner was not happy about the whole thing, but the damage had already been done by the time we got involved, and they were happy that we managed to solve the problem with minimal extra expense, even though the end result varied from the original intent.

sinc...

You stated above:

“you scale up the ellipsoid to your average project elevation...”

Did you by chance really mean to say:

“you scale up the DEVELOPED SURFACE (of the LDP) to your average project elevation...”

It might seem like a minor detail, but in reality, it is a MAJOR difference. If you SCALE the ellipsoid, then by definition, you are creating a new DATUM, and no longer playing in NAD83. That would then require additional metadata to get from your “custom ellipsoid/datum” to NAD83 (and back).

Loyal