We can't both be right.

What are the differences between the leveling values and the GPS values?

Did they RTK it?

How long a level run, to what type of BM?

There is a certain amount of error to be expected in any operation.

Elevations on TBM's published to 0.01 for both, spot shots with 0.1 published values.

We can argue whether either one is really possible, but yes, they RTK'D it, and how closely we matched isn't the point.

I would love to do it their way, but my conscience won't let me, and they are obviously able to do it faster and cheaper.

Which way is right?

Idiots? Neither of you.

Depends on the accuracy your client needs.

The GPS guys have it easy, but your levels will be 'better'.

> Surveyors walking to site, and they are using network RTK to bring elevation to the site. I am working next door...
>
> I ran a loop from the nearest published bench mark.
>
> We are doing the same work, but which of us is an idiot?

Depends, how accurate do you need to be? How accurate do they need to be? We routinely bring elevations to project sites using RTK and we also bring elevations to sites by level loops. Neither is wrong, each project has it's own requirements.

Same requirements same specs... Same type of job.

I am not an absolute idiot, I know about apples, oranges, and even some grapes...

They just are willing to publish an elevation procured from a dance around the magic surveying pole. They publish what the box gives them.

I think it is great for them, but it isn't "best survey practices"...

why would using a different tool make one or both of you an idiot? The question is what is the necessary precision? Using RTK derived elevations makes sense in many applications.

Correction: my understanding is that it isn't normal acceptable practice.

My question is whether that has changed in most minds here.

I ran a loop from the nearest published bench mark.

Be careful of one loop from a single bench mark, much better to tie from one bench through project control to another bench.

Different methods does not necessarily mean either of you are idiots.

Ya don't really know til ya do both 🙂

The level itself will be more "right" and closer to the 0.01' for the local site to flow downhill. As for the BM, ok, it's a number, might fit with the other local BMs, depends who is publishing it when.

The RTK gets you an idea of ellipsoid height, geoid height, and how close it that to the published BM ? With how many redundant observations?

Then there's the Helmert heights, that little subdivision up the learning curve a ways ...

Years ago, Loyal suggested I keep a database of all 3 of those things for my stompin' grounds. Still grokking it ... a height cadaster, kinda.

(BTW the level wins. If you are transferring heights with RTK only it is my learned opinion that you are ... of lesser foresight. There. How's that for diplomacy? )

Probably soon to be pickled!

As long as they both work within what is required and pay the same, it don't matter.

0.02

I see your point, I see companies doing projects with what I call questionable procedures. The motivation for this is sometimes laziness but mostly I think is profit. Some surveyors would say – “ it’s at their own risk” and to some extent I see this mentality but in reality unless the work results in an immediate and obvious error that causes a problem to non-surveyors than it’s perfectly acceptable to unknowing clients. Even if the work is substandard and shouldn’t be acceptable, that only seems to be known by other surveyors. Trying to tell a potential client or even an existing one that brand x doesn’t follow best practices just results in blank looks and often dismissal because of assumed competitive business. This all seems to stem from the magic box; it could be argued that GPS is probably the biggest single detriment to the surveying profession.

I can't say using RTK GPS to get elevation to a site is wrong. Maybe they check into bench marks all the time with good results. Maybe they checked into the same benchmark you used. How do you know the benchmark you started from is any better than their RTK results?

Same thing that has been said before but GPS is a tool, it can be misused just like a level can be misused.

Same as a single observation sideshot.

Static benchmarks are disappearing, and few government entities are putting them back.
GPS maybe the most accurate way to establish an elevation. If you have to level a mile up hills (both ways 🙂 ) GPS maybe more accurate.

I use GPS RTN a lot to do flood certificates. I try and check onto a benchmark, but sometimes that's not possible, because there are none.

We can't both be right. YES YOU CAN!

Ok, I'm no expert, but if the other guys have established known points around the neighborhood and they check the official benchmark and everything is within acceptable tolerance for the job, what's wrong with that??

You obviously didn't see them checking BM's, they probably didn't see you running your loop!

jmo

> Surveyors walking to site, and they are using network RTK to bring elevation to the site. I am working next door...
What Geoid model do they have loaded in their dc? What are the multipath conditions?

> I ran a loop from the nearest published bench mark.
Is that BM correct? Did you include a second BM to prove the first is correct? Certainly levelling, especially closed loop digital levelling, is the more precise method for relative positioning. But not all BMs are perfect relative to the network, even if they once were.

> We are doing the same work, but which of us is an idiot?
If you follow the procedures outlined in the Bill Hemmings "User Guidelines for Single Base Real Time GNSS Positioning" you can establish control point elevations relative to other nearby points that will agree with digital levelling within a couple of hundreths. And if you establish your base elevation by redundant observations to CORS (typically via OPUS) you should have a good position relative to the network. A least squares analysis will give you an idea how good.

With network RTK you should be able to achieve similar results but it will take more than a few epochs. If you are collecting only a few epochs, as for topo points, you should expect elevations to vary +/-0.05 in perfect conditions, and probably at least twice that in the real world. Good procedure with the network might include checking in to local brass monuments, or, if none are available, establishing a local check in point using redundant OPUS (or, perhaps, AUSPOS) observations.

Why not? I think the better question is, who is in a better position to prove they are not wrong. A closed level loop to a single benchmark only proves that your elevation is true relative to that single benchmark. Meanwhile the RTK elevations are not closed and nothing more than a side shot snap shot in time, based on a geoid model that may be less than perfect. Without multiple benchmarks being RTK'd in at the beginning and end of the work, there really is no way to prove the quality and validity of the elevation spit out by the magical algorithms.