What aspects of geodesy are important to you?

Getting paid.

> I imagine many of you encounter geodesy in your work. What do you consider the most important aspects?

I'm going to guess that you don't consider the map projections by which many land surveyors (myself included) use and respresent geodetic positions in everyday work to be geodesy proper. However, I think that competent use of standard map projections such as the SPCS is right at the top of my personal order of priorities.

In Texas, I consider it to be one of the fundamentals. So a surveyor or technician who doesn't have a functional understanding of how they are used is at a large practical disadvantage.

As far as geodesy goes, I'd say that those elements that pertain to the long-term maintenance of reproducible reference systems are most important to me as a land surveyor. That includes the details of how NAD83 or its successor will be realized and/or delivered to users, how to describe coordinate accuracy, and speculation about future trends that may pertain to surveying practice ten or twenty years from now.

I have managed to survive and sometimes prosper in land surveying with a pretty rudimentary knowledge of geodesy. That's why I would value such knowledge in a potential employee. I might only have a decade or less left in this profession, if I'm lucky. And I think I can probably get by in that period of time in my flat little world.

As time goes on, geodetic co-ordinates are becoming more and more popular and expected now and then. Still rarely, though. So far, I don't see the advantage to the end user in the small-scale projects that I am generally involved in but if only to avoid embarrassment, it's going to be necessary to provide my survey results in that kind of format, I suppose.

Howdy,

I do consider the SPCS to be important. Computations using the SPCS is usually covered in other course. The notion that SPCS and geodetic coordinates are unrelated is too common among students and some surveyors. As most conventional optical surveying provides 3D information, it is a shame that SPCS is 2D.

On a personal note, I used to work with Jim Stem (author of the SPCS manual NOS/NGS 5) who is also retired. The document is available here: www.ngs.noaa.gov/PUBS_LIB/ManualNOSNGS5.pdf The SPCS manual should be on every surveyor's bookshelf (or on their hard drive). Other than refreshing my memory in order to give presentations on the topic, I had not done any serious work with SPCS for decades.

As far as modern terrestrial reference frames, I agree that they are important. Anyone using an OPUS solution should understand time-dependent positioning. They should also understand why NAD83 CORS96 epoch 2002.0 coordinates differ from ITRF00 coordinates at a different epoch differ. They should understand the different types of heights and how they are derived and related. Statistics and their limitations are critical as well.

Even in a full semester length course, there are not enough hours to cover everything of interest.

Thanks for the response.

Cheers,

DMM

> So far, I don't see the advantage to the end user in the small-scale projects that I am generally involved in but if only to avoid embarrassment, it's going to be necessary to provide my survey results in that kind of format, I suppose.

Okay, I'll bite. You've been surveying boxloads of mining claims somewhere out in the hinterlands without bothering to put the whole works on some sort of geodetic coordinate system or grid coordinate system that is projected from geodetic coordinates? Why?

> As most conventional optical surveying provides 3D information, it is a shame that SPCS is 2D.

Well, we always do 3D surveys when we are combining GPS vectors with conventional measurements, both to get rigorous reductions of surface distances to the ellipsoid and as a condition for the simultaneous adjustment of the GPS vectors with the conventional measurements by least squares. The problem with reporting the height component as I see it is mainly that of third-party liability for misuse of elevations. My rule of thumb is that if you find yourself writing lengthy disclaimers, you may need to rethink what you're doing. Providing ellipsoid heights or NAVD88 elevations modeled from ellipsoid heights is a big problem in areas where flooding is an issue because Murphy's Law practically guarantees that they will be used inappropriately.

Since the 2D horizontal components of the SPCS describe the cadastrally relevant positions, the third dimension, height, is more of a potential bug than a feature.

I suppose 2/3 of a world coordinate is better than nothing. Assuming an elevation takes the risk out of doing a proper job. Is that an insurance companies idea?

2D coordinates really do suck! If yer lucky you'll be able to make another 10 years until retirement with out getting sued. Then the world can flip from flat to round, bout time since satellites been going round and round for 50 years or so. Stuck in a hundred year old time warp I suppose.

> I suppose 2/3 of a world coordinate is better than nothing. Assuming an elevation takes the risk out of doing a proper job. Is that an insurance companies idea?
>
> 2D coordinates really do suck! If yer lucky you'll be able to make another 10 year until retirement with out getting sued. Then the world can flip from flat to round, bout time since satellites been going round and round for 50 years or so. Stuck in a hundred year old time warp I suppose.

Well, that's disconnected from reality as usual, Leon, since land boundaries - you know, the things that land surveyors locate - are as a rule 2D entities. The height dimension is just hand-holding for those folks who don't have the means to determine the heights of things for themselves. Any *competent* surveyor won't be discomforted by not having the ellipsoid heights prechewed.

Yeah, but in 3D polar geometry you need the height from the center origin to calculate the distance between two points. I suppose you can fudge with some corny project wide combined average scale factor. But Geez most of us are working with GPS (ECEF) coordinates (3D for those of of touch with reality). So why cut off one leg of a three legged stool and pretend you're still doing competent work.

Maybe when you start throwing the word incompetent around you should avoid mirrors.

Also, part of the US is above 500 feet elevation. Competent measurements requires more than SPC's really provide. You really can't convert SPC's into world based 3D coordinates (to get to other useful systems) without the elevation. Most surveyors equipment these days collects the elevation. GPS won't work without a full 3D basic coordinate system. We can always hope to return to a chain and a compass I suppose.

Have a nice one!

Because nobody cares.

>
> Also, part of the US is above 500 feet elevation. Competent measurements requires more than SPC's really provide.

Really where?...j.k...I usually work in elevations from -3' to 400'. I also work on parcels sizes of <3 acres.

So I view GPS/geodosy a little different than you folks out west do...

> Because nobody cares.

I have to suspect that you haven't actually asked the right people. Being able to navigate to parcels and boundary corners with a handheld GPS or to plot them on georeferenced aerial imagery is a big deal that usually costs a surveyor nothing to make possible when GPS is being used on a project anyway. Adding value at a reasonable cost is a large part of what I thought professional practice was about.

> Yeah, but in 3D polar geometry you need the height from the center origin to calculate the distance between two points. I suppose you can fudge with some corny project wide combined average scale factor.

Well, the *competent* land surveyors I know wouldn't have any problem at all reproducing specific boundary corner positions described by just the horizontal components of the position, just the latitude and longitude. Naturally, the N and E values of the SPCS are exactly convertible to latitude and longitude. So what you're arguing is that you somehow can't do this, which frankly sounds at best unthoughtful and at worst, well ....

> Also, part of the US is above 500 feet elevation. Competent measurements requires more than SPC's really provide.

I get the idea that you have never actually thought about the problem. If the object of the exercise is to locate specific positions defined by the SPCS, the ellipsoid heights are merely an intermediate piece of information used in reducing distance measurements made at ground level to the ellipsoid. If a Combined Grid Factor varies by only a few parts per million across a survey and is provided, that can serve the identical purpose. Naturally, in a GPS-only solution, the ellipsoid height is a non-problem since if the WAAS-enabled autonomous height isn't good enough, a height from a differential solution from a CORS site will be.

For a larger project where the variation in Combined Grid Factor is large enough to warrant it, simply reciting the CGF at various monuments will accomplish the same thing, i.e. either tell the few surveyors with only conventional equipment how to reduce their surface measurements or provide a means of easily deriving the ellipsoid height of the point.

> You really can't convert SPC's into world based 3D coordinates (to get to other useful systems) without the elevation.

What your complaint really amounts to is wanting to have someone hold your hand while you gin up your own projection to keep from having to actually use a standard projection.

> Because nobody cares.

"Nobody" meaning "you", right?

You ever read Flatland? Oh, never mind, doubt you could comprehend Spaceland when firmly entrenched in Flatland.

Like I said before, two out of three ain't failing, barely passing.

With no need for geodesy (Flatlander), why even respond to the thread?

> Like I said before, two out of three ain't failing, barely passing.

Well, that's a genuinely idiotic way to start 2011. I'm embarrassed for you since you seem oblivious to the imbecility of your position.

We've been dicussing the problem of the 3D survey for quite a while. I'm working on a project right now that the client wants on the local SPCS. This area has a combined scale factor of 1.0007. Since it is partly a right-of-way project the 100' right of way needs to be shown on the plan as 99.93'. Or if the right-of-way is shown as 100' the ground distance is 100.07'. Not a big deal but that is quite a difference. The biggest problem with the job is that it is merging with a state parcel that was surveyed a few years before using surface distances and close to true north bearings. So I am showing a 2degree rotation and distances which are quite smaller.

I can see surveying going to a 3D based geodetic system. Heck I could do that right now; but imagine turning in a plat that way. Nothing would close for the reviewer-I guess I could say "trust me it's all good".

But SPC aren't the best solution either. My plats on the right-of-way project all have to be adjusted for area and I've got a boatload of info explaining everything on them (why I'm not matching the adjoining Certificate, why the areas won't calculate and how to adjust them, why the distances are so short, ect.).

Howdy,

FWIW, when I disparaged SPCS as 2D I did not acknowledge that the measurements used are inherently 3D. I do not doubt that accurate work can be done using SPCS coordinates when appropriate corrections are applied and careful work performed. Both NAD 83 and NAD 27 address the issue of heights in the context of corrections to the grid via the elevation and sea-level factors respectively.

The unfortunate aspect of the SPCS system is that however carefully the work is performed the outcome is a northing and easting coordinate with any height (in any system) loosely linked. As Mr. McMillan states, most of the work done to account for height differences is only used to correct distances to horizontal and then grid.

Decades ago when I worked with SPCS we would use reciprocal vertical angles, careful leveling of instruments and targets, careful measurement of instrument and reflector heights, attention to EDM corrections for curvature and refraction as well as atmospheric effects to determine heights in remote areas. Good results can be obtained over long (total) distances. Unfortunately, the effort involved is no longer cost effective.

Using GPS for the determination of heights is complex issue by itself. The deficiencies in tropospheric models and their impact on heighting remains problematic. OPUS-derived ellipsoid and NAVD88 heights are undoubtedly better than many other options especially with the improvements in geoid modeling (and the underlying data used in the modeling).

I am glad to read that 3D surface observations are being mixed with GPS vectors in adjustments. It is straightforward to do so. How important is it to practicing surveyors to understand the underlying mathematical relationships between LGH and ECEF systems?

My original post was intended to prompt conversation about what would be appropriate areas of study in a semester-length course in geodesy at the university level. Other ideas/suggestions are welcome.

Cheers,

DMM

I agree with Kent...THAT a better understanding and working knowledge of Mapping Projections is sorely needed within the profession in general.

However, I don't buy into the 2+1 dimensional paradigm that is (IMO) a conceptual holdover from the days of Transits/Chains (2 dimensional) + Leveling (1 dimensional). Modern Surveying really moved beyond that state of affairs back in the early 1970s when “low cost” EDMIs came into widespread use by Private Land Surveyors. The advent of GPS is now decades behind us, and that's a whole new ball game.

Personally...I think the very concept of a “stable” North American Plate is antiquated and somewhat disingenuous given our current knowledge of geodesy on a world-wide scale. I think we should move beyond that intellectual opiate into a more realistic (National) geodetic paradigm.

While it is TRUE that a good deal of CONUS is reasonably “stable” in relative terms, it is also true that even within these “stable” areas, there ARE varying degrees of intraplate crustal dynamics peeing in the spatial Wheaties.

Given the FACT that tweaking a “day of observation” ITRF2000 coordinate BACK to NAD83(CORS96) Epoch 2002.0000 uses the EXACT SAME velocity field model (currently HTDP v.3.0) as moving that same ITRF2000 coordinate to [say] ITRF2000 Epoch 2002.0000 (or 2011.0000 for that matter), the entire retrodiction/prediction process is still TOTALLY DEPENENT on the “accuracy” of the underlying velocity field model (which is internally expressed in ITRF2005).

If the NGS were to bag the whole NAD83 concept this winter, and publish the Multiyear Solution of the CORS in ITRF2008 Epoch 2010.0000, AND declare the NSRS to be coincident with that realization, then what would really be the problem?

Well, ALL of the published State Plane Coordinates on the CORS would CHANGE, BUT they going to change ANYWAY (just not as much). The National Grid Spatial Address' would also change, but again, they are changing anyway. Any existing NAD83xxxx coordinates could be transformed to ITRF2008 Epoch 2010.0000 with the SAME ACCURACY that they could be transformed to NAD83(CORS96a) or any other Datum for that matter.

When we go out and perform a X hour GPS observation, and compute that position using the National CORS (OPUS or otherwise), all we REALLY KNOW (to the extent that “know” anything), is where that point was THAT DAY! Retrodicting that position BACK to 2002.0000 amounts to simply applying a scientific model (that might might NOT even have a data point within 50-100 kilometers of ya) to extrapolate that position BACK to where it MIGHT HAVE BEEN 9 years ago. So we could go “back” to ITRF2000 Epoch 2002.0000 with just as much “accuracy” as we can go back to NAD83(CORS96) Epoch 2002.0000.

If we are going to hang our hat on retrodicted coordinate estimates (which I think is FINE [within reason]), then what is really the point of ADDING the additional transformation to NAD83xxxx? If the velocity field model is accurate, then it doesn't matter one iota WHEN you observe a given point, it would still transform BACK (or forward) to the same ITRFxxxx/NAD83xxxx Epoch whatever coordinate that was originally published.

I say...”publish” where your points are TODAY (ITRF2000, ITRF2005 or ITRF2008, Epoch 2011.xxxx) in geocentric Cartesian Coordinates (X,Y,Z in meters), and let the next guy use the LATEST velocity field model (HTDP/TDP v.X.x) to estimate where that point might be on THE DAY that he is THERE (in whatever ITRF realization is current/convenient for him).

Would all of this confuse some folks, well YEAH...but those folks are already confused, so what do we really have to lose.

As far as what Mapping Projection is used for the SURVEY/Plat/whatever goes...I DON'T Care, so long as you give me sufficient Metadata to follow in your footsteps.

Loyal