Vertical time dependent positioning

I don't think they are there, since they have tried to simulate NAVD88 there should be no time dependent vertical changes, however, we all know this isn't the case. 🙁

I see it jumping quite a bit between 99, 03, 09 and 12, but I think you are best to hold NAVD88 and not chase the verticals.

I just ran a test point through HTDP and it did change the ellipsoid height, but only by 4mm, which doesn't seem like much considering the location.

Since you are in LA how are the verticals for stability? Do you get a lot of subsidence?

4mm seems almost nothing, about .01'. Did they also calculate the "latest" ortho elevation using the GEOID for the time frame?

I see a train wreck coming here with many users when the "new" elevations are released.

I hope to be retired when the new horizontal and vertical datum is released...

Yes, South Louisiana has major subsidence issues, and the first train wreck already occurred. In 2005, less than three months after Hurricanes Katrina and Rita created a need for everything south of I-10 to be resurveyed, NGS revised Geoid03 to reflect the effect of subsidence and invalidated 90% of the vertical benchmarks in that region. Some of the benchmarks that were used to perform the survey leading up to that event had their elevations adjusted by over a foot - which is pretty significant this close to sea level, obviously.

At that time, our VRS network was only used for research and to provide CORS data. Since then it has expanded into a statewide RTN and is even mentioned in the statute concerning vertical control - http://c4gnet.lsu.edu/c4g/index.php?option=com_content&view=article&id=95:la-r-s-50-173-1-vertical-control-standards&catid=21&Itemid=198

NGS does not have a VTDP tool at this time. This is an area that was getting considerable attention and then several retirements that hit NGS a couple of years ago put a hole in that effort. At last weeks Geospatial Summit, Dr. Dru Smith ([email protected]), NGS Chief Geodesist did indicate that this was on their "honey-do" list. Drop him an e-mail and see where it stands.

Will do thanks.

HTDP & phantom vertical movement

Lee,

The “0.004 m” that you are (probably) seeing in (I assume) a NAD83(2011) Epoch XXXX.xxxx to [say] NAD83(2011) Epoch 2010.0000, is in fact a computational “artifact” related to the 14 parameter transformation between ITRF2008 and NAD83(2011).

The “short (less than exact) version" is something like this:

The HTDP velocity model is expressed in ITRF2008. ALL vertical velocities are SET to ZERO (relative to ITRF2008).

HTDP transforms a NAD83(2011) epoch 2015.1234 to ITRF2008 Epoch 2015.1234, and computes a ITRF2008 velocity from the velocity field model (which is of course expressed in ITRF2008 with ZERO vertical velocity relative to ITRF2008).

It then “moves” the ITRF2008 Epoch 2015.1234 to ITRF2008 Epoch 2010.0000 using the modeled ITRF2008 velocity (NO Vertical movement relative to ITRF2008).

It then transforms the ITRF2008 Epoch 2010.0000 to NAD83(2011) Epoch 2010.0000 (or between any other input Epochs).

HOWEVER...if you look at the ITRF2008-NAD83(2011) transformation values, you will see that:

1. The geocentric origin of NAD83(2010) “moves” relative to the ITRF2008 geocentric origin as a function of TIME (parameters 8,9,10).

2. The ROTATION between frames also varies across TIME (parameters 11,12,13).

3. The SCALE variance between ITRF2008 and NAD83(2011) ALSO CHANGES as a function of TIME (parameter 14).

SO...a “point” that is NOT moving vertically RELATIVE to ITRF2008, “APPEARS” (is in a mathematical sense based on the 14 parameter transformation), to be MOVING VERTICALLY in NAD83(2011). The magnitude of this “movement” is not only a function of TIME (duration between epochs), but also (to some degree) a function of WHERE you are in North America.

In essence, a POINT with ZERO vertical velocity relative to ITRF2008, Is (or would be) MOVING VERTICALLY relative to NAD83(2011).

This is kind of a Catch-22...

For more information, see:

http://www.ngs.noaa.gov/TOOLS/Htdp/Pearson_Snay_2012.pdf

Some of the benchmarks that were used to perform the survey leading up to that event had their elevations adjusted by over a foot

You have subsidence, I've got tight Geoid contours, in both cases GPS derived elevations have probably struggled to model the "real" world.

Over time it's been a bit like a bouncing ball, I made the good (lucky) decision not to try catching it some time ago and have stuck with NAVD88, hopefully the new gravity models will "fix" some of the issues.

And also hopefully people won't try and layer the new elevations over the NAVD88 elevations.

Subsidence is our challenge but there is something to be said for surveying at or near sea level. I've never had to deal with the challenges that mountainous terrain would present.

HTDP & phantom vertical movement

As mentioned previously, NGS does not yet have a VTDP tool. That being said you can estimate the vertical velocities in any area by looking at the "COMPUTED" velocity values on any CORS datasheet. For example the NAD 83 (2011) value at station at LSU - 1LSU (DF5754) is - 0.0040 m p/yr. Depending on where your project is relative to any set of stations and ignoring those that have PREDICTED velocities you can average the values in a reasonable way and enter those along with the average lateral values into HTDP. I suspect this will provide as good an estimate as anyone can reasonably expect until a more rigorous model is available from NGS. -- You can see an example -- 1LSU

HTDP & phantom vertical movement

Looking at the position and velocities for ENG6, the computed NAD83 vertical velocity is -0.0040m/year. http://www.ngs.noaa.gov/cgi-cors/corsage_2.prl

Is this number accounting for subsidence as well as tectonic movement? I seem to remember Cliff Mugnier saying the subsidence rate out there was closer to 1 cm.

HTDP & phantom vertical movement

And now you see some of the issues with building a VTDP tool - what are the real subsidence rates. Vertical movements are caused by a number of conditions - tectonic motions (e.g. earthquakes, post-glacial isostatic adjustment) as well as sediment loading and subsurface fluid withdrawal. I'm certainly no Roy Dokka (blessings and peace be upon him) but I understand that the last two are the greatest contributing factors to vertical change in the Gulf Coast area. If Cliff and the LA Center for GeoInfomatics have values that they say should be used then I would guess that's what you need to follow. If those values differ significantly from what NGS publishes on a nearby CORS then I would suggest that due diligence says you should contact Cliff and find out the rational for those differences. I did not spend too much time on this particular topic during my tenure at NGS but in my humble opinion the only way to get a really good handle on subsidence rates is either by systematic periodic releveling (regrettably very expensive and time consuming) or building a CORS network something like the one developed by the Harris-Gavleston Subsidence District - which takes time and resources.

While it's a bit hard to read in the map provided, there is a subsidence rate map for southern LA in the 1973(!) paper "Regional Investigations of Vertical Crustal Movements in the U.S. Using Precise Releveling and Mareograph Data." by S. Holdahl and N. Morrison -- Holdhal-Morrison
Certainly this investigation is dated but it does compare closely with the COMPUTED values of the CORS in the NGS network.

HTDP & phantom vertical movement

Thanks for the information. I miss Roy, he was both a great resource and a great guy.

Don't forget Shinkle and Dokka's NOAA TR 50

Don't forget the Kurt Shinkle, Roy Dokka NOAA Technical Report:

Rates of Vertical Displacement at Benchmarks in the Lower Mississippi Valley and the Northern Gulf Coast
Shinkle, K.D., R.K. Dokka, 2004, NOAA Technical Memorandum NOS NGS 50: Rates of Vertical Displacement at Benchmarks in the Lower Mississippi Valley and the Northern Gulf Coast, Silver Spring, MD, National Geodetic Survey.

http://www.ngs.noaa.gov/heightmod/NOAANOSNGSTR50.pdf

Lots of good info on subsidence monitoring and modeling is available. Expanding these efforts from local to regional or state size areas is problematic. Does the velocity derived from lengthy data sets at a geologically stable site (like an extensometer) reflect changes to less massive or deeply set monuments? Of course there is also uplift. Considering areas like S Louisiana where there is not only dewatering due to the levee systems, coastal faulting, and mass accumulation from transported sediments there are changes in mass and therefore the geoid...

Wow I see many PHD topics (but will there be the funding?).

Cheers,

DMM

Don't forget Shinkle and Dokka's NOAA TR 50

Thank you!

Don't forget Shinkle and Dokka's NOAA TR 50

Wow I see many PHD topics (but will there be the funding?).

It would be cool to see some PHD's out running long level loops. They could get some really good useful data;-)