If you do several GPS on Bench Mark Observations on this point and the data get's used in the next Geoid Model, it seems to me like your GPS elevation for any new observations, with the new Geoid Model would be relatively close to the published elevation. I like to find 2 marks on the same line and do simultanious observations on them, if I can't run a check line between them with a level.
Here are some photos of N 41 that I found on the Goe Cache Website. It may be suitable for GPS Observations.
ORTHO HGT: 3094.281(m) 0.020(m) [NAVD88 (Computed using GEOID12B)]
KN0233* NAVD 88 ORTHO HEIGHT - 3094.199 (meters) 10151.55 (feet) ADJUSTED
Leon's result was ABOVE the data sheet value, which does not argue for downward impact or settling, but could perhaps be consistent with slight frost heave.
I don't know how deep they put the posts in his area, but I've seen some around here ripped out by construction, etc., that weren't long enough, after allowing for reported projection above ground, as to put the bottom below the local maximum frost depth.
It looks like N 41 has a lot of concrete or plaster around it. Is that common for disks in bedrock? Did somebody cover up cracks in the rock to help preserve it? Or does it suggest that somebody found the disk loose and "fixed" it?
Bill93, post: 331858, member: 87 wrote: Leon's result was ABOVE the data sheet value, which does not argue for downward impact or settling, but could perhaps be consistent with slight frost heave.
I don't know how deep they put the posts in his area, but I've seen some around here ripped out by construction, etc., that weren't long enough, after allowing for reported projection above ground, as to put the bottom below the local maximum frost depth.
I wouldn't be surprised to find out the whole mountain range has moved up or down 8 cm since 1934. But until we have many years of GPS Height measurements on stable marks (like CORS GPS stations) we just don't know.
As far as my project goes I'm leaving everything on GEOID12B (no vertical adjustment of my data). It's only the relative difference in elevation of the ends of the tunnel (slope for open channel flow in a pipe) that matters. So distortion of the GEOID Model is of some concern but probably doesn't really matter.
After thinking about it, if a level loop is run I'll probably go through the tunnel (7130 feet long and 23 feet of drop) instead of over the top of the mountain (maybe 3 miles in length and over 500 feet change in elevation up and then back down). My best guess is the engineer is going to be fine with my GEIOD12B data. I'll suggest he look at his flow calculations at say 22 and also 24 feet of drop and see if it really makes and critical difference. If it does (it won't) well, we will level through the tunnel.
GeeOddMike, post: 331818, member: 677 wrote: Not to re-irritate the issue, I still feel that the monument shows signs of damage which could have impacted its height. There have been eighty years since the work was done.
Unfortunately, NGS policies on accepting level data requires that work done to replace, update or verify the published values require that second-order leveling be performed and all the associated paperwork. Not likely in this or many similar cases.
Getting back to the point of my earliest response, I still think there are issues with the determination of orthometric heights at these high elevations. Prof W. Featherstone of Curtin University in Australia has a very interesting paper on his web page on this issue. It is here: http://espace.library.curtin.edu.au/R?func=dbin-jump-full&local_base=gen01-era02&object_id=158292
I also point out the difficulty of geoid modeling in these high mountainous areas. As you mention there are large geoid slopes. Gravity reductions given lateral mass density variations and the like remain problematic.
Enjoy the paper. I especially recommend looking at his other papers. Looks of good stuff and clearly written as well.
Is there any way to download this paper without paying a hefty fee? It looks quite interesting...
I used his link and found a PDF icon to download it for free. If you follow the further links on that page, then you get into the subscription services.
Bill93, post: 331898, member: 87 wrote: I used his link and found a PDF icon to download it for free. If you follow the further links on that page, then you get into the subscription services.
Thank you! If I had only read the first line!!!
A great thread! Thanks to all.
LRDay, post: 331884, member: 571 wrote: I wouldn't be surprised to find out the whole mountain range has moved up or down 8 cm since 1934. But until we have many years of GPS Height measurements on stable marks (like CORS GPS stations) we just don't know.
As far as my project goes I'm leaving everything on GEOID12B (no vertical adjustment of my data). It's only the relative difference in elevation of the ends of the tunnel (slope for open channel flow in a pipe) that matters. So distortion of the GEOID Model is of some concern but probably doesn't really matter.
After thinking about it, if a level loop is run I'll probably go through the tunnel (7130 feet long and 23 feet of drop) instead of over the top of the mountain (maybe 3 miles in length and over 500 feet change in elevation up and then back down). My best guess is the engineer is going to be fine with my GEIOD12B data. I'll suggest he look at his flow calculations at say 22 and also 24 feet of drop and see if it really makes and critical difference. If it does (it won't) well, we will level through the tunnel.
Heck, 8cm is pretty good I think, but even if the mountain rose or sank the newest Geoid model should reflect it, after all, that's the whole point of them, to match existing conditions.
I think running a level loop is a good idea.
7130 feet and 23 feet of drop isn't too bad, shouldn't take all that long if they will pay for it.
And if the elevations don't need to be all that tight consider running the levels and check GPS points along the way, if they are checking to the geoid model corrected GPS numbers, then you might be good with a one way run.
MightyMoe, post: 331949, member: 700 wrote: even if the mountain rose or sank the newest Geoid model should reflect it, after all, that's the whole point of them, to match existing conditions.
If the mountain moved some small amount, it wouldn't move the true geoid nearly as much. The rest of the earth has a lot more influence on gravity than the mountain.
I could be wrong, but I think the new geoid model will be mostly derived from gravity measurements (airborne and ground), and the gps on benchmark data will be used as a check against problems and whether the gravity measurement grid is adequate. Some of the prior geoid updates have been based on the differences between GPS and old leveling data. If the bench mark moves up or down after leveling and you rely only on that difference, then you get an incorrect geoid value. Thus the gravity values are better to have. Maybe Dave or Mike will correct me.
Bill93, post: 331953, member: 87 wrote: If the mountain moved some small amount, it wouldn't move the true geoid nearly as much. The rest of the earth has a lot more influence on gravity than the mountain.
I could be wrong, but I think the new geoid model will be mostly derived from gravity measurements (airborne and ground), and the gps on benchmark data will be used as a check against problems and whether the gravity measurement grid is adequate. Some of the prior geoid updates have been based on the differences between GPS and old leveling data. If the bench mark moves up or down after leveling and you rely only on that difference, then you get an incorrect geoid value. Thus the gravity values are better to have. Maybe Dave or Mike will correct me.
As I understand it the new model will have a large shift from NAVD88 and will be gravity based. Here it looks like it's going back to NGVD29 numbers.
As far as there being true geoid number, they aren't there yet.
So if you want to get from the ellipsoid number (which is pretty much useless for anything more than a base for GPS) to the real world, NAVD88 is the standard at this point.
If there is a bench mark run across a mountain where the geoid numbers are changing faster than a foot per mile then it's probably the best solution to hold the existing bench marks. Even if the mountain shifted a inch or something since the 1930's the Geoid model isn't that refined to pick it up. That Leon was within 8cms is really good I think. I was a bit closer with my mountain Bench marks (3cm) and that made me very happy, however, I didn't use OPUS, I used CORS cause I've been down the OPUS rabbit hole too many times.
It has been really amusing to see articles that show climbers taking up GPS equipment to a mountain top and then proclaiming that the old number was 4 inches "off" because they used "GPS". Of course, by techno washing it the public thinks it's real good data, not like the old stuff that was used.
What are the chances that the Geoid Model on a mountain peak is all that accurate? Seeing the local one dropping 2 feet with the new data coming out makes the old guys look pretty good. And really, what does 9526.2' "above sea level" mean when it's 1500 miles to the nearest ocean. Is that number ever going to be real, is there anyway to actually measure it, or is it more an educated guess?
One thing is for sure, the old level runs sure look good at this point, we owe those guys a lot, their work has held up and it's really remarkable what they did.
Excuse the tardy reply to Bill's question. I am currently in Hawaii with only my iPad. I take the opportunity to expound beyond his question (prompted by my inability to yet adapt to the six hour time difference).
On the general issue of gravimetric geoid modeling see this freely available article on the subject:
http://mycoordinates.org/determination-of-local-gravimetric-geoid/all/1/
Remember when discussing the GPS on Benchmark campaigns that it is an effort to improve the fit of NAD 83 ellipsoid heights to NAVD 88 orthometric heights via the hybrid geoid model.
There are two geoid models developed and disseminated by the NGS: those given the name GEOID and an associated date (currently 12B) and those named USGG likewise with a date.
The models named GEOID are hybrid models that are created from the gravimetric models named USGG via datum transformations and surface fitting. The surface fitting process takes computed values of the geoid-ellipsoid separation from GPS-derived ellipsoid heights (h) and published NAVD 88 orthometric heights (H) using the relationship: h - H - N = 0 (or explicitly h - H = N).
Are the separationscomputed this way more accurate than the values from the GEOID model? Not necessarily in my view. They can be consistent and precise when both h and H are accurately determined and the points unchanged by physical processes. Can we say this when we use e.g. 1930s leveling and 2015 GPS?
Unfortunately, as we see in this thread, benchmark monuments are NOT monitored, they are subject to disturbance or destruction and undergo geophysical processes including subsidence, tectonics and GIA.
NAVD 88 has known deficiencies including tilts and offsets. It cannot be the basis of a future vertical reference frame. No significant releveling of NAVD 88 will take place. I understand that NGS is committed to developing tools like the current VERTCON to provide users with a way to relate the new and old vertical networks.
Will the GPS on BM observations aid in the creation of the new gravimetric model? Perhaps someone from NGS will address this issue explicitly. While an OPUS solution will provide an ITRF ellipsoid height they have a problem using h - H = N as NAVD 88 is not based on the zero surface of the geoid.
Recollect the NGS definition of the geoid: an equipotential surface of the earth's gravity field that best represents, in a least squares sense, global mean sea level.
There has been huge progress in the development of worldwide gravity models. Do a search of materials related to Earth Gravitational Model of 2008. Coupling this data source with current airborne and terrestrial gravity (absolute and relative) plus advances in data handling, algorithms and theory everything looks promising.
I agree with those who have noted, that the height determinations made using differential leveling are precise. If the physical monuments are unchanged from when they were determined, they represent the best way to determine height differences. The work of USC&GS/NGS level parties was exemplary. No similar levels of activity are likely to occur again.
Of course all appropriate corrections must be made to the field differences. I hope that those performing levels in higher elevations are aware that the differences in height determined during their leveling might not match the algebraic difference of the published heights of benchmarks. Take a look at the NGS Toolkit item, LVL-DH, and it's documentation to see the problem.
http://www.ngs.noaa.gov/TOOLS/LVLDH/lvldh.shtml
Aloha,
DMM
