Two Differing Elevation Certificates

I love your question. Waiting to see too....

I have come upon a similar situation.?ÿ The previous surveyor used "Leica Network" as hi base for elevations.?ÿ

I ran a FEMA reference mark 88 datum from about a half mile, to the site.?ÿ When the discrepancy showed 2.2 feet, I re ran the loop to be sure.?ÿ

The insurance company underwriting the flood policy, used my certificate for the the building in question and the other guys, for the two other buildings he surveyed.?ÿ ?ÿ ?ÿI showed lower grades for my buildings.?ÿ (deeper into the flood) the "other" survey showed buildings to be OUT of the flood plane.?ÿ ?ÿ ?ÿpersonally I this its best to use bench reference marks that appear on FEMA plats, but thats just my method.

Good luck?ÿ?ÿ

?ÿ

?ÿI either use OPUS or NGS Benchmarks. I have found some difference between NGS over a couple of feet. I was going to go exclusive OPUS, but since the NGS stuff is government cerifyed. And you have monument to fall back on. I decided the NGS Benchmarks are the best.

Most recent such project was within 300 feet of "THE" benchmark for the small town. ?ÿIt was known as RM1 on the prior FIRM. ?ÿCould string a few garden hoses together to carry that elevation to the site.

Thread hijack:?ÿ I've heard that the subtleties of a water hose level mean it isn't as accurate as a good optical level run. ?ÿ I note that the Coast and Geodetic Survey chose not to save a lot of money on equipment by using the water level, so I doubt it works as well. I can see that differences along the hose in the amount of dissolved air, surface tension/wetting giving difference in meniscus at the ends, and sun loading on part of the hose would bias the result.?ÿ There might be other factors?

Anyone have info on this?

Was it a two-hour OPUS solution? Do you have a second unit??ÿ Set one on your benchmark and the other on the NGS point the other firm used and measure static for two-hours and send to OPUS. Then post process the baseline yourself.?ÿ You can check the NGS monument and will have some redundancy and may be able to find/fix your or his error.?ÿ

The 0.1' vertical result came from the OPUS report?, or Javad's software??ÿ Care to upload the OPUS solution??ÿ

I don't own RTK network of otherwise, but have learned from everyone here, in order to get accuracy, you need to take multiple shots at different times with different constellations.?ÿ?ÿ

?ÿ

A nominal position would help sort out the error. ?ÿThere are common differences that point to datum shifts of certain flavors.

An OPUS position with a tie to a passive mark or two keeps us out of trouble. Using an RM from the FIRM Panel as the passive mark check prevents most datum errors as well...

You need to use the same Datum that the FIRM and FIS are based on.

Archive the project in your Triumph LS. Once archived you can go in and look at the DPOS report (found in the zipped archive file under reports). You should be able to see what the residuals were and which CORS stations were used. Attach that report to your elevation certificate. At least this way you have some data to back up your results unlike a VRS/network solution.?ÿ

locally, where we have several different and very robust networks in place, i've repeatedly seen network-based elevations that i wouldn't hesitate build anything from or insure anything against.

it's like anything else: i've also used the scope on an old leica as a defacto telephoto lens for my iphone... (won't be entering any of those photos in an art contest.)

if you understand the how/what/why of properly executing elevation work using a network as your basis (in addition to basic good practice re: gps in general), imo you can achieve repeatable results that are every bit as tight as running long sessions against opus or carrying hard levels across two counties.?ÿ?ÿ

my experience is based upon doing just that- running in control on network for some extremely elevation sensitive projects (one was a lomr-dependent multifamily project), then running some long-arse bench loops off established and varied control (NGS and static city control monuments). i also ran 2 hour sessions on a couple of them and the OPUS results came back +/-.02 from network results. that's as good as anything else i've ever done any other way.

i can't speak for areas with less robust networks.?ÿ and, of course, the practice of being in a hurry, cutting corners, or using a tool you don't even understand usually leads to less than ideal results- at least in my experience.

My $.02....I would steer away from using OPUS to establish/verify elevations but that's just based on my personal experience.

What Flyin Solo said about results from a robust RTN.

I know a number of surveyors in my area who have compared 15 -30 second RTN observations to post processed static networks and OPUS results and invariably the differences between two are in the +- 1 cm?ÿrange.

I set five RTN pairs with 20 second observations on this project then ran 3.5 miles of digital levels north/south across the entire control network.?ÿ?ÿ?ÿ Held one?ÿelevation at the south end and eight were within 0.03' vertically.?ÿ The lone outlier was 0.05'

?ÿ

Hence the reason we are going to a new geopotential datum to get away from unstable marks on the ground, different sources, etc. Even stable marks can differ significantly if they were on different level lines. I have recently seen marks that agree with neighbors, but disagree across a large river by 10 cm.?ÿ

When I want to bring in a "good" elevation, I recover at least four marks surrounding the project, and bring them in by static observations.?ÿ

In the "old" days, before geoid models, we had to survey enough benchmarks surrounding the project to model the geoid (with no gravity data to work with).?ÿI did a project a few weeks ago where the difference in geoidal separation from one end to the other (about 75 miles) was 21 feet. 3/4 of the project was on the great plains, and the western 1/4 was in the mountains. I would not have wanted to try and model that in the days before the geoid models were developed.?ÿ

I've only done 1 elev certificate in my career and that was only because I knew the elev. was well above any height that could be reached by a flood unless the whole county went under. IMO anyone who does these should carry high liability and charge for it. From what I've seen it's much the opposite. Accessing datum is not a simple matter. It takes a lot of surveying and study to determine the best estimate of datum particularly in areas with old leveling.?ÿ If 2022 datum catches on the amount of work required to do this will hopefully change.?ÿ The differences described in this thread don't surprise me in the least.?ÿ

From the Elevation Certificate Instructions

Item C2. A field survey is required for Items C2.aƒ??h. Most control networks will assign a unique identifier for each benchmark. For example, the National Geodetic Survey uses the Permanent Identifier (PID). For the benchmark utilized, provide the PID or other unique identifier assigned by the maintainer of the benchmark. For GPS survey, indicate the benchmark used for the base station, the Continuously Operating Reference Stations (CORS) sites used for an On-line Positioning User Service (OPUS) solution (also attach the OPUS report), or the name of the Real Time Network used. Also provide the vertical datum for the benchmark elevation.

All elevations for the certificate, including the elevations for Items C2.aƒ??h, must use the same datum on which the BFE is based. Show the conversion from the field survey datum used if it differs from the datum used for the BFE entered in Item B9 and indicate the conversion software used. Show the datum conversion, if applicable, in the Comments area of Section D.

For property experiencing ground subsidence, the most recent reference mark elevations must be used for determining building elevations. However, when subsidence is involved, the BFE should not be adjusted. Enter elevations in Items C2.aƒ??h to the nearest tenth of a foot (nearest tenth of a meter, in Puerto Rico).

Posted by: Squirltech

My $.02....I would steer away from using OPUS to establish/verify elevations but that's just based on my personal experience.

Why! It doesn't get any better than OPUS. NGS even knows it has bad benchmarks and wants to verify them with OPUS.

WSDOT (Washington State DOT) Maintains an excellent network of monuments; up and down all of the state highway corridors.?ÿ

Survey Monument Database

I subscribe to the WSRN (Washington State Reference Network) and my SOP is:

• I use my Leica Viva GNSS GS12, connected to the network and shoot a minimum of 5 shots, each, on at least 3 WSDOT mons
• I set a minimum of 3 temporary bench marks, on-site, and shoot each one at least 5 times
• I repeat on the 3 WSDOT mons
• I repeat on the 3 TBM's
• Set up the robot and shoot the 3 TBM's, collect all the appropriate data and maybe a little more
• This field procedure usually takes about an hour and a half, plus travel time

When I analyze the data, I tend to try to find a happy medium. Sometimes I throw out an outlier, most of the time the average is within a few hundredths of all of them. I don't do a lot of these, the most error I've seen is a couple tenths.

It's all about your comfort level; how comfortable are you, in saying: my elevation is the right one...

What Skeeter1996 said. Why wouldn't an OPUS solution work? I understand that when the 2020 datum is adopted physical benchmarks will be obsolete. At that time an OPUS solution would seem to be the ONLY acceptable vertical datum...

What David said, doubtful opus is the basis of the FIRM