Z coordinates don't match Benchmarks

[DonRodgers]

I am trying to use my centimeter-grade Trimble R2 with ESRI Collector on my Ipad to shoot XY and Z coordinates for civil engineering projects.

I have shot several USGS benchmarks with this equipment and I consistently get a Z value that is between 0.62 and 0.35 feet below the established elevations for these benchmarks.

I also collected hundreds of project-related points with this equipment including a professional-surveyor-established TBM. When I run a bench circuit with my traditional rod and level, shooting the NAD 83 NAVD 88 TBM and many hard-surface points I average 0.47 feet above the same points shot with the R2.

In both of these scenarios the R2 values are consistently approximately half a foot below the values that I determine to be “correct”. Usually 0.45 to 0.55 feet.

I am hoping (as I always do) that someone can point out a simple mistake that I am making and a simple solution to correct this.

[RobertBorchert]

Not certain but is there a setting that determines how high your device is above the ground when taking the reading.

Meaning your holding the device 4.62 feet above the ground and your setting are for 5 feet above the ground

It could also be the accuracy of your device based on how many satellites you can acquire, interference with the surroundings.  And how certain are you that your benchmarks are correct?

[DonRodgers]

Robert, thank you for your thoughtful response.
I am confident in the accuracy of the Benchmarks that I am using. All were established by government agencies and professional surveyors.
I know that I can go to Profile/Provider in my Collector app and change the antenna height. I currently have it set at 2.11 meters for my R2 per my Trimble support provider’s instruction. I have a “2 meter rod” that I use with the R2 which means that the base of the R2 antenna sits at 2 meters above the ground and the yellow ring around the antenna sits at 2.10 meters above the ground by my measurement. I wonder if I can reset my antenna ht. setting in my Collector profile to compensate for the (perceived) error I am experiencing?
I also have another rod that I can adjust to different calibrated heights. Wondering if leaving the setting as is and “playing with” the actual height would enable me to manipulate the data process to return an elevation value that matches the recorded value of my benchmark.
I would rather understand and use the correct coordinate systems and projections to achieve my goal, but after intermittently working on this for the past year I am under pressure to use this equipment productively.

[MiaogengZhang]

For Trimble R2 with iOS, Z value datum is based on your base station, such as NAD 83 2011 rather than NAVD 88. This could change in future. You can compare with your GNSS metadata if you have configured. Altitude at GNSS metadata is always HAE height. 

[DonRodgers]

Thank you. Can you be a little more descriptive regarding how to access my GNSS metadata?

[MiaogengZhang]

For GNSS/GPS metadata, you can look at Prepare for high-accuracy data collection—Collector for ArcGIS | Documentation 

[Jean-YvesLauture]

Don, 

Assuming your R2 is outputting coordinates in Nad83 2011, have you compared ellipsoidal height (HAE) of the R2 vs the ellipsoidal height of your benchmark? (It would be height in NAVD88 + undulation = HAE. 

If they match, then it would mean that your R2 is not applying the Geoid12B (or new GEOID18) model to give you orthometric height in the NAVD88 vertical datum. It might be using a coarse grid for MSL (Mean Sea Level).

I would recommend debugging this by first comparing ellipsoidal heights. 

[DonRodgers]

Thank you for your response. I will have to do some reading and research to completely understand what you are saying.

[Jean-YvesLauture]

Elevations can be a pain to understand. Here is an article that might help a bit: https://eos-gnss.com/elevation-for-beginners 

(This is expressed in very simple terms, and trust me, there’s much more to it).

In your case, the Geoid12B or the new Geoid18 model is not applied to your ellipsoidal height to get your elevations in the NAVD88 vertical datum (that are shown in your benchmark spec sheet). In ArcGIS, since this is not done in the field, you can convert your Z value back in the office in ArcGIS. You can search here on GeoNet for references on how to do this. Just search for GEOID12B topics for example. Then, your values should match.