GPS Elevation height

Question

I want to calculate the height of an object based on nearest ground surface. I have GPS coordinates and GPS elevation(which I assume is ellipsoid height) as an input in WGS84 coordinate system. I read through some basics and it seems I have to convert this ellipsoid height to orthometric height which I guess is roughly equivalent to Above Mean Sea Level (AMSL), correct?

I downloaded NOAA's VDatum for ellipsoid height conversion.. Attaching screenshot herewith.

I'm not 100% sure if 122.6m ellipsoid height for GPS location 47.662511,-122.152067 is correct or not?

Please also let me know if I want to get accurate height am I following correct approach?

I may have to again convert NAVD88 to AMSL for more accuracy right?

Answer 1

There's not really a way to tell unless you can find some reference data to confirm what either the NAVD88 / approx MSL / height relative to geoid OR the ellipsoidal height is.

One way is to find a topographic map with elevation contours for the area to get a general idea. That would give you NAVD88 or NGVD29 elevations.

I pulled some NGS (National Geodetic Survey) control mark datasheets. There's nothing exactly at the location so I was pulling datasheets for marks that were within a few minutes away (several kilometers). Their NAVD88 elevations were around 15 - 25 m. The separation between the ellipsoid surface and the geoid/NAVD88 surface is around 23 m, which you can see in the VDATUM results.

Several searches to datasheets are available at the NGS Survey Marks and Datasheets page. I used the radial search and you'll want to look for "any vertical control." Note that the searches sometimes want postive-west longitude values and/or DMS (degrees minutes seconds) rather than decimal degree values.

GPS "elevation" should be either ellipsoidal (that's natively calculated), but many GPS devices do have access to a geoid model and can convert to elevations / gravity-related heights.

I would suggest looking through your data to see if you can match up something close to a survey mark to compare the z values.

Note: GPS height values are not as accurate as the 'horizontal' (latitude-longitude) values due to the geometry of the GPS satellite constellation. If the GPS data wasn't RTK'ed (real-time kinematic) or post-processed, it could be up to a few meters off at least.

Answer 2

It appears as if the solution is using geoid12b as this is displayed in the lower left hand corner of the Vertcon application dialog.

After perusing the DEM, and Topographic maps I have included below, it appears as if the height of the location coordinates you provided are above mean sea level. You can see the closed 120 meter contour on both the DEM, and the Topo. This is where the coordinate of your location plots out.

Beyond that, quality of the measurement data will be the determining factor. Unless the GPS position has been established with some sort of correction, whether real time, or through post processing, you could have a significant vertical error.

Here are some images that may help you decide if what you have is reasonable.

This is a digital elevation model with the position you provided located in the center of the image. I drew 5 meter contours over the DEM

---

This is the topographic image of the same area.

@mkennedy beat me to it, and I agree with her answer.