3
$\begingroup$

CNN's July 24, 2023 There is a ‘gravity hole’ in the Indian Ocean, and scientists now think they know why links to the Geophysical Research Letter (Pal & Ghosh 2023) How the Indian Ocean Geoid Low Was Formed.

I just want to make sure I understand this situation.

If I call "sea level" the surface that a GPS system (using the WGS84 ellipsoid for reference) would read zero elevation, and I put my GPS system in the bottom of a rubber raft floating in the middle of the Indian Ocean Geoid Low, then on average it would read roughly -106 meters elevation, or 106 meters below "sea level"?

Have I got that right?

$\endgroup$
1

1 Answer 1

3
$\begingroup$

Is the surface of the Indian ocean really 106 meters below sea level?

Obviously not; it is at sea level. It is 100+ meters below the reference ellipsoid. Nor is the spot some hundreds of kilometers east of the Indian Ocean Low well above sea level. It too is also at sea level. The reference ellipsoid is a model that is intended to represent worldwide mean sea level. There is no such thing as worldwide mean sea level. All models are wrong but some remain useful.

If I call "sea level" the surface that a GPS system (using the WGS84 ellipsoid for reference) ...

There's a famous joke about a patient who goes to see the doctor and says "it hurts when I do this <<bonk>>". The doctor's response: "Don't do that then". So don't pretend that sea level is the WGS84 ellipsoid. Older GPS receivers did just that. Newer ones have a somewhat crude model of the geoid in their software.

A geoid (an equipotential surface of Earth gravitational potential energy plus centrifugal potential energy meant to approximate mean sea level that somehow accounts for tides) is a much better approximation of mean sea level than is any reference ellipsoid. That is after all what a geoid is designed to approximate. A geoid is not perfect; there's about a 20 cm difference in mean sea level on the Pacific side vs the Caribbean side of the Panama Canal, mostly due to density differences and sustained winds. But that's a lot better than the 200+ meter difference in the height above the ellipsoid between the Indian Ocean Low and the western Pacific high.

$\endgroup$
6
  • 1
    $\begingroup$ "Newer ones have a somewhat crude model of the geoid in their software." I'm having a problem finding support for that assertion. Most people use the GPS in their phone, with an app; it might be an option that you can turn on or off, but wherever I look on the internet I still see statements that GPS zero elevation is the WGS84 ellipsoid. Perhaps the term GPS , while it used to be well defined, now means different things to different people or in different contexts. $\endgroup$
    – uhoh
    Commented Apr 6 at 16:07
  • 2
    $\begingroup$ cf. eos-gnss.com/knowledge-base/articles/elevation-for-beginners $\endgroup$
    – uhoh
    Commented Apr 6 at 16:14
  • 2
    $\begingroup$ @uhoh gis.stackexchange.com/questions/304256/… , for example $\endgroup$ Commented Apr 6 at 20:11
  • 1
    $\begingroup$ @DavidHammen Yep, there it is - in my old Electronics SE question from my Raspberry Pi days links to u-BLOX NEO-6 GPS module protocol spec which indeed shows field #9 is elevation relative to mean sea level and field #11 seems to contain the geoid offset used for that calculation. $\endgroup$
    – uhoh
    Commented Apr 6 at 23:15
  • 1
    $\begingroup$ @DavidHammen In the late 1990's I worked for a company that designed and made various chips, including 56k modem chips (remember those?) some early RF-CMOS cell phone chips and were pushing into the portable GPS market. At that time it was explained to me GPS elevation was relative to WGS84. I guess he was just explaining the primary calculation. I don't know if on-chip (or in-module) geoids were implemented then and they forgot to tell me about it or not. Anyway, I'm convinced - 1) any modern GNSS device displaying elevation will use some geoid offset, and 2) that's what "sea level" means... $\endgroup$
    – uhoh
    Commented Apr 6 at 23:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.