Differences in SRTM data

[s.savastano]

Hi Marcus (@mengdahl).
yes I agree with you but the weird thing is that if I used the lat lon of a pixel belonging to the sea and then I calculate the difference N between geoid and ellipsoid

using this calculator
https://earth-info.nga.mil/GandG/update/index.php?dir=wgs84&action=egm96-geoid-calc
the value is 43m and not 55m, as visualized here

So, in my opinion, it should be very useful to understand which reference is used in SNAP and which Vertical Datum is adopted (not EGM96 I guess…).

S Savastano

[jun_lu]

Sorry Savastano, I didn’t fully understand your question before and wasted your time. Yes, you are right that EGM96 based SRTM DEM is converted to ellipsoid based DEM in SNAP and used for all position related calculation. The DEM output in Terrain Correction operator is not the original EGM96 based DEM, instead it is the ellipsoid based DEM. It is our fault that we didn’t make it clear. Hope my explanation above answers your question.

[s.savastano]

No problem at all @jun_lu.
your explanation is perfect and thanks to you again for your support. I appreciate that a lot!

For @mengdahl:
Marcus you were right, I forgot to consider the latitude in negative being in W and so recalculating the geoid for that pixel belonging to the sea, we have

exactly the value 55m more or less.
Thank you again to all of you!

S Savastano

[suribabu]

Nice explanation Jun_lu.
So, we can compare GNSS (WGS84) vertical component to PS-InSAR LOS component right. Because both have the same DEM.