underfire;
I created the images using the basic FR24 kml data in Google Earth (free version), with modifications as described in my post.
Method:
You can right-click-copy each/any of the FR24 waypoints and paste the text into Notepad to see both lat-long & altitude, which follows right after the lat-long data.
You can also create sloping lines to simulate descent paths by modifying the height (in meters), at one end of the line thus created. A slope can be calculated using trig functions. For the 3° descent path on the SXM10 I created a line from the runway threshold extended to a point 10NM out and saved it in Google Earth so it appeared in the Places list on the left side of the map. I right-click-copied that line and pasted the text into Notepad. I calculated the height of the line (ASL) using trig:
tan (3°) x d = h, or, 0.052407 x 60000 = 3144.46ft, or 958.4m ASL.
In Notepad, I changed the line's end height of 0m to 958m then Selected-All and pasted the changed kml over top of the original one in the Places list, (it actually gets inserted just below the waypoint over which you paste).
Because GE defaults to "Clamped to Ground" when drawing any lines, I right-clicked on the line to get the drop-down menu, and selected "Properties > Altitude". I selected "Absolute" which releases it from "clamped" and becomes a slope from a height of 3144ft (the correct height 10NM out on a 3° slope) to about 12', the height of the SXM airport. To be precise one could add that 12' (4m) in and program the end of the line to 962m, but none of this is that accurate, as discussed elsewhere. It takes a little experimentation and intuition.
To be clear, I concur with other posters regarding "no warnings" at that stage of flight. I believe that if you flew these aircraft you'd know that and the fact that GPWS glide slope warnings do not monitor RNAV approaches. As stated, we'll see what the investigation says; I think there are some interesting theories here.