Sorry if I am being picky and as a non-pilot I don't know how relevant these comments are, but I think that there is a basic error in the altitude/distance graph in the site linked to by BackPacker in post 713. This shows the final point at an altitude of zero and I am assuming that the originator of the graph merely added 200 feet to all the altitudes shown on the data from OpenATC.com.
It was reported early on in this thread that the QNH at the time of the crash was 1027hPa. Since the SBS altitude data is related to standard pressure, it was under-reporting the altitude by 14 times 27feet, i.e. 378 feet. Since the altitude data, anyway, is reported only in multiples of 25 feet, this points the altitude at the final position report as 175 feet +/- 10 feet or so. It is quite likely that the final position reported was not the impact point but just that at which the SBS box stopped receiving updates. This was likely due to the low height. For instance, I cannot follow aircraft all the way down to either 27L or 27R at Heathrow due to the intervening reservoirs (my whereabouts will be obvious from my name).
If we assume that all altitudes in the graph should be shifted up by about 175 feet, and adjust the graph accordingly it shows a somewhat different pattern. The graph shows the aircraft crossing the G/S at about 10 kilometres, staying above it for a time before steadily falling below it. Making the altitude adjustment I mentioned, the pattern shows the aircraft always above the GS but getting steadily closer until reaching it at the penultimate data point (the 75'/250' one). At this point, the sudden increase in the sink rate takes it sharply below the G/S until impact.
As a non-pilot, I am not sure if this is what HundredPercentPlease is discussing in post 720.;
I'm sorry that I can't post a revised graph here (I have to go out). However, if anyone would like a copy of my Excel file, please pm me and I will do so asap. If, as a non-pilot I am talking gibberish then I apologise. The same if this post is irrelevant.
SFS