DJ77 said:
Precise calculations using the last ACARS position message coordinates (as reported by BEA) show that the crosstrack error (XTK) was 2.95 nm left of UN873 centerline.
At further risk of being a pedant here I note from the BEA report's ACARS transcripts that the report was in 100ths of a degree latitude and lognitude.
That is a six tenths to 12 tenths of a mile possible error from the truncation or round off process that took place. You quoted an unjustified level of precision there. However, even with that error band it was very slightly to the left of track when you'd expect any pilots even minimally weather aware to be somewhere else.
I also note that altitude, which would be nice to know at this time, was left off the report.
Further I did not see in the report a definitive statement whether the ACARS antenna was an omnidirectional component of the SatCom antenna or whether it had to be precisely aimed at the satellite. That would give an important resolution to the 4 minute give or take a little ambiguity about when the problems started that exists.
If the antenna is omnidirectional or in an omnidirectional mode for ACARS then for transmissions to cease at 0214 one might hazard to guess they were already in trouble at 0210. If the ACARS antenna requires accurate aiming then they were not in trouble until 0214 when they'd have been nearly out of the storm unless they'd become disoriented and turned sharply.
I've been yelled at on this list for presuming both of these antenna propositions. With what I know about SatCom I am inclined to suggest ACARS transmissions would not need a directional antenna to work.
JD-EE