Originally Posted by
UAL T38 Phlyer
I have wondered the same thing.
Their premise: the pings stayed the same distance from a geosynchronous satellite over the Indian ocean.
But I wonder: how accurate? Plus or minus how much?
As the +/- tolerance gets bigger, the two arcs from last radar contact (based on the satellite) get wider.
If they get wide enough, they would merge, and literally the entire Indian Ocean is your search area.
DME tolerance is +/- 3%. Are satellite pings the same......or worse?
BTO was on order of 20 us error which mapped to around a 3 km width of the search arc. Inmarsat was very forthcoming and released raw data and there was a great deal of professional and academic independent analysis. Short story is the arcs in the Indian ocean were extremely accurate (the plane was almost certainly within 10 km of the arc) but they were infrequent, and the last arc was received quite a while before the crash.
The doppler motion analysis was not very good (they tried to look at the relativistic / motion shift of the pings to determine direction of travel.)
There was a really nice quite scientific article in aviation week IIRC a year ago or so, I'll try to dig it up and post if you are interested. It addressed those questions and more and was a great read.