greg99

Actually, while I obviously wasn't there, my bet is that the separation issue isn't the dependent approach limitations that you're citing in the order.

My guess is that the 757 was sent around because the heavy 747 (real heavy, not the "faux" heavy 757) was overtaking from above, and the 757 would potentially be behind and below the 747, at risk from its descending wake vortices.

At SFO, the parallel dependent approaches are typically used in somewhat but not fully crummy weather conditions (e.g., relatively heavy haze or a light cloud cover), such that pilots can't make a legal visual approach, but not so bad that they need to do a true ILS approach, with the resulting extreme limitations on flow rate. The parallel dependent approaches allow an ILS approach but reduced separation ("staggered"), which helps tremendously.

A couple of reasons why I think that - (a) the weather looks pristine, with minimal haze and no whitecaps on the bay, and (b) the instructions for the ILS & PDA approaches to 28L/R actually put the onus on the aircraft arriving on 28R (which would have been the CX 747) for wake turbulence avoidance: http://flightaware.com/resources/air...P/PRM+AAUP/pdf

On that short of final, I could also imagine the SFO tower controller deciding that it was safer to send the UA 757 around, because the 757's pilot had probably landed at SFO hundreds of times, had no language issues (which CX may or may not have had), was presumably lighter and had more juice to do a late GA than the 747, and hadn't just flown 13 hours.

Then again, my flight time is solely in a 172 and never @ SFO, so who knows if I'm correct. I did stay at a Holiday Inn last night, though...

BTW, that was an *awesome* video...

Greg