lkdiver

…the ability of the pilot to adjust the swoop based on experience makes any given swoop not random at all. Sure. I am only trying to show that the AMOUNT of adjustment in small number of cases may be much larger than in most cases. Recent swooping incidents clearly show that significant number of pilots do NOT have the ability to adjust the swoop correctly.

Let’s assume that we have an idealized swooper. His swoop is affected exactly by seven independent variables: swoop starting height, amount of front riser input, harness input, altitude density, turbulence, duration of front riser input and rate of front riser release (I don’t know what the actual number of variables affecting swoop is – substitute whatever variables and number you like). Let’s further assume that each variable will affect the total altitude loss by exactly plus or minus 10 feet, and that the total error is a simple arithmetic sum of all individual errors. That means that in this ideal case there are 128 possible error combinations, from -70 to +70 feet with following distribution: -70 feet 1x -50 feet 7x -30 feet 21x -10 feet 35x +10 feet 35x +30 feet 21x +50 feet 7x +70 feet 1x From this simple calculation we can derive following conclusions: - Even when individual variables are tightly controlled (+- 10 feet for starting height seems pretty good), if there is a lot of them, the resulting variability is significant (+- 70 feet) - Most of the time random offsets will cancel out. In this case 112 attempts out of 128 will be +- 30 feet (87.5% of the time). - Clearly this can give jumper a false sense of security that he is much better than he really is, luring him into lowering the swoop initiation height. - Once in a blue moon (on average once out of 128 swoops), and what must seem like for no good reason at all, he will be facing deviation of -70 feet (a full 40 feet lower than his ‘normal’ low deviation of -30 feet). It is quite reasonable to assume that correcting this unusually large deviation successfully requires taking action much earlier than usual – probably in the stage where the jumper is concentrating on other things – and by the time he notices this it is too late. It is VERY important to realize that on this particular jump our swooper did not do anything out of ordinary, did not make any larger mistakes than usual – just all of those normal random little mistakes lined up in one direction. - Of course this very low swoop happens on average once in 128 jumps, but this does not say anything about WHEN it will happen. It can happen on the next jump, or easily not for hundreds of jumps. Obviously, in a real world distribution for each variable will not be binary, but more like normal distribution. This will have an unfortunate effect that while on average total error will be smaller, maximum possible error will be even larger than in this example. The effect described here is independent of wing loading, however wing loading does play a significant role: lower wing loading translates into lower speed. Lower speed means smaller consequences on impact, and provides additional benefit of having more total time to react to the problem. There may be other factors in play as well – for example in general swoop on more loaded canopy starts higher, making it more difficult to estimate correct initiation height. It would also seem that effect described here is unavoidable. Even executing 200 ‘perfect’ swoops does not mean that one is out of danger. The only hope for prevention is to understand this phenomenon, leave significant margin for error (much larger than seems necessary) and train to recognize signs of things going wrong early. -Lubo.