There's clearly a world of difference between see-and-avoid in a maritime setting (large, slow visual targets confined to a plane), and see-and-avoid in an aviation setting (tiny, fast, moving in 3 dimensions).
Regarding pilot error... Sure, but most fatal accidents involve things such as flying into bad weather, handling errors, engine mismanagement etc...
puts them in the top-10 causes, at 9:
Loss of Control Inflight
Controlled Flight Into Terrain
System Component Failure – Powerplant
Low Altitude Operations
Unknown or Undetermined
Other
Fuel Related
System Component Failure – Non-Powerplant
Midair Collisions
Windshear or Thunderstorm
Overall, mid-air collisions only account for a relatively small fraction of fatalities and are only one of the ways in which we can screw up. I wouldn't blame myself necessarily for a collision, but I would if I'd left the carb-heat off inappropriately or flew into IMC.
I'll do some more reading before further comment.
My point about midairs is that it is an class of incident that ranks the highest in my book for irrecoverability and that it is a threat that is present every time I go flying regardless of weather or how incompetently I behave. For those two reasons, it warrants active countermeasures and real-time attention - and 'lookout' is free/within the grasp of every pilot out there.
A very interesting report and I'm going to adjust my view and approach on a few things as a result of this thread:
1) I have a red-tinted lens over my fin-tip strobe light, I'll be changing it to clear asap.
2) Earlier advice to fly as high as possible as often as possible and at 'odd' levels is clearly a good idea - not just because other a/c are easier to see against the darker ground but because the reduced risk of birdstrikes makes it good sense as well. I'm also going to stay well clear of balloons in the future - objects like that seem to act as a magnet to airborne traffic free to navigate towards them (an airprox with an RF4 2nm from a balloon was one of my near misses).
3) I'm going to look at the potential to change the fuselage side colour of my a/c away from pure white (maybe using vinyl add-on) - this may help increase conspicuity to other a/c at the same height - in that horizon-zone between terrain and sky/cloud.
4) I will concede that TCAS for GA can help, though I still question its relative 'defensive' value wrt lookout for sub 1500Lb a/c - on the basis of cost, onboard power and space requirements and detraction from cockpit attention time. Again - if you can afford it, fine. I already have a mode S - which should be helping to alert of my position to others, both TCAS equipped and ATC of course.
5) I will try and take a passenger EVERY time I go flying.
6) Ask for a traffic or deconfliction service every time I can. My experience is that ATC generally seem to plead 'too busy to provide' it, (esp around W London) but that is no excuse for me not to ask.
7) I'm going to spend a bit more time trying to see what's just under the cowling - that's the area that I'm about to descend into near an airfield . . .
[edited for formatting - Peter]
**Quite an informative paper on the subject, here: http://humansystems.arc.nasa.gov/flightcognition/Publications/Colvin_ISAP05.pdf
"our data suggest that the relatively low (though unacceptable) rate of mid-air collisions in general aviation aircraft not equipped with TCAS is as much a function of the “big sky” as it is of effective visual scanning"
I think that means - look out of the cockpit as much as possible and rely less on technique**
I've only just seen this post and read the paper, but I don't understand what you mean by 'rely less on technique'. The paper does show that some pilots have better scans than others, but it doesn't make the leap to determining the proportion of traffic that an average pilot might see. I can think of some reasons this may have been difficult to do, but I find it slightly surprising that it wasn't discussed.
My reading of the paper was that it's a good job the sky's so big, because pilots' lookout is so crappy that if it wasn't, we'd be having midairs all the time. They do encourage pilots to keep a good lookout - they're hardly going to do otherwise - and they argue that most of us could do better.
There are always caveats - a lot of my window time is to do with looking for landmarks for navigation. In view of this research: http://www.bbc.co.uk/news/health-21466529 - would I see an aircraft if I was actually looking for a steeple?
The US statistics 2007-2009 list 16 fatal accidents due to midair collision out of a total of 611 fatal accidents overall. I'm not clear whether they count a collision between two aircraft as being one or two accidents. That said, these numbers are much smaller than the most likely causes of death (loss of control - 250 deaths). All deaths are irrecoverable, so I don't think it makes sense to claim that midairs should be of greater concern than loss of control, because they're proportionately more deadly.
Whoops - http://www.ntsb.gov/doclib/reports/2011/ara1101.pdf page 37
I think they're good ideas, Europaboy.
Is it possible to do quizzes on this website, where you get to submit your own answer before seeing everybody elses?
A game I used to play when I was small: "Think of a flower - any flower..." Most people really do think of a rose. "Think of a vegetable - any vegetable..." Most people really do think of a carrot.
I would be really interested to see what 'odd' cruising level most people think of, when prompted.
That's very funny 
I would be really interested to see what 'odd' cruising level most people think of, when prompted.
"Obviously" not whole 1000s or whole 500s, so I fly at x300 / x700 usually, VFR, Class G.
It's true that most midairs are fatal for all involved, but then most in-flight cockpit fires are too.
What can I do about a cockpit fire? Inspect the aircraft after any avionics work; that's for sure! How many owners are going to do that?
But these things are very rare.
Obviously one should look out, and in general the more automation you have (an autopilot, especially) the more you are able to look outside. But lookout is limited in effectiveness. That's all I am trying to say. I am happy to take anybody for a flight and they can work out their own personal % of the traffic showing on TCAS which they can spot visually.
Another thing which is a bit of eye opener is that flying under the UK "Traffic Service" there is quite a lot of traffic not reported to you by ATC, which on the face of it should be. In reality the controller is not obliged to report all he knows about, and he is always covered by the "controller workload" get-out. He merely needs to make a reasonable effort. But maybe there is traffic he can't see on his radar? I don't know if this is true but maybe he is not allowed to report a SSR-only target, which would explain quite a lot because all that a TCAS sees are SSR targets.
Oh Peter, you've spoiled it.
I would have predicted that most people would fly at x300 and x700, because most people pick those as the 'most random' numbers.
http://scienceblogs.com/cognitivedaily/2007/02/05/is-17-the-most-random-number/
gives support to '7' but not 3, which I vaguely recall from somewhere else.
Personally I fly at x000 and x500, because I figure everybody else is at 2700.
The Australian report was first issued before the days of GPS meaning before we had super accurate lateral navigation....apart from flying at an "odd" level, an offset of say 0.3nm to the right is also probably a good idea...this presumably helps avoid the climbing/descending traffic for which your odd level will not help....
Also interesting that in one of the CAIR (confidential aviation incident report), the pilot's response was a hard bank to the LEFT....I often wonder whether the fact that Austarlians and Brits among others drive on the left negatively influences our instinctive responses.... AQ
This is why one should never pick lottery numbers using one's head. The chance of winning the jackpot is the same, but you will share the prize with about 10k people
I guess one turns LEFT because one can see better in that direction - in a plane; in a helicopter it's the opposite.
