PDA

View Full Version : Survey: Pilot Weather Related Decision Making


jarjar01
24th Feb 2013, 21:30
Hello,

I am conducting a survey on VFR pilot weather related decision making for a post graduate study. If you have a spare 10-15min it would be great if you could help – the survey comprises of assessing 5 in-flight images.

This survey is targeting pilots, therefore only complete this survey if you have some flying experiences. Please click on the link below.

Click Here: In-flight Decision Making (http://edu.surveygizmo.com/s3/1099238/VFR-Flight-into-IMC-Anchor)


Thanks very much for your help.


Stephen

Permission was sought and given from PPruneAdmin to post this survey.

ABZ777
24th Feb 2013, 21:48
Done ! Will be interesting to see your results

robin
24th Feb 2013, 22:44
Done

I would suggest that you add a comments box at the end to give qualitive information

I would say that my graphics aren't good enough to give a meaningful answer to some of the questions and I would have liked to point out my reasoning

RatherBeFlying
24th Feb 2013, 23:03
You need real photos -- not screen shots generated by a simulator. There's 2 altimeters that have been deliberately blanked out.

You also need a map showing the destination to give perspective to the weather forecast.

I abandoned the survey.

flybymike
24th Feb 2013, 23:35
There is an assumption that some of the flights would even have been commenced VFR in the first place!

abgd
24th Feb 2013, 23:54
I also felt that the quiz was lacking in realism. I can understand why you used computer-generated images, but would suggest that you could have gotten more realistic ones. For example I was uncertain whether the first picture was flying over sea or flying over forest, at first. Some of the flight simulator scenery packs would make for much better images.

Trees in Canada are much taller than trees in the UK. I can't comment on trees in New-Zealand or on FS, but I have been caught out misjudging distances whilst walking in Canada. This will certainly have affected my guesstimates of altitude and visibility.

I might not choose to fly in some of the weather you've shown, but if you're truly 'nearly at' your destination airfield, the question isn't whether it's safe to continue but whether it's safer to land than to turn back. To be honest, I don't particularly enjoy flying along just under cloudbase but I wouldn't have felt particularly threatened by any of the pictures you provided. I would have worried about embedded cumulo-nimbus in some of them.

All of your terrain was basically unlandable and reasonably flat, so I was less concerned about cloudbase than I normally would be. Given an engine failure the chances are you're ending up in the trees whatever altitude you're flying at, though at 3000 feet you may have just that bit longer to work out how to restart the engine. The chances of being caught between rising ground and cloudbase weren't particularly high.

I found your assessment-of-cloudbase questions quite interesting. The decisionmaking questions were harder to answer, because ordinarily you have so many more factors to take into consideration than a simple picture can give.

flybymike
25th Feb 2013, 08:31
I might not choose to fly in some of the weather you've shown, but if you're truly 'nearly at' your destination airfield, the question isn't whether it's safe to continue but whether it's safer to land than to turn back.
Yes, this was a major factor in my answer to your questions.

Unusual Attitude
25th Feb 2013, 08:32
Likewise I had to give up, no way of judging height without which you cant guage vis or cloudbase, suggest you might want to use one of the photoreal scenery packs or similar.

Regards

UA

wsmempson
25th Feb 2013, 09:00
In addition, although the altimeters were blanked out, in several shots the VSI was indicating +500fpm climb, which skews the perspective.

There is no indication given in your preamble as to whether we should be taking that into account or not?

Jonty
25th Feb 2013, 09:44
If you do not include the hight of the aircraft in your photos, you cannot gauge the hight of the cloud base.

Also, using computer images to gauge depth is next to impossible.

Gertrude the Wombat
25th Feb 2013, 09:48
Looked at the first three, I wouldn't have taken off for a VFR flight in those forecast conditions - I'd have planned IFR.

I personally do not find it easy to judge how far above me the cloud base is - usually I don't know where it is for sure until I'm level with it.

So I didn't complete the survey - there's no option for "I wouldn't have taken off".

dublinpilot
25th Feb 2013, 13:26
I think that the basis for the survey is flawed, so I haven't completed.

Your assumption is that Flight Sim's weather system gives a realistic perspective. I have always found that MS perspective is nothing like reality when it comes to weather.

It's impossible to compare visibility in MS Sim to real world weather, and the clarity of the terrain vs real world clarity gives totally false clues.

If you want a valid result, you should use real photographs.

dp

gasax
25th Feb 2013, 13:36
I have to confess to reading this thread and then looking at the survey.

It is immediately noticeable that a simple compture graphic simply does not give the information required to make a judgement.

Critically pilots have an immediate reference to cloud height - and to some extent visibility by knowing roughly what height they are at. They also look at the windows at the side - both laterally and downwards and those views offer a lot more information.

Finally we know that the US must be a very ugly country, as virtually all US manufacturers block the view outside with a massive panel - something my aircraft does not have...... So my forward view is at least twice as big!

david viewing
25th Feb 2013, 14:06
For a UK audience the inclusion of a 3800' cloudbase forecast in several of the examples is deeply unconvincing - that would rank as the 'best day of the year' here!

It's also deeply unlikely in UK that a 3800' forecast would turn into the sort of weather depicted here, unless crossing high ground or someplace unrelated to the forecast.

On the other hand I found the visuals quite realistic and comparable with the 'turn back now' sort of scenarios that are familiar here when a 2000' cloudbase is regarded as a gift from heaven.

I suspect most of my answers were based on the percieved granularity of the surface.

clunk1001
25th Feb 2013, 15:52
If your post grad study is looking at genuine in flight decision making based on flight conditions then you are way off course using these images.

I've been flying long enough (20 years) and using flight sims long enough (PC based through to level-C) to know you cannot establish any reasonable situational awareness or assess conditions based on 3D rendering and graphics with a limited field of view, certainly not from a pc game, and probably not from any still image capture.

Even if you had images or photos which conveyed the conditions accurately, decisions made in the air especially decisions relating to weather (both subconscious and conscious) come from continual assessment of conditions all around you - at a very basic level "Is it getting better? Is it getting worse?" (weather on track, off track, behind, towards alternate etc).

You simply couldn't (and woudn't) make a decision from your information. I suggest you might want to rethink your research method. All you'll get from this questionnare is "how inaccurate is the Microsofts Flight sim engine at portraying weather conditions and altitude". And I can already tell you that - 'very'.

thing
25th Feb 2013, 16:39
I did it for a larf but as others have said, there are no proper visual clues. You wouldn't rely on the information and pictures that you gave to make an assessment.

Or is it really an assessment on how awkward pilots are...?:)

Crash one
25th Feb 2013, 16:54
Images unrealistic, guestimate of cloudbases. I can't see the point without better resolution. Did it anyway just for a larf.

Romeo Tango
25th Feb 2013, 17:28
Difficult to judge weather from those images if I don't know my height

bucket_and_spade
25th Feb 2013, 18:46
Agreed. Computer-generated images (fidelity unknown) are no good for this survey. Sorry, but I abandoned the survey after looking at the first image.

Hopefully constructive feedback.

AN2 Driver
25th Feb 2013, 18:54
one more thing. Which forecast would give you a visibility of more than 9999? No TAF surely.

From the numbers and over flat territory most of those flights would have probably been possible in VFR but not much fun.

I agree that it would have been much better to use real pics. I am professionally involved with flight simulation and weather (both separately and jointly) and I am pretty clear that most sims can not display visibility accurately, far from it.

Personally, I have a feeling that this study has a different goal than accuracy and that is why I completed it. I won't elaborate but I think I can see where he is coming from and if I am right, then I guess he is onto something which will make most comments here insignificant, because it is not about what it appears.

robin
25th Feb 2013, 19:56
I put a very low confidence on each question. That was because the information was completely inadequate but you have no way of making that point.

Graphics poor and I cannot understand why the altimeter was blanked - how many times do we fly in VFR without a working altimeter??? Totally artificial exercise by, what would appear, a non-pilot

thing
25th Feb 2013, 19:59
I'm with AN2 on this; there's more to this than meets the eye.

flyme273
26th Feb 2013, 08:43
I found realism sufficient for the purpose. Allowing the altimeter would have made it too easy. Of course one has to make assumptions. How far to the airfield, does the otherwise flat terrain suddenly change? Any windmills or high masts to spoilt your day. Is finding the airfield an issue? E.g. a green field in a green forest?
Assuming some runway line up aids e.g. railway, river or GPS, quite low altitude and visibility can be safely accepted. I found all the examples suitable for continuing.
Good exercise. Let’s have some more.:)
Flyme.

Flyingmac
26th Feb 2013, 10:18
I can't see this cobbled together 'survey' producing any useful data.

This might be useful to the original poster:-
http://www.caa.co.uk/docs/33/20130121SSL23.pdf

clunk1001
26th Feb 2013, 11:14
An2 I agree. I originally wondered whether this test was in actual fact specifically designed to find people like FlyMe273 :)

I found realism sufficient for the purpose.
Allowing the altimeter would have made it too easy

Good exercise. Let’s have some more

Ok, FlyMe I'll give you one - you are in a plane and the engine starts making a 'funny' noise. Do you :

A) Continue, because on long journeys it's nice to break up the monotonous drone of Continental engines with a rattle or clunk or two. My favouritest noise is 'bang, splutter' (second only to the sound of two bricks being bashed together).

B) Continue, because I like funny noises. They make me laugh. I feel that my passengers feel more reassured when I laugh, out loud, a lot...

C) I don't have enough information to make a decision, yet.

Sometime, not making a decision, is the best decision to make.

And perhaps that is the point of the test?



....ok now I'm intrigued.

flyme273
26th Feb 2013, 12:42
Dear Clunk,

In my opinion you are expecting too much from this simple survey. This survey is in the form of a "flash-card" presentation. These types of surveys can never truely represent real life situations. Sure it could be further developed. I think the authors will find this out when they receive a wide scatter of answers.

Per the engine noise: I would:-
1. select carb heat
2. move mixture to full rich
3. fuel pump on, change tanks
4. mags select each in turn
5. adjust throttle
6. steer towards landing area.http://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/icons/mpangel.gif

flyme.

The500man
26th Feb 2013, 12:46
I just completed the survey before reading the thread and my feeling was pretty much the same as most other posters here. I think I used my experience of flight simulator flying more than actual flying experience to guess the altitude from which everything else was based. Ground texture is the only thing you can really use for altitude when you don't have an altimeter. In fact you can use the cloud base as well once you have confirmed its height (not a good idea really).

It did raise the question though about how accurately I could guess the altitude with real imagery. At low levels it's fairly easy, but looking at video stills from 2000 feet and up (where I know the altitude), I could just as easily guess it wrong.

rats404
26th Feb 2013, 13:33
I abandoned the survey because the images didn't provide information good enough to allow me to make a meaningful response to the questions.

robin
26th Feb 2013, 13:38
Given that the original intent of the survey is for VFR flight, does the person who set the questions not understand the use of the instrument scan.

We have an altimeter or two that help us judge things. Add to that our stereoscopic vision gives us perspective.

A computer-generated 2d image of dubious quality does not equate to real-world conditions and removes most of the cues we use on each and every flight.

Sorry, but I don't think there is enough information available in the survey to make meaningful judgements. But I would like to see the conclusions they offer.

Johnm
26th Feb 2013, 14:23
This survey is about as scientific as snake oil as a cure for cancer.

Steve6443
26th Feb 2013, 15:35
Per the engine noise: I would:-
1. select carb heat
2. move mixture to full rich
3. fuel pump on, change tanks
4. mags select each in turn
5. adjust throttle
6. steer towards landing area.

flyme.

You forgot the most critical point on the checklist....

1: Brown underpants time.....:E

DeeCee
26th Feb 2013, 15:40
I'm also with AN2 on this. I completed the survey after reading the comments and think most are getting a little bogged down with the aeroplane details. I would be very interested in the results.

rsuggitt
26th Feb 2013, 15:42
I agree with everyone who want to read the altimeter. If I don't know my altitude, I'm not going to start guessing the cloudbase. This survey doesn't really assess one's ability to make weather-related decisions, it's about how well one can estimate altitude from a poor picture.

hhobbit
26th Feb 2013, 15:53
just about what everyone else said.

abgd
26th Feb 2013, 16:26
@Robin:

Stereoscopic vision is useful only to about 6 metres or so. Thankfully we have other depth cues available.

I suspect I know what the study is looking into but I'm still not convinced that the results are going to be valid.

Lokki
26th Feb 2013, 16:40
I agree, probably not enough info to make a 'real life' judgement call, but it was fun to do it anyway wasn't it?

Especially when we all expect we'll be told we'll get 100% at the end of it...;)

24Carrot
26th Feb 2013, 16:51
OK, assuming everybody who is going to do it has done it by now:
Actually the altimeter was not blanked out, just the needles were.:hmm:
Was that part of the test?:confused:

Gertrude the Wombat
26th Feb 2013, 19:58
Was that part of the test?
I've got two altimeters and two static sources. If all that lot fails I'm guessing that there's some buttons I can press on the transponder to get it to tell me flight level (I shouldn't have to guess, I should know this), failing which I can get a radar service to read it back to me. All in all a complete failure of altimetry seems slightly unlikely and is waaaay down my list of technical failures to worry about.

flyinkiwi
26th Feb 2013, 21:22
I hope you all addressed your concerns to the student's supervisor...

Ebbie 2003
26th Feb 2013, 22:41
I quit part way through after a the second scenario - the graphics are not too clear, generally if the winds are that light (I am used to 20kts as a typical), the spread on the temp/dew point so big, the cloud that high I am happy to fly.

If it was to test decision making I would have thought a track of three hours of metars and a TAF with a "go no go" would have been a more precise way to gauge decision making.

Would be intereting to see the spread based on total hours of the participants - would it show the fabled over-confidence in the 250-350 hour range.

Using the Internet and site like this to select while maybe generating a larger number of participants will of course tend toward a self selecting sample which could/should slew results for a number of reasons which would compromise any statistical analysis.

Heston
27th Feb 2013, 08:48
Never complete surveys written by someone who uses the greengrocers' apostrophe...:}

Sorry, couldn't resist that, and I'm afraid I'm with the others who say its impossible to answer based on the images presented.

Interestingly I was "invited" to a seminar recently that involved watching video of the road in front of a moving car - the audience were supposed to comment on the hazards ahead. I couldn't do it because I don't drive by looking at a small square of available view in front and it all just felt totally unrealistic, I couldn't judge distances or speed.

Its why flight simmers who come to their first real flying lessons are so hard to teach. They need to un-learn their reliance on the instruments to tell them what is going on and to look out of the window. They've developed this reliance on the instruments in the game because the visuals outside the window on the screen just don't work well enough. My limited experience with pro simulators would suggest this is mostly down to the restricted field of view that a tv or pc screen offers.

And to the OP - do go and discuss the comments here with your supervisor. There's lots of useful comment here.

pfeinstein
1st Mar 2013, 15:00
Done; please feedback the input you've received. It'll be interesting to see!