AF 447 Search to resume
If upset was at 02:10 (or a little before), the aircraft simply could not have traveled anywhere near to the edge of a 40nm radius of interest (see Post #1349) in 4 minutes of rapidly descending and assumedly uncontrolled flight. You would conclude that the hull will be found close-in to the LKP.
I strongly tend to agree.
Back then I did a rough calculation or should I say guestimate about the dissipation of speed assuming a high altitude stall and an impact with low forward velocity. Depending on when the initiatial decelleration (possibly due to climb) occured the resulting distance of travel after 2:10 was roughly between 20 - 25nm. Taking into account a possible course reversal the turn alone would consume between 6 and 12nm. So this would point to a likely range of the probable area between LKP and maybe 15nm south of LKP. Extrapolating the drift funnel to the south would point to a location pretty much close to but to the south of LKP.
Wild guess:
If I were asked to put a bet, I would start at LKP or maybe 5nm to the south of LKP and go in concentric circles from there. If after a Radius of 15nm nothing is found I would be rather surprised.
Wind Data
Thanks SaturnV, have checked out the link. From what I can gather, did the drift models just look at surface current effects ? What I had in mind was just focusing on the galley section and using wind effects only (eg assume object attained x% of average wind speed each day with x between 25% and 70%).
I imagine that's what the submerged parameter in the BEA analysis is about, but for currents.
My French isn't reliable but did their model combine both wind and current, eg an object 50% submerged got 50% of the current effect and 50% of that due to wind, or was it just factoring in the area subject to the current forces and drag, with sub-models/assumptions used to determine the actual physical effects of both. ?
What I was hankering after was whether the wind-only backtrack on the galley section gave any course near the pollution spot. A simple parameterised model should be do-able if the daily wind vectors are known.
I imagine that's what the submerged parameter in the BEA analysis is about, but for currents.
My French isn't reliable but did their model combine both wind and current, eg an object 50% submerged got 50% of the current effect and 50% of that due to wind, or was it just factoring in the area subject to the current forces and drag, with sub-models/assumptions used to determine the actual physical effects of both. ?
What I was hankering after was whether the wind-only backtrack on the galley section gave any course near the pollution spot. A simple parameterised model should be do-able if the daily wind vectors are known.
Join Date: Mar 2001
Location: us
Posts: 694
Likes: 0
Received 0 Likes
on
0 Posts
Mr. Optimistic,
That link was to a set of a powerpoint slides, with no further explanation.
As the original language is English, I assume the simulation was likely prepared by an entity in the UK or the U.S., possibly Woods Hole.
I also assume that the drift was calculated based on the differing degrees of immersion, and the variable contributions over time of wind (both direction and velocity) and current (both direction and velocity) in steering an object under various immersion scenarios.
As the different drift paths all converge on the same line in the period before recovery, I assume that reflects better on-site data regarding wind and current in the search area, and thus reducing the error, or spread, that is likely present in the calculations for the period immediately following AF 447 crashing into the sea.
That link was to a set of a powerpoint slides, with no further explanation.
As the original language is English, I assume the simulation was likely prepared by an entity in the UK or the U.S., possibly Woods Hole.
I also assume that the drift was calculated based on the differing degrees of immersion, and the variable contributions over time of wind (both direction and velocity) and current (both direction and velocity) in steering an object under various immersion scenarios.
As the different drift paths all converge on the same line in the period before recovery, I assume that reflects better on-site data regarding wind and current in the search area, and thus reducing the error, or spread, that is likely present in the calculations for the period immediately following AF 447 crashing into the sea.
Join Date: May 2010
Location: MA, USA
Posts: 126
Likes: 0
Received 0 Likes
on
0 Posts
Drift Committee Report
SaturnV, Mr. Optimistic, Slats11, et al:
I have been trying to follow your discussion of drift, but it is not clear from your questions or responses that you are referring to the full 144 page report of the Drift Committee that proposed the Phase 3 search area. The document is available here (the report was originally written in English and the French language branch of the BEA site does not offer a French translation):
http://www.bea.aero/en/enquetes/flig...oup.report.pdf
The document is quite mathematical, but it contains a lot of raw data that is free to be reinterpreted. I believe that the members of the committee (perhaps individually) are reconsidering their own interpretation, in view of the fact that the aircraft was not found in the predicted box.
Likely you are aware of this report, so please accept my apology in advance if I'm repeating the obvious.
I have been trying to follow your discussion of drift, but it is not clear from your questions or responses that you are referring to the full 144 page report of the Drift Committee that proposed the Phase 3 search area. The document is available here (the report was originally written in English and the French language branch of the BEA site does not offer a French translation):
http://www.bea.aero/en/enquetes/flig...oup.report.pdf
The document is quite mathematical, but it contains a lot of raw data that is free to be reinterpreted. I believe that the members of the committee (perhaps individually) are reconsidering their own interpretation, in view of the fact that the aircraft was not found in the predicted box.
Likely you are aware of this report, so please accept my apology in advance if I'm repeating the obvious.
Last edited by auv-ee; 6th Dec 2010 at 02:29.
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes
on
0 Posts
GreatBear, even if the plane had power and was more or less under control until it's last ACARS message I suspect it could not be outside the 40 mile limit after a controlled glide into the water.
For that matter given its momentum I bet you could describe a figure that is slightly smaller than a simple 40 mile radius. Such a figure might go backwards along the flight path as little as 20 miles.
If BEA figures out a sane figure of that sort it might reduce the expense of a thorough search somewhat.
For that matter given its momentum I bet you could describe a figure that is slightly smaller than a simple 40 mile radius. Such a figure might go backwards along the flight path as little as 20 miles.
If BEA figures out a sane figure of that sort it might reduce the expense of a thorough search somewhat.
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,454
Likes: 0
Received 0 Likes
on
0 Posts
I read a substantial portion of the Drift Group Report when it was presented on this thread.
My impression (and only an impression) after reading this report is that each member of the group approached the problem as best they could within their specialties and then wrote up their reports without substantial interaction.
The reports were then bundled together into a master report, and a summary was generated to assemble them into some sort of framework.
It is not clear that all members of the group gathered together at some point to generate the summary and I didn't see evidence of an attempt to fuse diverse approaches to the problem to create new analytic tools.
Now the foregoing is NOT a condemnation of the report or the work of the contributors. I merely think that this was an impossibly difficult task and the widely separated members did the best they could with insufficient data.
If I am correct, the function of the report was to highlight the best information available to plan a search and to logically justify the expenditure of the substantial funds required for the search. I suspect the POL spot emerged late in the development of the report and there was not adequate time to revise the already presented papers.
Again, the foregoing is just my personal impression, and someone else may have gathered an entirely different impression of the report.
Feel free to disagree. I'm curious if others received the same impression from the report.
My impression (and only an impression) after reading this report is that each member of the group approached the problem as best they could within their specialties and then wrote up their reports without substantial interaction.
The reports were then bundled together into a master report, and a summary was generated to assemble them into some sort of framework.
It is not clear that all members of the group gathered together at some point to generate the summary and I didn't see evidence of an attempt to fuse diverse approaches to the problem to create new analytic tools.
Now the foregoing is NOT a condemnation of the report or the work of the contributors. I merely think that this was an impossibly difficult task and the widely separated members did the best they could with insufficient data.
If I am correct, the function of the report was to highlight the best information available to plan a search and to logically justify the expenditure of the substantial funds required for the search. I suspect the POL spot emerged late in the development of the report and there was not adequate time to revise the already presented papers.
Again, the foregoing is just my personal impression, and someone else may have gathered an entirely different impression of the report.
Feel free to disagree. I'm curious if others received the same impression from the report.
Join Date: May 2010
Location: Le Shed on the Tropic of Capricorn
Age: 62
Posts: 22
Likes: 0
Received 0 Likes
on
0 Posts
Northerly Drift
If anyone has the raw data of the locations and the discovery times of the individual bodies, it would be interesting to play around with this.
Slats 11
Table 1 pg 34 of the drift report gives the time and location of the retrieved bodies. I used this data and got a northerly drift of north = -0.000103399474 x time^2 + 0.042572523906 x time. Where time = hours from the time of the slick pic ie time from 2:14.5 less 30.03 hrs. If one ignores the first 2 bodies it becomes north = -0.000088568370 x time^2 + 0.040144268083 x time. There is no x^0 ie no intercept value, since the curve is focused on the slick.
FF
Edit. Centres impact at 2° 37.4’ N, 30° 30.6’ W
Last edited by FluidFlow; 6th Dec 2010 at 19:18.
Join Date: Jun 2003
Location: Camel jockey
Posts: 183
Likes: 0
Received 0 Likes
on
0 Posts
GreatBear, even if the plane had power and was more or less under control until it's last ACARS message I suspect it could not be outside the 40 mile limit after a controlled glide into the water.
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes
on
0 Posts
bai bota, I was simply ignoring that message and showing that even in that case the radius declared for the search should be good and conservative.
So unless some power beyond our understanding gave the plane a big push it's not likely to be outside that 40 nm radius. The only exception would be if the 0210 report was in error way more than you'd get from GPS.
So unless some power beyond our understanding gave the plane a big push it's not likely to be outside that 40 nm radius. The only exception would be if the 0210 report was in error way more than you'd get from GPS.
Join Date: Jun 2009
Location: in a plasma cocoon
Age: 53
Posts: 244
Likes: 0
Received 0 Likes
on
0 Posts
drift group
Machinbird, as I see it, indeed there were several teams implementing different models with different physical basis and data assimilation capabilities, but at one point, the BEA had to estimate a search area (the output of the work for phase III) using all the different backdrifting estimates and their accuracy level. For this latter, the work to quantify the errors between the real trajectories of the buoys and the simulated ones is interesting (the mercator team produced a refined model through this process). The only thing I regret is the absence of montecarlo simulations of the backdrifted trajectories (to get a better idea of the sensitivity of the different models) instead of single trajectories with fixed parameters.
Join Date: Apr 2010
Location: Spice Islands
Age: 58
Posts: 114
Likes: 0
Received 0 Likes
on
0 Posts
I'm going to do something a little different here.
For reasons arrived at after a LOT of reading, calculating, and "half-informed guesswork", I posit the folowing impact location:
N 2 deg, 37.2'. W 30 deg, 30.5'
After the aircraft remains are found (fingers crossed for that), if I'm close I'll post my complex (and probably deranged) calculations. Hopefully some others on here will do the same.
For reasons arrived at after a LOT of reading, calculating, and "half-informed guesswork", I posit the folowing impact location:
N 2 deg, 37.2'. W 30 deg, 30.5'
After the aircraft remains are found (fingers crossed for that), if I'm close I'll post my complex (and probably deranged) calculations. Hopefully some others on here will do the same.
Bit of a tease SA
Guess we will have to wait and see then -are we about to open a 'book' on this ?. My thoughts are that a physical model would have to be piecewise linear, so I am doubtful about quadratic curve fits as a basis of extrapolation. The manifold variables make the signal to noise ratio poor, and as has been suggested, filtering of the input data is probably necessary. Trouble is I am all theory and no practice so these words are probably only adding to the noise too !
AUV-ee, thanks for the link. Have skimmed it and am a bit surprised. Throwing Kalman filters around is all well and good but I wonder if that isn't just convoluting the noise. Why not get a pristine galley section, inflict equivalent damage, lob it in the sea, measure currents and winds, watch its motion and then apply the estimated environment.
AUV-ee, thanks for the link. Have skimmed it and am a bit surprised. Throwing Kalman filters around is all well and good but I wonder if that isn't just convoluting the noise. Why not get a pristine galley section, inflict equivalent damage, lob it in the sea, measure currents and winds, watch its motion and then apply the estimated environment.
Last edited by Mr Optimistic; 6th Dec 2010 at 19:21.
Join Date: May 2010
Location: MA, USA
Posts: 126
Likes: 0
Received 0 Likes
on
0 Posts
Drift Report
Originally Posted by Mr Optimistic
Have skimmed it and am a bit surprised. Throwing Kalman filters around is all well and good but I wonder if that isn't just convoluting the noise. Why not get a pristine galley section, inflict equivalent damage, lob it in the sea, measure currents and winds, watch its motion and then apply the estimated environment.
There has been talk of experimenting with some model parts, as you suggest, but I don't know if that effort happened. I'm not sure that it would help anyway, because the wind and current during the critical days are really as unknown as the effect they would have on the debris.
Guest
Posts: n/a
I could have my 3:1 model of the Vertical Stabilizer ready to ship within six weeks or so. I was going to drop it from a Balloon, but a drift test would be easier on the structure. The Galley? I have those all day long. Could make a replica quickly.
Join Date: Jan 2005
Location: France
Posts: 2,315
Likes: 0
Received 0 Likes
on
0 Posts
bearfoil,
Know something...?
Sometimes, all your jokey interjections and idle musings, in what are at least halfway serious discussions, are just not helpful.
You're polluting, not contributing.
Why?
Know something...?
Sometimes, all your jokey interjections and idle musings, in what are at least halfway serious discussions, are just not helpful.
You're polluting, not contributing.
Why?
Thanks auv-ee. I had not previously seen the complete report and was not aware that it was in the public domain. I have only skimmed over it, but a lot of it is well beyond me I am afraid. It is certainly a very sophisticated attempt - perhaps even too complicated.
I plotted the location of the bodies when found against the time of discovery.
7780 3.57 9056 3.65 9504 3.7 9542 3.73 9554 3.74 9566 3.75 9640 3.73 9740 3.75 10963 4.04 10990 4.03 11015 4.06 11059 4.06 11110 4.08 11141 4.09 11970 4.25 12001 4.25 12047 4.25 12134 4.27 12161 4.27 12194 4.31 12246 4.31 12330 4.35 12330 4.35 12369 4.41 The first figure is the duration of time (minutes) that the body was in the water - from 0215 on the 1st until discovery. The 2nd figure is latitude (decimal).
The 1st body found (1155 on the 6th) is a bit of an outlier. The graph is reasonably linear for the bodies found on the subsequent days. However this 1st body is too far north. Possibly there was an error recording the position, although this is unlikely given the significance of this finding.
If you ignore this 1st body and then do a linear extrapolation from the subsequent bodies, you end up around 1.6N - I don't think anyone believes the aircraft will be found that far south.
So this suggests that the drift changed sometime around the 6th and the 7th, complicating the backtrack.
The other evidence for this change is they didn't find any bodies on the 8th until 1658 UTC. They found one on the 6th at 3.57 N. They found 7 on the 7th between 3.65 and 3.75 N. This limited data would suggest a fairly gradual drift. On the 8th they presumably went out to the same area and didn't find any bodies until late in the afternoon at 4.04N. This suggests they did not find bodies the next day where they expected them to be - the drift had changed (increased) carrying them 15 or so miles further north from the previous day. Maybe the weather was bad in the morning and precluded searching earlier in the day. If the weather was OK however, a likely explanation for not finding bodies early in the morning was that they were not where they expected them to be, and it took them time to re-establish contact with the group. 15 miles does not sound like much, but it is when looking for a body in the open ocean a long way off shore when aircraft have limited time overhead. It could easily take hours to re-establish contact. To me, this is evidence that the current was likely non-linear over this interval.
I have played around with a few graphs, and have also come up with a figure of about 2 30' N. Interestingly this is close (sort of) to Fluid Flow and Sam - however I suspect that mine is more of a "guess" than their estimate. The data is just too limited.
The authors of this drift analysis struggled also. Lots of assumptions. Look at the comment that "the impact could be further east if we have underestimated the wind effect on the bodies". For all their considerable best efforts, the conclusions are only as good as the assumptions and estimates that went into it.
I plotted the location of the bodies when found against the time of discovery.
7780 3.57 9056 3.65 9504 3.7 9542 3.73 9554 3.74 9566 3.75 9640 3.73 9740 3.75 10963 4.04 10990 4.03 11015 4.06 11059 4.06 11110 4.08 11141 4.09 11970 4.25 12001 4.25 12047 4.25 12134 4.27 12161 4.27 12194 4.31 12246 4.31 12330 4.35 12330 4.35 12369 4.41 The first figure is the duration of time (minutes) that the body was in the water - from 0215 on the 1st until discovery. The 2nd figure is latitude (decimal).
The 1st body found (1155 on the 6th) is a bit of an outlier. The graph is reasonably linear for the bodies found on the subsequent days. However this 1st body is too far north. Possibly there was an error recording the position, although this is unlikely given the significance of this finding.
If you ignore this 1st body and then do a linear extrapolation from the subsequent bodies, you end up around 1.6N - I don't think anyone believes the aircraft will be found that far south.
So this suggests that the drift changed sometime around the 6th and the 7th, complicating the backtrack.
The other evidence for this change is they didn't find any bodies on the 8th until 1658 UTC. They found one on the 6th at 3.57 N. They found 7 on the 7th between 3.65 and 3.75 N. This limited data would suggest a fairly gradual drift. On the 8th they presumably went out to the same area and didn't find any bodies until late in the afternoon at 4.04N. This suggests they did not find bodies the next day where they expected them to be - the drift had changed (increased) carrying them 15 or so miles further north from the previous day. Maybe the weather was bad in the morning and precluded searching earlier in the day. If the weather was OK however, a likely explanation for not finding bodies early in the morning was that they were not where they expected them to be, and it took them time to re-establish contact with the group. 15 miles does not sound like much, but it is when looking for a body in the open ocean a long way off shore when aircraft have limited time overhead. It could easily take hours to re-establish contact. To me, this is evidence that the current was likely non-linear over this interval.
I have played around with a few graphs, and have also come up with a figure of about 2 30' N. Interestingly this is close (sort of) to Fluid Flow and Sam - however I suspect that mine is more of a "guess" than their estimate. The data is just too limited.
The authors of this drift analysis struggled also. Lots of assumptions. Look at the comment that "the impact could be further east if we have underestimated the wind effect on the bodies". For all their considerable best efforts, the conclusions are only as good as the assumptions and estimates that went into it.
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes
on
0 Posts
In the process of checking positions of debris found in the BEA's Drift Group Report I found that some of the plots were using a position 4NM further north than where it should have been. Another item that I believe has wrongly recorded co-ordinates is the Vertical Stabilizer. There is a "sighted" and a "recovered position which shows that the V/S moved SSW at 1.9 knots between these two positions. I don't believe that was probable, which then leads to which of the two positions is correct?
Having got over those two problems, I plotted a few postions and started to consider the conflict that happened on the northern edge of the 40NM radius circle on 5 and 6 June 2009. It then dawned on me that the Singaporean registered tanker "Ursula" [correct spelling] 9VFJ6 passed through this area, and the chance of debris being sighted would be increased considerably if there was a local eddy and a lot of "stuff" was circulating in a relatively small area. During daylight hours there would only be the Mate on watch on the bridge, and basically just one pair of eyes keeping a casual lookout, but at the time was probably keeping a better lookout than normal following a Radio Nav Warning to shipping regarding the missing aircraft. Then having sighted a body(?) noted its position and then did something about reporting it to RCC, with the ship continuing on its passage without any deviation.
So still considering the SSW drift of the V/S, it became apparent that a localised eddy could help explain the conflicts in timings in this small area, and hence a new graphic which uses some data from a buoy that passed through the area a few days later to show another possibility. Though not shown, because it doesn't really matter, is the starting point which is close to the Pollution Spot. Buoy 246 had a daily drift of about 19NM when it moved NNE after passing just to the south of the LKP, and I have used that sort of drift rate on that part of the track. Everything then slows down when the southern edge of the Counter Current pushing to the west near TASIL causes an abrupt direction change to the SW leading into the clockwise eddy and slowly picks up as the current finishes up heading north to 3Z and 9Z etc..
Note:: The timings and positions don't always match, but then that is to be expected after 6 days.
Having got over those two problems, I plotted a few postions and started to consider the conflict that happened on the northern edge of the 40NM radius circle on 5 and 6 June 2009. It then dawned on me that the Singaporean registered tanker "Ursula" [correct spelling] 9VFJ6 passed through this area, and the chance of debris being sighted would be increased considerably if there was a local eddy and a lot of "stuff" was circulating in a relatively small area. During daylight hours there would only be the Mate on watch on the bridge, and basically just one pair of eyes keeping a casual lookout, but at the time was probably keeping a better lookout than normal following a Radio Nav Warning to shipping regarding the missing aircraft. Then having sighted a body(?) noted its position and then did something about reporting it to RCC, with the ship continuing on its passage without any deviation.
So still considering the SSW drift of the V/S, it became apparent that a localised eddy could help explain the conflicts in timings in this small area, and hence a new graphic which uses some data from a buoy that passed through the area a few days later to show another possibility. Though not shown, because it doesn't really matter, is the starting point which is close to the Pollution Spot. Buoy 246 had a daily drift of about 19NM when it moved NNE after passing just to the south of the LKP, and I have used that sort of drift rate on that part of the track. Everything then slows down when the southern edge of the Counter Current pushing to the west near TASIL causes an abrupt direction change to the SW leading into the clockwise eddy and slowly picks up as the current finishes up heading north to 3Z and 9Z etc..
Note:: The timings and positions don't always match, but then that is to be expected after 6 days.
Last edited by Jetdriver; 18th Dec 2010 at 17:04.
Join Date: Nov 2005
Location: Harvest, Alabama
Posts: 109
Likes: 0
Received 0 Likes
on
0 Posts
Bia Botal writes:
I'm not certain of the question asked, but I always ASSuMEd that message was a 'Depress' message. Not that the cabin was not keeping up (normal descent / increasing pressure / lowered cabin altitude) with a rapid descent (Double Negative), but rather that the cabin was climbing at a triggered alarm rate. From a hull/system breach, or, more likely, the bleed supply insufficiency (idling/compressor-stalled/flameout engines).
This has me a little confused, wasn't the last ACARS message a advisory regarding cabin vertical speed. Wouldn't this suggest that the event was well underway, i was always of the impression that this indicated that the pressurization system was warning that it could not keep up with the rate of decent and had little or nothing to do with blocked pitot, am i wrong? if so then i must have missed that discussion some where, if i am not then surely this would indicate that the end was near and so the possible position of the aircraft.
I'm not certain of the question asked, but I always ASSuMEd that message was a 'Depress' message. Not that the cabin was not keeping up (normal descent / increasing pressure / lowered cabin altitude) with a rapid descent (Double Negative), but rather that the cabin was climbing at a triggered alarm rate. From a hull/system breach, or, more likely, the bleed supply insufficiency (idling/compressor-stalled/flameout engines).