Advanced search  
Pages: 1 2 3 [4] 5 6 ... 10   Go Down

Author Topic: LaPook Hypothesis: Box Search around 157-337  (Read 137128 times)

Martin X. Moleski, SJ

  • Administrator
  • *
  • Posts: 2962
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #45 on: June 24, 2012, 04:34:00 PM »

I've tried to plot out your search pattern on Google Earth (maybe you've already done this?) with the start of the search at Howland  Island.

Nice work, Andrew!  I've been dying for that kind of drawing, but didn't feel capable of making it myself.

Very helpful!
LTM,

           Marty
           TIGHAR #2359A
 
Logged

Gary LaPook

  • T5
  • *****
  • Posts: 1624
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #46 on: June 25, 2012, 02:33:50 PM »

Gary,

I've tried to plot out your search pattern on Google Earth (maybe you've already done this?) with the start of the search at Howland  Island.

Looking at the legs, they end up between 40nm and about 50nm (on the diagonal when they reach the corners) miles away from Howland, on the first 4 legs, then out to 90 miles at the end of the 5th leg, 110 nm at the end of leg 6.  I presume that you do not think they would abandon their search at this point, so I've added the 7th leg of 160 nm.  Since you only give them about an hour before the "END" I presume that you think they ran out of fuel some 3/4 of the way down leg 7. 

Abeam Howland on the LOP during leg 7, they are some 80nm away from Howland.  With visibility at 20 nm, that means that they can see all the way out to the 100 nm point along the LOP.

Since we know that they didn't find Howland, and that also presumably means that they didn't find Baker Isl, it follows that they must have started their search at least 100 nm up the LOP from Howland, or down the LOP from Baker Island, and perhaps as far away as 120 nm from Baker on a diagonal if they misjudged where the LOP was, which seems unlikely since most everyone seems to think that Fred was good enough to get to the Advanced LOP through Howland.



Andrew
You have shown the efficacy of the expanding square search pattern since they would have had to have been a very unlikely number of miles of course to not have been able to find Howland with this search, IF THEY HAD FOUR HOURS OF FUEL.

gl
Logged

Andrew M McKenna

  • Administrator
  • *
  • Posts: 666
  • Here I am during the Maid of Harlech Survey.
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #47 on: June 25, 2012, 02:50:14 PM »

You didn't answer my question, for starters, and I'm not exactly sure what you are trying to say.

Are you saying that if they had flown this search, they would have found Howland?

I'm just working off your calculations for the expanding search, which has an END time at 23hrs plus. 

Are you now proposing a different end time, and if so when do you calculate the END, and why?

amck
Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #48 on: June 25, 2012, 04:15:08 PM »

I am still confused as to where they actually started their box search from (if they did)
Where AE and FN thought they were? Image 2
Where the Monte Carlo simulation predicts they really were? Image 1
Some unknown point along on the LOP? Andrews splendid map

If they did start a search pattern, which I am sure they would have, then they must have been hopelessly off-course for the box search pattern to fail to put them onto Howland or Baker.

This must be the place
 
Logged

John Ousterhout

  • T4
  • ****
  • Posts: 487
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #49 on: June 25, 2012, 08:32:15 PM »

Jeff,
You're asking the right question.  We don't know where they might have been when the started their search.  Neither did they.  That's the whole point of a search pattern - it assumes you're "close", so the initial part of the pattern concentrates near where you currently are.  They thought they were at the coordinates of Howland, but did not see it.  They almost certainly knew there was some potential error in their navigation, so they would likely have known they were within a few miles of the precise location they were trying to reach.  They might have even known that the location of Howland Island wasn't quite where their information told them it should be.  In any case, they should not have been surprised to find empty ocean when they arrived at their best guess Howland location.  That's when they would start their search pattern.  That's also the point where they would have expected to use DF to steer towards the island/Itasca.
They may have used a box search pattern with too many miles between legs, thinking the island would be easy to spot.  They may have made some irrational decision about what sort of search pattern to fly, resulting in "splashed and sunk" in some unexpected location.  We just don't have enough information to know.
Cheers,
JohnO
 
Logged

Andrew M McKenna

  • Administrator
  • *
  • Posts: 666
  • Here I am during the Maid of Harlech Survey.
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #50 on: June 26, 2012, 02:41:29 AM »

Or they may have done what they said they were doing, flying north and then south on the LOP, looking for Howland, eventually finding Nikumaroro.

amck
Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #51 on: June 26, 2012, 05:35:06 AM »

Or they may have done what they said they were doing, flying north and then south on the LOP, looking for Howland, eventually finding Nikumaroro.

amck

It's a struggle to apply logic to this situation Andrew. An expanding search pattern as described must put them onto Howland or Baker. It didn't.
 
The best I can read into that is, they were close but couldn't see both islands, both?

They were nowhere near Howland or Baker. An expanding search pattern began so far away that they ran out of gas before getting within visual range of either.

Or, as you said, it was a search based upon the LOP and, running North and South. Maybe as a result of FN's calculation of where they were and, the best search pattern FN thought would be needed to nail Howland as a result of his calculation of where they were?

So, ran out of gas on an expanding search pattern or, run out of gas running North and South, or, running North and South led them to gardner?

Place your bets
This must be the place
 
Logged

John Ousterhout

  • T4
  • ****
  • Posts: 487
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #52 on: June 26, 2012, 08:57:30 AM »

What I've gleaned from some trivial personal experience and reading a lot of postings and reports is that Howland Island might have been harder to spot than expected.  AE/FN might have assumed they would be able to spot it at a distance of 20 miles, so chose to use 40 mile spacing on their search pattern and simply missed seeing it.  That's too far for anyone on the island to hear their engines.  That doesn't explain the radio silence, although the radio had been of little use to them so far.
If Fred was confident they were somewhere on or very close to the LOP, then the search pattern I imagine might be an elongated rectangle, not an expanding square, and not straying far from the LOP.  They might have arrived at the calculated LOP, flown north on it for a reasonable distance, turned around and flown south on it for some more reasonable distance, then offset their track 40 miles to East or West and turned back north, parallel to the initial LOP.  On the next southern leg, they might have been 40 miles on the other side of "the line", and simply missed spotting the island(s).  It's a big ocean.
If Fred's navigation was terribly far off-course to the south on the LOP, such an elongated rectangle search might bring Gardner island within sight.  Hard to imagine Fred being that far off course, but it seems like a slightly more reasonable scenario to me than simply following "the line" from Howland to Gardner.  If they arrived at Gardner, I don't believe it was by intentionally navigating there.  If they had the ability to navigate that accurately, they would seem to me much more likely to use it to fly to Howland.  Then again, it also seems reasonable to me that they would be squawking on the radio the whole time they were trying to find a safe place to land.
Cheers,
JohnO
 
Logged

Andrew M McKenna

  • Administrator
  • *
  • Posts: 666
  • Here I am during the Maid of Harlech Survey.
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #53 on: June 26, 2012, 09:43:32 AM »

I agree, given what we know, Fred "should" have been able to find Howland.  But he didn't.  That is a fact, so something must have gone terribly wrong, and it is hard to know exactly what.  Whatever it was, it is outside the norm, so trying to apply "norms" to this situation - search patterns, logical navigational solutions, "would have" statements - isn't going to solve it because they didn't get there, we have to look outside the norms to something unexpected.

Simply running out of gas as Gary would have us believe, is certainly one thing that could have gone wrong for them.

However, let's put it in context with other events we know happened.

We know that there were many post loss radio signals that were considered credible at the time, enough so that the Navy based their entire initial search on them.  If only one of the receptions was authentic, she had to be on land somewhere.  The Navy decided to go "search" the land and that was the initial tasking of their search.

We know that there was the skeleton of a castaway found on Niku with items, that would indicate that they were not a Pacific islander, that it looks like that castaway had a surplus US Navy sextant of the type Noonan liked to have with him as a back up, and that Gallager and the higher ups in the PISS system suspected (at least for a while) the castaway of being Mrs. Putnam.

The forensic analysis of those bones indicates a likelihood of being a european female of stature similar to AE.

There is a body of island lore about an aircraft wreck being there before the colonists arrived in 1930.

Apparent pre war aircraft material has been found there that is at least similar in nature to what we'd expect of the Electra parts.

The stuff at the 7 site, Etc. 

So, is initiating an expanding square search pattern and running out of gas the simplest answer given the context of some of the other things we know happened?  I think not.

None of these things are proven to be conclusively related to AE's disappearance, but they fit a hypotheses - an unproven thing - that TIGHAR has developed in order to try to understand all these oddities.  They do not fit with the hypothesis that she was flying an expanding square search and ran out of gas.

I think of it like a crime scene.  There are a lot of broken bits of evidence, some eyewitness accounts, rumors, odd stuff that may not initially make sense or seem related.  As the investigators of that crime scene, our job is to try to filter the important stuff, find out how things are related to each other, and integrate what's left into a coherent story of how the crime was committed and see if you can prove your hypotheses.  Right now we've got a pretty interesting story built around the evidence, but we're still trying to get the confession and close the case.  Hopefully this July will be the big break in the case.

Crashed and sank leaves a lot on the table (or swept under the rug?) that still needs to be explained.

Just my 2 cents.

amck
Logged

Jeff Carter

  • T2
  • **
  • Posts: 78
Question about Box Search Procedure
« Reply #54 on: June 27, 2012, 10:59:57 AM »

In reading "American Air Navigator", Mattingly (1944) on https://sites.google.com/site/fredienoonan/resources/american-air-navigator-mattingly-1944, I noticed the example problem shows the navigator starting the fixed square search from the plane's current location.  I thought the navigator first flew back to his best guess of the island's location and then started the fixed search. 

If Noonan had flown down the LOP as far south as required by his possible error limits, and then he decides to institute a fixed square search, doesn't Noonan first backtrack and fly north back up the LOP to his best calculation of Howland's location, and then institute a fixed square search?  Or does he start the fixed square search in his current location?

I also note Mattingly's example used a conservative V of only 5 nautical miles, although obviously it depends on atmospheric conditions.

Thanks.
Logged

Gary LaPook

  • T5
  • *****
  • Posts: 1624
Re: Question about Box Search Procedure
« Reply #55 on: June 27, 2012, 11:09:09 PM »

In reading "American Air Navigator", Mattingly (1944) on https://sites.google.com/site/fredienoonan/resources/american-air-navigator-mattingly-1944, I noticed the example problem shows the navigator starting the fixed square search from the plane's current location.  I thought the navigator first flew back to his best guess of the island's location and then started the fixed search. 

If Noonan had flown down the LOP as far south as required by his possible error limits, and then he decides to institute a fixed square search, doesn't Noonan first backtrack and fly north back up the LOP to his best calculation of Howland's location, and then institute a fixed square search?  Or does he start the fixed square search in his current location?

I also note Mattingly's example used a conservative V of only 5 nautical miles, although obviously it depends on atmospheric conditions.

Thanks.
Yes, you want to start your square search pattern at you best estimate of being over the destination so I have to disagree with Mattingly as do the other manuals available on my website here. All these manuals, including Mattingly, assume you got to the target location solely by DR. Where Mattingly goes wrong is to recommend flying beyond the DR position of the target by 20 to 30 minutes, 50 to 75 NM at a ground speed of 150 knots and twice that at a GS of 300 knots. If you started flying a search pattern there then it would take you a long time before you were back in the high probability area near the original DR position of the target. If you think about it, he is making that recommendation on the assumption the the DR is off by your being short of your target when it is twice as likely that you missed it off to the right or to the left.

It is a different proposition when you arrive at the target location by following an LOP because this eliminates any left-right error in the DR and you are left with only the along course, along LOP, error. In this case you would continue to fly the LOP and search beyond the target location by the size of your maximum estimate of the existing DR error. I have shown that when making the normal LOP approach to an island, the maximum DR error should not have exceeded 46 NM each way along the LOP, a total of 92 NM. Even using Ric's theory that they did not fly the standard LOP approach but went straight in to Howland then turned north to search along the LOP, the same north and south 46 NM legs from the Howland position would need to be searched. The first leg would take them 46 NM north and then the southbound leg would be 92 NM long taking past where Howland should be and ending up 46 NM south-southeast of Howland. In this case it would make sense to go back up the LOP those 46 NM to the closest estimate of where Howland should be found and to start the search pattern there. What makes this interesting, however, is that Baker is about 40 NM south-southeast of Howland so they could start a search pattern at the end of the 46 NM leg in order to find Baker and then it is a trivial flight from there to Howland. But, on the other hand, Howland should be easier to spot than Baker, it was bigger and had a ship next to it sending out smoke it still might have made more sense to spend the twenty minutes going back to the location of Howland prior to starting a search pattern a matter of judgment, these two choices are pretty evenly balanced.

gl
Logged

Andrew M McKenna

  • Administrator
  • *
  • Posts: 666
  • Here I am during the Maid of Harlech Survey.
Re: Question about Box Search Procedure
« Reply #56 on: June 28, 2012, 02:26:47 AM »


It is a different proposition when you arrive at the target location by following an LOP because this eliminates any left-right error in the DR and you are left with only the along course, along LOP, error. In this case you would continue to fly the LOP and search beyond the target location by the size of your maximum estimate of the existing DR error...

gl

Gary. 

If we agree that Noonan was able to navigate to the LOP through Howland, why would they choose to do anything other than search the LOP?  Why take a one dimensional problem - flying a line - and turn it into a two dimensional problem - the expanding square search?  This makes no sense to me.

As you've pointed out, all they have is latitude error to deal with, not longitude error, why induce it?  Howland has to be on the LOP, so why do anything other than search the LOP?  Isn't it better to extend the legs of the LOP search rather than burn fuel going east and west when there is no east / error?

And since they didn't find Howland, and by inference Baker, by flying the LOP, they had to be significantly farther N or S of Howland / Baker to not find it.  Using your 20 NM visibility, they had to be intersecting the LOP more than 66nm north of Howland or south of Baker.

You still haven't clued us in on the part of the LaPook hypotheses that explains how they got so far off course in the first place.

Andrew
Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #57 on: June 28, 2012, 06:43:42 AM »

Would it have something to do with the offset approach to Howland island? e.g. Planning to intercept the LOP at Baker island so, a turn to the left at the LOP would point you in the direction of Howland. So all you would have to do is fly the distance from Baker Island to Howland and Bingo, you're on top of Howland. Of course this relies on you intercepting the LOP at the point you planned to which may be a llittle different from the actual point?
I'm no expert in navigation but, I read and learn from Gary et al...

This must be the place
 
Logged

John Ousterhout

  • T4
  • ****
  • Posts: 487
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #58 on: June 28, 2012, 07:32:38 AM »

Andrew asks "... Isn't it better to extend the legs of the LOP search rather than burn fuel going east and west when there is no east / error?"
As I understand it (and hopefully Gary will kindly provide better analysis),  there is still some error to locating the LOP.  It's just a much smaller error than the north/south error.  That simplifies the subsequent search pattern to (mostly) 1 dimension.  If you're confident that your offset approach has placed your arrival on the line north of your target, then your search pattern is simplified even more, giving you high confidence which way to turn upon arriving at the approximate advanced LOP.  Assuming you arrive north of the target, you fly the 337 heading (accounting for cross winds), and scan the sea ahead and to the sides, covering a swath of ocean several miles to the sides as well as along the line.  I assume at some point a pilot/navigator would decide to turn the plane around and fly back "up" the line, thinking that they missed seeing the island due to being too far east or west of the line.  Call it a "modified box pattern".
Cheers,
JohnO
 
Logged

Chris Owens

  • T2
  • **
  • Posts: 64
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #59 on: June 29, 2012, 12:03:50 AM »

Take a look at this video, shot from a helicopter by Ted Waitt.  When I see how hard it is to spot Howland from a few miles out, I find it positively chilling to think of AE and FN flying a search pattern.  They really could have been right on it and still not seen it.

http://www.youtube.com/watch?v=m9c3yZ0xeHw
Logged
Pages: 1 2 3 [4] 5 6 ... 10   Go Up
 

Copyright 2020 by TIGHAR, a non-profit foundation. No portion of the TIGHAR Website may be reproduced by xerographic, photographic, digital or any other means for any purpose. No portion of the TIGHAR Website may be stored in a retrieval system, copied, transmitted or transferred in any form or by any means, whether electronic, mechanical, digital, photographic, magnetic or otherwise, for any purpose without the express, written permission of TIGHAR. All rights reserved.

Contact us at: info@tighar.org • Phone: 610-467-1937 • Membership formwebmaster@tighar.org

Powered by MySQL SMF 2.0.15 | SMF © 2017, Simple Machines Powered by PHP