Advanced search  
Pages: 1 ... 3 4 [5] 6 7 ... 10   Go Down

Author Topic: LaPook Hypothesis: Box Search around 157-337  (Read 137141 times)

Andrew M McKenna

  • Administrator
  • *
  • Posts: 666
  • Here I am during the Maid of Harlech Survey.
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #60 on: June 29, 2012, 03:27:21 AM »

OK, so here you are, Fred, flying east towards Howland.  Let's just say you haven't been able to get a celestial sighting for quite some time due to an overcast, and you are eagerly awaiting the dawn to get a sun shot.

Up comes the sun, and bing, you get your first sunshot from which you have already computed the time difference calculations between sunrise at Howland, so presto, you know exactly how many miles from your position to the advanced LOP, plus or minus what - 10% of the distance plus the uncertainty of the sun shot which I thing has been estimated at 10nm for a celestial sighting if you are good - Gary help me out here.

So, at sunrise at Howland, when Earhart is calling saying "200 miles out" we might expect that the error in finding the exact LOP through Howland is going to be 20nm plus the 10mn uncertainty in the sun shot = 30nm?  Seems like a lot for a 200 nm distance flight, am I getting this right?

But, as Gary points out, Noonan could have derived the same LOP through Howland for at least another hour after sunrise (I think someone else said 3 hours after sunrise), and if so, they would have been not 200 nm out, but another 130nm closer to the LOP, so lets assume another sun shot 70 nm out, so our error might shrink to 17 nm.  Gary had it figured down to 14 nm, 7 nm on either side of the LOP, in a previous post #504 on this page http://tighar.org/smf/index.php/topic,555.msg11163.html#msg11163,

And, given that 90 minutes go by between the 200 miles out, and the "we must be on you" message, the 200 miles out message, would seem to be fairly accurate.

So, then when you hit the LOP, you are likely within 8.5mn of the actual LOP, well within visual spotting distance of an island, at least in theory.

So, if I know I'm flying on a line and that Howland is within 17 and perhaps 8.5 or even 7 miles of my line, why fly 40 mn off the the east and west of that line looking for something that I know isn't there?

Assuming Noonan chose to be North of the target is just that, an assumption.  Might make sense then, and in our minds now, but given that they were expecting DF assistance, and the fact that AE says they are flying "north and south" on the line 157 - 337, it does seem plausible that they did not employ the offset landfall method and were barreling straight in.  Why on earth would she say she was flying "north and south" if they had used the offset method?  Why are we going to take a "should have - would have statement" offered in hindsight over a primary contemporaneous statement from AE herself?  As I said before, something they did had to be out of the norm, otherwise they "would have - should have" made it to Howland, and we'd not be discussing it here.

It is Gary who is suggesting the expanding square search pattern, not a "modified box pattern", and I'm looking forward to hearing his reasoning why that would be employed over simply searching the LOP 157 - 337, as indicated in the radio traffic since it seems to me that the margin of error for the LOP would be pretty reasonable to deal with. 

They had to hit the LOP well away from their target - beyond search pattern coverage - otherwise they would have found Howland / Baker.  The LaPook hypotheses has yet to suggest why they were so far off, what went wrong?

Andrew

« Last Edit: June 29, 2012, 03:36:44 AM by Andrew M McKenna »
Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #61 on: June 29, 2012, 06:22:48 AM »

This is exactly the point that I have the most difficulty with Andrew...

They had to hit the LOP well away from their target - beyond search pattern coverage - otherwise they would have found Howland / Baker

I have read, on this forum and from the internet how good Freddie Noonan was at navigation, Pan Am clippers, ships and so on. I agree, he was probably the best at the time. And yet, here we are discussing the probability of a search pattern that may have been implemented failing to locate Howland. From this I put forward 2 scenarios:

They simply ran out of gas
They were so far of course the search pattern missed Howland

Notice how both are dependent on each other!
This must be the place
 
Logged

John Ousterhout

  • T4
  • ****
  • Posts: 487
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #62 on: June 29, 2012, 08:13:37 AM »

"...Why on earth would she say she was flying "north and south"?"
It occurs to me that her poor understanding of DF theory might have included the idea that it would be helpful to a DF operator on Itasca to know what direction she was moving, thinking it would help avoid the ambiguity of a reciprocal bearing.  I don't like this theory because it would have been more helpful to simply state that she was flying 'south on the line for the next 2 minutes, please take a bearing on us', if her understanding of DF was as poor as I've presented.  We just don't 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 #63 on: June 29, 2012, 09:10:28 AM »


They simply ran out of gas
They were so far of course the search pattern missed Howland

Notice how both are dependent on each other!

Yes, I suppose they are interdependent.  If they run out of gas first, their search missed Howland, and if their search misses Howland, the'll eventually run out of gas.

Where they intersected the LOP is the big question, and of course we can't know that, but we do have the Monte Carlo analysis that suggests a probability of putting them south, perhaps significantly. 

The post loss signals would indicate that they made it to land somewhere, relatively intact.  Unless you want to completely dismiss that body of data, as Gary prefers, you have to think about what land is within range, and how would they have ended up there.  If they had hit the LOP well north of Howland, and their subsequent search didn't find Howland, they also wouldn't have found any other land, as there isn't any up there.

I'm with the folks who think they were trying to fly the 157 - 337 LOP searching for Howland, and ended up finding Niku instead.  Requires that they intersect the LOP well south of Howland, but evidently it happened given the Monte Carlo analysis, post loss signals, and other lines of evidence like the castaway, sextant box, aircraft material, radio triangulation, and etc. TIGHAR is attempting to prove are related.

Andrew

Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #64 on: June 29, 2012, 10:18:36 AM »

Where they intersected the LOP is the big question, and of course we can't know that, but we do have the Monte Carlo analysis that suggests a probability of putting them south, perhaps significantly. 

And of course Andrew, there is also the area THEY thought they were in to consider. If FN sets a corrective course from where THEY thought they were when in actual fact they were at the location the Monte Carlo analysis puts them then, it simply puts them even significantly further away from Howland ???
This must be the place
 
Logged

Jeff Carter

  • T2
  • **
  • Posts: 78
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #65 on: June 29, 2012, 10:48:26 AM »


I have read, on this forum and from the internet how good Freddie Noonan was at navigation, Pan Am clippers, ships and so on. I agree, he was probably the best at the time. And yet, here we are discussing the probability of a search pattern that may have been implemented failing to locate Howland. From this I put forward 2 scenarios:

They simply ran out of gas
They were so far of course the search pattern missed Howland


And, there is a third scenario, they flew a well-structured, textbook-correct search pattern with visibility factor ("V") too high, and simply missed spotting the island even though they passed within "V" distance of the island. 

Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #66 on: June 29, 2012, 12:09:31 PM »

Wasn't there some ambiguity over the exact location of Howland Island on maps and charts available at the time, 5 + nm as well ?
This must be the place
 
« Last Edit: February 27, 2013, 11:46:23 AM by J. Nevill »
Logged

C.W. Herndon

  • T5
  • *****
  • Posts: 634
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #67 on: June 29, 2012, 12:53:48 PM »

Jeff, in one of the archived forum threads, and I have not figured out how to copy them yet, there was this:

"Date: Sat, 1 Dec 2007
From: Randy Jacobson
Subject: Landing on Gardner

The discrepency in Howland's position back in 1937 was on the order of 6 miles, not 62 miles.

The true position is 48', 6" N, 176* 38' 12"W
The reported position is 49' 0"N, 176* 43' 09"W, as stated in the American Practical Navigator, 1936 edition.

Both positions were known to Bill Miller, who coordinated AE's first flight attempt,....."
Woody (former 3316R)
"the watcher"
 
« Last Edit: July 03, 2012, 06:35:03 PM by Martin X. Moleski, SJ »
Logged

Jeff Victor Hayden

  • T5
  • *****
  • Posts: 1387
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #68 on: June 29, 2012, 05:31:46 PM »

Which of these would have known that Woody, if any.
Paul Mantz, Amelia Earhart, Harry Manning and Fred Noonan
This must be the place
 
Logged

C.W. Herndon

  • T5
  • *****
  • Posts: 634
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #69 on: June 29, 2012, 06:44:26 PM »

Jeff, here is the best information I could find. From what I can tell from the reference--nobody knows if any of the crew was made aware of the differences in location for Howland. Apparently Clarence Williams didn't know since he used the wrong location on his strip map flight plan, at least the one for the first attempt.

http://tighar.org/Projects/Earhart/Archives/Research/ResearchPapers/Worldflight/2ndattemptroute.html
Woody (former 3316R)
"the watcher"
 
« Last Edit: June 29, 2012, 06:58:50 PM by C.W. Herndon »
Logged

Gary LaPook

  • T5
  • *****
  • Posts: 1624
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #70 on: July 02, 2012, 05:54:24 AM »


They simply ran out of gas
They were so far of course the search pattern missed Howland

Notice how both are dependent on each other!

Yes, I suppose they are interdependent.  If they run out of gas first, their search missed Howland, and if their search misses Howland, the'll eventually run out of gas.

Where they intersected the LOP is the big question, and of course we can't know that, but we do have the Monte Carlo analysis that suggests a probability of putting them south, perhaps significantly. 

The post loss signals would indicate that they made it to land somewhere, relatively intact.  Unless you want to completely dismiss that body of data, as Gary prefers, you have to think about what land is within range, and how would they have ended up there.  If they had hit the LOP well north of Howland, and their subsequent search didn't find Howland, they also wouldn't have found any other land, as there isn't any up there.

I'm with the folks who think they were trying to fly the 157 - 337 LOP searching for Howland, and ended up finding Niku instead.  Requires that they intersect the LOP well south of Howland, but evidently it happened given the Monte Carlo analysis, post loss signals, and other lines of evidence like the castaway, sextant box, aircraft material, radio triangulation, and etc. TIGHAR is attempting to prove are related.

Andrew
Here is the reason why they didn't fly so far south as to hit Gardner.

I have put the same information up many times before and everyone just ignores it, they just go past it like it wasn't there, kinda like whistling past the graveyard. See:

https://tighar.org/smf/index.php/topic,383.msg5653.html#msg5653

https://tighar.org/smf/index.php/topic,592.msg9901.html#msg9901

https://tighar.org/smf/index.php/topic,555.msg12109.html#msg12109

https://tighar.org/smf/index.php/topic,533.msg7166.html#msg7166

https://tighar.org/smf/index.php/topic,592.msg9901.html#msg9901

https://tighar.org/smf/index.php/topic,320.msg6668.html#msg6668

gl

Logged

Gary LaPook

  • T5
  • *****
  • Posts: 1624
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #71 on: July 02, 2012, 06:13:38 AM »



Up comes the sun, and bing, you get your first sunshot from which you have already computed the time difference calculations between sunrise at Howland, so presto, you know exactly how many miles from your position to the advanced LOP, plus or minus what - 10% of the distance plus the uncertainty of the sun shot which I thing has been estimated at 10nm for a celestial sighting if you are good - Gary help me out here.

So, at sunrise at Howland, when Earhart is calling saying "200 miles out" we might expect that the error in finding the exact LOP through Howland is going to be 20nm plus the 10mn uncertainty in the sun shot = 30nm?  Seems like a lot for a 200 nm distance flight, am I getting this right?
By Jove, you've got it!
Quote

But, as Gary points out, Noonan could have derived the same LOP through Howland for at least another hour after sunrise (I think someone else said 3 hours after sunrise), and if so, they would have been not 200 nm out, but another 130nm closer to the LOP, so lets assume another sun shot 70 nm out, so our error might shrink to 17 nm.  Gary had it figured down to 14 nm, 7 nm on either side of the LOP, in a previous post #504 on this page http://tighar.org/smf/index.php/topic,555.msg11163.html#msg11163, ...

Andrew
Well I thought you got it. Using your example of the sun observation being taken 70 NM prior to intercepting (and no additional observations) the line that runs parallel to the sun line, 157-337°, and passes through Howland, at the time of intercepting the line the uncertainty extends 14 NM on each side of the line, a total of a 28 NM wide band. The original uncertainty of plus and minus 7 NM in the observation plus the additional 7 NM in each direction due to the 70 NM DR leg (10% of 70 NM.) And the DR uncertainty continues to grow at the same rate, 13 knots in every direction (10% of the 130 K approximate true airspeed.) If they were first on the line at 1912 Z after flying 70 NM (your example) then at 1912 Z the uncertainty extended 14 NM on both sides of the line. AT 2012 Z, one hour later, the uncertainty had grown to 27 NM on both sides of the line so would not be accurate enough to assure spotting the island. And, in fact the 14 NM uncertainty at 1912 Z may have been enough to keep them from seeing Howland. (I mean to look further into the visibility of the island issue in another post.)
gl
« Last Edit: July 02, 2012, 08:05:46 PM by Gary LaPook »
Logged

JNev

  • T5
  • *****
  • Posts: 778
  • It's a GOOD thing to be in the cornfield...
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #72 on: July 02, 2012, 09:49:05 AM »

The ambiguity grows by the minute out there under celestial and DR, doesn't it?  Wow.

I like what Marty recently noted - any landfall at Gardner would likely have included a great deal of seredipitous blundering!

Kinda goes back to 'how can you find point b if you don't know where point a is' I think, if I'm still following you right Gary.  Excellent points.

LTM -
- Jeff Neville

Former Member 3074R
 
Logged

Jeff Carter

  • T2
  • **
  • Posts: 78
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #73 on: July 02, 2012, 10:54:29 AM »


 AT 2012 Z, one hour later, the uncertainty had grown to 27 NM on both sides of the line so would not be accurate enough to assure spotting the island. And, in fact the 14 NM uncertainty at 1912 Z may have been enough to keep them form seeing Howland. (I mean to look further into the visibility of the island issue in another post.)
gl

FWIW, I can't find any evidence that Noonan or Earhart had any real-world experience with search patterns.  I know Noonan could calculate a search pattern in his sleep, and Earhart could fly it no problem.   But, if Noonan ever had to fly a search pattern with PanAm, he would have had a large, well-trained crew trying for a visual sighting, and the crew would have been looking for large islands like Wake, Midway, etc. 

But did either one have any experience in estimating "V" visibility?  Had either one ever flown a search pattern to try to spot a ship (Itasca)?  Especially with a pilot in the left seat with limited visibility, and a navigator in the rear distracted with double-checking navigation calculations?  Had either FN or AE ever visited Howland Island, or received any kind of detailed briefing of how small and flat it was?

Logged

Chris Johnson

  • T5
  • *****
  • Posts: 1069
  • Trying to give a fig but would settle for $100,000
Re: LaPook Hypothesis: Box Search around 157-337
« Reply #74 on: July 03, 2012, 04:06:16 AM »

Jeff,

good points to which i'm sure GLP will have some kind of informed answer to stimulate discussion  ;D
Logged
Pages: 1 ... 3 4 [5] 6 7 ... 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