Print

[Irvine John Donald]

Hi Harry

In the video of the Lae take off you see the little sort of dust that is thought to be the antenna snapping?  Would the broken piece still be attached so the two broken ends could be repaired?  Did anyone at Lae examine the runway after takeoff or even see what we did? 

[Martin X. Moleski, SJ]

With respect to radio communications, They (AE and Itasca) were successful in communicating with each other on 7500 kHz.  Why didn't they continue on that frequency?

The Itasca could only transmit in Morse Code (CW) on 7500 kcs.

AE and FN had refused to learn Morse Code and planned on doing everything with telephony.

[Martin X. Moleski, SJ]

... Did anyone at Lae examine the runway after takeoff or even see what we did?

This is the second time I am posting the link to a short article on the Lost Antenna in this thread.

You should be able to tell that it is a link, Irvine, because it is in a different color from the rest of this post.

If you run your cursor over the link, then click on it, it will open up the article in the wiki.

There you should be able to read the following:

R.E. Fullenwider (TIGHAR #0126) "spent some time in Lae during World War Two courtesy of Uncle Sam." As he remembers it, the old-timers there often said they hadn’t been surprised when Earhart was lost because "she left part of her trailing wire antenna laying on the runway" (TIGHAR Tracks, December 31, 1993).

TIGHAR has not been able to verify that anecdote.

[Harry Howe, Jr.]

Erik:
How would Howland have looked from 1000 feet for three cases A. approaching Howland from about 270 degrees, 50 miles out, 20 miles out, 5 miles out?   B. Same parameters (sp?) but approaching from SSE (157 to 337) and C. Approaching from NNW (337 to 157?

[Irvine John Donald]

Marty, I thank you for that post and the link.

I apologize for using an ipad as a computer. Great device but it has shortcomings especially in this editor. Very difficult to edit as the ipad doesnt let me scroll in the text window. Therefore if i hit "quote" the quoted text disappears off into the bottom of the text window and i cant getbto the bottom to edit. This is no fault of the editor as it happens with many forums i am part of.

Using the ipad however is no excuse for me not seeing that quoted article. I had not seen that one and look forward to reading it.

As you say it wasn't verified so can't be accepted as fact. Thats unfortunate. It still seems to me that so many coincidental anecdotes, scientific facts and hard evidence collected by TIGHAR over the years isn't yet enough to say the theory has been proven.

[Gary LaPook]

With all the arguments and speculation over the accuracy needed to be able to find Gardner from LOP , I have not really seen any mention of the 'vertical height' being a benefit.   This is especially true for the navigational accuracy (or inaccuracy) that would be needed to reliably find Gardner.  From looking below, it doesn't appear that a whole lot of science would be needed!

Here's food for thought...

If I am not mistaken, the elevation of NW end of the island combined with the height of the buka trees would put a total vertical height of 100' feet or so.  Combine that with a horizontal width of 1/4 - 1/2 mile profile from the NW end.  This gives you an object sticking up from the horizon with dimensions approximately 100' x 2500' in size.    It's not a ship blowing smoke, but it would seem hard to miss an object that size.

I am assuming that they had binoculars, but even if not, I would guess that an object of that size on the horizon would be fairly noticable even from far distances.  With or without cloud cover.  Does anyone by chance have any real pictures of a similar island from open ocean? 

Just for kicks, I created a vertical object 100' x 2500' in google earth, drew an approximate line from Gardner to Howland, and then estimated flying altitude that might seem reasonable. Here is what it might look like from a sampling of several different pespectives.  Of course this is just a simulation, but you get the idea.

Gardner from 50 miles out on LOP at 3000' altitude


Gardner from 25 miles out on LOP at 1500' altitude


Gardner from 10 miles out on LOP at 1000' altitude


It would seem to me that Gardner would be fairly easy to spot, even with crude DR techniques. 

Comments?

-----------------------------------

Seeing a tree 100 feet high from 50 NM away would be like seeing a dime laying in  the end zone from the 59 yard line. (If you are using statute miles then it would be like seeing that dime from the 52 yard line. )

We have no weather observations at Gardner for July 2, 1937 but we have another source we can use to evaluate the ability of Earhart to spot that island. According to the U.S. Navy Climatic Atlas of the World, Volume V, South Pacific Ocean (1979), page 183, in July in the vicinity of Gardner, 90% of time the visibility is less than 25 NM; 60% of the time less than 20 NM; 45% of the time less than 15 NM; and 32% of the time less than 10 NM.

So their views of Gardner from 25 and 50 miles away are much more likely to be as depicted in the attached photo than in your simulations.

The clouds depicted are scattered, about one octa (1/8th coverage). According to the same source, the cloud coverage in the vicinity of Gardner in July is greater than two octas 60% of the time and greater than one octa 80% of the time!

Still think it would have been easy for them to spot the island from far away?


(BTW, this photo also illustrates why you can't determine an LOP at sunrise. For you to time an accurate sunrise you must be able to see the actual blue sea horizon edge of the earth to measure the edge of the sun against and to take an accurate time. A one minute error in timing of a celestial observation results in a fifteen nautical mile error in the resulting computed longitude. This photo was taken on a day with really good weather but the clouds are obstructing the sea horizon so a sunrise observation would not have been possible and there is reason to believe that the weather was worse north and west of Howland on July 2nd. See: https://sites.google.com/site/fredienoonan/discussions/the-myth-of-the-sunrise-lop )
gl

[Mark Petersen]

As you can see, the three methods approach the question from completely different directions.  If the answers to all three questions overlap to some degree there is a high probability that the correct answer is within that area.

My apologies for posting so late, but I just came across this thread and found the premise very intriguing.  Working backwards obviously won't prove the Niku theory to skeptics, but if one assumes that the theory is correct, then working backwards may yield some interesting location/time information if the theory is correct.  What's intriguing is that this is information that, by its very nature, is something that can't be answered by any other means.

The overlap is the key.  If there is a lot of overlap, then then the location and time information is too loose and won't help to tie down the specifics.  If there is no overlap, then it points to an accuracy problem with one or more of the 3 questions.  But if there is a small degree of overlap, then bingo, I would think that this would allow us to be able to narrow down the time/location datum considerably.

There is still confusion about the LOP.
.
.
8. He has Amelia fly northwest first and at 20:00 she tries to use her loop antenna, but no luck.  Fred has her turn around and they retrace their steps, DRing southeast and still hoping that they'll find Howland.

Do we know for certain that FN had AE fly NW?  I realize that it fits the theory and is the logical thing to do, but are there other scenarios that might have FN and AE flying SE along the LOP without having traveled NW first?  I have an idea about this, but I don't want to derail this thread so I'll post it in another thread.   

Thanks yet again Ric.  So when the radio signals reported by Itasca are strength 5 and they believe she must be almost overhead, she is in fact 150 miles away?  This would mean that if the Monte Carlo simulation is correct then the distance they report her away is accurate also?

Yyyyyyep.

One thing that I never understood about the Monte Carlo simulation (https://tighar.org/smf/index.php/topic,169.msg1051.html#msg1051) is why it puts the 10E so far off of the LOP.  My meager knowledge of navigation (gleaned only from this forum) tells me that one would assume that FN would be able to measure the dawnline with a high degree of accuracy.  After dawn, he would also be able to measure and account for wind drift fairly accurately.  So his DRing to get to the LOP should be fairly accurate and with a high probability, even though his location on that line might be very inaccurate.  The Monte Carlo simulation would imply that this isn't the case though.   From the other thread (in my link above), I don't think that the Monte Carlo simulation incorporates the recent 3105 donut information, so it could be freshened up if someone were so inclined.

Note, this doesn't change the radio propogation numbers and therefore the received signal strength that is being discussed.  But might it not be possible that the 10E were actually further south of Howland and therefore closer to Gardner?  This would give them more fuel reserves to reach the island, and also more fuel to run the radio after a possible landing. 

[Martin X. Moleski, SJ]

Do we know for certain that FN had AE fly NW?  I realize that it fits the theory and is the logical thing to do, but are there other scenarios that might have FN and AE flying SE along the LOP without having traveled NW first?

As Tonto is said to have said to the Lone Ranger, "What you mean 'we,' white man?" 

I believe this is the natural reading of the last transmission: "We are flying north and south on the line 157-337."  337 is NNW; 157 is SSE.  She said "north" before "south."  Did she mean what she said?  Or was she just indicating a general plan to search the line both ways?

You get to make up your own mind whether the last transmission satisfies your own standards of certitude.

One thing that I never understood about the Monte Carlo simulation (https://tighar.org/smf/index.php/topic,169.msg1051.html#msg1051) is why it puts the 10E so far off of the LOP.

Wild guess: it puts the 10E so far off the LOP because if the aircraft had reached the broad band of LOPs that pass within visible range of Howland, they would have landed safely.  I don't know whether Randy coded it that way or whether it is the consequence of other assumptions he made.

It seems to me that any reconstruction has to incorporate some "zone of exclusion": the area that the plane must not have reached because, if it had gotten inside that zone, AE and FN would have seen Howland or Baker.  Calculating the size of the zone involves many imponderables:

• Fred's skill as a navigator.
• The quality of FN's instruments and chronometer.
• AE's pilotage--how well did she follow Fred's directions?
• Visibility on that morning at 1000' feet ASL.
• Visibility of Itasca's smoke signal (hotly disputed).
• Accuracy of charts.
• Accuracy of compasses.
• Field of vision.
• Visual acuity after a night at high altitude without oxygen.

Whether you come up with a large zone or a small one, we must agree that they didn't enter that region--for whatever reason.

Note, this doesn't change the radio propagation numbers and therefore the received signal strength that is being discussed.  But might it not be possible that the 10E were actually further south of Howland and therefore closer to Gardner?  This would give them more fuel reserves to reach the island, and also more fuel to run the radio after a possible landing.

The Monte Carlo simulation is an estimate of probabilities based on millions of variations in the variables.  It does not say that the couldn't have been somewhere other than in the most probable region.  All it says is that the further you get away from the mass of particular instances that clump together, the less likely that outcome seems from the standpoint of the assumptions made in construction of the simulation.  Strange things do happen.  People do get royal flushes in poker every now and again.  The longest of long shots occasionally wins a race and pays off ridiculously huge amounts of money. 

Which is a long-winded way of saying, "Yes, it's possible that they hit the band of LOPs that pass through the neighborhood of Howland and Baker well south of those too islands instead of west of the band of LOPs."

On this drawing, the letter "A" represents the whole cloud of possible routes that would have brought them close to Howland and Baker, but just a little too far west to catch sight of them, while still placing them on a LOP that comes close enough to Niku for them to find it.  "B" represents the whole cloud of possible routes that are within the Howland-Baker zone in terms of longitude (East-West location) but too far south in terms of latitude to find Howland and Baker after searching northward for a length of time that is unknown to us.

It seems to me that these are the two most likely scenarios.  In either case, we have to judge that Fred's navigation was off (east-west or north-south).

[Chuck Varney]

With respect to radio communications, They (AE and Itasca) were successful in communicating with each other on 7500 kHz.  Why didn't they continue on that frequency?

The Itasca could only transmit in Morse Code (CW) on 7500 kcs.

AE and FN had refused to learn Morse Code and planned on doing everything with telephony.

Two-way communication between Itasca and the Electra was not possible on 7500 kHz--regardless of transmission mode--because AE's Western Electric 13C transmitter could not transmit on 7500 kHz.

Chuck

[Martin X. Moleski, SJ]

With respect to radio communications, They (AE and Itasca) were successful in communicating with each other on 7500 kHz.  Why didn't they continue on that frequency?

The Itasca could only transmit in Morse Code (CW) on 7500 kcs.

AE and FN had refused to learn Morse Code and planned on doing everything with telephony.

Two-way communication between Itasca and the Electra was not possible on 7500 kHz--regardless of transmission mode--because AE's Western Electric 13C transmitter could not transmit on 7500 kHz.

Technically, you are correct in your narrow definition of "on 7500 kHz."

But she could have transmitted on 3105 kHz and listened on 7500 kHz, enabling two-way communication, if she had known Morse Code or if they had their 7500 kHz transmitter set up for voice operations.

[Ric Gillespie]

With respect to radio communications, They (AE and Itasca) were successful in comunicating with each other on 7500 kHz. 

No they weren't.  Earhart asked for a "long count" on 7500.  A long count, as you probably know, is counting from 1 to 10 and back to 1 slowly by voice.  Itasca could not send voice on 7500. They had told Earhart that in a communication several days previous. She apparently was not paying attention.  They did what they could.  They sent the Morse code letter "A" (dit dah) - the pre-arranged letter for signals for Itasca.  (The USS Ontario was supposed to send the letter "N.") Earhart hear the "A"s but could not reply in Morse and didn't even try.  In short, Earhart heard the signals but there was no communication.

[Erik]

Erik:
How would Howland have looked from 1000 feet for three cases A. approaching Howland from about 270 degrees, 50 miles out, 20 miles out, 5 miles out?   B. Same parameters (sp?) but approaching from SSE (157 to 337) and C. Approaching from NNW (337 to 157?

Good point. 

Howland's vertical profile (15' or so max.) would be so minimal that it wouldn't be high enough to distrupt the natural horizon's profile. 

It appears it would take being as close as 1 mile and as low as 100' feet before the height of the island would even be a factor. 

More likely, one would need to be in a rowboat a hundred yards offshore before the natural horizon would be disrupted by Howland's vertical height.

[Mark Petersen]

As Tonto is said to have said to the Lone Ranger, "What you mean 'we,' white man?" 

We as in Tighar members.  I know who the Lone Ranger is, but I won't speculate on the identity of Tonto.   

I believe this is the natural reading of the last transmission: "We are flying north and south on the line 157-337."  337 is NNW; 157 is SSE.  She said "north" before "south."  Did she mean what she said?  Or was she just indicating a general plan to search the line both ways?

Ahh, okay I can see how one can interpret that.  She apparently also used the word 'AND' indicating that both directions had been traveled.  I used the word apparently, because as mentioned in Tighar Tracks, the radio logs are a bit of a mess and the exact wording might not have been recorded properly. 

Wild guess: it puts the 10E so far off the LOP because if the aircraft had reached the broad band of LOPs that pass within visible range of Howland, they would have landed safely.  I don't know whether Randy coded it that way or whether it is the consequence of other assumptions he made.

It would be interesting to see what probability the Monte Carlo Simulation assigned to the accuracy of the dawnline measurement vs the probability of the dead reckoning needed to hit the north-south LOP position.  As a fairly new Tighar member, I'm not aware of the early discussions that took place at the time that the Simulation was generated.  But there have been a large number of relatively new forum discussions about the LOP that all lead to the premise that navigating to the LOP itself with a morning dawnline should be easy, but hitting a north-south point on the LOP is difficult. 

This isn't to say that the Monte Carlo Simulation is wrong.  It might be the case that the simulation took into account the high accuracy of a dawnline measurement and the combined weight of all of the other factors still end up giving us the probability that we have (of hitting well SW of the target).  It might be the case then that the Simulation is telling us some important clues, such as FN might have made a simple mistake or was using the inaccurate charts that have been much discussed.  Also as you have pointed out the Monte Carlo Simulation is only giving us the probability of various outcomes, not necessarily the outcome that actually occurred.

The Monte Carlo simulation is an estimate of probabilities based on millions of variations in the variables.  It does not say that the couldn't have been somewhere other than in the most probable region.  All it says is that the further you get away from the mass of particular instances that clump together, the less likely that outcome seems from the standpoint of the assumptions made in construction of the simulation.  Strange things do happen.  People do get royal flushes in poker every now and again.  The longest of long shots occasionally wins a race and pays off ridiculously huge amounts of money. 

Exactly.  The thought has also occurred to me that this whole discussion of working backwards from our 3 assumptions, could in itself be food for another Monte Carlo Simulation.  It shouldn't be that difficult for someone to drum up a simulation in Matlab.  Even MS Excel has a Monte Carlo Simulation option I believe. 

Incidentally, one other thing about Monte Carlo Simulations.  They are very useful in "what-if" scenarios whereby various parameters are changed to gauge their impacts.   From that standpoint it becomes a living mathematical model that can be used for endless "what-if" scenarios.  Perhaps Tighar should revisit this tool and perhaps use it on an ongoing basis as new data is found.

Which is a long-winded way of saying, "Yes, it's possible that they hit the band of LOPs that pass through the neighborhood of Howland and Baker well south of those too islands instead of west of the band of LOPs."

On this drawing, the letter "A" represents the whole cloud of possible routes that would have brought them close to Howland and Baker, but just a little too far west to catch sight of them, while still placing them on a LOP that comes close enough to Niku for them to find it.  "B" represents the whole cloud of possible routes that are within the Howland-Baker zone in terms of longitude (East-West location) but too far south in terms of latitude to find Howland and Baker after searching northward for a length of time that is unknown to us.

It seems to me that these are the two most likely scenarios.  In either case, we have to judge that Fred's navigation was off (east-west or north-south).

Nice graphic.  This is the crux of the error isn't it?  Speaking as a novice, it's easy for me to believe that Fred's navigation was well off in a north-south direction because of the overcast that likely prohibited the course corrections needed for the overnight flight.  Simultaneously, it's more difficult to believe that Fred would have blown the dawnline measurement.  So both of those together seem to strengthen the notion that the flight was on the LOP probably well south (probably 150 miles south per the Strength 5 radio reception and 3105 donut discussions).   But most of the Tighar assumptions seem to be that the flight was actually much further west, which is what has me puzzled.

[Erik]

Still think it would have been easy for them to spot the island from far away?

I guess I should have been more clear in the original intent of my posting.  Perhaps using the word 'easier' as opposed to 'easy' would have been a better choice.

The point was that any vertical obstruction on an otherwise featureless horizon would be a benefit to any observer, regardless of visibility, cloud cover, fatigue, etc.  The greater magnitude of that vertical disturbance, the greater the benefit.

The larger and taller the object is would only increase the ability for naked eye to detect it when scanning the natural, unobstructed horizon.  Gardner certainly fits that category.  Its vertical profile (from the NW) is approximately 100' high by nearly 1/2 mile wide.

That was the intent of the simulation was to demonstrate the 'trigonometry' that would affect the visual cues while scanning the horizon's naturally flat profile.  It would certainly benefit the observer vs. a situation with no vertical obstructions at all.

[Irvine John Donald]

The navigation issue is part of what hurt their chances to find Howland. Another question for our navigators and experts. If a navigator flies long enough with one pilot does the navigator make allowances for that pilots flying habits?  For instance when flying a long distance without distinct land marks would a pilot tend to veer a bit more to the right or left.  At the end of a long leg would they have to fly 50 miles toward the target due to that pilots natural "drift"?  Would a navigator ever factor that in?  Would an aircraft have a tendency to "pull" in one direction more than another?  Without celestial shots and at night can you tell what you're drift is?