Print

[Martin X. Moleski, SJ]

Thread drift happens.  We're starting to go around in circles (26 December 2010) because people are commenting about how to interpret the results of the aerial search of Gardner Island on 9 July 1937.

I plan to do my level best to move posts out of other threads into this one if that is the main focus of the post.  That means that this thread is going to be a little jumbled up, but I hope, over the long run, that it will become the place to ask and answer questions about the aerial search.

Some essential reading:

• Lt. John O. Lambrecht
• Lambrecht's Report
• Colorado Search
• "An Answering Wave: Why the Navy Didn’t Find Amelia"
• Randy Jacobson: "The Colorado Search"
• "What did the Navy searchers see when they flew over Nikumaroro in 1937?"
• "An Aerial Tour of Nikumaroro"

[Mark Petersen]

Here it is.  If anyone knows how to adjust the default focal length that Google Earth uses, please let me know as I think that's the one thing that is keeping it from fitting more closely.  As I mentioned though, it does show the compass heading as being W rather than N (as mentioned in the Tighar archives), my best guess of the altitude that Lambrecht was flying was 1000 ft +- 300.  But that's just a guess.  I'm a newbie with Google Earth though and I'm sure others can do better.

[Bill Lloyd]

Here it is.  If anyone knows how to adjust the default focal length that Google Earth uses, please let me know as I think that's the one thing that is keeping it from fitting more closely.  As I mentioned though, it does show the compass heading as being W rather than N (as mentioned in the Tighar archives), my best guess of the altitude that Lambrecht was flying was 1000 ft +- 300.  But that's just a guess.  I'm a newbie with Google Earth though and I'm sure others can do better.

Mark, I did the same thing a couple of weeks ago and my estimate of altitude was 2000-2500ft. It appears to me that you are on the correct heading but a bit low. Just my guess.

[Erik]

...If anyone knows how to adjust the default focal length that Google Earth uses...my best guess of the altitude that Lambrecht was flying was 1000 ft +- 300....

Mark/Bill

Lining up photos like this is a very tricky business!  

In this particular case, focal length is not as important as is the azimuth or axis along which the photo was taken (or LOP so to speak   ).  We also don't know if any enlarging or cropping was done during the photo developing process, making it even more difficult.  

We are dealing with an object that is in a 2 dimensional plane.  Since Nikumaroro island is at sea level and relatively no elevation, it makes it very difficult to line it up in a 3-D world, not to mention anomalies that can occur with digital imaging and aspect ratios once a photo has been 'computerized'.

The best we can do at this point is to find the axis upon which the photo was taken.  From there we can make an educated guess where and how high the photo was taken.

In the image below you can see the original shoreline in yellow draped onto google earth, along with a yellow dot with representing a real-world position on the ground.  You'll notice how the shoreline does not match exactly and the yellow dot is not framed correctly.  In the third image, if we rotate google earth onto the axis in which the photo was taken, we now see that the shoreline and dot line up much better.

To create this yourself in google earth using the Snapshot View -> Properties.  Enter these coordinates to get a representation of what the axis looks like from a birds-eye.  
Latitude: 4°40'59.90"S
Longitude: 174°30'37.98"W
Range: 3339m
Heading: -67.000000°
Tilt: 86.000000°

Click HERE for full size image.

[Erik]

Below is an example of the axis along which the photo was most likely taken.  It is possible that the photo was taken from an altitude of 2000', but it would have been nearly 3.5 miles offshore to maintain the axis.  I would hedge my bets that the photo was taken somewhere in the neighborhood of 800' +/-.

[Ted G Campbell]

Keep in mind that this picture probably is not out the windshield of the aircraft by Lambrecht, he's busy flying the plane, it most likely was taken out a side window by his crewmate.
Ted Campbell

[Ric Gillespie]

Keep in mind that this picture probably is not out the windshield of the aircraft by Lambrecht, he's busy flying the plane, it most likely was taken out a side window by his crewmate.

The Vought O3U-3 was a two-place, tandem seating, open-cockpit biplane.  The photo had to have been taken by the observer in the rear cockpit.  A photo taken from the front cockpit would show wires and struts in the foreground.

[Bill Lloyd]

Below is an example of the axis along which the photo was most likely taken.  It is possible that the photo was taken from an altitude of 2000', but it would have been nearly 3.5 miles offshore to maintain the axis.  I would hedge my bets that the photo was taken somewhere in the neighborhood of 800' +/-.

You may be correct. When Lambrecht left Mckean bound for Gardner his flight plan appears to have been 1000’ altitude direct route to Gardner on a heading of 198 degrees true at about 78 statute miles.  If, upon arrival at Gardner on this heading, his observer took the photo looking out the starboard side of the airplane, the probable direction of the camera would be 90 degrees to the heading of the plane or about 288 degrees +/-.  It would be logical to assume that the photo was taken at 1000’ since this was the arrival altitude and the probable course of action was then to descend to search the island.  There would be no reason to climb to a higher altitude.

That would be the logical assumption except while looking at the black and white photo, my sense of height above the ground (AGL) tells me that the photographer just seems to be higher than 1000’.  A reference point would be the beach which measure about 65 feet in width or about the width of a four lane highway. 

[Erik]

You may be correct. When Lambrecht left Mckean bound for Gardner his flight plan appears to have been 1000’ altitude direct route to Gardner on a heading of 198 degrees true at about 78 statute miles.  If, upon arrival at Gardner on this heading, his observer took the photo looking out the starboard side of the airplane, the probable direction of the camera would be 90 degrees to the heading of the plane or about 288 degrees +/-.  It would be logical to assume that the photo was taken at 1000’ since this was the arrival altitude and the probable course of action was then to descend to search the island.  There would be no reason to climb to a higher altitude.

That would be the logical assumption except while looking at the black and white photo, my sense of height above the ground (AGL) tells me that the photographer just seems to be higher than 1000’.  A reference point would be the beach which measure about 65 feet in width or about the width of a four lane highway. 

Good call!

The 198 flight path heading is almost a perfect ninety-degree angle to the 293 photo axis heading being assumed.  The 1000' altitude fits nicely too.  Remember though it was reported that they were 'zooming' up-and-down to atttract attention and create noise.  Perhaps an elevation change of +/- 250 feet would allow enough noise without too much abrupt flight distruption.  Puting a window of 750'-1250' - just a thought.  I also remember somewhere I read that a 400' altitude was being used.  Perhaps they were in a shallow decent down to that level.  400' would not leave much room for 'zooming' though.

I agree that the photo looks a bit higher than 800' too.  Perhaps the tall, narrowness of it is creating an optical illusion.  1000' altitude on the projected photo axis puts the plane (camera) at approximately ~1 mile off shore.   That seems reasonable to me.

Another view showing flight path, photo axis, and photo outline together.

[Bill Mangus]

Perhaps the picture was taken after the "repeated circling and zooming failed to elicit an answering wave. . . ." as the flight completed their last circle of the island and prepared to head SE to Caroldelet Reef.  I'd think getting a picture would be a low or last priority, waiting until after making one low pass (50') to wake people up, distrub the birds and conduct the rest of the search from 400-500'.  Ric's book describs the search in chapter 20; "As in the case of the subsequent search of the rest of the Phoenix Islands one cirlce at fifty feet around M'Kean aroused the birds to such an extent that further inspection had to be made from an altitude of at least 400 feet"  (Lanbrecht, "Aircraft Search for Earhart Plane).

He seems to be saying that the same procedure was followed on initial arrival at every search location.  I'd think one circle or even a pass down the center of the lagoon would be loud enough to arouse anyone capable of responding, but whether or not they'd be seen from 400-500' on subsequent passes/circles is difficult to say.  The picture would be almost an afterthought, waiting until the search was completed.  I'd want my eyes looking at the ground until I was sure there was nothing to be seen, then worry about the photography.

[Erik]

Beautiful lineup--thanks!

Thank you...

The Vought O3U-3 was a two-place, tandem seating, open-cockpit biplane.  The photo had to have been taken by the observer in the rear cockpit.  A photo taken from the front cockpit would show wires and struts in the foreground.

Any word on why the photo is cropped so tall and narrow?  At first I thought it was perhaps to eliminate the struts and guy wires, or even possibly I thought about some type of unusual 'panoramic' film being  used by the military.  Why crop out the island like that?

It would be interesting to find out if the carrier ships had photo developing labs on board.  I would imagine they did, so that real-time analysis of the photos could be made.

PS = Did you happen to notice the small 'dot' at the far top-right corner of the island in the original photo?  Too far away from Norwich City and awfully close to Nessis.  It appears to be just offshore and not part of the island.  Interesting and coincidental...

[Erik]

Perhaps the picture was taken after the "repeated circling and zooming failed to elicit an answering wave. . . ." as the flight completed their last circle of the island and prepared to head SE to Caroldelet Reef.  I'd think getting a picture would be a low or last priority, waiting until after making one low pass (50') to wake people up, distrub the birds and conduct the rest of the search from 400-500'.  Ric's book describs the search in chapter 20; "As in the case of the subsequent search of the rest of the Phoenix Islands one cirlce at fifty feet around M'Kean aroused the birds to such an extent that further inspection had to be made from an altitude of at least 400 feet"  (Lanbrecht, "Aircraft Search for Earhart Plane).

He seems to be saying that the same procedure was followed on initial arrival at every search location.  I'd think one circle or even a pass down the center of the lagoon would be loud enough to arouse anyone capable of responding, but whether or not they'd be seen from 400-500' on subsequent passes/circles is difficult to say.  The picture would be almost an afterthought, waiting until the search was completed.  I'd want my eyes looking at the ground until I was sure there was nothing to be seen, then worry about the photography.

Completely understand.   I could see an arguement for both ways though.

One thing that is not completely clear to me is why they would not have taken several bunches of pictures of each reef/island back then?  It would seem obvious that they had a developing lab of some sort aboard ship.  Then, after the flight, the photos could have been developed in a matter of hours, analyzed, and additional visits could have been ordered if they saw something that warranted additional merit - such as 'signs of previous habitation'.  Why not take a picture and analyze it greater detail later under more sterile conditions?  You would essentially be buying yourself unlimited viewing time and as many eyes as needed by trained photo interpruters.

[Daniel Paul Cotts]

One thing that is not completely clear to me is why they would not have taken several bunches of pictures of each reef/island back then?

Maybe air search had not progressed to the point where that was standard procedure. Ah, the could have beens! Looking at the Luke Field inventory shows a flare pistol and fourteen flare rounds. Items #23 and #33. Maybe they were left behind for the second flight. Maybe they were not recovered from the plane after landing. We will never know. Back to the regular thread - I'm in complete awe of the photographic manipulation above.

[Mark Petersen]

Hi Erik,

Nice analysis!  I think that 1 mile offshore at 1000' seems about right to me too.  For what it's worth what I keyed off of was the angle of the lagoon that is off in the distance (see the circled areas below).  I placed the viewer at what seemed like a reasonable distance from Niku and then changed the elevation to get the correct angle of the lagoon in the circled areas because I knew that it wouldn't be affected as much by focal length as the shoreline (areas of Niku closest to the viewer).  If you position the viewer at 1 mile offshore and at 1000' how well does the shoreline matchup? 

[Erik]

Hi Erik,

Nice analysis!  I think that 1 mile offshore at 1000' seems about right to me too.  For what it's worth what I keyed off of was the angle of the lagoon that is off in the distance (see the circled areas below).  I placed the viewer at what seemed like a reasonable distance from Niku and then changed the elevation to get the correct angle of the lagoon in the circled areas because I knew that it wouldn't be affected as much by focal length as the shoreline (areas of Niku closest to the viewer).  If you position the viewer at 1 mile offshore and at 1000' how well does the shoreline matchup? 

Thanks Mark,

Lining up imagery like this requires aligning all three axes, vertical tilt, horizontal tilt, and rotation.  Pitch, roll and yaw to pilots.  Doing this is very hard.   

Finding out where the photo was taken, latitude, longitude and altitude is the second part of the equation.  Once we have the axis from the first part down, we can use additional clues to find the second part.

The second part may be more challenging.  I'm suspecting that the photo was cropped.  The narrowness of it is one clue and the fact that there is a high likelihood of the wing struts being in the way would be a good indicator that the left hand side of the photo was cropped.  The analysis that I did in google earth shows a slight shift of the photo axis towards the left hand side - another good clue.

In google earth (GE) could you go to Snapshot View -> Properties, copy-paste all the numbers defining the viewpoint? 

It will look something like this:
Latitude: 4°40'59.90"S
Longitude: 174°30'37.98"W
Range: 3339m
Heading: -67.000000°
Tilt: 86.000000°