Robert Ballard, the man who found the RMS Titanic, is leading a search for the remains of Amelia Earhart decades after she vanished without a trace while flying across the Pacific. According to the New York Times, his interest in the search was sparked by the Bevington Photo — an image taken on Nikumaroro Island a few years after Earhart disappeared that shows what may be the landing gear of a Lockheed Electra sticking up above the surface of the water.
Dr. Ballard is a legend in oceanography. In addition to his discovery of the Titanic in 1985, he discovered the Bismark in 1989, the Yorktown in 1998, and the wreck of President Kennedy’s PT 109 in 2002. In-between Bismark and Yorktown he also found the time to respond to a letter from one earnest little boy who was considering a career in oceanography and had been inspired by his work. While said child disappointingly grew up to be an IT journalist instead, he always kept the letter. Ballard has brought considerably more firepower to the investigation of what happened to Earhart. The investigation he is conducting could finally answer the question of whether she crash-landed on Nikumororo.
ExtremeTech has covered Earhart’s life and disappearance in some detail in the past. Our analysis of the current state of the search is below. It has been updated to conform to the latest available evidence in addition to Dr. Ballard’s decision to conduct his own research discussed above.
Much of the data on Earhart’s final resting place has been compiled over the course of decades by TIGHAR, The International Group for Historic Aircraft Recovery. Since 1989, TIGHAR has been compiling data, evaluating evidence, and assembling information on where Earhart’s Lockheed Electra 10E might have gone down, and where the wreck is now. There are two things you need to know before diving into the rest of this story:
1) I have chosen some of the highlights and most interesting details, not to cherry-pick the evidence for the Nikumaroro landing theory, but because TIGHAR has produced several hundred PDFs of evidence, analysis, historical documents, and records. I highly recommend that anyone who finds this topic interesting dive into that body of research. A shorter version for those without several hours to spare is also available.
2) All of the evidence we’re going to discuss is circumstantial. The strength of the argument is that there is an absolutely mind-boggling amount of circumstantial evidence: multiple reports from settlers and navy personnel of recent habitation, including one such report from the navy pilot who overflew Gardner Island in the week following Earhart’s crash. A body recovered from the island (along with American shoes) in the 1930s, before being lost in World War II. Photographs of what appears to be airplane wreckage. Villagers and one-time settlers who report seeing and even using bits of metal and material that would match said airplane wreckage. The discovery of a water catchment system in 1944 that could’ve been built using materials carried aboard Earhart’s Electra.
After a review of the evidence TIGHAR has found to date, as well as the aforementioned forensic report, I personally find the Nikumaroro hypothesis extremely credible. But readers should be aware that whether the Lockheed Electra 10E possessed sufficient fuel to reach the island is a matter of debate and depends, to some extent, on how you read the radio transmissions the Itasca picked up.
Earhart’s Last Flight
Amelia Earhart and navigator Fred Noonan took off from Lae, New Guinea on July 2, 1937. They were flying a specially outfitted Lockheed Electra 10E. But they had made several changes to the plane’s equipment, particularly its antenna layouts and structure, that may have critically harmed communication between the plane and its ship tender, the United States Coast Guard Cutter Itasca. The Itasca was standing by at Howland Island to signal the Electra and guide it in.
Earhart’s planned route and the deviation TIGHAR believes she most likely took.
Unfortunately, due to the aforementioned communications problems, the Lockheed and the cutter never made contact. While the ship could hear Earhart, it could not respond to her transmissions on the same frequency. Earhart’s last transmission stated the following:
“We are running on line 157-337… We are running on line.”
What this means is fairly simple: Noonan, as navigator, had taken the aircraft’s line of position relative to the sun’s location as it rose. Overcast conditions had prevented him from using celestial navigation overnight, but using a single body for a line position only told Earhart part of where she was. The slideshow below contains additional information about the Electra, its position, possible damage to the radio antenna that would have compounded her problems communicating with the Itasca, and data on what being “on line” meant.
In theory, she could’ve been anywhere along that course, from directly on top of Howland Island or Baker to far off-course in either direction. No further transmissions were received by the Itasca and prevailing wisdom for decades has been that Earhart’s Electra ran out of fuel and landed in the ocean.
TIGHAR has compiled evidence that illustrates otherwise and backed it up with formidable analysis. Let’s consider some of the highlights.
Betty’s Notebook and the Bevington Photo
July 1937: A 15-year old girl named Betty is listening to her family’s radio, tuned to “short wave” frequencies. Her father had run a 60-foot antenna, and the set was capable of picking up transmissions from all over the world. She hears — or believes she hears — the words “This is Amelia Earhart.”
This was scarcely uncommon at the time. But what separates Betty from a number of other witnesses is that she kept a notebook of everything she heard and when she heard it. More than sixty years later, TIGHAR researchers sat down with her to determine whether her claims were possible. At first glance, this might seem farcical. Surface craft and radio stations within a few hundred miles of Gardner/Nikumaroro could only pick up rough triangulations and estimates on Earhart’s transmissions. How could a teenager be hearing her in St. Petersburg, Florida?
The answer lies in the hardware of the day. Earhart was broadcasting on a Western Electric 13C radio and broadcasting on 3105, 6210, and 500 Kilocycles. Her radio, however, lacked any form of harmonic suppression equipment — meaning that when she broadcast on these frequencies, she was simultaneously and inadvertently broadcasting at much higher frequencies as well. Furthermore, Earhart’s aircraft antenna had been modified in a manner that inadvertently boosted the harmonic signal emanating from the aircraft.
All of this would be moot, however, if the radio set Betty’s family owned was incapable of receiving the harmonic signals Earhart would have been broadcasting. That radio — a Zenith 1000Z Stratosphere — was one of the most capable and powerful consumer sets available on the market from 1935-1938. Not only was that particular model capable of receiving a harmonic frequency of Earhart’s broadcasts, but the nature of what Betty heard — faint, scratchy, intermittent audio — matches the statistical probability of what she’d have been likely to hear. At the most likely frequencies, the signal-to-noise (SNR) ratio was only barely high enough to permit intermittent contact.
But there’s another side to Betty’s notebook — the question of what she heard. And this is where the story gets a little chilling. The slideshow below contains multiple pages from Betty’s notebook, both in her hand and with her text transcribed. We recommend opening images in a separate window for easier reading.
Included in her transcribed notes is the phrase “New York,” abbreviated “N.Y.” Earhart, and the unidentified male voice Betty also documented, continually repeat this phrase. It’s the last thing Earhart says, even as she’s apparently abandoning the plane as the tide rolls in. Betty also recorded the time of day when she heard Earhart. Our best tidal predictions for the area suggest that, in fact, Earhart would’ve landed at low tide, but been forced to leave the plane within a matter of hours.
There was no known reason for Earhart to be frantically repeating the word “New York,” but what if she wasn’t? What if, instead, the word was Norwich? On November 30, 1929, the tramp steamer Norwich City ran aground on Nikumaroro. The wreck was a known hazard and prominent marker for decades after. Earhart surely would’ve known of it and recognized it on her approach to what she’d have identified as Gardner Island.
Fast forward several months. Cadet Officer Eric Bevington visits Gardner Island. While there, he snaps what’s become known as the Bevington Photo — with the Norwich City in the background, and a strange object jutting up from the water to the far left.
The Bevington Photo. SS Norwich City in the background.
TIGHAR was able to hunt down the original photo negatives and perform forensic analysis on them. What emerged was this result:
Compare that with the landing gear of a Lockheed Model 10 Landing Gear Installation 40650:
And finally, look at what happens when we superimpose one on top of the other:
Knowing something about how the landing gear assembly tended to fail on Electra’s allowed for better modeling.
This PDF details how the TIGHAR group found detailed photographic evidence of precisely how a Lockheed Model 10 Landing Gear installation shears and fails when subjected to certain kinds of lateral stress. If you compare the angles of failure (Page 14/15), the mangled landing gear winds up bent in similar angles to what’s seen above.
A scrap of metal found by TIGHAR, which probably belongs to Earhart’s Lockheed Electra 10E.
The most recent physical evidence was uncovered in 2014, triggering some additional coverage on the disaster. TIGHAR found a piece of scrap metal that has the same “fingerprint” as custom-made aluminum sheets that replaced some windows on Earhart’s customized Electra. The piece of scrap metal is the right size, and its rivets and other hallmarks identify it as almost certainly coming from Earhart’s plane. Nor is it the only artifact of note: We know that the British recovered a sextant box in 1940. While they dismissed the idea that the box could have contained an aviator’s sextant, we now know that Fred Noonan was in the habit of carrying both a mariner’s sextant and a bubble octant. The latter is used in aviation, while the former was carried as a backup tool.
The box the British found was stenciled with the number 3500 and otherwise marked with a second number, 1542. TIGHAR believes this refers to the manufacturer’s serial number for the device and the Naval Observatory code for validating that each device was calibrated properly. As they detail here, the serial number and NO number, if that’s what they were, fit perfectly into a sequence of other sextant boxes that were confirmed to have been processed by the Naval Observatory.
TIGHAR met with metallurgical engineers at Massachusetts Materials Research on July 11, 2019, to evaluate the nature of the aluminum fragment they found. A section of the scrap shows distinct signs of being cold-worked (as in, repeatedly stressed). Photos of the Electra show a similar area that corresponds to the metal patch.
Again, is this proof? No. It’s just a piece of aluminum that appears to be of the same type (2024 ALCLAD) as that used in the Electra, with a size and shape that allow it to fit over a patch that we know had been applied to the Electra Earhart was flying. It’s not proof. It’s just (another) hell of a coincidence.
Setbacks in the Sonar Search
After decades of work, and boots-on-the-ground searches for artifacts, the TIGHAR team elected to use side-scan sonar to search the area of Nikumaroro where the Bevington Object was seen and where the debris field from the Lockheed Electra might have washed up. Initially, the team thought it had captured data that showed clear signs of a fuselage-shaped object in the sonar data. Unfortunately, subsequent analysis and a later 2014 expedition appear to have confirmed that the object TIGHAR thought it captured via sonar was actually a coral ridge. The 2014 expedition was beset by problems, including a last-minute shipping cost issue that prevented the team from shipping a backup ROV and the failure of the ROV they were able to bring.
Further analysis of the anomalous readings captured in the 2012 data revealed that they weren’t as anomalous as they appeared to be, due to variation in how side-scan sonar “sees” objects on the seafloor. (See page 37 of this PDF for further details.)
After it analyzed the 2012 data, the TIGHAR team hoped to find definitive proof for their theory during the 2014 trip. That didn’t happen, though technical difficulties prevented the group from performing anything like the data-gathering it originally intended to do. TIGHAR’s response to these setbacks was to straightforwardly admit that the anomaly they thought they had found might well be a coral ridge, and to acknowledge that there’s no sign of debris from the Electra at depths accessible to Scuba divers.
Ballard’s investigation is a vastly better opportunity to search than the resources TIGHAR was able to bring to bear. That doesn’t mean he’ll find the wreckage — even if it’s there, these kinds of searches are difficult and time-consuming — but the odds are better with Ballard.
Did We Find Her Body?
A forensic analysis of remains first found on Gardner Island in 1940 (now known as Nikumaroro) suggests they were mis-categorized by the original examining doctor, D. W. Hoodless. Nikumaroro is the only island Earhart could’ve reached, given the fuel reserves in her aircraft at the time of her disappearance. The discovery of a skeleton on that island in 1940 was seen, even at the time, as a clue to where she might have landed — until Dr. Hoodless concluded the bones belonged to a man some 5’5″ in height (1.65m). Not only was Earhart female, but she was also significantly taller, at 5’8″.
This new report, published in the journal Forensic Anthropology, builds on a report from 2016 by Richard Janz, professor emeritus of anthropology and director emeritus of UT’s Forensic Anthropology Center, and represents a deeper dive into the bone measurements taken by Dr. Hoodless in 1940 (the bones themselves were lost during the war). Janz is not the first anthropologist to argue Hoodless might have been mistaken in his evaluation of the skeleton, but he’s the first to apply modern computer modeling technology to the problem. He also extensively discusses an article published by Cross and Wright in 2015, which dismissed the idea that the bones in question might have belonged to Earhart.
The full text of the article is available online, but we’ll hit the basics. Forensic anthropology wasn’t nearly as developed in the 1940s as it is today. Even prominent leading scholars of that era are now known to have made significant mistakes in the classification of remains, including mis-mapping of male/female gender ratios. While Hoodless was a doctor, he had no training in forensic anthropology and is known to have relied on Pearson’s formula from 1899 when making his own estimates of the individual’s height. But Pearson’s data on estimating the height of an individual, which was gathered from individuals born in the early 19th century, didn’t reflect the impact better nutrition had on both male and female height towards the end of that era. The limited sample size Pearson worked with also made his conclusions less accurate.
In addition to the inaccurate information in size, extensive exposure to the elements and scavenging by coconut crabs (both recorded by Hoodless) had also weathered the bones, making it much more difficult to sex them accurately. None of the bones had been buried; the skull was buried briefly after it was initially found. Crucial forensic evidence, like the woman’s shoe sole found with the bones, was lost when they were. This last bit is infuriating, given that expeditions decades later found another shoe sole fragment that could’ve been combined with the first to determine if both came from the same pair of shoes, and whether they resembled footwear Earhart was known to wear.
Janz also used photographic analysis of Earhart at various points throughout her life to provide other guidelines for her height and weight, including the length of certain specific bones in her arm. Photos like the one below allowed forensic anthropologists to first measure the exact height of the Mobile Lubricant container by finding a still-existing container, then use that knowledge to create a scale for the rest of the image, including the length of her arm. The black dots in the image reflect the specific location of the bones mentioned in the caption.
Based on the data in this image, Earhart’s humerus is estimated at 321.1mm, while her radius was 243.7mm. The bone measurements reported by Hoodless for the skeleton found on Nikumaroro were 325 and 245mm, respectively. Estimates of Amelia Earhart’s tibia length based on her inseam and known stature yield an expected length of 372mm.
The bones are consistent with Earhart in all respects we know or can reasonably infer. Her height is entirely consistent with the bones. The skull measurements are at least suggestive of female. But most convincing is the similarity of the bone lengths to the reconstructed lengths of Earhart’s bones. Likelihood ratios of 84–154 would not qualify as a positive identification by the criteria of modern forensic practice, where likelihood ratios are often millions or more. They do qualify as what is often called the preponderance of the evidence, that is, it is more likely than not the Nikumaroro bones were (or are, if they still exist) those of Amelia Earhart. If the bones do not belong to Amelia Earhart, then they are from someone very similar to her. And, as we have seen, a random individual has a very low probability of possessing that degree of similarity…
From a forensic perspective the most parsimonious scenario is that the bones are those of Amelia Earhart. She was known to have been in the area of Nikumaroro Island, she went missing, and human remains were discovered which are entirely consistent with her and inconsistent with most other people.
Is this the final word? No. Absolute proof will only come via DNA identification from bones that no longer exist, the wreckage of the Lockheed Electra (near Nikumaroro or another location), or some form of incontrovertible written testimony from Earhart or her navigator, Fred Noonan.
A Multi-Decade Puzzle
In this write-up, I’ve had to pick and choose what I think are the most tantalizing pieces of the Earhart investigation. But the data TIGHAR has collected, sifted, and analyzed is orders of magnitude larger than I’ve even had room to mention. The fact that a man with Dr. Ballard’s reputation has been convinced to spend his time is evidence that the theory isn’t just a crackpot one. From on-site artifact recovery, tidal patterns, photographic analysis, sonar scans, and even trips to analyze the wreckage of other Lockheed Electras, this sprawling research project would’ve been impossible without access to the technology of the modern era. Even the mathematics used to estimate debris field scattering and the optimum methods for searching large debris fields with side-scan sonar depend on techniques developed within living memory.
Part of what makes Earhart’s story so compelling is that it took place at a time when air travel was rapidly shrinking the globe, yet technologies like radio were still in their infancy. The idea of losing an aircraft because of a mismatch between transmission frequencies, or a plane that couldn’t instantly coordinate its own position on Earth, is foreign to us thanks to eighty years of technological advancement. It’s easy to forget that when Earhart went missing, there were no policies for conducting search and rescue operations with aircraft, because few ships had aircraft, for example. There’s also evidence that the Itasca and other search vessels were given incorrect information about the capabilities of the Electra. Initially, the Itasca assumed that Earhart could still transmit if the plane had landed in the water (this was not the case). All of these miscommunications and problems stole precious time from the search.
TIGHAR hasn’t proved its theory — only discovering the wreckage of the Electra or incontrovertible evidence of Earhart’s personal presence on Nikumaroro would do that. But despite the loss of the sonar data, the overall body of evidence is significant. And it’s not particularly surprising that the team has had relatively slow progress — it’s not cheap to hire ships and a crew to do research on a South Pacific island, and any search of the ocean around Nikumaroro requires the cooperation of Mother Nature. If the seas are rough, there’s no way to perform the necessary photography.
At the same time, however, recent events give Earhart’s case fresh resonance. The loss of flight MH370 in 2014 drove home the fact that while we may think we have perfected the art of tracking objects in flight, our ability to do so is sharply limited. All credible evidence suggests that MH370 deviated sharply from its intended flight plan before crashing into the Indian Ocean, but there’s no explanation as to why this occurred. With the black boxes having long since fallen silent, it’s possible we’ll never find the plane.
Finding Amelia Earhart’s Lockheed Electra wouldn’t tell us what went wrong that July morning — the plane would almost certainly be too battered and broken to yield much forensic evidence and there were no black boxes in 1937. But it would lay to rest the question of what happened to one of aviation’s pioneering spirits and most driven pilots, thanks to the efforts of a determined group of researchers and the painstaking work of decades.