Read the…

Experience the thrills and defeats of an authentic deep sea adventure as you immerse yourself in the CATALYST 2 Mission Log. More>>

Waitt Institute

The Waitt Institute is a non-profit research organization that serves as an exploration catalyst, enabling scientific pioneers to transform the ways in which discoveries are made. More »


Search Grid

The proposed search grid and the bathymetric characteristics of the area…

Search Grid Orientation
The sunrise azimuth at Howland Island of 067 produced the perpendicular Line of Position of 337-157 degrees. Since the sun azimuth is in true degrees, the 337-157 degree Line of Position (LOP) should also be in true degrees. However, in practice, AE may have flown along the LOP on a magetic heading of 337-157 degrees. Our final search grid is oriented to a 337-157 degrees magnetic LOP, reflecting what AE likely flew. The difference in search grid coverage between orienting the grids using true or magnetic degrees, is not significant.

Final Search Grid

Standard Grid
This is the final search grid. Terminal area maneuvering, as well as wind effects on the flight’s EON point are contained in these search grids.

Standard Grid p68.jpg

Final Search Grid

Bathymetric Grid

Bathymetric Grid P69.jpg

This NOAA image includes the bathymetric characteristics of the search area.

Back To Top


Search Strategy Considerations

Things to look for in the sonar data

Debris Field
Following the Natal-Dakar flight segment, AE describes a project conducted during the World Flight in which air samples were being taken, in a study of upper air microorganisms.75

These were performed by AE and FN for Dr. Fred C. Meier of the Department of Agriculture, and were similar to experiments carried out on some of Charles Lindbergh’s flights. The collection devices were aluminum cylinders, and AE reported having a supply aboard exceeding a dozen broomstick-sized collection cylinders. These were mentioned in AE’s logs and the Luke Field aircraft inventory established following the initial westbound World Flight attempt and crash at Luke Field.

These cylinders, if distributed post-impact, would be characteristic shapes in a debris field that could be used to assist in identifying the wreckage.

In Situ Documentation
Ideally the entire structure should be documented as it is found, from as many angles as possible for 360 degrees around the wreckage. Interest areas include

Condition of fuselage, wings, engines, propellers, flaps, and control surfaces.

Impact damage is important to computing impact dynamics.

Debris field mapping and geo-referencing of anything found in the debris field.

If possible, cockpit documentation of controls, instrument indications, and items found in the cockpit are valuable to the crash analysis.

Accident investigators should be on-site to observe and record information prior to recovery, and to examine wreckage as it is recovered.

Aircraft Views and Dimensions

Dimensional Data

Below are basic dimensional data for a Lockheed 10A airframe. AE’s Lockheed 10E had larger engines and other modifications.

Figure 11.jpg

Figure 11 - From Purdue's Collection - Basic Data - From Lockheed, Burbank, CA.

Figure 12.jpg

Figure 12 - From Purdue's Collection - Three-view Drawings


Below are basic images for general structural reference


Lockheed Electra 10E



Lockheed Electra 10E

Back To Top


Long, Elgen M. and Marie K.

Amelia Earhart - Mystery Solved (Simon & Schuster, 1999)

Long’s research is by far, the most complete, comprehensive, and accurate work done to locate this accident aircraft. Documentation of the entire mission is very thorough, and where literary license is taken to craft the publication, it does not materially detract from the research work.

Rather than reiterate Long’s extensive analysis, the following are significant points as they relate to our work in review, analysis, and aircraft localization:

Valid assessment of headwind at 23 knots, 26.5 mph throughout mission.

We investigate reduced headwind and increased crosswind effects on final position.

These effects are contained in the Primary Search Grids.

Flight path Lae-Howland passes through three points that are updated.

No evidence the flight was at the point reported at 0519 GMT.

Long’s path through this point may simply conform to tradition.

No evidence the flight passed directly over the SS Myrtlebank.

This point is 40-60nm north of track from Lae to Howland.

There is no evidence the flight did not overfly SS Myrtlebank, and the final End-of-Navigation point(s) should not be affected by the small lateral track deviation abeam Nauru Island, if indeed, AE was slightly closer to Nauru and did fly closer to SS Myrtlebank.

No evidence the flight was north of track, or Howland, at any time

This conclusion appears to have come from CDR Thompson of Itasca

Distance and time averages are valid assessments, remarkably accurate for the analysis methods used, and comprise a very credible basis for track plot and terminal area arrival at Howland.

Validates an arrival near Howland with insufficient fuel to exit the Howland area

Long creates a reference to “Itasca standard time (IST)” in addition to the already confusing Howland Standard Time and GMT references.

Long advances a compounding 10% of navigation distance error model, that, while somewhat subjective and lacking of a more rigorous analytical conclusion, is a reasonable approach to a location methodology.


Swenson and Culick

Swenson, G., Culick, F.E.C.

Analysis of Amelia Earhart’s Final Flight - July 2, 1937

This research is well done. The engineering assessments are based on L487 and wind tunnel testing performed in 1935 and contained in a test report, GALCIT Report No. 161P. Standard aerodynamics equations are applied to determine performance.

Assumptions underlying this research include:

1. The magnitude of headwinds and their constant velocity throughout the entire flight.

This value, from Long, was 26.5 mph.

2. A constant true air speed flown throughout the entire flight.

This value, From Long, was 161.5 mph.

3. Initial fuel load.

This initial value was reported by Chater and Collopy.

Swenson and Culick, et al, applied adjustments for temperature and volume to arrive at an initial fuel load of 1080 gallons, vice the value of 1100 gallons reported by Collopy.

4. Flight Path.

The authors used the path constructed by Long.

5. Altitudes.

The authors used In-flight position reports as the basis for reconstructing the vertical flight path profile between Lae and Howland Island.

6. Ship Sighting.

The authors addressed the relative reliability in the 1030 ship sighting report by AE

They concluded that the vessel observed was either the USS Ontario, or SS Myrtlebank.

Their assessment is based on assumed aircraft ground speed and time to the ship sighting, and a projection forward to the Howland area arrival time. This helps to establish a time abeam Nauru Island, in their path reconstruction.

The authors’ work is very credible. They presented a good baseline fuel consumption analysis, and created numerous alternate scenarios as functions of headwind, fuel consumption and error tolerance. Their assumptions for headwind, aircraft true air speed, initial fuel load, fuel consumption and endurance are appropriate given the scarcity of facts. Their conclusions are valuable and interesting in assessing boundary values for mission parameters.

The authors’ choice to use Long’s 26.5 mph headwinds is prudent and replicated by virtually all researchers. Similarly, assuming aircraft true airspeed of 161.5 mph is considered valid.

Swenson and Culick and Long plot AE’s flight path at the ship sighting as passing over the assumed ship’s position, which very slightly affects distance, timing, and position.

Swenson and Culick do not project this path point, slightly north of the great circle direct route from Lae to Howland (Path A), to an end point north of Howland, as Long does.

Swenson and Culick’s path from the ship sighting converges to Path A as it proceeds direct to Howland from the ship sighting point.

The Swenson and Culick, et al, conclusions below, are validated as follows

  1. Initial fuel load and preflight planning should have enabled flight for 20 hours 38 minutes.
  2. Actual mission time to initial arrival near Howland was 19 hours 12 minutes.
  3. This should have allowed post-arrival endurance of 1 hour 16 minutes.
  4. The ship sighted was SS Myrtlebank, based on assumed average headwind and aircraft ground speed and time, and projecting those parameters forward to an estimated arrival at Howland at 1912 GMT.
  5. AE was within 100 miles [units not specified] of Howland based on radio strength.
  6. “…AE’s flight…ended in the ocean short of her intended landing place.”

Safford, Laurance

Earhart’s Flight Into Yesterday - The Facts with the Fiction (Paladwr Press, 2003)

Captain Laurance Safford passed away before this book was published. Co-editors Cameron Warren and Bob Payne salvaged the original manuscript and its supporting exhibits, presenting the work in this publication. Most of Safford’s work involves the communications in the planning and search phases of the mission. Safford devotes only 38 of 199 pages to the actual Lae to Howland mission segment. Most of his work is with radio logs and communications, coordination, control (operational as well as administrative), and the search effort.

This is no surprise as Captain Safford’s Navy career was in Cryptology, and Intelligence.

Safford’s conclusion (p115) is that AE crashed at N 01.00 degrees and E 178.00 degrees, with a 95% probability of a final position within 100 miles [units not specified] of this location.

This is approximately 325 miles [units not specified] west of Howland Island.

Safford’s inclusion of logs, messages, radio communications, and the attention to command and control issues associated with the mission planning and conduct of search operations, is valuable in adding background detail to our overall analysis.

Perhaps the most valuable information from Safford concerns the Itasca search patterns, and search decisions, made by its commanding officer, CDR Thompson. In his messaged assessment to COMDESRON 2, CDR Thompson concluded that AE was within 250 miles [units not specified] of Itasca, based on signal strength, and went down within 250 miles [units not specified] of Howland Island between 337 degrees and 45 degrees true and not nearer than 30 miles [units not specified].

CDR Thompson and Itasca assumed AE had laterally deviated north and had overflown Howland.

Safford is critical of many elements of this mission and his work does a credible job of detailing errors and inconsistencies.


Nesbit, Roy

Missing Believed Killed (Sutton Publishing LTD, 2002)

In this work detailing the accounts of famous missing persons, the author devotes 34 pages in a total 173 pages to AE’s life and final flight. The book is an account of 5 accidents involving famous people.

The author details the Electra aircraft from Lockheed documents, including interesting details concerning fuel tank arrangements and capacities, previous flight segments, aircraft weights and speeds, flight times and position.

The author depicts a flight path directly over Nauru Island, assuming this path from the AE report of seeing Nauru Island’s lights. This path discounts the involvement of USS Ontario, and SS Myrtlebank, but validates identifying lights on Nauru Island.

Most interesting and valuable are the author’s references to celestial navigation, and the effect on the Lae-Howland flight from various aspects of celestial navigation, including navigation errors. The author is an experienced aircraft navigator, with experience near the era of AE’s World Flight and with the USAAF in WWII.

The author “recreates” an assumed series of actions taken inside the Electra, by Fred Noonan and centered on celestial navigation, during the final portion of the flight.

This recreation begins with the AE In-flight position report of 200 miles out [units not specified] at 1745 GMT, and includes a proposed resolution of this position, with the next report at 1815 of 100 miles out [units not specified].

The author generally concludes that these reports are consistent with an increasing accuracy of navigation provided by the fixing of position based on sunrise. Further, the author discusses the Line of Position, how it is used, and how it may have been used by Noonan, if he used such a technique at all. No conclusions are provided.

The author details (p26) one source of navigation error in using a sun fix at sunrise. The error arises in defining the sunrise time, and angle to the sun itself, at the time of first sighting of the rising sun.

“On the sea, the angle is essentially zero, however, in an aircraft at altitude, the occupants view the sun rise at an earlier time than if viewed from the sea surface. This difference is accounted for by correction factors in sight reduction tables.” Failing to correct a sun shot for this angular value, according to the author [no mile units specified], results in a 31-mile error at 1000 feet, 44 miles at 2000 feet, 70 miles at 5000 feet.

This error produces an aircraft position that is closer to Howland than the actual aircraft position. Further, the author contends that the “200 miles out” report was more accurate than the “100 miles out” report, if this error were made.

The author concludes this error was made, and that AE was flying north and south along a Sun Line of Position, located at least 31 miles [units not specified] west of Howland Island.

In general, the author publishes interesting aircraft information, and covers the celestial navigation issues and error potential very well.

The work concludes that a sun shot error produced a final position at least 31 miles west of Howland Island, and that AE had flown north and south along a 337-157 line through this position.


Pellegreno, Ann Holtgren

World Flight - The Earhart Trail (The Iowa State University Press, 1971)

Valuable data from this 1967 Commemorative Flight includes references to climb speed of 100-120 mph, and 20 minutes time to climb to 1000 feet after a gross weight takeoff. This reference is made twice and the author comments that this is normal performance for her Electra.

The author cites several cruise performance values which provide good comparisons for AE mission analysis, in speed and fuel consumption, despite flying an Electra model 10A with smaller engines, but effectively the same horsepower-to-weight ratio, as for AE’s Electra 10E.

The author also cites en route winds throughout the flight from Lae to Nauru Island, indicating useful information about the behavior of en route winds in this area.

Pellegrino cites work published by Polhemus (p208) in which Polhemus calculates AE’s initial fuel at Lae at 900 gallons, and that AE executed a direct (great circle) flight path to Howland. Polhemus estimates AE’s final position “…in the vicinity of Howland Island….”

Near Howland Island, as Pellegreno was flying on the Line of Position heading 157 degrees at 1905 GMT (p160), a squall appeared over where Howland Island should be. The flight adjusted course slightly to avoid the squall, but continued to pursue visual acquisition of the island.

With pilot Pellegreno flying, and two dedicated observers (one in the cockpit right seat and one in the cabin), Howland Island could not be found until approximately 1957 GMT, when the person in the cabin spotted what he thought was land. They had less than 20 minutes remaining fuel on station to devote to the search for the Island, and as Pellegreno later said, “we nearly missed it.” This, after searching for nearly an hour.

They were approximately 10-12 miles [units not specified] north of Howland Island at the moment they visually acquired the island.

Pelllegreno’s account of her thoughts and feelings upon arriving and not seeing Howland, then conducting a protracted search with limited fuel resources, is extremely interesting as a human factors and operational comparison to what may have occurred on AE’s mission. Pellegreno writes a compelling narrative here, one that can not help but evoke a sense of urgency, desperation, and elevated tension.
Pellegreno’s flight had the advantage of better navigation equipment, a third set of human eyes, a nearby ship providing good DF bearings, and the luxury of having departed Nauru Island, with a Canton Island destination. With all of these advantages, they nearly missed visually acquiring Howland Island.

This account demonstrates the great challenge attempted by Amelia and Fred, and provides a good assessment of the difficulty in visually acquiring tiny Howland Island.


Finch, Linda

No Limits (World Flight, Inc., 1996)

This account of preparations for a 1997 Commemorative Flight details most known facts and assumptions concerning AE’s flight from Lae to Howland Island.


Strippel, Dick

Amelia Earhart - The Myth and the Reality (Exposition Press, 1972)

The author concludes that initial fuel load was 980 gallons, based on two calculations

A gross weight and takeoff distance analysis for Lae’s grass airfield in 1937, that results in a possible takeoff weight and fuel load.

The author concludes the ship sighted was USS Ontario because had it been SS Myrtlebank, the position and time error would have exceeded Noonan’s standard performance for navigation accuracy.

The author recounts a number of scenario theories, possible errors, and the effects of those errors.

The only actual position statement in this work is from Captain J.S. Dowell of the USS Lexington (p156) who concludes “…at 2030 the plane landed on the sea to the northwest of Howland Island, within 120 miles [units not specified] of the island.”

The author’s Appendices contain interesting and useful information regarding aircraft configuration, performance, and some details of the declassified messages and logs contained in national archives.


Gillespie, Ric

Finding Amelia - The True Story of the Earhart Disappearance (Naval Institute Press, 2006)

The International Group for Historic Aircraft Recovery (TIGHAR) and the author have compiled a comprehensive and useful website, and this publication, including a resource CD containing AE-related information, research and data.

This work supports an alternate theory that AE landed the Electra on Gardner Island in the 1937 Phoenix Island Group. Gardner Island is now Nikumaroro Island in the Republic of Kiribati, approximately 400 statute miles southeast of Howland Island.

This theory emanates from essentially the immediate four-day period following the disappearance of AE, information for up to two weeks following the disappearance of AE, and multiple expeditions to Nikumaroro by TIGHAR personnel during which artifacts were found that are claimed to possibly be linked to the AE mission. These artifacts have not yet been validated or documented as coming from AE’s mission, however, the discoveries are interesting.

Main support for the theory comes from analysis of radio transmissions allegedly made by AE, and received by experienced radio operators at Honolulu, Wake Island, and Midway Island radio operator stations.

The signals and attempted direction finding (DF) bearings from these three stations converge somewhat close to Gardner Island, lending to TIGHAR’s theory that AE crash landed the Electra on Gardner or very near it, making landfall and transmitting radio calls.

Numerous challenges exist in these theories, not the least of which is whether or not the Electra could have flown to Gardner Island at all, or transmit any signal with inoperative engines and generators, or do so following an off-field landing or water ditching.

Interestingly, the best DF bearings on good, strong radio signals in 1937 contained some directional variance under the best conditions. If the DF bearing signals received by Honolulu, and Midway Island are adjusted by a 10 degree variability in directional reliability, and in the direction of common sense toward the area most likely containing the Electra, and the Wake Island bearing is given a +/-10 degree azimuth variance since it was reported as a strong signal and bearing, the area bounded by the convergence of these adjusted signal directions is a centroid approximately 90-123 nm southwest of Howland Island, and 260nm northwest of Gardner Island. Questions remain concerning whether or not the Electra could transmit these radio signals following a water ditching, if AE had any backup or portable radio transmitting equipment aboard the Electra on the Lae-Howland mission segment, or if a life raft was aboard the Electra, which AE may have occupied while transmitting and drifting towards Gardner Island. The research of many investigators indicates that flying a total of more than 4 hours fuel after 1912 GMT, is not likely. If AE’s last transmission was at 2013 GMT, an hour after arriving at Howland, and they commenced a divert to Gardner Island, then AE would have had to arrive at Howland Island with more than 5 hours fuel remaining.

Swenson and Culick’s thorough aerodynamic analysis precludes such a fuel state, and other researchers corroborate these findings.

However, the author makes some compelling arguments for TIGHAR’s theories, discusses interesting discoveries made on Nikomororo Island, and provides evidence to consider TIGHAR’s alternative theories.