To the Aviation community: As a private pilot of some thousands of hours of flight experience, as a radio amateur, and as a consulting physicist, in the 1990's I undertook an investigation of the problems -mainly radio communications - which Amelia Earhart encountered. on her fatal flight. I found numerous radio-technical operational errors, and some misfortunes As well, that were pivotal in the outcome. The following is my summary of these details. -------------------------------- But first, some /////////acronyms///////// The technology of 1937 now seems archaic, but it’s vital here to understand it: AE-- Amelia Earhart, pilot and radio operator. CW--“Continuous Wave”-- radio transmission that can be switched on and off at transmitter, resulting as a tone at the listener’s ear as a series of dot and dash pulses of tone, to form letters and words; slowly but VERY reliably. DF-- “Direction Finding”; receiving radio transmissions with a small loop that is rotated to determine the direction at which the transmitter is located. DF steer-- land or ship station operator hears a radio signal transmitted from a remote vessel or aircraft, applies his loop antenna to deduce the direction from which the transmission arrived, then verbally transmits instructions to that remote vessel or aircraft radio receiver on steering to said land station or ship position. This works best at radio frequencies of 500 kHz and lower. HF-- high frequency radio transmissions; above perhaps 1,500kHz HFDF, HF-DF-- high frequency direction finding. Direction finding using radio waves of a frequency above 1,000 kHz (above one megacycle), later proven to be sporadic and unreliable over long distances due to “skipping” and “sky waves”. KCS-- kilocycles per second (now kHz, kilohertz), the operating frequency of the radio transmitter and the collaborating radio receiver. LF—low frequency radio waves, generally of frequencies below 500 kHz. MF-- medium frequency radio transmissions; from perhaps 500kHz to 1,500kHz QSO-- a completed 2-way radio conversation. QSL-- acknowledgement that a two-way QSO indeed successfully occurred! \\\\\\\\\\\\end acronyms \\\\\\\\\\\\\\ -------------------------------------------------- ---------------------------- Amelia Earhart (AE) loss - OR -Radio Propagation //Use & misuse of MF and HF ---------------------------- Hopefully, the truth about her disappearance is now a little closer, but more importantly, I sensed poor communications procedures. She was not the first to fly that area using radio communication. In "200 meters and down", it is noted that radio amateur Robert F Wilson[1] accompanied pilot Dr. Richard Light (both of Yale University on a 30,000 mile "round the world" flight. On return to New York they remarked that amateur radio contact was "invaluable". AE's round-the-world flight in her Lockheed 10E "Electra" included a false start in March, 1937 westbound to Honolulu. With Army help, the Electra had been fitted with supercharged engines to carry a greater fuel load. Amelia convinced President Roosevelt to apply WPA money to scrape out a landing strip on Howland Island, to complete the critical link across the Pacific. Near Howland Island were the British Phoenix Islands to the south and the British Marshalls to the west. The area was politically 'hot'. The League of Nations had given Japan a mandate to occupy the Solomon islands several hundred miles to the northwest, Saipan being their principle station. Japan busied those islands with civilian activities, but considered them out of bounds to Westerners. ---------------------------------------- [1] DeSoto, Clinton, B., "Two Hundred Meters and Down", The American Radio Relay League, West Hartford, Conn. 1936,1981, Page158. It is also noted that in 1935 the schooner "Kincajou" toured this area with K6BAZ (Kenneth L. King) operating that ship's radio, WOFV, as well as operating on Howland Island as K8XJI. These are not connected with Amelia's flight, but show the general level of radio activity in that Pacific area at that time. -2 The US Navy provided several ships with radio equipment for communications radio relays and weather reports for leg to Howland Island. The Electra was also fitted with a rubber raft and box-kite antenna. The Coast Guard Cutter Itasca (CGC-50) was to be anchored near Howland to provide her terminal guidance. The minesweeper Whiporwill (AM-35) was to steam midway between Honolulu and Howland. The cutter Shoshone (CGC-52) would be on line possessions (sic) providing navigation and communications support for a Pan American clipper as well as the Electra, and was to take Fred Noonan back to the US (Amelia did not need Noonan's astro-navigation beyond Lae when westbound. The ocean going tug Ontario (ATO-13) was enroute to American Samoa from Lae. These ships were pulled from routine duties in March, 1937 for this arrangement. The plan was that on her way from Honolulu the Itasca would establish HF voice communication for guiding "DF steers", also by voice on 3105 kHz or 6210 kHz[2]. The guidance was to be derived from HF or MF direction-finding equipment carried by the Itasca. Generally, AE was not as well prepared for this ambitious flight as one would hope. But her husband, publisher Putnam, optimistically pressed her onward to get to the US by July 4th. AE mused that such pressure was to her dislike[3]. The March, 1937 flight was aborted after the first hop. Her Honolulu takeoff ended in a "ground-loop" which damaged her Electra. Personnel and aircraft were returned to Oakland, California for repairs and a repeat try. In June, 1937 they (Putnam always had a hand in her planning) opted for an eastward route due to weather pattern changes. After repairs and improvements, the flight to Miami would serve to shake down the Electra. When that Lockheed Electra left Miami, it was not properly prepared for detailed conventional two-way communication and cooperative direction - finding. Earlier in 1936, AE was commissioned for a dollar-a-year by the Department of Commerce to test a new direction finder. I find this to be crucial. Though it gave her access to new navigational devices, it also gave the impetus to experiment within broad limits on her own, as evidenced by subsequent telegrams to the Itasca. In Miami, Amelia and Putnam, (despite repeated expert advice to the contrary) abandoned the 250' trailing wire antenna and later its LF tuning coil, as they deemed them expendable ((Burke)[4]. (BOTH are vital for contemporary 500 kHz LF transmission and reception for distant DF steers via a collaborating surface station.) While still in Miami, AE instead resolved to routinely transmit her position every 45 minutes, expecting to use the Electra’s radio receivers only for obtaining HFDF radio bearings. The flights over South America, Africa, India and New Guinea were without incident, except for Noonan's boredom on this unplanned long way around to the Howland leg, the only leg where his navigation was vitally needed. I have found no documentation stating that she practiced HFDF successfully in these legs. Articles sent back by AE to Putnam, published and "arranged" by Putnam in 1938 says essentially nothing about HFDF! These were intended to be for newspaper consumption. It was reported that AE shipped notes, etc. back as she progressed. Somewhere in those notes should be HF DF success/fail information. Only in Lae did her documented activities reflect concern in this (HFDF) matter. -------------------------- [2] Strippel, page 125: This frequency pair, harmonically related, became a standard in the '30's for civilian and private flight operations worldwide. They probably gave Amelia comfort in that she could talk to most any aviation facility with them. However, such standardization (worldwide) made it a little problematic when you would prefer to have a "clear channel". [3] Loomis, Page 30, p 59. [4] Burke, John, "Winged Legend", G. P. Putnam & Son, 1970 (#B/E12bu), p 216. See also Pellegreno, page 205. Strippel, Appendix 1A (Installed Radio Gear Block Diagram. -3 Written descriptions of Amelia and Fred's activities in Lae during the days and hours just before takeoff include comments of frustration and uncertainty about all these matters. In reading those comments, I could not sense that they were planning realistic and prudent alternatives for MF and HF ground wave DF-steers without LF capability. They presumed that their MF and HF radio communications would be reliable for this purpose. AE dispatched a radio message to Thompson on the Itasca to the effect that (quoted from the Itasca log): "(My) HOMING DEVICE COVERS FROM 200-1500 AND 2400-1800 KILOCYCLES ANY FREQUENCIES NOT REPEAT NOT NEAR THE BANDS ENDS SUITABLE. SUGGEST ONTARIO STAND BY ON 400 KCS TO TRANSMIT LETTER N FIVE MINUTES ON REQUEST WITH STATION CALL LETTERS REPEATED TWICE END EVERY MINUTE. SWAN TRANSMIT VOICE 9 MEGACYCLES OR IF I (am) UNABLE (to) RECEIVE BE READY ON 900 KCS. ITASCA TRANSMIT LETTER A, POSITION , OWN CALL LETTERS AS ABOVE ON HALF HOUR 7.5 MEGACYCLES. POSITION (of the) SHIPS AND (the time of) OUR LEAVING WILL DETERMINE BROADCAST TIMES SPECIFICALLY. IF FREQUENCIES MENTIONED (are) UNSUITABLE (for) NIGHT WORK INFORM ME (at) LAE. I WILL GIVE (a) LONG CALL BY VOICE (on) 3105 KCS (on the) QUARTER HOUR (and also) QUARTER TO (each hour). (The 1500 KCS limit was later corrected by AE to be really 1400KCS). "REPORT IN ENGLISH NOT CODE ESPECIALLY WHILE (I am) FLYING (. I) WILL BROADCAST QUARTER PAST HOUR GCT [GREENWICH CIVIL TIME] (now called UT)." The Itasca replied: "---ITASCA TRANSMITTERS CALIBRATED 7500 6210 3105 500 425 KCS CW AND EITHER CW OR MCW. ITASCA DIRECTION FINDER FREQUENCY RANGE 550 TO 275 KCS." (I find this exchange to be unsettling: AE implied that her DF work would suffice, since there was obviously no official HFDF capability on the Itasca. Itasca clearly states that 275-550 KCS was available (and nothing else). Two additional DF systems were involved. One was a 7500 beacon arrangement comprising a transmitted "A" on a separate frequency of 7500 KCS. I presume that AE was testing (her CAB contract?)this frequency on the basis that it should be particularly good for long distances. It was transmitted only once during the final moments by the Itasca). The second was a transportable HFDF system Putnam had borrowed from the Navy, including operator Cipriani) and working specifically on the 3105 and 6210 KCS communications bands, to fill the void left by the Itasca's only LFDF capability. More experimentation, in my opinion, on the basis of the CAB contract. Analysis of the theoretical fuel capacity of the Electra, along with an eyewitness account of the takeoff[5] from Lae as "bounding from the airport perimeter road and skimming the water low enough to kick up a spray" does not rule out the possibility of a full fuel load of 1150 gallons. Other researchers quote 1100 gallons, and also estimate 20 hours duration while holding an airspeed of 145 mph. Departure was 10am Lae time, (which is also 00:00 GCT, as specified by Putnam to minimize confusion later in the flight). Fuel consumption rate tapered from 70 gallons/hr on takeoff to 38 gallons per hour in economy cruise at the end of the flight. The first part of the Howland leg included communications between Lae radio and Amelia. Later, their position was to be determined by celestial navigation by night and Sun-Line-Of-Position by day, presuming either no clouds or flight over the clouds. Ground speed and direction was estimated by drift meter when the ocean surface was visible during daylight hours. Since Amelia flew in an overcast much of the night, it can be presumed that their night navigation was poor, amounting to essentially "dead reckoning" (assumed course and ground speed). Early over-water navigation was unpredictable. In 1925, Naval Commander John Rodgers, an excellent naval navigator, led a flight of three flying boats from California to Honolulu. His flight became lost at sea for 10 days! Later, he said "I wasn't good enough as a navigator. We missed the whole of the Hawaiian Islands"[6]! (He did not have radio navigation aids then. Noonan's performance in 1937 may not have been much better when radio DF-steers did not materialize. In my conversation with a WW II Air navigator that ferried military aircraft from the US west coast to Hawaii, I learned the following. "Line of Position" (solar navigation) is based on the fact that, say, that one's UT time of dawn will pin down one's meridian of longitude. Likewise, the sun's elevation and UT data pair are used to compute a line (which is approximately north-south but tilted per season and latitude) called the "sun line-of-position", or LOP. Knowing that, one can, say, fly eastward while measuring the sun elevation from time to time and until the LOP (which moves eastward with the aircraft). Chronograph accuracy is critical. It was estimated that one minute of time error would create four miles of position error. When the LOP intercepts the destination, one turns to fly along that very line (generally north or south) in the direction which where one will "fly down the offset" until seeing the destination island! ------------------------ [5] Pellegreno, page 144. New Guinean pilot Bertie Heath recounts in 1967: "I saw her silver plane move slowly down the unpaved runway. It must have been 3,000 feet long at that time (terminating at the shoreline). (on takeoff) When her plane reached the roadway that had a high crest on it and ran across the runway near the seaward end, it bounced into the air, went over the drop off and flew so low over the water that the propellers were throwing spray. "She continued out to sea for several miles before climbing on course slowly. The wind was calm and the dust from where she had hit the crown of that dirt road didn't disperse quickly, just sort of hung there." -4- The scenario is as follows: Transoceanic flights were scheduled to leave before nightfall, spending the major part of the flight generally heading toward their destination with no great apprehension as to where they were. Comes the dawn, pilot and navigator were very busy getting on top of the clouds and get some good sun shots. If cloud cover or haze was a problem, it was nip-and-tuck whether they should stay high and above the clouds to get good sun shots or descend to look for the island. Presumably, AE would arrive near Howland in the morning when the sun was low and LOP accuracy was good. Hopefully, they should find their destination island before noon and before fuel crisis. It was known that Noonan's Chronometer was acting up since he tried on more than one occasion to set it to WWV while at the Lae airport the night before departure[6a]. Nautical (and early aeronautical) navigators knew that there were two major problems with this LOP technique. The first was that statistics makes it equally probable that you should fly north and south on reaching the LOP. To do both can be very time consuming, flying 50 miles north, and then 100 miles south, then 150 miles north, etc., while the LOP angle tilts throughout the day and whose accuracy diminishes when the sun is overhead. Navigators accordingly chose a course deliberately offset a little to the north (or south; their choice) of their destination, so that when their sun LOP was on that of the island, they flew up this offset. A mandatory reserve for this was provided. Similarly, the time of takeoff was critical as they had to always plan to arrive during local morning. A second problem was that the success of the LOP method depends on the meridian of the destination to be known more precisely than the flight visibility (limit to which one can see and identify a small island) at an altitude where one would ultimately search for the destination. I believe this to be a contributing factor since maps of the day were not all updated to eliminate a 5+ mile offset to the east of the true Howland meridian from that indicated on old maps of that area. Various radio transmissions from Amelia were logged during that day-and-night crossing of the International Date Line (July 2/Lae=July 1/US to July 2/US). These were chiefly her progress reports and "in the clouds" weather comments (not appropriate for stellar navigation). When one examines the detail of radio transmissions logged by the Itasca, it records that she was talking (transmitting), but not listening (receiving), as planned. Perhaps the most detailed (in my opinion) communications accounts are given by Burke, Strippel[7] and Goerner[8]. Ann Pellegreno[9]. In her duplicate Electra, Ann retraced Amelia's route in 1967, providing navigation, timing and fuel consumption insights. Analyses by Loomis[10] (path and progress throughout the night) and also Strippel[11] (aircraft technical data in his Appendix 1) do not add any convincing clues or evidence of her aftermath or fate. The analysis by Loomis[12] of reported documented radio transmissions imply that Amelia made reasonable progress toward or a little to the north of Howland Island during the night. She also crossed the International Date Line, turning her pm July 2, 1937 into pm July 1, 1937 (Howland). There was no arrangement with British radio stations in the Gilbert Islands (called "Kiribati" after 1979) for communication at this point. -5 Since the trailing wire antenna was not available, no Itasca DF was possible. Amelia and the Itasca crew were not fully aware that only close-in two-way radio communication between the Electra and the Itasca crew at Howland was feasible. The Itasca radio crew verified that radio propagation was reasonable through the night by two-way test transmissions with other ships in that Pacific area. There were several 3105 kHz routine reports heard from AE throughout the night. One was at he Island of Nauru[13] heard enroute between Lae and the Gilbert Islands ("Kiribati"); "weak but readable at times". Later, the Itasca radio operator on watch would hear her "weak but partly readable" periodic transmissions :15 and :45 past the hour. ------------------------ [6] Lt Myron F. Eddy. "Aeronautics" No.22, Jan 29, 1941 (a tutorial of limited distribution by (then)) "National Aeronautics Council, 37 W. 47th St. NY, NY. [6a] Pellegreno, page 195 -196. [7] Strippel, Dick, "Amelia Earhart", Exposition Press, Jericho, NY, 1972, (#B/E12s). See especially appendix A on installed radio gear and pages 119- 132, etc. for radio specs and usage. [8] Goerner, Fred, "The Search for Amelia Earhart", Doubleday & Co., Garden City, New York (#B/E12G), pages 71 to 78. This book is well known for its "captured and buried hypothesis". [9] Pellegreno, Ann Holtgren, "World Flight, the Earhart trail”, Iowa State University Press, Ames Iowa 50010, 1971 (#629.109/P38w). [10] Loomis, Vincent, with Jeffrey Ethell, " Amelia Earhart, the Final Story", Random House, (#629.130924 \ L863a), Chapter 4. [11] Strippel, Dick, "Amelia Earhart, the myth and the reality", Exposition Press, Jericho, New York, 1972, (#B/E12s), enroute S-Levels, pp 63 - 65 and especially pp 128 - 133; aircraft and radio installation, Appendices. [12] Loomis, Vincent V "with" Jeffrey L. Ethell, "Amelia Earhart, the final story" Random House, New York, (#629.130924/L863a). Page 103. Author Loomis reports results of his "hard information" type intelligence analysis of radio transmissions to develop a possible flight course. Unfortunately hard information is sparse here where it is most needed, namely after her futile departure from Howland. Loomis then develops the "ditched and captured by Japanese and taken to Japan" hypothesis. [13] Loomis, Pages 100 - 101. -6 After each Electra transmission, the Itasca radio operator gave a routine reply, but such replies were not acknowledged by AE. This introduces the possibility that AE's 3105 kHz receiver was either not very sensitive[14], though she planned to use it only for DF (listening) purposes. AE had only had one good HF receiver[15]. The 7,500 kHz signal was to be made available at an appointed time. In those days, experts believed that both day and night HF-DF and long distance communication was always possible at 7,000 kHz (40m) & higher frequencies. Experimental HF-DF equipment[16] was on board the Itasca, carried on shore for the AE arrival. Tragically, no functional use other than misleading confusion came of it[17]. Putnam may have been overly optimistic about the efficiency of that HF-DF. Today we now know that sky waves, with their varying phase and polarization make "nulls" poor to non-existent at 7,500 kHz. Voice messages were warbled, but intelligible after a fashion, but HFDF resolved to be useful only close- in (less than 25 miles, I estimate), defeating the promised long distance advantages. I suspect that the 7,500 kHz equipment and its white hopes were among AE's CAB test cases. The 500 KCS DF process suffered very little of these anomalies, but AE had removed (with Putnam's concurrence) the LF trailing antenna in Miami and its tuning coil in Lae -. MF DF was possible on 3105 kHz, but her cooperative efforts were not adequate (microphone transmit button not held down long enough). Myers, p 95-96 indicates that the Itasca HF-DF set unit was portable and was deployed on land using batteries borrowed from the Itasca's gun turret. Cipriani's attempts at HF-DF fixes on Amelia's short transmissions on 3105 or 6210 kHz were a failure. The 3105 kHz receiving apparatus on the Itasca could not provide timely direction steering information[18], requiring as much as two minutes for a "fix". In preparation for this voyage, AE abandoned the CW key. This mode provides a slow but reliable communications mode in difficult situations, but requires some skill to master. As a result, those of us skilled in this communications art can only feel frustrated as we read the facts recounted by all authors about these final hours. For instance, the Itasca radio watch crew asked by voice (on 3105 kHz) for long wave (400 kHz or 500 kHz) transmissions from the Electra for more accurate DF data, not knowing that the 250' long wire was not available for that task[19]. ////////////////To here17:47 14 Dec,1996//////////// ------------------------- [14] Communications analysts should take note of this point. Did she not confirm (not QSL) at all as the Itasca log and other logs imply, or did the station loggers always use the linguistic implication that they were the only one that spoke, it being implied that they need not log QSL's if they occurred. A semantics point, I admit, but I still wonder! In think that her MF/HF receiver had to be stone dead on 3105 and 6110 kHz for the logs to be literally accurate as written. Another speculation is that, in her excitement she was working "split frequency"; transmitting on 3105kHz (and later 6210kHz) while listening on 7,500kHz or other frequency. There is evidence that the latter occurred at times. Having flown myself since 1964, I observed then pilots without a ham background did accept this method, especially when the ground station in question was transmitting continuous signals for one reason or another. The 7,500 kHz signal transmitted by the Itasca by previous agreement might have purveyed this impression. In other words, she perceived the Itasca as an aeronautical nav-aid station rather than the routine traffic radio communications station that it effectively was. If the Itasca crew were briefed on this contrast, and were they able to get her to transmit on 3105 kHz for long periods for DF and to ultimately "talk her in" on 7,500 kHz, history would have been different.--- An alternative interpretation is that the logs of the Itasca were falsified, and by persons not expert in air communications practices and procedures. -7- ------------------------------------------- [15] Strippel, page 120, 122-123. The Electra had a Western Electric Model 13A 50 Watt Transmitter, capable of emitting 500 kHz or in two bands covering 2500 to 6500 kHz. The controls were modified so that a 3-position switch would allow either 500 kHz, nominal 2500 kHz or nominal 6,000 kHz transmission. The transmitter antenna was a "flat top" running from the top of the cabin to either rudder. The matching receiver was a Western Electric Model 20A. The original receiving antenna was separate and located under the fuselage on short posts. For this voyage, it was disconnected and a T-R relay installed so that the top antenna could be used for receive as well as transmit (ever have a T-R relay hang up?). A Bendix MN-20 DF loop was added above the cabin. (In addition, the configuration she embarked with had a reel- out, reel- in trailing wire antenna as well as a loading coil for 500 kHz. (This Trailing Wire Antenna was removed in Miami and the loading coil was removed in Lae. In the same action, she sent the one and only CW key they had back to Western Electric!) Burke, p 220-221; from 8am to 10am Howland time she searched the area for Howland Island. I speculate that she was transmitting on 3105 and listening elsewhere, perhaps to the Itasca 3,105 emission with the DF loop and/or listening to the 7,500 signal with the experimental (Army?) equipment. [16] Goerner, page 208 Loomis, p 61, indicates the Itasca unit was a US Army experimental portable HF DF receiver. Skip was not understood then. Some Itasca crew members took it on shore, using batteries for portable operation. [17] Loomis, Page 102. Whereas Amelia was under the impression that 7,500 kHz emissions would provide a homing beacon to receive, in reality her Electra's HF receiver could not tune above about 6,000 kHz. One is obliged to presume that the "black box" she carried aboard provided this service. [18] Loomis, page 102. Early radio Direction finding ("DF") equipment and techniques developed from low frequencies upwards. In 1937, the Navy's 500 kHz DF system was very good. The most common LF-DF steers were made on 375kHz, however. -8- Amelia had expected directional steering information ("DF-steer"), to be broadcasted by voice from the Itasca, to back up Noonan's solar navigation. Howland was a slim strip, 1/2 by 2 miles and all alone. (Only Baker Island, slightly larger, lay 30 odd miles south). She apparently had flown according to Noonan's navigation [21] onto the sun line-of-position, then north and south, back and forth. This extensive flight along the nebulous sun line of position was wearying her. Even when relatively close (probably less than 100 miles from what I could read in ALL accounts[20]) she never actually acknowledged (QSL'd) a voice radio transmission from the Itasca[22]. Rather, she demanded that the Itasca continue 7,500 kHz transmission for her own independent HF-DF steering[23] ------------------------------------- [19] Goerner, page 209, also Strippel, page 126 (Electra's DF loop covered 200-1500 kHz and 2400-4800 kHz) [20] Strippel, page 129. [21] Pellegreno, pages 206 - 207. [22] Strippel, page 64. Pellegreno, p 199. [23] Strippel, page 129. It was not clear as to how she was receiving the 7,500 kHz signal, as the WE 20A did not go that high. A "black box" was brought on board during voyage preparation (Miami?). That may have been another HF DF set. The Bendix MN-20 DF loop was reputed to work at 200 to 1500 kHz and 2400 to 4800 kHz. (page 126). p 130. The Itasca had three transmitters on board; types T-10 (2,000 - 4,050 kHz), T-14 (125 - 500 kHz) and T-16 (4,000 - 18,000 kHz). -9- During the final minutes of navigation by the line-of-position method, it takes a lot of confidence by the pilot to strictly adhere to the navigator's instructions; especially on final approach (the last ten miles). Thirty years later, even Ann Pellegreno [24] expressed serious doubt about her navigator's commands in the final minutes of her replicated flight and approach to Howland with a similar Lockheed in 1967. Clouds, haze, or both always posed a dilemma for the pilot[25]. Should he (she) fly low to look for the island or fly high to get another fix on the sun to verify that they have maintaining the LOP course to the island. All of this requires a sharp mind, a good memory and more than a little luck. The Army HF-DF radio capabilities reputed in 1937 and the US Navy participation may have played a mind game on Amelia and Noonan, lulling them into a "piece of cake" mentality with the promise of a nice morning HF QSO spiced with verbal DF steers while inbound to Howland. This mind set could have caused them to reduce their targeted offset along the island LOP to come hypothetically closer to Howland for better signal strength. When that signal strength pickup did not occur, the subsequent mind boggling fatigue likely reduced their sharpness needed to solve the communications dilemma. Towards the end of that attempted and missed approach, the Itasca radio crew begged her to make long transmissions on 3105 kHz in order that they could manipulate their direction finding equipment. However, she only talked briefly and whistled a little more, for a far short shorter time than the needed two minutes[26],[27]. ------------------------------- [24] Pellegreno. Ann Holtgren, "World Flight, The Earhart Trail", Iowa State University Press, Ames Iowa, 1971, (#629.109/P38w), pp160-162. [25] Myers, p 51-53; implies that the Electra's arrival at he Itasca at dawn was premature, since lights might be visible and flares were dropped in vain hope of recognition; that by the time the light was good, her allotted loitering time had expired. [26] Pellegreno, p 199. -10- Any pilot that has made instrument approaches can empathize with weary Amelia at this point[28]. Mechanically fly the prescribed approach down to minimum altitude! Look for landmarks . . NONE! Fly out the clock. Turn back; trace the reverse course; NOTHING. One climbs to verify position. Look some more… NOTHING! (despite the visual smoke of the Itasca[29]. Hopes of a HF-DF steer on 7,500 kHz was still with them according to the corresponding fragments of the Itasca log. There is a hint of north/south dilemma in their comments. Amelia, the perfectionist that she was, probably pursued this search to a frustrating degree. Of the 1150 gallons maximum fuel load at the outset, 950 should have been expended[30] at the time of her departure from the Howland environs. An interesting fact pronounced later in Ann Pellegreno's book is that she (Ann) made a 2500 mile hop from Honolulu to California with 800 Gallons of fuel, whereas in 1937, Amelia took off from Lae with 1100 gallons of fuel[31]. This implies that Amelia could have flown for several hours more after missing Howland! The speculations on her quantity of fuel remaining at this point and what she subsequently did with it are many!. Fuel consumption data in Burke's Appendix 1 supports this contention. Speculation has it those in situations of abundant fuel reserve can head for areas of denser island population. For a low fuel situation, this is a meager to hopeless choice. Baker island, no bigger or better, lies a few miles to the south. For all intents, it is one and the same target as Howland. The Phoenix islands lie perhaps 300 miles to the south, probably within range, but only if pre-planned. For a greater fuel-remaining scenario, the Gilbert islands that were over flown 600 miles west during the night hours, would be within reach. The dense Marshall Island group, (Japanese possessions, "friendly" in 1937) offered high probability of landfall, attainable under careful planning, and opens the possibility that they deliberately went there for other purposes. --------------------------------- [27] Strippel, page 132: evidence of the comedy of errors; removing the trailing wire AND the 500 kHz loading coil, insistence on the change out of the night (3105, good) frequency to the day (6210, bad, conflicting with 7,500 kHz service), dogged pursuit of the 7,500 kHz system despite HF-DF failure on all counts. A final insult lies in the fact that the British at Tarawa (Gilberts) was up and running on those frequencies and could have been immensely helpful. (See also [39]). [28] Strippel, page 155-156. Amelia had her problems in mastering the Electra. Due to its high performance and her totally civilian experience, Paul Mantz had to coach her through its demanding qualities. She learned to fly it "by the numbers", which is robot-like. This could be a problem when fatigued or when "conditions" depart from the "numbers"; like the radio communications plan, for instance. It is common knowledge among todays pilots that the worst part of a flight is ALWAYS at its end, the "approach", when unforeseen incidents spoil the day. [29] Pellegreno, p 200. [30] Strippel, Page 117 - 118 and his detailed Figure 1 Table, pp 120-121. [31] Strippel, Appendix 1A. Bottom notes. -11- At the last, she revealed (transmitting again "in the blind") that she would from that moment forward communicate only on 6210 kHz[32], in reality ruling out further 2-way voice communication[33]. Amelia's confident flight continuance and switching to 6210 kHz support that her intentions were of further and likely distant flight.[34] The Itasca begged her to remain on 3105 kHz where two-way communication was feasible, but she never acknowledged, and she likely never heard that plea[35]. Hypothetical positions [36] imply that Amelia was a bit north of Howland. Aircraft data[37] imply that Amelia could have had significant fuel remaining at this point, at least a few hours and perhaps more. Her report of "fuel running low"[38] may have been in the context of LOITERING fuel remaining (that portion of her total fuel on board allocated solely to survey the Howland environs to sight it), and not total fuel remaining to exhaustion, which was likely to be 200 gallons? Prudent pilots would not rule out a retreat to the Marshall or Gilbert Islands and look for a more favorable ditching site near an inhabited island. The Loomis predictions would place Amelia somewhat north of that course, causing her ditching to more likely be in the Japanese held Marshall Islands. Later, Nauru Island radio picked up some 6210 kHz transmissions which might have been Amelia, "speech not interpreted, excited, without hum of airplane in background"[39]. ------------------------- [32] Goerner, Page 209, Pellegreno, page 200. [33] Strippel, p 130; The Itasca had only one transmitter for frequencies above 4,000 kHz. The 7,500 kHz "beacon" service tied it up! Perhaps the Itasca crew never though of sending voice over it. [34] Loomis, p103-105; In earlier discussions, Amelia told friends Vidal and Miller that if she did not find Howland Island, she would fly back to the Gilberts and ditch on a smooth beach. [35] Strippel, page 65, 129 [36] Loomis, page 104. [37] Strippel, pages 167 - 169 and Pellegreno, page 204, corroborated by Ann on page 174; "Oakland was 2461 miles from Honolulu. Fuel on board was 800 gallons for 20 hours, 20 minutes. Estimated time enroute was 17 hours 20 minutes". (Amelia departed Lae with 1150 gallons of fuel.) [38] Strippel, page 129. [39] Pellegreno, page 207. -12- Still later, Myers[40] gives a dreamy and perhaps not credible account of radio transmissions by Amelia for days afterwards. Radio amateurs at that time reported hearing what they thought was Amelia's voice transmissions, but there was no good way to verify and to use this information under those conditions[41]. It seems that naval radio operators or authorities considered these radio amateurs to be intruders or bunglers, and would not place any faith in their reports or "assistance". ------------------ [40] Myers, Robert, "Stand by to die", 1985 (#B/F12/M996s) Lighthouse Writer's Guild, P.O. Box 51277, Pacific Grove, California. pp 57 - 69. (See [41]). [41] Myers, pp 54-56, (through a ghost writer), alleges that while a 15-year old youth, he knew Amelia in Oakland during her early 1937 flight preparations. He claims that earlier, he saw the voids of her Electra being filled with ping-pong balls to facilitate flotation in the event of ditching. Over nine successive days (2-11 July, 1937) Myers claims to have monitored her transmissions with his father's all-wave Philco superheterodyne radio set. It is difficult to judge his credibility, as his story is rendered in a primitive and sentimental fashion. Myers claims to have heard Amelia issuing position reports throughout her flight, especially one overnight on 3105 to 3110 kHz reporting to be 500 miles from the Itasca. As dawn came she reported seeing Baker Island at 6:10am. Around 6:15, she talked to the Itasca radio operator. Around 6:30 she reported another ("Canon" or” Canton") Island. {PS 17Mar92: we now think it is "Gardner"} Subsequently, Amelia followed her declared convention that she would transmit on 6210 kHz. Notes on a sketch (ex-post-facto?) by Myers indicate a search in a 157-337 line was followed by a ditching south of Howland in the general Hull Island area with comments of "into the wind", "water calm and smooth". Another part of his record indicates that the ditching was on the east side of a unidentified island, that Noonan was slightly injured, that the aircraft was oriented a little nose-down, that "the propellers were a mess (bent)", that she was worried about "a battleship I saw steaming toward you (Howland)", but that she and Fred were alive and that they were "staying with the plane". Myers claims that her Electra was well suited to float for days(ping-pong balls in all cavities) and that she issued daily pleas for rescue and then reported a rescue by a Japanese war vessel on July 11th, 1937!. The above hypothesis ties into that of Loomis if one allows that it was a Japanese aircraft that flew her and Noonan back to Saipan with an emergency landing in the Mili Atoll. Myers' account is at the edge of credibility. It is now well known that the capricious "skip" character of 6 MHz radio waves can fully account for the bizarre series of events that followed. That is, some U.S. receiving stations could have easily heard most transmissions from that area of the Pacific Ocean whereas the search ships did not hear those transmissions due to "skip". The U.S. radio amateurs reporting on the same frequency of such interceptions (at dusk or night) were scolded as being out of order, and also neither credible nor welcome by the Navy and others. -13- Present-day speculation centers about identifying the most likely place of ditching or landing. Possible final scenario choices include any combination of the following behaviors while flying over the endless and featureless murky grey Pacific seascape in search of the tiny Howland and Baker Island dots and other landing or ditching locations. That hypothetical 200 gallons fuel reserve was good for about five more hours of flight or about 700 miles[42], which includes the Marshall or Gilbert Islands, provided she conserved fuel and did not meander in navigation; all a feasible "long shot" under those combined conditions. Subsequent speculation and documented hearsay would have it that they ditched near a strange island. Still further speculation is that this unfortunate pair of whites were picked up by authorities and mistaken for despots or beachcombers, were badly treated and perished leaving a highly questionable grave. Another speculation is that the Japanese military found them, and interned them as spies, eventually disposing of them[43]. Most practical speculators conclude that they simply ditched at sea, where they launched a raft whereon they waited for rescue[44]. *************************************** SUMMARY: I feel that the Itasca radio crew was not properly prepared for the scheduling and guidance discipline vis-a-vis aeronautical conventions and practices required in this mission, even for 1937. The delay (from March to June) from the original schedule may have "broken up their act" so that less resources were available to support the actual flight. The Itasca crew did not plan for possibility of emergency voice modulation of ALL their emitted frequencies (a common practice today in aviation Nav-Aids, which partially compensates for the tendency of pilots to "broadcast and monitor" rather than "2-way QSO and QSL" as is the practice of commercial and amateur radio operators). One gets the impression on reading the documented Itasca radio logs, in that morning encounter, that NO frequency was worked productively. From the point of view of today's safety conscious mind set, both Amelia and Fred Noonan, a good experienced navigator, appear to be unprepared to deal with communications problems that combined the practices of ships with those of aircraft[45]. Their intra-cabin communication was ostensibly by notes passed back and forth on a stick over the top of a fuselage fuel tank (no cabin interphone). Amelia operated the radios, so Fred had little opportunity to help. Though alternate lower HF and MF frequencies were available, neither Amelia nor the Itasca crew made good procedural use of them. It is not unreasonable that Fred should partake of that responsibility, especially for steering on the low altitude search for the Itasca. They could have, for instance, asked the Itasca to talk over the 7,500 kHz transmission or switch it off and on if they heard her engines, etc. ----------------------------------- [42] Loomis, p 130; Putnam, in his urging the Japanese to search the Marshall Island area, knew that a 500 mile trip by the Electra from Howland was possible. [43] Loomis, p 112 claims that the new Japanese generation (1985) has interest in pursuing the truth in this matter. [44] Myers, p57-69. -14- Amelia or Fred did not plan for the contingency of poor to nonexistent DF steers. The correspondingly small accuracy circle demanded for the sun LOP technique (to account for poor transmitter and receiver performance plus noise, less than 50 miles radius required for MF and HF day ground wave worse conditions at sea). Its ancillary demands to achieve such precision include good weather and accurate wind forecasts, or prudent options in their absence (Charles A. Lindbergh, "CAL.", was a practiced and meticulous planner in this regard). Recall that the leg they were flying was about 2500 miles long. Forget dead reckoning to Howland. Even a 100 mile offset is marginally too small. In tracking that sun LOP, Noonan's sun elevation observation had to be accurate to less than 30 minutes of arc or 1/2 degree (one minute of arc is one nautical mile) while underway (a 30-second average of steady course manual sight averaging was common in those days). And all of this while Amelia was trying to make sense of and/or steer by what she heard or did not hear over HF radio. It appears that her trust in radio contact eroded slowly. She transitioned gradually into realizing the futility of her search for Howland. Flight to her alternate - distant as it may be - was her only choice. This does not detract from the ambitions, significance and bravery of their actions. Simply put, as every modern pilot will agree, "their luck ran out". The ultimate conclusion (in my opinion, but you may deduce your own) is that with perhaps four hours of fuel remaining, a plaintive flight occurred, perhaps first to the to the south (Baker and the Phoenix Islands) and then later to the north toward the Gilberts or Marshalls to fuel exhaustion and finally ditching. Landfall, if any, would have been either early in the Phoenix or later near the Gilbert Islands ("Kiribati") where they would have waited for rescue by naval search ships. Barrier reefs perplex all coastal operations around such Pacific islands as noted later by Thor Heyerdahl in "Kon-Tiki"[46], especially in stormy seas. In retrospect, CAL would have made the Phoenix Island group his first visual landfall with no less than six hours of fuel reserve for locating the Phoenix Islands, successive navigation to Howland environs as well as a retreat to the Gilberts when Howland did not appear after limited loitering. At that time, he would also have had radio aboard operated by his wife, Ann, including CW. ----------------------- [45] Strippel: Page 119 (vehicle transmitting and receiving frequency capabilities). Page 125 (craft frequency use conventions and options). -15- ---------------------------------- [46]Heyderhal, Thor, "KON-TIKI, across the pacific by raft", Rand McNally & Company, New York, San Francisco, 1950. Chapter 7 describes their extreme difficulty in surviving their crossing of barrier reefs of the group they encountered (Puka Puka in the Tuamotu Archipelago) which could tear their raft apart in seconds. Large waves emitted from distant storms were always a threatening force. Only by extremely skillful seamanship in raft sailing, were they able to explore the perimeters of several more islands island to find a safe passage to an uninhabited land. Navy search personnel, both air and sea, reported storms northwest of Howland toward the Marshall Islands on July 2, 1937 and for days afterward. ------------------------------------ -COMMUNICATIONS CRITIQUE EPILOGUE (re Amateur Radio): 1// The apparent lack of CW and voice communications skills underscores one important mission of amateur radio today; namely to educate our youth with information and communications practices (5 WPM CW will do), as such knowledge will serve them handsomely for the rest of their natural lives[47]. Were any amateurs or "Elmers" to know Amelia in her youth, history may have been changed. 2// The ill feeling toward amateur radio operators, such as occurred in the early days of the search might be abated if we could tie together any and all radio log information - albeit 1/2 century later. Our logs may have some merit after all!. 3// Today, we have a better pool of scientific knowledge of propagation and known aircraft characteristics through the research of such persons as Loomis, Strippel, Pellegreno, and even the reports of Myers! The amateur radio community can quantify the aspects of radio propagation and logged (re QST & other logged sources) so that one can predict their most probable location of Amelia's Electra at the time that their fuel was about to expire, and whether there were any transmissions by her thereafter. Aeronautical and political speculators can then project from that time and place as to what is most likely to be true. 4// Who knows, maybe a global ARES may result! -16- ------------------------------------- [47] Strippel, on page 132, elaborates on the naivete of all involved parties regarding communications capabilities and procedures. These include such "trivial" (not so to a practiced ham) items such as communications and DF loss without a loading coil, nonspecific frequency terminology (is "7500" to be kHz or meters?), fear of CW, and of all things, Amelia's opinion that she did not expect anyone to hear her most of the time while she gave position reports over the ocean (i.e. she held no value for the QSL concept). She bypassed a chance to talk with the British station on Tarawa Island. Any one of our respective ham proteges would have been "working them all"! Our proteges would use good procedures, and be "welcome on the band", thereby turning irritation into opportunity. [48] Mary S. Lovell "Sound of Wings" p235. -------------------------------------- *********************************** POSTSCRIPTS In September, 1988, I looked over the '37 -'38 issues of QST to find only one note: "..... (a ham) was on the 'Colorado' early July during the search for Amelia". I could find no other direct reference to the incident. In a 1938 QST, I found a verbal account of the radio outfitting for Howard Hughes' 'around the world flight' in a very similar Electra. This aircraft had DUAL transceivers and planned HAM schedules on 40m and 20m CW. Howard did it right! Note: Locally, W8OZA says that perhaps more pertinent SWL reports of that era might be found in found in "Radio News", edited by Hugo Gernsback. Perhaps someone around here can find one or more of these at OSU or the Ohio Historical Society. ***1996Update*** Reviewed a few more books. Lovell's "The sound of wings" clarifies that AE intended to only transmit on :15 and :45 by voice, her position and comments. The remainder of the time she intended to use the radios for DF information. The Army HF DF set was only on the Itasca and later on shore on Howland. It was not capable of producing quick DF data. Lovell supports the "ditch at sea nearby" scenario. ***1995 ***** 26 June: AOPA Pilot announces that Linda Finch will reconstruct a Lockheed 10E and retrace Amelia's trip. I am sending her a copy of this summary for reference. June 10: A new book, "Lone Star-Search for Amelia Earhart" by Randall Brink. 1993: TIGHAR's search ended up with a piece of sheet metal.. Their search site (Gardner Island) was indicated by an analysis of radio message logs of the era. 00:02 17 Mar,1992 The Island identified by Gillespie, Gannon and Willi of TIGHAR is Gardner Island, now named Nikumaroro Island. ~~~~~~~ In addition, on re-examining my notes from various references, those of Robert Myers (1985) mentions Hull Island (near Gardner). //////////////// ** 15 March, 1992: Just saw in Columbus Dispatch that Gilbert found an aircraft sheet metal ruminant and shoe on the Island of Nikumaroro; press conference in Washington on 16 March 1992... I decided to 'publish' this article on the CCS BBS on the communications aspects of this disappearance. ///////////////// ** A 1987 book "Eyewitness" by Thomas E Devine details his 1960-63 investigation of Saipan burial sites.. One hypothetical burial site was not investigated. (Loomis and others hypothesize that Earhart and Noonan were found by the Japanese, flown to Saipan, and then executed.) October 22, 1988, W8EDR, Columbus, Ohio AMELIAER.ART, VERSION 3 (With interleaved footnotes & W8OZA QST refs October, 1988) //////////// October, 1988: I recently read "200 meters and down", and noted the enthusiasm about HF that was developing in the '30's. Amelia Earhart had problems with HF in her last 1937 flight. Much of what has been written was done just after a new "fact" or "discovery", thus prejudicing the thinking of the moment. No "revelation" having occurred recently, I feel one can reexamine the circumstances with objectivity. ------ - . . . W 8 E D R "Andy"-------------. . . . - . - 25 Dec 87 (Version 1): 5 Jan 88 AJC (Version 2): Jun-Sept 88 (Version 3) Readers; Please send comments on BBS [W8EDR] or LL [(614)876-5108. 28 August, 1989: On Tom Willi's interest, I've reproduced a recent copy. 25 August, A slightly edited (took out some negative conception Amelia's communication capabilities) version reproduced for Walter Burnham. Angelo Campanella. 13:52 25 Aug,1995 //////////