Jeff Wise

If you were leading a high-profile international aircraft investigation, in command of the world’s most qualified technical experts and in possession of all the relevant data, would you bother listening to a rag-tag band of internet commenters, few of whom actually work in the space or aviation industry, and none of whom have access to all the data?

Most likely, you’d say: certainly not! But as time goes by, and the puzzle remains curiously impenetrable, you might find it worthwhile to pay a listen to what the amateurs were saying. You might even abandon some of your own conclusions and adopt theirs instead.

This appears to be the case in the search for Malaysia Airlines Flight 370, which disappeared en route from Kuala Lumpur to Beijing back in March. From the beginning, the authorities running the investigation — first, Malaysia’s Ministry of Transport, and later the Australian Transportation Safety Board (ATSB) — held their cards close to the chest, releasing very little information about the missing plane and maintaining a posture of absolute conviction.  The investigators’ self-confidence reached its apex in April, when their methodology led them to an area of ocean where underwater accoustic signals seemed to be coming from pingers attached to the plane’s black boxes. Officials assured the press that the plane would be found in “days, if not hours.” But then it wasn’t. A scan of the seabed found nothing; the pingers were a red herring (perhaps literally!). Back to square one.

Meanwhile, on the internet, a group of amateur enthusiasts had come together from all around the world to trade ideas and information about the missing flight. The group, which came to call itself the Independent Group (IG), emerged from various online comment threads and eventually grew to about a dozen individuals. This was a truly spontaneous, self-assembling crowd: there was no vetting of credentials, no heirarchy of any kind. (Full disclosure: I count myself among this group.) Basically, if you seemed to know what you were talking about and could comport yourself in a collegial fashion, you were accepted into the crowd.

While the mainstream press was reporting the ATSB’s pronouncements as received wisdom, the IG was raising red flags. IG members were among the most vocal critics of the ATSB’s contention that the accoustic pings probably came from black-box pingers. And later, after a public outcry led Inmarsat to release a trove of data received from the aircraft, and the ATSB issued a report explaining how it had come to identify its current search ear, the IG dove into the new information with abandon, quickly identifying holes in the data and weaknesses in the official approach. In a pair of papers, the group recommended its own search area, hundreds of miles to the southwest of the ATSB’s officially designated  zone.

Today, the ATSB has released an update to its earlier report, explaining why it has decided to reassess its conclusions and move its search zone to a new area — one that overlaps, as it turns out, with the IG’s recommended area. (In the graphic above, the white bracket shows the ATSB area; I’ve added a yellow dot to show the IG area.) Needless to say, this has caused elation within the ranks of the IG, who see the move as vindication of their methods, and indeed validation of their combined efforts over the last few months.

A few observations on the new report:

Continue reading In Search for Missing Airliner, Peanut Gallery Shows the Way

It’s been more than six months since MH370 vanished, and in some ways we know no more now than we did in late March: no new clues have emerged, no more data has been discovered.  In a sense, though, we have come a very long way. For one thing, we now understand how many of the “breaking news” developments that occurred in the early days were actually untrue. (There were no wild altitude swings, no “fighter plane-like” maneuvering, and probably no cell-tower connection with the first officer’s phone.) What’s more, thanks no doubt to a drumbeat of public pressure, the authorities have released a tremendous amount of data and provided useful explanations of how that data is being interpreted. And finally, a spontaneous collaboration between technical experts and enthusiasts around the world has provided a trove of insight into avionics, aerodynamics, satellite communications, and a whole host of other topics that collectively shed light on what might and what might not have taken place on the night of March 7/8, 2014.

While a great deal of information has become available, it has not always been easy to find; much of it, for instance, has been exchanged via email chains and Dropbox accounts. For my part, I often find myself rummaging through emails and folders looking for information that I’m pretty sure I’ve seen, but can’t remember where. So what I’d like to do with this post is try to aggregate some of the most basic facts — a set of canonical values, if you will, of the basic data on MH370. Necessarily, some of this data comes with implicit assumptions attached, so as far as possible I’ll try to make these assumptions explicit.

Okay, on to the data. What we know now:

The bedrock data. In the wake of MH370’s data, there were numerous news reports concerning information leaked by anonymous sources from within the investigation and elsewhere that have subsequently been either disproven or inadequately verified. For the purposes of the present discussion, the following are considered the bedrock sources of information upon which our understanding of the incident can be built — the “Holy Trinity” of MH370 data:

1. Up to 17:21: radio communications, ACARS, transponder, ADS-B
2. 17:22-18:22: military radar track. This information is of uncertain provenance but has been endorsed by the governments of both Malaysia and Australia. Furthermore, it plausibly connects the prior and following data sets.
3. 18:25-0:19: Inmarsat data, especially BFO and BTO values. There is some discussion as to how this data is best interpreted, but the numbers themselves are assumed to have been received and recorded by Inmarsat from MH370 via their 3F-1 satellite. The “ping rings” in particular are derived through relatively simple mathematics and should be regarded as established fact unless someone comes up with a specific mechanism by which some other result could be obtained.

Timeline. Courtesy of Richard Godfrey and Don Thompson, here is a basic timeline of MH370’s disappearance (all times UTC):

• 16:41:43 MH370 departs runway at KUL runway 32R
• 17:01:14 MH370 flight crew report top of climb at 35,000 feeet
• 17:07:48.907 Last acknowledged DATA-2 ACARS message sent from plane
• 17:19:29 Last radio voice transmission
• 17:21:04 Plane passes over IGARI waypoint
• 17:21:13 MH370 disappears from air traffic control (secondary) radar screens
• 18:22 Last primary radar fix
• 18:25:27 Inmarsat log-on request initiated by aircraft
• 0:19 Final transmission from aircraft to satellite

A more complete table of values, including the location of the plane at each point in time, can be found here, courtesy of the inimitable Paul Sladen. And Don Thompson has created an impressively detailed breakdown of the sequence of events, with a special focus on radio communications between the aircraft, ground, and satellite, here.

More stuff after the jump…

Continue reading What We Know Now About MH370

 

In what must as the most significant act of rapprochement in the history of underwater aircraft-accident investigation, it was learned today that the Fugro Equator, a bathymetry vessel contracted to map the seabed as part of the Australian government’s efforts to locate missing Malaysia Airlines Flight 370, has begun exploring the area that the Independent Group has identified as the jet’s most likely resting place. The chart above, supplied by Geoff Hyman [update 10/3/14: I’ve replaced it with a newer image from Don Thompson], shows the course of the Equator’s recent progress in red. The center of the Independent Group’s recommended area is marked with a blue cross. For months, the Australian Transport Safety Board has been focussing on an area hundreds of miles to the northeast, on the basis that flight routes that best fit Inmarsat data are most likely to terminate in such an area. The Independent Group has taken a different approach, and focused on routes that are both psychologically plausible and comport better with the way that airliners are actually flown. It is not known if the Equator’s current search zone indicates that the ATSB has come around to the independent expert’s line or reasoning, or is simply being thorough in mapping the entire zone where the plane might have come to rest.

Yesterday the “Independent Group” (IG) of technical experts looking into the disappearance of Malaysia Airlines flight 370 (of which I am a part) released a new report which made the case that the official search area now being scoured by undersea robots is not where the plane most likely crashed. The reason, the group explained, is that the Australian Transport and Safety Board has relied on a statistical model in which hundreds of possible paths were generated, then winnowed down to include only those that fit the timing and frequency data from the seven handshake pings; this resulted in a distribution whose greatest density coincides with the current search area. The Independent Group, in contrast, began by asking what possible routes most closely match the flight speeds and altitudes that a pilot would most likely choose:

The ATSB analysis used two basic analysis techniques referred to as “Data Driven” and “Flight path/mode driven”… While we agree that these statistical methods are reasonable techniques, both tend to overlook or minimize likely human factors in favor of pure mathematical statistics. This ATSB approach appears to have resulted in a conclusion that the most likely average speed was approximately 400 kts (Appendix A). However, 400 kts is not consistent with standard operating procedure (typically 35,000 feet and 470-480 kts), nor is it consistent with the likely speed a pilot would choose in a decompression scenario (10,000 feet and 250-300 kts). A speed of 400 kts may minimize the BTO and BFO errors for a given set of assumptions, but the errors can also be shown to be very small for other speeds. Given all the tolerances and uncertainties, we believe it is important to consider human factors with more weight… B777 pilots consistently tell us that under normal conditions, the preferred cruise attitude would be 35,000 feet and the TAS would be approximately 470-480 kts. We believe this is the most likely case for MH370, and note that the last ADS-B data available indicated that MH370 was at 35,000 feet and 471 kts at that time.

As can be seen in the chart above, the differing approaches result in search areas that are some 500 miles apart. The full report can be found online here.

UPDATE 9/12/14: Richard Godfrey has pointed out that a recent report from the ATSB  shows that the seabed-mapping effort has recently been extended some 200 nautical miles toward the IG search area:

 

 

At a press conference in Canberra today, Australian Deputy Prime Minister Warren Truss state that “further refinement of satellite data” indicated that missing Malaysia Airlines Flight MH370 had turned south earlier than investigators had originally thought. This implied, he said, that the plane had most likely would up further to the south than previously estimated.

The previous assumption was laid out in a report released in June by the Australian Transport Safety Board (ATSB), which included the map shown here. The document described a methodology for determining the search area which suggested that the plane did not make a single turn to the south and then fly on a straight-ahead course into the southern ocean, but rather lingered near western Sumatra for the better part of an hour.

After the report was issued, a loose coalition of experts from around the world called the Independent Group (of which I’m a part) released a statement which questioned the ATSB’s methodology, and in particular pointed out that signal data related to an attempted satellite phone call at 18.40 UTC indicated that the plane was already established on a course to the south. This fact allows the range of possible flight paths to be narrowed considerably.

As fate would have it, Truss’ announcement came just one day after the Independent Group issued a follow-up statement reminding the authorities that its own analysis suggested a search area futher to the south.

“The data is nothing new, but the fact that the Australian government has chosen to issue this statement is very interesting,” says Independent Group member Victor Iannello.

Truss was vague as to where the priority search area had shifted, saying it remained “within the search area” previously laid out in the southern Indian Ocean. This area, however, is more than 1,500 miles long.

It’s been two months since I last posted about MH370, so I think I’m overdue for an update. The big news that’s happened in the meantime is that on June 26th, the Australian Transport Safety Board (ATSB) released a report that laid out in admirable detail what the authorities felt they knew about the circumstance of MH370’s disappearance and how they had come to narrow down its likely location to the current search area. We now have a much clearer understanding of just what Inmarsat’s data reveals about the last four hours of the flight.

In the report, the ATSB explains that sometime after the plane vanished from radar screens at 18:22 GMT, whoever was in control  most likely became incapacitated and the plane flew on autopilot until it ran out of fuel and crashed into the depths of the Indian Ocean some time after 0:19 GMT. Its impact point, according to the ATSB’s calculations, was most likely somewhere in a region 1,000 miles off the coast of Western Australia.

As you’ll notice, that last sentence is extremely vague. The reason is that, as we now understand, the data is incapable of telling where the plane went with any degree of certainty. That is to say, you cannot recreate the airplane’s flight path using Inmarsat data alone. This is kind of a shocker, because for months now, Inmarsat has been telling the public that their mathematical wizardry had allowed them essentially to solve for the plane’s final location. This turns out to be false. For any given flight path, we can now calculate the expected BTO and BFO values; but given a set of BFO and BTO values, we cannot derive a unique flight path.

Take a look at the route chart at the top of this post, which comes from page 31 of the ATSB report. It shows the actual path of a Malaysia Airline jet that flew on the same day at MH370, as well as an “estimated” path generated from the BFO and BTO data recorded from that same airpline. If you’re like me, when you first saw this chart, you assumed that Inmarsat had decoded the BFO and BTO data so thoroughly that they could generate specific flight paths like this one. Give them BFO and BTO data, and they could draw you a line on the map; no wonder they say they know exactly where to spend $56 million scouring the sea bed. But if you read the accompanying text, what it says is: “Using only the starting location and an equivalent number, and approximate time spacing, of BFO and BTO values as the accident flight, predicted paths were created and compared against the actual flight paths.” Reading between the lines, the ATSB was able to generate not just the route illustrated here, but dozens or possibly hundreds, and chose this one as the most promising for MH021.

There’s another disappointing realization embedded in the ATSB report. Continue reading What Could Have Happened to MH370?

A hundred days have passed since MH370 went missing — and while air and sea search operations have been put on hold, hope springs eternal. Today, the BBC is reporting that Inmarsat remains confident that its analysis of the satellite data will lead to the plane, saying that the authorities never searched the area of highest probability because they were distracted by the underwater acoustic pings that turned out not to have come from MH370’s black boxes. Once a new search gets underway, it will explore an area that conforms much better to the likely speed and heading of the missing plane:

By modelling a flight with a constant speed and a constant heading consistent with the plane being flown by autopilot – the team found one flight path that lined up with all its data. “We can identify a path that matches exactly with all those frequency measurements and with the timing measurements and lands on the final arc at a particular location, which then gives us a sort of a hotspot area on the final arc where we believe the most likely area is,” said Mr Ashton.

Unfortunately, it will be several months before such a search of this new area can get underway, since the survey of the ocean floor will be required to figure out how deep it is and what kind of underwater technology should be used. Meanwhile, a spokesman for the Australian organization leading the search described a more complex and ambiguous state of affairs, telling the AFP that experts were still struggling to narrow down the highest-probability search area, taking into consideration not just the satellite data but also “aircraft performance data [and] a range of other information.”

What other information? Your guess is as good as mine. As I wrote last week in Slate, Inmarsat has by now leaked enough clues about MH370’s electronic Inmarsat “handshakes” that outsiders can now understand why, mathematically, the plane must have gone south. Yet we have not the slightest hint of what sequence of events might have taken it there. We don’t even know how it could have navigated southward. An airliner like the 777 doesn’t just wing off in random directions like a paper airplane; its Flight Management System would have been following a series of waypoints or a compass heading. Yet its range of possible courses doesn’t seem to match up with any particular heading or waypoint. (The last search area matched up with a flight route that tracked waypoints between the Cocos Islands and Australia, which is likely one of the reasons it seemed so appealing to authorities, but as we now know, that came up empty.)

MH370 looks to be a unique case not just in aviation history. No machine this big, no group of human beings this large, vanished so completely and so mysteriously since the advent of modern technology. What’s more, MH370 didn’t just disappear once, but three times.

The first disappearance, of course, was when it vanished from air traffic controllers’ screens in the early morning hours of March 8, apparently after someone turned off its transponder and automatic status-reporting equipment, and took a hard left turn. Based on the speed and precision of its navigation, the plane almost certainly was under human control.

The second disappearance occurred about an hour later, as the plane slipped beyond the range of military radar. Minutes later, some kind of unknown event caused the plane to transmit a mysterious triple burst of electronic signals to the Inmarsat satellite. At around the same time, the plane took another radical course change, pivoting from a northwest heading toward mainland Asia to a southwestern course that would take it over western Indonesia and out into the open ocean. Based on the slim evidence of subsequent Inmarsat pings, the plane seems to have flown in a simple straight line, so it may not have been under human control at that point.

Then it disappeared a third and final time, this time leaving not a single clue. Continue reading The Triple-Disappearing Airplane

Two weeks ago, after months of mounting public pressure, Inmarsat and the Malaysian government finally released the raw satellite data that had been received from the missing Malaysia Airlines Flight 370. Most of the data dump proved unrevealing. But tucked away amid 47 pages of detailed communications logs and explanatory notes was a two-sentence description of the plane’s electronics system that turned out to be a doozy. Combined with previously released data, publicly available information, and a little vector mathematics, it has proved sufficient to lift the veil on Inmarsat’s calculations and reveal the ultimate fate of the plane.

The story goes back to March 25, when Malaysian authorities announced that an analysis of the data had determined that the plane must have wound up in the southern Indian Ocean. An explanatory document released at the time purported to back up that claim with charts and numbers, but as I’ve written earlier, it in fact was so obtuse that it didn’t really clarify anything at all. The message’s subtext was basically: Trust us, we know what we’re doing. But the subsequent behavior of the search officials—who, among other things, promised that they’d located the plane underwater but then came up empty-handed—left little room for confidence. Many, including me, wondered whether the authorities were hiding something, or else trying to conceal how little they knew. Continue reading Slate: Where the Missing Plane Went

UPDATE 5/21/14: The families of missing MH370 passengers have released a fascinating document presenting their own analysis of the preliminary report issued by the Malaysian government’s Ministry of Transport, including their own assessment of what we know and what we’d like to know. Link: Analysis of the Preliminary Report on MH370 Incident, May 20 2014

by Michael Exner

[Note: The totality of what we know about the fate of missing Malaysia Airlines Flight 370 consists of a series of electronic handshake “pings” that were received by an Inmarsat satellite in the hours before the plane disappeared for good. Unfortunately, the authorities have steadfastly refused to release the full data set to the public, and there is an ongoing dispute between Inmarsat and Malaysia as to who exactly has the data and who is authorized to release it. According to CNN, a source within Inmarsat has said that the company released satellite ping data amounting to just 14 numbers to Malaysian authorities, along with documentation explaining their methods for analyzing the data. Here Michael Exner, Chairman of the Board of Radiometrics Corporation, weights in on that claim. — Jeff Wise]

In fact, we know there are at least 51 numbers, and here’s why. The BFO chart [released on March 25 by the Malaysian government as page two of “Annex I” accompanying the Inmarsat report — ed] shows 12 times and 12 frequencies. That’s 24 numbers.

Then, on April 29, there was a photo in Beijing that showed that there were more handshakes, and ACARS messages that preceded the first handshake on the BFO chart, but there were no BFO values given in the Beijing meeting with the families. But the fact that they had the angles proves they had the times and the BFO values. Thus, we know of at least 17 events for which they have Time, BFO and Angle (or time delays). That is 51 numbers total.

We have assembled the following data from two sources. The “BFO Data” provided in the March 25th AAIB ANNEX I Chart and the photo taken in Beijing on April 29, 2014.

Note that the numbers above represent our best estimates based on digitized paper graphs and photos. The true resolution is less than that inferred by the number of digits. Apparently, the statements about “…only 14 numbers…” are in reference to the last 7 BFO frequencies and last 7 elevation angles, which the official investigation team is focused on. But all the data from the other events are also valuable for the calibration of the other data. Those first 10 events are also very important. Continue reading Guest Post: What We Know About MH370, and What We Would Like to Know