For the past six months, we’ve gotten conflicting information from different experts about what most likely happened during MH 370’s final hours. Evidence that the pilot had performed suicide runs out over the Indian Ocean on a trajectory not-too-dissimilar from the one MH 370 is believed to have followed, combined with one aircraft expert’s testimony that the flaps on the craft appeared to be deployed, suggested that the plane might have ditched. This would also have explained why the aircraft has been so difficult to find — an aircraft that glides into the ocean, doesn’t break up into nearly as much wreckage, and the pieces that are left are more likely to sink quickly.
Now, the Australian Transport Safety Bureau (ATSB) has thrown a bucket of water on these conclusions, stating that their own analysis shows MH 370 descending out of control. The board bases this conclusion on an analysis of the Burst Frequent Offset (BFO) data recorded by the orbiting Inmarsat satellite. The BFO is described as follows:
A function of the Doppler shifts imparted on the communication signal due to the motion of the satellite and the aircraft. The relationship is more complicated than a direct Doppler calculation because the aircraft software contains Doppler compensation that offsets the Doppler shift due to the aircraft motion. Although the aircraft attempts to compensate for its own motion, it does this under the assumption that the communications satellite is in notional geostationary orbit and it does not include the vertical component of the aircraft velocity.
Here’s what the BFO plot looks like for MH 370 — and as you can see, the BFO plot increases in a linear fashion from 18:39 GMT to 0:11.
At 0:19, Inmarsat recorded a partial handshake attempt of the kind that would be generated if the aircraft had lost power due to fuel exhaustion. Emergency power would have then kicked in, reactivating the satellite link. Note, however, that the burst frequency offset for MH 370 is now wildly different — and both of the dramatically different BFO values were recorded within the space of a minute. This BFO data is why investigators ultimately concluded the aircraft was on a southerly track — a conclusion borne out by the discovery of multiple pieces of debris in the more than two years since MH370 vanished.
If the trend had continued following its previous trajectory, the 0:19 value should have been 260Hz. Instead, the recorded offset values were 182KHz at 00:19:29 and -2Hz at 00:19:37. The ATSB report steps through several alternate explanations of this dramatic shift, including changes in the aircraft’s speed, warm-up drift as the crystal oscillator inside the SDU (Satellite Data Unit), and the inherent margin of error within the BFO metric itself. None of these metrics, even if combined and assumed to be at maximum impact, explain the dramatic decrease in the aircraft’s BFO in its last minutes of flight. The final point to be considered is the rate at which the aircraft is descending.
A descending aircraft in the position MH 370 is assumed to have been in would have lost 1.7Hz of BFO per 100 feet/minute of descent. The problem here is that the logon request at 00:19:29 and the log-on acknowledgement sent at 00:19:37 are so different from each other, it’s impossible to solve the problem with anything but an extremely high rate of descent, as shown in the tables below:
With a descent rate between 14,500 feet and 25,000 feet per minute, the aircraft wouldn’t have stayed in the air very long. This data was fed into a Boeing simulator specifically designed to model aircraft systems with a high degree of fidelity. The various paths shown below model the trajectories the aircraft might have taken, depending on its speed, turbulence level, remaining fuel, electrical configuration, and initial altitude.
These various paths show just how difficult it is to constrain our understanding of MH370’s final moments based on the information we have. In some simulations, the aircraft remained aloft for up to 20 minutes after flameout, while in others it plunged more or less straight down. Several tracks show MH370 looping back on its own flight path, or spiraling outwards (though generally within a 15 nautical mile point of the search area arc). These findings indicate that the 40 nautical mile (roughly 46 terrestrial mile) width of the search arc was sufficient to incorporate all of the uncontrolled descent paths fed into the system — but cautions that even simulators like Boeing’s can’t capture every detail. Specifically:
And, of course, we still can’t find the damn plane.
To understand how oceanic currents carried artifacts to the locations where they were found, CSIRO (the Commonwealth Scientific and Industrial Research Organization) conducted field tests on how aircraft debris, specifically, is affected by waves and wind. Replicas of certain MH 370 debris, like the flaperon found on Reunion Island, were constructed to match its buoyancy and weight, then tested in open water under various weather conditions. The flaperons were found to consistently orient with their raised trailing edges into the wind, allowing waves to propel them in the same direction as the wind. This produced a model that better fit the locations where debris has been found within the relevant time elapsed since the loss of the aircraft.
Simulation of debris drift patterns positing different impact points among the most likely arcs similarly suggest that the original search area corresponds well to the debris located to date. If the aircraft had crashed farther south along the arc, the bulk of the debris should have washed up in Western Australia or Tasmania, where no debris has been found so far. Similarly, if the aircraft had impacted farther north along the arc, debris from the crash should have reached Africa more quickly than it did.
As of this writing, the MH 370 investigators have received more than 20 pieces of debris that are considered to be of interest to the MH 370 disaster. The slideshow below steps through each individual piece and its current status.
Ultimately, we may never find enough of MH 370 to construct a plausible model of what happened to the aircraft. Unless we find a major debris field and sections of the wreckage, the bits and pieces located to date can only offer fragmentary clues about the aircraft’s final hours. CNN reports that their analyst, Mary Schiavo, believes this data points towards two early hypotheses: “namely, that a fire on board the plane incapacitated everyone through smoke or fumes, or that a rapid decompression, perhaps because of a breach in a window, led to their deaths hours before the plane ran out of fuel and spiraled downward.”
Malaysian Transport Minister Datuk Seri Liow Tiong Lai told reporters in August that there was no evidence that pilot Zaharie Ahmad Shah had attempted to commit suicide, saying, “The simulator was used by the pilot for trial and error in many areas. There are thousands of simulations to many destinations.” Absent some critical new findings, we may never be able to answer this question — but the ATSB’s data definitely suggests that the pilot wasn’t in control of the aircraft when it ran out of fuel and slammed into the ocean.
The various countries involved in the ongoing search for MH 370 are meeting to discuss the ongoing search, whether to extend the initial search area, and how to do so (if at all). Earlier this year, it looked as if the search would end once the initial search grid was completely scanned. But if new analysis can identify a compelling alternative area, it’s possible the search could be extended. While we covered a Canadian expert who claimed the wing flap appeared to have been deployed, the ATSB has published an extensive report on why they don’t feel this is the most likely explanation.
Both the flaperon recovered in mid-2015 and the right outboard flap recovered in June 2016 are from the right wing and would have sat adjacent to each other, as shown above. By analyzing both fragments, the ATSB believes the damage is best explained if the flaps were closed at the moment of impact. It’s much less clear how the wing structures would have been damaged in the same way had the flaps been deployed. The debris analysis and flaperon discussions are in two separate PDFs, located here and here.