Analysis: MH370 Preliminary Report

Steve Winter provides an initial analysis of the documents released by the Malaysian authorities on 1 May, including the ICAO-specified Preliminary Report.
MH370 Radar and Satellite Tracks 20140501In addition to the report, a number of other documents were released, including some maps of the estimated aircraft flight paths, and a chronology of “Actions Taken” by the air traffic control authorities after the disappearance of the aircraft from secondary MH370 Satellite Tracks 20140501radar.
As expected, the preliminary report itself is largely unremarkable. It provides a few more specifics on the times of key events, such as the time at which the aircraft disappeared from the Kuala Lumpur Air Traffic Control Centre (KLATCC) screens.
It does finally confirm that the aircraft disappeared after a handoff had been initiated to the Ho Chi Minh ATCC (HCMATCC) in Vietnam, and that the last ACARS contact (at 01:07 MYT, all times in this article will be local Malaysian times) was prior to the loss of secondary radar contact (at 01:21 MYT).
However, the report is very vague about that times that events were determined: for instance, it is noted that the aircraft was observed over the IGARI waypoint, but it is unclear whether the determination that the aircraft disappeared from the radar screen at 01:21:13 was detected at the time, or by subsequent analysis.
The “Actions Taken” document helps clarify some of these uncertainties, but not all of them. It would appear, however, that KLATCC was unaware that there was a problem with the aircraft until HCMATCC queried them some 17 minutes later.
There is little mention of the Royal Malaysian Air Force’s (RMAF) capabilities: only that a radar playback indicated a possible aircraft turning back over the Malaysian Peninsula. Since it is not mentioned in the report, it has to be concluded that no attempt was made to coordinate with the RMAF at the time of the aircraft’s disappearance, which may well turn out to be a critical factor in the loss of the aircraft. There is no mention of the reports of aircraft altitude from the air defence radars.
The report does confirm that seven satellite “pings” were subsequently received and that “the investigation continues” into this.
The report concludes with a comment about commercial aircraft operations “over remote areas”, and a vague recommendation to ICAO about “introducing a standard for real time tracking of commercial air transport aircraft.”
As has been noted elsewhere, the implied comparison with the loss of AF447 is misleading: MH370 was under radar and ADS-B (which the report fails to mention) coverage at the time of its disappearance and this position is known very accurately. A better recommendation would be to ask whether there should be improved procedures for maintaining contact with aircraft when handing them over from one ATCC to another.
The released maps provide more information about the radar and satellite tracks than have been previously produced and are worth remarking upon.

  • The first satellite “ping” occurred only five minutes after the last air defence radar contact at 02:22 MYT, so the estimated satellite position is quite tightly constrained.
  • However, the displayed initial estimated point seems somewhat inaccurate, as it implies that the aircraft performed a turn to the NW immediately after leaving radar coverage. On the basis of the presented information, it seems much more likely that the aircraft would have continued on its heading and that the initial satellite point should be located further south.
  • This would potentially have the effect of skewing the entire estimate satellite trajectory (and thus any impact point) by a number of miles.
  • The maps also show the probability areas for the aircraft impact, corresponding with the final “ping” at 08:19 MYT. The areas are shown as the “Highest Probability Area” to the north, centred on the Zenith Plateau which was the focus of the initial underwater search; the “Lowest Probability Area” to the south of that, and finally, the “Mid Probability Area” to the south of that. This seems curious, as a simple probability distribution would be expected to have the “Mid Probability Area” adjacent to the “Highest Probability Area”. Perhaps this is a typo, but it also makes one wonder why there is no probability area to the north of the Highest Probability Area.
  • It is also worth noting that the extrapolated trajectories would reach the Australian coast, possibly even Perth, but there have no published reports of trajectory analysis on the basis of likely waypoints.
  • This all indicates that further refinement of the satellite data would be useful and that the final resting place of MH370 may well be within a much wider search area.

The document labelled, “Actions Taken Between 01:38 and 06:14 on Saturday 8 March”, is much more revealing than the preliminary report itself. It provides significant previously-unpublished details about the ATC actions taken following the initial disappearance of the aircraft.
In particular, the document shows the significant confusion and delays after HCMATCC queried KLATCC about the whereabouts of MH370, at 01:38 MYT.
Specific points of importance in the document include:

  • Confirmation that HCM did have initial radar contact with the aircraft (which was lost at BITOD), but had no voice contact.
  • Confirmation that despite multiple attempts to contact the aircraft from other aircraft no response was received (there had been earlier reports of a garbled response to another aircraft). It took more than 35 minutes from the loss of secondary radar contact before HCM reported “officially no contact” (at 01:57 MYT).
  • There was considerable delay and confusion about whether the aircraft had entered Cambodian airspace, including a statement from the Malaysian Airline OPS that they were in contact with the aircraft and that it was in Cambodian airspace. This confusion appears to have lasted for about 30 minutes before confirmation was received from Cambodia that they were not aware of the aircraft.
  • Confusion as to whether the airline was in touch with the aircraft (presumably via ACARS) persisted for another 70 minutes until the airline OPS finally stated that they only had a projected position for the aircraft.
  • KLATCC continued to contact other ATCCs for information on the aircraft for more than five hours after the disappearance.
  • Finally, nearly six hours after the disappearance, KLATCC queried HCM if SAR had been activated. It is apparent from this request that KLATCC was unaware of the possibility that the aircraft had turned back and had been tracked by RMAF air defense radar (there is also no evidence that KLATCC attempted to contact the RMAF).

In summary, there are no great revelations about the disappearance of the aircraft in the preliminary report. The most useful information is in the additional documents, which confirm some details about events at the time.
However, the most telling statements are those that reveal the considerable confusion and delay that occurred in the minutes and hours after the initial disappearance of MH370: time which we may well eventually conclude, if used more effectively, could have led to a significantly different outcome for the aircraft and those onboard. In addition, the satellite analysis indicates a significantly larger possible impact area, together with a statement that the analysis is continuing to be refined.
Steve Winter is an independent ATM consultant and an Engineering Fellow with Raytheon Company. He was formerly Chief Technologist for NATS.
Read More: AF447 crash probe urges tracking mandate A ICAO High Level Safety Conference identified issues surrounding coordination between Atlantico and Dakar Flight Information Region (FIRs) as contributing to the excessive delay in alerting the search and rescue (SAR) services to the disappearance of AF447.


  1. Dear Sirs, Thank you for producing the report amidst all the stress involved.
    I would like add a piece of speculation as a non-flying scientist, just from the point of view of logics.
    Several people on dicussion fora have wondered why a potential malicious flying person would have chosen just that complicated route, among all possibilities.
    But the reasoning is turned round if we assume that ANY fuzz with the aeroplane, even for some days, might have already fulfilled the persons’ needs, if it is combined with sufficient destruction of ecvidence about who did it. Maybe they did not care where it will land.
    Motivation, in that case, might be e.g. just distracting a part of the global news coverage from something else which was planned to happen at the same time on the globe elsewhere.
    Thorough background checks for some passengers with currently war-bound nationalities – or their claimed nationalities – have not been published, as yet.

  2. Steve, very coherent anf thoughtful analysis. Might I suggest you take a look at the conversations at

  3. Don –
    Yes, I have been following the discussion on Duncan Steel’s web-site regularly (though not right before writing this article). I think this highlights an issue regarding the “satellite ping” investigation: by refusing to release this information, the investigators are missing an opportunity to engage the tremendous resources of a wider community. This lack of transparency extends to the question of what type of peer review is the analysis being subject to, to ensure its correctness. I can only guess that this opacity is for proprietary reasons. However, given the technical complexity of the analysis, it seems there is plenty of opportunity for error.
    I do think the media should push for release of the analysis so that we can be confident of its correctness.

  4. Really amazing that the official investigators refuse to take advantage of the considerable scientific resourses available. The refusal to disclose the raw data is a crime.

  5. Very well said. Thank you for very coherently, analyzing the questionable details in the official reports.
    I would like to add an eyebrow raising issue about the way the satellite pings were analysed:
    The simplest way to calibrate the offset would be- to operate a SATCOM device on the ground for 7 hours and ping at the same measured pings instances, starting at the initial vector of the satellite in its wobbling course. Then, subtract them from the measured pings frequencies. This would not only contain the effect of wobbling, but also other noises.
    As one of the first action done in investigations of accidents, is real simulation. It’s strange that non such proposed measurement, I just described, was done, or reported.
    (Since we talk calibration offsets, they have nothing to do with the movement of the plane, Their value would be the component of the calibration offset measured in KL, in respect to plain of wobbling of the satellite, every new predicted ping location).
    Another issue which does not concern the pings, but smells a lot of hide-up, that need attention.
    Your comment or provision of links concerning the issue will be appreciated:
    The issue is missing reports, or, discarding reports concerning mobile phone communication.
    1. There should be quite a significant number of people who would conduct telephone conversations before take-off (in every flight).
    2. According to some internet search, about 30% of passengers would not turn their phones off during the whole flight, crew inclusive (intentionally).
    3. Normally, when people disappear, among the initial pieces of information coming from passengers would be their last contact with their lost relative.
    4. Operators provide local service for local subscribers, and roaming services for international passengers. That means, operators, in a few countries, would be able to provide logs of actual calls as well as connectivity (handshakes), ie reception.
    Knowing that there were neither calls nor reception, may tell us, there was something wrong, already from the time of boarding and the whole investigation could take another route.

Comments are closed.