MH370 and Other Investigations

Civilian radar data after transponder was disabled. (Click to enlarge.)

We know that MH370 was captured by civilian and military radar sites before and after the transponder was disabled at 17:21 UTC. However, Malaysian authorities have chosen to release these data sets only as low resolution images that have imprecise position information with few timestamps. The DSTG did choose to publish in its Bayesian analysis the speed and track data that was derived from radar data that was provided to them by Malaysia. However, the DSTG presented the speed and track data after applying a Kalman filter to remove noise. It’s unknown whether that Kalman filter produced artefacts in the graphical presentation of that data.

We now have what we believe are the data sets for the primary surveillance radar (PSR) and secondary surveillance radar (SSR) from Malaysian civilian radar assets. The PSR data is of particular significance because it provides additional insight about how MH370 was flown after the transponder was disabled at 17:21 UTC. The data was publicly released by fellow IG member Mike Exner. The military radar data remains unavailable.

The data begins at 17:30:33 when the civilian radar installation at Kota Bharu Airport (WMKC) detected MH370 traveling back towards the Malay peninsula about 58 NM from shore. The last radar target was captured by the civilian radar installation at Butterworth Airfield (WMKB) after MH370 had passed to the south of Penang Island and was tracking northwest up the Malacca Strait towards Pulau Perak.

Some initial observations about the data:

1. The PSR data is similar to the civilian radar data that was graphically presented in the Factual Information (FI) from March 2015. However, while the last civilian radar capture in the FI was at 17:51:47, the new data set has captures until 18:00:51.
2. The path derived from the Kota Bharu radar data is not straight. More analysis is required to determine if this waviness indicates that there were pilot inputs from manual flying, pilot inputs to the selected heading with autopilot engaged, or inaccuracies of the radar data.
3. The path was tangent to a 5 NM radius for both Kota Bharu and Penang Airports. This may indicate that these airports might have been displayed as fixes in the navigational display (ND) with a radius of 5 NM and used as navigational references.
4. After passing to the south of Penang Island, the plane first tracked towards 301ºT, and then changed to 291ºT, which aligned with Pulau Perak and roughly towards VAMPI.
5. The groundspeed data as derived from the radar data is noisy, reflecting uncertainty in the value of the timestamp as well as the range and azimuth for each capture. In light of the uncertainty, the average speed was calculated for five of the six segments of radar captures, and shown by the red line in the figure below. (The time interval of the shortest segment was only 24 s, and deemed to short to calculate the speed with a useful level of precision.) The average speed for the second and third segments are 527 knots and 532 knots, respectively, which suggests the plane was flying close to Mmo=0.87. For instance, with a tailwind of 12 knots and a temperature offset of ISA+10.3K, a groundspeed of 527 knots converts to M0.87. At the Mmo/Vmo crossover altitude of 30,500 ft, a groundspeed of 532 knots converts to M0.86. This suggests that after the aircraft flew past Kota Bharu, it was at the upper end of its operating speed range, and possibly at times beyond it.

Calculated groundspeed as derived from the civilian radar data. (Click to enlarge.)

I know that independent investigators that contribute here and elsewhere will continue to analyze the data to better understand how MH370 was flown before it completely disappeared from all radar sites.

Update on April 12, 2018: The plot of groundspeed was updated by removing the trend lines and replacing them with average speeds over segments. In light of the noise on the speed calculations, this is more appropriate. The estimated peak groundspeed reduced from 545 knots to 532 knots. The corresponding text in (5) was also updated to reflect this change.

Update 2 on April 12, 2018: Here is an Excel file for those wishing to see the basis for my calculations. Please let me know if corrections are required.

Ocean Infinity’s search progress, from Richard Cole.

Recent Activity

Seabed Constructor, the vessel operated by Ocean Infinity to scan the seabed in search of MH370, is returning to port in Fremantle, Western Australia, to refuel, change crews, and resupply. Constructor is completing the second of three or four swings, each swing lasting about six weeks. So far, there have been no promising sonar “contacts” that might represent the debris field of the missing aircraft.

There remains about 3,000 sq km of seabed to search in the area that the ATSB and CSIRO designated as a priority. After that, the extended search area along the 7th arc would require scanning about 46,600 sq km to reach north to around 29S latitude if the width of the search was 25 NM on either side of the 7th arc. That will require more than one additional swing to complete. In fact, it could prove challenging to complete with even two additional swings, depending on the weather and how well the eight autonomous underwater vehicles (AUVs) perform. Although not publicly stated, there are indications are that at least one of the AUVs is having technical problems.

What We Know So Far

Any scenario that leads to a particular location (a “warm spot”) is based on a set of assumptions, and the failure to find the debris field in proximity to this location means that one or more of those assumptions are false. So we can be fairly certain that the large, blurry objects seen in the French satellite images were not from MH370, as the corresponding impact locations calculated by CSIRO were searched without success. Also searched was the warm spot that was calculated by assuming that the aircraft flew until fuel exhaustion on a path towards the South Pole. Unless an interesting contact was found but not yet disclosed, this scenario also can be dismissed.

In the coming weeks, other scenarios will be searched, including the impact site near 30S latitude that is based on two floating debris fields that were spotted during the aerial surveillance, and discussed at length in a previous post.

Reasons Why the Debris Field Has Not Yet Been Found

Although the area already scanned by Seabed Constructor was designated the highest probability by the ATSB and CSIRO, there are reasons why endpoints outside of this area are still possible.

• A descent at 18:40 followed by a holding pattern, excursion, or other “loiter” before the turn to the south could mean the plane impacted along the 7th arc to the north of the priority area that has been searched. The last radar target was captured at 18:22, and after 18:28, the next ping arc derived from the BTO data is known at 19:41. There is simply no way to be sure of the path of the plane during this interval.
• A shift in oscillator frequency of the satellite data unit (SDU) of the SATCOM, which would change the value of the fixed frequency bias (FFB) that is used to convert the location and speed data into a BFO that can be compared with the measured BFO values. In a nutshell, if the FFB shifted by +7 Hz after the power up at 18:25, endpoints as far north as 27S are allowed by the BTO and BFO data. It turns out there is an effect called “retrace” that causes oscillators that are powered down, cooled, and powered up to shift in frequency, and there are indications that a retrace shift of about -4 Hz occurred while 9M-MRO was on the ground at KLIA before the MH370 flight. A similar shift, but in the opposite direction (up) might have occurred due to the inflight power cycling.
• Pilot inputs after 19:41 might have altered the path. The continuous, smooth progression of the BTO and BFO data suggests automated flight with few or no pilot inputs until fuel exhaustion. However, there is a remote possibility that the smooth progression of values was produced by a more complicated path that by chance replicated the simplest of paths.
• There is also the possibility that the previous search was as the correct latitude along the 7th arc, but the width of +/- 25 NM from the 7th arc was not sufficient. The final two BFO values indicate a steep, increasing descent that if continued would mean the plane impacted close to the 7th arc. The debris is also consistent with a high-energy impact. However, it is possible, albeit unlikely, that a skilled pilot carefully recovered from the high-speed descent, regained altitude, and glided for some distance beyond 25 NM.
• Although some of the area north of the priority search area was searched by aerial surveillance in the weeks following the disappearance, the search area was large and the coverage was spread thin. Also, some debris was seen from air, but never recovered due to the distance of ships supporting the search effort.

Simulation of Seabed Constructor’s Search Pattern

Finally, Richard Cole, who has carefully been tracking and analyzing the search patterns of Seabed Constructor, has produced a short video which shows the path of the vessel and how it relates to the launch and recovery of the AUVs. Richard is quite talented at extracting a lot of information from small amounts of data, and this video, like all his work, is commendable.

Status of current underwater search. (Click on image to enlarge.)

Recent Activity

After a short stop in Fremantle for to re-fuel, re-supply, and change crew, Seabed Constructor, operated by Ocean Infinity, is back searching for the wreckage of MH370. Ocean Infinity is under contract with Malaysia to use its team of eight autonomous underwater vehicles (AUVs) to scan the seabed in the Southern Indian Ocean (SIO) in search of wreckage from the aircraft. Under the terms of the contract, Ocean Infinity will only be paid if the wreckage is found. The search is occurring in multiple six-week long “swings”, of which the first swing has been completed, and the second swing is just beginning. Subsequent swings will also require a stop in Fremantle for servicing.

For the first swing, Ocean Infinity began by searching the 25,000 sq km of seabed that ATSB and CSIRO have designated as the priority area (shown in white in the figure). So far, Constructor has scanned about 7,500 sq km of seabed, including 5,000 sq km within the priority area that was designated by CSIRO as the “primary area” (solid white). There remains about 20,000 sq km of the priority area that is unscanned (translucent white). Beyond that is the extended search area, which reaches to about 29S latitude along the 7th arc (translucent green), and is expected to be searched at a width of +/- 25 NM from the 7th arc. Under ideal conditions, the eight AUVs are capable of scanning about 1,200 sq km of seabed each day. Recognizing the possibility of weather and operational constraints,  a more realistic expectation might be about 25,000 sq km per swing. However, until Ocean Infinity gains more operational experience, it is difficult to predict what scan rates are realistically achievable.

With the sparse and imprecise evidence we have, it is impossible to assign a high level of certainty to any impact site, as the satellite data and the drift models allow a broad range of possibilities. So, it becomes a numbers game–the more area searched, the higher probability of finding the wreckage. However, within that broad range, there are some “warm spots” that are based on assumptions about navigation inputs and other evidence.

What We Know So Far

In the previous post, I estimated the probability of finding the wreckage as 67%, assuming all of the priority and extended areas are scanned. (This probability will vary some depending on how far north the search reaches.) Considering that only 5,000 sq km of that area were scanned in the first swing, and assuming that there are equal probabilities within that total area, the probability of finding the debris field within the primary area would be about 4.4%. Considering this low percentage, it should come as no surprise that the wreckage has not yet been found, and we are far from the point of re-thinking the search strategy.

Within the area searched so far, there are three warm spots that CSIRO has designated as priorities, based on satellite images of objects that could have been MH370 debris, and from drift models that estimated the points of impact from the location of these objects. Last August, the highest priority location (CSIRO Priority 1) was described by CSIRO’s David Griffin in these words: We think it is possible to identify a most likely location of the aircraft, with unprecedented precision and certainty. Unfortunately, all three of these locations have now been scanned with negative results. Unless positive news is being withheld, the confidence expressed by CSIRO was unfounded. This is not a total surprise: The objects captured by the satellite images had too much surface area to likely be from MH370, and the location of the potential impact sites were  not consistent with the high speed descent suggested by the final BFO values.

Two other warm spots have been at least partially searched in the first swing. The first is an impact location near 34.7S latitude that Inmarsat derived by minimizing the BFO error. More recently, Bobby Ulich proposed a location near 34.8S latitude that was based on a path of constant true heading (CTH). We should know soon whether or not these warm spots are completely eliminated.

Another warm spot that should be searched during the current swing is based on a great circle path between waypoints BEDAX and the South Pole. I first proposed this path in August 2014, and I still consider it to be among the best possibilities because of the excellent fit of the BTO and BFO data, and because of the simplicity of navigating in the direction of true south. That said, despite the attractiveness of this scenario, we don’t know whether the aircraft was navigated in this manner, so it remains one of many other possibilities.

As shown in the figure above, there are warm spots that reach as far north as 27S latitude that are based on certain navigational inputs. Although the match to the BFO data is not as good for paths ending that far north, the BFO error is still well within what was recorded for previous flights of the 9M-MRO airframe. The drift models also favor an impact point further south than 27S. However, for debris discovered on the beaches of Eastern Africa, there could have been a considerable delay between the time of discovery and the time of arrival near the shore, and this uncertainty reduces the accuracy of the drift models.

In a nutshell, although the previous search swing has eliminated some possibilities, we are still very early in the search process, and it is much too early to draw any conclusions.

Unknown Activities of Seabed Constructor

The figure below from Richard Cole shows the recent behavior of Seabed Constructor. At the end of the last swing, Constructor returned to the outer leg of the primary search area, which had been previously scanned. After following the pattern of a 5-km circle, it retraced what we believe was part of a previous path of an AUV, and then disabled its AIS data, which made it impossible to remotely track. When the AIS was eventually re-enabled three days later, Constructor had left the search area, and was traveling back to Fremantle. What activities occurred during these three days is not known.

Seabed Constructor’s path, as adapted from the work of Richard Cole. (Click on image to enlarge.)

At the start of the search for the current swing, Constructor again returned to the southern end of the outer leg of the primary search area, and seems to be actively searching the seabed in this location. The activities in the current area are likely related to activities that occurred when the AIS was disabled during the last swing.

Some possibilities that have been proposed by others to explain the behavior are:

• Constructor is re-scanning areas that had poor quality or missing data either because of malfunctioning equipment or challenging terrain
• There are one or more promising points of interest that are under being comprehensively investigated
• A search is underway to locate equipment that was lost in the previous swing
• Some combination of the previous possibilities

Whatever the reason for the unexplained behavior, it is noteworthy that there was no reference to the behavior in either of the last two weekly updates from Malaysia. As Malaysia has two observers on Seabed Constructor, Malaysia is certainly aware of the surrounding circumstances. Malaysia’s decision to omit pertinent information in the weekly reports further erodes the public’s confidence in the Malaysian-led investigation. Credibility is not possible without transparency.

What many of us have been encouraging has finally transpired–the seabed search for the wreckage of MH370 has been re-started. The search vessel Seabed Constructor has just arrived in the new search area, outfitted with a team of eight autonomous underwater vehicles (AUVs). Ocean Infinity, the company under contract with Malaysia to conduct the search, has agreed to start by searching the 25,000 square kilometers identified by the ATSB and CSIRO as most likely. Included in that area are three locations that CSIRO has identified as high priority, as determined from satellite images of floating objects and complex drift models. Last August, the highest priority location was described by CSIRO’s David Griffin in these words: We think it is possible to identify a most likely location of the aircraft, with unprecedented precision and certainty. This location is 35.6[degrees south], 92.8 [degrees east]. 

At Ocean Infinity’s touted scan rate of 1,200 square kilometers per day, the entire 25,000 square kilometers would be completed in 21 days of searching, and the highest priority area of 5,000 square kilometers would be completed in less than a week.

The nominal location of the 7th arc that is shown in the figure above is a based on the assumption that the last transmission from the aircraft occurred at 20,000 ft, and our best estimate of the final BTO value is 18390 μs. The final two BTO values that were used for the best estimate occurred when the SATCOM of MH370 initiated a log-on to Inmarsat’s satellite network at 00:19 UTC on March 8, 2014, minutes after the engines stopped due to fuel exhaustion. (The re-boot of the SATCOM likely occurred after the APU automatically started and briefly supplied electrical power.)

I performed a statistical analysis of previous log-on events that occurred on March 7, 2014, including those that occurred on flight MH371 from Beijing to Kuala Lumpur. Using the results of this analysis, the final two BTO values from MH370 were first corrected and then appropriately weighted based on their respective uncertainties in order to arrive at the best estimate of 18390 μs.  The procedure was briefly described in a previous comment of mine.

Also shown in the figure are two other arcs that are positioned at +/- 25 NM from the nominal location of the 7th arc. These might serve as limits for some parts of the search. The figure shows that the +/-25 NM limits do not correspond to the boundaries of the 25,000 square kilometer area that was previously identified. In fact, the highest priority location identified by CSIRO (labeled CSIRO Priority 1) falls slightly outside of the 25-NM outer limit.

If not found in the initial 25,000 square kilometer area, the contract with Ocean Infinity indicates that the search will continue further northeast along the 7th arc. Likely, the search will continue along the 7th arc as far northeast as time and weather permit.

I often get asked whether I believe this search will succeed in finding the wreckage of MH370. I long ago arrived at the conclusion that based on the evidence we have, it is impossible to determine any one location with a high level of certainty, and I stopped trying. The satellite data and the drift models allow a broad range of possible impact sites. Within that range, there are at best some “warm spots” that are based on assumptions about navigation inputs. So, it becomes a numbers game–the more area searched, the higher probability of finding the wreckage. I subjectively believe there is a 33% chance of finding the wreckage in the first 25,000 square kilometers. If there is time and money to search at +/- 25 NM from the 7th arc all the way to a latitude of 26S, I subjectively put the chances of success at around 67%. That might seem like bad odds, but realistically, that’s higher than they’ve ever been.

The highest priority location identified by CSIRO is about 66 NM from Seabed Constructor’s present location, and might be reached within the next day. We’ll all be watching.

[Don Thompson reminds me that the data from an AUV mission is available only after the AUV is recovered after the completion of a dive, which could last 2+ days, based on the endurance of the batteries. It might take another 18 hours to analyze the data. That means that although the AUVs could reach “CSIRO Priority 1” by tomorrow, we would not know until Wednesday or Thursday whether or not the debris field was found.]

A new article by Hal Hodson on the search for MH370 was today published in the The Economist, and gives more details surrounding Ocean Infinity and its exploration technology. The article discloses that:

• Host vessel Seabed Constructor, owned by Swire and under lease by Ocean Infinity (OI), has been fitted with eight underwater autonomous vehicles (AUVs) for the search.
• The search will be conducted under the basis of “no find, no fee”, which means that OI will bear the economic cost of not finding the wreckage.
• Even though the contract with Malaysia has not yet been signed, Ocean Infinity will proceed with the search in order to take advantage of the favorable weather in the Southern Indian Ocean in January and February.
• The expected scan rate that is achievable using eight AUVs is 1200 sq km per day.
• Some additional testing of the scanning capability of the AUVs will be performed en route between the imminent departure from Durban, South Africa, and the arrival to the search area.
• The scanning will begin in the area designated by the ATSB as most likely (the 25,000 sq km) around 35S. If unsuccessful, the search will proceed towards 30S latitude.
• The advice to proceed north towards 30S latitude came from independent experts. (Readers here might be able to guess the names of the independent experts that have advised OI.)
• Rather than communicating with the autonomous surface vehicles (ASVs), the AUVs will communicate with the host vessel to periodically recalibrate the onboard inertial guidance system.
• If the flight data recorder (FDR) is found, it will be recovered and surrendered to the Australian authorities.
• Recovery of wreckage would require a separate agreement with the Malaysian authorities.

For readers of this blog, there are few new facts presented. Probably the most significant new fact is OI’s decision to start the search without a signed agreement.  The article is helpful in that it will provide useful information to a larger, broader audience, and will increase the overall awareness of the new search. There will also be renewed questions as to why Malaysia has delayed signing the agreement with OI.

Update on Jan 3, 2018: Malaysian Transport Minister Liow was asked about recent developments regarding Ocean Infinity and the renewed search for MH370. He replied that the parties were in final negotiations, and there would be an announcement next week. The fact that he offered no stipulations for reaching an agreement, which has been the pattern in the past when Malaysia has wanted to stall the negotiation, is very encouraging.

JUST IN: Malaysia's Transport Minister @liowtionglai says he will be announcing details of negotiations for US company Ocean Infinity to resume the search of #MH370 next week. This after reports of Ocean Infinity sending a search vessel to Australian waters. pic.twitter.com/jBDKBTV0xh

— Sumisha Naidu (@SumishaCNA) January 3, 2018

Update on Jan 5, 2018. Channel News Asia is reporting that Malaysia has accepted Ocean Infinity’s offer to continue the search on a “no cure, no fee” basis. The information was sent to the families of passengers on in an email. (Malaysia in the past has informed the next-of-kin of new developments before releasing details to the public.)

Update on Jan 10, 2018. As widely reported, the agreement between Ocean Infinity has been finalized in a signing ceremony. The tiered payment terms are linked to where the debris field is found, and ranges from $20 million if found in the highest priority, 5,000 sq km area, to $70 million if found beyond the 25,000 sq km area. Here is the complete statement from Minister of Transport Liow:

Figure 1. Location of possible debris from MH370 (red) detected during the surface search (green) on March 29, 2014. Possible flight paths aligned with waypoints also shown (black). (Click on image to enlarge.)

Introduction

Australia’s CSIRO recommends that the next subsea search for MH370 be carried out in a 25,000 sq km area that extends along the 7th arc as far north as 32.6S latitude. The recommendation is based on extensive drift analyses as well as satellite imagery which may have captured debris from the impact.

Despite the work of CSIRO, there are reasons to believe that the wreckage from MH370 is along the 7th arc, but at latitudes to the north of 32.6S latitude. Some of these reasons are:

1. The absence of debris discovered along the shores of Western Australia is better explained by an impact to the north of 32.6S latitude.
2. The timing and location of the fragment of the engine cowling with the letters “Roy” that was discovered in Mossel Bay, South Africa, in December 2015, is better explained for an impact north of 32.6S latitude.
3. The particular impact points that CSIRO considers to be most likely are located at latitudes along the 7th arc that have already been searched to about 19 NM from the arc. With the increasing descent rates indicated by the final two BFO values, the impact likely occurred within that distance from the arc.

On the other hand, CSIRO has provided reasons why they believe the impact could NOT be north of 32.6S latitude. These are:

1. For latitudes north of 32.6S, debris should have arrived in Madagascar, Mozambique, and other locations in Eastern Africa, before the discovery of debris was reported. But this could easily be explained by the delay between when debris arrived near a location, and when it is discovered and reported to authorities. In fact, the circumstances surrounding Blaine Gibson’s discovery of the horizontal stabilizer fragment with the letters “No Step” suggests the arrival of that part in Mozambique well before it was discovered.
2. Reconstructed paths terminating along the 7th arc to the north of 32.6S latitude are less consistent with the satellite data. Yet, if we allow for pilot input between the last radar capture at 18:22 and the second handshake at 19:41, and if we assume automated flight after 19:41, there are reconstructed paths that match the satellite signaling data within allowable error limits. In fact, there is a range of possible paths that terminate to the north of 32.6S, and these were the focus of my last post. Within this range, there are some waypoint-derived paths that might deserve extra attention.
3. If an impact had occurred north of 32.6S, the floating debris would have been detected by the surface search that was conducted by aerial surveillance in the weeks following the crash.

In this article, we take a closer look at the last item in the list.

Many believe that CSIRO’s estimate of search efficiency is over-stated, and it is possible that debris was missed during the surface search. Here, we consider the possibility that floating objects from MH370 were in fact detected by the aerial search and photographed, but those images were ignored after path reconstruction models placed the probable impact site further north, and then further south along the 7th arc. The images captured by the surface search demand a new level of attention in light of the prospect for re-starting the subsea search along the 7th arc in areas that will likely include latitudes to the north of 32.6S.

Obtaining the Images

Large areas of the ocean were surveilled by the Royal New Zealand Air Force (RNZAF) using P-3 Orion aircraft as part of the overall surface search effort. One image from March 28, 2014, which included an object described as a “blue panel”, received some attention because of the object’s resemblance to a flaperon. This object was discussed in an article by Bernard Lagan, who flew with the New Zealand Orion crew on March 28. (The video that accompanies the article is worth watching.)

Curiosity surrounding this debris prompted IG member Brian Anderson in October 2015 to request more information about this and other images collected by the RNZAF search that could be debris from MH370. His request was fulfilled, and he received a large set of images that were made publicly available on Duncan Steel’s website. For convenience, here is a compressed file (1 GB) of the all the images.

Brian continued to query the RNZAF, and was able to obtain more information about the coordinates of the images. IG member Don Thompson was also able to extract the position and timing from the meta data embedded in some of the images. IG member Richard Godfrey in turn helped assemble the data for the completed set of images and also provided some descriptions of the images. The time and position data that we have for all the images is compiled in this Excel file.

As far as we know, none of the objects identified in the surface search on March 29, 2014, were recovered by ship, so the relationship to MH370 remains unknown.

What the Images Show

The images taken on March 29, 2014, captured a variety of objects, some of which could be floating debris from the impact of MH370. The surface search also found objects that are without a doubt not from MH370, including fishing articles such as nets, floats, and marker buoys. Some of the more interesting images from March 29 are shown in Figures 2 – 5,  including two debris fields of small fragments, an object that resembles a suitcase, one that resembles a portion of a panel with wires, and a rectangular box that resembled a cargo package.

Figure 2. Enlarged images of debris fields on March 29, 2014. Left: 28.3927S, 97.7750E from image 5832. Right: 29.19333S, 95.1250E from image 5853.

Figure 3. Object resembling a suitcase near 28.3176, 97.8234E on March 29, 2014. (Enlarged from image 5833.)

Figure 4. Object resembling a portion of a panel with wires near 28.8867S, 96.0844E on March 29, 2014. (Enlarged from image 5847.)

Figure 5. Object resembling a rectangular cargo package near 29.1936S, 95.1094E on March 29, 2014. (Enlarged from image 5854.)

Drift Analysis

As only 21 days elapsed between the impact of MH370 and the discovery of the debris, it should be possible to “backtrack” the objects to March 8 to determine a potential point of impact. To simplify this procedure, I looked at the forward drift results for a point of impact on the 7th arc at 29.7S latitude, which corresponds to the path aligned with Wilkins Runway (YWKS), Antarctica, and is also the closest path to the objects among the four waypoint-derived great circle paths that I examined in the last article.

Using the CSIRO results for drift of low windage debris from 30S latitude, I selected 31 particles that were positioned within a 15-NM radius of the assumed impact point at 29.7S latitude, among the hundreds of particles that CSIRO studied over a much larger area. The selection of multiple particles within a circular area allows for a diversity of drift paths which reflect the uncertainty in the impact point and the stochastic nature of ocean drift. Ideally, more than 31 particles would be modeled as we know the crash produced many more, but the geographic spread of the injected particles in the available CSIRO data sets limited the number of particles I could select and get meaningful results about a particular impact point. The selected particles (green) among the total particles available (translucent blue) are shown in Figure 6.

Figure 6. Selected particles (green) within a 15-NM radius (white) of the 7th arc crossing (intersection of black and blue lines) at 29.7S latitude. (Click on image to enlarge.)

The drift of the particles is shown in the following video for the time period between March 8 and March 29, 2014. The direction of the particles is from the impact at 29.7S latitude (white circle) and towards the detected debris (red dots). The video also shows how the paths of the particles diverge so that after 3 weeks, the particles are separated by more than the 150-NM distance between the two detected debris fields (images 5832 and 5853). The drift results indicate that the detected objects are consistent with an impact on March 8, 2014, along the 7th arc at 29.7S latitude. (The video is best seen in “full screen” mode by selecting the four arrows next to the Vimeo logo.)

Implications for New Search

Unless there is evidence that objects detected on March 29, 2014, are not from MH370, we have to consider the possibility that one or more objects are from MH370. This greatly increases the probability that the impact was further north than the 25,000 sq km area recommended by CSIRO for the new seabed search. Ocean Infinity seems committed to searching beyond the first 25,000 sq km, and based on the results presented here, a hot spot around 30S latitude deserves special attention.

I have asked the ATSB and CSIRO for more information about these “contacts”. The ATSB had record of the detection of some of the objects on March 29, 2014, but no record of the detection of the two debris fields. Meanwhile, David Griffin of CSIRO has agreed to do some “backtrack” drift calculations to see where along the 7th arc the debris would have originated if the debris was from MH370. As these results are very relevant to future search efforts, I will make the results available when I can.

Update on April 5, 2018

On November 29, 2017, I received an email from David Griffin in which agreed that the objects spotted by the RNZAF on March 29, 2014, could have been from an impact on the 7th arc around 29.7S latitude. At that time, I had no permission to share his email. I now have that permission, and what follows is the email in its entirety:

Hi Victor,

First, a few responses to your post (which I thought made some good points):

1. The 25,000km^2 search area was not recommended by CSIRO. It came out of the First Principles Review. CSIRO was just one voice in the room.
2. Roy: “better explained” – well, sort of, see below
3. “could NOT be north of 32.6S” – if you check the reports I think you will find words like “less likely” are what we said.
4. I won’t comment on what the images may or may not be. I leave that to others. CSIRO’s role is to comment on how floating things move.
5. Your use of the kmz files: well done! I think you’ve done much of the job. See below for additional info.
6. I think you have made a good point that the items seen by the RNZAF were NOT all conclusively proven NOT to be from MH370.

Next, two asides:

1. The surface search targeted regions of ocean where items from the 7^th arc may have drifted to. So backtracking any items seen should, by and large, conclude that they were near the 7^th arc on 8 March. So anyone (one of your commenters, I think) who thinks that backtracking to the arc supports the idea that the items were MH370-related has missed the point. You are correct that the only point of doing the backtracking is to identify what area on the arc was the potential crash site.
2. Forward tracking and backward tracking are mirror images when we do not include any random numbers (which is normally the case). The difference is sensitivity to starting point. Assume for the moment that the NZ photos are all bits of plane. If I backtrack them with our (guaranteed-imperfect) model, I fully expect that many would go wide of the 7^th arc. Long story short: its simpler to use forward tracks, as you have done.

Finally, some fresh results:

I have now used both our models (BRAN2015 and BRAN2016, as used in our reports) to investigate the 29.7S crash site scenario. This is a “poor man’s ensemble” (just 2 members) but I think it really helps people not to over-interpret results when confronted with two models. Confucious said: “man with 2 watches does not know the time”. I say “man with just one watch thinks he knows the time exactly when he might not”.

The flow field

Trajectories from a line of points on the 7^th arc (~30.5S to 29.5S on 8 March) up til 29 March:

http://www.marine.csiro.au/~griffin/MH370/br15_MH370_93103_tp3l1p2d_bh_arc7_305295_0/20140329.html

http://www.marine.csiro.au/~griffin/MH370/br16_MH370_93103_tp3l1p2d_bh_arc7_305295_0/20140329.html

It is around 19 March (click PREV to go back to then) that the line of points starts to stretch apart. You can see that this is because of the fanning-out of the sea surface height contours. 28.5S 96.5S is a saddle point in the sea level. Images 5847 and 5846 were near this saddle point. Neither model is keen to make trajectories go right there (corollary: backtracks will not come back). Wind and Stokes drift are what make it possible (in both reality and the modelling). Images west or east are more easily reached, from points on the arc that are not far apart, suggesting that it is indeed plausible (taking model imperfections into account) that all the photographed debris items were from a single origin (near 29.7 or 29.8S) on 8 March.

SST images

I have also looked at the satellite sea surface temperature imagery, and overlain this on modelled trajectories with three levels of windage. The debris will be somewhere between the 1.2% and 3% windage dots, while the deep ocean features will move more like the zero windage dots.

The images for 28 Mar

http://www.marine.csiro.au/~griffin/MH370/br15_L3S-1d_MH370_93103_tpb_arc7_305295_0/2014032812.html

and 30 Mar are clearer than 29^th, and show (like others before) that there was a strong temperature front at about 29S, very close to the image locations. Currents tend to flow mostly along temperature fronts, not across them. This happens in the model as well as in the real world to a pretty good degree. But fronts are also often slightly convergent at the surface (water sinking) which is why buoyant material accumulates at fronts. So you can see that this provides a plausible explanation of why there were many debris items here – MH370-related or not.

Roy

I would not claim that BRAN2015 trans-Indian trajectories starting at 29.7S are very consistent with the finding of Roy in Dec 2015. It is only a very small fraction of the trajectories that go down there. Enough to say to it is possible – certainly – but not really a very conclusive result. See:

http://www.marine.csiro.au/~griffin/MH370/br15_MH370_IOCC_tp3l1p2dp_rw2_297986/index.html

Other items

March-April 2015 is when debris from a crash at 29.7S would (according to BRAN2015) have started to wash up on Madagascar, Tanzania and Mozambique. http://www.marine.csiro.au/~griffin/MH370/br15_MH370_IOCC_tp3l1p2dp_rw2_297986/20150312.html

Flaperon

Nov-Dec 2014 is the earliest that the flaperon might have arrived at Reunion if the crash was at 29.7S according to BRAN2015. 29 July 2015 is well after this earliest time but still perfectly plausible:

http://www.marine.csiro.au/~griffin/MH370/br15_MH370_IOCC_tp3l1p2dpf10_20_99_297986/20150729.html

So not particularly helpful.

I talked to Craig Longmuir at AMSA today about this, so have copied him in. He may have something to add about the photos and their interpretation.

So, as with the Pleiades images, these RNZAF sightings, if MH370-related, do potentially lead us to a crash site. But Pleiades and RNZAF are mutually-exclusive. Same with what was sometimes referred to as “the Esky lid”.

David

Figure 1. Great circle paths that match the satellite data after 19:41. (Click on image to enlarge.)

Introduction

In the last post, I presented how the simulation data found on the home computer of Captain Zaharie Shah suggests that the recovered data were from a single flight session on Feb 2, 2014, in which the aircraft takes off from Kuala Lumpur International Airport (KLIA), flies northwest over the Malacca Strait, flies past the Andaman and Nicobar Islands, turns to the south, and exhausts its fuel in the Southern Indian Ocean. The simulated flight could have represented a diversion of flight MH150 from KLIA to Jeddah, Saudi Arabia, which the captain commanded two days later on Feb 4, 2014.

The alignment of data set 10N (the northernmost), 45S1 (just after fuel exhaustion in the SIO), and Pegasus Field (NZPG) at McMurdo Station, Antarctica, suggests that the simulation user might have selected autopilot and used LNAV mode. If LNAV mode is selected, the aircraft follows a great circle path between the starting and ending waypoints of the active leg. This raises the possibility that McMurdo was used as a final waypoint for navigation with the expectation that fuel would be exhausted in the SIO, well before reaching Antarctica.

In this article, I reconstruct flight paths with the assumption that MH370 was flown in automated flight in a similar way as in the simulation session. In particular, I reconstructed flight paths using the following two criteria:

1. After 19:41, MH370 was flown in under autopilot, and was following a great circle path in LNAV mode and speed was controlled using the autothrottle.
2. In addition to (1), after 19:41, MH370 was following a great circle path that leads to an airport in Antarctica.

What this article does not consider is how drift analyses based on the location and timing of recovered debris from MH370 affects the impact probabilities along the 7th arc. A review of the available drift analyses needs to be considered along with any path reconstruction studies before a new search area can be recommended with any level of confidence.

Flight Paths Along Great Circles after 19:41

Figure 1 shows the path (green) of MH370 as captured by Malaysian civil and military radar. After the last radar capture at 18:22, all we have to help us reconstruct possible paths are the Burst Timing Offsets (BTOs) and Burst Frequency Offsets (BFOs) from communication between the aircraft, Inmarsat’s F1 satellite above the Indian Ocean, and Inmarsat’s Ground Earth Station (GES) at Perth. The BTO values indicate how far the aircraft was from the satellite, and therefore possible locations that satisfy the BTO values form an arc when plotted. Shown in Figure 1 are the arcs at 19:41 (the 2nd arc) and 00:19 (the 7th arc). While the BTO values are indicative of position, the BFO values are indicative of the speed (horizontal and vertical) and track of the aircraft. The BFO nvalues tell us, for instance, that MH370 was traveling south after 19:41, and also tell us the aircraft was in an increasingly steep descent at 00:19.

The radar path ends as MH370 was traveling northwest in the Malacca Strait. Yet we know from the satellite data that the flight ended in the SIO, and the satellite data after 19:41 shows a progression of values consistent with automated flight. The details of how MH370 might have been flown between 18:22 and 19:41 is still the subject of much debate, and it is possible that multiple maneuvers occurred in this time period. (I’ll be presenting some thoughts on this in the next article, benefiting from some new insights.) For now, we start our analysis at 19:41 and don’t consider the time period between 18:22 and 19:41, but also recognize that some of the paths presented below may be eliminated by constraints imposed when the time period 18:22 – 19:41 is considered.

The reconstructed paths starting at 19:41 and ending at 00:19 are shown in white in Figure 1 for paths at a constant pressure altitude of 35,000 ft. Wind and temperature data from GDAS were used to relate Mach number to ground speed. Each path corresponds to a specific track angle at 19:41. (The track angle will in general vary along each path as would be expected along a great circle.) Initial track angles between 162°T and 192°T were considered, and these paths cross the 7th arc over a range of latitudes between 22S and 40S. For each initial track angle, the position at 19:41 was found that minimized the RMS error for the BTO at the times 19:41, 20:41, 21:41, 22:41, and 00:11. (The position at 19:41 was not constrained to fall exactly on the 2nd arc.) The corresponding RMS error for the BFO values at times 19:41, 20:41, 21:41, 22:41, 23:14, and 00:11 was also recorded.

For automated control of speed, two autothrottle modes were considered for the time period between 19:41 and 00:11: constant Mach number, and Long Range Cruise (LRC), in which the Mach number decreases as fuel is burned and the weight decreases. For the time period between 19:41 and 00:11, a speed is chosen so that the aircraft exactly crosses the 7th arc at 00:19 .

The LRC speed schedule also serves as a proxy for ECON speed, for which the speed also varies with aircraft weight. ECON offers better cost efficiency than LRC because the relative value of time and fuel can be adjusted through a Cost Index (CI) parameter, and because the Mach number is adjusted for wind. However, I chose to not consider ECON speed here because the speed profile would be similar to LRC (depending on the CI), and the exact methodology is not generally available to the public for calculating Mach number as a function of weight, altitude, Cost Index, and wind. (Bobby Ulich has just published a model that makes excellent progress in this regard.)  If ECON speed had been considered, it is possible that the BTO values for the same of the paths would have marginally improved.

Figure 2 shows the BTO and BFO errors across a range of latitudes for crossing the 7th arc. Some observations are:

Figure 2. Match to satellite data after 19:41 for great circle paths. (Click on image to enlarge.)

• For latitudes south of 34S, the LRC speed is too slow, and constant Mach number results in a better BTO fit.
• For latitudes north of 34S, the speed reduction offered by LRC speed results in a better BTO fit than for constant Mach number.
• For the paths at constant Mach number, the Mach number varies between 0.801 for a crossing at 22S, to 0.842 for a crossing at 40S.
• Constraining the BTO error to less than 32 μs eliminates paths crossing the 7th arc north of around 26.6S.
• Constraining the BFO error to less than 7 Hz eliminates paths crossing the 7th arc south of 39S latitude and north of 28S latitude.
• The minimum BTO error for LRC speed occurs for a crossing of the 7th arc around 33S latitude
• The minimum BFO error occurs for a crossing of the 7th arc of around 35S latitude.

Flight Paths Leading to Waypoints Past 7th Arc

From among the family of reconstructed paths that follow great circles between 19:41 and 00:19, we consider three paths that align with three airports in Antarctica: South Pole (NZSP), Pegasus Field-McMurdo (NZPG), and Wilkins Runway (YWKS). In August 2014, I first considered a path towards the South Pole that might have occurred after a possible landing at Banda Aceh airport (WITT). Although I have long abandoned the possibility of a landing, the scenario of a holding pattern near Banda Aceh followed by a cruise on a due south course remains an interesting possibility. More recently, the path towards NZPG was  investigated in a paper I co-authored with Richard Godfrey, and subsequently Richard proposed the YWKS destination in a separate paper.

The three paths to airports in Antarctica are shown in Figure 1 as black lines that extend past the 7th arc. The coordinates as the paths cross the 7th arc are also shown.

Added Nov 10, 2017: Additionally, a fourth path is shown which aligns with 45S, 104E, which are the coordinates from the final data set found on the captain’s home computer. This case is included to represent the scenario where MH370 was flown towards the location where fuel exhaustion was simulated on the captain’s home computer.

The match of the four paths to the satellite data is shown in Figure 2. Of the four, the path to NZSP crosses the 7th arc closest to the latitude where the BFO and LRC BTO errors are at their minimum values. (Lower BTO errors occur for constant Mach paths crossing at more southern latitudes. However, any path requiring Mach numbers faster than LRC is unlikely to have enough fuel.)

More on the Path Towards the South Pole

The flight towards the South Pole is interesting because there are several ways that the autopilot might be used to create this flight. In LNAV mode, a pilot could enter a custom waypoint with a latitude of 90S and any longitude. Or, he could enter the built-in waypoint for the South Pole, which is SPOLE. Or, if it’s available in his waypoint database, he could enter the waypoint for the runway serving the South Pole, which is NZSP. Any of these methods would cause the aircraft to follow a path that closely follows a great circle to the South Pole. Also, if a pilot wanted to reach as far south as possible, a path towards the South Pole using LNAV would be an obvious selection.

Another procedure would be to use TRK SEL mode, with a value of 180° and the NORM/TRUE switch set to TRUE. Although this would produce similar results as the method using LNAV, it is possible that the aircraft could at times deviate from the required track, and the path could deviate from a great circle. For instance, a wind gust or turbulence could momentarily cause the track to deviate. Although the autopilot would correct for the deviation and bring the track back to the target value of 180°T, the error in path that accumulated during the track deviation would not be corrected. By contrast, in LNAV mode, deviations from the great circle path are continuously corrected so path errors don’t accumulate.

There is also close alignment between this path towards the South Pole and waypoint BEDAX. I’ll discuss this more in a future article.

Figure 3 shows an exploded view of the search area, showing the boundary of what was previously searched (yellow), and where CSIRO proposes to search next (green). The new search area extends about 25 NM to the northeast of the 7th arc at 35,000 ft (blue), and about 27.5 NM to the southwest. The three impact sites proposed by CSIRO are also shown, ordered by their priority. In this part of the arc, the width of the searched area is about 19 NM to either side of the 7th arc. The highest priority impact site is 23 NM to the southeast of the 7th arc, and falls within the proposed search area, which extends between 19 NM and 27.5 NM from the 7th arc.

Figure 3. Exploded view of search area proposed by CSIRO. (Click on image to enlarge.)

The path that extends to NZSP (white) is also shown in Figure 3, which runs along 93.7E longitude. At the point of crossing the arc, the width searched was only about 6.5 NM to the northwest and 15.6 NM to the southeast. The fact that this part of the arc was only narrowly searched presents an interesting opportunity to search in the future.

In the future articles, I’ll present more thoughts on how the MH370 aircraft might have been flown between 18:22 and 19:41, and the implications for possible impact sites along the 7th arc.

Update on November 10, 2017

Here is a CSV file with data for the great circle paths, including the position and track at 19:41, position at 00:19, and speed mode. Included are the data for four paths that align with waypoints past the 7th arc. The four waypoints are the South Pole (NZSP), Wilkins Runway (YWKS), Pegasus Field-McMurdo (NZPG), and the fuel exhaustion position from the simulator data (45S, 104E).

Update on November 11, 2017

In Figure 4 below, I have plotted the paths to the four waypoints (NZSP, YWKS, NZPG, 45S) on the same plot that was generated by CSIRO to show the cumulative probability of detection of low-windage debris by the surface search for various impact points along the arc. The calculated probabilities include the drift that might have occurred between the time and location of the impact and the time and location of the search. It can be seen that there are impact points along the 7th arc and north of 33S where the probability of detection is significantly less than 100%, especially if the impact was to the northwest of the arc.

Figure 4. Efficiency of surface search shown with selected great circle paths. (Adapted from CSIRO.) (Click on image to enlarge.)

Update on November 14, 2017

IG Member Brian Anderson has reminded us that “There are a number of RNZAF photos of interesting flotsam (debris), in areas that may now be much more significant. Unfortunately none was ever recovered.” Two of the more interesting photographs are found below. The first photo (Figure 5) is an unknown object that could be part of an aircraft. The second photo (Figure 6) is a field of floating debris.

Figure 5. Aerial photograph (5849) of unknown object on March 29, 2014, at 28.8866S, 96.0844E. (Click on image to enlarge.)

Figure 6. Aerial photograph (5832) of debris field on March 29, 2014, at 28.3927S, 97.7750E. (Click on image to enlarge.)

Both photographs were taken on March 29, 2014, at coordinates not far from where the great circle path to 45S,104E crosses the 7th arc near latitude 28.3S.

In order to determine if these objects are consistent with the expected drift of debris from an impact near 28.3S on March 8, 2014, I used the CSIRO-generated drift results for an impact on the arc near 28S latitude, where the drift model was seeded with objects within an approximate +/- 0.5 deg square area. The drift model results are shown in Figure 7 for the calculated position of debris on March 29, 2014.

Figure 7. Drift results from CSIRO for an impact along the 7th arc near 28S latitude. (Click on image to enlarge.)

The results are shown for low windage debris (red) and high windage debris (green), and should be representative of a range of objects produced by the impact. Both the unknown object and the debris field are found in the general vicinity of where the model predicts objects would drift for the modeled impact location. This makes the possible impact site of 28.3S even more interesting.

Simulator path (red) and recent MH150 flights (colored circles). (Click on image to enlarge.)

Introduction

Within weeks after MH370’s disappearance on March 8, 2014, news stories revealed that Captain Zaharie Shah deleted data from his home flight simulator in the weeks prior to the disappearance, and FBI investigators at Quantico, Virginia, were assisting in recovering the data. Months later, new reports surfaced stating that the captain had used his home simulator to practice flights to the Southern Indian Ocean, and then deleted those flights from his home computer. Despite these reports, there was no official confirmation from Malaysia that suspicious data was found on the captain’s home computer.

We now know that the recovered data sets from the captain’s computer were included in a report compiled by the Royal Malaysian Police (RMP). The report was eventually obtained by several French media organizations, and large portions of the report were later made public.

On August 14, 2016, the MH370 Independent Group (IG) released a preliminary assessment of the simulator data contained in the RMP report. The data sets were created by Microsoft Flight Simulator (MSFS) software, and are in the form of fragments of “flight files”, which a user may create during a simulation session to record the state of the session for future reference or to resume the session at a future time. The particular flight files of interest were among hundreds of others found on the computer on several drives; however, the files of interest were deleted and recovered by investigators from a “volume shadow” on a single drive that was found disconnected from the computer. The grouping of the files in the volume shadow makes the set unique among the all flight files that were found.

The recovered flight files include flight and navigation parameters that are “snapshots in time”, and are associated with six unique coordinates. If these coordinates were all from a single simulation, it suggests that a user simulated a flight of a B777-200LR aircraft with a departure from Kuala Lumpur International Airport (KLIA), a climb and cruise over the Malacca Strait and then the Andaman Sea, followed by a turn to the south, and a termination from fuel exhaustion in the Southern Indian Ocean (SIO) near 45S 104E. The IG found  that if a great circle path that connects the final points is extended past the final point, the great circle would cross McMurdo Station, Antarctica. (McMurdo is the largest and most populated research station in Antarctica.) This raises the possibility that McMurdo was used as a final waypoint for navigation with the expectation that fuel would be exhausted in the SIO, well before reaching Antarctica.

The simulator data was found on what the Malaysian investigators refer to as drive MK25, where Microsoft Flight Simulator (MSFS) 2004, also known as FS9, was installed. (The main drive of the computer is referred to as MK26, where Microsoft Flight Simulator X, also known as FSX, was installed.)

The fragments of the flight files were found in a volume shadow dated February 3, 2014. A volume shadow is periodically created by the Windows operating system so that files can be restored to a particular date. This means that the file fragments found in the volume shadow were created on or before February 3, and then over-written on or after February 3.

Of the eight total flight file fragments that were recovered, three were for an aircraft on the ground at Kuala Lumpur International Airport (KLIA), and five were for an aircraft flying. Based on fuel levels and other indicators, the data sets in which the aircraft is flying can be related to one of the data sets for the aircraft at KLIA, i.e., six data sets appeared to be related to a single flight. The remaining two data sets at KLIA show fuel levels that are not consistent with the other data sets. These fuel levels may reflect intermediate values before the final takeoff fuel levels were selected.

There has been much discussion as to whether the six data sets are from a single simulation session, and whether the simulator was functioning normally when the data sets were saved. On November 29, 2016, I co-authored a paper with MSFS expert Yves Guillaume that investigated these and other aspects. To replicate the software used by the captain, Yves and I independently installed FS9 with the B777-LR model upgrade from Phoenix Simulation Software (PSS). After substantial testing, we discovered that:

• The data points in the Andaman Sea share some of the same unique values as the data points in the SIO, suggesting the flight files came from the same simulated flight.
• The parameters related to fuel and flight dynamics show that the position of the aircraft was manually changed along the flight path.
• The data files were manually created and saved after certain parameters were manually changed.
• The simulator appears to have been fully functional, and the rates of climb and other flight parameters after fuel exhaustion can be explained and repeated.

After this paper was published, I created a simulation that was initialized using the recovered values from the data set just after fuel exhaustion. I was able to show that it was possible to manually fly the plane for about 27 seconds and achieve the position, altitude, heading, pitch, and bank that were close to the values contained in the subsequent, final data set. This was further confirmation that the captain’s simulator was properly functioning when the data sets were created. Here’s a video of this short flight.

With the recent release of the ATSB’s report on the operational search, we have learned more about the simulator data found on the captain’s computer. Now three years and seven months after the disappearance, for the first time, the existence of the simulator data has been officially acknowledged in a public document. Although the ATSB does not speculate about whether the captain was responsible for the diversion, the report does reveal that in April 2014, the ATSB considered the simulator data as relevant evidence in defining the search area.

Date and Duration of the Simulator Session

From the ATSB’s report, we learned that on April 19, 2014, the Australian Federal Police provided the ATSB with the recovered simulator data. The report states that The simulator data was a partial reconstruction of a flight simulator session from 2 February 2014. Based on the February 3 date of the volume shadow, we already knew that the simulation was created on or before February 3. However, I was curious to know how the ATSB determined that the exact date of the simulation session was February 2 since the information about date and time was not included in the file fragments that were provided in the RMP report.

After some email exchanges with the ATSB, I was surprised to learn that the ATSB has additional data values from the recovered flight files that were omitted from the data sets that were provided in the RMP report. In particular, there is a section of the flight files in which the date and time of the simulation session are stored. These data values tie the date of the simulation to February 2, 2014. Also, the time values show that the chronological order of the data sets matches a flight departing KLIA, flying over the Malacca Strait, continuing past the Andaman Islands, turning to the south, and exhausting fuel in the SIO, in the same order that the progressively depleting fuel levels suggest. The time values also indicate that the flight session lasted for about one hour. This confirms that the position and fuel levels were modified during the simulation, just as Yves Guillaume and I had concluded in our paper.

The newly released information regarding the date and duration of the simulator session means that it is almost certain that the recovered data sets were from a single flight session.

However, as the signed date of this part of the RMP report is May 15, 2014, it remains a mystery as to why the Malaysians omitted important data values that were already available to the ATSB on April 19, 2014.

Simulated Flight Path

Using our previous notation, we refer to the identification of each data set by its latitude, i.e., as 2N, 3N, 5N, 10N, 45S1, and 45S2. The position of the first four coordinates are shown above in the figure, where the position for data set 2N is Runway 32R at KLIA.

After takeoff, the next coordinate is 3N, which is in the Malacca Strait on airway R467 between waypoints AGOSA and GUNIP. The altitude for 3N is 23,247 ft.

The next is coordinate 5N, which is between GUNIP and TASEK near airway B466. The exact position suggests the aircraft has already left airway B466; the track, which can be calculated from the components of the velocity presented in the 5N data set, is 317 deg, which is exactly towards waypoint VAMPI on N571. (The value for the heading is 315 deg, which differs slightly from the track due to wind.) The altitude at coordinate 5N is 32,246 ft.

The final point for the northwest portion of the path is coordinate 10N, at an altitude of 40,003 ft, and to the west of the Andaman and Nicobar Islands. The plane is at a heading of 256 deg, and is banked at 20 deg and turning left towards the south. The coordinates for 10N place it near airway N877 between waypoints LAGOG and DOTEN. The position suggests that the aircraft left the airway about 30 NM before DOTEN.

The final two coordinates, 45S1 and 45S2, are in the SIO near 45S, 104E, and are separated by a distance of about 2.5 NM. Despite the short distance, the altitudes at 45S1 and 45S2 are 37,561 ft and 4,000 ft, respectively. The data set for 45S2 shows unmistakable evidence that the altitude was manually changed from 37,654 ft to 4,000 ft just before the data set was saved, which is consistent with the large change in altitude over a short distance.

The track of 5N towards VAMPI on N571, and the proximity of 10N to airway N877, together suggest that in the simulation session, after taking off and flying to GUNIP, the aircraft joined N571 and followed the waypoints VAMPI, MEKAR, NILAM, IGOGU, LAGOG, and DOTEN. The great circle alignment of 10N, 45S1, and Pegasus Field (NZPG) at McMurdo Station, Antarctica, means the session might have simulated a great circle path towards McMurdo Station until fuel exhaustion was reached in the SIO. (Aircraft that are flying in LNAV mode follow great circle paths between waypoints.)

Similarities to Flight MH150

We know that the recovered simulator coordinates describe a flight departing KLIA that is not consistent with a flight to Beijing: the initial path of the simulation is to the northwest, flying over the Malacca Strait, and the initial path to Beijing is to the northeast, and flies above the South China Sea. I searched for whether there might be a scheduled flight that better matched the simulation coordinates 2N, 3N, 5N, and 10N. In particular, I searched for scheduled flights that had a departure that followed a route from KLIA to GUNIP, and then to either TASEK or VAMPI, and would require a fuel loading similar to the 68,424 kg that is inferred from data set 2N.

After considering the list of Malaysia Air’s departing flights from Kuala Lumpur, I discovered that MH150, from Kuala Lumpur to Jeddah, Saudi Arabia, met the fuel and path criteria. Next, I checked whether the captain had flown any MH150 flights in that time period. I was surprised to learn that he commanded MH150 on February 4, 2014, which appeared to be very close to the date that the simulation session was created. In November 2016, I privately communicated my findings to other independent investigators, although I thought it was still too speculative and controversial to present in a public forum. Months later, on April 30, 2017, I noted in a blog comment:

The flight might have been a simulation of a flight from KLIA to Jeddah with a diversion, as the fuel loading and SID are consistent with that. Also, Zaharie Shah was scheduled to captain MH150 to Jeddah on Feb 4, 2014. The Shadow Copy Set containing the deleted flight file fragments was dated Feb 3, 2014.

Now that we know that the exact date of the simulation session was February 2, 2014, the link between the simulation session and MH150 is less speculative. The authors of the recent ATSB report had similar thoughts:

On the day the simulation was conducted the PIC was on a rostered day of leave. The following day the PIC was rostered to fly from Kuala Lumpur to Denpasar, Bali and return the same day. On 4 February 2014 the PIC was rostered to fly from Kuala Lumpur to Jeddah, Saudi Arabia. The first three data points recovered from the simulator were consistent with the route from Kuala Lumpur to Jeddah. In the weeks between the Jeddah flight and the accident flight the PIC was rostered to fly return flights from Kuala Lumpur to Denpasar, Beijing, Melbourne and then Denpasar again.

I was curious to see just how close the path suggested by the simulator coordinates resembled previous paths taken by MH150. For the comparison, I used the ADS-B data available from FlightRadar24 for eight consecutive flights between Aug 17, 2017 and Sep 21, 2017. (The available data for all the flights end around waypoint LAGOG, probably due to the range limitations of the receivers in the vicinity.)  The MH150 flight paths are plotted above in the figure.

For seven of the eight flights, the aircraft departed KLIA and flew over waypoint GUNIP. For the seven flights that flew over GUNIP, six followed airway N571 towards VAMPI and continued to follow N571 towards LAGOG. One of the six that reached GUNIP, one flew towards TASEK on airway B466, but rejoined N571 near IGOGU. (Some of the flights show deviations from airways, which are probably weather related.). The similarity between typical routes taken by MH150 and the simulator data is evident.

We can only speculate as to why, just two days before he commanded MH150 from Kuala Lumpur to Jeddah, and five weeks before the disappearance of MH370, the captain used his home computer to simulate a flight of MH150 that was diverted to the SIO.

In Part 2, we further investigate the simulator data and how it may give us clues about MH370.

Updates on October 13 and 15, 2017. In the figure, the dates in the legend were corrected so that all refer to year 2017. Also, the label on the airway between LAGOG and DOTEN was corrected to be N877, and the label of the airway between GUNIP and TASEK was corrected to be B466. Thank you to Twitter handle @BKKDiet and blog contributors @Mick Gilbert  and @TBill for finding these errors.

The Australian Transport Safety Bureau (ATSB) has released a report entitled “The Operational Search for MH370”. It is a long document (440 pages) that is meant to provide final documentation of all of the ATSB’s activities related to this incident. There are no new conclusions, although we can gain some new insights:

• MH370 flew over or near IFR waypoints a cluster of waypoints near Kota Bharu called ABTOK, KADAX, and GOLUD and later PUKAR as it flew towards Penang. This implies that MH370 was following waypoints after the transponder stopped operating after the turn back.
• The registration of the first officer’s cell phone on a tower on Penang Island is officially acknowledged for the first time. A footnote citation says, “This information was obtained by the Royal Malaysian Police and reported to the Ministry of Transport Malaysia. Though a formal report was not available to the ATSB, information relevant to the search was shared.” Of course, this begs the question as to why the ATSB only learned of this information after the RMP report was leaked.
• The two sources of the primary surveillance radar (PSR) data revealed to be the civilian radar at Kota Bharu and the military radar on Penang Island on Western Hill. The military radar captures are described as “not continuous” with no further explanation.
• After passing Penang Island, the report says the “Radar data shows the aircraft then headed to the northwest, eventually aligning with published air route N571 from IFR waypoint VAMPI. The validity of this section of the radar data was verified using the track of a commercial flight that followed N571 about 33 NM behind MH370.” This implies the radar captures shown to the NOK in Beijing on March 21, 2014, at the Lido Hotel, are valid. The performance of the military radar was verified by comparing the civil radar data to the military data as another commercial aircraft, likely to be EK343, trailed 33 NM behind MH370.
• The data recovered from the captain’s home flight simulator is discussed for the first time in an official report. An overview of the data is presented, suggesting a flight from Kuala Lumpur up the Strait of Malacca, and then towards the Southern Indian Ocean (SIO), ending in fuel exhaustion. The flight path recovered from the simulator did not match the MH370 flight paths that were reconstructed from the Inmarsat satellite data.
• The ATSB pegs the date of the simulator session as February 2, 2014. We know that the deleted file fragments were found in a Shadow Volume with the date February 3, 2014, so likely the session was created on or shortly before that date. It is not explained how the ATSB can be sure the session was created on exactly February 2, 2014, but this would be significant.
• There is mention that the last data point in the SIO suggested there was a user input of an altitude of 4,000 ft. (The evidence that the simulator’s user manually changed the altitude and other parameters during the flight was first presented in a paper by me and Yves Guillaume.)
• There is acknowledgement that the simulator data shows a beginning sequence that is similar to the flight the captain flew from Kuala Lumpur to Jeddah, Saudi Arabia, on February 4, 2014. (The significance of the similarities in path and timing between the simulator data and the captain’s flight MH150 to Jeddah was first discussed on this blog.)
• Although the ATSB does not speculate that the captain was responsible for the diversion, and although the simulator data was not deemed useful for reconstructing possible flight paths of MH370, based on the presentation of this data in the report, it is clear that the ATSB considers the simulator data to be significant evidence.
• Based on examination of the debris and a detailed study of the final BFO data, the ATSB believes that MH370’s flight ended in a steep, uncontrolled descent. This will serve to limit the distance from the 7th arc for future searches.
• Although there was an attempt by the official investigators to discern information about the crash site from an investigation of the marine ecology attached to the recovered debris, all results were inconclusive.
• No new drift analyses are presented. The ATSB reaffirms its belief that the most likely crash site is 35.6S, 92.8E, based on the drift analyses by CSIRO.

So, although there are no new conclusions in this report, there are some interesting new pieces of information. It is also important to note that Malaysia chose to omit key pieces of evidence from the Factual Information (FI) released in March 2015 that are presented in the new ATSB report. These pieces of evidence include details about the radar data, information regarding the simulator data found on the captain’s home computer, and the data related to the registration of the first officer’s cell phone as the aircraft flew near Penang. Although these omissions have been discussed in detail on this blog, perhaps with the release of the ATSB report, more will question why Malaysia chose to not disclose, and even deny the existence of, important evidence.

On a final note, the ATSB chose to acknowledge the contributions of some of the independent investigators, including many that comment here. The ATSB was kind enough to give a special recognition to Blaine Gibson:

The ATSB acknowledges the extensive contributions that many individuals and groups have made during the underwater search for MH370. Many contributors have provided credible, alternate and independent approaches and analysis of the limited data available. In particular, the ‘MH370 Independent Group’ comprised of scientists, researchers and individuals who have cooperated across continents to advance the search for MH370. The ATSB is grateful for their work collectively and individually including Duncan Steel, Mike Exner, Victor Iannello, Don Thompson, and Richard Godfrey. The ATSB also acknowledges the extensive and detailed contributions provided by Simon Hardy, Bobby Ulich and Robin Stevens.

The search for MH370 was significantly advanced after the first debris from the aircraft was found on La Reunion Island in July 2015. The subsequent efforts of Blaine Gibson in searching for and locating MH370 debris on east African coastlines did much to raise public awareness of the importance of the MH370 debris which led to many more items of debris being handed in. Mr Gibson met and communicated with ATSB during his 2015-2016 search expeditions and he is acknowledged for his outstanding efforts in communicating his debris finds to Malaysia, ATSB, the next of kin and the wider world.

Blaine Gibson (right) and Nick Connite (left) in Madagascar with two new pieces possibly from MH370.

In a newly released report, private investigator Blaine Gibson reveals details surrounding two new pieces that could be from MH370. The pieces were discovered by residents of Madagascar in September 2016, and delivered to Malagasy authorities on August 16, 2017.

Blaine writes that “for the protection of those involved, we decided not to make this report public until the debris was safely delivered to Malaysia.” That transfer was supposed to occur imminently. However, with the assassination of Hon Consul Zahid Raza, who served as a diplomat for Malaysia in Madagascar, the transfer has been delayed, and Blaine has decided to now release his report.

In the report, Blaine explains that “under the agreement between the two countries, debris is supposed to be collected by Hon. Zahid Raza, the Honorary Malaysian Consul in Madagascar, and delivered by private courier to Malaysia. On August 24 the Hon. Zahid Raza was assassinated in Antananarivo. The debris is still safely in the hands of the Madagascar Authorities. However new arrangements must be made for the collection and delivery of debris.”

Although not mentioned in the report, Blaine told me that during a trip to Madagascar, death threats were made to him and others for continuing to collect debris related to MH370. Blaine told me about these threats when I met with him on August 3, which was three weeks before Mr Raza was assassinated. The link between custody of the debris and the slain diplomat was first discussed in a previous post.

Blaine attributes the new discoveries to be the “result of the 370 families’ debris search and awareness efforts and travel to Madagascar.”

Twenty-seven photographs of the two pieces are available for download.

Of the two parts, the larger one was found on Maroantsetra Beach on Antongil Bay in September 2016. Blaine estimates the dimensions to be about 27 inches long, 12 inches wide, and 2 3/4 inches thick, and composed of a composite honeycomb structure. However, there is a 3 1/2-inch strip attached with fasteners that have not yet been identified as aviation related.

The two new pieces. The smaller piece (left) is more likely from an aircraft than the larger piece (right).

The smaller part was found on Antsiraka Beach around September 2016. Blaine estimates the dimensions to be about 12 inches long, 12 inches wide, and 3/8 inches thick, and composed of a composite honeycomb structure. This piece appears to have a higher possibility of being aviation related.

The pieces remain in the custody of Malagasy officials until new arrangements are made to transfer the pieces to Malaysia.

Blaine also has released new maps that show where it is in Madagascar that MH370 debris is predicted to wash ashore. The maps were created using computer models developed by Dr Charitha Pattiaratchi of the University of Western Australia. Blaine acknowledges that Dr Pattiaratchi and CSIRO’s Dr David Griffin have helped guide search efforts.

Location of MH370 debris as predicted by computer models (white) and where two new pieces were found (red).

The new report, debris photographs, and maps are available for download as a collection of files.

Update on Aug 29, 2017. The fasteners on the larger of the two pieces have been identified to be aviation related. Thank you to Annette Mansfield and Mike Exner for that information.