Vessel Details for the 2025 Search for MH370

Content researched and created by Don Thompson

Armada 78-06. Credit: Wei-Lin Chen, @taiwanshipswatchers

Ocean Infinity previously conducted a search for MH370 in 2018 using a fleet of Kongsberg Hugin 6000 Autonomous Underwater Vehicles (Hugin AUVs). The AUVs were operated from Seabed Constructor, a vessel designed to support subsea construction activity. Seabed Constructor’s roster of equipment was impressive, a truly multi-purpose subsea vessel: eight Hugin AUVs, acoustic comms and positioning system, two 5000-6000m depth rated (tethered) ROVs, autonomous surface vessels, and a crane able to lift from 3000m deep.

In early 2021 Seabed Constructor’s lease was terminated, along with Normand Frontier, as Ocean Infinity prepared to introduce its fleet of Armada 78 vessels.

The Armada 78 vessels, at 78m length, are smaller than Seabed Constructor. The fleet comprises four distinct operating configurations, none of which involve permanent AUV mobilisation.  However, in the context of an MH370 search mission, an Armada 78 can operate Hugin AUVs deployed in a self-contained ISO 668 form factor hangar/garage.

Container module for deploying Hugin AUVs. Image credit: H Henriksen

As of mid-January 2025, four Armada 78 vessels are operating in European waters, one has been operating out of San Diego, California, while three underwent further completions work in Singapore following the experience of those vessels operated in Europe.

While not permanently mobilised to operate AUVs, some Armada 78 vessels have been configured to deploy AUVs: Armada 78-07, from San Diego, to the Clarion-Clipperton Zone, Pacific Ocean; Armada 78-05, surveying pipeline/connector routes in the North Sea and Norwegian Sea; and, Armada 78-06, to the Ryukyu Trench, east of Taiwan, operating out of Kaohsiung port.

Recently, through December 2024-January 2025, much discussion has ensued, speculating on which of the three vessels based in Singapore may be a candidate to conduct a renewed search for MH370 in 2025. One commentator even asserted that Armada 78-08 had been designated as the vessel to conduct the search. Through December 2024 and January 2025 Armada 78-04, 78-06, and 78-08 separately departed Singapore.

Armada 78-04 departed 14th December 2024 fully configured for geotechnical operations with its CPT and vibrocore machines in place and without AUVs. In this configuration, it was not a candidate to conduct the search. After departure from Singapore, Armada 78-04 passed through the Sunda Straits, crossed the Indian Ocean and proceeded to the Atlantic with enroute replenishment and bunkering stops at Port Louis, Mauritius, and Cape Town, South Africa.

Armada 78-06 departed for a task in deep waters east of Taiwan. At departure, it was mobilised with two containerised AUV garages on the aft deck. Upon its return to Singapore, the vessel moored at the ST Engineering facility before departing again on January 19, 2025, and broadcasting Mauritius as its destination. Armada 78-06 passed through the Sunda Straits before establishing a course across the Indian Ocean for Mauritius.

Armada 78-08 had departed for Cape Town with a replenishment and bunkering stop at Port Louis, Mauritius. The vessel is mobilised to operate two Saab Seaeye Leopard ROVs using a LARS over the forward moonpool. These ROVs do not have the ability to operate at the depths found in the MH370 search area. No containerised AUV garages were deployed on the vessel at departure from Singapore.

Following a brief stop in Cape Town, Armada 78-08 departed, this time eastbound to the small port of Mossel Bay where a containerised AUV garage was waiting on the quayside to be loaded onto the vessel. The curious folk of Mossel Bay recorded, on social media, the missing piece of the puzzle to complete the picture of vessel preparation for the MH370 search: Armada 78-08 is ferrying a third AUV to Port Louis where it will be transferred to Armada 78-06 and transit will begin to the search area.

We suspect that delivery of the third container AUV garage was delayed in transshipment to A78-08 at Cape Town. By repositioning Armada 78-08 to Mossel Bay while the intermodal container was delivered by road to the port, any overall delay to the schedule was minimised. Armada 78-08 and 78-06 should arrive at Port Louis within a short time of each other.

While the period required to fully mobilise the additional, third, containerised AUV garage on Armada 78-06 is unknown to us, an estimation of sailing time Mauritius to 7th arc search area is approximately 10 days. Assuming a rendezvous of Armada 78-06 and Armada 78-08 around February 5, and assuming a week to outfit and resupply Armada 78-06, the search might begin around February 22, and perhaps sooner if the time to outfit and re-supply Armada 78-06 occurs faster than one week.

[VI comment: Here are some additional remarks to refute claims that are circulating.]

First, there is no evidence that Malaysia will not sign a contract with OI to search again. Yes, the negotiations seem to be taking longer than any of us would like, and yes, this might reduce the duration of the search compared to the last search, but considering only 15,000 km2 of area is proposed, the shorter duration of the search might not matter. The claim that Malaysia is not interested in searching again is at best premature, and most likely it is false.

Second, there is much speculation about the trajectory of Armada 78-08, which some believe is the vessel designated for the search. More likely, that is not the selected vessel for the search. As Don deduces in the article above, Armada 78-06 is more suitably designed and outfitted for the search.

Thirdly, there is no reason to believe that the search area will be very much different than what OI proposed in March of last year (2024), as shown at the top of the prior post. That area is largely consistent with the Zone 2 search recommendation proposed in Ulich and Iannello (2023). We also hope that OI will have the willingness and capability to search the steep slope at around 33 km (18 NM) due south of our best estimate of where MH370 crossed the 7th arc. This area, centered around S34.53° E93.84°, was not previously searched by either the ATSB or OI, and remains a high priority despite the challenging terrain, as discussed in this previous post.

So, although there are still many unknowns regarding the search, including the precise start date, I still believe it is likely that a search contract will be signed and the search will proceed without many surprises.

Tags: , , , ,

77 Responses to “Vessel Details for the 2025 Search for MH370”

  1. airlandseaman says:

    Nice summary Don. And Victor’s comments are spot on.

  2. Mick Gilbert says:

    Mmm … properly researched, cogently reasoned, an absence of incendiary hysteria. Definitely not a Los tres idiotas que ladran production.

    Thank you Don and Victor.

  3. Victor Iannello says:

    Most of you are probably aware of the collision between a CRJ-700 commuter jet on short final to RY33 at DCA and a BlackHawk military helicopter passing over the centerline to the runway.

    There are two references that I believe summarize what we know quite well:

    Philip Greenspun’s blog: https://philip.greenspun.com/blog/2025/01/29/reagan-national-airport-blackhawk-crj-crash/

    Juan Browne’s YouTube channel (blancolirio): https://www.youtube.com/watch?v=ouDAnO8eMf8

  4. TBill says:

    @Victor
    I believe correction on Juan Brown’s, the copter was based in Fort Belvoir and not Joint Base Bolling/Anacostia. Fort Belvoir is (near me) in Virginia a bit south of DCA, doing training flight reportedly. The crash itself was near Joint Base Anacostia/Bolling near DCA. I do not think we know the training flight plan for the BlackHawk but presumably going up and down the Potomac River visual rule flight paths.

  5. Victor Iannello says:

    @TBill: Thank you.

    The CRJ was on an (possibly ILS) approach to RY01, circling to RY33. There were three CRJs before AA5342 (AA5038, AA5608, and AA5170) that all followed similar trajectories. Nothing looks out of the ordinary for the jet.

    It’s not clear why visual separation was not maintained by the helicopter. We need to know why the “holes in the Swiss cheese” aligned”, but this requires an understanding of the standard procedures for military helicopter operations near DCA.

  6. Barry Carlson says:

    @Victor,

    My understanding is that the helicopter – PAT25, when requested by the Control Tower if they had a visual on the CRJ, replied on UHF – “PAT25, Have traffic in sight. Request visual separation.”

  7. TBill says:

    @Victor
    One last comment. I suspect everyone is hearing the same news, but local discussion this was apparently a night vision goggle training flight. So that has visual flight implications of (1) goggles restricting peripheral vision, and (2) the lighting is altered vision of course and reportedly not always ideal for VFR. Of course I would not know if the copter occupants were all wearing night vision goggles, but that seems to be implication for this type of flight training.

  8. @370Location says:

    Re CRJ: When two craft are on a direct collision course, they appear to each other to be static on the horizon, only growing in size. My father who was a private pilot and avid fisherman taught me this when I was spotting for him, to watch out for the craft that were moving but not moving.

    Mick West used his custom perspective sim software to load the ADSB for the crash. He shows how the next plane on final for runway 01 could easily be mistaken for the CRJ on final to runway 33. However, the sim CRJ appears to be moving against the sky from the Blackhawk perspective, and passes above it.

    Most likely, this is ADSB reporting inaccuracy due to the altimeters having slightly different barometric settings or maybe military depends on GPS estimates.

    I agree with others the helicopter crew mistook the CRJ for a static planet or a tower, and were indeed distracted by the much brighter landing lights aimed in their direction.

    Regardlesss, the Potomac crash is another great tragedy, and I don’t want to be assigning blame.

  9. Victor Iannello says:

    ADSB data shows that AA4514, an Embraer E175LR, landing at DCA on RY 19 (the reciprocal to RY 01) about 24 hours before AA5342 collided with the BlackHawk helicopter, aborted a landing, allegedly due to a helicopter crossing in front of it on final approach. The ADSB data shows the plane executing a go-around on the first approach at around 1500 ft MSL before successfully landing on the second approach.

    https://www.flightradar24.com/data/flights/aa4514#38e9549a

  10. Mick Gilbert says:

    @Barry Carlson

    G’day Barry,

    I think that the Blackhawk was on the DCA Twr Helo VHF frequency, 134.350, not UHF as has been suggested from some quarters.

    @Victor Iannello

    According to one of the least reliable commentators on matters, aviation it was all down to the Blackhawk flying a “covert mission”, https://youtu.be/93OmWojjBeE?t=110

  11. Andrew says:

    @Victor

    According to FR24, the last Mode S transmissions from the helicopter show it was flying level at 400 ft (pressure altitude) shortly before the collision. The QNH at the time was 29.90″ or 1012.5 hPa, so the helicopter’s altitude was just under 400 ft AMSL. The FAA’s Helicopter Route Chart shows that helicopters flying on Route 4 should be at or below 200 ft AMSL north of Wilson Bridge, presumably so they stay below the approach paths of aircraft landing at DCA.

    FAA Helicopter Route Chart Baltimore-Washington

  12. Victor Iannello says:

    @Andrew: I think it is a case of the holes in the Swiss cheese lining up, with one of those holes the altitude of the helicopter.

    Here are two more videos from sensible people, both commercial pilots and former military pilots:

    Juan Browne (most recent): https://www.youtube.com/watch?v=_3gD_lnBNu0
    Hoover (Pilot Debrief): https://www.youtube.com/watch?v=RzQe6W7vcu4

    This video is from Dan Gryder, who tends to sometimes speculate more than the facts warrant, but often provides interesting insights. He places the blame 100% on the controller for switching the runway from 01 to 33 while the helicopter was flying published Rt 1 to Rt 4.

    https://www.youtube.com/watch?v=tRnk_ycXYQI

  13. Victor Iannello says:

    @Mick Gilbert: Sad.

  14. Andrew says:

    @Victor

    Thanks, I agree there are several factors that are likely to have contributed to the accident. The two videos by Juan Browne and Hoover provide some great insights. The change to Runway 33 might well have been a contributing factor, but I think Dan Gryder’s claim is premature, to say the least.

  15. TBill says:

    @Victor et al
    According to WTOP news radio here, the training mission was a drill for continuity of leadership, in other words, practice for evacuating national leaders from DC in event of national emergency.

  16. Victor Iannello says:

    @370Locations said: Most likely, this is ADSB reporting inaccuracy due to the altimeters having slightly different barometric settings or maybe military depends on GPS estimates.

    The altitude reported by ADS-B and Mode C transponders is the pressure altitude, which is independent of the altimeter setting. The (local) altimeter setting only changes what altitude is indicated on the aircraft and ATC displays.

  17. Adi says:

    Simple question: how is PAT25 supposed to visually distinguish a CRJ (specifically) vs. a larger plane among the row of planes lining up to land (on either runway) AT NIGHT?

    Simple answer: it cannot.

    This accident is 95% on the controller for:
    a) Making a moronic assumption, that PAT25 has the ability to dynamically make the above determination at night and then visually avoid conflict at night
    b) Making the best case assumption that once PAT25 says “we got it,” the ATC can wash their hands off active management of any potential conflict
    c) An almost criminal lack of direct AND URGENT conflict resolution directives to BOTH aircraft, the moment CA showed up on their screen, complete with audio blaring.

    When you are in the business of being possibly the last man standing to avoid mass casualty, you DO NOT make convenient assumptions. You act and resolve. He had 16 seconds to avoid 67 deaths.

    I would ascribe 40% blame to PAT25 for the obvious reasons: shooting over their assigned altitude barrier, and more importantly for being over-confident in their ability to spot a CRJ in pitch darkness. He should’ve immediately called the controller out and said, “uh, no – I can’t spot a CRJ, but I might be seeing several planes lined up. Can you please advise.”

    And yes, I’m aware that 90%+40% is way above 100%. Trust me, in this business (I manage one in an equally busy airspace further Northeast), we exist on our ability to stamp down risk.

    Best,
    Adi

  18. Kenyon says:

    @Don, thank you (and @Victor’s commentary) for pulling the Vessel Details summary together, I find it quite informative, nice work you two, as always. This helps to bring focus to some expectations and challenges related to the search for MH370.

    During the 2018 search there were (as expected in the SIO at these depths) technical issues and delays. It would seem possible that the strong complement of Seabed Constructor’s on-board capabilities and capacities could have reduced schedule delay risks. I’m curious (generally speaking) if the capabilities of the Armada 78 class offers equivalent schedule risk mitigation capabilities. Time to launch & recover (AUV/ROV), headcount for on-board tech crew, tech gear, rough sea capabilities etc…

    @all, thoughts and prayers to all impacted by the two recent aviation incidents in the US. Heartbreaking. I’m hopeful the lessons learned are properly investigated and will save lives in years to come.

  19. 370Location says:

    @Victor:

    Thanks for the pressure altitude correction.
    NWS says the barometer was 29.90 at the crash, vs standard 29.92 inHg.
    That appears to be a difference of about +20 ft.
    The Potomac is at mean sea level there.

  20. Victor Iannello says:

    @Adi: Thanks for your comment. I won’t even try at this point to assign relative blame. However, the controller could have (should have?) told the BlackHawk the relative direction of the CRJ, e.g., 11 o’clock. This greatly helps the pilot to identify the correct contact and helps to avoid confusion in busy airspace.

    Would you be willing to share what your role is in managing air safety?

  21. TBill says:

    Re: DCA Crash
    To quote Sen Tim Kaine, there are going to be a whole lot of questions (regarding the airspace design, management, and so on). DCA has always been questionable location, but it is such a convenient quick in/quick out airport close to DC that it is very popular. Part of the original idea was IAD Dulles was to take growth but it suffers from DCA popularity and growth there. Newly completed is a very expensive subway Silver Line Metro all the way out to Dulles via Tysons Corners from DC, but not too much use yet. Of course, since COVID most of gov’t offices have been work from home, so subway and commuter train ridership has been rock bottom, with perhaps change now. They are talking gambling casinos in Tysons, presumably to beef up Silver Line ridership and grow Tysons more to what they were hoping the Silver Line would enable. Meanwhile locally we are mourning the Northern Virginia ice skating students and parents and coaches lost. My neighbor is retired civilian admin from Ft Belvoir, but had worked in the night vision dept.

  22. 370Location says:

    @Adi:
    “lack of direct AND URGENT conflict resolution directives to BOTH aircraft”

    This might be the crux of avoiding the collision. Previous near-misses at DCA were when the aircraft on final approach aborted their landing due to helo conflict. ATC advised PAT25 which took responsibility, but CRJ apparently wasn’t notified, and on a collision course would have seen a stationary flashing light that looked like something on the ground. Sadly, another hole in the Swiss cheese.

  23. Barry Carlson says:

    @Mick Gilbert,

    G’day Mick.

    Re PAT25 tx frequency.. My “faux pas” for not checking the AIP and relying on someone else’s comment.

    Seems to me that a number of other “mis steps” were contributors to this totally preventable accident. The most obvious to me being the landing separation between sequenced aircraft for RWY 01 being that tight that the TWR was relying on some aircraft to accept a deviation to RWY 33.

    Collisions occur when the relative movement observed by each participant of the other is effectively none. True for 2D and 3D situations.

  24. Andrew says:

    @Barry Carlson
    @Mick Gilbert

    The latest NTSB Briefing (No. 3) (08:42 onwards) makes very clear the helicopter was using the UHF frequency.

  25. Mick Gilbert says:

    @Barry Carlson

    As it turns out, Barry, the Blackhawk was on UHF.

    @Andrew

    Thanks for that clarification and reference.

  26. Barry Carlson says:

    @Andrew
    @Mick Gilbert

    Thanks for that clarification Andrew. I had just watched the latest briefing by the NTSB and came to this blog to retract my previously described “faux pas”. Transpires that my original source was correct.

    Thanks Mick for raising the matter, which confirms to me that the TWR was effectively receiving information concurrently on VHF and UHF, but transmitted responses were selectively made either on VHF or UHF. So situational awareness by General Aviation and Military aircraft pilots operating in close proximity was at the TWR’s discretion.

    The holes in this Swiss Cheese model, sadly lined up.

    The terminology, “Cleared to Land” was seriously flawed.

  27. Andrew says:

    @Barry Carlson

    Just to clarify, my understanding is that all tower transmissions were on VHF and UHF. Both aircraft could hear everything the tower said, but they could not hear each other.

  28. Barry Carlson says:

    @All,

    My apologies for misrepresenting the TWR’s transmissions as being selective.

    However, when operating in close proximity in controlled airspace, being unaware of the other (Military or Civil) aircraft’s transmissions doesn’t aid in situational awareness.

  29. Victor Iannello says:

    AIS data indicated that Armada 08-06 should arrive in Mauritius tomorrow (Feb 4) and 08-08 on Wednesday (Feb 5). That’s consistent with the timeline presented in the article above. Assuming one week to reconfigure and resupply Armada 08-06, the search could begin around Feb 22, and possibly earlier if the turn around operations in port take less than a week.

  30. Don Thompson says:

    @All,

    A78-06 has arrived in Port Louis, Mauritius. Port movement updates state that A78-06 is due to remain until Sat 8th February. It has berthed at Terminal II, Quay 3 and A78-08 is due to tie up adjacent to A78-06 on arrival.

    A78-08 is due into Port Louis 2025-02-05T0230Z

    Both vessels are scheduled for load container operations, provisions, crew change.

  31. @370Location says:

    On a slight MH370 digression, I had previously asked for assistance with figuring the probability that the Java acoustic event was geological vs MH370. Well, perhaps the best way to get answers is to try anyway. I realized from the latest breakthroughs in AI that these models being based on evolved statistical weighting might pretend to be the expert I needed.

    I see others have tried AI for MH370, but always with some hallucination or strong bias. I was looking for details on how to impartially calculate the probabilities. My exchange turned into the AI listing what background data was needed, so I provided it. I stuck with factual evidence and left the structural organization and weighting factors up to the latest ChatGPT. It turned into writing a paper, which I repeatedly fact checked and while providing more background details on the hydroacoustic and seismic detections, and the physics involved. ChatGPT did not initially pick up on the close proximity to the 7th Arc being significant until I mentioned that and then provided estimates of seismic epicenter accuracy. I also described the 7th Arc, damage found, flight altitudes, crash timing, and estimates for floating and sinking time when those were raised as factors in the probability.

    Both probabilities were independently low numbers until I asked if the result could be shown in percentages that might incorporate some unknown non-geological event estimate and sum to 100%. Spoiler, 0.1% geological vs 95% MH370 seabed impact in this summary:

    https://drive.google.com/file/d/13ZGbKDq0cwQc4zRH3X1h5Rh3Zhi3lb22/

    I had asked it to stick to unbiased scientific methods, and in that spirit asked for a summary of how the weights, factors and intermediate data were derived with enough formulaic detail that a knowledgable scientist could reproduce similar results given the same factual evidence. After a hint not to repeat the content of the paper, this is the result:

    https://drive.google.com/file/d/13RyL7KVp6vD7qBocRHqKKaaBcnlsVcMK/

    As I wasn’t correcting methodology report, I believe it may have hallucinated that it actually checked H08 signal strength against known thresholds. It also provided very little detail about how the calculation was derived. When I asked earlier, I got a treatise on Bayesian combinatorics. I suspect this is a sign that it is not at all aware of how it is generating its results, even though the underlying neural network weightings and language models are capable of impressive results, especially with programmatic constructions. I have seen OpenAI write a snake game and play it with minimal prompting or errors. It may also be very good at telling a human what it wants to hear regardless of how it is prompted. Perhaps someone has suggestions on how to conduct a counter-experiment by giving it the same evidence, but somehow with prompting bias for a geological event

    I’m hoping true experts here could advise on how accurate these probability results might be, and perhaps a better method for deriving them.

    My intuition says the odds are astronomical that this was a random event unrelated to MH370, but I’m sure anything better than 50/50 should be more than enough to initiate a search of such a specific site.

  32. Adi says:

    @all,

    My prior post was in utter frustration at what happened. It is easy to apportion blame after the fact – like I did above. Yet, it’s hard to get over the fact that this event was so avoidable. I do know that one of things the NTSB has requested is specific historical data. They’ll (this part is a guess) want to observe the patterns of tangent or chord intersections between that now infamous Helo corridor hugging the east bank of the Potomac, and the Runway 33 approach. They’ll probably find a lot more near misses that fell through the cracks because everyone got comfortable with 100 puny feet of vertical separation due to outdated FAA guidelines on vis-sep, over-reliance on Helo maneuverability, and too much trust in human ability to see and avoid.

    That controller was probably well within the rules he was trained on. He just missed common sense, which is unfortunately NEVER optional.

    @370Location, I tend to agree. Since there wasn’t the foresight to actively anticipate the conflict 60 seconds prior to impact, and since the controller knew he was the only one who had the full picture, those last 16 seconds were such a lost opportunity to redirect BOTH aircraft.

    @Victor Iannello, you are correct. No point in ascribing blame. The sad part is that the DCA team does (at least as of a couple years ago) have a good reputation in our circles, which makes me shudder about some of the others in our country – and abroad – with a less that stellar reputation. I would rather not get into specifics of my own background; suffice to say I and my team get the most stress out of Helo pilots flying their corporate bosses in for investor conferences (which all happen to converge geographically and timing-wise) or foreign leader-ferrying aircraft (big and small) who think they can land right on top of a certain building they get invited to once in a while.

    Let’s hope the NTSB patiently drills down to root causes, and more importantly, the FAA understands that slim margins are BAD in aviation; with or without advanced technology.

    On a different – and positive note – let’s hope that contract gets signed quick, and the MH370 search can restart before the seasonal window turns.

    Best,
    Adi

  33. Andrew says:

    @Adi

    RE: “That controller was probably well within the rules he was trained on.”

    I certainly hope the NTSB addresses that issue. My view is the US ATC system has become far too reliant on visual separation procedures to cope with the volume of traffic. ATC is under considerable pressure to keep traffic flowing, to the point where visual separation procedures are applied in situations where they are not appropriate.

    In this case, I suspect ATC routinely approved visual separation between low-level helicopter traffic and aircraft on approach to DCA at night, without fully appreciating the difficulty that pilots have in accurately identifying aircraft on approach, particularly to Runway 33. That issue obviously needs to be addressed. I can’t see the FAA banning the use of visual separation at night, but I would hope they prohibit the use of such procedures when Runway 33 is in use at DCA, with similar restrictions at other airports.

    I’m sure there are many who share your frustration. NTSB board member Todd Inman became quite emotional towards the end of the third media briefing, when he was asked a question about spending time with the victims’ families. He specifically mentions the 200-plus NTSB recommendations that have not yet been addressed by the regulators and that people need to take action to improve safety. His frustration was palpable.

  34. Don Thompson says:

    @Andrew, @all

    I concur with everything you wrote above.

    And reiterate: [Inman] specifically mentions the 200-plus NTSB recommendations that have not yet been addressed by the regulators and that people need to take action to improve safety. His frustration was palpable.

    The vertical separation dictated for the helicopter corridors is, in practice, dangerous. NTSB should make a review of fixed wing near misses with traffic on Route 1 and Route 4 part of the investigation.

  35. Don Thompson says:

    Armada vessel update.

    Armada 78-08 has now arrived at Port Louis, Mauritius, at Terminal II, Quay 3 adjacent to A78-06.

    A78-08 is scheduled to depart Thursday 6th Feb.

  36. TBill says:

    Runway 33.
    This video courtesy YouTuber pilot Gary Baumgardner shows a past actual (daytime) cockpit video and discussion of Runway 33; reveals the unorthodox nature of it. Requires pilots to change landing plans below 1000-ft, when they should be stable on approach. Then the pilot can choose how far to swing out over DC and how sharp to turn into Runway 33. Apparently the pilot who photographed this unauthorized video has since retired (wink, wink). Sure enough this random video taken one day includes ATC warning about helicopters in the vicinity.

    https://www.youtube.com/watch?v=LKJklQRyKxo

    I have been unable so far to understand how many years Runway33 has been used in this manner. I had thought one report said relatively new runway.

  37. Victor Iannello says:

    @TBill said: Requires pilots to change landing plans below 1000-ft, when they should be stable on approach.

    There is a little more nuance than this statement. If on an instrument approach with a circle to another runway, the circling minimum is often less than 1000 ft AGL. There is also a circling maneuvering radius criteria, which for the CRJ-700 (Cat B), is 1.5 NM.

    For DCA, an ILS approach for RY 01 circling to RY 33 would have a minimum descent altitude (MDA) of 900 ft MSL (886 ft AGL) for a Cat B aircraft. So, the aircraft would descend to 900 ft on the ILS approach and level off. Upon getting to within 1.5 NM of the field (and seeing the landing environment), a circling maneuver to RY 33 can begin. For a Cat A aircraft, the MDA is 720 ft MSL (706 ft AGL) and the circling maneuver would begin 1.3 NM from the field.

    I have a friend that was a First Officer for PSA (the same company as the accident flight) and flew the CRJ into DCA many times. The circling approach to RY 33 has the reputation of being fun but challenging.

  38. Barry Carlson says:

    @TBill,

    Thanks for sharing the RWY33 approach video.

    Great daytime scenery. Just flick the switch to night-time and imagine the street, housing and vehicle lighting pollution, while trying to focus on the RWY33 approach lighting. This accident has been a long time waiting to happen.

    Re: MH370,

    The MSL weather in the South Indian Ocean area is currently somewhat different to what was previously considered to be normal. The following link is to the Australian Bureau of Meteorology MSL Analysis –

    http://www.bom.gov.au/australia/charts/indian_ocean.shtml

    In short, the the mid latitude High pressure systems are weaker than normal, while the Southern Ocean Low pressure systems are deeper. The Tropics are active and the Tropical cyclone season is already underway.

    NB – Look for the link to the print version in the above chart; for a high definition version.

  39. Andrew says:

    @Victor
    @TBill

    Also bear in mind there is no single definition of “stabilised approach”. The FAA issued guidance over 20 years ago, largely based on recommendations made by The Flight Safety Foundation (FSF). Those recommendations specified two “gates” for a stabilised approach: 1,000 ft in IMC and 500 ft in VMC, at which point the aircraft should be on the correct flight path, at the correct speed and rate of descent, configured for landing with an appropriate power setting, and with all checklists complete. The FSF recommendations also state that in the case of a circling approach, the wings should be level on final when the aircraft reaches 300 ft AAL.

    Most operators adopted similar criteria in their stabilised approach policies, but there is some variation between operators.

  40. Mick Gilbert says:

    A brief update on Armada 78-08, at around 0400 UTC 6 February -08 repositioned from Quay 2 to Quay 1, most likely to bunker. -08’s destination is now listed as Cape Town.

    Armada 78-06 remains tied up at Quay 2.

  41. Victor Iannello says:

    @Mick Gilbert: Seems like Don was right about 78-06 being the more likely search vessel.

  42. Andrew says:

    @Victor

    RE: “There is also a circling maneuvering radius criteria, which for the CRJ-700 (Cat B), is 1.5 NM.”

    FWIW, at DCA they use expanded circling approach protected areas, as noted by the “C” symbol next to the charted circling minima. For a Cat B aircraft the circling radius is 1.7 NM, where the MDA is 1,000 ft or less.

    The expanded protected areas were introduced in 2013 after years of lobbying by the aviation industry, which felt the standard TERPS circling areas were too small for large jet aircraft. In contrast, the circling radius for a Cat B aircraft under the ICAO PANS-OPS criteria is 2.66 NM.

    In any case, I doubt that many aircraft fly a full-on circling approach to Runway 33 off the ILS to Runway 1. I suspect they mostly fly a visual approach to Runway 33 after an initial clearance for the Mount Vernon Visual to Runway 1, or after an early break-off from the ILS.

  43. Andrew says:

    Many here will be familiar with the “swiss cheese” model of accident causation in high-risk industries. The father of that model, Professor James Reason, passed away yesterday, 5 February.

  44. Victor Iannello says:

    @Andrew: Thanks. I was being lazy for not consulting the Terminal Procedure Publication (TPP) for the expanded radius because I tend to associate it with airports at higher elevations with a corresponding increase in true airspeed. For the Cat A aircraft I fly, at sea level, there is no difference between the standard and expanded radius. However, I do see that for Cats B, C, and D (but not E), the expanded radius is greater than the standard radius at sea level.

    The point I was trying to make to @TBill is that for approaches to DCA’s RY 33, and for circling approaches in general, the aircraft could be maneuvering below 1000 ft AGL. In fact, if the MDA is higher than 1000 ft AGL, it requires a steady descent to stay within the radius and land. This happened to me the other day while landing at Roanoke (KROA). I was on an instrument approach to RY34, circling to RY06. The circling MDA for RY 34 for Cat A is 1465 ft AGL which means I started the circling maneuver well above pattern altitude.

    I think your expectation that aircraft landing at DCA would typically fly a visual rather than a (instrument) circling approach to RY 33 is consistent with reality.

  45. Andrew says:

    @Victor

    RE: “The point I was trying to make…is that for approaches to DCA’s RY 33, and for circling approaches in general, the aircraft could be maneuvering below 1000 ft AGL.”

    Absolutely. Stabilised approach criteria are not intended to preclude manoeuvring below 1,000 ft AGL, if it is required for a ‘normal’ approach to the runway in question. They are intended to provide a set of parameters that pilots can use to to judge when an approach has gone awry and should be abandoned.

  46. Don Thompson says:

    There remains some confusion in certain quarters over which vessel is to conduct the forthcoming MH370 search, even despite what I have set out above in the article leading this post. Spoiler: it remains Armada 78 06.

    It appears that a fundamental absence of any detailed understanding for the Armada 78 vessel configurations may be at the root of this confusion.

    In a comment post made Feb 6th by another site’s author (URL ending comment-3347) a selection of images of Armada 78 06 and 78 08 are linked.

    The third image of the selection is purported by the writer to show ‘Armada 7808 is equipped with 2 AUVs, crane and davits designed for search operations‘. That statement is emphatically incorrect: Armada 78 08 is equipped with a Kongsberg built LARS (the green structure over the forward moonpool) that operates two Saab Seaeye Leopard eWROVs. These eROVs operate only to a depth of 3000m therefore unsuitable for 7th arc depths. The Leopard ROVs are parked in the open under the LARS during vessel transits. Armada 78 08 has no AUVs embarked at this time (Thu 6th Feb).

    The fourth image of the selection is purported by the writer to show ‘Armada 7806 is equipped with a gantry over a moon pool in the deck and is capable of handling a ROV for wreckage filming, recovery and salvage operations’‘ That statement is emphatically incorrect: the Seaonics LARS rig constructed over the aft moonpool is intended to operate a, yet to be commissioned, geotechnical seafloor survey system. Armada 78 06’s Seaonics LARS is not intended to operate an ROV such as the 4500m-6000m rated Schilling or Kystdesign WROVs which, for example, are deployed on Armada 78 01. Besides, recording of a wreckage/debris field may be undertaken by Hugin AUV camera systems to produce a mosaic of the area.

    For the fifth image of the selection the writer describes ‘Armada 7806 had a blue container loaded this morning‘. Indeed it did. That blue container was loaded on Armada 78 08 in Mossel Bay, together with the transportable AUV container hangar, and ferried to Port Louis where the two containers were transferred to Armada 78 06 bringing its complement of AUVs to three in total. The blue container likely provides a supporting/servicing cabin for AUV operations.

    Where the MPA ‘Port Situation’ advice states, in relation to Armada 78 08, an activity ‘ld ships spares‘ this implies that Armada 78 08 will remove from Port Louis all parts dismounted from Armada 78 06 when installing the additional containerised AUV hangar and the support/service cabin.

    Direct, scheduled, Air Mauritius flights operate weekly from Kuala Lumpur only on Tuesday and Friday. Any Malaysia observer that may be assigned to the search would find the Friday service a timely connection to join Armada 78 06 departure on Saturday.

    Should any of the images linked above become unavailable, the selection is archived here

    It is not that I want Armada 78 06 to be the vessel that conducts the search, while others appear to be insistent that Armada 78 08 will be the vessel to conduct the search. It is that the observed configuration of Armada 78 06 confirms it is the vessel configured and equipped to conduct the search.

    Similarly, it is not that I don’t want some statistical data mining of the WSPR spot archive to discriminate the position of an aircraft, it is that observation of received WSPR signals and experimentation in conditions where RADAR-like characteristics might be reliably discriminated have shown that absolutely nothing can be described as ‘detectable’.

    And it is not that I want the flotsam known as ‘tataly-antsiraka’ and ‘broken-O’ to be identified as originating from Vestas Wind, it is that careful and deliberate observations show that those items of flotsam are not landing gear trunnion doors or anything related to a Boeing 777 but they are items that are consistent in construction with decking and the bulkhead separating the steering gear compartment and OBM station/sleeping compartment of the VO65 yacht Vestas Wind grounded on the Cargados Carajos Shoals, November 2014. One particular individual did not want those items examined by the VO65 designer and intervened to stop Farr Design doing so.

  47. Victor Iannello says:

    @Don Thompson: Once again, thank you for the facts.

  48. Don Thompson says:

    Armada 78 08 has updated its AIS with a ETA to match the destination broadcast by the vessel over recent days.

    Destination: Cape Town
    ETA: 2025-02-18T1900Z

    Bunkering and loading of equipment removed from Armada 78 06 (coded as ‘ld ships spares’ in MPA Port Situation information) must be almost complete in preparation for imminent departure.

    By a process of elimination only Armada 78 06 remains in Port Louis ready to make the voyage east to the 7th arc search area.

  49. TBill says:

    @Victor
    Re: 25-hr CVR
    Apparently USA’s NTSB 25-CVR proposal was officially adopted…”CVR technology is about to become even more useful thanks to a provision of the Federal Aviation Administration Reauthorization Act of 2024, which was passed by Congress in May. The act requires all commercial aircraft in the U.S. – both new and (surprisingly) existing – to be equipped with 25-hour cockpit recording devices.”

    Last January, I had submitted public comment in support of the NTSB proposal based on my experience following the MH370 accident. My comments below:

    https://www.regulations.gov/comment/FAA-2023-2270-0094

  50. airlandseaman says:

    Here is an informative report on the AA5342 PAT25 midair:

    https://bit.ly/413vmLZ

    There were several sources of altitude data for PAT25 ranging from 200 feet to 400 feet, noting however that the resolution for some was only 100 feet. The data from the CRJ FDR was more precise: 325±25 feet.

    To me, it sounds like PAT25 may have been slightly above the CRJ when the PF the CRJ saw PAT25 and tried to climb to avoid it. That certainly could make sense if you are down at 325 feet.

  51. Joseph Coleman says:

    Looking at the Berth plan for Port Louis, Armada 78 06 is set to Shift to QE tomorrow for Fueling.

  52. Don Thompson says:

    @Joseph Coleman

    Glad you could make sense of that ‘Port Situation‘ document, its formatting and presentation is appallingly poor!

    Yes, bunkering from Vivo Energy followed by departure. Whether departure is to direct to the 7th arc search area or a local test/validation area remains to be seen.

    There is a tropical cyclone moving east to west across the ocean but the forecast shows it waning.

  53. Don Thompson says:

    @Joseph Coleman,

    Actually, Armada 78 06 has just now (Friday, 20:20UTC) completed the move to Quay E.

  54. Joseph Coleman says:

    @Don Thompson

    I noticed the movement a little while after I posted. They are 4 hours ahead of us in that neck of the woods. They’re Starting nice and early, no messing about. Go Ocean Infinity Go.

  55. sk999 says:

    While we wait for 78 06 to make its next move, I am going to provide a link to some work on the radar that I did almost a year ago. Jean Luc Marchand, Capt. Blelly, and a certain trise5631 wrote a document regarding certain data extracted from the Geoscience Australia website that they interpreted as being the primary military primary radar data provided to the ATSB. Most importantly, they were able to reconstruct time stamps for the positional data. I compared this reconstructed track with the civilian ADSB, secondary, and primary radar data in more detail than they had done. Bottom line – the reconstruction looks good and now supersedes some work I had done several years ago.

    https://docs.google.com/document/d/1k8KJqb2zGfcA5eiztqMMsvaorldUUan2Lu-dXEVMr54/edit?tab=t.0

  56. TBill says:

    @Joseph Coleman
    @Don Thompson
    As I recall these new ships were said to run on ammonia as fuel?

  57. Don Thompson says:

    @TBill

    At present, the Armada 78’s generators consume ‘clean’ diesel.

  58. Ben John says:

    Armada A78 is a lean crewed vessel with facilities to host up to
    sixteen marine and payload personnel. It is prepared for future fuels with two large tank spaces, fuel cell rooms and a DNV Ammonia. Ready notation prepared for future upgrade. Checkout Armada 78m specs PDF web link:

    https://oceaninfinity.com/wp-content/uploads/2023/08/OI-ARMADA78-Spec-Sheet-2023.pdf

  59. Don Thompson says:

    airlandseaman wrote about the PAT25-JIA5342 collision.

    Mike makes an important point about the capability of technology to support decision making and post-flight analysis in an area of traffic density operating with such minimal separations. This is separate from the issue of DFDAUs, that may provide recoverable data, and DFDRs.

    The ATC monitoring of altitude via secondary surveillance radar or ADS-B relies on an altitude parameter quantised to 100ft or 25ft increments (per Eurocontrol monoradar and ADS-B target report specifications, & likely, the FAA’s equivalents). The eJet’s transponder was transmitting altitude at 25ft increments, the UH-60 at 100ft increments. This is a transponder capability. While the crew has visibility of a radio derived altitude at these altitudes, baro alt is the primary parameter used on the ground (enhanced surveillance options include geo/GNSS derived altitude).

    The Baltimore-Washington Helicopter Route Chart states:

    DESCRIPTION OF ROUTES AND ALTITUDES
    ALL ROUTES MAY BE ALTERED AT PILOT’S REQUEST OR AS DIRECTED BY ATC.
    ROUTE ALTITUDES ARE MAXIMUM

    Maximum altitude for Route 4 north of the Wilson Bridge, encompassing the site of the collison, is 200ft. However, the controller’s console view of altitude, presented on their screen, cannot accurately discriminate whether a helicopter is clipping boat wakes in the river or broaching the maximum altitude and risking a collision above 200ft. The controller has barely enough information to be a ‘controller‘ or at least an assistant, only lateral position is of adequate resolution to assist. That the UH-60 pilot had been given authority to continue and maintain visual separation isn’t an excuse, it’s a symptom of what the system delivered.

    It’s not as if the FAA hasn’t quantified and codified the surveillance systems’ performance, it has. In the Washington-Baltimore area, the fallback has been airmanship and Mk.1 eyeball. I remain to be convinced that’s a satisfactory solution, hence, much misalignment between the cheese slices.

    The thought that techbros propose, generally, to introduce unmanned air vehicles as flying cabs into urban airspace is just insane.

  60. Andrew says:

    @TBill

    RE: “The act requires all commercial aircraft in the U.S. – both new and (surprisingly) existing – to be equipped with 25-hour cockpit recording devices.”

    The retrofit requirement is surprising, given that neither ICAO nor the European regulator (EASA) adopted such a requirement in their own rules. In 2018, the NTSB recommended the retrofit of CVRs capable of recording the last 25 hours of audio, but the FAA did not adopt that recommendation in its 2023 NPRM on the matter. NTSB lobbying clearly prevailed!

    The FAA isn’t required to issue a final rule on the retrofit requirement until 16 May 2027. It will be interesting to see if the requirement survives or is ultimately dropped to harmonise US regulations with the ICAO standards and the regulations adopted by EASA.

  61. TBill says:

    @Andrew
    If CVR25 retrofit stands, does it mean only U.S. aircraft or does it apply globally? In any case my message is, those living outside USA may want to move forward on recommending retrofit for their specific countries, where it makes sense (for those aircraft where it is feasible to make/schedule the switch).

  62. TBill says:

    @sk999
    I am very interested having just downloaded your digitized for the purpose of putting it on a SkyVector flight path. The first thing I notice, looking at the IGARI turn back, the former digitized path drops as low as approx 200 kts which it suggestive of steep climb during turn, whereas the MTB shows more constant speed. For my purpose I do not really need the accuracy, but will you update your digitized path?

  63. Andrew says:

    @TBill

    The retrofit requirement would apply to US-registered aircraft. I think it would also apply to foreign-registered aircraft operated by US carriers under a dry-lease arrangement, for example.

  64. Barry Carlson says:

    Re: MH370

    Armada 78-06 departed Port Louis at 2025-02-08 17:41Z and at 1929Z was 20NM north of Mauritius heading 095&degT at 2KTS. The AIS is indicating “Offshore Mauritius”, but I suspect the SE trade winds are freshening and the easterly swell is rising as tropical cyclone “Vince” with central pressure of abt. 946hPa at 1800Z was at 20&deg 30’S 071&deg 12’E moving in WSW direction at abt. 10KTS with winds of up to 215KTS. TC Vince is expected to start turning slowly to a southerly track over the next 12 hours.

    If Armada 78-06 is bound for the MH370 LEP, I suspect that progress is going to be slow over the next 48 hours.

  65. Barry Carlson says:

    Update;

    Armada 78-06 appears to be undertaking trials of gear, and has recently launched a boat (fitted with AIS) that is operating close by, i.e. <500 meters.

  66. Barry Carlson says:

    An animated chart of meteorological conditions including predicted TC track(s) is at:-

    https://zoom.earth/maps/pressure/#view=-20.61,70.992,6z/model=gfs/overlays=wind

  67. Don Thompson says:

    @Barry Carlson,

    I expect that Armada 78-06 is undertaking trials with the third AUV that has been mobilised in Port Louis. At present it’s operating over reasonably shallow water, circa 500m depth, on the Mascarene Plateau so depth testing AUVs isn’t the objective (they’re not fresh out of Kongsberg, Horten).

    AUV launch requires the vessel to guide the AUV via acoustic comms and positioning to a final GNSS referenced INS alignment when the DVL establishes with the seafloor. The vessel’s indicated speed is consistent with that activity.

    This time around, AUV operations will be coordinated between the vessel and the RCC in Woolston. To the best of my knowledge, and that is only by observation over recent months, this is the first time that an Armada 78 has operated three AUVs. A78-07 and A78-05 have both operated with a single AUV, A78-06 previously with two.

    It’s been reported to me that a small craft has been observed close to A78-06, possibly the fast recovery boat that you and I have discussed off-line.

    As yet, there is no confirmation of the inception of a contract with Malaysia. That ‘milestone’ may be hastened with a demonstration of AUV operating trials by Armada 78 06.

  68. Don Thompson says:

    @Barry,

    Ah-ha, you’ve also noticed the small craft. Extended time drafting my comment while you posted a second + third.

    That small craft is configured on each Armada 78 that has operated AUVs (-05, -06, and -07). It’s used in recovery of the AUV back to the hangar’s stinger. Another reason Armada 78-08 was never the ‘assigned’ vessel for the search.

  69. sk999 says:

    TBill,

    That drop in speed was was an artifact of the DSTG’s modeling of the turnback (which I never understood). I just used the MTB version of the data (with time stamps) directly, and while my speed determinations are more sensible, there is still has a gap at the turnback that is not modeled correctly. So the MTB version of the data is the update, even if it may not be what you want.

  70. Barry Carlson says:

    @Don Thompson,

    Thanks Don for the update. Currently, at 08T0100Z I have the 78-06 heading 261°T at 1.7 knots – not far from the earlier position. The small boat hasn’t updated on Marine Traffic and it may be back onboard.

    NB. My previous mention of the max windspeed of TC Vince gave the speed in Knots, when it should have been 215 km/h. A significant difference!

    Apologies for the miscellaneous ‘&deg’ in my previous post.

  71. David says:

    @Barry Carlson. For interest, wave height, top section of:
    https://www.surf-forecast.com/weather_maps/Indian-Ocean?hr=3&over=none&type=htsgw

    Set to IST, UTC + 5½. Next 12 days projected.

    Also wind guide, though getting east the below is more detailed:
    https://www.surf-forecast.com/weather_maps/Western-Australia?hr=3&over=pressure_arrows&type=wind

    The 3rd selection, wind energy, is new to me.

  72. David says:

    @Barry. Last line, ‘wave’ vice ‘wind’.

  73. Don Thompson says:

    @Barry Carlson,

    After reviewing the Zoom Earth and earth.nullschool.net weather forecasts can I ask if you consider it feasible that Armada 78 06 could reliably navigate a way through the remnants of these two tropical cyclones?

    Assuming the trials activity concludes satisfactorily it may be Monday Feb 12th before the vessel departs waters around Mauritius and a further five-six days before crossing east of the E70º line of longitude where TC Vince is expected to be moving southward.

    More generally: the satisfactory conclusion of the trials activity would be a milestone indicating the vessel and equipment’s readiness to proceed to the 7th arc search site. That’s likely a good time for an exchange of signatures.

    One might think the combined minds of a lawyer, an ‘editor’, and a project manager would be sufficiently aware of how business is conducted to comprehend that a publicly voiced, tacit, MoU would progress to paper and signatures when the parties have completed preparations and are ready to deliver.

  74. Victor Iannello says:

    @Don Thompson: The fact that Armada 78-06 is conducting sea trials is a positive development. Despite the reported drama, it looks like OI and Malaysia are closer to a signed agreement for the search. The signing of the agreement just before the search is exactly what occurred last time.

  75. Barry Carlson says:

    @Don Thompson,

    Re – “… can I ask if you consider it feasible that Armada 78 06 could reliably navigate a way through the remnants of these two tropical cyclones?”

    Tropical Cyclones are at their most dangerous when tracking to the West, but on curvature i.e. turning to the South (S. Hemisphere) their intensity will slowly decrease and after reaching 25°S they are treated as an ex sub-tropical Low and will blend into the middle latitude pressure systems tracking Eastwards. Due to their relatively small physical size, the effect they have on the sea surrounding them is limited, though within 60NM of the TC centre, sea states can be high and confused.

    On passage from Mauritius to the MH370 LEP, the wind will be normally from the SE with a mean wind speed of 13 knots, less at night and up to 20 knots during the heat of the day. For vessels of the Armada class, a comfortable speed will in these conditions will probably be in the 8 to 10 knots region.

  76. Ben John says:

    Maybe Armada 7806 should consider leaving Mauritian waters soon & head due east for ~7 days (~1500nm @ 9kts) to avoid rough weather & forecast movements of Cyclone Vince & Cyclone Taliah then they could head safely southwest towards #MH370 search areas…

    https://www.windy.com/?-22.147,83.936,4,i:pressure

Leave a Reply