Mike Eichhorn publishes his latest paper on the MH370 floating debris analysis. This paper presents the results of a Two-Way Particle-Tracking Model (PTM) to detect possible crash sites of MH370. Mike emphasises the defined leeway coefficient has a strong influence on the detection of possible crash sites. Mike agrees with the statement of Dr. Alec Duncan at Curtin University’s Centre for Marine Science and Technology and concludes: “You have to accept the satellite data, but if it eventually transpires that there is a problem with that data, our location (near the Maldives) is the first place you would search.”
The paper can be downloaded here
@Mike Eichhorn
An excellent paper, that is well argued and nicely presented.
In my view, the satellite data is a given and therefore the crash location of MH370 is near the 7th Arc. In this case, your finding that the crash latitude is between 20°S and 28°S is very revealing. I admit that in my paper, I only considered the windage fully with the case of the Flaperon, but with a set leeway. All other MH370 floating debris was considered to be generic and drift similar to an undrogued drifter, again with a set leeway.
I fully accept the point that both you and Dr. Alec Duncan make, that if it eventually transpires there is something wrong with the satellite data, then the location west of the Maldives is the first place to look.
@Mike Eichhorn
Taking a closer look at your Figure 2, the most obvious MH370 crash location is near the intersection of the 7th Arc with the maximum fuel range, which you show is around 34°S. Of course, if the flight was conducted in a less than optimum fuel efficient manner, the crash latitude would not be as far south as 34°S. However, the fuel endurance until around 00:17:30 UTC indicates a near optimum fuel usage for a typical cruise altitude and cruise Mach speed, with a fuel endurance of 5.92 hours after the last Malaysian military radar point at 18:22:12 UTC.
It would be interesting to see the results for a Leeway Coefficient of 1.4% and 1.6% and how they compare with 34°S. David Griffin used a Leeway Coefficient of 1.2%, but in Table 1 of your paper you show other analysts have used Leeway Coefficients of up to 1.8% (Gao), 2.5% (Jansen) or even 3.39% (Nesterov).
https://www.dropbox.com/s/iu43ey84ykke0tn/MH370_Report_Mike_Eichhorn%20Figure%202%20Leeway%201.2%25.png?dl=0
@ Richard
Your assumption is correct: a higher leeway coefficient results in a crash site more southwards. The question is if such a high leeway coefficient is acceptable for the flat shape of the most debris found.
I will repeat the backwards-in-time simulations with a leeway coefficient of 1.4, 1.6 %. It is also interesting to analyze a crash site more northwards where the MH370-CAPTIO group is looking.
For this, can you send me four additional crash-sites more northwards for the forward-in-time simulations, please?
@Mike Eichhorn
As requested, here are four additional crash-sites more northwards for the forward-in-time simulations:
8.0000°S 108.0494°E
12.0000°S 107.2494°E
16.0000°S 106.1028°E
25.0000°S 101.9185°E
@Mike. Intuitively we imagine that the leeway factor is attributable to a combination of freeboard (projection above water) and draft or drag. No doubt would play a significant role and I believe various data from SAR experiments and experience demonstrate that drift speed is affected by an object’s windage.
However, I’m given to understand that this is by no means the whole story. The effect of the wind on water is to cause the entire surface film to start moving. The speed of movement, and the angle relative to wind alters as you go deeper, with very substantial changes over even the first 5cms depth. Thus even a spec of dust on the surface, with zero windage, still experiences leeway due to the movement of the surface film. Objects with very minimal draft (like flat sections of honeycomb composite structure) would barely project above the surface but still be subject to the high leeway velocity of the surface film (as much as 5% of windspeed for the top 10mm). As I understand it, there is a time lag as the surface film gains momentum and also a lag following change of wind direction. Higher wind speeds leading to rougher sea state lead to churn, which paradoxically breaks up this surface film movement and reduces effective leeway.
If the foregoing is correct (and I’d appreciate an expert commentary), then it may be that we are grossly under-estimating the leeway of the majority of debris items that have been recovered – most of which are of the flat, highly buoyant composite sort. We are also almost certainly wrong to assume zero offset to wind angle.
An expert discussion of the topic, the degree of uncertainty/disagreement involved, and its application to drift modelling can be found here
https://www.frontiersin.org/articles/10.3389/fmars.2020.00305/full
Another experimental example here, exhibiting undrogued drifter leeway of 2.3% of windspeed (also cites “typical” values uppermost layer 3%-5% from oil spill studies). Offset to the right of wind (in northern hemisphere) would be reversed in the southern hemisphere and angles in the region of ~20 (10-40) degrees relative to wind vector are commonly found.
https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1002/2015GL066733
Needless to say, a much higher leeway combined with directional offset to left of wind would predict that debris items originated further south. And would also help explain absence of debris in Australia.
@Paul Smithson
Many thanks for alerting us to the 2020 research on “Depth-Dependent Correction for Wind-Driven Drift Current in Particle Tracking Applications” by Mirjam van der Mheen, Charitha Pattiaratchi et al.
You raise some important issues around the assumptions, models and interpretation of the results of the many drift studies that have been published in relation to MH370.
I would be interested in hearing Mike’s views on this subject.
David Griffin may care to give an expert comment.
Here is a fairly comprehensive treatment of the theory, and state of knowledge/modelling techniques for near-surface currents. Again, this highlights the variability in mechanisms dependent on depth, with quite particular conditions pertaining to the very top layer (see discussion on “slippery water”!). To me, the clear take home from this is that deterministic application of ocean surface current models cannot be expected to have good predictive accuracy for drift of items floating in the top few cms. Overlaying that with a real-time linear leeway factor proportional to wind would improve the model, but even that fails to represent faithfully the more complex interaction of various factors at work that have different inertial characteristics and are affected by vertical mixing with changing sea state or diurnal effects.
factorhttps://www.tandfonline.com/doi/full/10.1080/1755876X.2021.1903221#
And finally, an empirical study on drift during high wind events: “The wind-driven component of the drifter velocities exhibits a rotation to the right with depth between the velocities measured by undrogued and drogued drifters. We find that the average wind-driven velocity of undrogued drifters (drogued drifters) is ∼3.4 %–6.0 % (∼2.3 %–4.1 %) of the wind speed and is deflected ∼5–55∘ (∼30–85∘) to the right of the wind, reaching higher deflection angles at higher wind speeds. Results provide new insight on the vertical shear present in wind-driven surface currents under high winds, which have vital implications for any surface transport problem.
https://os.copernicus.org/articles/15/1627/2019/