Question about Tyrannosaur running speed estimations paper and its questionable findings

[Chaos Soahc]

Hello,

I found an interesting paper from 2018  

Usami, Yoshiyuki, and Ryuta Kinugasa. "A possibility of fast running of Tyrannosaurus." DEStech Transactions on Engineering and Technology Research 118.amma (2017): 126.

It found rather high speeds for tyrannosaurus to be possible, and I don't know what to make of it, with speeds opf 10-13 meters per second being possible for decently sized T. Rex specimens. Does anyone have any thoughts or insights on what to make of this paper, is its findings completely absurd? There are some interesting points made at least particularly about elastic elements in the legs reducing stress during locomotion and potentially increasing its top speed. can anyone help me on this one????

[Mickey Mortimer]

Where did you find this available online? The only pdf I found was corrupt and closed the page immediately.

I did find this earlier paper on the topic by the first author though- https://www.scitepress.org/papers/2014/50317/50317.pdf

Usami, 2014. A possibility of fast running of Tyrannosaurus rex by the result of evolutionary computation. Proceedings of the International Conferenceon Evolutionary Computation Theory and Applications. 145-152.

Mickey Mortimer

[Franco Sancarlo]

Dowloaded here it is

--
You received this message because you are subscribed to the Google Groups "Dinosaur Mailing Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to DinosaurMailingG...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/DinosaurMailingGroup/d927e6d2-7cde-4b1d-9545-8c916e81d0bdn%40googlegroups.com.

[Mickey Mortimer]

Nope. That's the single-authored 2014 paper I mentioned and linked to.

Mickey Mortimer

[Chaos Soahc]

I found it on google scholar here maybe I can send a link https://scholar.archive.org/work/ok35xh6wivdfdihloyolqgmrr4/access/wayback/http://dpi-proceedings.com/index.php/dtetr/article/download/13347/12869 

[Chaos Soahc]

Please tell me if it worked, and if it did work please tell you're thoughts on it?

[Franco Sancarlo]

Seems to be almost the same paper

To view this discussion visit https://groups.google.com/d/msgid/DinosaurMailingGroup/f57d7191-5bef-4ba5-9806-539e09aa742bn%40googlegroups.com.

[Gregory Paul]

A quite good paper that has gone ignored, although it is in an obscure location. 

The best conclusion of the analysis is that it is not possible to accurately restore the top speeds of potentially fast running extinct exotic animals. 

We can without simulations predict the speed of an extinct gracile ungulate being around and about the speed of similar ungulates of today. 

Anything that has rectigrade elephantine limbs that have short nearly immobile hindfeet like sauropods, stegosaurs, unitatheres are going to be about as fast as elephants, which in timed races can do a walk-run amble without a suspended phase of about 15 mph. That includes the juveniles, watch the John Wayne flick Hatari! 

The fatal defect of trying to use biomechanics to estimate top speeds is that one is simulating a simulation, not observing the actual animal. The latter being the only way to see what the top speed really is or was. Simulations have to make assumptions that may not match the reality of the animal, which is a product of evolutionary selective factors that may have allowed it to exceed human expectations and calculations. 

I could not convince Paul McCready, who developed the first successful human powered aircraft, that Quetz could fly at over 70 kg. Now we know those things weighed 100s of kilograms. 

Who would via simulations estimate that some marine mammals could dive thousands of feet if they were all extinct? After all, mammals are high metabolic rate beasts that have to surface frequently to get all that oxygen they need. 

There is that vid a paleo has posted showing three big white rhinos trotting along, one trips and does an aerial flip that would back a gymnast proud, and gets up and trots along. So much for all those nonsensical calculations that Tyrannosaurus would break its bones if it ran too fast. 

If an animal had flexed limbs with a long foot on a mobile ankle that it could push off with into a suspended phase it could fast run. How fast, who knows. We do know that arch predators are animal athletes built for speed. We know that the giant avepods were built like running birds and mammals, not elephants. They may have had a lot of white fibers in their leg muscles, including that big honking caudofemoralis, to produce intense ambush burst speed. So trying to outrun one probably not the best idea even if a top Olympic athlete. Giant ceratopsids and hadrosaurs should have been in the rhino speed range (again, see Hatari for the best full galloping rhino sequences ever taken, they are online) -- http://www.gspauldino.com/Forelimb.pdf; http://www.gspauldino.com/GaiaNeoceratopsian.pdf). Ceratopsids seem adapted for short burst charges.

GSPaul

To view this discussion visit https://groups.google.com/d/msgid/DinosaurMailingGroup/CADJSBGqPNOi_k%3DHB9oZrSfgtMVVhTzvWtp_Xr-9TnO__RVWa-g%40mail.gmail.com.

[John Hutchinson]

The paper (2014 one and 2018 one self-published) has some pretty huge flaws, already pointed out in earlier studies we did (and even buried in the same Usami papers). A key one is that the centre of mass is placed at the hip joint. This is an implausible assumption that, as noted back in 2004, biases the analysis to obtain the lowest possible muscle mass and thereby maximal speed estimate. But this COM position is rejected by plausible reconstructions. The COM was in front of the hips. So as the 2014 paper acknowledges, with a more realistic COM the speed estimate drops a lot. There are other glaring problems like the very generous maximal limb extensor muscle mass, which is implausible for an animal whose skeleton makes it clear its hindlimbs were relatively less muscular than an ostrich's. This was all addressed from 2002-2011. The Usami paper does not show that models cannot estimate maximal speeds. It is better seen as an example of why it's a bad idea to model something using inputs that are unrealistic and bias the analysis toward extreme results and conclusions.

It's a model, and that might make people feel it's easy to dismiss, but models are everywhere in science. Philosophers of science have made it very clear what a model is. Even when not called a model explicitly. Phylogeny, morphometrics, diversity/disparity curves and extinctions, any statistics, etc. All are models. One of the few things palaeontologists do that is not really modelling is descriptive osteology. Climate change or epidemiology are models used today, for example. Models are used for a reason; mainly because reality is complex. Models, if quantitative, make it very explicit how reality is being quantified. They expose assumptions that an intuitive approach glosses over. They address complex interactions of variables that intuition can fail to grasp or can conceal. Their assumptions can be discussed and evaluated. They can be compared with reality on an apples-to-apples basis to evaluate them. Qualitative functional morphology is also a modelling approach but less explicit and more subjective than quantitative biomechanics; all too easy to be slippery with the logic or even non-reproducible. But one has to understand a model to evaluate it. Which inputs are least well known and how important they are. In the end, with these simple models (before 2017's much more complex one by Sellers et al. in PeerJ) the results hinge on two things that assumptions impact: how large the "flexor" moments were about the limb joints, and how much extensor muscle mass could there have been for each joint. The former is most greatly impacted by posture and COM position. Posture has plausible ranges, and COM has been extensively analysed and bounded in numerous studies based on comparative anatomy and 3D body shape. Other inputs like muscle moment arms and pennation angles have been bounded pretty confidently, or even have little impact on results. It's not hard to bound reasonable vs. unreasonable boundaries. Simple models make it clear, too, what is missing from them (because of complex or poorly understood phenomena), and that can be evaluated in follow-on studies. Anyway, with repeated studies the basic conclusions of our 2002 study continue to be upheld; T. rex was not very "fast" (a highly subjective number, but a lower bound of 11m/s for "fast" can be justified; 20m/s would be extremely fast but still often suggested in the past). All estimates have ranges of uncertainty since these are fossils. Nihilism that a method cannot work is not valid; it emerges from a lack of understanding, and probably a bias (worst case examples today are climate-change-denial and anti-vaccination agendas). Overall, it's the same old story; anecdotes don't falsify the results.

--John Hutchinson

To view this discussion visit https://groups.google.com/d/msgid/DinosaurMailingGroup/347890855.150332.1778350294189%40mail.yahoo.com.