A Swim Test for Spinosaurus: Challenging the Semi-Aquatic Hypothesis

I bet you’re thinking to yourself: really, another Rex Machina post on spinosaurs? Doesn’t this topic get old?

Look guys, I can’t help it. Spinosaurs are so cool, and we’re learning more about them every day. I’m actually having trouble keeping up with all the new studies being released, so sorry–there’s more where this came from. (Admit it, though–you love it.)

Lots of debate has gone on regarding Spinosaurus’ appearance, lifestyle, and preferred environment. In recent years, hypotheses arguing it was primarily aquatic have been popularized, but are far from being settled. A new study published in PeerJ by Donald M. Henderson, PhD–Curator of Dinosaurs at the Royal Tyrrell Museum–sought to look at these hypotheses and contemporary reconstructions of these theropods to better examine their feasibility. Using computer simulations to simulate their abilities to float and submerge in water, Henderson’s results challenge the current zeitgeist around Spinosaurus’ morphology and lifestyle.

Why Give Spinosaurus a Swim Test, Anyway?

In 2014, Ibrahim et al. published a groundbreaking new study on Spinosaurus. Their composite reconstruction of the theropod brought a new idea to the table–that Spinosaurus was an animal with short limbs, and spent most of its time in water. While a novel hypothesis, it has drawn much skepticism since it was first put forward. Multiple objections over the reconstruction built for the study have been raised by experts around the globe. Gregory S. Paul, independent scientist, author, and paleoartist, sums up these concerns well in his own response to the 2014 study:

[They] have taken an assortment of incomplete spinosaur remains found over an enormous territory of North Africa spanning 3000 km, found in sediments whose exact temporal correspondence is not certain, and presumed that they represent one genus and species. This despite the specimens being limited in overlapping comparative material, being of different ontogenetic stages, and there being some uncertainty concerning the individuality of some specimens…

Ibrahim et al. reconstruction of Spinosaurus.
A rendering of the composite Ibrahim et al. reconstruction. Courtesy Wikimedia Commons (CC BY 4.0)

These objections to Ibrahim et al.’s proposed Spinosaurus morphology, as well as the assertions about its lifestyle, prompted Henderson’s investigation. Besides the animal’s dense hind limb bones and associations with aquatic environments, its other known features don’t appear well-adapted for aquatic living. For example, its tail and dorsal sail were both likely stiff, supporting terrestrial balance and locomotion over traversing lagoons or rivers. As well, the placement of its femur in relation to its hips–would not have helped the animal swim. Yet, Ibrahim et al. still maintain that Spinosaurus was able to overcome these problems. Henderson’s study set out to resolve these disparities.

Model(ing) Dinosaurs

In order to test the Ibrahim et al. Spinosaurus, computer models of the animal were developed based on its illustration and description within that study. Then, using a slicing method Henderson developed, the 3D geometry of the model was built. Body tissue density was applied and modified in different regions of the model in order to reflect particular features or aspects of the animal. For comparative analysis, models of five other theropods were generated. These species were chosen to represent the overall diversity of body size in theropods. As well, each has an abundance of fossil material from which reliable whole-body reconstructions can be produced. These included: 

  • Coelophysis bauri
  • Struthiomimus altus
  • Allosaurus fragilis
  • Tyrannosaurus rex
  • Baryonyx (Suchomimus) tenerensis

Figure from the Henderson paper of each of the theropod models used in the simulation.
Each of the theropod models used for the study. A) Coelophysis, B) Struthiomimus, C) Allosaurus, D) Baryonyx (Suchomimus), E) Spinosaurus, F) Tyrannosaurus rex. From Henderson (2018) [CC BY 4.0]
Choosing this range of species would help in understanding how body size impacted the ability of theropods to float across time and lineages. The changes that occur in the body shapes of this group over time (the trunk region deepening and shortening, hind limbs becoming more massive, etc.) could better be accounted for and compared against Spinosaurus. This would help in identifying anything about Spinosaurus‘ morphology that was more adapted to aquatic life than its cousins.

Models of the American alligator (Alligator mississippiensis) and the emperor penguin (Aptenodytes forsteri) were also included. Their models acted as validation of the simulation’s intended use and function. If the computer-simulated models mirrored what we know of these animals’ flotation abilities, the simulation should work as intended. Once this validation was complete, Henderson ran all the models through the simulation to determine flotation and lateral stability when immersed in water. 

We All Float On, All Right…

When looking at the “non-floating theropods” (any of the dinosaur models that weren’t Spinosaurus), the simulation showed their centers of mass (CMs) were located just ahead of their hip sockets. As a result, they were balanced in the water, without much of a tendency to tip forwards or backwards. Similar CM positioning and the resulting balance ability also appear in the Spinosaurus model built. This contradicts arguments that Spinosaurus’ CM was further back because of its shortened limbs and wouldn’t be great at moving on land.

An artistic rendering of Coelophysis swimming. By vasix on deviantart.
An artistic rendering of Coelophysis wading across a swamp/lake with its young. Credit: Vasika Yasanjith Udurawane (CC BY-NC-ND 3.0)

The test results also showed that the floating states of all the theropods were independent of size. In other words, the proportions of the animals’ bodies exposed above the water line was the same across the models. The only exception was Coelophysis, whose model floated with the body tipped forward more than the others. Henderson explains the possible reasoning for this:

This may be related to two aspects of the body shape of Coelophysis. The much more attenuated, and slender axial body, with less of the body mass concentrated about the hips, and the much longer neck, which will not only represent a larger fraction of the total body mass, but in combination with the head, will also exert a stronger turning moment on the body.

Coelophysis was a thin, slender theropod with less of its body mass centered on its hips. This would have altered its ability to float in comparison with its cousins. In fact, both the Coelophysis and T. rex had to do some adjustment in the water, but could elevate their snouts above the surface.

Sails, Balance, and Legs

In analyzing the test results, Henderson looked to understand the impact of Spinosaurus’ anatomical features on flotation and buoyancy. To determine the effect of its sail, the Spinosaurus model was compared with data from the Baryonyx (Suchomimus) model. Henderson determined that, as Ibrahim et al. argue, the sails of spinosaurids stay visible while the animal is immersed in water. Additionally, the position and mass of the Baryonyx (Suchomimus) sail was shown to have a minor affect on its overall CM. Scaling up, it’s reasonable to think infer the same is true for Spinosaurus, eliminating its sail as a factor in its immersion and balance.

Image of an alligator submerged in water.
Alligators–masters of floating and general aquatic tomfoolery.

Henderson also modeled a cross-section of Spinosaurus to look at its ability to stay upright while in the water. It was compared with the alligator model, since alligators are expert floaters with the ability for passive self-righting. If knocked to one side, they can right themselves easily. In an animal argued to be primarily aquatic, it’s not unreasonable to anticipate similar abilities. However, the data shows otherwise. 

In the simulation, the Spinosaurus model showed it would have been easy to tip onto its side while in water. It would not have been able to passively self-right its body, either. As a result, it would have constantly needed to kick its arms and legs in order to remain upright. 

Henderson investigated this further by running tests on models of an Allosaurus hind limb. Ibrahim et al. have argued Spinosaurus’ dense hind limbs played a role in keeping the animal stable and submerged. If significant enough inferences could be made about the impact of Spinosaurus’ bone density on flotation in relation to the Allosaurus model, it might explain some of the inconsistencies the alligator comparison brought to light.

Image of a ship tipped over.
Some of the trouble Spinosaurus might have found itself in when trying to maintain stability in the water…

Looking at bone density and overall mass, Henderson determined that just one of Allosaurus’ hind limbs would have represented 4-5% of its overall body mass. As well, its bone density would have been about 50% more than water. In comparison, Spinosaurus’ leg would have had twice that mass with an overall body mass seven times greater. Assuming there were similar flesh-to-bone ratios in both theropods, any increased bone mass in Spinosaurus would have been an even smaller fraction of total body mass than Allosaurus. Therefore, Spinosaurus’ dense hind limb bones might not have had any impact on its ability to float and submerge in water.

So, it Wasn’t a Good Swimmer…?

Based on Henderson’s findings, there is more support to the idea that Spinosaurus–as is currently understood and reconstructed–was not particularly well-adapted for aquatic life. Ibrahim et al.’s reconstruction would have been able to float with its head above water, but would have needed to constantly move its limbs to stay upright. As well, becoming submerged would have required it to deflate its lungs significantly. In reality, it would have been unable to become fully immersed in water, which–when compared to extant semi-aquatic animals–would have seriously impacted its ability to effectively hunt underwater prey. This challenges many of the assertions laid out by Ibrahim et al. and opens the door for more investigation into the lifestyles of these ever-mysterious theropods. 

Thoughts from the Experts

In a weird stroke of fate (and/or luck), I had a brief opportunity to discuss this new study with Dr. Thomas R. Holtz, Jr., vertebrate paleontologist and University of Maryland lecturer. An expert in theropod dinosaurs, Dr. Holtz was gracious enough to provide the following thoughts on Henderson’s study:

The main takeaways from the new paper are:

–The anatomy of Spinosaurus is such that it probably was not an excellent swimmer.

–Even using the new proportions, it could still function as a bipedal strider.

–But (and this is clear in the graphs, but Henderson does not emphasize it) it would not have been as effective a terrestrial strider as the rest of non-avian theropods: it has only 1/2 to 1/3 of the muscle mass involved in the hind limbs as others.

So, I agree that the paper shows it wasn’t an excellent aquatic animal, but despite the text and the press, it ALSO shows it wasn’t an excellent strider, either!

Major issues, though, have to do with the model used. The rib cage is restored as being Allosaurus-like. But what little we know at present (from Stromer’s papers, for instance) suggests a rounder rib cage than most typical theropods. (Sadly, because the Ibrahim et al. paper was so brief, we don’t know too much about what their specimen shows!) So a newer model incorporating this information will be needed to see if it is equally unstable as a floater: we simply don’t know at present.

So, what are your thoughts on this new study? I’d love to hear them! Leave a comment below, and let’s start a discussion.

 

Header image courtesy of Francisco Delrio, DeviantArt (CC BY-NC-ND 3.0)

Special thanks to Dr. Holtz for his generosity in allowing me the opportunity to discuss this new study with him. You can find Dr. Holtz on Twitter @TomHoltzPaleo–go check him out, he’s a lot of fun, and very informative!

References

Henderson DM. (2018) “A buoyancy, balance and stability challenge to the hypothesis of a semi-aquatic Spinosaurus Stromer, 1915 (Dinosauria: Theropoda).” PeerJ 6:e5409. Accessed September 5, 2018. https://doi.org/10.7717/peerj.5409

Ibrahim, N., P. C. Sereno, C. Dal Sasso, S. Maganuco, M. Fabbri, D. M. Martill, S. Zouhri, N. Myhrvold, and D. A. Iurino. (2014) “Semiaquatic Adaptations in a Giant Predatory Dinosaur.” Science345, no. 6204: 1613-616. Accessed September 5, 2018. doi:10.1126/science.1258750.

4 thoughts on “A Swim Test for Spinosaurus: Challenging the Semi-Aquatic Hypothesis”

  1. Howd I get here?

    Anyway, the 2020 model – the “salamander” propelling tail shape, plus the “dorsal fin” crest for stability, plus the piscivorous teeth – I think says S. would have swum underwater to get its fish dinners, not on the surface. “Semi” aquatic only means it can crawl out onto land to lay eggs and travel between lakes.

    Same would go for Dimetrodon, Edaphosaurus and cousins. (I’ll put it in Turquoise Energy News #144, probably out tomorrow.)

    1. This post is out of date, which is why I haven’t addressed the 2020 study. What id say though is that the new Spinosaurus materials don’t necessarily invalidate Henderson’s findings; rather, they are important new data that needs to be tested within the validated models. Particularly, I’m interested in whether the new tail would have produced enough mass to keep Spinosaurus neutrally buoyant and upright in the water—-a major issue of contention w/in Henderson’s paper, along with the animal’s center of gravity. I imagine the 2020 findings essentially just slap the new tail onto the prior reconstruction done in 2014 by Ibrahim et al, which Henderson has directly modeled. Yes, the comparison to newts and other animals is compelling, but there are issues of scale and proportionality that complicate a direct 1:1 comparison. I definitely can say at this point that Spinosaurus was spending time in water, but the extent of that time and its Physical abilities in a fluvial environment need to be more thoroughly explored.

      1. Hello, As far as I read; there are several paleontologists; such as Mr. Handerson (in a recent interview he raised his disagreement with the tail’s proposed functionality), Michael Milbourne (he constructed the Tail akin to basilisks/sailfin lizards; John Hutchinson (the original individual that deducted the fsac-kk could be chimera), scott hartman (he, just as now; published recent post that agreeing the fsac-kk is a chimera), as well as David Hone. Additonally Mark Witton; even in his recent tweet he said there are upcoming publications that’ll challange the nazar’s newt tail swimming/propolsion theory.
        I want to ask; what do you think about these; as well as the nazar’s recent reddit post about the Spinosaurus could likely be 10 – 12 tons ?

        1. Hello Curious One!

          When it comes to the new Spino tail, I tend to skew conservative when discussing its morphology and functionality. My biggest hold-up right now is that the tail shape was essentially tested in isolation, and furthermore that I think Ibrahim was too quick on reaching their conclusion. The tail may be more hydrodynamic than other theropod tails, but to me that does not necessarily mean it was a tool for swimming. And they haven’t slapped that same tail onto the animal and tested its full swimming ability in context. I’m very intrigued how such a tail–if indeed analogous to newts–holds up as an exponentially scaled appendage. There’s also the fact that the new research doesn’t refute Henderson’s models to any high degree, so the “swim test” needs to be done again with these new features/dimensions added in.

          As far as the chimera status of the fossil and Ibrahim’s latest weight estimates, I’m holding off judgment. Given the history of this neotype, I’ve harbored suspicions that the chances of it being the same specimen dug up years apart may be slim. And if the specimen is indeed chimeric, then those weight estimates are off and need to be adjusted. I hope this answered your question!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.