Definitely an interesting read, and certainly adds to reconstruction of gaits and speed for extinct animals. I am not as familiar with quadrupeds as bipeds, but I did a quick run over for the math suggests that to achieve a speed of approx. 8 m/s a hippo with a mid-stance hip height of 1.2m would need a relative stride length of about 3.5, although this is assuming the use of the Thulborn (1990) running formula given the presence of an airborne phase. What is also quite interesting is the DF being near identical to elephants. While I am not very familiar the range of motion in the limbs of a hippo, if the high DF and trotting gait is to be maintained, the limbs will have to demonstrate either a broader range of motion, increase stride frequency, increase vertical oscillation, or some combination of these. A quick look at the material says that the stride frequency is what is being increased, and possibly vertical oscillation given the presence of an airborne phase. Notably, from my understanding, both increase loading and are fairly demanding on muscle. In regards to a higher stride frequency/cadence, the limb and associated muscle has to move through a stride cycle more frequently and thus incur a greater energetic cost. Similarly with vertical oscillation, while pushing off the ground demonstrates the presence of a reasonable GRF, if that energy is directed in a more vertical direction rather than horizontal, it is doing less to move the animal forward, and in turn that energy is lost as additional muscular action must be done to help stabilize the animal when it lands and in turn launch itself during the next stride. Gist of it is pretty much that for the type of locomotion the hippo is engaged in, for work done and energy expenditure compared to result (in this case the animal's speed), it is not as efficient a result of energy in/work done vs result when compared to other animals. Evidently though, in terms of absolute energy in and trying to go as fast as possible this is the type of locomotion a hippo engages in.
Ultimately this looks like a case of an animal using the fastest kind of locomotion available to it even though it is less efficient than simply switching to another type of gait which may place less demand on muscle; this inability to switch gaits being the result of some kind of inability on the part of the animal. Examples of "grounded running" or race walking immediately come to mind. Both concluding hypotheses presented by the authors are quite interesting, and looking over the publication I would guess that both hypotheses are correct and are limiting the hippo's top speed. This certainly an interesting example of animal doing something that a relatively simplified model would have a hard time predicting, although I would also be careful with a lot of my discussion here since I have not taken the time to quantify some of these observations. I know that the Tyrannosaurus speed discussion was mentioned, and I am plenty happy to discuss what this publication adds to the discussion, as well as just general discussion on findings, methodology, gaits, or whatever may be relevant (although I am assuming it should take place in the corresponding thread?).
Best,
Adrian
Some references I found that are useful, at least in regards to vertical motion and oscillation
Adams, D., Pozzi, F., Willy, R. W., Carrol, A., & Zeni, J. (2018). ALTERING CADENCE OR VERTICAL OSCILLATION DURING RUNNING: EFFECTS ON RUNNING RELATED INJURY FACTORS. International journal of sports physical therapy, 13(4), 633–642.