Could (large) pterosaurs really fly?

Perhaps it might seem a little strange that the first real post of a research group whose sole purpose is to study flight in pterosaurs is asking the question “could pterosaurs actually fly?” There is, however, a little logic to this when some of the more recent headlines about pterosaurs in popular science media read “Giant Pterosaurs Couldn’t Fly, Study Suggests” (National Geographic) and “Were pterosaurs too big to fly?” (New Scientist). Thus it seems this is a good pace to start.

As with any media coverage it is always advisable to get to the source of the headline and the study in question was that of Professor Katsufumi Sato and his extensive list of co-authors, published in the online journal PLoS one entitled “Scaling of Soaring Seabirds and Implications for Flight Abilities of Giant Pterosaurs.” For those not familiar with this article their research centres around attaching accelerometers on to five procellariiform species (encompassing shearwaters, petrals and albatross) and recorded the frequency of wing flaps during take off and brief episodes of flapping flight between periods of gliding. This is itself is a novel idea and the paper should be rightly credited with being one of the few examples where data was collected from wild animals in the field, a rather formidable task and one that, without some technological advancements, would be quite impossible. Higher frequencies were associated with take off while lower frequencies were recorded when the animals were actually airborne. To quote from their abstract “In these species, high and low flapping frequencies were found to scale with body mass (mass^-0.30 and mass^-0.18) in a manner similar to the predictions from biomechanical flight models (mass^-1/3 and mass^-1/6).”

So far so good, but were Sato et al. really trying to argue that giant pterosaurs couldn’t fly? By plotting the regression lines of both high and low flapping frequencies against body mass an intersection point is found (as should be clearly illustrated in the figure below) corresponding to the point at which an albatross-like animal would lack the power margin to keep itself airborne if not supported by favourable wind conditions. Any animals heavier than this would simply be unable to flap fast enough to increase their flight speed. The magic number, the point of intersection between the two lines, was 41kg (402.21N), which corresponded to a wing span of 5.1m by their estimates. Thus, their conclusion goes that pterosaurs either larger or heavier than this would be unable to remain airborne in the absence of either favourable wind conditions or thermals.

High wing beat frequencies (red) associated with take off and lower wing beat frequencies (blue) during flight (figure from Sato et al 2009).

While this data here is certainly useful for putting an upper limit on flapping flight in birds (or more specifically procellariiforms), particularly as previous values have been estimates, e.g. 12 – 15 kg (Pennycuick 1986), we run into a number of problems when applying this data to pterosaurs. Firstly the giant Quetzalcoatlus was not the only pterosaur to exceed this wing span or weight, in fact many “medium-large” taxa including Pteranodon, Anhanguera and Coloborhynchus, to name a few, also grew to wing spans of up to 5m and, based on fossil material that I have seen, some almost certainly got much larger than this.  This point has also been noted before in the figures of Templin and Chatterjee (2004) where larger pterosaurs would have certainly occupy a range in excess of the “mass limit for continuous level flight.”

Secondly while Quetzalcoatlus is the iconic “giant” pterosaur its bauplan differs remarkably from that of the albatross (aspect ratio of ~8 versus 19, wing load of 224N/m^2 versus 155N/m^2) and so a comparison between the two is difficult not least because Quetzalcoatlus has been suggested to thermal soar (rather than dynamically soar), a point Sato et al acknowledge. Other larger pterosaurs, however, were all “albatross-like” and found in marine settings leading to the suggestion that they would have had to inhabit areas of special environmental conditions in order to fly (e.g. the roaring forties of today). Their global distribution in the fossil record and finds on shorelines or sheltered settings rather than exposed on bare islands like albatross today suggest that this was not the case.

Perhaps the most obviously point to consider (and I’ve saved it until last) is that pterosaurs are not birds, even the ones with wing profiles that resemble them. By this I mean that pterosaur muscle configuration, physiology, air-sac development and wing structure, to name a few, are all very different to birds. While all flying animals are constrained by a number of universal factors (e.g. drag reduction, lighter bodies) differences in physiology and structure radically alter the way in which animals fly. Birds and bats for example do not share a maximum or minimum size for flapping flight and in turn insects are radically different to both of these again. Consider that the size range in bats is between 1.9g and 1.5kg while in birds it is much larger 1.5g – 15kg (Altringham 2001). As bats increase in size the mechanical power need for flight also increases along with wing inertia and thus they sacrifice both agility and manoeuvrability which in turn influences what the animal can catch. Bats also appear to link breathing and echolocation pulse emission to the wing beat cycle and it is possible that at a certain size (as wing beat frequency decreases) this period may become too infrequent for effective prey capture (see Altringham 2001). Kilpatrick (1994) also reported on the structural differences between birds and bats where the breaking stress of their humeri was found to be 125MPa and 75MPa respectively. These are just a couple of the many examples that can be readily found and differences in structure and function explain why we do not find albatross-sized bats, Pteranodon-sized birds (Argentavis magnificens aside) or bird-sized insects outside the Carboniferous when environmental conditions were significantly different from today.

Without an in depth comparison of birds and pterosaurs to account for their differences in morphology, physiology and structure it is highly problematic to suggest that pterosaurs above a wing span of 5m simply couldn’t fly. All large pterosaurs are fully equipped for flight as their global distribution attests to and even in the more terrestrial taxa there is no indication that the wings were becoming redundant. It seems appears unlikely that these forms were restricted to areas of very specific environmental conditions (with the possible exception of really big pterosaur that must have required thermals).

Thus in summary while I disagree with the conclusions of Sato et al, generally applying data from a number of procellariiform birds directly to larger pterosaurs, it is nice to finally have some experimental data on the upper range of size of certain birds. The take home message from this is, of course, that if this data is not generally applicable to pterosaurs then it raises the interesting point that whatever pterosaurs did to attain these large sizes, they did it very different to albatross-like birds.

Altringham, J.D. 2001. Flightt.. In: Bats, biology and behaviour. Altringham, J.D. (eds). Oxford University Press, 262pp.

Chatterjee, S. and Templin, J., 2004. Posture, locomotion and paleoecology of pterosaurs. Geological Society of America Special Publication: 376, 1-64.

Kilpatrick, S.J. 1994. Scale effects on the stresses and safety factors in the wing bones of birds and bats. Journal of Experimental Biology: 190, 195-215.

Pennycuick, C.J. 1986. Mechanical constraints on the evolution of flight. Memoirs of the California Academy of Sciences: 8, 83-98.

Sato, K., Sakamoto, K.Q., Watanuki, Y., Takahashi, A., Katsumata, N., Bost, C-A., Weimerskirch, H. 2009. Scaling of Soaring Seabirds and Implications for Flight Abilities of Giant Pterosaurs. PLoS One: 4, 1-6.