Wednesday, October 5, 2011

Dinosaur Revolution: Anhanguera

This is actually a cross-post from H2VP, but it seems quite relevant here, as well. Those of you that watched episodes 3 and 4 of Dinosaur Revolution (which aired exactly one week ago) saw the sequence focusing on the large pterodactyloid pterosaur, Anhanguera. This was one of the sequences I had the most input on, so I thought it might be fun to chat briefly about some scientific background that inspired the sequence.

Parental care in pterosaurs?
The sequence opens with a mother Anhanguera visiting her nest of offspring, whom she then kicks out of the test for their first flights (these end poorly for the first two babies, but the "hero" character survives to fly another day). We have relatively little evidence regarding the specifics of parental care in pterosaurs. What we do have is good evidence that pterosaur babies were able to fly very early in life, and that the eggs were of a soft-shelled structure, which implies that the eggs were buried in foliage rather than brooded in the manner of birds. This manner of egg-laying alone does not tell us much about parental care - "leathery" eggs are laid by some taxa that do guard young (crocodilians) and many that do not (most squamates, though some of those guard nests and young, too). However, the fact that baby pterosaurs were so well developed, and likely able to fly early in life, is at least suggestive that there was not an extended period of parental care. Baby pterosaurs probably set off from the nest relatively early (possibly immediately). Check out Darren Naish's blog post from February on pterosaur babies and eggs for more.

Pterosaur Locomotion
There are three major types of motion shown in the sequence: ground locomotion, launching, and flight. As it turns out, the first two of these are really the same "mode" of movement. After speaking with David Krentz, he and I thought it would be interesting to show the baby Anhanguera hopping in a saltatorial fashion. There are no trackways that show this mode of locomotion in pterosaurs, but we also don't have any trackways that can reliably be mapped to ornithocheirids yet, and the limb proportions of ornithocheirds like Anhanguera are consistent with a saltatorial method of movement. This observation is noted in the paper that Mark Witton and I published in 2010. It is published in the highly acclaimed, open access journal PLoS ONE, and is freely available here.

The takeoff mechanism features the quadrupedal launch model that I proposed in 2008, and which was further used to make predictions about pterosaur ecology by Mark and I in the PLoS ONE paper. Julia Molnar generated a wonderful animation of quadrupedal launch for Anhanguera, and it has appeared across multiple venues, including National Geographic. She was subsequently kind enough to make it freely available on YouTube. I have inserted the video below. You can also pull it up by clicking here.

I have written about quadrupedal launch on other web resources previously, so I won't belabor the point here. In short, takeoff acceleration in animals tends to be generated mostly by the walking limbs, rather than the wings. As such, takeoff is really a form of running or leaping (usually the latter). The strengths of the limb bones in bending and torsion, particularly with regards to the moments sustained for leaping, are therefore highly indicative of launch mode. Pterosaurs turn out to be much more bat-like in this regard than bird-like: they had forelimbs which were much stronger than the hind limbs across a wide range of body sizes. By contrast, large birds have stronger hind limb elements (particularly the femur) when compared with the forelimb elements. Giant pterosaurs, such as Quetzalcoatlus, had very long, thin hind limb elements, which argues against a bird-like launch. However, because pterosaurs walked on their folded wings, as well, the incredibly robust forelimb musculature and structure could provide most of the launch power (and sustain the resulting forces) during a quadrupedal launch. Since pterosaurs were quadrupedal while moving on the ground to begin with, this is actually the most simple model, as well. Modern bats, particularly vampire bats and New Zealand short-tailed bats, use a quadrupedal launch.

There are a host of other problems with a bipedal launch in pterosaurs, including problems with angle of attack of the wing, trailing edge flutter, Wagner effects, insufficient height and time, and pitching instabilities. Depending on interest, I may do a summary of these observations at a later date.

There is not much to say specifically about the flight patterns. One nice thing was that the flapping amplitudes used were pretty reasonable. The wing cycles are probably a bit too large in some cases (particularly the Quetzalcoatlus models that do flybys), but it's usually much worse. The problem here is that large animals actually tend to fly with more shallow wing strokes, especially if they have high aspect ratio wings. This tends to make a distant albatross etc. seem a bit smaller to us than it really is, and the same happens when doing pterosaur models - they just don't look as huge if you model them correctly. In Clash of the Dinosaurs, I tried very hard to get the animators to reduce the flapping amplitude of the Quetzalcoatlus model to no avail - there was a general feeling from those working on the show that the giant size didn't come across with the lower-amplitude wingbeats that are predicted by anatomy and fluid mechanics. Oh well, such is life.