Monday, 11 June 2012

How much did that pterosaur weigh?

So a forever perennial problem in palaeontology is estimating the mass of an extinct animal. It's important to know for about a million reasons including ecology, but it's especially important in terms of locomotion and biomechanics. There are of course several methods to estimating the mass of extinct animals, primarily focusing on dinosaurs. Recently, this made the news with the so called 'Dinosaurs and lasers' paper, which you can read about here. Of course, I'm more interested in the mass of pterosaurs, which I'm working on for my MSc thesis (kind of). In flying animals, mass is even more important because it largely dictates whether an animal is able to achieve lift or not. If it's too heavy, it's not flying. Period.

So how is pterosaur mass estimated anyways?
There are lots of different pterosaur mass estimates out there, and several different ways of doing it, but I'm going to focus on 3 (more like 2) main ways.

The first method relies on estimating the volume of the pterosaur, and multiplying it by a density. This was first done pre-computer and relied on estimating the volumes of different portions of the body on paper (by estimating how much muscle there would have been). Bramwell and Whitfield (1974) did this first for Pteranodon and multiplied the volume by an overall density of 1000 kg/m3, to get a mass of 16.6 kg. A similar method was used by Brower and Veinus (1981), but with a density of 900 kg/m3, to get a mass of 14.94 kg. This method is good for a first try, but there are some holes in it. It doesn't account for the huge variation in density between regions of the body, and it relies heavily on using birds as a modern analogue for pterosaurs... which is a big problem. But more on that another day perhaps (or check out Witton and Habib 2010 for details).

The second method is a lot like the first one, but it uses a computer programme to estimate the volumes. Henderson (2010) applied this method to many pterosaur genera, by making 3D computerised models of each animal. This method allowed him to apply a different density to different areas, such as a much lower density to the neck and skull, where there is heavy pneumatisation. He was also able to subtract the volumes of cavities such as the lungs from the equation. This gave him a mass of 18.6 kg for Pteranodon longiceps. He verified this method using birds, and found it was pretty accurate. However, it also relies heavily on the idea that birds are a good modern analogue for pterosaurs... no good!

The third method is definitely the most interesting in my opinion (although I might be biased since it's similar to my project). In birds and mammals, there is a relationship between the skeletal mass and the total mass (Prange et al. 1979). That's pretty cool since with fossils, we basically only have the bones. Mark Witton (2008) used this method to estimate pterosaur masses by first estimating the skeletal mass, then applying the relationship seen in birds to get the total mass. To do this, he first estimated the volume of each bone in a pterosaur skeleton by simplifying it to assume it is a geologic shape. For example, a wing phalanx (the long bones that make up the fourth finger and therefore the wing in a pterosaur) is assumed to be the shape of a cylinder. Knowing the cortical thickness, length, and radius, the volume can be calculated. Once the volume is multiplied by density, you have the mass of the bone! This method is good since it relies precisely on what you have in the fossil record: bones. However, again, it relies heavily on using birds as a modern analogue, and it does some over simplification. For example, a cortical thickness of 0.7 mm was used for the entire bone, when the thickness can very from 0.6-2.4mm in one section. Big difference.

Without saying too much about what I'm doing, I'm working on a new, more accurate method of estimating bone mass using CT scans. This has shown me that the cortical thickness can very A LOT within one bone (specifically a phalanx), and 0.7 mm is a pretty small number for the whole bone. Basically, I'm finding that the previous mass estimates for single bones are underestimated. And underestimated by a fair bit (although not as much as I had thought at first, which I was struggling to explain so I'm glad I figured that out). Does this mean that Witton's entire skeletal estimate is underestimated? Not necessarily... More likely it is further evidence that we can't directly use relationships seen in birds to estimate things in pterosaurs. They aren't the same! Just because a bird did it one way doesn't mean a pterosaur did. They are different animals! Stop assuming they are the same!


Bramwell, C. D. & Whitfield, G. R. 1974. Biomechanics of Pteranodon. Philos. T. Roy. Soc. B 267, 503-81.
Brower, J. C. & Veinus, J. 1981. Allometry in pterosaurs. University of Kansas Paleontological Contributions Paper, 1-32.
Henderson, D. M. 2010. Pterosaur body mass estimates from three-dimensional mathematical slicing. J. Vertebr. Paleontol. 30, 768-785.
Prange, H. D., Anderson, J. F., & Rahn, H. 1979. Scaling of skeletal mass to body mass in birds and mammals. Am. Nat. 113, 103-122.
Witton, M. P. 2008. A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana Reihe B 28, 143-158.
Witton, M. P. & Habib, M. B. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PloS One 5, e13982.