[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Great in the air, not so good underwater

is an
There is an alternate hypothesis that accounts for the larger size of flapping fliers in the past, without resorting to educational conjecture about interesting novel aerodynamic systems.

The density hypothesis is indeed interesting (see further notes below). However, the launch differences that Jim and I were discussing are not as conjectural as they might seem. Forelimb-assisted launch is reasonably well known in bats at this point, and the different launch modes of modern birds have known implications (though some were only rigorously tested in the few years). It is quite apparent now that pterosaurs were quadrapedal. The forelimb-assisted launch is, granted, much less certain, but there is significant evidence to suggest that it was used (not to mention being quite intuitive). I'll leave it to Jim to reply to that subject more specifically (if he chooses), since he has worked on the problem. I'm putting together a project to more quantitatively test the forelimb-assisted launch ability hypothesis for pterosaurs. So more on that in the future.

Simply put, a denser atmosphere increased the lift available to these organisms. Even the seemingly small density increase (12-15%) concomitant with the increased partial pressure scenarios of O2 at various geologic periods that have been proposed since the early 1990’s may have significantly increased aerodynamic performance in still air; even if O2 related increases in available mass-specific power are ignored. [Names to google include R. Berner, J.B. Graham, R. Dudley, and R. Seymour, among others.]

All good names, and it is indeed a long-running and interesting hypothesis. I am skeptical, however, that such a difference in partial pressure of O2 is the best explanation for the large size of pterosaurs. This is the case for several reasons:

1) It is not clear how much of a performance increase a 12-15% difference in O2 partial pressure would really have for large soaring vertebrates (see below). I think the advantage may have been exaggerated to some extent.

2) If the atmosphere did, indeed, make a significant difference, then I would expect to see noticeable and measurable differences in planform between modern birds and advanced Cretaceous forms. So far, I have not seen anything to suggest that the rules were different for Cretaceous birds (for example, we don't see higher wing loadings in arboreal enantiornithine analogs of modern passerines).

3) Adaptive differences in launch actually seem more parsimonious, in a sense, than evoking atmospheric effects. It seems odd to me to assume that large-bodied pterosaurs, despite being anatomically capable of using forelimb-assisted launch, used a less efficient launch system instead (less efficient for them, in any case). By contrast, simply invoking the same launch cycle used by some living quadrapedal flyers solves the problem in one fell swoop (more or less).

I can offer some sketchy, unpublished empirical observations--

1). I found that if the wings of carpenter bees (X. virginica) are
trimmed such that the lengths of the primaries and the secondaries are equal to 90% of the original lengths
of the secondaries, bees cannot even (< 2 seconds)
achieve lift-off.

It's been some time since I looked at insect wings. Can you remind me again what the markers are to distinguish "primaries" and "secondaries" in insect wings? (or is that another terminology for forewing and hindwing?)

In air of
~1.5-2 atmospheres pressure, some even (incredibly!) manage controlled,
sustained (>10 seconds) hovering and lift-off is effectively 100%.
(I use “hovering” here in the strict, still-air, flight-kinematics
sense, which has no relation to the stationary soaring observed in
larger birds.) I offer this to show that hyper-dense air can, as intuition would indicate, significantly reduce constraints relative to wing-loading (lift) and wing shape (control).

Very cool experiment. I do note, however, that 1.5-2 atmospheres is a lot of pressure (relatively speaking).

2). When testing the vertical ascent capabilities of various drosophilids,
I found that on average, 90% of a given sample of wild-type D.
melanogaster would starve if required to fly (at sea level pressure)
up a vertical tube (~6.5 cm inside diameter, ~1.8m in height,
and fluon-coated to prevent “cheating”) to obtain food. Pressurizing the
tube to only 1.2 atmospheres reduced the starvation rate to ~10%. My
(tentative) conclusion was that, due to unsteady-state effects, the
response of aerodynamic performance to flight medium density was
non-linear, at least in small fliers.

Seems like a decent (if somewhat speculative) conclusion to me. I suspect, however, that the major difference you saw is likely to be limited to small-bodied flyers. In particular, I agree that the relationship is non-linear, and I suspect the curve plateaus at higher Re. Just like you pointed out, the unsteady-state dynamics are probably playing a huge role in your result.

As small fliers are more
vulnerable to viscous effects than large fliers, it is my opinion that
birds may receive even larger relative benefits from small density
increases than insects.

I'm not sure about this. Granted, since birds fly in a more inertial-dominated flow regime, it seems like density changes could matter quite a bit. However, I suspect that the small insects are more sensitive to Re changes. For one thing, I would think that they should respond more strongly to alterations of drag regime than a bird. Small insects get more useful momentum flux out of drag, for one thing (though Drosophilia flies around a Re of 100, where vortex generation is still the primary source of momentum flux and L/D ratios are still well above 1) The other issue, of course, is that tracheal diffusion rates are a rate-limiting step for insects. As R. Dudley emphasizes in his text on insect flight, increases in oxygen partial pressure have a profound physiological effect for insects. Vertebrates get a boost as well, but to a more limited degree.

Testing the load-carrying capacity of
pigeon-sized birds at increased pressures might clarify the
relationship between size, flight medium density and performance in flapping fliers,
just as variable-density wind tunnels were once used to manipulate the Re
number when designing airplane wings. Astonishingly (to me), this has apparently never been done, although it appears straightforward.

That does seem like a good idea.

On the peer-reviewed level--

R. Dudley, P. Chai, and others performed experiments within the last decade with hummingbirds and reduced flight-medium density which in my opinion can only be classified as elegant. Of particular interest is confirmation of the intuitive perception that wing-stroke amplitude and frequency increase as flight-medium density decreases. The corollary is that increased atmospheric density reduces wing-stroke amplitude/frequency, with obvious implications for take-off scenarios in large animals with long wings.

Yes, they're quite nice studies. One advantage of the particular approach they took (using a Helium mixture) is that viscosity was kept relatively constant at varying air densities. The question, however, is how much of a change in stroke amplitude and/or frequency larger flyers demonstrate when density increases. I suspect that while atmospheric changes would have had an effect on ancient flyers, most of those changes would be compensated for with relatively minor changes in planform and/or kinematics. The difference in maximum observed size between pterosaurs and birds is well over 2x however, and that seems like a larger difference than can be accounted for with a 12% O2 partial pressure jump.

In any case, very interesting stuff (thanks for sharing the information on your insect flight experiments!), and there is obviously plenty of work still to be done. I suspect that large pterosaurs would have been perfectly viable in today's conditions, but we'll see what further data shows.


--Mike H.