Water takeoff in a pelican (part 2 with reference to pterosaur water takeoffs)

Mike Habib, famous for his pterosaur forelimb take-off scenario, sent a short note recently for which my reply here is aided with illustrations.

M. Habib wrote, “Incidentally, pelicans do not launch using wing power. The launch using a series of pushes from the hindlimbs, which are essentially leaps. Their launch is basically saltatorial. Most of the power is provided by the final leap, as evidenced by the fact that the last “splash” is the largest – they displace more water, not less, as the feet push, so the wings are not picking up more power as they go. Animals do not launch like airplanes.”

Earlier I posted a pelican take-off composite photo that stitched together several scenes from a video from YouTube here (redux in figure 1) in which a Sea of Cortez pelican became airborne from a floating configuration.

Pelican take-off sequence from water.

Figure 1. Pelican take-off sequence from the Sea of Cortez. Click to enlarge. Full video linked above in text.

Thankfully there’s another YouTube video newly posted in slow-motion here that is even more instructive.

In both videos, you’ll note that the pelican slowly unfolds its wings (as in figure 1) preparing for flight. With one mighty downstroke it becomes airborne (#4 and #5 in figure 1, note the belly is clear of the water). Beneath the waves no doubt the feet have provided some thrust, as we can see by their position as they rise above the waves in the first (slow-motion) 7 seconds (#5 in figure 1) fully extended posteriorly. The wings continue to beat in sync with the feet cycling back, pulling the water backward, using every limb in its power to achieve greater airspeed. But note from the start (#4 in figure 1), the pelican is airborne and flying. Wings are dry. Feet are too, except for occasional dips (leaps? or simply gaining traction?). To my eye the foot motion looks more like a pull (swimming stroke, paddling, thrust from the entire interaction of foot and water), than a push (as in leaping, thrust only from the moment of last contact with the substrate).

Not sure, given the relative sizes of the feet and wings, how M. Habib can say, “pelicans do not launch using wing power.” Looks to me like they use everything they have, with the majority contribution from the enormous wings (and who knows how much added headwind  from the breeze). Certainly it takes five and a half foot contacts with the water before the pelican has enough airspeed to rise above the “ground (in this case water) effect” in which the wings benefit from more lift in closer association with the ground (water). And the last splash was the least, not the greatest in both videos.

The key seems to be: one downstroke and everything but the feet leave the water.

All of these musings on pelican water takeoff refer back to my thoughts on convergent water take off in pterosaurs first mentioned earlier and duplicated below (figure 2).

Pterosaur water launch

Figure 2. Ornithocheirid pterosaur water launch sequence in the pattern of a pelican launch. LIke ducks, geese and pelicans, pterosaur probably floated high in the water. Here the wings rise first and unfold in an unhurried fashion, keeping dry and unencumbered by swirling waters. Then the legs run furiously, like a Jesus lizard, but with such tiny feet, they were not much help in generating forward motion. The huge wings, however, did create great drafts of air, thrusting the pterosaur forward until sufficient airspeed was attained, as in the pelican. Considering the pelican’s abilities, perhaps the relatively lighter pterosaur could have arisen free of the water on the first downstroke, without much fuss from the hind limbs, especially aided by a light marine headwind.

Another video of a skimming pelican here brings to mind Nyctosaurus and Pteranodon feeding activity over the Niobrara Sea.

And yet another video of an origami Pteranodon here. Amazing. I better stop there. YouTube can be fascinating and distracting.

9 thoughts on “Water takeoff in a pelican (part 2 with reference to pterosaur water takeoffs)

  1. That slo-mo pelican is quite clearly pushing off with its feet while the wings are being positioned for the down stroke. It’s just mostly hidden behind the wave, but the body’s motion betrays what’s happening.

  2. I agree. That provides the first inkling of forward motion, as in many birds that simply leap into the air and start flapping. The first downbeat lifts the body off the water, which is key. It all works together.

  3. Dave, if you look more carefully in the video, you’ll see that the wings are coming down *after* the feet push in each stroke phase. They are not timed together. In this case, the animal launched at the penultimate foot touch; the last one is basically and after-thought and that’s why the splash is smaller, but the primary launch occurred one stroke earlier, and that’s the largest water displacement.

    You said that “To my eye the foot motion looks more like a pull (swimming stroke, paddling, thrust from the entire interaction of foot and water), than a push (as in leaping, thrust only from the moment of last contact with the substrate).”

    This is not a meaningful distinction. A leap does generate all of the force from the last moment of contact. If this were so, leapers would not have long distal limb segments. They have long limbs so that the force can be exerted on the substrate over a longer period of time, decreasing power requirements for the same leaping height/distance (the area where the muscle goes has to be robust to withstand muscle forces; these segments tend to be proximal and often shorter). The pelican is using an altered swimming stroke to produce a series of leaps. Drag-based pushing from the feet is essentially the same thing as leaping from a solid substrate, except the animal loses some energy to slippage (in fact, I have a single equation that accurately models both types of leaping – the same math does both because the actions are essentially the same).

    Yes, the pelican uses everything it has, but the data available indicate that the wings provide the minority of the power until the immediate post-launch phase where the wings take over. That switch is very quick, but the initial launch power is still technically provided by the feet. To understand why, we have to look at bit at fluid dynamics:

    1) The fluid forces are proportional to the densities of the media. The wings are moving air, the feet are moving water. Water is 800+ times denser than air, so all else being equal, the feet are effectively 800x larger during water launch.

    2) Wings do not produce much lift starting at zero speed; they have to build circulation first. This requires either a lot of room, or a high (albeit brief) acceleration. Animals typically use the latter approach, but this requires a propulsion system other than the flapping wings to jump-start the system. Drag gets animals out of the gait the fastest. Pelicans use the feet to push first for the same reason that fish use drag to fast start and lobsters use drag to jet backwards.

    3) There is a way to allow the “enormous wings” to provide a lot of the initial launch power, and that’s by having them push on the water directly to launch. Some ducks do this, as do osprey. These animals water launch more rapidly (and often steeply) and gain more wing clearance. They are quad launching.

    Most birds biped launch from the water as the pelican does; note the pelican has much larger feet and more robust hindlimbs than, for example, an ornithocheirid pterosaur.

    This is not something you can just eyeball. You have to understand the dynamics. We’ve spoken about this before, at length, but apparently I have not been clear so you will have to explain to me what part of the dynamic is confusing. Good resources on animal takeoff and the basics of animal flight include Colin Pennycuick’s 2008 “Modelling the Flying Bird” and Robert Dudley’s 2002 “Biomechanics of Insect Flight”.

  4. Question: What sort of numbers are you attaching to your water launch scenario? You have some vague associations in the figure legend about feet moving “fast” and the wings producing “great drafts of air”, but no solid numbers on what this means. I therefore can’t tell what kinds of forces you are attributing to each limb during each phase.

  5. There is some doubt that the pelican feet move first underwater in the slo-mo pelican video. If so, then only by ten percent or so. To my eye the feet and and wings are moving simultaneously, with both reaching maximum extension, downstroke at about the same time, with the feet slightly leading, granted. It’s a rhythm. But the sound effects for both motions would largely overlap.

    On that point here is a video of some wood ducks swimming really fast. http://www.youtube.com/watch?v=txdSJhsjBCc

    The pelican will never achieve the airspeed necessary for flight by just paddling, or in this case paddle/galloping.

    The pelican is also going to stay in the water if it doesn’t produce downdraft and back draft with its wings. And the evidence shows it does this with one wing sweep.

    So the question is, can a pelican without feet take off from water? Someday a YouTube video will show this, but today videos of pigeons without toes are present there.

    I think we can agree a pelican without wings will never become airborne.

    The numbers I will never have for you. That may comfort you, but we’re really arguing about how many angels can stand on the head of a pin. Even so, theory is fun and I look forward, Mike, to your paper showing the graphs you propose and I endorse.

    —> Mike, Don’t make me break out the Jesus lizard, which, as everyone knows, actually walks on water — without — the benefit of webbed feet, which many pterosaurs seem to have. And, as everyone knows by now, pterosaurs are lizards.

  6. “There is some doubt that the pelican feet move first underwater in the slo-mo pelican video. If so, then only by ten percent or so. To my eye the feet and and wings are moving simultaneously, with both reaching maximum extension, downstroke at about the same time, with the feet slightly leading, granted. It’s a rhythm. But the sound effects for both motions would largely overlap. ”

    Yes, the timing of the feet and the wings matter. The feet lead slightly because they have to generate the bulk of the initial force (and therefore the burst acceleration required for jump-starting the wings).

    “The pelican will never achieve the airspeed necessary for flight by just paddling, or in this case paddle/galloping. ”

    What data are you basing this on? What is the required speed, and how much force do the hindlimbs generate? You cannot just through around assertions like this without either measurements to back them up, or at least a solid prediction from well-tested theory (i.e. numbers).

    More to the point, you are still missing what I’ve explained about the dynamics of launch. It’s not about reaching steady state speed on a runway as an airplane does. Animals launch largely by producing a very brief, powerful acceleration (or a quick series of them, in the case of a pelican). This provides a moment of high velocity (which is all they need), a powerful acceleration to jump start circulation on the wings, and clearance for the first wing stroke cycle. You’re trying to make the pelican an airplane and assuming that it is a simple summation problem (force of legs + force of wings to get the magic speed number) and that’s not how it works.

    “The pelican is also going to stay in the water if it doesn’t produce downdraft and back draft with its wings. And the evidence shows it does this with one wing sweep. ”

    Wings are not paddles. They do not work by pushing on the air like an oar, and contrary to your assertion, the birds for which the forces have been measured do not launch with a “wing sweep”. They launch with an unloading of the hindlimbs. See work by Earls, Tobalske, Rayner, etc. Even hummingbirds get more than half the force from the hindlimbs. Other birds generate upwards of 80-90% of the launch force with the legs, even though “by eye” the wings engage just after the feet. To understand why that is, you have to understand fluid dynamics.

    What is ironic is that an animal can do *exactly* what you are suggesting (combine drag-based propulsion from the wings and feet to get a combined super-push) *if* the wings push on the water, instead. In other words, if it quad launches, which is exactly what some ducks do, and what osprey and sea eagles do. It is also, roughly speaking, what I suggest pterosaurs may have done to get out of the water. So I really can’t figure out where your problem with that idea lies, except that you seem to really like bipedalism.

    “Mike, Don’t make me break out the Jesus lizard, which, as everyone knows, actually walks on water — without — the benefit of webbed feet, which many pterosaurs seem to have. And, as everyone knows by now, pterosaurs are lizards. ”

    Basilisks really aren’t relevant here. They do not “walk on water”. They briefly sprint on the surface by pushing on a surface tension held concavity, made possible by their small size, long feet, and special fringes on the toes (not webbing, as you correctly noted). It is not a paddling dynamic, and if the hindlimbs of a basilisk are truly submerged then they cease to run on the surface and resort to swimming (which they do rather well). I worked for a herpetology department that raised basilisks for 6 years; the dynamics of their water running do not work they way you think. I recommend becoming familiar with this paper: http://www.pnas.org/content/101/48/16784.long

    Incidentally, there is a way for a pterosaur to shed ring vortices from the surface in a manner very roughly similar to what basilisks do (not using surface tension, but rather more typical drag-based propulsion), but with the wing finger joint, not the feet – which again, would make it a quadrupedal (and in this case, saltatorial) takeoff.

    What we are arguing about are physics, which can be calculated. Application of theory is not the same as an idle philosophical speculation this is not “how many angels can stand on the head of a pin”, but rather “what do the physics allow?”. That may not tell us exactly what did happen, but it can tell us what is physically plausible, and what is not.

    The results are the numbers, not the graphs. I think your chart idea was a great one, but that doesn’t make any difference to the results. It might, however, make them more clear to a general audience.

    • I said:
      “The pelican will never achieve the airspeed necessary for flight by just paddling, or in this case paddle/galloping. ”

      You said:
      What data are you basing this on? What is the required speed, and how much force do the hindlimbs generate? You cannot just through around assertions like this without either measurements to back them up, or at least a solid prediction from well-tested theory (i.e. numbers).

      I say:
      Observation: No bird takes off from water by simply paddling while extending its wings to the flight configuration. No bird can paddle as fast as it can fly.

      You said:
      More to the point, you are still missing what I’ve explained about the dynamics of launch. It’s not about reaching steady state speed on a runway as an airplane does. Animals launch largely by producing a very brief, powerful acceleration (or a quick series of them, in the case of a pelican). This provides a moment of high velocity (which is all they need), a powerful acceleration to jump start circulation on the wings, and clearance for the first wing stroke cycle. You’re trying to make the pelican an airplane and assuming that it is a simple summation problem (force of legs + force of wings to get the magic speed number) and that’s not how it works.

      I say:
      Observation: No bird can leap from the water using its legs alone (webbed or not) without using its wings for added thrust. I understand that leaping makes one airborne for a short time, during which the wings can be extended. On that we agree. But then, the second half of the leap brings one back down to the substrate, except if wings are employed to take over and accelerate greatly, that initial velocity, which isn’t much initially. I know, there’s no numbers here, but the pelican does become airborne after that initial downstroke. The subsequent foot touches to water might accelerate the pelican, but only minimally compared to the large sweep of the wings. I’m saying the pelican could fly without those foot touches. They are inconsequential. You’re saying, I think, that those foot touches are essential.

      I said:
      “The pelican is also going to stay in the water if it doesn’t produce downdraft and back draft with its wings. And the evidence shows it does this with one wing sweep. ”
      You said:
      Wings are not paddles. They do not work by pushing on the air like an oar, and contrary to your assertion, the birds for which the forces have been measured do not launch with a “wing sweep”. They launch with an unloading of the hindlimbs. See work by Earls, Tobalske, Rayner, etc. Even hummingbirds get more than half the force from the hindlimbs. Other birds generate upwards of 80-90% of the launch force with the legs, even though “by eye” the wings engage just after the feet. To understand why that is, you have to understand fluid dynamics.
      I say:
      You’re talking about the moment before the wings unfold and downstroke. I’m talking about the moment after.

      You said:
      What is ironic is that an animal can do *exactly* what you are suggesting (combine drag-based propulsion from the wings and feet to get a combined super-push) *if* the wings push on the water, instead. In other words, if it quad launches, which is exactly what some ducks do, and what osprey and sea eagles do. It is also, roughly speaking, what I suggest pterosaurs may have done to get out of the water. So I really can’t figure out where your problem with that idea lies, except that you seem to really like bipedalism.

      I say:
      If a bird or pterosaur spanks the water and that helps it get airborne I have no trouble with that. Some birds fly underwater, then become airborne. http://www.youtube.com/watch?v=ZtowVki4Ij8. Note they use their feet to change direction, but their wings continue to provide thrust underwater. When going in a straight line, the feet fold as if in air. The wings are the superior thrust providers in this case. If a pterosaur is going to get its wings clear of the water to become airborne (which is the ultimate goal) then forelimb leaping with the wings folded (they can’t unfold until your pterosaur is a sufficient height above the sea in your scenario) just isn’t going to produce anything more than splashing, as Mark Witton’s illustration shows. There’s no happy ending there. Opening up the pterosaur wings like a pelican, whether free of the water or floating upon the surface, appears to be a more valid alternative because it lets the wings catch a head wind and, undaunted by the drag of water, flap at a sufficient rate to become airborne.

  7. “No bird takes off from water by simply paddling while extending its wings to the flight configuration. No bird can paddle as fast as it can fly.” –Dave

    Response: You are missing the point. It doesn’t have to paddle as fast as it can fly, because the problem isn’t getting up to steady state flight speed. The problem is getting a powerful acceleration from rest along with clearance for the wings. I emphasize, again: *animals do not launch like airplanes*

    “But then, the second half of the leap brings one back down to the substrate, except if wings are employed to take over and accelerate greatly, that initial velocity, which isn’t much initially. I know, there’s no numbers here, but the pelican does become airborne after that initial downstroke. The subsequent foot touches to water might accelerate the pelican, but only minimally compared to the large sweep of the wings. I’m saying the pelican could fly without those foot touches. They are inconsequential. You’re saying, I think, that those foot touches are essential.” — Dave

    Response: Apparently my point is still being lost. Wings do not produce much force from a standing start, unless they can be used as drag-based propulsion elements by contacting the substrate directly. Otherwise, the hindlimbs have to do most of the work. And yes, I am saying that the initial pushes from the feet are critical. I offer, as evidence, actual measurements of launching birds (again, see Earls 2000 or Tobalske et al. 2004). Now, this has not been directly measured yet for water launch, but the same basic dynamics should hold, and based upon real, careful, experimental results, the hindlimbs produce 80-90% of the launch force. You can “say” that the foot pushes are inconsequential all you want, but the data indicate otherwise. You cannot just make numbers up. If you are going to argue that theory would predict something different in this case, then you must provide a quantitative counter-theory. It is entirely unconvincing to eyeball what you think is happening from some YouTube videos.

    “You’re talking about the moment before the wings unfold and downstroke. I’m talking about the moment after.” –Dave

    Response: I am talking about the entire launch cycle in land birds, and at least the early launch cycle in water birds. What you are still missing is that the initial unfold and downstroke cannot produce substantial force without a “kick start”, which is provided by leaping. As a result, it is the leap that truly initiates the takeoff, and not the wings, unless the wings also push on the water. If you are talking about the phase during which the wings have taken over locomotor power, then you are referring to the climb out *after* launch, which is different. (Note that climb out can effectively begin with the feet still close enough to touch the water; that’s not relevant here. The initial force production to initiate the launch cycle is the critical component, and that will require a push off the substrate. The best option is to use the wings and feet together to push on the water, but most birds are not anatomically capable of this. Some of them are, however, and do so. Fishing bats also do this, and pterosaurs were anatomically capable of it. I suggest they did so).

    “Some birds fly underwater, then become airborne. http://www.youtube.com/watch?v=ZtowVki4Ij8. Note they use their feet to change direction, but their wings continue to provide thrust underwater.”

    Response: Yes, aquaflying is most interesting (see my paper in Biological Journal of the Linnean Society, in 2010, on this subject), but that’s not the same dynamic as a quad water launch. Still, nice vid :)

    “If a pterosaur is going to get its wings clear of the water to become airborne (which is the ultimate goal) then forelimb leaping with the wings folded (they can’t unfold until your pterosaur is a sufficient height above the sea in your scenario) just isn’t going to produce anything more than splashing, as Mark Witton’s illustration shows. There’s no happy ending there. Opening up the pterosaur wings like a pelican, whether free of the water or floating upon the surface, appears to be a more valid alternative because it lets the wings catch a head wind and, undaunted by the drag of water, flap at a sufficient rate to become airborne.” –Dave

    Response: As best I can tell, you’re just making things up on the fly at this point. You can’t intuit your way through this. Mark’s illustration is a schematic image of what the kinematics would roughly look like; it is not a result of it’s own and has no bearing on the physics. The ultimate goal is not to just get the wings clear of the water. If that was the end game then no birds would quad launch from water. The ultimate goal is to get a large initial acceleration, along with some height, and *then* get the wings clear.

    You can speak colloquially about “happy endings” but you have provided no data or calculations at all here. What do you mean to “catch a head wind”? How do you know it would flap at a “sufficient rate to become airborne”? You still seem to be missing my repeated point that the wings are not terribly useful from a standing start. If you are going to make suppositions about flapping rates and force production then you need to CALCULATE the results of your scenarios. Or, if that is not feasible, you need to find experiments that have measured the effects and provided quantification that way. You also need to consider angle of attack, initial floating position, and contact areas. All of these problems should be addressed quantitatively.

    As luck would have it, I have done these calculations. As for ground launch, a quad water launch is better in every respect than a bipedal one except for the added upstroke time. Therefore, we have a relatively simple, testable question: would a marine pterosaur (say, Anhanguera for example) have enough air time on the last bound to raise the wings and begin a wingbeat cycle? We know how upstroke time scales with wing shape and body size in birds and bats, so we can get a good rough estimate for a pterosaur. We know about how long the animal would be in the air because ballistic equations are easy, and we have a rough idea of the power production by the muscles of the brachium and pectoral complex thanks to work by Marden in the mid-90’s, plus a decent idea of the contact areas for the feet and folded wings.

    Answer: Calculations indicate that it would have about 3x the time it needs for the first upstroke, being conservative; so plenty of leeway. At that point, everything else is solved: the wings will get good circulation right away because of the high quad-push acceleration (something Robert Dudley noted when I had him check my reasoning on this; he’s written literally an entire book on animal flight biomechanics), and the height gain is much greater than for bipedal launch.

    Not sure what else there is for me to say at this point. I don’t understand your fixation with bipedalism (yes, I know you think pterosaur ancestors were bipedal; that doesn’t really matter here). Unless you can provide a quantified counter-argument, I’ll have to consider quadrupedal water takeoff to be the more plausible model for now.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.