I was thinking about Flash, the superhero, or the little boy in the Incredibles.

There is one Yahoo answer that doesn't answer a lot. Especially, I don't think surface tension would help a lot for a human to run over water, I think one would have to build on the inertial effect of the water.

There is one empirical approach based on figuring out the speed at which bare-footed water skiing is done, but I wasn't able to find a decent number. Still the difference may be that a hypothetical runner would have to propulse herself over the water, which may or may not make the thing more difficult.

Hence, I raise the question here.

  • $\begingroup$ One problem I suspect will come up when comparing to bare-footed water skiing is that bare footed skiing pushes water in the direction of motion, creating a high pressure - and higher density - area making it easier to get the required thrust. Whereas when running the water is pushed opposite to the direction of motion, creating a low pressure area ahead of the test subject and making it even harder to get the required thrust. When running fast enough we might be able to jump over the low pressure area, but we will never get a high pressure area to land on. $\endgroup$
    – Poseidaan
    Nov 21, 2020 at 19:33

1 Answer 1


Unsurprisingly this has been the subject of several scientific papers. In particular Google for papers by J. W. Glasheen and T. A. McMahon. They studied the basilisk lizard, but their results can be extrapolated to humans. It's debatable how reliable such a large extrapolation is, but the result is that the required speed is so far beyond human ability that we can safely conclude it's impossible without some artificial aid.

There's a summary of the results from the papers in this article and a more general summary here. The conclusions are that you have to run at a speed of 20–30 m/s, which doesn't sound too bad, but you'd need to generate a mechanical power of 12 kW to do it. Trained athletes can just about manage half a kW, and most of us would struggle to generate 200 W.

  • 2
    $\begingroup$ Perhaps you should specify a kW of mechanical power, because it's quite common to expend over a kW of chemical energy - a difference that might matter to some calorie counters. $\endgroup$ Aug 21, 2012 at 12:16
  • 2
    $\begingroup$ kW is a unit of power not energy. I think a human body produces about 100W of heat at rest. Obviously this will rise if you're exercising, but I'd guess only by a factor of two or three. If you were generating 12kW you would be nicely roasted. $\endgroup$ Aug 21, 2012 at 14:22
  • 1
    $\begingroup$ 100W of thermal power would correspond roughly to a 100 lb individual sleeping, which burns 89 calories per hour. That same individual doing intense biking can burn 1000 calories per hour, or 1.2 kW, but the bike would see a mechanical power of 200 W. I think that the 12 kW in your answer is mechanical power, necessitating not one, but two orders of magnitude higher energy expenditure than the intense biking example. $\endgroup$ Aug 21, 2012 at 14:46
  • 4
    $\begingroup$ Ok, I think you have a fair point. If you were generating 12kW of mechanical energy you'd be be producing around 120kW of heat as well and you'd probably be too heavily charred to be worth eating. $\endgroup$ Aug 21, 2012 at 14:53
  • $\begingroup$ For reference purposes 20 m/s is about 72 km/hr. The fastest recorded speed for a human running is just under 45 km/hr by Usain Bolt and that was a peak speed during a 100m dash. So 72 km/hr is outside of the possible by quite along way. $\endgroup$ Apr 11, 2018 at 11:46

Not the answer you're looking for? Browse other questions tagged or ask your own question.