This article has been hanging around my neck like an albatross. I’ve been sent it by several readers, but it’s not terribly exciting, and I’m reporting on it more as a duty (and because it’s on a felid) rather than to impart any scientific information. (UPDATE: after writing this piece, I decided that the results are more interesting than I imply here.]
The results, taken from a new paper in Nature (reference below) are about how cheetahs (Acinonyx jubatus) run when pursuing prey. There’s not much new save for the quantitative data, but the paper’s publication (and the ensuing publicity in places like The New York Times) appears to derive largely from its use of gadgetry: special collars fitted on cheetahs that give data on their location (within 0.2 m), their speed, and their acceleration.
Here’s an example of the publicity in the journal itself (read the “speed test in wild cheetah” box). As Matthew commented to me:
BTW you might want to poke fun at the splash that Nature had on its front page (see attachment). Talk about
the bleedin’ obvious. As a pal responded on Twitter, “I knew that when I was 8 years old”. They’ve changed it since.
Matthew adds this: “Even the opening two paras of the accompanying news item are really stupid.” Here they are, and yes, we sort of knew this already:
The cheetah crouches in the undergrowth. When a young antelope strays a little too far from the herd, the cheetah explodes out of the bush — and, with a burst of speed unrivalled in the natural world, brings down its next meal.
Or so we have assumed. But the first study to collect data on the animal’s movements in the wild reveals that, contrary to popular opinion, a cheetah’s sheer speed is not its only weapon when it comes to hunting. Its success as a predator also hinges on its lightning reflexes and its ability to accelerate faster than a Ferrari.
Well, the Ferrari part is interesting, I suppose, but what’s new here? I’ll distill it down to a few points.
First, the methodology. The authors attached special, solar-powered collars to five wild cheetahs in Botswana, and monitored 367 runs at prey over a period of 17 months. For each run they tracked the animal using GPS technology, and measured its speed and acceleration. Here are the highlights:
- 94 of the hunts, or 26%, were successful. Most of the prey were impalas (75% of prey items), but cheetahs also took warthogs and Thompson’s gazelle.
- Cheetahs hunt by stalking. Starting from either a slow walk or a stationary position, they accelerate rapidly. This much we already knew.
- The average run length was 173 meters, and the average frequency of a run after prey was 1.3 times per day. That means that they have a successful hunt about once every three days.
- They can run fast: the already-published top speed is 29 meters/ sec (104.4 km per hour or 64.8 mph), substantially faster than racing speeds for greyhounds (18 m/sec), horses (19 m/sec) and humans (12 m/sec; Usain Bolt’s 100 m record). The collar data substantiate this, showing a top speed of 25.9 m/sec (58 m/hr or 93 km/hr). That’s important because it’s hard to measure running speed in captive cheetahs (they’re trained to follow a bait over a predetermined distance), so we didn’t know how close the “captive” speed was to speeds attained in nature. The mean top running speed during a chase, though, was about half that: 14.9 m/sec (53.6 kph or 33.3 mph).
- The rates of acceleration and decleration are extreme. The acceleration rates aren’t expressed in any terms I really understand: the highest was 100 W/kg (can some reader translate that in to m/sec/sec?). That rate is about four times that of Usain Bolt’s acceleration in his world-record 100m dash. But they can speed up by 3 meters/sec and slow down by 4 m/sec (6.7 mph and 8.9 mph respectively) in a single stride. They can accelerate, then, faster than a Ferrari (at least according to the New York Times), and decelerate from their average chase speed to a standstill in less than four strides. That’s pretty amazing.
- The deceleration often occurs right before turning, enabling them to maneuver with great dexterity, something you’ll know if you’ve seen videos of cheetahs hunting. (I’ve put one below.)
- Finally, their ability to speed up and slow down rapidly, as well as to corner tightly, comes, according to the authors, from their “ridged footpads,” nonretractable claws (see picture below), their flexible spines, and their tremendous muscle mass. According to the authors, “the locomotor (limb and back) muscle accounts for 45 ± 4% of body mass in captive cheetahs. . . major propulsive muscles such as the hamstrings. . at the hip and gastrocnemius at the tarsus have 64% and 60% longer movement arms, respectively, than in the greyhound. . “
The photos below show the collars, the forces impinging on hunting cheetahs, their nonretractable claws, and the posture they assume during deceleration.
After rereading the article and writing this, I guess the findings aren’t so humdrum after all!
Finally, you can see it all in action in this National Geographic video of a hunt:
___________
Wilson, A. M., J. C. Lowe, K. Roskilly, P. E. Hudson, K. A. Golabek, and J. W. McNutt. 2013. Locomotion dynamics of hunting in wild cheetahs. 498:185-189.
“…acceleration rates aren’t expressed in any terms I really undertand: the highest was 100 W/kg (can some reader translate that in to m/sec?”
I can’t, but ‘(m/sec}/sec’ maybe?
Cool article. I’m confused by one bit in the summary, however:
…suggests that the frequency of a run was (367 runs / 5 cheetahs / 17 months / roughly 30 days) approximately 0.14/day. That would make for some hungry cheetahs.
But later it’s said:
Is the number of total runs off by an order of magnitude? Or are the two quotes talking about different things? Is there not a typo there, and what’s at issue is cheetah post-worship jogs?
The authors say that the collar didn’t work properly on many of the runs, and perhaps they didn’t analyze data from many runs.
“…acceleration rates aren’t expressed in any terms I really undertand: the highest was 100 W/kg (can some reader translate that in to m/sec?”
A watt is a kg·m2/ s3, so it’s 100 m2/ s3. It’s not so much a measure of acceleration as power per unit mass.
And that then gives us the acceleration, when we know their speed – divide the m2/s3 by m/s, and you get the acceleration in m/s2. eg at 3m/s after the first stride, they’re accelerating at about 33m/s2, or about 3.4g. If, that is, the power per unit mass is what actually goes into their kinetic energy, rather than the total power their muscles are producing – a lot of which will end up as heat, of course.
That is impressive. I wonder what kind of lateral acceleration they can achieve.
Haven’t had time to read and watch everything yet. Were the collars equipped with accelerometers or just GPS?
The steep bank angle in photo “b” above suggests that their cornering acceleration exceeds one gee.
In the BBC podcast mentioned below the researcher says its about 1.5 gee
That is as good, or a little better, lateral acceleration than a top MotoGP rider on a top MotoGP bike.
It’a hypothesized that the limiting factor in cheetah runs is net heat production, not available power from musculoskeletal equipment nor delivery rates of fuel and oxygen by cardiovascular equipment.
IIRC, it was heat dissipation while hunting on the veldt that was the adaptive advantage in our being hairless apes.
/@
Wikipedia offers four competing theories of human hairlessness, none of them a slam-dunk winner as far as I can see, and all of them supported largely by just-so stories rather than concrete evidence.
Darwin apparently favored sexual selection as the explanation.
Yeah… I should have put an “is posited as” in there…
/@
This brings to mind the riddle, “What is the difference between a cheetah and a cheater?”
Cheetahs, while impressive hunters, as mentioned in the article, only bring down prey about once every three days. But this doesn’t mean they get to eat their prey. After bringing down an animal, they’re exhausted and are unable to defend their kills against lions, leopards, hyenas, or other opportunistic predators. Add to that the shrinking habitat and their overspecialization, cheetahs are likely to be doomed to extinction. As you can see, cheetahs will never prosper.
That’s a creative answer, but it’s not the traditional one – and it’s treason that doesn’t prosper.
The answer is: The cheetah is a running cat and the cheater is a cunning rat.
“Treason doth never prosper: what’s the reason? Why if it prosper, none dare call it treason.”
– John Harrington (1561-1612, of Kelston, a courtier, known as “Queen Elizabeth’s saucy godson”, popularly known as the inventor of the flush toilet)
That’s the quote I had in mind – I was just too lazy to look it up for a correct citation.
The trouble for me is that none of those measures are expressing an acceleration, but are proxies that needs knowledge of cheeta traits. W/kg is a power density and m/s a speed.
Maybe a Ferrari owner can do this without putting constraints on a cheetah. (And considerably faster to boot.) The paper is paywalled (of course) but the video tells me what I need:
– At top speed a cheetah step frequency is ~ 3.5 steps/s.
Assuming constant step frequency to get an oom estimate, a cheetah can accelerate with ~ 3 * 3.5 ~ 10 m/s^2 and deaccelerate with ~ 14 m/s^2.
– Check: Mean top running speed during chase is given as ~ 15/s^2. They can then decelerate to a standstill in a little over 3.5 steps, or 4 strides.
– As comparison, a Ferrari can accelerate to ~ 100 km/h in ~ 3.9 s. [ http://en.wikipedia.org/wiki/Ferrari_California ] That is an average acceleration of a = v/t ~ 100/4 km/h/s ~ 25 km/h/s ~ 7 m/s^2.
– Assuming instead constant top acceleration, without the step frequency constraint, to get the average power density. Given an acceleration of 10 m/s^2, the time to hit top speed at ~ 100 km/h ~ 28 m/s is t = a/v ~ 10/28 ~ 0.4 s. The distance covered is then s = 1/2*at^2 ~ 5*0.4^2 ~ 0.8 m.
Check: A distance of 0.8 m would be ~ 1-2 steps or a time of ~ 0.3 – 0.6 s, in this estimate.
The work done is W = int(F*ds) = ma*s, giving the power density P = W/t/m = a*s/t ~ 10*0.8/0.4 ~ 20 W/kg.
Assuming I haven’t made a mistake, but I would expect an oom estimate, the step length constraint from having a skeleton (a stride length) still puts a lot of power demand on the agile cheetah.
“Mean top running speed during chase is given as ~ 15/s^2.” Huh. I meant 15 m/s, natch.
Yeah, I made a big mistake. Somehow I switched the v = at expression around. Here is the now much better estimate:
– Assuming instead constant top acceleration, without the step frequency constraint, to get the average power density. Given an acceleration of 10 m/s^2, the time to hit top speed at ~ 100 km/h ~ 28 m/s is t = v/a ~ 28/10 ~ 3 s. The distance covered is then s = 1/2*at^2 ~ 5*3^2 ~ 45 m.
Check: A distance of 45 m would be equivalent to ~ 3*3.5 ~ 10 – 11 steps.
The work done is W = int(F*ds) = ma*s, giving the power density P = W/t/m = a*s/t ~ 10*45/3 ~ 150 W/kg.
So the cheetah *is* an agile, mean, acceleration machine.
Just to make clear – I think this is a pretty cool study technically (putting collars on wild cheetahs!), and it puts some precise numbers on stuff that we had all intuited from watching wildlife documentaries.
My snark (also expressed on Twitter) was about a) the obvious description on the Nature website and b) the silly introduction to the accompanying news article, which suggests that we all thought that cheetahs were ambush predators that run in straight lines. That’s what was humdrum/daft.
The other aspect is the eternal issue of publishing in the Big Magazines (Nature, Science and Cell, which Jerry has done – all three! – and I haven’t, so am therefore bitter and twisted). Big predators (including fossils) are eyeball bait, so tend to be favoured. If this study had been done on some other organism (say a pangolin) it wouldn’t have got within sniffing distance of Nature. Which is just the way it is. After all, if it had been on African Dogs, it wouldn’t have got within sniffing distance of WEIT…
No doubt about it, high performance and beauty sell.
There was a discussion and interview on this research on the BBC ‘Material World podcast dated 13 June
http://www.bbc.co.uk/podcasts/series/material
this includes some more details of the Horizon wandering cat survey, which used miniaturized versions of the radio collars
The remark about muscle mass certainly rings true. I once saw a cheetah start to chase some impalas, starting from a few metres in front of me and running directly away. The size of her bunched haunches on a basically slender animal was quite extraordinary.
I would think that rate of acceleration would eventually give the cheetah a concussion…or do they have an adaptation?
I don’t think so. Even an average condition human can withstand that level of acceleration for a couple of minutes without any serious issues.
Something that has always impressed me as much as the cheetah’s speed is how it is able to keep its head so still even when moving flat out, and violently changing direction, over uneven ground. Very well engineered for high speed chases.
Any experts reading here know if cheetah’s use constant bearing for target interception like most animals, or do they use constant absolute target direction like bats do? The cheetah’s prey often moves rather erratically.
http://www.sciencedaily.com/releases/2013/06/130617104608.htm
This is a report of a fast tricky robot designed like a cat. Sort of fits in here.
The acceleration rates aren’t expressed in any terms I really understand: the highest was 100 W/kg (can some reader translate that in to m/sec/sec?)
With the following incantation, I asked my online assistant (wolframalpha) to convert this to an equivalent acceleration, at the cheetah’s average sprint speed:
“(100 W/kg / 15 m/s) in m/sec/sec”
He answered 6.667 m/sec/sec, and also helpfully noted that this is ~0.88x the acceleration of a 2002 Ferrari Enzo (from 0-100 km/h).
I ma surprised that there is no link to this video.
Nat Geo put many hours and bucks into getting these images, I have never seen any like it.
Some things this video makes easy to see
1) At top speed, no more than one foot is on the ground for more than milliseconds.
2) For a substantial portion of each stride, the cheetah is airborne. They are flying along the ground.
3) Mentioned above, the body flexes in several directions, but the head and the cheetah’s visual track does not vary a millimeter.
4) When they extend their front legs, you can see the wind resistance lift the fur on their paws.
http://www.youtube.com/watch?v=THA_5cqAfCQ