Nicole Rust on Nostr: Fact check gem of the day: The hare/lynx story in the textbooks is all wrong!! The ...
Fact check gem of the day: The hare/lynx story in the textbooks is all wrong!!
The 101 description of how properties can "emerge" in a complex system through interaction of its parts often relies on the observation that hare and lynx populations oscillate up and down with a phase shift, and this can be described by the Lokta/Volterra coupled differential equations. The gist is that the numbers of lynx around depend on the hares to be eaten, but the numbers of hares depend on the numbers of lynx around to eat them. From the math, the oscillation emerges; it's not coming from either of the parts alone.
This is a story that is in *many* (many!) textbooks and books (including mine, but it's still in revision - whew!). The problem is that it's an inaccurate description of the data insofar as the phase is off: the equations predict that the peak of the lynx will follow the peak of the hare but the data say the opposite. (Oops).
This inspired the hilarious title of one hilarious paper, "Do hares eat lynx?"
https://www.journals.uchicago.edu/doi/abs/10.1086/282870?journalCode=an
The presumed resolution has to do with the nature of the data: the animal numbers were counted via the pelt yield of trappers in Canada. Trappers are a third source to consider. When trappers are incorporated into the model, the phase shift is well-predicted:
https://core.ac.uk/download/pdf/188134744.pdf
The 101 description of how properties can "emerge" in a complex system through interaction of its parts often relies on the observation that hare and lynx populations oscillate up and down with a phase shift, and this can be described by the Lokta/Volterra coupled differential equations. The gist is that the numbers of lynx around depend on the hares to be eaten, but the numbers of hares depend on the numbers of lynx around to eat them. From the math, the oscillation emerges; it's not coming from either of the parts alone.
This is a story that is in *many* (many!) textbooks and books (including mine, but it's still in revision - whew!). The problem is that it's an inaccurate description of the data insofar as the phase is off: the equations predict that the peak of the lynx will follow the peak of the hare but the data say the opposite. (Oops).
This inspired the hilarious title of one hilarious paper, "Do hares eat lynx?"
https://www.journals.uchicago.edu/doi/abs/10.1086/282870?journalCode=an
The presumed resolution has to do with the nature of the data: the animal numbers were counted via the pelt yield of trappers in Canada. Trappers are a third source to consider. When trappers are incorporated into the model, the phase shift is well-predicted:
https://core.ac.uk/download/pdf/188134744.pdf