How the dance of predation upends ecology models

Not all predators are equal, and neither are prey. When lots of genetic variety allows two species to co-evolve rapidly, reports a new study, both their population numbers and their evolutionary fitness levels fluctuate, defying traditional ecological models.

The primordial two-step of predation and predator-avoidance has driven evolution since long before our hominid ancestors abandoned knuckle-walking for the upright posture that let us scan the savannas for dangerous cats, who were meanwhile learning more effective ways to flatten themselves and stalk prey.

A hundred years ago, when ecologists began to study how predation shaped populations, they found that many relationships followed a logical cycle: rising prey populations sustain rises in predator populations, which cause prey numbers to drop, thereby sustaining fewer predators, whose decline allows prey populations to grow again, and so on.

These undulating predator populations, when graphed, lag about a quarter of a cycle behind the populations they preyed upon, according to the venerable Lotka-Volterra ecological model. Soon after prey populations rise, predator populations will rise to eat them. So, when hares are hopping through a forest at maximum density, lynx levels will be moderate, and rising. And when lynxes are peering tuftily down from the greatest possible number of tree branches, hare numbers will be moderate and falling. 

This model assumes the lynxes and hares in question are all pretty much interchangeable. Which happens to be true enough, in the case of lynxes and hares, to make them the poster animals for this model.

But elsewhere in forests – and oceans and deserts and peat bogs – individuals often survive thanks to specific strengths, be they fleet feet, an astute mind, or a hard beak. Evolution depends on these differences.

Squirrels with extra-brilliant memories for buried nuts, for example, have an advantage in life, all else being equal. But during peak-owl moments, extra-skittish squirrels may survive to bear far more pups than those with strong memories. In other words, species can evolve over the course of their population cycles, changing how those cycles interact.

"Changes in allele frequencies (and associated phenotypes) can occur at the same rate as changes in population densities or spatial distributions, and alter the ecological processes," explains a report published Tuesday in the Proceedings of the National Academy of Sciences.

The mathematical biologists behind the report already knew this – that evolution and ecological cycles can occur over the same time scales. But, exploring how these two kinds of shifts interact, they made a surprising discovery: when specific prey and predator communities are co-evolving, the interplay between their populations can be reversed.

"I hoped to find something new, but I didn't expect to find this," says Michael Cortez, the study's lead author and a National Science Foundation post-doctoral fellow at Georgia Institute of Technology.

Dr. Cortez took data gathered in past studies, of phages eating cholera bacteria, mink eating muskrats, and gyrfalcons eating rock ptarmigans, and plugged them into one of two ecological math models, which include fitness levels as variable.

Out popped the surprise: in all of those relationships, prey populations actually rose in response to predator increases. In fact, they followed them by between a quarter cycle and a half cycle.

Why would this happen?

"The prediction would be that the prey are eating the predators, but that's completely wrong," says Cortez.

The mechanism behind it, which depends not only on populations' fluctuating numbers, but also on their fluctuating levels of evolutionary fitness, is quite a bit more complex than the Lotka-Volterra model. 

Imagine a community in which a small population of "fit," hard-to-catch prey – perhaps lightning-fast mice – interact with a large population of not-very-effective predators, like visually impaired hawks. Because of the imbalance, the hawk population drops, with only enough food to sustain the very sharpest-eyed among them. The disappearance of so many predators allows the swift mice to become numerous, and at this point we see a reversal of the Lotka-Volterra model: a population peak of inept predators actually leads to a population peak of adept prey.

This abundance of mice, of course, gives the few surviving sharp-eyed hawks fodder to multiply, thus creating a large population of adept hunters. And this population drives the mouse population back down. Now, because the mice are outnumbered by terribly keen-eyed hunters, their genius for lickety-split scampering stops being much of a defense.

At this point – and this is the tricky part – it actually behooves the evolving community of mice to trade in their speed for another quality, perhaps higher litter size. The population of these slow mice is quickly driven down by the hawkish hawks, at which point, surrounded by defenseless prey, the hawks no longer need to be quite so keen-eyed. So they revert to their previous myopia, in exchange for for some more advantageous trait, such as strong memory. The abundance of vision-impaired raptors makes it worthwhile for the mice to re-evolve great running skills, and the cycle begins again.

For this reversal to happen, the relationship needs to foster a brisk pace of evolution. Cortez' team identified three main conditions that enable this:

• Predators' skill sets need to be full of trade-offs. Meaning that, for hawks' hunting aptitude to be as elastic as it is in the scenario above, an improvement in vision must lead to a loss in some other quality – which is related to survival but not hunting – and vice versa.

• The relationship between the two species needs to favor "extreme phenotypes," like lightning-fast scampering or enormous litters, rather than a stable balance of decent speed and moderate litter sizes.

• The defenses of fit prey animals, like fabulous speed, must be effective deterrents against weak hunters, like half-blind hawks.

The new model, a so-called "clockwise" pattern of prey and predator populations, is likely to change how ecologists predict population dynamics. "I think the important part to take from it, is that when studying populations and their fluctuations over time, it's important not just to measure the total numbers of things, but also to measure their different phenotypes," says Cortez.

"The most exciting part to me, is that you have this 90- or 100-year old prediction from some of the founders of mathematic ecology, who predicted that predator and prey cycles should have a certain shape," he adds. "This has been consistent until maybe 10 years ago. My work has sort of shown that, when you have evolution or genetic diversity in popultions, you can actually completely reverse that. To people who have spend their careers on this, it is pretty mind-boggling that you can have adaptation reversing an ecological dynamic."