Natural selection from the top

When a dam formed a new lake in Venezuela, hills became islands, and ecologists could see what happens when 'top predators' vacate an ecosystem, writes Steve Connor

In the environmental debate, many of us have overlooked one argument that goes to the very roots of the issue. Not so ecologists. They are split, fundamentally, between the "bottom-uppers" who think that the lifeforms at the bottom of the food chain (basically, plants) dictate the state of the environment, and the less fashionable "top-downers" who take the view that it's the big predators at the top of the food chain that ultimately mould an ecosystem.

Like the polarised positions of the "big enders" and "little enders" in Gulliver's Travels (who argued over which end of the egg to break), there was not much anyone could do to decide who was right. The debate was stifled by a lack of hard evidence and a dearth of scientific predictions that could be tested by experiments. However, that is changing, following the first results of a unique study into environmental degradation. And it seems that the top-downers could be proven right after all.

The story starts in 1986 with the flooding of a number of valleys as part of a vast hydroelectric scheme in Venezuela. The dam created Lake Guri, covering 4,300 square kilometres, and turned hundreds of former hilltops into islands. The islands quickly lost many of their original species, with the smallest losing about three-quarters. Furthermore, within four years all the islands had lost their top predators – the jaguar, puma and harpy eagle.

It was a perfect "natural" experiment into what happens when top predators are suddenly taken out of the environmental equation. John Terborgh, from the Centre for Tropical Conservation at Duke University in Durham, North Carolina, thought it was too good an opportunity to miss. With an international team including Madhu Rao of Cambridge University and Luis Balbas from Venezuela, he set about studying what has happened to these isolated communities over the past 15 years.

Traditional, bottom-up ecologists argue that plant growth is enhanced through light and warmth, and the availability of moisture and nutrients. Herbivores exert top-down forces on the plants, and the bottom-uppers argue that these plant-eaters are kept in check by the quality of the forage and the in-built plant defences, such as toxins, which render vegetation unpalatable or indigestible.

Top-downers, however, argue that the world is green because predators regulate the number of herbivores in an ecosystem, thereby limiting the damage that these plant eaters do to the environment.

In a paper published in last week's issue of Science, the scientists make a prediction to test which one of these ideas is right: "If all relevant predators could be excluded from a sufficiently large experimental area, the top-down model would predict that consumer populations would expand, whereas the bottom-up model would predict little change in consumer numbers." Lake Guri provided a perfect opportunity to see which was right.

The scientists built up inventories of animals and plants from a dozen islands – which they classified as small, medium or large – and compared them with natural inventories from two mainland sites. They were interested in different "trophic" levels, which means observing who is feeding on whom. On small islands there were three trophic categories: eaters of invertebrates (such as lizards, spiders and birds); consumers of seeds (small rodents); and herbivores (howler monkeys, common iguanas and leaf-cutter ants).

Bigger islands also have larger animals, such as armadillos, agoutis (a big rodent) and capuchin monkeys. But none had any of the large predators that would normally have preyed on these animals. The small area of land of even the largest island is just too small to sustain a breeding population of large predators, says Terborgh: "A space of just a few acres simply doesn't support a self-sustaining population of any predator that feeds on vertebrates. The resulting communities are thus highly aberrant, consisting of a suite of consumers without predators."

The effect of removing this top-down pressure varied from island to island, but essentially consisted of a population explosion of certain creatures that benefited from the new composition of species in the ecosystem. For example, the numbers of plant-eating animals increased by between 10 and 100-fold. "Some islands have monkeys living on them at densities well above combined densities in the most abundant primate communities ever discovered on Earth," says Terborgh. "It is something off the scale, never before seen. Similarly, we have other herbivores that are in extreme abundance in leaf-cutter ants and common iguanas, and various kinds of small rodents," he says.

What was particularly striking was how differently events turned out on each island. Jared Diamond, the award-winning author and evolutionist at the University of California in Los Angeles, says that the findings point to the inherent random- ness of ecological development. "It reminds me of a kaleidoscope that gets shaken up hundreds of times, with some general tendencies emerging among the resulting patterns but with fascinating differences of detail because of effects of chance, yielding bizarre communities that would never exist in equilibrium."

On one island, a herd of capybaras have turned the island's low-lying vegetation into bare ground covered in their own dung. On another island, a population of howler monkeys has soared, and as a result fertility rates (the number of offspring per female) have gone down. On another island, capuchin monkeys have decimated the bird population by 90 per cent, and on yet another, agoutis have created a landscape full of certain plants whose seeds they like to bury.

"Those species that do manage to persist on small islands tend to become far more abundant than populations of the same species on larger, more species-rich islands," says Dr Diamond. "Part of the explanation for increased abundance is ecological release from competition: when competing species are removed, the resources that would have gone to them become available to the persisting species. The other part is escape of prey from control by predators or parasitoids that limit their abundance on larger islands."

One of the key findings of the study was that in the absence of top-down regulation by big predators, a new form of bottom-up regulation begins to take place. An example is the self-limiting nature of the howler monkey's fertility. On one small island of just 1.5 acres, the fertility rate was 0.125 births per female per annum, compared with 0.5 births per annum among howler females living on a large, 864-acre island.

As a result of what the scientists call a "trophic cascade" (herbivores becoming hyperabundant when no predators are there to cull them), the vegetation begins to change from edible plants to inedible or toxic species. "Hyperabundant herbivores threaten to reduce species-rich forests to an odd collection of herbivore-resistant plants," wrote the scientists in their paper.

"The trees there before the flooding are now dying off at much higher than normal rates," says Terborgh. "The combination of rodents and leaf-cutter ants and all these other herbivores are just eliminating all the young plants that appear. The vegetation is in a state of collapse. Plant species that can survive under this massive onslaught by herbivores are the ones that are incredibly tough and terribly toxic," he says.

This most unnatural experiment in habitat fragmentation also tells us that dividing up the natural environment into ever-smaller parcels of land – which is what happens when roads are built through a rainforest – will inevitably result in massive degradation. Dr Diamond points out that habitat fragmentation on a grand scale is how much of the world formed after the last Ice Age 13,000 years ago. "Melting glaciers raised sea levels, and drowned low-lying land bridges and carved off edges of continents into islands such as Britain and Japan. But fragmentation is not just a feature of the last Ice Age, it is also the fate befalling most natural habitats today; hence the practical interest [in this study]."

Taking away the top predators in a natural community, as has happened in many parts of the world from Europe to North America, could also have far-reaching consequences only now being realised.

The simple lesson for those who would manage species: make sure you keep the top predators. It's a very good way of keeping nature natural.