Unnatural selection

By carefully removing species, humans can help avert disaster in imperiled ecosystems

A little human meddling may prevent the mess of extinctions that can ripple through disturbed ecosystems.

CONSTRUCTIVE INTERFERENCE Modeling interactions in the Chesapeake Bay food web reveals how meddling affects the entire network. While removal of some species (large red dots) may cause an extinction cascade, removal of other players (green) can shore up the network, limiting damage. For example, taking out species F results in an increase in species i, which indirectly helps species S. S. Sahasrabudhe and A.E. Motter/Nature Communications

A new analysis of how perturbations propagate through a network of organisms reveals that when an ecosystem is already off-kilter, proactively removing particular species can halt the cascade of destruction that may follow. The approach, described online January 25 in Nature Communications, could help well-defined areas such as islands deal with the effects of invasive species.

“At the end of the day, methods based on inflicting locally controlled damage — despite being damaging — can have a positive effect on the entire network,” says study coauthor Adilson Motter of Northwestern University in Evanston, Ill.

To understand how secondary perturbation for the sake of restoration plays out, Motter, an expert in complex networks, and his Northwestern colleague Sagar Sahasrabudhe developed an algorithm that takes into account two classic ecological models of species interactions. By modeling who eats whom and how the exchange of biomass leads to changes in population levels over time, the researchers could identify particular species whose removal or suppression would contain damage. Tactics such as birth control for deer that are overrunning an area, or encouraged fishing to bring down a species’ numbers, may prevent damage to multiple other species, the analysis suggests.

Human interference to compensate for prior human interference has had some success. The feral pigs that were introduced to Santiago Island in the Galápagos, a few years after Darwin visited the archipelago in 1835, were finally cleared from the island in 2000, by which time their numbers had reached more than 18,000. But predicting how such meddling may affect the big picture isn’t always obvious. Much as an initial interference (such as the introduction of an invasive species) can cause indirect damage, secondary meddling can have surprising, but positive, consequences, says Motter.

In fact, the dynamic, indirect nature of networks means that such interference can prevent the majority of secondary extinctions, the researchers report.

“Some extinctions you cannot prevent, but most are caused by this speedy perturbation wave sweeping the network,” Motter says. “Complex systems are always subject to failure; the important thing is to prevent the propagation of local failure.”

Motter notes that the team isn’t advocating large-scale abolition of any species. But for contained spaces such as islands, lakes or parks — places that often grapple with invasive or otherwise out-of-whack distributions of critters — the approach may be quite useful.

“This is a significant advance over prior related work,” says computational ecologist Jennifer Dunne of the Santa Fe Institute in New Mexico, noting that the work is adept at combining the complexity of ecosystem interactions. “If we make bad decisions it doesn’t just affect ecosystems; it affects us,” she says. “So it behooves us to get beyond overly simplistic ways to confront and manage perturbations.” Additional interactions between species — such as pollinator-plant relationships, or how an animal such as a beaver may alter a habitat  — should be incorporated into future modeling efforts, she says.

While more work is certainly needed, the approach provides a potentially helpful tool to conservationists, says Josh Donlan, executive director of the organization Advanced Conservation Strategies in Midway, Utah, and a visiting fellow at Cornell University. “These are exactly the kinds of questions that theoretical ecologists should be asking, that add value to what’s happening on the ground,” he says.

More Stories from Science News on Math