Predictions of the loss of animal and plant diversity around the world are common under models of future climate change. But a new study shows that because these climate models don't account for species competition and movement, they could grossly（很，非常） underestimate future extinctions. "We have really sophisticated meteorological models for predicting climate change," says ecologist Mark Urban, the study's lead author. "But in real life, animals move around, they compete, they parasitize each other, and they eat each other. The majority of our predictions don't include these important interactions."

Plenty of experimental studies have shown that species are already moving in response to climate change, says Urban, assistant professor of ecology and evolutionary¹ biology at the University of Connecticut. For example, as temperatures rise over time, animals and plants that can't take the heat are moving to higher altitudes where temperatures are cooler.

But not all species can disperse²（分散，传播） fast enough to get to these more suitable places before they die off, Urban says. And if they do make it to these better habitats, they may be out-competed by the species that are already there -- or the ones that got there first.

With coauthors Josh Tewksbury and Kimberly Sheldon of the University of Washington, Urban created a mathematical model that takes into account the varying rates of migration³ and the different intensities⁴ of competition seen in ecological⁵ communities. The goal was to predict just how successful species within these communities would be at shifting to completely new habitats.

Their results showed that animals and plants that can adjust to climate change will have a competitive advantage over those that don't.

Animals with small geographic⁶ ranges, specific habitat needs and difficulty dispersing⁷ are likely to go extinct under climate change, their model shows. Further, these animals are more likely to be overrun by other species that can tolerate a wider range of habitats.

"When a species has a small range, it's more likely to be out-competed by others," Urban says. "It's not about how fast you can move, but how fast you move relative to your competitors."