Warming puts species on collision course

Amphibians demonstrate the important role that the ability to shift habitat, or disperse, plays in responding to climate change, as some disperse only a few meters a generation, while others can disperse a kilometre or more. Credit: R. Semlitch/D. L. Drake

Scientists might be significantly underestimating how many species will die out due to climate change, said Mark Urban from the University of Connecticut this week. That’s because current methods for predicting how organisms will fare as the world warms overlook how they interact with each other. However, Urban and two University of Washington scientists have instead developed prediction models that show greater extinction risks when they considered how organisms compete and move. “Because species interact and differ in dispersal ability, we might be vastly underestimating climate change impacts on biodiversity,” Urban told Simple Climate. “This means that current predictions underlying biodiversity threats used by governments and conservation organizations could be conservative.”

Urban is what’s known as a “community ecologist”, meaning that he researches how different organisms living closely together influence each other. “We know that species interact in nature,” he explained. “And we also know that they disperse at quite different rates. For instance, I work with amphibians and there are some amphibians that disperse only a few meters a generation, and then others that can disperse a kilometre or more.” Those relationships can be just as important to an animal or plant as physical properties like temperature, the ecologist noted. “We have some inkling that species interactions can interfere with climate change responses, but to date few scientists had tackled this issue.” he said.

Instead, existing predictions start by looking at what the climate’s like where a species currently lives. After examining how climate models forecast that will change, scientists can predict whether the species will be able to move to locations that have the kind of conditions it needs. But this method overlooks factors like species that have never met before potentially coming together as they shift ranges. “This is a fine initial approach, but we need to explore more sophisticated models that account for species interactions and dispersal differences,” Urban said.

Competing interests

Urban and colleagues’ model assumes a thermal gradient from 10 °C to 30 °C which could simulate a mountainside as pictured or a larger flat terrain running north-south. It then allows the gradient to warm by 4 °C in 100 years. Species lettered a-g each are each best suited to different temperature ranges. Reproduced with permission from: Mark C. Urban, Josh J. Tewksbury and Kimberly S. Sheldon, “On a collision course: competition and dispersal differences create no-analogue communities and cause extinctions during climate change”, Proc. R. Soc. B, 2012, DOI: 10.1098/rspb.2011.2367, data supplement, figure S1A, Royal Society Publishing.

To get a prediction that included competition, Urban considered how well species might manage at different temperatures in mathematical equations commonly used to understand population changes. Such equations multiply the size of populations by different fixed numbers to give a value indicating how much they will change due to competition. However, rather than multiplying by a fixed number, Urban used a value that could change to represent differences in how well the different species adapt to different conditions. “The way this works is that two species adapted to survive well in a habitat compete relatively evenly,” Urban explained. “However, if one of a pair of species is much more poorly adapted to that temperature than another, then it faces stronger competition.”

From that behaviour, Urban’s team created a model described in a paper published on Wednesday in the biological research journal Proceedings of the Royal Society Paper B. Their model contained forty different species with different responses to temperatures, based as much as possible on measurements. They placed these species in a landscape with a 20°C temperature gradient that might be seen over the climb up a 3000 metre high mountain or on a journey away from the equator across 15° of latitude.

Ecological car wreck

Mark Urban, assistant professor of ecology and evolutionary biology at the University of Connecticut, stands under a sheet of aufeis in the Brooks Range of mountains in Alaska, a region where some of the most dramatic climatic changes of anywhere in the world have been seen. Credit: Heidi Golden

“We find that competition and dispersal variation matter tremendously for the effect of climate on species,” Urban said. “Without competition, as current predictions commonly assume, there are very few extinctions. However, with competition, many species become extinct. The resilience of the community all but disappears. Extinctions become more prevalent when species as a whole disperse poorly, but also when they differ in their dispersal. The reason is that the fast moving species outcompete those species in front of them. Like cars on a single lane, you cannot move faster than the car in front, or else you collide.”

Previous estimates for the number of species likely to die out due to climate change include between three and eight in twenty among ground-inhabiting organisms. More recently, researchers have predicted around one in ten of all types of creature could go extinct by 2100. While Urban’s team’s work indicates these numbers may be too low, he cannot say by how much. “Our model is very general which makes it good for testing the sensitivity of predictions to different assumptions,” he explained. “The downside of this approach is that the model does not apply to any specific system.”

Therefore, among the next research items on Urban’s to-do-list are to determine how true these effects are in reality, and build models based on natural communities to make more specific predictions. Another follow-up will be to look at how other types of relationship, like those between predator and prey or host and parasite, affect climate change responses. But one of the biggest challenges they face in doing studies on the role these factors play is a lack of measurements to start from. They note data on the differences in how fast species move in particular hasn’t been recorded – and now are hoping other scientists will help them collect it. “At this point, we just don’t have a lot of data, especially on the dispersal differences among species,” Urban said.