When a giant asteroid struck the tip of the Yucatan Peninsula in Mexico at the end of the Cretaceous Period roughly 65 million years ago, the dinosaurs of North America were doomed no matter what. The impact almost certainly triggered a plant die-off that led to mass extinctions among animals big and small.
Using statistical models of the food webs in several different sites before and after this catastrophe, a University of Chicago graduate student calculated that changes to the structure of North American ecosystems leading up to this event actually made things worse, causing even more species to go extinct. Validated by evidence from the fossil record, this method could be used to develop models that predict the impact of human activity on the environment in the future.
Jonathan Mitchell, a PhD candidate in Committee on Evolutionary Biology, developed models of possible food webs for 17 different sites, 10 from the 13 million years leading up to the asteroid impact and seven from the following two million years. Along with colleagues Peter Roopnarine from the Field Museum of Natural History and Kenneth Angielczyk from the California Academy of Scientists, he built computer models to simulate what would happen to hundreds of different combinations of animal species if ten percent of plant life was wiped out, then 20 percent, and so on. Their findings were published in the Proceedings of the National Academy of Sciences.
When they compared their predictions to data from the fossil record, it correctly predicted 78 percent of the animals that actually went extinct, and 100 percent of those that survived. Mitchell said these simulations showed them how changes to ecosystems leading up to the asteroid made them more fragile.
“It has to do with the distribution of the different organisms in different guilds [groups of similar animals in the same location]. Specifically there were fewer freshwater taxa, and there were more large herbivores,” he said. “As you wipe out a large herbivore, especially a large dinosaur herbivore, you’re actually hitting a lot of guilds all at once.”
For instance, throughout its life a triceratops can serve as prey for many different predators. Small predators can eat a young triceratops; medium-sized predators can eat a juvenile, and only the largest predators can eat an adult. If triceratops herds die off in a given area, it eliminates a food source for a wide range of species. Because the ecosystems of the late Cretaceous period had less species diversity and more large herbivores like the triceratops, they were more susceptible to the effects of a catastrophic impact on plant life.
Mitchell says that testing such models on dinosaur species is useful for two reasons. The first is that they can compare their predictions with known data from the fossil record about what actually went extinct. This allows them to make adjustments and improve the model. The second is that it will give researchers a greater understanding of how natural processes of ecosystem decline and species extinction work without influence from humans.
“This is a hot topic in modern conservation ecology. We want to know how humans are affecting the environment, and how random changes are affecting it,” Mitchell said. “The problem is that all the models you use in the modern world are confounded because you can’t tease those two things apart. We don’t know what ecosystems would be collapsing just because things are evolving, or because of what we’re doing.”
Figuring out how food webs worked without human activities, such as burning fossil fuels, chopping down trees and polluting rivers, can give researchers a baseline they can compare to changes in modern ecosystems. This can help build a case for intervening to mitigate future environmental damage. It’s too late for the dinosaurs, but it doesn’t have to be for today’s animals.
Mitchell, J., Roopnarine, P., & Angielczyk, K. (2012). Late Cretaceous restructuring of terrestrial communities facilitated the end-Cretaceous mass extinction in North America Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.1202196109