Crunching the Climate Change Numbers
Earlier this week, an international team of 42 researchers, including John Tipton, assistant professor of mathematical sciences at the University of Arkansas, published an analysis of changes in vegetation after the last ice age in the journal Science. They found that rising temperatures caused dramatic changes in Earth’s vegetation, which suggests that warming from climate change could have a similar effect in the future.
The team, which was led by researchers at the University of Arizona, looked at previously published data from 594 sites, covering every continent except Antarctica. In order to compare the different records, they classified the changes in vegetation as either low, moderate or large.
When the time came to crunch the numbers, the researchers enlisted Tipton’s help. Tipton studies statistical data science, or, as he puts it, “putting the science into the math.” Tipton helped the team find the right statistical tool to analyze the data.
“They needed a model that respects their scientific understanding of how vegetation is influenced by climate,” he said. “We had to make sure the statistical model is appropriate for data and the question they wanted to answer. If you use the wrong model, you end up with results that don’t answer the question of interest.”
Tipton recommended the team use an approach called “ordered logistic spline regression,” and with this model, the researchers concluded that regions with the largest changes in temperature also experienced the largest changes in vegetation. Based on this, they predicted that climate change could have a similar effect, causing changes in vegetation that could affect global biodiversity, agriculture, carbon storage and recreation.
“We can learn a lot about the future by learning about what happened in the past,” said Tipton. He also explained that interdisciplinary projects like this one demonstrate the power of combining the abstract, quantifiable power of math with the observable phenomena studied by ecologists and geologists. “This speaks to the power of expert and domain knowledge,” he said.