North American forests are full of nonnative insects—more than 450 species, by the latest available count. Most have done no obvious damage, but the few that have—such as the emerald ash borer, which is killing off its namesake trees, and the hemlock woolly adelgid, which is devastating eastern hemlocks—have remade entire landscapes, doing untold ecological and economic harm. People have tried for decades to understand why some introduced insects become pernicious invasive species while others apparently remain innocuous, but these efforts have been mostly fruitless. Predicting which path an organism will follow “is the holy grail of invasion biology,” says forest entomologist Kamal Gandhi.

Gandhi, who works at the University of Georgia, is on a team of scientists who have taken what they say is a first step toward that prize. Concentrating on the nonnative insects known to attack North America’s conifer trees, they used newly available data and computer models to uncover several patterns that reliably predicted which nonnative insects were likely to be damaging—and which tree species would be the victims. The researchers say their findings, published this past fall in Ecology and Evolution, can immediately be used to prevent potential new damaging invaders from arriving and could provide a template for predicting other invasive species. “This was the proof of concept,” says team member Matt Ayres, an ecologist at Dartmouth College. “We’re on the trail to the grail.”

The team focused on the 58 known nonnative specialists that feed only on a certain few species of trees—in this case, one or more of 49 species of North American conifers (the order that includes pines, junipers and redwoods). Of those 58 introduced insects, six have done widespread damage, becoming problematic invasive species. The researchers first built a database of the ecological traits of the trees and of the native and nonnative insects that feed on them. The team used this data to build a series of models that incorporated the impact of insect and tree traits, the evolutionary history of the trees, and the presence or absence of native insects that fed on them.

One critical piece of information used in the study was a conifer phylogeny—that is, a genetic history of when the various conifer families, genera and species diverged from one another. When the researchers incorporated the phylogeny into a model, they found a Goldilocks-type relationship between the North American trees most damaged by nonnative insects and the ones those insects fed on in their place of origin: both sets of trees were neither too closely nor too distantly related. “There’s a sour spot right in the middle,” says Nathan Havill, a U.S. Forest Service entomologist on the team. “The trees are far apart enough that they don’t have the defenses against a specialist herbivore, but they’re not so far apart that a herbivore couldn’t recognize them as food.”

A second model showed that the trees most likely to be damaged by a nonnative insect were shade-tolerant and drought-intolerant. The researchers hypothesize this link occurs because trees growing in damp, shady places are less able to repair themselves than plants that grow in brighter conditions. “Their photosynthesis is limited,” says entomologist Dan Herms, a co-author of the study and vice president of research and development at the Davey Tree Expert Company. “They don’t have a strong ability to tolerate being defoliated” by ravenous pests. These trees often do have defenses that are well honed against native insects. But if a new insect arrives that is able to circumvent those defenses, the trees are easily overcome. A third model, meanwhile, showed that conifers in general were more likely to withstand a nonnative insect when they already had defenses against a closely related native one.

What the models did not show, surprisingly, was any relationship between the invasive insects’ traits (such as the number of eggs they laid or how well they were able to disperse) and their deadliness. “I thought insect life-history traits would matter,” Gandhi says, noting that it seems intuitive to expect the damaging invaders would have traits in common. Most past efforts to predict invasive species in forests had focused on insect traits, she says—perhaps explaining why those efforts failed to turn up useful patterns.

Taken together, the models affirm the great importance of shared history. Researchers have long suspected the impact of nonnative insects in North America’s forests depended on the relationships between the native trees, the nonnative insects and the relatives of both, says Michael Donoghue, an evolutionary biologist at Yale University, who was not involved in the new study. “That idea had been around in the literature, with very little evidence to back it up,” he says. Now “they’re saying, ‘Oh, wait—we find actual evidence.’”

When the researchers combined their new models, they found they could retroactively predict which nonnative insects would become damaging invaders with more than 90 percent accuracy—which gives the scientists confidence these models could predict future problematic invaders. The team is now working on such predictions, evaluating which specialist insects might be deadly if they arrive in North America and which tree species are likely to be vulnerable. The team is also looking to replicate its work with studies of North American flowering trees and of generalist insects that are capable of feeding on many types of plants. Other groups might eventually go even further. “Even though this study looks at only one specific group in one part of the world, it’s really an important contribution,” says Eckehard Brockerhoff, an ecologist at the Swiss Federal Institute for Forest, Snow and Landscape Research, who was also not involved in the recent paper. “I think it will act as a template for other studies.”

Angela Mech, a Western Carolina University entomologist who led the work, agrees. “Folks have been trying to unlock this door for a long time,” she says. “This is just the beginning.”