The Bird Watcher
By Christina Dierkes, Featuring: Steve Matthews
Stephen Matthews’ research focuses on understanding the responses of ecological systems in a changing world.
Climate change is not only likely to affect our weather, our water levels, and our health, but also the variety of birds we see in our backyards and city parks. Changing environmental conditions will affect where these birds live, raise their offspring, and spend their winters, and wildlife managers across the Great Lakes region and beyond will have to consider these potential changes when deciding on the best way to manage the wooded areas in their care.
Dr. Steve Matthews is one of the researchers providing wildlife managers, other forestry officials and the general public with tools to make informed decisions about the potential future of birds in the Eastern United States, including the Great Lakes region. Matthews’ research focuses on how bird species are distributed across a landscape, and how those patterns may change in terms of climate change projections.
“The general focus of a lot of my work is in developing species distribution models and trying to understand what environmental conditions species are most associated with in terms of climate, tree sites and elevations, for example,” Matthews, Research Assistant Professor in the School of Environment & Natural Resources at The Ohio State University, explains. “I am also interested in how those bird habitats may change under different climate change scenarios,” he adds.
To determine these habitat changes, Matthews and his collaborators Louis Iverson, Anantha Prasad, and Matthew Peters of the US Forest Service use statistical models that analyze current distributions of both the birds of interest and of the trees in these habitats under current environmental conditions. Once the model has determined the relationship between the current climate, the tree species present in a certain area, and the birds that live there, the researchers can use projected climate data from the Intergovernmental Panel on Climate Change (IPCC) to get an estimate of how climate change may affect bird populations in the future.
“Using the RandomForest algorithm, which is the modeling approach that we use to identify the connections between climate, trees and birds, we can say that a species appears to be associated with spring-like temperatures, for example, but there’s also an important link to a tree species of a certain type,” Matthews explains. “This means that if we can build those models and we get those estimates, then we can basically swap the climate out with future projections. The advantage of using trees as a predictor for birds as well is that we can then also have an idea of how those trees’ habitats can change,” he adds.
Models like these are already used to estimate the impact of decisions about how to manage forests and wildlife populations in light of climate change, from determining what trees would fit best into a reforestation project to which bird species are going to be most impacted by a changing forest composition, but Matthews and his colleagues suggest that predictions that are based on climate alone do not tell the whole story.
“We rely on those models to identify the important connections between species under current conditions, but as climate or habitat changes, it’s important to understand that these connections may break or shift,” Matthews says. “So we want to be able to quantify how species’ habitat may change, but we also have to keep in mind that other factors come into play. For example, on the tree side, red maple is projected to decline under climate change, that’s what the profile would suggest. We also know that it has adaptable capacities and can grow under a variety of conditions, so our decline projections should be taken in the context that the species might be quite adaptable.”
Such potential adaptability in turn may allow tree species to provide more than the projected habitat for associated birds. It also suggests that some bird species may be similarly adaptable to changing environmental conditions. This idea is not taken into account by current statistical models, which therefore may not provide the full picture of how a species (be it bird or tree) may respond to certain environmental conditions.
To improve the predictive powers of current models, Matthews and his collaborators at both Ohio State University and the US Forest Service have developed “modification factors” that allow for input of variables beyond climate change. These “ModFacs” consider species’ life history characteristics as well as additional distributional summaries. The third component of their multi-model framework makes a detailed quantitative assessment of potential species expansion or move beyond current habitats, specialized local knowledge and current land use (such as agriculture), and are applied to standard models to better inform management decisions.
Support for wildlife and forestry managers has recently become more important, Matthews says. “The approach we take is that a list of species is informative, but there are other components that need to be included,” he explains. “These species distributions aren’t static, and they have never been static. So in the long term, changes are likely to occur, and we’re going to have to build in an idea of how adaptable species may be to those habitat changes. One of the key results is that by looking at how the climate may change and how the species’ habitat may be stressed, the individual characteristics of a species’ distribution might change as well.”
The most recent tool Matthews and his collaborators have developed is a vulnerability assessment that tries to lay out the ways in which, as a species changes, “you can adapt how areas are currently managed to identify the ecosystems that may be most vulnerable”, Matthews says. The assessment tries to pull all of that information together, to show the data that are likely to be important, and to present the information in an easily understandable but not too simplified manner. “Then the managers on the ground, the ones that are most familiar with the situation, they are the ones that are in the best position to make the important decisions,” Matthews concludes.
How statistical models can assist managers in making these important decisions is going to be one focus of Matthews’ Global Change, Local Impact webinar on September 29 as well. Matthews will also introduce another currently available tool to visualize changes in bird species distribution under climate change, not just for wildlife managers but also for the interested public: an online Climate Change Bird Atlas that provides distribution data for 150 bird species in the Eastern United States, both under current conditions and under various climate change scenarios, allowing wildlife managers concerned with protecting a particular species to understand how climate change may affect that species’ future distribution in their region of concern.
More information about Dr. Matthews’ webinar, as well as the upcoming webinars on climate change impacts in the Great Lakes region, is available at the webinar page. The Global Change, Local Impact webinar series is a multi-departmental effort within Ohio State University, led by Ohio Sea Grant, OSU Extension, and nine other OSU departments, to help localize the climate change issue for Ohioans and Great Lakes residents.