Removing Dams Can Restore Wildlife Habitat at Small Cost

By applying new mathematical techniques to river ecology, a University of Maryland biology professor has found that removing dams to reconnect rivers in a watershed like Oregon’s Willamette River could result in significant wildlife habitat restoration benefits at a comparatively small economic cost.

William Fagan, associate professor of biology at the University of Maryland, presented his findings at the Society for Conservation Biology’s Annual meeting, last week at Columbia University.

Fagan’s study looks at the connectivity of an entire watershed, a critical element in the survival of many species of freshwater fish. “Rivers have a unique ecology, because of the branching geometry they exhibit,” says Fagan. “My analyses call attention to the importance of connectivity in river systems.”

That importance was highlighted in another of Fagan’s recent studies, where he demonstrated that for fishes native to the Sonoran Desert, riverine species with fragmented distributions are more prone to extinction than those with more continuous distributions.

Restoring River Connectivity
While the effects of breaking up wildlife habitat on land with development and other human manipulation have been heavily studied, there has been far less research on what happens to migratory species in freshwater systems when their habitat is fragmented. “Fragmentation has different consequences in freshwater systems than in terrestrial habitats,” says Fagan.

For example, migratory fish, such as Pacific salmon in the northwest, need different areas of the river system to complete their life cycles. Pacific salmon spawn upstream in fresh water, but develop in the ocean. Consequently, an individual fish must make two passages through the same riverine system — one downstream as a juvenile, and one upstream when it’s ready to mate.

“Dams can block or hinder those movements, and consequently there is much discussion about the benefits of reconnecting rivers by removing dams,” Fagan says. “However, most of the focus is on removing dams one at a time. My colleagues and I looked at would happen if some dams were removed in a whole watershed and areas of the river habitat were reconnected. We wanted to see what kind of balancing act exists between economic costs of removing dams and the benefits of restoration when the analyses are done at the landscape scale.”

Using a system of dams in the Willamette River in Oregon, a habitat for Pacific salmon, as a test case, Fagan and his colleagues explored the economic costs and environmental benefits by applying new math analysis techniques, termed combinatorial multiobjective programming for stream networks, to the question.

“Because of the hierarchical geometry of river systems, you can sometimes get a lot of restoration action with a relatively small economic loss if you consider dam removal systematically at the watershed scale,” Fagan says. “Restoration benefits would include increased population sizes of key species of migratory fish, increased dispersal of other species throughout the landscape and return of the natural flow.”

Sonoran Desert Fish Show Effects of Fragmentation
In the study that shows the importance connected river habitat for some riverine species, Fagan used mathematical analyses to examine more than 160 years of data on Sonoran Desert fish species in the Colorado River. The goal – to see if patterns of species’ distribution in the past could be a guide as to which species today will be more prone to extinction because of human manipulation of their river habitats.

One such endangered species is the Colorado pikeminnow. A large fish — an adult ranges between four and six feet in length — the Colorado pikeminnow must migrate hundreds of kilometers to carry out its life cycle. Damming and diversion of its river habitat and widespread introduction of sport fishes such as bass, which feed on the young pikeminnow, are threatening the native Colorado pike-minnow with extinction.

Using mathematical and statistical techniques to analyze data from thousands of historical records and maps, made as far back as the 1840’s by the U.S.-Mexican Boundary Commission, along with data from the past 20 years, Fagan found that the species most likely to have gone extinct since historic times are those that had more highly fragmented distributions.

“Species that historically have had fragmented distributions have had a tougher time than those that had more cohesive distributions. We show that those species that were historically fragmented have suffered more loss,” says Fagan. “That implies that today, those species whose distributions more fragmented because of human activity are more at risk for extinction.”

Collaborators on the studies were: Dam Removal – Michael Kuby, Arizona State University; Charles ReVelle, Johns Hopkins University; William Graf, University of South Carolina; Desert Fish, funded by the National Science Foundation – Peter Unmack, Arizona State University; Craig Aumann, Christina Kennedy, University of Maryland.