Ancient Landslide Merged Trout Populations
Like many rivers in the western United States, California's Eel River has two types of steelhead trout that swim upstream from the ocean to spawn. Today, these typically standoffish fish clans spawn in separate stretches of the river and at different times of year, but they are more closely related than are comparable groups in other western rivers. Scientists have long wondered why. Now they may have an answer. A massive landslide blocked access to spawning grounds for centuries, forcing many generations of trout from the two groups to interbreed on the downstream side of the natural dam.
The hills on either side of the Eel River in northern California are made of easily eroded rocks. Covered in loose sediment and prone to small landslides, they have slopes that typically range between 30° and 35°. So when Benjamin Mackey, a geomorphologist at the California Institute of Technology in Pasadena, and his colleagues studied high-resolution topographical maps of the Eel River valley as part of another research project, the flat terraces that line part of the valley between 239 and 244 meters above sea level—or about 140 meters above river level—stood out starkly in the consistently sloped terrain. Most of those terraces are a few meters thick, Mackey says.
Fieldwork revealed that in some places, particularly in creek beds and gullies where water flows down the hills, the terraces are composed of layered sediments, just like those deposited in a river delta year after year. The hillside deposits indicate that at some time in the past, a deep lake filled part of the Eel River valley, Mackey says. The terraces, which lie at one consistent elevation, are like a big bathtub ring.
That lake formed, the researchers contend, when a large landslide blocked the river. Mackey and his colleagues suggest that the landslide originated on the southwestern flank of 1213-meter-tall Nefus Peak. Pointing to a 100-meter-deep scar on the peak's heavily forested slope, they estimate that 36 million cubic meters of rock—enough to fill the Superdome in New Orleans more than 10 times—broke loose and rumbled down to form a 140-meter-tall natural dam.
By carbon-dating bits of charcoal taken from the layered portions of the terraces, the researchers found that the charcoal—tiny remnants of an ancient wildfire—may have been deposited along the lakeshore sometime between 22,600 and 25,700 years ago, suggesting that the lake was in place around that time. Results of the team's analyses appear online today in the Proceedings of the National Academy of Sciences.
The temporary lake theory explains a lot. Fish geneticists estimate that the summer-run and winter-run trout in the Eel River trout last interbred during the period when Mackey and his colleagues suggest the lake would have existed. The time span also roughly matches an extended interval when Eel River sediments are missing from offshore deposits because they got trapped behind the dam, they note. Previously, scientists had suggested sediments were missing due to short-term changes in ocean currents, Mackey says.
Because none of the data are precise, it's tough to estimate how long the landslide dam persisted, Mackey says. But all of the signs suggest that the dam lasted at least several centuries and possibly for a few thousand years.
"You typically don't find lake sediments high in a river valley," says Harvey Kelsey, a geomorphologist at Humboldt State University in Arcata, California. And the researchers' notion that the landslide slipped off Nefus Peak is convincing, he says. "Not finding the terraces downstream of there, that's the 'smoking gun.' "
"The idea that geological events can drive biological evolution is a major concept in the environmental sciences," says David Montgomery, a geomorphologist at the University of Washington, Seattle. He calls the genetic divergence of the Eel River's steelhead trout, a process that began once the dam eroded away and the fish began to spawn in separate parts of the river again, "an exciting snapshot of evolution in action."
"The team's study," Montgomery adds, "is also a tremendous example of how access to high-resolution digital topographic data can open new vistas on landscapes we thought we knew well."