Reconstructing Rivers


By reverse-engineering a meandering stream, researchers have shown two ingredients to be very important: vegetation to reinforce banks and prevent erosion, and sand to build point bars and block off cut-off channels and chutes. This knowledge will help stream restoration efforts in the future. Credit: Zina Deretsky, National Science Foundation

Many rivers naturally twist and turn through their terrain. But just what conditions create and sustain this curvature has eluded scientists for decades—until now. For the first time, researchers have combined the right ingredients to replicate that meandering in the lab, according to a new study in the Proceedings of the National Academy of Sciences. Their effort may lead to improved strategies for river restoration and management.

“It’s important to have a river that is sinuous and migrating,” says Christian Braudrick, a geomorphologist at the University of California at Berkeley and lead author of the study, particularly one that covers a variety of terrain and has deep cut banks and pools. That type of construction, he adds, is crucial for many species of wildlife—including half of the 1,200 species currently listed as endangered or threatened, according to a figure quoted on the American Rivers website.

To build their river, Braudrick and his team carved a straight—with the exception of one lone crook—56-foot long channel through an oversized and slightly sloped sandbox in a lab. Then, they planted alfalfa sprouts to simulate grasses and trees along its banks (an idea they got from related experiments at the University of Minnesota) and, after the water began to flow, introduced coarse and fine sediments: namely, sand to represent gravel, and tiny bits of plastic to act as sand. (The lightweight material was chosen to mimic fine sediment’s vulnerability to the powerful currents of natural rivers.) Finally, the concoction was left to mix and migrate for a total of more than 130 hours, the equivalent to 5-7 years in the wild.


Christian Braudrick and colleagues are the first to build a scaled down meandering stream in the lab that successfully coursed through its flood plain for 130 hours (representing 5 to 7 years of real time in the wild). The substrate seen behind Braudrick is composed of sand to represent real-life gravel; white light-weight plastic for sand, and alfalfa sprouts for deep-rooting vegetation. Credit: Zina Deretsky, National Science Foundation

As the hours passed, the team’s train set-size model began to build regularly spaced bends until it looked much like a miniature version of the fish-friendly rivers in the wild. The alfalfa and plastic bits appeared to support this healthy meandering: Vegetation reinforced the river's banks and prevented erosion while the simulated sand (i.e., the plastic bits) blocked the formation of offshoot channels.

The importance of the faux sand was especially surprising to the scientists. Prevailing wisdom has discouraged its use in restoration of some rivers because it can cover up the coarser gravel sediment in which salmon lay their eggs. But the team witnessed that sand can also be beneficial by plugging the path of chutes—that is, runaway channels that threaten to leave the river’s main thruway wider and shallower. The simulated sand also proved valuable in building sediment deposits, called “point bars,” found inside a river’s natural bend.

Braudrick and his team hope the lessons they’ve learned can eventually be applied to making repairs to real waterways, as well as determining how best to mitigate the effects of climate and land use changes on rivers. “I think the most important part of our work is not what we have accomplished so far, but what we can do in future,” notes Braudrick. “Now we can ask specific questions about restoration possibilities, testing them in our self-formed channel before [attempting them in the wild].”