More on the Braudrick et al. work published last week: NPR's Science Friday made a nice little video, in which Bill Dietrich used the phrase "recipe for a river." Click on the image above to see it.
The recipe includes two sizes and densities of sediment, one quartz sand and the other plastic. And stopping the run no less than eleven times to replant alfalfa outside the bankfull channel, then waiting 7-10 days for it to grow.
After reading the paper and the suppliments in PNAS, I have a lot of questions. Two authors have replied to emails I sent (thanks), but I don't have the details yet. I'm most interested in why both lithic and thermoplastic media were used, and suspect its because plants don't grow well in the plastic media, something we've tried.
Here's video of an Emriver run we made in 2007. After setting conditions to simply visualize meander initiation from a straight channel, and getting the video we wanted, we cranked up the flow and injected dye to make interesting patterns. Strictly for fun. The patterns we got are sort of the opposite of what Braudrick et al. were looking for--a highly braided channel. Those guys were looking for a single thread channel with point bars connected to the floodplain.
I might argue such rivers aren't as common as we'd like to believe, and lots of gravel bed rivers tend more towards a braided form. But that's for another post.
UPDATE:This technical memo by Bill Dietrich for Stillwater Sciences covers details of modeling efforts leading up to the Braudrick paper.
UPDATE: A nicely done news article by Phil Berardelli in Science, with quotes from the authors and David Montgomery.