Pulsing Flow in Steep Alluvial Streams

Channel bores observed in laboratory simulated floods in alluvial-bed channels resemble roll waves, but are apparently related to the existence and quasiperiodic breaking of trains of antidune waves associated with antidunes that in these experiments moved downstream. Experiments with rigid-wall and alluvial bank channels, rigid-bed channels, and varied inlet conditions show that the bores are not artifacts of the apparatus, and in these experiments always occurred in the antidune flow regime in which the antidunes migrated downstream. Release by sequential breaking from the upstream end of water stored in trains of antidune waves over antidunes apparently both initiates and nurtures the bores. Field observations of pulsing flow under comparable flow conditions suggests that thesee bores are a bona fide natural phenomenon and not a laboratory curiosity.