Knowledge is presently lacking of the crucial spatio-temporal link between the instant of streambank failure and the entrainment and transport of the sediment comprising a failed block. Many blocks come to rest at the bank toe, but bank-normal motion of blocks may be accelerated by rotation or by elevated pore pressures at the block base, resulting in motion further into the channel. Conversely, motion may be arrested by impact and breakup, or retarded, prior to reaching the bank toe. Current modeling approaches tend to fall into two classes. After the instant of failure the failed block is either completely removed, entering the water column immediately, or segregated into its constitutive sediment fractions and held in a virtual "tank". Erosion of a particular size class from the bank or bank toe does not recommence until the size class has been completely removed from the "tank". In either case, the predicted spatio-temporal dynamics of a simulated stream are likely to be at best accelerated and at worst distorted. In response to these shortcomings, this paper presents preliminary results from experiments that investigate the effect of block size, fall height and apparent cohesion on the fate of failed blocks. The experimental apparatus permits blocks of up to 0.008 m~3 to be lifted to a height of 2.5 m and then dropped through a trap-door. Thus far, experiments have been limited to the zero-rotation case with impact onto a solid horizontal surface, but later experiments will incorporate block rotation and vary the properties of the receiving area. Controlled sequential photographs have been analyzed to extract surficial clast sizes and to produce fallout maps.
展开▼
