EXPERIMENTAL STUDY OF THE INFLUENCE OF LARGE BLOCKS ON FLUVIAL RESPONSE TO BASELEVEL FALL

Much recent work has shown that sediment size distribution significantly influences river incision, yet few studies have explored the role of the largest sediment grains in changing the dynamics of fluvial erosion. Recent numerical modeling supports the hypothesis that large, hillslope-derived blocks have the potential to inhibit fluvial erosion through a combination of bed cover and reduction of bed shear stress. The presence of blocks may inhibit channel adjustment to baselevel lowering, make knickpoint retreat rates both slower and more variable, and preclude accurate reconstruction of baselevel history. To test this hypothesis experimentally, we use a narrow, quasi-1-D bedrock flume to study river profile response to baselevel fall in the presence of large block delivery. We compare control experiments without blocks to experiments in which blocks are added at a rate proportional to the bed lowering rate. Preliminary results indicate that knickpoint height, shape, and retreat rates are relatively steady in the absence of blocks. However, when blocks are added to the channel in response to rapid bed erosion, knickpoint height and shape become highly variable, and knickpoint retreat rates are both ~50% slower and much more variable than in the control case. The importance of blocks as erosion inhibitors depends strongly on block erodibility and the baselevel lowering rate applied to the flume. We compare our flume results with a simple numerical model, and find that while blocks inhibit fluvial adjustment to baselevel fall in both cases, the flume results exhibit variability that is not captured by the numerical model. Together, experimental and numerical results show that block delivery to rapidly eroding rivers is in many cases an important and poorly understood control on knickpoint retreat, channel response to perturbations, and landscape evolution.