Abstract Alongside hydraulic processes, animal burrowing can erode significant amounts of sediment from riverbanks and contribute to their mass failure. The interaction between hydraulic and burrowing effects has not been investigated, despite its potential importance in augmenting or reducing independent abiotic and biotic effects. To investigate this, we completed a series of flume experiments in which the erosion of simulated clay riverbanks was measured under three different flow conditions and in the presence and absence of a model burrowing animal, the signal crayfish ( Pacifastacus leniusculus ). Flow conditions were purposefully chosen to ensure hydraulic erosion and to reflect crayfishes' ability to navigate the flume substrate without hindrance, with some constraint and with significant difficulty. Total bank erosion was partitioned by mass according to erosion mechanism (burrow construction, mass failure and diffuse erosion) to derive quantitative measures of the relative importance of abiotic and biotic drivers under each experimental treatment. Bank erosion was dominated by crayfish burrowing under low flow conditions but by abiotic effects at high flows. However, erosion was greatest in the mid‐range of flow conditions tested, where moderate levels of erosion associated with burrowing and moderate hydraulic erosion combined to erode more sediment, exceeding the sum of the maximum observed biotic and abiotic components by 25%. Interactive effects were attributed to changes to crayfish burrowing behaviour in response to the flow and crayfishes' conditioning of the clay surface, which enhanced erosion when banks were exposed to hydraulic stresses. These results demonstrate for the first time that biologically driven erosion and geophysical erosion interact, in this case resulting in greater erosion than the sum of the processes operating independently. Such interactive effects likely occur across many zoogeomorphic systems. This is significant because it means that sediment mobilisation in biogeomorphic systems is unlikely to simply be the sum of the independent effects.
Sanders et al. (Fri,) studied this question.