Modelling the Influence of Riparian Vegetation on River Bank Erosion

Despite the recognition that riparian vegetation influences riverbank stability, many of the mechanical and hydrological mechanisms involved are yet to be fully quantified. In particular, although empirical research (e.g. Abernethy and Rutherfurd, 2001; Simon and Collison, 2002) has advanced our knowledge of the relative importance of each of these mechanisms in certain environments, results are not necessarily transferable. This is because vegetation influences bank stability via a complex suite of process mechanisms that vary in effectiveness as a function of environmental characteristics. The net effects on bank stability at an individual site are therefore difficult to predict. In the face of such complexity, numerical modelling provides a useful adjunct to empirically derived data. Modelling provides a means of analysing and isolating the influence of different vegetation assemblages, across a variety of different bank material types and physiographic settings. Herein we report preliminary results from an integrated field monitoring and numerical modelling study undertaken at six special study sites within the United Kingdom. At each site, field data have been collected to parameterise coupled simulations of pore water seepage dynamics and bank failure processes. The requisite data include bank slope surveys, in situ geotechnical testing, laboratory analyses of sedimentary and hydraulic properties of the bank materials, root tensile strength testing, and surveys of sub-surface root networks. These data are supplemented by archived hydrological and climatological data sets, enabling analyses to be performed for a range of specific flow events at each site. For each event, saturated/unsaturated flow within the riverbank is modelled using a finite element code (SEEP/W v. 5) for the seepage analysis in transient conditions (Geo-Slope International, 2001). In these simulations, model boundary conditions are adjusted to account for the hydrological effects of vegetation by adjusting the amount of rainfall infiltrating into the bank during storms. The pore water pressure data computed at each time step in the simulation are then used, together with the observed geotechnical data adjusted for the mechanical effects of root reinforcement and vegetation surcharge, in a limit equilibrium river bank stability analysis (SLOPE/W v. 5). In the stability analysis, the geometry of the most probable failure mechanism and overall factor of safety (a measure of net bank stability) are computed (Geo-Slope International, 2001). The study sites were chosen to include a wide range of river and bank characteristics (bank morphology, sediment types, vegetation assemblages, hydrological regime and climatic setting), so changes in dominant bank erosion mechanisms and stability as a function of changing environmental characteristics at each site can be computed and isolated. Moreover, the parameterised models are used to undertake controlled numerical experiments to quantify the effects on bank stability and failure mechanisms of changes in riparian vegetation.