Exploring controls on the spatial distribution of bedrock exposure in a mixed bedrock-alluvial river system

Abstract

In mixed bedrock-alluvial river systems, the riverbed sediment cover depends on interactions between sediment supply, channel morphology and flow characteristics. The controls and dynamics of bedrock exposure are not well understood due to the complexity of interactions between hydraulics and sediment processes. These processes constantly modify the spatial location of bedrock and alluvial reaches over a timescale of hundreds of years. This study aims to understand how channel and sediment properties affect the spatial distribution of bedrock exposure in mixed bedrock and alluvial systems using a combination of numerical modelling and field data. The River Carron in the Scottish Highlands was chosen as the study area as it is a mixed bedrock-alluvial river system with available data on the percentage of sediment cover along the main channel. A numerical model – Network Sediment Transporter from Landlab – was used to simulate sediment transport and track sediment parcels in the River Carron. The river network is represented as a graph made up of nodes and links, in which links are reaches of 100 m. The model was updated to simulate sediment input in every timestep and to calculate the percentage of sediment cover in each reach. A sensitivity analysis of the model was peformed to evaluate how the input parameter values of initial sediment cover, flow depth, grain size, sediment input, and timestep length affect the sediment cover when the simulations achieved a steady state. Field data on topography, bankfull depth and width and grain size were used to estimate initial model input values. We ran the model in spin-up time to freely adjust parameters such as the spatial distribution of bed sediment grain size. The percentage of sediment cover resulting from the model was compared with field data. Scenarios were created by modifying sediment input, flow depth and grain size parameter values in the simulations to determine how these parameters affect the spatial distribution of bedrock exposure of the system. We found that the model did not represent the mixed bedrock-alluvial reaches, as it mostly separated the reaches into either 100% bedrock or 100% alluvial reaches. Numerical modelling is a useful approach to explore controls on the spatial distribution of bedrock exposure, given that data, particularly temporal data, from bedrock rivers are still scarce.