A self-limiting bank erosion mechanism? Inferring temporal variations in bank form and skin drag from high resolution topographic data

Fluvial bank erosion rates are often quantified by assuming that the erosion rate is a function of the excess (above a critical threshold) boundary shear stress applied by the flow. Research has shown that the form roughness induced by natural topographic bank features, such as slumps, spurs and embayments, is the dominant component of the spatially-averaged total shear stress, meaning that form roughness provides an important control on bank erosion rates. However, measuring the relative components of the total shear stress for a natural system is not straightforward. In this paper we employ the method of Kean and Smith (2006a,b) to partition the form and skin drag components of river bank roughness using a time series (2005-2011) of high-resolution topographic surveys of an eroding bank of the Cecina River in central Italy. This method approximates the form drag component of the roughness along a longitudinal bank profile as a series of user defined Gaussian curves. The extracted metrics are used in conjunction with an estimate of the outer region flow velocity to partition the form and skin drag components of the total boundary shear stress according to the Kean and Smith analytical solution. The relative magnitude of the form and skin shear stress at each survey date is analysed alongside DEMs of difference to reveal that intense episodes of erosion are followed by periods of quiescence. We show that this is due to the protection offered by increased form drag roughness following erosion. We conceptualise the dynamic feedbacks that exist between river discharge, bank erosion processes and bank form roughness, into a simple model of the self-limiting nature of river bank erosion