A data-driven, uncertainty-bound estimation technique for bedload transport rates is developed based on passive sensing devices. The model converts sediment samples to a mass in transit for each instantaneous discharge according to impacts detected and a Monte Carlo simulation of the load determined at random from the particle size distribution. Using impact count data autogenically produces a supply-limited, location-specific and high-resolution time-series of bedload rates, while the probabilistic approach inherently accommodates the stochastic nature of bedload transport. Application to the River Avon (Devon, U.K.) provides cross-sectional bedload rate estimates within the bounds of experimental data and calibrated to observed field behaviour. This new procedure offers an alternative ‘class’ of bedload estimation to existing approaches and has the potential for wide-ranging applications in river management and restoration, while contributing to the integration of ‘big data’ into a progressive agenda for hydrogeomorphology research.
展开▼
