A reduced-complexity model for sediment transport and step-pool morphology

Abstract:

Step-pools are common morphological units in steep catchments. They are characterized by channel-spanning steps created by boulders, with downstream pools due to the tumbling water flow. Assessing stability of steps and describing sediment transport trends in these channels remain partially an open issue. Sediment transport and bed morphology in steep streams interact in a complex, non-linear way that makes difficult to predict them both with a satisfactory degree of accuracy. Wide grain-size distribution, complex boundary conditions, coupling with hillslopes, various flow resistance components are only few of the reasons why sediment transport estimates in steep streams differ from measurements by orders of magnitude.

In this work we model bed morphology and sediment transport in step-pool channels with a reduced-complexity (RC) approach. We develop a 2-D cellular-automaton sandpile-like model in which entrainment, transport and deposition of particles are modeled at a grain scale. Intuitive rules, requiring few parameters, are implemented, based on physical principles applied in a stochastic way. Probability of entrainment and deposition of grains (and, as a consequence, particle travel distances and resting times) are set up as function of flow condition and bed topography. Sediment input is fed at the beginning of the channel at constant or intermittent rate.

Despite its relative simplicity, our model yields realistic results in terms of sediment transport trends and particles travel distances. Phases of aggradation and degradation can be observed in the system even under a constant input. Sediment yield at the channel outlet shows intermittency as observed in natural streams. A stepped morphology is obtained in the simulations and the density of steps is inversely proportional to the width of the channel, as confirmed by previous studies. Our results encourage the development of RC approaches, as complementary tools to more sophisticated models: they provide a realistic description of complex morphological systems and help to better understand the physical principles that rule their dynamics.