Inicio  /  Water  /  Vol: 10 Par: 4 (2018)  /  Artículo
ARTÍCULO
TITULO

Comparing the Sensitivity of Bank Retreat to Changes in Biophysical Conditions between Two Contrasting River Reaches Using a Coupled Morphodynamic Model

Yannick Y. Rousseau    
Pascale M. Biron and Marco J. Van de Wiel    

Resumen

Morphodynamic models of river meandering patterns and dynamics are based on the premise that the integration of biophysical processes matching those operating in natural rivers should result in a better fit with observations. Only a few morphodynamic models have been applied to natural rivers, typically along short reaches, and the relative importance of biophysical parameters remains largely unknown in these cases. Here, a series of numerical simulations were run using the hydrodynamic solver TELEMAC-2D, coupled to an advanced physics-based geotechnical module, to verify if sensitivity to key biophysical conditions differs substantially between two natural meandering reaches of different scale and geomorphological context. The model was calibrated against observed measurements of bank retreat for a 1.5 km semi-alluvial meandering reach incised into glacial till (Medway Creek, Ontario, Canada) and an 8.6 km long sinuous alluvial reach of the St. François River (Quebec, Canada). The two river reaches have contrasting bed and bank composition, and they differ in width by one order of magnitude. Calibration was performed to quantify and contrast the contribution of key geotechnical parameters, such as bank cohesion, to bank retreat. Results indicate that the sensitivity to key geotechnical parameters is dependent on the biophysical context and highly variable at the sub-reach scale. The homogeneous sand-bed St. François River is less sensitive to cohesion and friction angle than the more complex Medway Creek, flowing through glacial-till deposits. The latter highlights the limits of physics-based models for practical purposes, as the amount and spatial resolution of biophysical parameters required to improve the agreement between simulation results and observations may justify the use of a reduced complexity modelling approach.