Author(s): Stephane Bertin, Heide Friedrich
Linked Author(s): Heide Friedrich
Keywords: Roughness, remote sensing, photogrammetry, DEM, statistical analysis
Abstract: Stable fluvial armours, shaped by surface coarsening during selective sediment transport, received considerable attention over the years. Stable armouring is important in river engineering studies. For example, a classical problem is the riverbed degradation downstream of a dam. Our knowledge of whether stable armours can develop with limited sediment supply is still insufficient, yet this is a condition found in many natural gravel-bed rivers. Practically, given a suitable timeframe that allows sediment transport to reduce to approximately zero under a constant flow of water, stable armours can be re-created in laboratory flumes, allowing controlled studies with precise measurements. This study examines the extent to which armour structure is replicable under identical flow and bulk sediment composition. Two sets of experiments were performed using two different bulk sediment mixtures. Unstructured gravel beds were prepared in a laboratory flume and were water-worked successively with two constant discharges until the formation of stable armours. No sediment was fed and selective sediment transport prevailed. Grain-scale digital elevation models (DEMs), as well as bed-surface and bedload compositions, were obtained to quantify the changes due to armouring and to identify the formative parameters. We found bed structure to be more responsive to changes in flow discharge than bed-surface composition, and both armour composition and surface structure were unique given identical formative parameters. This finding is significant as it shows that bed composition alone is not sufficient to describe armour roughness. We discuss the relationships between a fully-developed stable armour and the flow and sediment forming it. The relationships examine the effects of varying the formative parameters onto the armour properties (e. g. , composition and roughness), which were extended with the addition of extensive data from previous research
Year: 2017