EP33A-1055
Particle mobility and bed surface adjustments on episodic sediment supply experiments

Wednesday, 16 December 2015
Poster Hall (Moscone South)
Carles Ferrer-Boix and Marwan A Hassan, University of British Columbia, Vancouver, BC, Canada
Abstract:
This research aims to explore how episodic sediment supply affects particle mobility and bed surface adjustments in mountain streams. We conducted a set of runs in a 1 m-wide, 18 m-long tilting flume. Seven consecutive runs, each lasting 40 hours, were conducted under constant flow (65 l/s) but varying sediment supply rates for a total duration of 280 hours. The feed rate for the runs was as follow: no feed (runs 1 and 7), constant feed of 2.1 g/m/s (runs 2 and 6), one pulse of 83 g/m/s (run 3), four pulses of 83 g/m/s (run 4) and two pulses of 83 g/m/s (run 5). The total mass of sediment supplied during each of runs 2-6 was 300 kg. The feed texture was identical to that of the original mixture (Dmin = 0.5 mm, Dmax = 64 mm, Dg = 5.65 mm and sg = 3.05) with a bed slope of 0.0218 m/m. Bed surface images of a 2 m-long reach in the middle of the flume were processed. Bed surface areas covered by particle sizes coarser than 5.66 mm were automatically identified. Thus, we can easily obtain fractional particle mobility, i.e. how much bed area covered by a particular grain size changed at a given time.

Preliminary analyses of the experiments show that the bed surface texture systematically adjusts to each change in the sediment supply. Thus, (i) bed surface gradually coarsens during no feed runs 1 and 7, (ii) suddenly fines and subsequently coarsens after each episodic sediment supply event (runs 3-5) and (iii) remains approximately constant during runs 2 and 6 under constant feed conditions. Surface coarsening during run 1 was accompanied by the formation of bed structures and particle clusters. However, the amplitude of the changes of the bed surface texture is relatively small compared to the texture obtained after the first 40 h under no feed in run 1. This does not imply that the bed particles remain immobile. Conversely, bed particles on the surface move, bed structures loosen while the bed surface texture maintains, overall, the same grain size distribution. Preliminary results of run 3 (one sediment pulse between t = 0 h and t = 1 h) illustrate an asymptotic trend in the mobility of the bed: 63%, 75%, 80%, 83% and 86% of the bed area changed at t = 2 h, t = 7 h, t = 10 h, t = 20 h and t = 40h compared the bed surface at the beginning of the run, respectively.