EP54A-07
Exploring the role of flood transience in coarse bed load sediment transport

Friday, 18 December 2015: 17:30
2005 (Moscone West)
Colin B Phillips1, Michael Bliss Singer2, Kimberly M Hill3 and Chris Paola3, (1)University of Minnesota Twin Cities, Minneapolis, MN, United States, (2)University of St Andrews, St Andrews, United Kingdom, (3)University of Minnesota, Minneapolis, MN, United States
Abstract:
The rate of bed load transport under steady flow is known to vary both spatially and temporally due to various hydrologic and granular phenomena. Grain size distributions and riverbed properties (packing, imbrication, etc.) are known to affect flux for a particular value of applied flow stress, while hydrology is mainly assumed to control the magnitude of the applied bed stress above the threshold for bed material entrainment. The prediction of bed load sediment transport in field settings is further complicated by the inherent transience in flood hydrology, but little is known about how such flood transience influences bed load flux over a range of applied bed stress. Here we investigate the role of flood transience for gravel bed load transport through controlled laboratory experiments in a 28 m long 0.5 meter wide flume. We explore transient flow as the combination of unsteady and intermittent flow, where unsteady flow varies in magnitude over a given duration, and intermittent flow is characterized by turning the flow on and off. We systematically vary these details of flood hydrographs from one experiment to the next, and monitor the bed load as it varies with water discharge in real time by measuring sediment flux and tracking particles. We find that even with a narrow unimodal grain size distribution and constant sediment supply we observe hysteresis in bed load flux, different thresholds for entrainment and distrainment for the rising and falling limbs of a flood, and a threshold of entrainment that can vary one flood hydrograph to the next. Despite these complex phenomena we find that the total bed load transported for each flood plots along a linear trend with the integrated excess stress, consistent with prior field results. These results suggest that while the effects of transient flow and the shape of the hydrograph are measurable, they are second-order compared to the integrated excess stress.