Abstract
The purpose of tracking routes and dispersal rates of bedload is to show the response of sediment to discharge events, as driven by meteorological phenomenon. The environment chosen for this bedload tracing experiment is Potters Creek, which is an alluvial fan adjacent to Franz Josef in Westland, where 285 tracer particles (ranging from 81 – 1,236 g, average 395.7 g) had radio frequency identification tags placed in each clast to trace individual mobilisation rates. Over 9 months the maximum transfer distance was 2,144 m, with evidence from a small cohort of recovered particles indicating an abrasion rate equivalent to 7 g per 500 metres travelled. Compared to their original mass, semi-schist clasts experienced higher quantities of abrasion (6.6%) compared to higher grade schist clasts (2%), as did tracer particles that were transferred from the proximal zone (10.2%) compared to mid-fan (3.7%) or distal zone (1.14%) transported sediment. The median sediment texture (D50) decreases downstream from 99 mm in the proximal zone to 39 mm in the distal zone. Sediment has accumulated in the mid-fan zone due to a lateral injection of sediment from a tributary, increasing the D50, which requires stream competency and capacity to adapt for transfer to occur. Despite a D50 increase in the mid-fan the sediment size does reduce throughout the catchment, showing evidence of rounding and fining downstream. Axial comminution shows that the rate of fining from each axis is higher at the upstream sites where slope is highest, where axis lengths are on average 2 times higher here compared to downstream reaches. Topographic cross sections show that near the tributary is an aggrading site, causing the alluvial fan to develop a concave profile in the mid-fan zone, these cross sections also show aggradation and degradation zones alternate downstream. Depth-duration frequency analysis of precipitation data during this experiment shows that there have been 113 events that have exceedance return periods of between 2 – 50 years for durations ranging from 0.5 hours – 72 hours, with one additional event exceeding a 100-year event. Maximum discharge recorded during this experiment was 53.54 m3s-1 and with 12 other events recorded that exceed 5 m3s-1, and these events are of a magnitude to episodically transfer sediment downstream. The identification of rainfall patterns show that quick and intense rainfall events contribute to rapid discharge increase, where critical shear strengths can be overcome rapidly and disperse sediment episodically and that has been captured by the experimental design.