Sediment delivery through a vetiver buffer strip as affected by interactions between flow hydraulics and surface relief of deposited sediment
MetadataShow full item record
Vegetated buffer strips (VBS) are widely employed to reduce fluxes of eroding soil and associated chemicals, from hillslopes into waterways. However, there are a number of inconsistencies in reported research and the associated recommendations for use of VBS. The VBS efficiency is time-dependent and changes as sediment deposition builds up, adding to the complexity of the situation None of the current erosion models handles deposition in the zone of sediment accumulation upstream of the buffer, except when this is at a sufficiently early stage of net deposition, for its effect on flow to be negligible. This research extends the current understanding of the physical processes involved in sediment reduction and is used to improve the predictive ability of an erosion /deposition model. Experiments were carried out in the Griffith University Tilting-Flume Simulated Rainfall facility. Replicate experiments were conducted at 1, 3 and 5% slopes using a dense vetiver strip of 0.3 m length inserted into the flume. Surface flow was introduced in front of the buffer and the water surface elevations in front of, and behind, the buffer were recorded using thin boards covered by water-soluble dye. A Vertosol was then introduced as a slurry into the surface flow and the inflow and outflow of sediment was measured, together with the runoff rate. Sediment deposition in front of the buffer was measured using small tags introduced into the flow at different distances, and times, in front of the buffer. At the end of the sediment addition (~20 minutes), the elevation of the new water surface was again measured, using dyed boards. Surface elevation of the deposited sediment, both across and along the flume, were later recorded with dyed boards and with a surface relief device. The collected outflow and deposited sediment samples were analysed for particle size distribution. The vetiver strip caused a region of enhanced flow depth, upstream of the buffer. The region increased in depth and decreased in length with increasing slope. Deposition started at the beginning of the enhanced water depth and subsequently moved towards the buffer as the surface of the deposit approached the water surface and flow reached critical values. As slope increased, sediment was deposited closer to the buffer, moving into the buffer itself at 5% slope. Deposition rates varied from 0 to 0.3 g m.s-1. Sediment loads in the outflow increased slightly with time and were primarily in the 0.002 0.02 mm size range compared to the input sediment, which was primarily in the 0.02 2.00 mm range. Buffering action resulted in the mean deposition of 88.7 (±1.0), 89.3 (±2.1) and 94.0 (±0.5) % of the added sediment at 5, 3 and 1% slopes respectively, with the remaining 6 to 13 % lost in the runoff. Particle size of the deposited sediment increased slightly in coarseness towards from the buffer, but overall, was similar in size range to the added sediment. Surface relief measurements indicated non-uniform deposition with some rilling. Water depths, sediment concentrations and rate of deposition were simulated at different times during the deposition process using an erosion/deposition model and predictions compared with data from the flume experiments. Modifications to this model are on-going using these data and further flume runs are underway using two additional soil types. Chemical analyses of the runoff and sediment samples are also on-going.
Where waters meet
Applied Hydrology (Drainage, Flooding, Irrigation, Quality, etc.)