Sediment transport in urban runoff over grassed areas

Abstract

Grassed areas, such as grass filter strips, and grassed swales, have been used extensively for the protection against polluted urban runoffs. In order to design and maintain these systems, it is important to understand and model sediment behaviour in overland flow through grass. The existing models are very limited, particularly for urban runoff conditions (i.e. low concentrations of fine sediment). A laboratory investigation was carried-out aiming at development of a new method for assessment of sediment transport in non-submerged, steady, and uniform flow through grass for low to moderate concentrations of fine sediment. Two groups of experiments were conducted in order to study sediment deposition and detachment in controlled environment, named as the ‘deposition’ and the ‘detachment’ experiments, respectively. Artificial turf (Astroturf) was used to simulate natural grass, where no infiltration was allowed. Experiments were conducted for different flume slopes, grass densities, flow rates, sediment inflows, and sediment densities. Concentration and size of particles in the flow along the grass strip were monitored. During 21 deposition experiments, it was observed that concentration of sediment in runoff decreases exponentially with the distance down the strip, where small particles travel much further than coarse ones. During 4 detachment experiments, it was found that sediment detachment is negligible compared to deposition. The Kentucky model, developed mainly for rural runoff conditions, failed to assess the measured sediment transport rates. Dimensional analysis was carried out to establish the main non-dimensional parameters for assessment of particle trapping in grass. Using the results of this analysis and the laboratory data a new simplified relationship was established between particle fall number and percentage of particles trapped in the grass.