Assessment of flow conditions in a new vortex-type stormwater retention pond using a physical model

The stormwater retention pond is a best management practice used for the improvement of runoff water quality before it discharges into larger surface waterbodies. A vortex-type retention pond, called the Nautilus PondTM, is a new design approach for stormwater retention ponds that is expected to produce an internal flow pattern in the pond that is more conducive to removal of sediments from runoff. Since many existing stormwater retention ponds were originally designed only for flood control, most of the ponds are subject to large dead zones, severe short-circuiting and short retention times, which can limit sediment retention in the ponds. In this study, the robustness of the design of the Nautilus PondTM was evaluated by assessing its residence time distribution (RTD) characteristics, flow pattern and sediment deposition patterns under various conditions of flow in the pond. The study was carried out in two physical scale models of a simplified Nautilus PondTM: one with a scale ratio of 1:30.775 for an aspect ratio of 100:2, and the other with a scale ratio of 1:13.289 for a pond of 50:2 aspect ratio. The aspect ratio is the ratio of the pond diameter at its water surface (top width) to the depth of flow, 2 m at corresponding design flow rates, in the pond. First, the RTD characteristics and flow patterns in the ponds were investigated using tracer mass recovery and flow visualization tests allowing different times for steady flow development (flow development time) for the design flows corresponding to 4 m3/s in the 100:2 prototype pond and 1 m3/s in the 50:2 pond. Then, tracer tests were carried out at different flow rates to investigate its effects on the RTD characteristics in both model ponds. The deposition patterns of approximately 50 micron sediment particles (prototype size) were also observed. The best position of a berm around the pond outlet was determined for the 100:2 pond by comparing the RTD characteristics and the sediment deposition patterns in the pond for three different positions of the berm. The residence time distribution characteristics and the sediment deposition pattern were also assessed for the 50:2 pond with a berm placed in a position equivalent to the best position identified in the 100:2 pond tests. It was found that the RTD curves at design flow rates of 4 m3/s and 1 m3/s for different flow development times were very similar to each other for both pond aspect ratios; the flow development time was found to have little effect on the flow characteristics of the ponds. The average baffle factors, short-circuiting indices and Morril dispersion indices were 0.41, 0.20 and 4.1, respectively, for the 100:2 pond aspect ratio, whereas these were 0.23, 0.05 and 8.6 for the 50:2 pond. The flow rate was found to have a significant effect on the RTD characteristics of both ponds. There were multiple peaks in the RTD curves for the lower flow rates tested for the 100:2 pond. This was thought to be a result of the low inflow momentum and high aspect ratio of the pond. As the flow rate was increased, the residence time distribution curve had a single, lower peak. In both ponds, an increase of flow rate caused the baffle factor and short-circuiting index to decrease and the Morril dispersion index to increase indicating that the inflow spent a shorter time in the pond. The sediment deposition pattern tests in both ponds without the berm around the outlet showed that a higher quantity of sediments deposited in the outer peripheral region of 100:2 pond. The 50:2 pond deposited a small amount of sediment along the periphery due to the high velocity inflow jet and lower aspect ratio of the pond. The best position of the berm among those tested was found to be at the 60% of pond bed radius from the center. Though the RTD characteristics for the 100:2 pond with different berm positions were very similar to each other, the 100:2 pond with the berm position at 60% of pond bed radius deposited most of the sediments outside the berm. The RTD characteristics in both ponds showed significant improvement with a berm at the 60% of radius position compared to the ponds without a berm. This improvement was more significant for the 50:2 pond than for the 100:2 pond. Further, the sediment deposition pattern in 100:2 pond with berm at 60% of bed radius showed that the larger sized sediment particles mainly deposited outside the berm and the finer particles deposited inside the berm. The 50:2 pond did not show any significant difference in particle size distribution of the sediments deposited inside and outside of the berm