Factors affecting scour of previously captured sediment from stormwater catchbasin sumps

Humberto Avila, Robert Pitt and S. Rocky Durrans


The sediment-retaining performance in conventional catchbasin sumps has been reported to be in the wide range between 14 and 99% (Metcalf & Eddy 1977); obviously, the higher performance is obtained by combining low flowrates, large particle sizes, and high specific gravities. Typically, about 30% of the total stormwater particulates are captured during actual rainfall tests (Pitt 1985). The accumulation rate, or sediment-retaining performance, depends on the size and geometry of the device, the flow rate, sediment size, and specific gravity of the sediment. In the same way, scour phenomenon includes all those parameters previously mentioned, in addition to the depth of the water protection layer and the consolidation of the sediment bed due to the aging phenomenon.

In order to evaluate the importance of the parameters and their interactions on the phenomenon of scour or migration of sediment out of a conventional inlet catchbasin, an experimental design was developed and analyzed with four parameters: flow rate, sediment size, overlying water protection depth, and specific gravity of the sediment. A 2-dimensional Computational Fluid Dynamic (CFD) model was implemented in Fluent 6.2, using the Eulerian multiphase model, with which it is possible to include two phases: an upper layer of water and a submerged dense layer of sediment. When examining the loss of sediment after 1,000 sec of continuous flow (16 min), the results showed that the expected parameters, such as flow rate and sediment size, are important factors that explain sediment scour. Also, the water protection depth over the sediment was identified as an important factor which is related to the extent of exposure of the sediment layer to the in-flowing water. However, it was also found that specific gravity was not as important as the sediment particle size, or the water protection in the prediction of sediment scour.

In the case of 10 L/s (130 GPM) flows, an important difference of shear stress magnitude as a function of the sediment elevation occurs. This flow is a typical high flow that would occur about once a year in the southeastern US from a 0.4 ha paved parking area, a typical drainage area for the catchbasin sump being modeled. The increasing rate of shear stress at different elevations is also consistent with the development of the velocity field. The protecting water layer above the sediment bed becomes important for the smaller flow rates. Particle sizes smaller than 600 µm are exposed to initial suspension at 0.6 m, and particles smaller than 100 µm would be exposed to initial suspension at 0.8 and 1.0 m at a 10 L/s flow rate.

These scour observations are similar to what has been observed during field tests of manholes in the past. The next stage in this research will be directly measuring the velocity fields in a laboratory full-sized manhole to confirm these calculations, followed by selected laboratory scour tests for further confirmation. Finally, the results will be implemented in the WinSLAMM stormwater model to better consider sediment scour from small hydrodynamic devices.

Permanent link: