First flush, power law and particle separation diagrams for urban storm-water suspended particulates

Commensurate with development of in situ storm-water control and treatment is the need to quantify the delivery and granulometry of the suspended p articulate fraction in storm water. Consistent with this need, this study examined the so-called "first flush" phenomenon for suspended particles with a measured range from 2 to 75 mum (typically <50 mum), the appropriateness of a single-versus multiple-power-law model of particle-number density (PND), and the application of process selection diagrams for particle separation. In comparison to a defined concentration "first flush" during the early portion of the examined rainfall-runoff events, results indicate that a disproportionately high and in some cases a continuous suspended particle delivery phenomenon that followed the hydrology of the event occur-red. Such results suggest that the entire event may require treatment, not solely the commonly designated "first flush" based on indices such as suspended solids. While a single-power law reasonably represented granulometric characteristics of suspended storm-water particulates, and in theory a continuously size-based power law is the most accurate representation; within the given suspended particle-size range a multiple-power law provided reasonable simplicity and accuracy for total PND, surface area, and particulate volume. Despite a wide range of hydrologic conditions for a series of nine rainfall-runoff events examined process selection diagrams based on the number-volume mean size (I v) and total PND led to a similar conclusion. Based on the combination of I-nv and PND in the urban catchment sedimentation in a typical transportation land use drainage facility is capable of removing 90% of particulate matter by mass within a detention time of 120 min.