Potential effects of structural controls and street sweeping on stormwater loads to the lower Charles River, Massachusetts
The water quality of the lower Charles River is periodically impaired by combined sewer overflows (CSOs) and non-CSO stormwater runoff. This study examined the potential non-CSO load reductions of suspended solids, fecal coliform bacteria, total phosphorus, and total lead that could reasonably be achieved by implementation of stormwater best management practices, including both structural controls and systematic street sweeping. Structural controls were grouped by major physical or chemical process; these included infiltration-filtration (physical separation), biofiltration-bioretention (biological mechanisms), or detention-retention (physical settling). For each of these categories, upper and lower quartiles, median, and average removal efficiencies were compiled from three national databases of structural control performance. Removal efficiencies obtained indicated a wide range of performance. Removal was generally greatest for infiltration-filtration controls and suspended solids, and least for biofiltration-bioretention controls and fecal coliform bacteria. Street sweeping has received renewed interest as a water-quality control practice because of reported improvements in sweeper technology and the recognition that opportunities for implementing structural controls are limited in highly urbanized areas. The Stormwater Management Model that was developed by the U.S. Geological Survey for the lower Charles River Watershed was modified to simulate the effects of street sweeping in a single-family land-use basin. Constituent buildup and washoff variable values were calibrated to observed annual and storm-event loads. Once calibrated, the street sweeping model was applied to various permutations of four sweeper efficiencies and six sweeping frequencies that ranged from every day to once every 30 days. Reduction of constituent loads to the lower Charles River by the combined hypothetical practices of structural controls and street sweeping was estimated for a range of removal efficiencies because of their inherent variability and uncertainty. This range of efficiencies, with upper and lower estimates, provides reasonable bounds on the load that could be removed by the practices examined. The upper estimated load reduction from combined street sweeping and structural controls, as a percentage of the total non-CSO load entering the lower Charles River downstream of Watertown Dam, was 44 percent for suspended solids, 34 percent for total lead, 14 percent for total phosphorus, and 17 percent for fecal coliform bacteria. The lower estimated load reduction from combined street sweeping and structural controls from non-CSO sources downstream of Watertown Dam, was 14 percent for suspended solids, 11 percent for total lead, 4.9 percent for total phosphorus, and 7.5 percent for fecal coliform bacteria. Load reductions by these combined management practices can be a small as 1.4 percent for total phosphorus to about 4 percent for the other constituents if the total load above Watertown Dam is added to the load from below the dam. Although the reductions in stormwater loads to the lower Charles River from the control practices examined appear to be minor, these practices would likely provide water-quality benefits to portions of the river during those times that they are most impaired-during and immediately after storms. It should also be recognized that only direct measurements of changes in stormwater loads before and after implementation of control practices can provide definitive evidence of the beneficial effects of these practices on water-quality conditions in the lower Charles River.