Three permeable pavements were evaluated for their ability to improve the quality of stormwater runoff over a 22-month period in Madison, Wisconsin. Using a lined system with no internal water storage, permeable interlocking concrete pavers (PICP), pervious concrete (PC), and porous asphalt (PA) were able to significantly remove sediment and sediment-bound pollutant loads from runoff originating from an asphalt parking lot five times larger than the receiving permeable pavement area. Reductions in total suspended solids were similar for all three surfaces at approximately 60 percent. Clogging occurred after approximately one year, primarily due to winter sand application that led to high sediment load in spring runoff. Winter road salt application resulted in high chloride load that was initially attenuated in all three permeable pavements but later released during subsequent spring runoff events. Total phosphorus load was reduced by nearly 20 percent for PICP and PA, and 43 percent for PC. These values were likely tempered by the export of dissolved phosphorus observed in PICP and PA, but not PC. Average removal efficiencies for metals were 40, 42, and 49 percent in PA, PICP, and PC, respectively. A median pH of 10.2 in PC effluent could explain elevated removal efficiency of phosphorus and select metals in PC over PICP and PA (median = 7.5 and 7.8, respectfully) through enhanced precipitation. Elevated pH values in PC may also have led to higher removal efficiencies for select metals than PICP or PA. The environmental benefits as well as potential unintended consequences of stormwater practices like permeable pavement that utilize infiltration as a form of treatment warrant consideration in management of urban runoff.