Bioretention is the go-to stormwater control measure in many urban areas. Bioretention success results from the interaction of the loading (flow and pollutants), the treatment media, and the vegetation. Working with the City of Lancaster, PA, Penn State University and Franklin and Marshall College researchers tested infiltration and nutrient removal performance in 100 bioretention systems throughout the city, and did a deep dive into understanding the performance of two systems in Brandon Park where salt loadings in the winter were different due to their proximity to the road. From the study of 100 basins, seasonally, the infiltration rate was higher in the summer, as was the resistance to penetration, likely due to the impact of the densely-packed roots in most of these systems and not media compaction. The soil moisture and electrical conductivity, as expected, were higher in the winter with soil moisture increasing as the electrical conductivity increased. Weak but significant correlations were found between soil moisture and number of plant species and also with canopy height. Due to the vegetation, as expected, nutrient removal from the water was seen in the summer, but not in the winter after the vegetation was harvested. The soil moisture sensors in the two Brandon Park bioretention systems showed that, in the winter, the soil moisture content changed as a function of rainfall and not snowfall. This resulted in the delay of the signal of the deicing salts until there was sufficient rain to move water into the basin, which may be 6 – 7 days. The field results also supported prior laboratory tests indicating that deicing salt was “cleaning” the bioretention media of previously-trapped metals during the winter months. To ensure that the system is robust, the designer needs to understand and address the interactions of pollutant loading, media selection, and plant palette.