AbstractForested filter zones (FFZ) are being used more frequently for remediation of agricultural non‐point source pollution. The objective of this study was to determine the effects of short‐term dispersal (1–2 yr) of agricultural runoff on the denitrification potential of the soil microbial population and denitrification rates, to a depth of 1 m, in forest soils in two small watersheds (W1 and W2) in the Piedmont of North Carolina. Each watershed consisted of a field and a FFZ. Denitrification potential was measured in a series of soil slurry incubations of soils from inside the FFZ that received agricultural runoff and from soils immediately adjacent to the FFZ that received no runoff (control). Soils were amended with both glucose and nitrate (G + NO3) to ensure adequate supply of substrate and energy source. Denitrification rates were measured at ambient C conditions in a similar incubation with only NO3‐N amendment (NO3). We measured NO3‐N disappearance in both incubations and reported loss as a percentage of initial concentrations. For the FFZ soils, >80% of the added NO3‐N was lost in the G + NO3 incubation from soils from the upper 50 cm in W1 and from the upper 30 cm in W2. In control soils, high levels of NO3‐N loss were observed in only the upper 20 cm of the profile in W1, and in W2 surface soils had significantly lower denitrification potential than FFZ soils at all depths. Denitrification potential was greatly enhanced (P = 0.05) throughout the entire first 100 cm in the first FFZ and in the surface 40 cm in the second FFZ. Denitrification rates under ambient C conditions were higher (>40%) in the surface 20 cm of the profile of the FFZ in W1, compared with the unexposed control (∼20%), but no enhancement was observed on W2. Exposure of soil to agricultural runoff had a significant impact on the soil microbial community. Denitrification potential in subsoil was limited by the absence of denitrifiers in unexposed soils, but subsoils exposed to agricultural runoff had a significant denitrifier population. The fact that higher denitrification potential did not translate to higher denitrification rates in these incubations indicates that C availability limited the denitrification process at all depths in these Piedmont forest soils.