Short-term pulses of increased N leaching typically follow the harvest of forest stands, but the magnitude of these pulses after whole-tree (WT) and conventional bole-only (BO) harvests remain difficult to predict on many sites. In this study we measured leaching for up to 6 and 8 years post-harvest on two Douglas-fir sites: a high productivity site with a silt loam soil and a low productivity site with a loamy sand soil. Whole-tree and bole-only harvest treatments were compared on each site, and presence/absence of 5 years of vegetation control (VC) was tested at one site. The magnitude and duration of post-harvest N-leaching pulses differed by site. At the high-productivity site, N leaching between years 3 and 8 post-harvest totaled 250 and 94 kg N ha^-1 in BO and WT harvest treatments with VC, respectively. At the low-productivity site, N leaching totaled 32 and 17 kg N ha^-1 from year 3 through year 6 post-harvest in BO and WT harvest treatments with VC, respectively. In both BO and WT treatments, annual N leaching did not fall below 5 kg N ha^-1 until year 7 at the high-productivity site and until years 5 (BO) and 4 (WT) at the low-productivity site. In unharvested stands at both sites, annual N leaching was 2 kg N ha^-1 or less each year. Cumulative amounts of N leached among site/treatment combinations were small compared to the soil total N pool (0.5 to 2.5%). The N leaching patterns among treatments at these two sites suggest that differences in soil C:N ratio (i.e., N status), the amount of post-harvest vegetation regrowth, and the N content of harvest residues influenced the amount of N leached. Although the maximum post-harvest N leaching pulse of 250 kg N ha^-1 was much higher than those reported in most comparable studies, this level of N leaching is unlikely to be reached under operational conditions, where VC is less intensive.