Changes in soil carbon stocks following forest harvest can be an important component of ecosystem and landscape-scale C budgets in systems managed for bioenergy or carbon-trading markets. However, these changes are characterized less often and with less certainty than easier-to-measure aboveground stocks. We sampled soils prior to the whole-tree harvest of Watershed 5 at the Hubbard Brook Experimental Forest in 1983, and again in years 3, 8, and 15 following harvest. The repeated measures of total soil C in this stand show no net change in the O horizon over 15 years, though mixing with the mineral soil reduced observed O horizon C in disturbed areas in post-harvest years 3 and 8. Mineral soil C decreased by 15% (20 Mg ha−1) relative to pre-harvest levels by year 8, with no recovery in soil C stocks by year 15. Proportional changes in N stocks were similar. The loss of mineral soil C offset two-thirds of the C accumulation in aboveground biomass over the same 15 years, leading to near-zero net C accumulation post-harvest, after also accounting for the decomposition of slash and roots. If this result is broadly representative, and the extent of forest harvesting is expanded to meet demand for bioenergy or to manage ecosystem carbon sequestration, then it will take substantially longer than previously assumed to offset harvest- or bioenergy-related carbon dioxide emissions with carbon uptake during forest regrowth.
