Mineral Soil Carbon in Managed Hardwood Forests of the Northeastern US

New England is characterized by extensive forest cover and large reservoirs of soil carbon (C). In northern hardwood forests, mineral soil C can account for up to 50% of total ecosystem C. There has been an increasing demand for forests to serve both as a C sink and a renewable energy source, and effective management of the ecosystem C balance relies on accurate modeling of each compartment of the ecosystem. However, the dynamics of soil C storage with respect to forest use are variable and poorly understood, particularly in mineral soils. For example, current regional models assume C pools after forest harvesting do not change, while some studies suggest that belowground mineral soil C pools can be affected by disturbances at the soil surface. We quantified mineral soil C pools in previously clear-cut stands in seven research or protected forests across New York, New Hampshire, Massachusetts, and Vermont. The ages of the sites sampled ranged from recently cleared to those with no disturbance history, with 21 forest stands represented in the study. Within each research forest studied, physical parameters such as soil type, forest type, slope and land-use history (aside from forest harvest) did not vary between the stands of different ages. Soil samples were collected to a depth of 60 cm below the mineral-organic boundary using a gas-powered augur and 9.5-cm diameter drill bit. Samples were collected in 10-cm increments in shallow mineral soil and 15-cm increments from 30-60 cm depth. Carbon, nitrogen (N), pH, texture and soil mineralogy were measured across the regional sites. At Bartlett Experimental Forest (BEF) in New Hampshire, mineral soil biogeochemistry in cut and uncut sites was studied at a finer scale. Measurements included soil temperature to 55 cm depth, carbon compound analyses using Py-GCMS and soil microbial messenger RNA extractions from mineral soil. Finally, we simulated C dynamics after harvesting by building a model in Stella, with a particular interest in the role that priming effects may play if C is transported from organic to mineral soil layers after forest harvest. Laboratory analyses were conducted at Dartmouth College and at the University of New Hampshire. For the regional study, mineral soil C and N concentrations, and in some cases, pools were highest at locations that had never been harvested. Although sites represented different stages of succession after clearing, there were no significant patterns over time since harvest. At BEF, soil temperature at 55 cm depth in a recently cleared stand was on average 1.5° C higher than surrounding forested sites between June and September, and shallower depths had greater temperature discrepancies. Our model, which was parameterized using published field data from Bartlett and Hubbard Brook forests, showed that inputs of labile C to mineral soil after harvest could prime the decomposition of preexisting mineral soil C and account for up to 40% of the observed difference in C pools between harvested and undisturbed sites.