PhD Dissertation Abstract
For decades, secondary forests of the southeastern US accounted for ⅙ to ½ of the northern hemisphere terrestrial carbon sink. Ecological theory predicts that carbon uptake will slow as forests age and the southeastern US carbon sink will diminish, however long-term observations of forest carbon accumulation that might confirm or refute this pattern are rare and largely focus on live plant biomass. Detrital pools, which are characteristically large and slow to change, have an important role in storing carbon after live aboveground biomass is harvested or dies, buffering carbon losses to mortality. My dissertation compared the contribution of live and dead carbon pools to the carbon inventory of a secondary forest in the southeastern US.
In earning my Ecology Ph.D. from Duke University, I worked at a USDA Forest Service research site, the Long-Term Calhoun Soil-Ecosystem Experiment, where long-term measurements of plants, litter and soils have allowed documentation of the biogeochemical and ecosystem changes associated with old-field forest development. My research asked how carbon storage changed as the ecosystem progressed through different phases of development, and if the whole-ecosystem carbon inventory was buffered by accumulations of detrital carbon during periods of high aboveground mortality. My work combined the historical data with new experiments to address three questions related to long-term carbon balance.
First, I asked how much woody detritus accumulates in successional southeastern forests, and for how long. I inventoried four decay classes of coarse woody detritus, extracted decaying taproots to ~2m depth, and matched logs to historical stem maps to estimate time of death and decay rate. I found that despite rapid decomposition rates in the subtropical climate, woody detritus served as a large carbon storage pool for 5-10 years following large mortality events. (Mobley et al. 2012 CJFR)
Second, I asked how ecosystem development influenced soil organic matter pools. I measured changes in concentration and quality of dissolved organic matter through 2m of soil profile. I combined density and size fractionation and stable and radio-isotopic analyses of archived 60cm mineral soils to isolate soil fractions with different C and N concentrations, levels of microbial processing, and residence times. I found opposite patterns of increasing surface soil C and decreasing deep soil C that resulted in zero net SOC change over fifty years. The loss of deep soil carbon suggested either a priming effect of tree roots on soil organic matter, or a >50 year lagged land use legacy effect on deep soil carbon. (Mobley et al. 2014 GCB)
Finally, I asked how slow detrital pools contribute to whole-ecosystem carbon storage over time. I combined woody detritus and soil results with long-term aboveground biomass data to estimate whole ecosystem carbon storage over 50 years. I found that while carbon in live biomass, plant detritus, and shallow surface soils accumulated during ecosystem development, deeper soils lost carbon. This counters the idea that ecosystem recovery from cultivation includes recovery of soil carbon storage. Furthermore, carbon in dead biomass was insufficient to counter high tree mortality and rapid decomposition following self-thinning, such that the ecosystem switched from net C sink to net C source around age 35.
My work shows that large accumulations of carbon above ground do not guarantee similar changes below, and that in this aboveground biomass-dominated ecosystem, management strategies that maintain the ecosystem in an early successional state are likely the most effective for maximizing whole-ecosystem carbon storage. Allowing the ecosystem to undergo further development from an even-aged to a shifting mosaic steady-state forest will result in a lower magnitude of total ecosystem carbon storage.