In one chapter of my dissertation, I employed some very fun tools, like heavy liquids (at one point I made sand grains float!) and radioactive isotopes, to figure out what old-field forest succession does to soil carbon and nitrogen. What I learned is that shallow soils and deeper subsoils respond very differently. When the forest was planted in the old cotton fields, soil carbon and nitrogen began to decrease, because the trees were using the nitrogen and the carbon left over from the crops was decomposing faster than the forest was adding new dead leaves and roots. After about 20 years, after the forest started to accumulate a thick layer of leaf litter on the ground, the surface mineral soil started to increase in carbon. It wasn’t until the forest basically stopped growing after about 30 years (individual trees were still growing, but so many others were dying that the forest as a whole stopped accumulating biomass) that soil nitrogen stopped going down. Deeper in the soil, soil carbon decreased for about 40 years after the forest was planted. What we think happened there is that the forest’s deep roots lowered the water table, which allowed carbon that was previously too wet to decompose to now do so. At the same time, an abundance of little tree roots that die quickly and release their carbon provided energy to soil microbes to do the work decomposing that old carbon and releasing the associated nitrogen. So, 50 years after the loblolly pine forest was planted in an old cotton field, only the leaf litter layer has accumulated carbon. The mineral soil down to 60 cm below that has not changed, because the loss of carbon in subsoil canceled the gain in surface soil. The lesson is that it takes a long time (well over 50 years) for reforestation to result in an increase in mineral soil carbon.
This study seeks to investigate the dynamics of dead plant carbon over fifty years of old-field forest development at the Calhoun Long Term Soil-Ecosystem Experiment (LTSE) in South Carolina, USA. Emphasis is on the transition phase of the forest, which is less well studied than the establishment and early thinning phase or the steady state phase. At the Calhoun LTSE, the biogeochemical and ecosystem changes associated with old field forest development have been documented through repeated tree measurements and deep soil sampling, and archiving of those soils, which now allow us to examine changes that have occurred over the course of forest development to date.
In this dissertation, I first quantify the accumulation of woody detritus on the surface of the soil as well as in the soil profile over fifty years, and estimate the mean residence times of that detrital carbon storage. Knowing that large accumulations of C-rich organic matter have piled onto the soil surface, the latter chapters of my dissertation investigate how that forest-derived organic carbon has been incorporated into mineral soils. I do this first by examining concentrations of dissolved organic carbon and other constituents in soil solutions throughout the ecosystem profile and then by quantifying changes in solid state soil carbon quantity and quality, both in bulk soils and in soil fractions that are thought to have different C sources, stabilities, and residence times. To conclude this dissertation, I present the 50-year C budget of the Calhoun LTSE, including live and dead plant carbon pools, to quantify the increasing importance of detrital C to the ecosystem over time.
This exceptional long term soil ecosystem study shows that 50 years of pine forest development on a former cotton field have not increased mineral soil carbon storage. Tree biomass accumulated rapidly from the time seedlings were planted through the establishment phase, followed by accumulations of leaf litter and woody detritus. Large quantities of dissolved organic carbon leached from the O-horizons into mineral soils. The response of mineral soil C stocks to this flood of C inputs varied by depth. The most surficial soil (0-7.5cm), saw a large, but lagged, increase in soil organic carbon (SOC) concentration over time, an accumulation almost entirely due to an increase of light fraction, particulate organic matter. Yet in the deepest soils sampled, soil carbon content declined over time, and in fact the loss of SOC in deep soils was sufficient to negate all of the C gains in shallower soils. This deep soil organic matter was apparently lost from a poorly understood, exchangeable pool of SOM. This loss of deep SOC, and lack of change in total SOC, flies in the face of the general understanding of field to forest conversions resulting in net increases in soil carbon. These long term observations provide evidence that the loss of soil carbon was due to priming of SOM decomposition by enhanced transpiration, C inputs, and N demand by the growing trees. These results suggest that large accumulations of carbon aboveground do not guarantee similar changes below.
Mobley, Megan Leigh (2011): An Ecosystem Approach to Dead Plant Carbon over 50 years of Old-Field Forest Development. PhD Dissertation, Duke University, Durham, North Carolina. DOI: 10161/4992