PS 22-23 - Estimating carbon derived from deadwood in soil pools over eight years of wood decomposition

Background/Question/Methods

Quantifying the proportion and timing of carbon (C) transferred to soil from deadwood during decomposition is critical to terrestrial ecosystem C modelling. As wood decomposes, substantial proportions of C may be transferred to soil. However, little is known regarding the ultimate amounts incorporated into soil and how much is unprotected in labile and surface soil pools versus protected in mineral-associated and subsoil pools. Differences in C transferred to soil from deadwood of species with varying biochemical properties are also poorly understood. Soil collected to 10 cm under decaying logs of trembling aspen, paper birch, and loblolly pine with distinct δ¹³C signatures was analyzed for C isotopes to determine the amounts of soil C derived from wood after 4 and 6 years of deployment in a South Carolina forest. Additionally, 0-20 cm and 20-30 cm samples collected beneath pine logs after 8 years were analyzed to assess C movement from wood into surface and subsoil. Labile C in soil was distinguished from protected, mineral-associated C by separating soils into free light (FLF), occluded light (OCC) and heavy (HF) fractions using density separation. A two-endpoint mixing model was used to estimate the proportion of C in each fraction derived from decaying logs.

Results/Conclusions

After 4 years, the proportion of soil C to 10 cm depth derived from aspen logs was significantly greater than that for birch and pine. Surface soil C in FLF derived from the wood above for aspen (11.7%) was more than twofold that of the other species. Across species, wood C in FLF (7.41%) and OCC (5.52%) accounted for more wood C in soil than the protected HF. Similarly, large amounts of wood C were found in the FLF (5.98%) and OCC (4.60%) in year 6, however, more wood C was present in soils under pine, with amounts of wood C in surface soil under aspen and birch declining from year 4 values. At 8 years, wood C in surface and subsoil occurred mainly in the HF. Our findings suggest that the early years of deadwood decay lead to wood C transfer to the surface labile and occluded C pools, particularly from faster decomposing species, and this C may be more vulnerable to further decomposition associated with climate change and forest management impacts. In the longer term, susceptibility to further release is lowered as more wood C is transferred to the protected, mineral associated fraction in both surface and subsoils.