Abstract
Understanding soil organic carbon (SOC) responses to land-use changes requires knowledge of the sizes and mean residence times (MRT) of specific identifiable SOC pools over a range of decomposability. We examined pool sizes and kinetics of active and slow pool carbon (C) for tropical forest and grassland ecosystems on Barro Colorado Island, Panama, using long-term incubations (180 days) of soil and stable C isotopes. Chemical fractionation (acid hydrolysis) was applied to assess the magnitude of non-hydrolysable pool C (NHC). Incubation revealed that both grassland and forest soil contained a small proportion of active pool C (<1%), with MRT of ~6 days. Forest and grassland soil apparently did not differ considerably with respect to their labile pool substrate quality. The MRT of slow pool C in the upper soil layer (0–10 cm) did not differ between forest and grassland, and was approximately 15 years. In contrast, changes in vegetation cover resulted in significantly shorter MRT of slow pool C under grassland (29 years) as compared to forest (53 years) in the subsoil (30–40 cm). The faster slow pool turnover rate is probably associated with a loss of 30% total C in grassland subsoil compared to the forest. The NHC expressed as a percentage of total C varied between 54% and 64% in the surface soil and decreased with depth to ~30%. Grassland NHC had considerably longer MRTs (120 to 320 years) as compared to slow pool C. However, the functional significance of the NHC pool is not clear, indicating that this approach must be applied cautiously.
Published Version (Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have