Decomposition of soil and plant carbon from pasture systems after 9 years of exposure to elevated CO2: impact on C cycling and modeling.

Elevated atmospheric CO2 may alter decomposition rates through changes in plant material quality and through its impact on soil microbial activity. This study examines whether plant material produced under elevated CO2 decomposes differently from plant material produced under ambient CO2. Moreover, a long-term experiment offered a unique opportunity to evaluate assumptions about C cycling under elevated CO2 made in coupled climate-soil organic matter (SOM) models. Trifolium repens and Lolium perenne plant materials, produced under elevated (60 Pa) and ambient CO2 at two levels of N fertilizer (140 vs. 560 kg ha-1 yr-1), were incubated in soil for 90 days. Soils and plant materials used for the incubation had been exposed to ambient and elevated CO2 under free air carbon dioxide enrichment conditions and had received the N fertilizer for 9 years. The rate of decomposition of L. perenne and T. repens plant materials was unaffected by elevated atmospheric CO2 and rate of N fertilization. Increases in L. perenne plant material C : N ratio under elevated CO2 did not affect decomposition rates of the plant material. If under prolonged elevated CO2 changes in soil microbial dynamics had occurred, they were not reflected in the rate of decomposition of the plant material. Only soil respiration under L. perenne, with or without incorporation of plant material, from the low-N fertilization treatment was enhanced after exposure to elevated CO2. This increase in soil respiration was not reflected in an increase in the microbial biomass of the L. perenne soil. The contribution of old and newly sequestered C to soil respiration, as revealed by the 13C-CO2 signature, reflected the turnover times of SOM-C pools as described by multipool SOM models. The results do not confirm the assumption of a negative feedback induced in the C cycle following an increase in CO2, as used in coupled climate-SOM models. Moreover, this study showed no evidence for a positive feedback in the C cycle following additional N fertilization. [ABSTRACT FROM AUTHOR]