Soil carbon pools, plant biomarkers and mean carbon residence time after afforestation of grassland with three tree species

Afforestation of grassland has been globally identified as being an important means for creating a sink for atmospheric carbon (C). However, the impact of afforestation on soil C is still poorly understood, due to the paucity of well designed long-term experiments and the lack of investigation into the response of different soil C fractions to afforestation. In addition, little is known about the origins of soil C and soil organic matter (SOM) stability after afforestation. In a retrospective study, we measured C mass in the soil light and heavy fractions in the first 10 years after afforestation of grassland with Eucalyptus nitens, Pinus radiata and Cupressus macrocarpa. The results suggest that C mass in the soil heavy fraction remained stable, but the C mass in the light fraction decreased at year 5 under three species. Soil δ13C analysis showed that the decrease in the light fraction may be due to reduced C inputs from grassland species litter and low inputs from the still young trees. After the initial reduction, the recovery of soil C in the light fraction depended on tree species. At year 10, an increase of 33% in light fraction soil C was observed at the 0–30 cm depth under E. nitens, compared to that under the original grassland (year 0). Planting P. radiata restored light fraction soil C to the original level under grassland, whereas planting C. macrocarpa led to a decrease of 33%. We concluded that the increase of light fraction soil C between year 5 and 10 is most likely due to C input from tree residues. Most of the increased C was derived from root turnover under pine and from both root and leaf turnover under E. nitens, as indicated by plant C biomarkers such as lignin-derived phenols and suberin and cutin-derived compounds in the 0–5 cm soil layer. Modelling of soil ∆14C‰ suggested that SOM had a greater mean residence time at year 10 than year 0 and 5 due to increased relative abundance of recalcitrant plant biopolymers.

► The soil carbon mass in the light fraction decreased at year 5 after afforestation of grassland.
► After the reduction at year 5, the recovery of light fraction soil carbon at year 10 depended on species.
► Most of the increased soil carbon between year 5 and 10 was derived from root turnover under pine.
► Under E. nitens, both the root and leaf of trees increased soil light fraction carbon between year 5 and 10.
► Soil carbon had a greater turnover time at year 10 than year 0 due to increased recalcitrant biopolymers.