Contrasting responses of soil respiration to litter manipulation in subtropical Mytilaria laosensis and Cunninghamia lanceolata plantations

Abstract

C.lanceolata plots. The average Rsin litter exclusion, litter transplant and control treatments were 1.79, 2.18, 2.45μmol CO2m-2s-1in M.laosensis plots and 1.77, 2.29, 1.90μmol CO2m-2 s-1 in C.lanceolata plots, respectively. Overall, litter exclusion and transplant significantly decreased the annual fluxes under M. laosensis plots by 29. 8% and 14. 2%, respectively. Litter exclusion caused a slight decline in annual flux but litter transplant markedly increased the annual flux by 37. 8% in C.lanceolata plots. Using the decay constants of the two litter species in a reciprocal litter transplant decomposition experiment, priming effects were calculated as the amount of litter-derived respiration minus decomposed carbon from litter. Results showed that M.laosensis litter aroused higher priming effect than C.lanceolata litter, suggesting that the quantity and quality of litter input caused by different tree species could result in different below-ground responses. Correlations showed that Rswere exponentially correlated with the soil temperature and negatively correlated with the soil moisture under both M.laosensis and C.lanceolata plots. Non-liner regressions of soil temperature and soil moisture together accounted for 68.9%-77.0% and 36.2%-63.8% of monthly variation in soil respiration in M.laosensis and C.lanceolata plots, respectively, indicating that other factors were responsible for interpreting variations in Rs. The average Rswas significantly correlated to soil dissolved organic carbon concentration (P = 0. 019) and microbial biomass carbon concentration (P =0.037), which suggests that litter manipulation induced changes in soil labile carbon caused significant effects on Rs. A one time-point measured soil PLFA profiles showed that the average Rswas significantly correlated with the biomass of main signature PLFAs such as fungi (18:1 w9c, 18:2 w6,9c), which may imply that the high quality of litter input could alter soil microbial biomass and thus affect Rs. The temperature sensitivity (Q10) of Rsdecreased to 1.86 and 2.05 in the litter exclusion and transplant plots compared to 2.14 in the control plots under M. laosensis plantations. However, the Q10in the litter exclusion, litter transplanting and control plots were 1.40, 1.52 and 1.52 in C. lanceolata plantations, respectively, which are lower than other studies in subtropical China. In summary, we concluded that the effects of litter manipulation on Rswere dependent on tree species, the contributions of litter to Rs.can be attributed to the quantity and quality of litter input, the changes in soil labile carbon and soil microbial biomass in subtropical forest plantations.