Distinct patterns in the diurnal and seasonal variability in four components of soil respiration in a temperate forest under free-air CO₂ enrichment

Abstract. Soil respiration (R_S) is a major flux in the global carbon (C) cycle. Responses of R_S to changing environmental conditions may exert a strong control on the residence time of C in terrestrial ecosystems and in turn influence the atmospheric concentration of greenhouse gases. Soil respiration consists of several components oxidizing soil C from different pools, age and chemistry. The mechanisms underlying the temporal variability of R_S components are poorly understood. In this study, we used the long-term whole-ecosystem ¹³C tracer at the Duke Forest Free Air CO₂ Enrichment site to separate forest R_S into its autotrophic (R_R) and heterotrophic components (R_H). The contribution of R_H to R_S was further partitioned into litter decomposition (R_L), and decomposition of soil organic matter (R_SOM) of two age classes – up to 8 yr old and SOM older than 8 yr. Soil respiration was generally dominated by R_SOM during the growing season (44% of daytime R_S), especially at night. The contribution of heterotrophic respiration (R_SOM and R_L) to R_S was not constant, indicating that the seasonal variability in R_R alone cannot explain seasonal variation in R_S. Although there was no diurnal variability in R_S, there were significant compensatory differences in the contribution of individual R_S components to daytime and nighttime rates. The average contribution of R_SOM to R_S was greater at night (54%) than during the day (44%). The average contribution of R_R to total R_S was ~30% during the day and ~34% during the night. In contrast, R_L constituted 26% of R_S during the day and only 12% at night. About 95% of the decomposition of soil C older than 8 yr (R_pre-tr) originated from R_SOM and showed more pronounced and consistent diurnal variability than any other R_S component; nighttime rates were on average 29% higher than daytime rates. In contrast, the decomposition of more recent, post-treatment C (R_pre-tr) did not vary diurnally. None of the diurnal variations in components of R_H could be explained by only temperature and moisture variations. Our results indicate that the variation observed in the components of R_S is the result of complex interaction between dominant biotic controls (e.g. plant activity, mineralization kinetics, competition for substrates) over abiotic controls (temperature, moisture). The interactions and controls among roots and other soil organisms that utilize C of different chemistry, accessibility and ages, results in the overall soil CO₂ efflux. Therefore understanding the controls on the components of R_S is necessary to elucidate the influence of ecosystem respiration on atmospheric C-pools at different time scales.