Are variations in heterotrophic soil respiration related to changes in substrate availability and microbial biomass carbon in the subtropical forests?

International audienceSoil temperature and moisture are widely-recognized controlling factors on heterotrophic soil respiration (R h), although they often explain only a portion of R h variability. How other soil physicochemical and microbial properties may contribute to R h variability has been less studied. We conducted field measurements on R h half-monthly and associated soil properties monthly for two years in four subtropical forests of southern China to assess influences of carbon availability and microbial properties on R h. R h in coniferous forest was significantly lower than that in the other three broadleaf species-dominated forests and exhibited obvious seasonal variations in the four forests (P < 0.05). Temperature was the primary factor influencing the seasonal variability of R h while moisture was not in these humid subtropical forests. The quantity and decomposability of dissolved organic carbon (DOC) were significantly important to R h variations, but the effect of DOC content on R h was confounded with temperature, as revealed by partial mantel test. Microbial biomass carbon (MBC) was significantly related to R h variations across forests during the warm season (P = 0.043). Our results suggest that DOC and MBC may be important when predicting R h under some conditions, and highlight the complexity by mutual effects of them with environmental factors on R h variations. With about 68-79 petagram (Pg) CO 2-C being released into the atmosphere each year, soil respiration is the second largest carbon (C) flux in the terrestrial C cycling 1,2. Because variations in soil respiration are most likely to upset the equilibrium of terrestrial C cycling 3,4 , soil respiration has received increasing concerns in recent decades 5. Total soil respiration (R s) consists of two components, i.e., autotrophic (R a) and heterotrophic (R h) respiration, with R a mainly being the CO 2 released by plant roots and R h by microbial activities in association with soil organic matter (SOM) decomposition 6,7. Averaged on the global scale, R h accounts for 54% of R s in forests 6. Temperature and water supply, interactively or separately, have been most frequently reported to control soil respiration 1,8,9 , and account for a certain portion of variations in soil respiration in previous studies 5,8,10. However, which factors contribute to the remaining proportion is still an active topic. Substrate availability is proposed as an important underlying factor influencing soil respiration 5,11 , and even overshadows the role of temperature in the processes of soil respiration in specific ecosystems 12,13. A positive substrate-R h relationship has been found under laboratory conditions, especially when soil microbes are energy-limited 14. The increase in CO 2 efflux rate with C additions may be attributed to reactivations of dormant microorganisms 15 , augments in microbial biomass and activity 14,16 , and shifts in microbial community composition and growth strategies 17. Substrate-dependent soil respiration was also noted in field studies 12,18 , but thi