Black carbon decomposition and incorporation into soil microbial biomass estimated by 14C labeling
Yakov Kuzyakov, Irina Subbotina, Haiqing Chen, Irina Bogomolova, Xingliang Xu
Department of Agroecosystem Research, BayCEER, University of Bayreuth, 95447 Bayreuth, Germany.
Incomplete combustion of organics such as vegetation or fossil fuel led to accumulation of charred products in the upper soil horizon. Such charred products, frequently called pyrogenic carbon or black carbon (BC), may act as an important long-term carbon (C) sink because its microbial decomposition and chemical transformation is probably very slow. Direct estimations of BC decomposition rates are absent because the BC content changes are too small for any relevant experimental period. Estimations based on CO2 efflux are also unsuitable because the contribution of BC to CO2 is too small compared to soil organic matter (SOM) and other sources.
We produced BC by charring 14C labeled residues of perennial ryegrass (Lolium perenne). We then incubated this 14C labeled BC in Ah of a Haplic Luvisol soil originated from loess or in loess for 3.2 years. The decomposition rates of BC were estimated based on 14CO2 sampled 44 times during the 3.2 years incubation period (1181 days). Additionally we introduced five repeated treatments with either 1) addition of glucose as an energy source for microorganisms to initiate cometabolic BC decomposition or 2) intensive mixing of the soil to check the effect of mechanical disturbance of aggregates on BC decomposition. Black carbon addition amounting to 20% of Corg of the soil or 200% of Corg of loess did not change total CO2 efflux from the soil and slightly decreased it from the loess. This shows a very low BC contribution to recent CO2 fluxes. The decomposition rates of BC calculated based on 14C in CO2 were similar in soil and in loess and amounted to 1.36 10−5 d−1 (=1.36 10−3% d−1). This corresponds to a decomposition of about 0.5% BC per year under optimal conditions. Considering about 10 times slower decomposition of BC under natural conditions, the mean residence time (MRT) of BC is about 2000 years, and the half-life is about 1400 years. Considering the short duration of the incubation and the typical decreasing decomposition rates with time, we conclude that the MRT of BC in soils is in the range of millennia.
The strong increase in BC decomposition rates (up to 6 times) after adding glucose and the decrease of this stimulation after 2 weeks in the soil (and after 3 months in loess) allowed us to conclude cometabolic BC decomposition. This was supported by higher stimulation of BC decomposition by glucose addition compared to mechanical disturbance as well as higher glucose effects in loess compared to the soil. The effect of mechanical disturbance was over within 2 weeks. The incorporation of BC into microorganisms (fumigation/extraction) after 624 days of incubation amounted to 2.6 and 1.5% of 14C input into soil and loess, respectively. The amount of BC in dissolved organic carbon (DOC) was below the detection limit (<0.01%) showing no BC decomposition products in water leached from the soil.
We conclude that applying 14C labeled BC opens new ways for very sensitive tracing of BC transformation products in released CO2, microbial biomass, DOC, and SOM pools with various properties.