Soil biotic legacy effects of agricultural practices influence microbial C use efficiency

International audience

إنجليزي. Increasing soil organic matter in cropped soils is a major challenge. Soil carbon in arable systems is mainly acquired through above and below-ground litter decomposition, with an efficiency depending both on litter chemical characteristics and on the composition and activity of the soil microbial communities; however, the interactions between these two factors remain poorly understood. To answer to this question, we selected fields from a same soil type which underwent different agricultural practices during 5 years. We tested whether the impacts of these practices on soil biotic characteristics will have legacy effects on litter-C microbial assimilation strategy. The study was conducted at the long-term experimental site SOERE ACBB, located in Northern France (49.873 N, 3.032 E), on a loamy soil (Orthic Luvisol, FAO classification) where six treatments varying by their tillage, N fertilization and crop residues management were set up in 2010 (Coudrain et al., 2016). We sampled in 2015 the 0-5 cm layer of the deep-tillage (CONV) and the reduced-tillage (RT) treatments. A laboratory incubation was performed during 29 days with, for each soil, three litter treatments, i.e. (i) no litter added (control), and uniformly (ii) 13C-labelled mature wheat (recalcitrant litter), and (iii) 13C-labelled flowering wheat residues (labile litter), mixed into the soil at the rate of 3 mg C g-1 dry soil. Dynamics of C-CO2 and 13C-CO2, microbial biomass C and 13C, and hydrolytic enzymatic activities were measured during the decomposition. Non amended RT soil (control) showed higher total C content, C-CO2 emitted, microbial biomass C and enzymatic activities than CONV soil. RT litter-amended treatments exhibited both higher carbon use efficiency (proportion of the litter-13C decomposed incorporated into 13C-microbial biomass) and enzymatic efficiency (amount of litter-13C mineralized per amount of enzymes produced), yet had lower or equivalent priming effects than CONV treatments. These results suggest legacy effect of agricultural practices such as reduced tillage on the microbial C assimilation strategy in soil. We observed also that differences between CONV and RT microbial activities and carbon use efficiency increased with the recalcitrance of the decomposing litter.