Soil carbon storage and mineralization rates are affected by carbon inputs rather than physical disturbance: Evidence from a 47-year tillage experiment

In spite of the large number of studies conducted on the drivers of soil organic carbon (SOC) stocks, there is still no consensus on the impact of tillage on the distribution and turnover rate of SOC in the soil profile. Few studies have characterized precisely the turnover of SOC using 13C natural tracing or simulated the SOC evolution per soil layer. In this study, we combined several approaches (diachronic analysis of SOC stocks, isotopic tracing and modelling) for characterizing the SOC evolution per soil layer in one of the oldest tillage experiments comparing no-till (NT), shallow till (ST) and full inversion tillage (FIT) combined with six crop managements. The new measurements made in 2017 reported in this paper confirm that i) tillage had no effect on SOC stocks integrated over the old ploughed layer (∼0−28 cm) or deeper (∼0−58 cm) and ii) reduced tillage affected the SOC distribution in the soil profile, with SOC storage in the upper layer (∼0−10 cm) offset by a SOC loss in the underlying layer (∼10−28 cm). The change in rotation (from C4 to C3 crops) in two crop management treatments allowed to quantify the decrease in C4 stocks during 19 years and calculate the specific mineralization rates relative to tillage treatments and soil layers. The mineralization rates did not vary significantly between tillage treatments in the whole (old) ploughed layer (∼0−28 cm) but varied according to depth and tillage. The highest rates were found in the layer 0−5 cm of NT or ST and the lowest rates in the layer 10−28 cm of the same tillage treatments. The rates were highly correlated with estimated C inputs and particulate organic matter contents in each layer, but not with tillage intensity. The evolution of total SOC and C4 stocks of each soil layer was simulated with AMG model during 47 years. The standard model (with a single mineralization rate) gave a good prediction of SOC evolution when applied to the whole profile (∼0−28 cm) but not for each individual layer. Including a relationship between C input and mineralization rate in the model allowed to well simulate the SOC evolution in all soil layers. This study shows that the main effect of a change in tillage on SOC storage is the change in the distribution of C input throughout the profile and the corresponding variation of the priming effect rather than the change in physical soil disturbance.