Le non labour augmente la séquestration du carbone par le sol en surface mais la diminue en profondeur
Les auteurs soulignent que selon les études, le non labour apparaît tantôt favorable tantôt défavorable à la séquestration du carbone dans le sol. Ils font donc une méta analyse pour tirer des conclusions globales.
L'étude analyse 69 expériences pour comparer les situations avec ou sans labour en terme de séquestration du carbone dans le sol.
L'arrêt du labour sur une parcelle autrefois labourée permet d'augmenter le carbone du sol de surface (10 cm de profondeur) de 3.15±2.42 tonnes par hectare, mais réduit de 3.30±1.61 tonnes par hectare le carbone dans les couches plus profondes (20 à 40 cm de profondeur). Globalement, l'adoption du non-labour ne permet pas d'augmenter significativement la séquestration du carbone dans les 40 premiers cm de sol.
L'augmentation du nombre d'espèces cultivées dans la rotation entraîne une réduction du carbone dans le sol, au contraire l'augmentation de la fréquence des mêmes espèces permet d'augmenter significativement le stock de carbone.
Le non labour n'est pas forcément un levier important pour augmenter la séquestration de carbone dans les sols agricoles.
Adopting no-tillage in agro-ecosystems has been widely recommended as a means of enhancing carbon (C) sequestration in soils. However, study results are inconsistent and varying from significant increase to significant decrease. It is unclear whether this variability is caused by environmental, or management factors or by sampling errors and analysis methodology. Using meta-analysis, we assessed the response of soil organic carbon (SOC) to conversion of management practice from conventional tillage (CT) to no-tillage (NT) based on global data from 69 paired-experiments, where soil sampling extended deeper than 40 cm. We found that cultivation of natural soils for more than 5 years, on average, resulted in soil C loss of more than 20 t ha−1, with no significant difference between CT and NT. Conversion from CT to NT changed distribution of C in the soil profile significantly, but did not increase the total SOC except in double cropping systems. After adopting NT, soil C increased by 3.15±2.42 t ha−1 (mean±95% confidence interval) in the surface 10cm of soil, but declined by 3.30±1.61 t ha−1 in the 20–40cm soil layer. Overall, adopting NT did not enhance soil total C stock down to 40 cm. Increased number of crop species in rotation resulted in less C accumulation in the surface soil and greater C loss in deeper layer. Increased crop frequency seemed to have the opposite effect and significantly increased soil C by 11% in the 0–60cm soil. Neither mean annual temperature and mean annual rainfall nor nitrogen fertilization and duration of adopting NT affected the response of soil C stock to the adoption of NT. Our results highlight that the role of adopting NT in sequestrating C is greatly regulated by cropping systems. Increasing cropping frequency might be a more efficient strategy to sequester C in agro-ecosystems. More information on the effects of increasing crop species and frequency on soil C input and decomposition processes is needed to further our understanding on the potential ability of C sequestration in agricultural soils.
