Soil organic carbon sequestration and its synergetic effect on crop yield in cropland under different fertilizations

Under the dual pressure of food demand growth and climate change intensification, how to promote soil organic carbon (SOC) sequestration and synergistically improve crop production via agricultural practice is not only a major strategic demand to ensure food security but also a key way to implement the "Four Per Mille Initiative" and "Carbon Neutrality". Fertilization, especially adding organic amendment, is the most effective practice to enhance SOC, but limited information is available on the underlying mechanism of SOC sequestration in intensive cropping systems. Moreover, in recent years, the application practices of both traditional (e.g., straw incorporation and organic substitution (substitution of chemical fertilizers by manure)) and emerging (e.g., biochar application) organic amendment have shown a great potential in enhancing crop yield and SOC. However, an understanding of how and to what extent organic substitution and straw incorporation impact SOC at different depths and crop yield remains limited. Moreover, previous results regarding the effects of biochar application on SOC and crop yield showed wide variation and high uncertainty. Therefore, it is essential to enhance the understanding of SOC sequestration mechanisms and crop yield response to those practices. In this thesis, firstly, we used a field experiment with double maize cropping to study the responses of OC in soil fractions to long-term fertilization and revealed the underlying mechanism of SOC sequestration in intensive cropping systems. Secondly, four organic substitution and one straw incorporation field experiments were conducted in China to quantify the effects of these practices on SOC at different depths and crop yield as well as their relationship. Finally, the effects of biochar application, as a new perspective of fertilization strategy evolution, on SOC and crop yield were evaluated via a global meta-analysis, and their key factors were determined using a boosted regression tree model