Variations in crop yield caused by different ratios of organic substitution are closely related to microbial ecological clusters in a fluvo-aquic soil

Context: Partial substitution of chemical fertilizer with organic fertilizer (PSCFOF) can improve soil quality and reduce the overreliance of crop yield on chemical fertilizers. However, the relationship between alterations in crop yield and soil microbial community in response to various substitution ratios is not comprehensively understood. Objectives: This study aimed to investigate the impact of PSCFOF on soil nutrient supply, enzyme activity, bacterial and fungal community, and crop yield. Methods: An 11-year (2009 - 2021) field experiment was carried out in a winter wheat-summer maize doublecropping system with four fertilization regimes (no fertilizer, NF; pure chemical fertilizer, NPK; 30% organic substitution, NPKLM; and 50% organic substitution, NPKHM). Co-occurrence network analysis was applied to examine the importance of ecological clusters and keystone species in controlling microbial activities and crop yields. Results: There were no significant differences in soil nutrient contents and annual crop yield between the NPK and NPKLM treatments. However, the NPKHM treatment resulted in significant reductions in crop yield by 9.49%16.78%, despite increasing soil organic matter (SOM), total nitrogen (TN), available nitrogen (AN), and available phosphorus (AP) by 9.05%, 7.26%, 11.67%, and 26.62%, respectively. This contrasting effect was primarily attributed to the influence of a specific ecological cluster (module 2), rather than the entire microbial community. Long-term 50% organic substitution would suppress keystone species within module 2, such as Aggregatilinea, Luteimonas, Usitatibacter and Microvirga. These species were specifically correlated with soil carbon enzyme activities, including beta-glucosidase, invertase, and dehydrogenase, and contributed significantly and positively to crop yield. Conclusions: Organic substitution at different ratios altered soil nutrient content, enzyme activities and microbial community in different intensities and directions. Long-term 30% organic substitution could preserve soil fertility and microbial community, leading to stable crop yield; whereas 50% organic substitution could inhibit keystone species, reduce extracellular enzyme activity and ultimately lead to a decrease in crop yield. Implications: Our results highlight the importance of ecological clusters within soil microbial co-occurrence network in regulating crop yield, presenting a new perspective on the variations in crop yield resulting from varying ratios of organic substitution.