Insights into the Driving Factors of Methane Emission from Double-Season Rice Field Under Different Fertilization Practices in South China

Paddy fields are the main agricultural source of greenhouse gas methane (CH4) emissions. To enhance rice yield, various fertilization practices have been employed in rice paddies. However, the key microbial and abiotic factors driving CH4 emissions under different fertilization practices in paddy fields remain largely uncharted. This study conducted field experiments in a traditional double-cropping rice area in South China, utilizing five different fertilization practices to investigate the key factors influencing CH4 emissions. High-throughput sequencing and PICRUSt2 functional prediction were employed to investigate the contributions of soil physicochemical properties, CH4-metabolizing microorganisms (methanogens and methanotrophs), and key genes (mcrA and pmoA) on CH4 emissions. The results showed that CH4 emission fluxes exhibited seasonal variations, with consistent patterns of change observed across all treatments for both early- and late-season rice. Compared to the no-fertilization (NF) treatment, cumulative CH4 emissions were lower in early-season rice with green manure (GM) and straw returning (SR) treatments, as well as in late-season rice with GM treatment, while rice yields were maintained at higher levels. High-throughput sequencing analysis revealed that potential methanogens were primarily distributed among four orders: Methanobacteriales, Methanocellales, Methanomicrobiales, and Methanosarcinales. Furthermore, there was a significant positive correlation between the relative abundance of the CH4-related key gene mcrA and these microorganisms. Functional analysis indicated that these potential methanogens primarily produce methane through the acetoclastic and hydrogenotrophic pathways. Aerobic CH4-oxidizing bacteria, predominantly from the genus Methylocystis, were detected in all the treatments, while the CH4 anaerobic-oxidizing archaea ANME-1b was only detected in chemical fertilization (CF) and cow manure (CM) treatments. Our random forest analysis revealed that the relative abundance of two methanogens (Methanocellales and Methanosarcinales) and two environmental factors (pH and DOC) had significant impacts on the cumulative CH4 emissions. The variance decomposition analysis highlighted the CH4-metabolizing microorganisms explained 50% of the variance in the cumulative CH4 emissions, suggesting that they are the key microbial factors driving CH4 emissions. These findings provide guidance for the development of rational measures to reduce CH4 emissions in paddy fields.