Methane Fluxes from a Rice-Wheat Rotation Agroecosystem in Eastern China at Seasonal and Diurnal Scales

CH₄ fluxes from a rice-wheat rotation agroecosystem in eastern China were continuously measured using the eddy covariance technique. The diurnal and seasonal variations of CH₄ flux and potential controlling factors in 2016 were analyzed using wavelet coherence, conditional Granger causality, correlation analysis and path analysis methods. CH₄ fluxes showed distinguishable diurnal variations with single peaks during 13:00-16:00 local time. At the diurnal timescale, gross primary productivity (GPP) regulates CH₄ fluxes after accounting for the effects of latent heat flux (LE), air temperature (TA), and soil temperature (TS) on CH₄ fluxes. LE mirrored the diurnal pattern of CH₄ fluxes when the effects of TA and TS on CH₄ fluxes were considered. Daily CH₄ fluxes exhibited large seasonal variations, with the largest daily CH₄ flux of 1191.78 mg C-CH₄ m^-2 d^-1 on 29 July 2016. The daily CH₄ fluxes were continuously low in the growing season of wheat, and sharply increased from very low values in late June to peaks in late July and early August, and then gradually decreased to low values at the end of the rice growing season in late November and early December. Correlation analysis and path analysis showed that seasonal variations of soil temperature, air temperature, and GPP had strong effects on daily CH₄ fluxes during pre-panicle initiation of the rice growing season, while soil temperature and leaf area index (LAI) had very strong effects on daily CH₄ fluxes during the post-panicle initiation stage. The total of CH₄ fluxes from the rice-wheat rotation agroecosystem into the atmosphere amounted to 58.08 ± 9.87 g C m^-2 in 2016, and the annual net carbon (C) budget and greenhouse gas (GHG) budget were 163.50 ± 9.87 g C m^-2 and 2322.53 ± 329.00 g CO₂eq m^-2, respectively. This study represents a comprehensive assessment of fluxes and drivers of CH₄ from a rice-wheat rotation agroecosystem at different timescales. Additionally, the consecutive data of CH₄ emission in this region will also useful for model calibration and validation.