Methane emissions from rice production: a synthesis of 24 eddy covariance sites

Rice production annually contributes up to 8% of anthropogenic CH ₄ emissions, creating a higher per-calorie climate burden than other grains. It is still crucial to understand how production management and underlying field conditions contribute to CH ₄ production and emission in rice fields, at different time scales and locations. Landscape-scale greenhouse gas emissions measurements are possible with the eddy covariance method. They offer a different perspective from chamber- or laboratory-scale measurements by covering a larger area, integrating across smaller-scale soil and management heterogeneity, and providing a largely continuous record. We have assembled a unique global dataset from 24 rice field sites measured with eddy covariance. The diversity of agronomic, edaphic, and climate conditions enables a unique perspective on CH ₄ flux dynamics from these important agro-ecosystems. The dataset contains 46 growing seasons and it has been gap-filled and analyzed following a consistent protocol from the Fluxnet-CH₄ project. The aims are (1) to quantify the range of CH4 emissions at different time scales (2) derive their key drivers and (3) to make the dataset available for further process-based modeling.

Emissions vary widely across locations: daily peak CH ₄ emissions range from 1-12 kg CH ₄ -C ha ^-1 (median 2.6 kg CH ₄ -C ha ^-1 ); seasonal emissions range from 4-606 kg CH ₄ -C ha ^-1 (median 90 kg CH ₄ -C ha ^-1 ). The median diurnal range (max-min daily flux) ranged from 0.11 to 2.3 kg CH ₄ -C ha ^-1 hr ^-1 (median 0.5 kg CH ₄ -C ha ^-1 hr ^-1 ). Seasonal emissions are correlated to N fertilization rate (R2=0.47 for sites not frequently drained) and can be reduced through frequent drainage (to 4-35 CH ₄ -C ha ^-1 ). Fertilization rates may be representative of a number of other agronomic factors, including higher biomass and yields; a yield-normalized analysis will follow. Irrigation and agronomic management (including drainage dynamics, fertilizer applications, residue management, and crop rotations) play a strong role in determining net CH ₄ emissions. These findings highlight the need for clear biological and ancillary data and meta-data to be included in the dataset and offer a basis for climate-smart rice production management.