Quantifying bubbling emission (ebullition) of methane from a rice paddy using high-time-resolution concentration data obtained during a closed-chamber measurement

Methane (CH₄) produced in rice-paddy soil is transported to the atmosphere either via the rice plants or by bubbling events (ebullition); however, little is known about the frequency and intensity of bubbling CH₄ emissions and the factors that affect them. We developed a method to quantify ebullition using high-time-resolution (~1 Hz) CH₄ concentration data obtained by closed-chamber measurements. Field measurements were conducted in a Japanese rice paddy at different rice growth stages: panicle formation (PF), booting (BT), and heading (HD). A dataset of 132 chamber measurements was used to develop and evaluate the method. A scripting file written in R programing language was used to automatically determine CH₄ emissions via the two pathways. Plant-mediated CH₄ emission intensity was constant during chamber deployment and was reflected as a steady linear increase in chamber [CH₄] with time or as a constant baseline in a flux time series. We found that the plant-mediated emission could be determined as the peak with the lowest flux intensity in the flux frequency distribution even if bubbling events occurred during the chamber deployment. The field measurement results in combination with established data processing protocols showed that at PF, ebullition contributed only 4% of the total emission, whereas it accounted for 32% and 60% of the total emission at BT and HD, respectively. In contrast, the plant-mediated flux variation among growth stages was smaller. Both ebullition and plant-mediated emissions correlated significantly with air temperature at HD, but the magnitude of the dependency was much higher for ebullition than for rice-mediated emission. These results demonstrate that ebullition occurs more frequently than has previously been thought, and the different transport pathways show varying degrees of dependency on plant phenological and environmental factors, thus underscoring the need to separately determine CH₄ emissions via each transport pathway.