Characterization of methane emissions from rice fields in Asia. II. Differences among irrigated, rainfed, and deepwater rice

Methane (CH4) emission rates were recorded automatically using the closed chamber technique in major rice-growing areas of Southeast Asia. The three experimental sites covered different ecosystems of wetland rice-irrigated, rainfed, and deepwater rice-using only mineral fertilizers (for this comparison). In Jakenan (Indonesia), the local water regime in rainfed rice encompassed a gradual increase (wet season) and a gradual decrease (dry season) in floodwater levels. Emission rates accumulated to 52 and 91 kg CH4 ha(-1) season(-1) corresponding to approximately 40% of emissions from irrigated rice in each season. Distinct drainage periods within the season can drastically reduce CH4 emissions to less than 30 kg CH4 ha(-1) season(-1) as shown in Los Banos (Philippines). The reduction effect of this water regime as compared with irrigated rice varied from 20% to 80% from season to season. Methane fluxes from deepwater rice in Prachinburi (Thailand) were lower than from irrigated rice but accumulated to equally high seasonal values, i.e., about 99 kg CH4 ha(-1) season(-1), due to longer seasons and assured periods of flooding. Rice ecosystems with continuous flooding were characterized by anaerobic conditions in the soil. These conditions commonly found in irrigated and deepwater rice favored CH4 emissions. Temporary aeration of flooded rice soils, which is generic in rainfed rice, reduced emission rates due to low CH4 production and high CH4 oxidation. Based on these findings and the global distribution of rice area, irrigated rice accounts globally for 70-80% of CH4 from the global rice area. Rainfed rice (about 15%) and deepwater rice (about 10%) have much lower shares. In turn, irrigated rice represents the most promising target for mitigation strategies. Proper water management could reduce CH4 emission without affecting yields.