Greenhouse gas emissions as affected by different water management practices in temperate rice paddies

Abstract The mitigation of methane (CH 4 ) and nitrous oxide (N 2 O) emissions from rice paddy fields is pivotal in minimizing the impact of rice production on global warming. The large majority of the world rice is cropped in continuously flooded paddies, where soil anaerobic conditions lead to the production and emission of significant amounts of CH 4 . In this work we evaluated the effectiveness of water management techniques alternative to the conventional flooding on the mitigation of CH 4 emissions from paddy soils, and verified whether any concurrent increase in N 2 O emissions can totally or partially offset their environmental benefit. Two alternative water management systems were compared to the conventional continuous flooding system (WFL): dry seeding with delayed flooding (DFL) and intermittent irrigation (DIR). Methane and N 2 O emissions were monitored at field-scale over two years including both rice cropping and fallow seasons, using non-steady-state closed chamber approach. The DFL system resulted in a 59% decrease (average of the two measured years) in total CH 4 emissions with respect to WFL, while DIR annulled CH 4 emissions. The effect of CH 4 mitigation of DFL with respect to WFL was mainly concentrated within the vegetative stage, while any significant flux from DIR was recorded throughout the growing and non-growing season. However, DIR resulted in the highest emission peaks and cumulative fluxes of N 2 O, almost totally occurred during the vegetative stage. In contrast, DFL and WFL showed N 2 O emissions that were 77 and 93% lower with respect to DIR, respectively. Total annual fluxes suggest that the adoption of alternative water management practices that involve dry seeding and subsequent delayed flooding or intermittent irrigation can contribute to significantly reduce the global warming potential of rice cropping systems by 56 and 83%, respectively with respect to continuous flooding.