Water management to mitigate the global warming potential of rice systems: A global meta-analysis.

Highlights • A meta-analysis on GHG emissions and rice yield as affected by water management. • Non-continuous flooding reduced CH 4 emissions by 53%, but it increased N 2 O emissions by 105%. • Severe soil drying reduced rice yield, but mild soil drying did not. • Non-continuous flooding reduced GWP by 44% and yield-scaled GWP by 42%. • IPCC underestimates the impact of multiple (i.e. ≥ 2) drying events on CH 4 emissions. Abstract Rice is a main staple food for roughly half of the world's population, but rice agriculture is also a main source of anthropogenic greenhouse gas (GHG) emissions. Many studies have reported that water management (e.g. alternate wetting and drying, intermittent irrigation, mid-season drain, aerobic rice) affects rice yields and methane (CH 4) and nitrous oxide (N 2 O) emissions from rice paddies. However, these studies span a variety of practices and vary in experimental design and results, making it difficult to determine their global response from individual experiments. Here we conducted a meta-analysis using 201 paired observations from 52 studies to assess the effects of water management practices on GHG emissions and rice yield. Overall, compared to continuous flooding, non-continuous flooding practices reduced CH 4 emissions by 53%, increased N 2 O emissions by 105%, and decreased yield by 3.6%. Importantly, N 2 O emissions were low, contributing, on average, 12% to the combined global warming potential (GWP; CH 4 + N 2 O). As a result, non-continuous flooding reduced GWP (-44%) and yield-scaled GWP (-42%). However, non-continuous flooding practices stimulated N 2 O emissions to a greater degree in soils with high organic carbon or with manure additions. The reduction in CH 4 emissions increased with the number of drying events, soil drying severity, and the number of unflooded days. Currently, Intergovernmental Panel on Climate Change (IPCC) scaling factors for single and multiple (≥ 2) drying events are 0.6 and 0.52. Based on this analysis using actual side-by- side field studies, we suggest changing these to 0.67 for a single event and 0.36 for multiple events. [ABSTRACT FROM AUTHOR]