Alternate wetting and drying (AWD) irrigation is a reliable and widespread water-saving technology for rice production. This review aimed to comprehensively assess the changes in water use, yield, nitrogen use efficiency (NUE), and greenhouse gas (GHG) emissions after AWD adoption and explore the optimal AWD water thresholds. Based on abundant data obtained from contrasting experiments between AWD and continuous flooding (CF) irrigation, the effects of AWD were quantified by stepwise multiple linear regression (MLR). The quantitative indices are the changes in total irrigation water (dIW), water productivity (delta WP), grain yields (delta GY), NUE indices, global warming potential (delta GWP), and yield-scaled global warming potential (delta YGWP) between AWD and CF. It was found that the delta(C-A) (i.e. the difference between the CF and AWD thresholds) and soil types were the most noteworthy indicators affecting IW, water productivity, yield and partial factor productivity of N (PFPN). Meanwhile, the delta(C-A) and fertilizer-N rates (NR) were two key indicators affecting GHG emissions. Thereafter, for different soil types, a group of optimization models were established to explore the optimal AWD water thresholds for optimizing delta IW and delta WP simultaneously without yield, NUE, and GHG emissions penalties. The results showed that the optimal upper limit of AWD (U-AWD) should be equal to the upper limit of the CF (U-CF), and the optimal lower limit of AWD (L-AWD) can be obtained by the model. Taking the clay paddy field as an example, the min delta IW were-41% to-8%, and the max delta WP were 11%-54% after adopting optimal AWD water managements. The corresponding delta GY, delta PFPN, delta GWP, delta YGWP were 0%-1%, 0%-1%,-66% to-15% and-52% to-12%, respectively. This review provides a novel way to estimate the performance of AWD and provide guidance for proper AWD water management.