Greenhouse gas (GHG) emissions from rice (Oryza sativa) systems have been correlated to water management practice, but to date, no study has directly evaluated three main GHGs (i.e., methane [CH4], nitrous oxide [N2O], and carbon dioxide [CO2]) under flood‐ and furrow‐irrigated conditions at the same time as affected by various fertilizer‐phosphorus (P) sources, in particular the reportedly slow‐release struvite‐P source. Therefore, the objective of this study was to evaluate the effect of water regime (flooded and furrow‐irrigated) and fertilizer‐P source (diammonium phosphate, chemically precipitated struvite, electrochemically precipitated struvite [ECST], triple superphosphate, and an unamended control) on GHG emissions and two‐ and three‐gas global warming potentials (GWP* and GWP, respectively) in the greenhouse. Methane emissions were 10 times greater (p < 0.05) under flooded (29.4 kg CH4 ha−1 season−1) than under furrow‐irrigated conditions (2.9 kg CH4 ha−1 season−1), and four times lower (p < 0.05) with ECST (3.4 kg CH4 ha−1 season−1) than other fertilizer‐P sources, while CO2 emissions were three times greater (p < 0.05) under furrow‐irrigated (23,428 kg CO2 ha−1 season−1) than under flooded (8290 kg CO2 ha−1 season−1) conditions. The GWP* under furrow‐irrigated conditions was almost 40% lower (p < 0.05) than under flooded conditions. Although N2O emissions were unaffected by fertilizer‐P source, the N2O contribution to GWP* was more than 80% under furrow‐irrigated conditions. Flood‐ and furrow‐irrigated water regimes require diversified approaches in GHG mitigation, where the best management for ECST needs to be more fully evaluated.