Prolonged flooding changes the oxidation–reduction status of soils, often enhancing phosphorus (P) release to overlying floodwater. We studied P release from unamended, gypsum-amended and biochar- amended soils under simulated snowmelt flooding (previously frozen, cold flooding at +4°C) and summer flooding (unfrozen, warm flooding at +22°C), using two soils, Fyala clay (FYL-Cl) and Neuenberg sandy loam (NBG-SL) from Manitoba, Canada. Amended and unamended soils were packed into vessels and flooded under cold and warm temperatures in the laboratory. Pore water and floodwater samples were taken weekly for 6 weeks after flooding (WAF) and thereafter biweekly for 10 WAF, and analyzed for dissolved reactive P (DRP), pH and cation concentrations. NBG-SL showed a significantly higher DRP concentration in pore water and floodwater despite its low Olsen P content. Redox potential (Eh) decreased slowly under cold compared to warm flooding; hence redox-induced P release was substantially lower under cold flooding. Gypsum amendment significantly decreased the floodwater DRP concentrations in NBG-SL by 38% and 35% under cold and warm flooding, respectively, but had no significant effect in FYL-Cl, which had low DRP concentrations (<1.2 mg L-1) throughout flooding period. Biochar amendment significantly increased floodwater DRP concentrations by 27 to 68% in FYL-Cl under cold and warm flooding, respectively, but had no significant effect in NBG-SL. The results indicate substantially less P release under cold, than under warm flooding. Gypsum was effective in reducing floodwater DRP concentrations only at high DRP concentrations; thus the effectiveness was greater under warm, than under cold flooding conditions.