This study aims at evaluating the effect of different wastewater composition on kinetics, capacities, and mechanisms of P sorption by carbonated bauxite residues (CBR). A series of batch experiments was performed to investigate P sorption behaviors from solutions prepared with different aqueous matrices (deionized water, tap water, and real wastewater) and different initial P concentrations (from 10 to 200 mg P/L). Also, a series of sequential P extractions was performed to investigate P fractionation of CBR before and after its use in P sorption experiments, and hence to elucidate the main P removal mechanisms. The results indicate that initial P concentration is the most influential parameter controlling kinetics, capacities, and mechanisms of P removal in batch experiments. Kinetic constant of P sorption increases exponentially with decreasing initial P concentration below 100 mg P/L, thus indicating a faster achievement of P sorption equilibrium. Equilibrium P sorption capacities increase linearly from about 0.2 to about 3.9 mg P/g CBR with increasing initial P concentration from 10 to 200 mg P/L, thus indicating that P saturation of CBR was not reached. Ca phosphate precipitation is the main P removal mechanism at higher initial P concentrations (> 10 mg P/L), whereas phosphate adsorption on CBR surface becomes more relevant over the total amount of P removed at lower initial P concentrations. Overall, the findings of this study allow to evaluate kinetic constants, sorption capacities, and removal mechanisms under different operating scenarios, thus providing crucial information for the design and operation of P treatment units.