This study reports the effects of four environmental factors on the amount of phosphorus adsorbed on suspended kaolin by applying the Box Behnken experimental design and response surface methodology. Two factors (e.g. the phosphorus and suspended sediment concentration) were screened out from the literature review about variables having significant correlation with the phosphorus adsorption. The other two factors (e.g. the turbulent kinetic energy and time interval of phosphorus loading) are new variables which are to the knowledge of the author have never been considered adequately in the previous studies of phosphorus adsorption. However, they potentially affect the amount of phosphorus adsorbed on sediment. The range of turbulent kinetic energy (47.9 to 707.9 cm/s2), the phosphorus concentration (3 to 6 mg/L), the suspended kaolin concentration (500 to 2100 mg/L), and the time interval of phosphorus loading (30 minutes to eight hours) are selected for the study in order to represent the condition in Brantas River, Indonesia (after the period of wet season but before the dry season starting from around March to May each year) which was reported to have a problem with high level of phosphorus and suspended sediment. The phosphorus adsorption are measured in 27 runs of laboratory experiments by using a mixing box with 48 Litre of water volume oscillated by a horizontal grid producing a relatively isotropic homogeneous turbulent in the horizontal plane. Samples of water were taken and measured for phosphorus concentration at the beginning and every two hours during the total of eight hours of experimental runs by using DR3900 spectrophotometer. The amount of phosphorus adsorbed on suspended kaolin was calculated as the difference between the amount of PO43-P loaded into the mixing box and the PO43-P concentration measured on the sampling time. The study shows that second order models provide the best fit to the adsorption data at two, four and eight hours of residence time (with R2=77.51%, 79.94%, 49.86%, respectively). A first order model with interaction terms provides the best fit to the data at six hour of residence time (R2 is 47.51%). Among the factors, the time interval of phosphorus loading has the most significant effect on phosphorus adsorption. The phosphorus concentration has greater effect at the beginning of the reaction and the effect is less significant in the later duration of observations. The suspended kaolin concentration has less effect in the beginning of reaction, but the effect increases as the duration of reactions increases. The turbulent kinetic energy shows a more significant contribution in its quadratic form in the first six hours of the experiment, and the effect diminishes in the last two hours of residence time.