Response　surface　methodology　(RSM)　was　applied　to　analyze　the　magnetic　flocculation　influencing　factors　and　removal　efficiency　for　micro-polluted　river　water　treatment.　Box-Behnken　experimental　design　was　applied.　The　poly　aluminum　chloride　(PAC),　magnetic　particle　(MP)　and　polyacrylamide　(PAM)　were　chosen　as　causal　factors.　The　effects　of　these　factors　on　turbidity,　orthophosphate　(PO43--P)　and　chemical　oxygen　demand　(COD)　removal　efficiency　were　analyzed　using　RSM.　The　pollutants　removals　in　river　water　were　developed　to　describe　by　regression　models.　The　results　showed　that　the　optimum　conditions　were　PAC　51.31　mg/L,　magnetic　particle　176.47　mg/L　and　PAM　0.10　mg/L.　Based　on　these　conditions,　the　turbidity,　orthophosphate　and　COD　removals　were　96.80%,　86.25%　and　51.00%,　respectively.　The　relative　errors　between　the　measured　data　and　the　predicted　values　were　less　than　2.12%.　It　confirms　that　the　statistic　models　are　reliable,　and　the　magnetic　flocculation　operating　conditions　for　contaminants　removals　in　river　water　are　appropriated.