Nanofiltration and reverse osmosis membranes are very susceptible to membrane fouling. Therefore extensive (advanced) pre-treatment must be applied to control productivity loss. The combination of extensive advanced pre-treatment with nanofiltration or reverse osmosis is defined as an integrated membrane system (IMS). Within the framework of a project cofunded by AWWARF and USEPA three very promising; IMS＇s were identified for surface water treatment. This paper describes a part of the pilot plant research carried out by the Water Supply Company of Overijssel and Kiwa on the combination of riverbank filtration/conventional ground water treatment and nanofiltration. At the site of the future production plant Vechterweerd surface water is abstracted via bank filtration. The raw water has a high colour and hardness. Moreover the water contains a number of synthetic organic chemicals originating from the river. The river bank filtrate is pretreated by a double aeration and rapid filtration steps. The nanofiltration plant is loaded with Hydranautics PVD-1 membranes. During a period of 20 months the productivity control and the biological stability of the water before and after nanofiltration is studied. The nanofiltration plant was operated at 80% recovery and hydrochloric acid was applied to avoid scaling. Use of anti-sealants was avoided to restrict biofouling. For the first 160 days MTC values showed a gradual decrease in combination with a gradual increase of the feed concentrate pressure drop. Since then an exponential increase in feed concentrate pressure drop was observed caused by a strong biafouling build up onto the feed spacer of the membrane modules. The biofouling proved to be related to the quality of the chemical pure hydrochloric: acid used for acidification of the nanofiltration feed. During the total run of 480 days 13 cleanings were applied Normally cleaning with sodium hydroxide had the best results. After the severe biofouling a mechanical cleaning with air combined with a sodium hydroxide rinse restored the MTC-value and feed concentrate pressure drop. During the last 100 days of operation the biofouling could be controlled by an anaerobic treatment with sodium bisulphite. During the first 140 days of a following experiment with TriSep membranes measures were taken to minimise the biogrowth onto the membrane surface The time between two cleanings was lengthened to about 3 months, although the biofouling couldn＇t he prevented completely. Anaerobic treatment in a early stage might control biofouling. Cleaning experiments with a detergent show promising results in restoring the original levels of MTC and pressure drop.