The main objective of this research was to evaluate the effectiveness of nanofiltration (NF) at full and bench scale for controlling AOC and BDOC, which are the main indicators of biological stability of the finished potable water. One of the major observations from full-scale operation was that nanofiltration was a very effective means to reduce BDOC, but conversely, did not reject a significant fraction of AOC. The high BDOC rejection by nanofiltration (NF) membranes at full scale can be explained by size exclusion, since a significant fraction of BDOC consists of compounds, such as humic and fulvic acids, which are larger than the pores of NF membranes (molecular-weight cutoff ≈200 daltons). The insignificant AOC rejection observed in full-scale systems was probably due to the low pH, high hardness, and high ionic strength (TDS) of the raw water. Bench scale tests using simulated waters clearly demonstrated that AOC removal by NF membranes decreases markedly with decreasing pH, and increasing hardness and ionic strength, implying that electrostatic repulsion plays a significant role in AOC removal mechanisms. These solution environments repress the electrostatic interaction between charged organic compounds and membranes, allowing passage of small molecular weight compounds and thus reducing AOC rejection.