Because of increasing need to balance health risks for pathogen control and disinfection by-products (DBP) formation in drinking water supplies, water utilities are forced to closely examine and optimize their disinfection practices. This research was designed to investigate the effects of independent variables of dissolved organic carbon (DOC), ferric chloride dosage, chlorine dose, and reaction time on trihalomethanes (THMs) formation in Terkos Lake Water (TLW) of Istanbul City. A statistically-based empirical model was developed for predicting THM formation during enhanced coagulation. The R-2 and F value of model were 0.762 and 460, respectively. The model was found to be statistically significant for all four variables, and model predictions appear to be most accurate for this study. A multiple linear model exhibited the best fit of data. It was observed that THM formation depended primarily on DOC removal. Model calibration, testing and validation were accomplished by using independent data set.