Dissolved oxygen, one of the most important water quality parameters, is a crucial parameter for the aquatic ecosystems. In this study, some advanced chemometric techniques included in a multi-layer perceptron, radial basis neural network, and two different adaptive neuro-fuzzy inference system methods are developed to model dissolved oxygen concentration. Moreover the estimations of these models are compared with the multiple linear regression. In this context, monthly mean quantities of the temperature, pH, electrical conductivity, discharge and dissolved oxygen data recorded at Broad River near Carlisle, SC in USA are used. The accuracy of the models is compared with one other by using determination coefficient, mean absolute error, root mean square error and mean absolute relative error statistics. Results indicate that radial basis neural network method performs better than the other methods in modelling monthly mean dissolved oxygen concentration. The temperature, pH, electrical conductivity, and discharge are found to be effective on dissolved oxygen concentration.