Yes / Natural draft cooling towers are subjected to dynamic air temperature and humidity ratio of air throughout the day. Their performances based on steady state model is restricted to single air temperature and humidity ratio of air values and thus are not accurate. In this work, a dynamic model for a natural draft cooling tower is developed. Air temperature and relative humidity data for Johannesburg were acquired to develop dynamic correlations for these two parameters and are embedded in the process model together with the dynamic model of the collection basin of the tower. A simple algorithm is proposed to solve the resulting boundary value problem. Finally, the common perception that seawater/river water should be used instead of freshwater as cooling medium is tested via optimisation where the operating costs of the process for two cooling mediums is minimised. The freshwater has been found to be cost effective cooling medium.