Fecal coliform bacteria (FCB) are indicative of the presence of organisms from intestinal tract of humans and other animals. Since FCB can enter the tissues of oysters as they feed by filtering water, oysters grown and harvested in the waters with high concentration of FCB may contain pathogens responsible for the transmission of waterborne diseases. As such, an accurate assessment of FCB count is critical for healthy practices of oyster harvesting. This study proposes to develop a method for accurately predicting the FCB count for use in oyster harvesting in Alabama coastal waters. A method to accurately predict Fecal Coliform levels has been developed that utilizes Bayesian statistics and dynamic time series modeling with salinity, temperature, river stage, and rainfall as regressors. Data from 1978 to the present was used to construct the models. Data were classified based on sampling stations, seasons, and on FCB response to the regressors. Such data were combined in a judicious manner and used in the development of the prior model. Events triggering abnormal behavior in FCB count were flagged for inclusion in the likelihood model. Principles of Bayesian theory were used to combine the prior and the likelihood models to provide an accurate assessment on the FCB count. The proposed approach outperforms the conventionally adopted mechanism for assessing FCB count for oyster harvesting.