A semi-continuous water-quality monitoring system was installed in Yunlin Offshore Industrial Park (YOIP), the largest industrial park in Taiwan, in 2007 to provide real-time water-quality information such as pH, water depth, dissolved oxygen, temperature, turbidity, conductivity, and chlorophyll. To interpret the large quantities of high-frequency data generated by this system, information theory was applied for data analysis and extraction of useful information for further coastal water-quality management. Information theory is a branch of applied mathematics that involves the quantification of information. Shannon entropy is a key measure of information that was calculated in this study to reveal the inherent uncertainty of water-quality information. The applicability of Shannon entropy for signaling possible coastal pollution events in the YOIP was explored and results showed that it provides new insight into the inherent uncertainty or randomness of the original data. Specially, when Shannon entropy was high, multiple instable readings were observed for turbidity and salinity. This indicates that Shannon entropy may be a useful new tool for exploratory data analysis. It can be used as a supplementary indicator along with the original environmental data to signify some episodes of water-quality degradation.