Water quality (WQ) is to serious environmental and health concern as a result of worldwide population growth. A predictive model for WQ is essential for the timely prevention and management of water pollution. This study proposes a novel Bio-Inspired Spiking Neural Network (SNN) for identifying water pollution in the environment. The data pre-processing stage finds the missing data and shows the duplicate WQ values through data imputation and data cleaning methods. Also, the data normalization where WQ data are normalised as mounting data into nominal data and transformed the better WQ data using data transformation. Then the WQ data are split as 80% train and 20% test using a data splitting technique. The Bio-Inspired SNN model is trained and tested on processes WQ data, enabling to accurately predict pollution. Implemented using Python software, the WQ data is chosen from WQ dataset and effectively evaluated pollution sources with an improved prediction accuracy of 93%.