Groundwater plays a vital role in meeting the demands of the water supply for irrigation, human consumption and industrial needs. Groundwater level simulation gets its importance in predicting and planning of water resource management systems. In this study, groundwater levels of the three major southern districts of India were taken for analysis, the districts are Tirunelveli, Thoothukudi and Kanniyakumari. These districts were situated in the banks of the river Thamirabarani. The historic groundwater levels from the year 1991 to 2019 were taken for this analysis. Quarterly data were considered for this research. So data sets for each district would be 29x4=116. These were feed into the artificial neural network tool and after several training activities a final model for each district is arrived. The predicted values were found to be much closer to the observed values. Co-relation co-efficient value R value of 0.99307, MSE of 0.27 and RMSE of 0.519 denotes the model is well-fitted one to predict future groundwater levels of Tirunelveli district. Similarly, Thoothukudi model showed MSE of 0.24, RMSE of 0.49 and R value equal to 0.9013. Kanniyakumari district model showed MSE of 0.33, RMSE of 0.109 and R value equal to 0.924.

Login to Download PDF

Please login with your Paper ID and password to access the full PDF.

🔑 Login to Download