Assessing Suitability of GP Modeling for Groundwater Level

Artificial Neural Network and other soft computing techniques have been widely used for modeling groundwater level changes. Emphasis has been laid by different researches in improving the quality of input data which significantly affects the final model. In this study Genetic Programming (GP) is used to model spatial variation of groundwater in Arjuna Nadhi sub basin region. For a limited list of monthly groundwater level data the result indicates that when information from neighbouring wells, which are selected on their appropriation, is incorporated, the modeling accuracy improves significantly. It is also concluded that each region/ zone needs individual modeling irrespective of their geographic proximity.