Feasibility Study on Application of Soft Computing Algorithms for Salinity Intrusion Mapping

In recent years, salinity intrusion is a major concern in coastal regions due to the increase in demand for groundwater. However, monitoring and estimating the status of saltwater intrusion is a challenging task. Nowadays, the application of soft computing approaches gaining potential in engineering problems because of the rational structure and capable of analyzing non-linear problems. Saltwater intrusion is a key factor to assess the quality of groundwater, particularly in inland aquaculture areas. This paper presents aims to provide a detailed review study on the potential application of soft computing techniques to assess saltwater intrusion. Soft computing techniques such as artificial neural networks (ANN), support vector machine (SVM), random forest (RF), particle swarm optimization (PSO), fuzzy logic (FL), and genetic algorithm (GA), show desirable prediction results and to some extent replaces the tradition regression models. This paper also explores the case study on application of extreme learning machines (ELM) to predict salinity levels in the Upputeru watershed in Andhra Pradesh.