A Hybrid Approach Based on Statistical Method and Meta-heuristic Optimization Algorithm for Coastal Aquifer Vulnerability Assessment

Abstract

One of the major shortcomings with GALDIT model is the difficulty in specifying numerical constant values for the rating and weighting system of parameters of interest. The frequency ratio (FR) and genetic algorithm (GA) methods were applied in this research for the first time to improve the rates and weights of the GALDIT model. FR model was used to modify the rates of this model. Additionally, genetic algorithm was used to optimize the weights of GALDIT model based on the hydrological characteristics of the aquifer and the values of TDS parameter. The correlation between hybrid models of GALDIT-FR and GALDIT-GA were obtained as 0.69 and 0.61, respectively, while this correlation was increased up to 0.76 after combining the rates modified by FR statistical method and optimal weights of the genetic algorithm. The results of this model showed that the northwest and west parts of the study area have the highest vulnerability to seawater intrusion. So, it was concluded that a combination of meta-heuristic algorithm and statistical method provides more accurate result in the study region.