Grid Search Based SVM Approach for Sea Level Rise

Abstract

Sea level rise is one of the most damaging impacts of climate change. Rising sea levels leads to loss of coastal wetlands, coastal flooding, degradation of coastal ecosystem, sinking of islands and general loss of quality of life. Countries which are located in low-lying areas as well as small islands are concerned that their land areas would be decreased due to inundation and coastal erosion. In this paper, a hybrid wavelet and grid search based support vector machine was used for statistical downscaling of sea level using climatic variables. The results of the hybrid model were compared with the observed sea level data and the results indicate that the hybrid wavelet and grid search based support vector machine can be used for future sea level projection. Thus the effect of sea level rise on low lying island can be studied with the model.