A fuzzy model integrating shoreline changes, NDVI and settlement influences for coastal zone human impact classification

Abstract

© 2019 Elsevier Ltd. Current approaches for obtaining shoreline change rates suffer from inability to give a specialist interpretation of the numerical results represented by velocities (m/yr). This study proposes a fuzzy model for coastal zone human impact classification that integrates shoreline changes, NDVI, and settlement influences to enhance numerical-linguistic fuzzy classification through Geographical Information System (GIS)'s graphical visualization prowess. The model output representing scores are numbers ranging from zero to one, which are convertible into fuzzy linguistic classification variables; i.e., low, moderate, and high on the one hand. On the other hand, use of GIS through NDVI (Normalized Difference Vegetation Index) provides enhancement through graphic visualization. Using Itamaraca Island in Brazil as an example, multi-temporal satellite images are processed to provide all the required input variables. The resulting output divides the entire island into five sectors representing both quantitative and qualitative outcomes (i.e., fuzzy classification composed of both scores and maps), showcasing the capability of the proposed approach to complement shoreline change analysis through physical (map) interpretation in addition to the frequently used numbers. The proposed fuzzy model is validated using random in-situ samples and high resolution image data that has been classified by a coastal geomorphology specialist. The accuracy of the interpretation show 81% of matches are achievable compared to the results of the fuzzy model. The final results delivered by the proposed fuzzy approach show the complex behavior of the local dynamics, thereby adding useful and substantial information for environmental issues and Integrated Coastal Zone Management.