A Comparison of Geostatistics and Fuzzy Application for Digital Elevation Model

This paper is continuation work of a research started in 2005. In the previous research, variety of fuzzy digital elevation models was constructed and the best fit model was selected in order to present an alternative in modelling height information. The previous study revealed that fuzzy digital elevation model gave satisfied result compared with the result from TIN (Triangulated Irregular Network). Fuzzy concepts and theories deal with ideas based on possibility approach. This fact leads us to the idea of attempting to apply geostatistical method in Digital Elevation Model as geostatistical theories are based on probability concepts. Kriging is one of the most well-known geostatistical methods and it is said to be an optimal interpolation. Therefore, kriging was selected for this study. There are various kriging techniques but in this study, ordinary kriging is our main concern. Kriging model construction and interpolation are our main focus. Our assumption was simple that we would like to compare resulting maps produced from different techniques such as fuzzy application, geostatistics and TIN. The resulting maps from fuzzy modelling and TIN were ready from our previous research. In this paper, we created a kriged interpolation map in order to include it to the final comparison. In additional, we discuss the strengths and weaknesses of different methods for better understanding of model selection. * Corresponding author