Teledetección en sistemas costerosDetección de la línea de costa en morfologías deltaicas

Résumé

The research presented in this Doctoral Thesis proposes an automatic methodology using an algorithm that allows the detection of the coastline in deltaic environments from medium resolution satellite images. The images have been acquired through one of the free access databases of the National Aeronautics and Space Administration of the United States (NASA). To assess the goodness of the methodology, images from the Landsat project were used, from which three of the sensors used on the different satellites of the project were selected. This made it possible to use a large amount of data over a wide time range. In order to evaluate the goodness of the proposed methodology, a statistical comparison was made between the results obtained using the proposed indices, and those obtained from other spectral indices commonly used in the literature to detect bodies of water. This comparison was made by applying each of the indices to three deltas of the Spanish Mediterranean (deltas of the rivers Guadalfeo, Adra and Ebro), which were selected, among other things, for being subject to microtidal conditions and for having climatic and morphological similarities. The results concluded that the methodology with one of the indices proposed in this Doctoral Thesis presented remarkable advantages in terms of efficiency and precision over the most commonly used indices for the detection of coastlines, thus contributing to the expansion of knowledge related to spectral indices in these areas of study.