Shoreline mapping with cellular automata and the shoreline progradation analysis in Shanghai, China from 1979 to 2008

Abstract

This paper presents a cellular automata (CA) algorithm to extract shorelines from remote sensing images by analysing the edge directional information of the images. Using this algorithm, the tide-coordinated shorelines along the entire coast in Shanghai Municipality of China were extracted and analysed using the multi-temporal Landsat TM images from 1979 to 2008. The shorelines of four sub-regions, including the mainland and three islands (Changxing, Hengsha and Chongming) were analysed along with six areas experiencing drastic shoreline changes. The results show a total progradation of 551.7 km2 along the coastal area of Shanghai over the past 30 years, due to both long-term sediment deposition and short-term land reclamation. Furthermore, both horizontal and vertical displacements along the shorelines were identified. Fractal analyses between the length of the shorelines and the spatial resolution of the images achieved goodness-of-fit (R2) values above 0.6 for the shorelines of the entire Shanghai as well as for each of the four subregions, indicating that the relationship between the length of the shorelines and the spatial resolution of the images accord with the power laws. The fractal dimension values indicate that the shorelines of both Changxing and Chongming Islands were getting regular. The paper also demonstrates that the CA-based shoreline extractor can detect shoreline information of both artificial and muddy coasts from remote sensing images. The shoreline extraction and change analysis tool is valuable not only for shoreline mapping but also for comprehensive coastal management.