Ismael Fernández Luque, Fernando J. Aguilar Torres, Manuel A. Aguilar Torres, José L. Pérez García, Andrés López Arenas
Journal of Coastal Research 28 (3), 683-699, (1 May 2012) https://doi.org/10.2112/JCOASTRES-D-11-00107.1
KEYWORDS: shoreline change, shoreline detection, shoreline analysis, shoreline definition, coastal erosion–accretion, remote sensing, LIDAR, digital elevation model, extrapolation method, cross-shore profile
FERNÁNDEZ LUQUE, I.; AGUILAR TORRES, F.J.; AGUILAR TORRES, M.A.; PÉREZ GARCÍA, J.L., and LÓPEZ ARENAS, A., 2012. A new, robust, and accurate method to extract tide-coordinated shorelines from coastal elevation models.
The extraction of highly accurate shoreline data is fundamental to carrying out accurate and reliable studies to enhance our understanding of coastal evolution and coastal vulnerability. In our case, shoreline extraction was needed to develop a method based on an extrapolation process because the most suitable height for datum-coordinated shoreline extraction along Spanish coastal areas turned out to be the orthometric datum origin, i.e., the origin of the vertical reference system in Spain. Because of the microtidal nature of the Mediterranean Sea, using this vertical datum is rather troublesome for remotely extracting ground points to apply to traditional shoreline-extraction methods based on interpolation procedures. Because of these difficulties, a new method for shoreline extraction, based on extrapolation from an iterative digital-elevation model, is presented in this article. The Elevation Gradient Trend Propagation method employs the local elevation gradient to estimate the shoreline position by extrapolating the slope until the zero-elevation contour, representing the modeled intersection of the vertical datum and the beach profile, is reached.
The proposed methodology was tested on a Light Detection and Ranging (LIDAR)–derived digital-elevation model, which comprised a coastal area of Almería (Mediterranean Sea, south Spain). The results obtained from the new approach were compared with those provided by the widely known Cross-Shore Profile (CSP) method.
A validation process was conducted for both methods to highlight their advantages and shortcomings. An alternative contour level of 0.4 m was employed as a ground truth because the zero-elevation contour was not available because LIDAR returns under the water surface were unavailable. The validation process showed that the proposed method was more robust and more suitable than CSP method was for microtidal coasts and for data that need to be extrapolated to reach the desired contour level. In addition, the influence of the starting point in applying the elevation extrapolation process was also proven.