CoastSat: A Google Earth Engine-enabled Python toolkit to extract shorelines from publicly available satellite imagery

TLDR

CoastSat is an open‑source Python toolkit that retrieves time‑series shoreline positions worldwide from over 30 years of publicly available satellite imagery. CoastSat aims to give coastal managers, engineers and scientists a user‑friendly toolkit to monitor and explore coastlines with ~10 m shoreline accuracy. It uses Google Earth Engine to retrieve and pre‑process Landsat and Sentinel‑2 imagery, then applies a supervised classification and sub‑pixel border segmentation algorithm to detect shorelines, with the code and examples hosted on GitHub.

Abstract

CoastSat is an open-source software toolkit written in Python that enables the user to obtain time-series of shoreline position at any sandy coastline worldwide from 30+ years (and growing) of publicly available satellite imagery. The toolkit exploits the capabilities of Google Earth Engine to efficiently retrieve Landsat and Sentinel-2 images cropped to any user-defined region of interest. The resulting images are pre-processed to remove cloudy pixels and enhance spatial resolution, before applying a robust and generic shoreline detection algorithm. This novel shoreline detection technique combines a supervised image classification and a sub-pixel resolution border segmentation to map the position of the shoreline with an accuracy of ~10 m. The purpose of CoastSat is to provide coastal managers, engineers and scientists a user-friendly and practical toolkit to monitor and explore their coastlines. The software is freely-available on GitHub (https://github.com/kvos/CoastSat) and is accompanied by guided examples (Jupyter Notebook) plus step-by-step README documentation.