By Dimitris Poursanidis
Seagrasses are traversing the epoch of intense anthropogenic impacts that significantly decrease their coverage and invaluable ecosystem services, necessitating accurate and adaptable, global-scale mapping and monitoring solutions. FORTH and DLR, in the framework of the ECOPOTENTIAL project, join forces for a new approach on satellite data analysis in the context of aquatic environment.
The combination of the cloud computing power of Google Earth Engine with the freely available Copernicus Sentinel-2 multispectral image archive, image composition approaches, and machine learning algorithms such as Support Vector Machines allow us to develop a methodological workflow for large-scale, high spatio-temporal mapping and monitoring of seagrass habitats, focused mainly on the endemic and protected Neptune’s seagrass (Posidonia oceanica).
The present workflow can be easily tuned to be used in new geographical areas, analyzing past satellite data such as the Landsat archive or the new data that are flow in the cloud systems from USGS and ESA. Also to be used with commercial data coming from high resolution satellite data providers but also from aerial imagery and drone systems.
The current implementation allows us to map 2510.1 km2of P. oceanica seagrasses in an area of 40,951 km2 between 0 and 40 m of depth in the Aegean and Ionian Seas (Greek territorial waters) using Support Vector Machines (SVM) to an image composite of 1045 Sentinel-2 frames at 10-m resolution.
The overall accuracy of P. oceanica seagrass habitats features an overall accuracy of 72% following validation by an independent field data set to reduce bias. We envision that the introduced flexible, time- and cost-efficient cloud-based chain will provide the crucial seasonal to interannual baseline mapping and monitoring of seagrass ecosystems in global scale, resolving gain and loss trends and assisting coastal conservation, management planning, and ultimately climate change mitigation.
The current workflow published in the Remote Sensing journal (https://www.mdpi.com/2072-4292/10/8/1227/pdf) uses well established methods in the context of aquatic habitat mapping. The source code, developed in the system of the Google Earth Engine, is not open yet as is still under improvements and tunings, but if you want to apply the method to your area, please contact Dimitris Poursanidis (firstname.lastname@example.org) and he will be happy to discuss further collaborations.
Traganos D., Aggarwal B., Poursanidis D., Topouzelis K., Chrysoulakis N., Reinartz P. (2018). Towards Global-Scale Seagrass Mapping and Monitoring Using Sentinel-2 on Google Earth Engine: The Case Study of the Aegean and Ionian Seas. Remote Sensing, 10(8), 1227. doi: 10.3390/rs10081227.