Towards Global-Scale Seagrass Mapping and Monitoring Using Sentinel-2 on Google Earth Engine: The Case Study of the Aegean and Ionian Seas

Dimosthenis Traganos,Bharat Aggarwal,Konstantinos Topouzelis,Peter Reinartz,Dimitris Poursanidis,Nektarios Chrysoulakis

doi:10.3390/rs10081227

Abstract

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. Here, we combine the cloud computing power of Google Earth Engine with the freely available Copernicus Sentinel-2 multispectral image archive, image composition, and machine learning approaches to develop a methodological workflow for large-scale, high spatiotemporal mapping and monitoring of seagrass habitats. The present workflow can be easily tuned to space, time and data input; here, we show its potential, mapping 2510.1 km2 of 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) after applying support vector machines to a composite of 1045 Sentinel-2 tiles 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.

Highlights

Seagrasses are marine flowering plants that hold important ecological roles in coastal ecosystems since they can form extensive meadows that support high biodiversity
We have developed a methodological workflow within Google Earth Engine which employs a plethora of universally used algorithms in coastal aquatic habitat remote sensing along with image composition and machine learning that could potentially be applied to map and monitor seagrasses globally
We demonstrate its power along with its issues in the Greek Seas, namely the Aegean and Ionian Sea; in a total coastline extent of 40,951 km2, we map 2510.1 km2 of Greek seagrasses (Posidonia oceanica species) between 0 and 40 m of depth applying Support Vector Machines in a pixel-based fashion on 1045 Sentinel-2 tiles

Summary

Introduction

Seagrasses are marine flowering plants that hold important ecological roles in coastal ecosystems since they can form extensive meadows that support high biodiversity Their habitats are found in temperate and tropical ranges [1]; the temperate bioregions include the temperate North Atlantic, North Pacific and Southern Oceans, and the Mediterranean. Seagrasses form a critical marine ecosystem for carbon storage, fisheries production, sediment accumulation, and stabilization [2] They contribute to the function of ocean ecosystems by providing an important nursery area for many species that support offshore fisheries and for adjacent habitats such as salt marshes, shellfish beds, coral reefs, and mangrove forests [3]. The ecosystem functions of the seagrass meadows include: maintenance of genetic variability, resilience of the coastal environment through protection from erosion, and carbon sequestration by removing carbon dioxide from the atmosphere and binding it as organic matter [4]. Carbon storage by seagrasses is essentially an effective removal of carbon dioxide from the ocean–atmosphere system which plays a significant role in the amelioration of climate change impacts [6]

Methods

Results

Discussion

Conclusion