Sentinel-1 Maritime Surveillance: Testing and Experiences with Long-term Monitoring

Sentinel-1 is the Synthetic Aperture Radar (SAR) satellite of the EU’s Copernicus program for Earth observation. Being a radar, it is very suitable for ship detection. Sentinel-1 produces routinely a large amount of images over the seas and coasts, in particular in Europe and the Arctic but also els...

Full description

Bibliographic Details
Main Authors: SANTAMARIA SERNA Carlos, STASOLLA Mattia, FERNANDEZ ARGUEDAS Virginia, ARGENTIERI Pietro, ALVAREZ ALVAREZ Marlene, VAN WIMERSMA GREIDANUS Herman
Language:English
Published: Publications Office of the European Union 2015
Subjects:
Online Access:https://publications.jrc.ec.europa.eu/repository/handle/JRC98532
https://doi.org/10.2788/090400
Description
Summary:Sentinel-1 is the Synthetic Aperture Radar (SAR) satellite of the EU’s Copernicus program for Earth observation. Being a radar, it is very suitable for ship detection. Sentinel-1 produces routinely a large amount of images over the seas and coasts, in particular in Europe and the Arctic but also elsewhere on the globe. Up to now, software to analyse satellite SAR images for ship detection was limited to handle a few images at the time, so was not up to the high production volume of Sentinel-1. This report describes improvements in JRC’s ship detector SUMO to fully automatically process large amounts of images. With this increased capability, time series of many months of Sentinel-1 images are analysed over two sites in the Arctic, revealing the seasonal relation between ship traffic and sea ice. Also a multi-month data set over the Mediterranean Sea, mapping the shipping patterns there, and a one-year data set over the Western Indian Ocean are analysed. The combination of the Sentinel-1 ship detections with ship positions derived from ship self-reporting systems (AIS and LRIT) results in the quantification and mapping of the non-reporting ship traffic, that is not seen on AIS or LRIT. Beyond mere detection, new methods are tested to estimate the size of the ships from their signature in the Sentinel-1 images, and to allow discrimination between tankers and cargo vessels even if they are of similar size. The occurrence of range ambiguities, and the occurrence SAR imaging effects on the apparent heading of the ship signatures are briefly discussed. The developments and experiences discussed here should further promote Sentinel-1 in becoming a tool for maritime surveillance to the benefit of maritime safety, security and sustainability, including spatial planning. JRC.G.3 - Maritime affairs