Investigating Polarization Effects on annual Sentinel-1 Ship Density Maps through a Ship Polarization Index (SPI)

Synthetic Aperture Radar (SAR) imagery acquired by the Sentinel-1 mission provides an effective solution for maritime surveillance because of its all-weather, day-and-night imaging capabilities. Although most studies have focused on vessel detection from individual SAR scenes, comparatively few have investigated the long-term characteristics of vessel density products derived from different radar polarizations.

This study analyzes one year (2025) of Sentinel-1 dual-polarization (VV and VH) observations to investigate annual vessel density patterns. A pilot area located west of the Strofades Islands (Ionian Sea, Greece) was selected, where mixed maritime activities, including commercial shipping, fishing, and recreational navigation, take place. A total of 185 Sentinel-1 acquisitions from both ascending and descending orbits were processed to generate annual VV and VH vessel density maps.

The two polarization products exhibited strong overall agreement, consistently identifying the principal vessel traffic corridors. However, distinct local differences were observed, with several areas showing pronounced density maxima in only one polarization. These discrepancies were most evident outside the major commercial routes, where smaller fishing vessels and recreational boats are likely to coexist with larger commercial ships.

To quantify these polarization-dependent differences, a Normalized Ship Polarization Index (NSPI) was introduced. The index follows the normalized difference formulation, using the difference between the VV and VH density maps in the numerator and their sum in the denominator. The results indicate that the combined interpretation of VV and VH annual density products provides complementary information on vessel activity and may improve the discrimination of maritime zones dominated by different vessel classes. The proposed methodology offers a novel approach for long-term SAR-based maritime traffic characterization.