Operational Detection of Rip Currents Using Shore-Based Microwave Radar Imagery and Artificial Intelligence

Rip currents are a leading cause of coastal drowning accidents worldwide, yet their detection remains challenging due to their significant spatial variability and intermittent nature. While traditional in-situ methods provide high-fidelity but localized insights, optical imagery is often constrained by specific illumination and weather windows. To extend monitoring capabilities across broader areas and diverse environmental conditions, shore-based microwave radar offers a robust alternative. This study investigates the detection and characterization of rip current signatures using time-series microwave radar imagery, focusing on the development of an automated operational technology. Radar imagery captures the observed area by recording variations in backscatter intensity, which are primarily driven by wave breaking and surface roughness. In X-band radar, small-scale surface scatterers, such as breaking gravity waves, facilitate Bragg scattering, which is significantly modulated by rip current dynamics in the surf and outer surf zones.

Our proposed framework adopts a two-stage approach. In the first stage, conventional image processing techniques, including temporal averaging and filtering, are employed to identify candidate rip current patterns from radar sequences. To enhance detection robustness and mitigate false alarms, the second stage introduces an artificial intelligence-based recognition model trained to discriminate rip current signatures from transient wave breaking and background noise. Comparative analyses demonstrate that this AI-assisted approach significantly improves detection consistency across varying sea states. By combining the physical interpretability of traditional image processing with the predictive power of AI, this framework enables near-real-time, continuous rip current monitoring. These results highlight the potential of intelligent microwave radar systems to support coastal safety applications, including early warning systems and real-time hazard mitigation.