Surf during typhoon events poses severe threats to coastal infrastructure and public safety. Traditional monitoring approaches, including in situ sensors and numerical simulations, face inherent limitations in capturing surf impacts—sensors are constrained by point-based measurements, while simulations require intensive computational resources for real-time monitoring. Video-based monitoring offers promising potential for continuous surf observation, yet the development of deep learning models for surf detection remains underexplored, primarily due to the lack of high-quality training datasets from typhoon events. To bridge this gap, we propose a lightweight YOLO (You Only Look Once) based framework for real-time surf detection. A novel dataset of 2855 labeled images with surf annotations, collected from five typhoon events at the Chongwu Tide Gauge Station, captures diverse scenarios such as daytime, nighttime, and extreme weather conditions. The proposed YOLOv6n model achieved 99.3% $mA{P}_{50}$ at 161.8 FPS, outperforming both other YOLO variants and traditional two-stage detectors in accuracy and computational efficiency. Scaling analysis further revealed that YOLO models with 2–5 M parameters provide an optimal trade-off between accuracy and computational efficiency. These findings demonstrate the effectiveness of YOLO-based video monitoring systems for real-time surf detection, offering a practical and reliable solution for coastal hazard monitoring under extreme weather conditions.

In recent years, typhoon activity over the South China Sea (SCS) has intensified, and interactions between typhoons and mesoscale eddies profoundly regulate the regional oceanic environment and air–sea energy exchange. To systematically investigate the position- and polarity-dependent eddy responses to typhoon forcing, we developed a typhoon–eddy spatial matching algorithm and analyzed the global mesoscale eddy dataset (2006–2020) combined with China Meteorological Administration (CMA) best-track typhoon records. Composite and correlation analyses were employed to examine variations in the eddy surface available potential energy (SAPE) and sea-surface temperature (SST) within a 7-day window before and after typhoon passage, with the typhoon power dissipation index (PDI) used to quantify storm intensity. Composite results reveal distinct dual-asymmetric responses: (1) Energetically, eddies on the left side of typhoon tracks exhibit overall weakening, with anticyclonic eddies (ACEs) showing more pronounced energy decay; in contrast, right-side eddies undergo significant intensification, and cyclonic eddies (CEs) display stronger enhancement than ACEs. (2) Thermally, all eddy types experience net cooling after typhoon passage, with right-side eddies showing stronger SST reductions than left-side ones, and CEs exhibiting more intense cooling than ACEs. Time-scale correlation analyses further demonstrate that the eddy energy change rate (EECR) of left-side CEs, right-side CEs, and right-side ACEs is positively correlated with PDI, whereas left-side ACEs show no significant correlation. For the SST change rate (SSTCR), all types of eddy events exhibit significant negative correlations with PDI, with weaker correlations for CEs and stronger correlations for ACEs. This study demonstrates that the track-relative position of tropical cyclones and the polarity of pre-existing mesoscale eddies exert a systematic control on the observed eddy responses to tropical cyclone forcing in the SCS. These results provide observational constraints on the asymmetric oceanic responses induced by tropical cyclones and offer insights into the interpretation of typhoon–ocean interaction diagnostics in marginal seas.