Dynamic Monitoring of Dust Transport and Its Impact on Maritime Visibility in the Red Sea Using Multi-Source Satellite Data and Advanced Deep Learning Approach

The passage of dust over the Red Sea poses considerable obstacles for maritime operations, impacting visibility, navigational safety, and climatic conditions. This study offers a thorough spatiotemporal investigation of dust transport variability from 2015 to 2023, employing multi-source satellite datasets (MODIS, CALIPSO, MERRA-2) and analytical methods, including GIS-based spatial modelling and deep learning algorithm. The integration of these approaches has enabled a high-resolution assessment of the link between wind speed dynamics and fluctuations in aerosol optical depth (AOD). The findings indicated significant interannual and seasonal fluctuations in dust transport strength, with peak AOD values noted in 2015, 2017, and 2022, corresponding with heightened Shamal winds, whereas diminished dust activity was documented in 2016, 2020, and 2023. Seasonal research revealed that dust transport reaches its zenith in spring over the northern Red Sea, summer dust storms prevail in the center basin, and autumn and winter dust activity migrated southward. A robust positive correlation (R² > 0.60) between wind speed and AOD indicates that wind-driven mechanisms were the principal drivers of dust mobilization, while deviations implied the impact of supplementary climatic factors. The Mann-Kendall trend analysis revealed a statistically significant reduction in visibility over the last ten years, highlighting the growing influence of dust storms on maritime safety. This research utilized convolutional neural networks (CNNs) and satellite-derived aerosol data to enhance dust transport predictions and risk evaluation for maritime navigation. The results highlighted the imperative for real-time monitoring systems and adaptive navigation tactics. The peak recorded AOD (1.40) and wind velocities (8.10 ms ^− 1 ) in 2022 underscore the necessity of establishing prediction models for marine risk mitigation in the Red Sea.