A multi-method comprehensive analysis of the taxonomy-based risk assessment of fishing vessel safety in Guangdong Province

Introduction Fishing vessel safety is critical for the sustainable development of fisheries in Guangdong Province, China. Methods This study systematically assessed the safety status of fishing vessels based on data from 687 accidents between 2019 and 2023, using a multi-method framework that combines the International Maritime Organization (IMO) taxonomy, Kernel Density Estimation (KDE), grid-based accident location statistics, and Exploratory Data Analysis (EDA). Results Key findings reveal that "Collision" accidents accounted for the highest proportion (47.16%), representing the primary accident type; while "Mechanical Damage/Failure" incidents occurred less frequently (7.28%), 98% resulted in severe casualties, highlighting their high hazard potential. Significant differences in vessel length, tonnage, and power output were observed among fishing vessels of different materials (steel vs. fiberglass). Spatial analysis indicates that the coastal areas from Yangjiang and Maoming to Zhanjiang, along with the Pearl River Estuary, constitute extremely high-risk zones. Accident distribution exhibits pronounced spatiotemporal clustering, such as a peak in August and the highest risk occurring at 9:00 AM. Discussion Based on these findings, the study proposes targeted measures including strengthening technical prevention and precise supervision, establishing a dynamic risk-tiered control mechanism, and constructing a data-driven long-term management system. This study not only deepens systematic understanding of fishing vessel safety risks in Guangdong Province but also provides replicable empirical evidence and decision support for relevant fisheries safety authorities.