Supporting data for:Characteristics and causal paths of refined oil depot accidents in China: A statistical and network-based analysis

This dataset contains the underlying data for the manuscript entitled “SCharacteristics and causal paths of refined oil depot accidents in China: A statistical and network-based analysis”. The data were compiled from publicly available sources, including accident investigation reports, books, and online databases, covering 470 refined oil depot accidents recorded in China between 1954 and 2023.The master data file (Supporting dataset for accident statistical analysis.csv) records China's refined oil product accidents and their data sources from 1954 to 2023, with each row representing an accident. The dataset includes accident characteristics, contributing factors, and detailed source information including official website URLs and published literature. The data contained in this file is sufficient to replicate the statistical results reported in the manuscript. The original accident information was obtained from public government websites, professional accident databases, and two copyrighted Chinese monographs; this minimal dataset allows full reproducibility without requiring access to the original publications.A second core data file (<code>co_occurrence_matrix.csv</code>) contains the directed, weighted adjacency matrix used specifically for the network analysis (UCINET) presented in the manuscript. This matrix quantifies the causal relationships between factors identified in the 31 oil spill-initiated fire and explosion accidents, serving as the direct input for calculating node centrality and identifying critical accident paths.This dataset supports the full reproducibility of the findings, from descriptive statistics to network-based path analysis, and facilitates further research on safety management and accident prevention in hazardous chemical storage facilities.