Abstract: In order to provide guidance for prevention and control of agricultural economic damage in condition of mountainous rainstorm, static risk analysis of rainstorm-agricultural disaster chain is conducted by using complex network theory. Firstly, considering that risk evolution process of storm-agricultural disasters in mountainous areas has a chain law, 39 disaster events were identified based on basic factors like agricultural production, agro-ecological environment and crop itself, and a complex network model was developed with disaster events as nodes and event development connection as connecting edge. Then, node access and clustering coefficients were used to analyze risk evolution process and to identify key nodes in network that caused damage to agricultural economy, and some chain-breaking measures and chain control recommendations were proposed. Finally, risks of six types of secondary disaster chains that caused agricultural economic damage were identified by using the shortest path method. The results show that the overall clustering coefficient is 0.04 with low degree of grouping. There is only a clear "single node-single node" simple transfer relationship between most risk events, and disaster network shows obvious characteristics of small world networks.

Key words: mountainous rainstorm, agricultural disaster chain, complex network, model building, risk analysis