Abstract: Apple and grape are the major cash-trees in China and their main producing areas are located in northern China. The common flood and drought disasters of apples and grapes are adopted as a case how to setup a comprehensive database for cash-trees’ disasters with powerfull abilities, which required by the comprehensive prevention technology and the decision-making service platform, for disaster monitoring and early warning, risk assessment,as well as risk transfer in this study. The database is constructed by plentiful data collections and managements, strict data quality controls and multi-source data formats requirements, as well as detailed design for database overall schemes and database interfaces, by involving the open-source MySQL database building techniques with the C/S architecture. Multi-source and multi-type data are well managed by the comprehensive disasters database, such as long-term meteorological and hydrological data, cash-trees’ disaster and orchard’s producing management information, remote sensing data and its derived information, soil properties and basic geographic information, and so on. These multidiscipline data have been stored in different ways managed by the database system. For example, the quantitative/qualitative attribute data has been input into the database tables, whereas the vector and raster datasets have been kept in the computer operation file system formats and indexed by the database table. For the efficient data management and sharing, a sub-system for database managing activities is developed by using the database software developing tools under the supports of GIS function middleware and Quartz open-source components. Within the sub-system, the Role-Based Access Control (RBAC) model has been introduce to assign the database user’s privileges and to control their behaviors for accessing the database tables. Meanwhile, many functions have been developed for task schedules, data collections and updates, data managements and storages, information statistical analyses and extractions, I/O ports accesses and model links from other platforms, as well as the basic data visualization abilities within the sub-system. Under the supports of the completely comprehensive disaster database, the early warning risk and disasters monitoring can been clearly illustrated in the cash-trees' producing areas by using the stored data and in terms of the preliminary disasters’ indicators. Therefor, the successful case study of building up this comprehensive disasters’ database contributes to promote other disaster database construction for cash-trees’ except for apple and grape, or other disasters except for floods and drought, even or other areas except for northern China.

Key words: Cash-trees, Apple, Grape, Multi-source data, Database, Disaster, Flood, Drought