Managing a mega-disaster: GIS applications, decision making and spatial data flow between local, state and federal levels in Hurricane Andrew disaster management.

摘要

The spatial information that maps provide is of vital importance to decision makers during disaster management operations. Despite this, maps are a scarce commodity on the local scene during response to natural and technological disasters. For problem solving during crises, current cartological thinking dictates that appropriate visualization tools are needed. Spatial databases existing at all levels of government and owned by many responding groups must become readily available at the local level. A potential solution is the application of computer technology, such as Geographic Information systems (GIS). This report is a data flow case study of the spatial information sharing which occurred in support of a relatively small network of ad-hoc GIS operations producing maps for disaster managers responding to Hurricane Andrew. That storm was a mega-disaster whose estimated costs total {dollar}30 billion. Cartographic concepts (map communication theory, visualization and spatial database models), elements of hazard theory (hazard management, decision making models) and federal government standards were employed to provide the theoretical focus for this research. Through analysis of interviews conducted in person or by telephone with over 80 disaster managers involved with Hurricane Andrew, the flow of spatial information between differing jurisdictions and private and volunteer organizatons was traced. Guided interviews were employed to investigate both formal and informal methods of communication about spatial data. This allowed the sort of in-depth probing imperative for detailed analysis of the communication structure at work in the response to Hurricane Andrew. The goal of this qualitative study was to make clear what people did and what maps they used to do it. Four main subjects were explored in the interviews: decisions, maps, geographic information systems and spatial data flow. Two generic models applicable to both natural and technological disasters resulted: (1) A Hazard Management Mapping Model, and (2) A Chronospatial Disaster Data Model for Rebuilding a Community. Recommendations are made for actions at the community level, changes to the structure of the Federal Response Plan, and future cartographic research.