Abstract
Geoinformatic surveillance for spatial and temporal hot-spot detection and prioritization is crucial in the 21st century. A hot spot may be any unusual phenomenon, anomaly, aberration, outbreak, elevated cluster, or critical area. Government agencies require hot-spot delineation and prioritization for monitoring, etiology, management, or early warning. Responsible factors may be natural, accidental, or intentional, with relevance to both infrastructure and security. This article describes multidisciplinary research based on novel methods for hot-spot detection and prioritization, driven by a diverse variety of case studies of interest to agencies, academia, and the private sector. These case studies concern critical societal issues, such as public health, ecosystem health, biodiversity and threats to biodiversity, emerging infectious disease, water management and conservation, carbon sources and sinks, persistent poverty, environmental justices, crop pathogens, invasive-species management, biosurveillance, biosecurity, disease biogeoinformatics, social networks, sensor networks, hospital networks and syndrome surveillance, video mining, early warning, tsunami inundation, remote sensing, and disaster management. Our approach has involved an innovation of the popular circle-based spatial scan statistic. In particular, it employs the notion of an upper level set (ULS) and is accordingly called the upper level set scan statistic system, pointing to the next generation of sophisticated analytical and computational systems, effective for the detection of arbitrarily shaped hot spots along spatiotemporal dimensions. It also involves a novel prioritization scheme based on multiple indicators and stakeholder criteria without having to reduce indicators to a single index using Hasse diagrams and partially ordered sets. It is accordingly called the poset prioritization and ranking system (see Patil & Taillie, 2004a, 2004b). The following Web sites have additional information. 1. http://www.stat.psu.edu/hotspots/ 2. http://www.stat.psu.edu/~gpp/ 3. http://www.digitalgovernment.org/news/stories/2004/1104/1104_hotspots_heyman.jsp